Biochimica et Biophysica Acta. 1075(1991) 162-168 © 1991ElsevierScience PublishersB.V. All rights reserved 0304-4165/91/$03.50 ADONIS 0304416591002398
162
BBAGEN 23586
Development of enzyme immunoassays for leukotrienes using acetylcholinesterase Catherine Antoine
1,3, J e a n - P a u l L e l l o u c h e 2, J a c q u e s M a c l o u f and Philippe Pradelles 1
1.3
J Sercice de Pharmacologic et d'Immunologie, DRIPP, Gifsur Ycette (France), 2 Sercice des Molecules Marquges, DBCM, Commissariat ~ l'Encrgie Atomique, CEN / Saclay, Gif sur Ycette (France) and 3 U 150 INSERM, H~pital Lariboisi~re, Paris (France)
(Received 10 January 1991)
Key words: Enzymeimmunoassay;Leukotriene; Acetylchalinesterase;(E. electrieus) We have developed sensitive solid phase enzyme immunoassays (EIA) to analyze quantitatively lenkotrienes (LTs) using acetylcholinesterase from Electrophorus electricus as a label for LTB4, LTC 4 and LTE 4. However, because of problems specific to LTs, we used different coupling procedures to prepare LTs conjugates necessary for the production of antibodies and for the preparation of enzymatic tracers. For the immunogens, all LTs were coupled to bovine serum albumin using glutaraldehyde (ethylene diamine was used to add an amino group to LTB4). Immunizations in rabbits were done following classical procedures. For the enzymatic tracers, succinimidyl 4-(N-maleimidomethyl)cyclobexane-l-carboxylate was selected to conjugate the LTs via their amino groups to acetylcholinesterase. Titers of the different antisera ranged from 1 : 30000 (LTE4) , 1 : 40000 (LTC 4) to 1 : 50000 (LTB 4) and sensitivities (ICs0) were 5.5 pg, 4.3 pg and 2.4 pg, respectively. Cross reactivities were also examined against other LTs. Sensitivities and specificities of the different systems were dependent on the conditions of incubation (temperature). Validation of the technique was done (i) after spiking known a m o u n t s of LTC 4 in plasma and measuring the substance added after prior extraction and purification, (ii) by analyzing the s u p e r u a t a n t of h u m a n neutrophils suspended in buffer or in plasma, (iii) by measuring LTE 4 in urine. Due to the background provided by these complex matrixes, quantitation was performed after addition of [aH]LTs for recovery, protein precipitation, extraction by Sep-Pak R and purification by HPLC. Measurement of LTs can be dune in biological fluids with the same ease and advantages as other enzyme immunoassays that we have previously developed for eicosanoids analysis.
Introduction Leukotrienes (LTs) represent a class of biologically active lipids that derive from the oxidative metabolism of arachidonic acid [1]. Clarification of their function in allergic, inflammatory disorders or cardiovascular diseases require accurate and reliable quantitative analytical methods. Characterization and quantitation of LTs can be achieved by the use of conventional bioassay although it cannot monitor accurately the production of a mixture of the various LTs without prior chro-
Abbreviations:ACHE,acetylcholinesterase;EIA, enzyme immunoassay; LT, leukotriene; PG, prostaglandin. Correspondence: P. Pradelles, Service de Pharmacologic et d'lmmunologie, DR1PP, CEN/Saclay, 91191 Gif sur Yvette, Cede× France.
matographic separation [2]. Gas c h r o m a t o g r a p h y / m a s s spectrometry is specific and fairly sensitive for LTB 4, however, it requires catalytic desulfuration for peptidoLTs prior to analysis [3]. Radioimmunoassay analysis have turned out to be the most convenient method to measure all LTs [4-6]. However, the obtention of antisera against the various LTs with sensitivities adapted to biological problems has remained difficult because of the limitations in availability of synthetic material necessary to generate antibodies as well as the availability of 3H tracers with specific radioactivity lira;ted to approx. 2.22 T B q / m m o l allowing current ser : ivities (ICs0) of 0.3 pmol [4-6] which are insufficient for some applications. In addition, increasing costs of disposal of radioactivity have become a serious problem for the extensive use of radiolabeled ligands such as in RIAs techniques. We have previously used successfully acetylcholin-
163 esterase from Electrophorus electricus as a label for most eicosanoids providing solid phase enzyme immunoassays with sensitivities equal to or superior to those achieved with ~ZSl radioactive tracers [7,8]. We have undertaken a similar approach for LTs B4, C~ and E 4. However, because of the specific problems raised by these molecules, we have combined a dual strategy of coupling to prepare the protein-LT conjugates necessary for the generation of antibodies and corresponding enzyme tracers. In this work, we will describe the procedures followed for the development of such assays and their validation. Materials and Methods
Materials Glutaraldehyde, anhydrous dimethylformamide (DMF) and methanol were from Merck, Darmstadt. F.R.G. N,N'-Dicyclohexylcarbodiimide (DCC), N-hydroxysucciifimide ester (NHS), S-acetylmercaptosuccinic anhydride, hydroxylamine, HCI, bovine serum albumin {fraction V) were from Sigma Chemicals (St. Louis, MO). Succinimidyl 4-(N-maleimidomethyl)cyclohexane-l-carboxylate (SMCC) was purchased from Pierce Chemicals (Rockford, IL). N-Succinimidyl-Saeetylthioacetate (SATA) was from Calbiochem (San Diego, CA). Two buffers will be used for the couplings throughout the experiments and will be referred to as 'phosphate buffer' (0.1 M potassium phosphate buffer, pH 7.4) and 'borate buffer' (0.1 M borate buffer, pH 9). Acetyleholinesterase (ACHE) was purified by a onestep affinity chromatography as described by Massouli6 and Bon [9]. The tetrameric form of the enzyme (G4 form) was obtained from the crude purified preparation by incubation with trypsin and its maleimidated form was prepared as described in Ref. 10. LTC 4 and LTE 4 were synthesized as described in Refs. 11 and 12. LTB a was a generous gift from Dr. J.E. Gleason (Smith Kline and French, Swedeland, PA). Preparation of a LTB4-NH2 derivative was as follows. LTB 4 (1 tzmol) in 200 ILl acetonitrile was transformed into its lactone after addition of 3 g l of a 3% solution of iso-butyi chloroformate and 2 t~l N,N-diisopropylethylamine (both from Kodak, Rochester, NY). After 45 rain at 22°C, the completion of the reaction was assessed by HPLC and alkaline-induced reversibility into the free carboxyl followed an addition of 0.1 M N a O H to a small aliquot of the solution. After evaporation to dryness, 200 tzl of ethylene diamine (Kodak) were added and the reaction performed overnight at 22°C in the dark. Identity of the derivative was first assessed by (3C/MS; in further reactions, the presence of a ninhydrin-positive stain co-migrating with the derivative by T L C was used.
Coupling procedures Production of antisera. The haptenic nature of LTs requires their coupling to an antigenic molecule (i.e., carrier) in order to elicit antibodies. Because peptidoLTs possess both amino and earboxylic functions to which they can be attached to the carrier the final selection of the method will depend on twe important criteria, the uniqueness of the function carried on thc LT (there is only one amino group/molecule of LT and three or two carboxylie groups for either LTCa or LTE 4) and the final yield of the coupling. In addition, we have used throughout these experiments bovine serum albumin as the carrier. All preliminary tests were performed on LTE 4 and the general approach extrapolated to LTC 4 and LTB 4. Coupling of LTE 4 by its carboxylic groups. LTE 4 (25 nmol) was reacted overnight at 22°C in the dark with 250 nmol of NHS and 25 nmol of DCC in 60 ~.1 of DMF. The ester is reacted with 2 nmol bovine serum albumin in solution in 300/zl borate buffer. Coupling of LTE 4 by its am#~o group. LTE 4 (25 nmol) and bovine serum albumin (2 nmol) are reacted overnight at 22°C in the dark with 0.06% glutaraldehyde in 200 p.I phosphate buffer. Coupling of L TE4 to thiolated albumbt. Bovine serum albumin (0.83 ~mol) in 2 ml of borate buffer is reacted with 40 mmol of S-acetylmercaptosuccinic anhydride for 30 min at 22°C. The thio-ester is hydrolyzed using 3 ml of a ! M hydroxylamine solution (pH 7.2) overnight at 22°C and the modified 'SH'-albumin purified by gel filtration on a G25 (Pharmacia) column (1 × 30 cm). Colorimetric analysis using the method of EIIman [13] at 414 nm revealed that approx, seven thiols are present per molecule albumin. LTE 4 (25 nmol), in 100 tzl of phosphate buffer is mixed with SMCC (25 nmol) in 5 /~1 DMF. After 30 min at 22°C, thiolated albumin (3 nmol) is added in 500/zl borate buffer for overnight reaction. The three conjugates were purified by gel filtration (GF 0.5 column 20 x 1 cm, IBF France) that allows separation of free LTE 4 from LTE 4 covalently linked to albumin. All fractions were measured for their immunoreactivity using an antiserum and enzyme tracer previously prepared (P. Pradelles and J. Maclouf, unpublished data *). Calculation of the percentage of bound vs. free LTE4 showed that the yield of coupling was most efficient using glutaraldehyde method ( > 86%), contrasting with less than 0.1% reaction with the two other methods. Because of the limited amount of our LTs supply, we did not investigate further these
* In preliminaryexperiments, a small amount of an immunogenof LTB4 (kindly provided by Drs. Rokach and Young, Merck Frosst, Montreal, Canada) allowed us to generate enough antiserum to perform the initial tests of immunoreactivirydescribed in Fig. 1.
164 unsuccessful coupling approaches. The glutaraldehyde method was thus selected for LTs B4, C 4 and E 4 using 1, 3.2 and 4.5/~mol of LT, respectively, and respecting the ratio of the different reagents for the coupling. hnmunization. This followed the procedure described by Va'itukaitis [14]. The conjugate (1 mg in 1 ml water) is emulsified with 1 ml of Freund's complete adjuvant (Difco, Detroit, MI) and injected subcutaneously to three rabbits. 6 weeks later, the first booster is performed using the same dose of conjugate and rabbits are bled on a weekly basis and the serum analyzed for titer and sensitivity. When the titer drops, a new booster is performed and the same protocol of follow-up is used. Antisera are kept at 4°C after addition of sodium azide (0.02% final).
Preparation of tracers Different conditions of coupling LTs to AChE were studied by using LTE 4 as a model. Coupling of thiolated L TE 4 to maleimidated enzyme. LTE 4 (8.5 nmol in 100/zl of borate buffer) is added to 17 nmol of SATA in 10/xl DMF and incubated at 22°C for 30 min. The resulting LT-thioester is hydrolysed using 100/zl of 1 M hydroxylamine (pH 7) for 30 min and added to the maleimidated G4 form (0.28 nmol in 800/xl 0.1 M phosphate buffer (pH 6) containing 5 mM EDTA); the reaction proceeds overnight at 22°C. Coupling of LTE 4 to the enzyme using SMCC. LTE 4 (100 nmol in 100/~1 phosphate buffer) is added to 100 nmol of SMCC in 10/zl DMF during 30 min at 22°C in the dark. The (34 form (0.3 nmol in 500 /zl borate buffer) is subsequently added and reacted overnight at 22°C. Each of these conjugates is subsequently purified by gel filtration on a Biogel A15m column (90 × 1.5 cm, Bio-Rad, Richmond, CA) [7]. For each conjugate, the fractions corresponding to the enzyme activity are pooled and their capacity to bind antibodies is further tested as well as displacement of bound activity by unlabeled LTs. The coupling using SMCC was equivalent to the thiolated LTE 4 method. However, the SMCC procedure is much simpler to perform and was subsequently retained to prepare subsequent tracers. Enzyme immunoassay Briefly, this method is a solid phase immunoassay by competition which uses AChE coupled to different LTs as label. All assays are performed in EIA buffer (0.1 M phosphate buffer (pH 7.4) containing 0.4 M NaCI, 1 mM EDTA, 0.1% bovine serum albumin and 0.01% sodium azide). Standards or samples (50 ~1) are added to each well of a 96-well microtiter plate (Nunc 96F with certificate, Denmark) which was previously coated with mouse anti-rabbit IgG antibody solution (2 p.g/well) to separate bound from free [15]. Tracers and specific antisera, are added (50/xl each) at appropriate
dilutions. After incubation at 4°C, 18-48 h, the plates are washed using a 0.05 M phosphate buffer (pH 7.4) containing 0.05% Tween 20, and an automatic washer (Washer 120, Labsystems Oy, Helsinki, Finland). Each well is filled with 200 /xl of the Ellman's reagent [7] consisting of enzymatic substrate (acetylthiocholine, 7.5 | 0 -4 M) and chromogen (5,5'-dithiobis(2-nitrobenzoic acid), 5 - 10 -4 M) in 0.01 M phosphate buffer. After shaking 1-3 h in the dark at 22°C, the absorbance of a yellow colored product in each well is measured at 414 nm using a Multiskan MC spectrophotometer (Flow laboratories). Results are expressed in terms of B/Bo × 100 where B and Bo represent the absorbance measured on the bound enzyme fraction in the presence or in the absence of LTs competitors, respectively. Fitting of the standard curves and calculations are done with a microcomputer (Apple lie) using a linear Iog-logit transformation [16]. Non specific binding (less than 0.1% of the total enzyme activity) is determined, replacing specific antibody by 50 /~! of EIA buffer. •
Purification of the samples Human blood is collected from healthy volunteers who had not taken any drug for at least 1 week before donation. It is anticoagulated with 3.8% sodium citrate (1 vol./9 vol. blood) in polypropylene tubes• Platelet poor plasma is obtained after centrifuging blood for 15 rain at 200 x g. Neutrophils are isolated using the procedure described by Haslett et al. [17] using a plasma PercoU gradient. Neutrophils are suspended in platelet poor plasma at 5 • 106/ml. Platelet-rich plasma is obtained from the neutrophil preparation and neutrophil concentrates are suspended in this medium at the same concentration as above (platelet/neutrophil = 40). Incubations are performed at 37°C after a 5 min temperature equilibration period: calcium ionophore A23187 in ethanol is added to the cells at a final concentration of 15 /~M without exceeding 0.4% of ethanol in the suspension. The cells are gently mixed for a 10 min period and the reaction quenched by addition of ice-cold methanol (3 vol./1 vol. cells); the mixture is left overnight at - 2 0 ° C after addition of 22000 dpm [3H]LTB4 and LTE 4 (Du Pont, Wilmington, DE) (30 and 90 pg, respectively) for quantitation. The supernatant is evaporated to dryness, resuspended in 10% methanol and passed through a C,s Sep-Pak R cartridge (Waters/Millipore, Milford, MA) which has been preconditioned with 5 ml EDTA 0.5% (w/v), 8 ml methanol and finally 10 ml of water. The biological material retained is washed with 10 mi of water and LTs are eluted with 3 ml of methanol. The eluate is dried under reduced pressure and dissolved in the HPLC solvent together with 30 ng of prostaglandin (PG)B 2 (see below). Separation of the LTs is carried out using a LDC (Riviera Beach, FL) HPLC system.
165 The samples are injected onto an Ultrasphere R column (5 p,m, 25 cm × 4.6 mm, Beckman, San Ramon, CA). The sample is eluted using methanol/water/acetic acid (65:35:0.05%. v / v ) adjusted to pH 5.7 with ammonium hydroxide at 1 ml/min. Using this system. PGB2, LTC4, B4 and E 4 elute at 10.4, 11, 19 and 22 m/n, respectively. Between each sample, the system is run 5 min using 100% methanol and re-equilibrated in the initial solvent. PGB 2 serves to assess the reproducibility of the HPLC system. Fractions corresponding to the elution of radioactive LTs are dried in vacuo and assayed d!rectly by EIA. The values are then corrected for extraction recovery. Urine is obtained from a pool of normal volunteers and stored aliquoted at -70°C. Prior to assay, the urine (see volume in the figure legend) is thawed, centrifuged and [3H]LTE4 added as indicated above. All subsequent steps (extraction, HPLC, EIA, calculations) are performed as described in the previous section. Results
Sensitivity and optimization Fig. 1 represents dose-response curves of selected bleedings from antisera against different LTs. The sensitivity is excellent for all systems with concentrations ranging from low picogram values (between 0.7 and 100 pg). Interassay reproducibility was assessed by calculating the coefficient of variation at 50% displacement of tracer on 12 different assays. The mean + S.D. was 5.5 + 0.8 pg, n = 12 and the C.V. 14% for LTE~. Similar values were obtained for other LTs. Because sensitivity is a very important criteria for the quantitatoo 9o 8o 7o
6(3
~ ac 3¢
o,
........
;
.......
pg
'1'o
......
',;;o
Fig. 1. Dose-res~nse emwesof LTB4(m). LTC4 (*) and LTE4 (0) antisera using corresponding LT-AChE tracers. Incubations were carried out at 4"C. Values in parenthesis represent the IC50 (pg). The final dilutions used ,%r the antisera were respectively 1:50000, 1:40~.-~0 and 1:30060.
TABLE I
Effi'ct of temperature on the scnsitiri~, of the diffi,rent EIA of LE~ IC~o. pg / well D:~sc-responsecurves were done at the indicated temperatures by incuhating the plates at 4°C or 22°C(i.e.. room temperature)during 18 h. Separationof bound from free and subsequentstepswere done as described in Materials and Methods. IC5~}~r /well) LTB4
2.4
LTCz
4.3
2.5 5.1
LTE4
5.5
14.5
t/on of LTs, it is worth being optimized. Various procedures can maximize sensitivity [18], among these, we found that the temperature of incubation appears to be important. Table I shows sensitivities (ICs0) obtained using overnight incubation at 22°C and at 4°C. Lowering the temperature of incubation improves significantly the sensitivity for LTE~ and to a minor extent LTC~: however, temperature has no effect on the LTB4 assay.
Specificity Cross react/v/ties of the anti-peptido LTs with other LTs are shown in Table It. As can be seen, for both systems, there is a minimal cross reactivity with respect to LTB 4. Conversely, the cross reactivity of the anti LTB4 antiserum is very low when tested against pept/do LTs. However, the specificity of the peptido-LTs antisera were found to be dramatically affected by the temperature. Incubation at 4°C decreased the specificity of both LTC 4 and LTE 4 vis ~ vis the other peptido-LTs, suggesting that temperature induces a conformational change of the LTs a n d / o r of the antibodies binding sites. Measurement o f L Ts in biological samples Variable amounts of LTC 4 were added to aliquots c ~ a same plasma pool. The samples are analyzed as described in Materials and Methods after addition of [3H]LTC4. In preliminary attempts, we tried to measure directly the LTs but basal values (i.e., plasma alone) were in the same range as when LTC 4 was added to plasma (not shown). We subsequently analyzed the samples after addition of tracer amounts of [3H]LTs prior preci0itation of the proteins, Sep-Pak extraction and HPLC purification (see Materials and Methods). The regression analysis of measured concentrations as a function of added is satisfactory even at low concentrations (Fig. 2). In early experiments, we added gamma glutamyl transpeptidase and leucyl amino peptidase to drive the conversion of residual LTC 4
166 TABLE 11
Effect of temperature on the relatit'e cross.reaction of the different antisera Cross reactions were determined by using the heterologous ligand to displace bound ligand. The percentage corresponds to tile relative mass of beterologous that gives 50% displacement. Incubating conditions were similar to Table I. N.D., not determined. LTB 4 (%)
LTC4 ( % )
LTE 4 (%)
4°C
22°C
4°C
22°C
4oC
22°C
LTB 4
100
100
< 0.1
< 0.l
< 0.1
< 0.l
LTC 4
< 0.1
< 0.1
100
100
79
22
LTD4
< 0.1
< 0.1
82
76
52
20
LTE 4
< 0.1
< 0,1
27
100
100 N.D.
5,6-di-HETE GSH 20-OH-LTB4 20-COOH-LTB4
0.07
0.07
< 0.01
< 0.01
2.7 < 0.01
7.4
N.D.
N.D.
N.D.
< 0.01
< 0.0
< 0.01
< 0.01
0.6
N.D.
N.D.
N.D.
N.D.
< 0.0l
N.D.
N.D.
N.D.
N.D.
into L T E 4 [19]. H o w e v e r , w e o b s e r v e d c o n s i s t e n t l y t h a t p l a s m a c o n t a i n s e n o u g h p e p t i d a s c s to c o m p l e t e t h e conversion within a few minutes. LTs were measured after challenge of human neutrophils stimulated with the calcium ionophore A23187 e i t h e r a l o n e in p l a s m a o r in t h e p r e s e n c e o f p l a t e l e t s . Analysis of the HPLC fractions containing LTE4 and LTB 4 shows a marked increase of both LTs when the n e u t r o p h i l s a r e s t i m u l a t e d in t h e p r e s e n c e o f p l a t e l e t s as c o m p a r e d to n e u t r o p h i l s a l o n e (Fig. 3). I n all c a s e s , t h e v a l u e s o b t a i n e d w e r e well a b o v e t h e b a c k g r o u n d p r o v i d e d by p l a s m a a l o n e (i.e., less t h a n 200 p g / m l f o r LTB, and E,).
,
PJo
+
.
(~i
,_________.____........._ PMNs
+ A 23187
(20.M)
/ [" Blank ~
~
~
~
~
~
ng/lO 6 neutrophils
Y = 0.92 x • o.on4 r -- 0.99 2.
B
P,.,,4s + A 23187
1.5
( 20 ~M)
Blank I
0
LTC4 foun(~ ( n g l r n l )
Fig. 2. Spiking of LTC 4 in human plasma. Known amounts of LTC 4 were added to human plasma (1 ml) and treated as described in Materials and Methods. The samples were then analyzed as LTE 4 (n = 3-6).
0.5 ng/I0e neutrophlls
Fig. 3. Human neutrophils were suspended at 5.106/ml in plasma in the absence or in the presence of platelets (10'~/ml). The incubations were done in 5 ml and the cells were challenged with the ionophore A23187 (20 ~M) during 10 rain. 2 vol. of ice-cold methanol were then added to the incubate and the mixture left overnight at -2{PC, The samples were treated as described in Materials and Methods,
167 1oo 9C
8(
~ 6o 5O
i
4C
~Co
~'o
ld0
i~o
2~o
UmOl creotlnlne Fig. 4. Specificity threshold in relation to urine volume. Volumes of
urine of 2-40 ml correspondingrespectivelyto 9.7-194 ~.mol creatinine were analyzed followingthe procedure described in Materials and Methods. Finally, LTE 4 was measured from increasing volumes of a urine pool. We observed that the concentration of LTE 4 is independent of the volume above a minimal amount of urine corresponding to 50 /zmol creatinine (Fig. 4). This finding suggests that the use of volumes of urine corresponding to creatininuria greater than 0.1 mmol overcomes the problem of non specific values.
Discussion We have succeeded in obtaining antibodies as well as in developing enzyme immunoassays for peptido-LTs by combining separate approaches to attach these haptens to macromolecules. From previous experience, we can ascertain that the success of these assays is due to the use of distinct coupling reagents to generate antibodies and to prepare the enzyme label. Preparation of the immunogen was mainly concerned with optimizing the coupling reaction because of the limited amount of material. Labeling of LTs by the enzyme was rather a compromise between the efficiency of the coupling and preservation of the enzymatic activity. LTs B 4, C 4 and E4 can be coupled to the enzyme using the same chemical groups than for the immunogen preparation. However, caution should be taken so that the enzyme activity is not impaired during the coupling procedure (e.g., homo-bifunctional reagent such as glutaraidehyde polymerize the enzyme and may lead to a complete loss of activity). Additionally, it is our experience that the coupling reagent should be different for the immunogen and for the tracer in order to minimize any undesirable recognition of the coupling moiety (i.e., spacer molecule) on the label by the antibodies. Therefore, distinct methods of coupling have been tried for
the preparation of the immunogen and of the tracer. Finally, the stoichiometry of the reaction of the enzyme with the LT should be ideally 1 : 1 in order to achieve the highest specific activity of the tracer and hence maximize sensitivity. The selection of plasma and urine to validate the technique was guided by the fact these matrices represent the extreme situations for an analyte to be quantirated when it is present in minute amounts. Although HPLC step was required to obtain specificity as previously observed for such measurements [19-21], the values obtained corresponded either to predicted ones when LTC4 was spiked in plasma or to what was reported by others for basal urinary levels of normal subjects (i.e., 0.01-0.03 n g / m m o l creatinine) [21]. In the cellular incubate, the values obtained for LTE 4 are quite compatible with what is known of the nearly total absence of synthetic capacity of peptido-LTs by neutrophils contrasting with the overall production in coincubation with cells such as platelets known to serve as acceptor cells in the process of transceUular biosynthesis [22,23]. The accompanying increase of LTB 4 in the presence of platelets is intriguing although the relatively long incubation ( > l0 min) could also generate variable amounts of 20-OH-LTB4 in one condition of incubation as compared to the other (i.e., with or without platelets). Concluding remarks Although LTs B4, C 4 and E~ may not be the relevant metabolites to evaluate the in vivo production of peptido-LTs, there are a number of in vitro situations when such assays are needed. In most systems, isolated cells only generate LTC4, whereas in all other in vitro situations, the overall synthesis of peptido-LTs can easily be estimated after enzymatic transformation of all peptido-LTs into LTE 4 [19]. In addition, some reports suggest that measurement of urinary LTE 4 may be of interest to assess the production of peptidoLTs under some pathological states [20,21,24]. 1"his technique should provide valuable tools to measure these substances.
Acknowledgements These studies were made possible by financial support from Commissariat ~ I'Energie Atomique and from INSERM and CNRS and are part of an INS E R M / I t a l i a n C N R project.
References 1 Lewis, R.A., Austen, K.F. and Soberman R.J. (1990) N. Engl. J. Med. 323, 645-655.
168 2 Folco, G.C. and Sala, A. (1986) in Biology of lcosanoids (Lagarde, M., ed.), Vo]. 152, pp. 217-226, Editions I.N.S.E.R.M., Paris. 3 Balazy, M. and Murphy, R.C. (1986) Anal. Chem. 58,1098-1101. 4 Levine, L., Morgan, R.A., Lewis, R.A., Austen, K.F., Clark, D.A., Marfat, A. and Corey, E.J. (1981) Proc. Natl. Acad. Sci. USA 78, 7612-7616. 5 Hayes, E.C., Lombardo, D.L., Girard, Y., Maycock, A.L., Rokach, J., Rosenthal. A.L., Young, R.N., Egan, R.W. and Zweering, H.J. (1983) J. Immunol. 131,429-433. 6 Wynalda, W.A., Brashler, J.R., Bach, M.K. Morton, D.R. and Fitzpatrick, F.A. (1984) Anal. Chem. 56, 1862-1865. 7 Pradelles, P., Grassi, J. and Maclouf, J. (1985) Anal. Chem. 57, I170-1173. 8 Pradelles, P., Grassi, J. and Maclouf, J. (1990) Methods Enzymol. 187, 24-34. 9 Massouli~, J. and Bon, S. (1976) Eur. J. Biochcm. 68, 531-539. 10 Mc Laughlin, L., Wei, Y., Stockmann, P.T., Leahy, K.M., Needleman, P., Grassi, J. and Pradellcs, P. (1987) Biochem. Biophys. Res. Commun. 144, 469-476. I l Corey, E.J., Clark, D.A., Goto, G., Maffat, A., Miokowski, C., Samuelsson, B. and Hammarstr6m, S. (1980) J. Am. Chem. Soc. 102, 1436-1439. 12 Rokach, J., Girard, Y., Guindon, Y., Atkinson, J.G., Larue, M., Young, R.N., Masson, P. and Holme, G. (1980) Tetrahedron Lett. 21, 1485-1488.
13 Ellmann, G.L., Courney, K.D., Audres, V. and Featherstoue, R.M. (1961) Bioehem. Pharmacol. 7, 88-95. 14 Vaitukailis, J., Robbins, J.B., Nieschlag, E. and Ross, T. (1971) J. Clin. EnducrinoL 33, 988-990. 15 M6treau, E., Pleau, J.M., Dardenne, M., Bach, J.F. and Pradelles, P. (1987) J. Immu:iol. Methods 102, 233-242. 16 Rodbard, O.X',- ~lidson, W. and Rayford, P (!969)J. Lab. Clin. Mcd. 74, 770-776. 17 Haslet, C., Guthrie, L.A., Kopaniac, M.M., Johnston, R.B. and Henson, P.M. (1985) Am. J. Pathol. 119, 101-110. 18 Ciabattoni, G. (1987) in Radioimmunoassay in Basic and Clinical Pharmacology (Patrono, C. and Peskar, B.A., eds.), pp. 181-191, Springer-Verlag, New York. 19 Heavey, D.J., Soberman, R.J., Lewis, R.A., Spur, B. and Austen, K.F. (1987) Prostaglandins 33, 693-708. 20 Taylor, G.W., Black, P., Turner, N., Taylor, I., Maltby, N.H., Fuller, R,W. and DolleD', D.T. (1989) Lancet i, 584-587. 21 Huher, M., K~istner, S., Sch61merich, J., Gerok, W. and Keppler, D. (1989) Eur. J. Clin. Invest. 19, 53-60. 22 Maclouf, J. and Murphy, R.C. (1988) J. Biol. Chem. 263,174-181. 23 Edenius, C., Heidvall, K. and Lindgren, J.A (1988) Eur. J. Biochem. 178, 81-86. 24 Tagari, P., Rasmussen, J.B., Delorme, D., GiTard, Y., Eriksson, L.-O., Charleson, S. and Ford-Hutchinson, A.W. (1990) Eieosanoids 3, 75-80.
162
BBAGEN 23586
Development of enzyme immunoassays for leukotrienes using acetylcholinesterase Catherine Antoine
1,3, J e a n - P a u l L e l l o u c h e 2, J a c q u e s M a c l o u f and Philippe Pradelles 1
1.3
J Sercice de Pharmacologic et d'Immunologie, DRIPP, Gifsur Ycette (France), 2 Sercice des Molecules Marquges, DBCM, Commissariat ~ l'Encrgie Atomique, CEN / Saclay, Gif sur Ycette (France) and 3 U 150 INSERM, H~pital Lariboisi~re, Paris (France)
(Received 10 January 1991)
Key words: Enzymeimmunoassay;Leukotriene; Acetylchalinesterase;(E. electrieus) We have developed sensitive solid phase enzyme immunoassays (EIA) to analyze quantitatively lenkotrienes (LTs) using acetylcholinesterase from Electrophorus electricus as a label for LTB4, LTC 4 and LTE 4. However, because of problems specific to LTs, we used different coupling procedures to prepare LTs conjugates necessary for the production of antibodies and for the preparation of enzymatic tracers. For the immunogens, all LTs were coupled to bovine serum albumin using glutaraldehyde (ethylene diamine was used to add an amino group to LTB4). Immunizations in rabbits were done following classical procedures. For the enzymatic tracers, succinimidyl 4-(N-maleimidomethyl)cyclobexane-l-carboxylate was selected to conjugate the LTs via their amino groups to acetylcholinesterase. Titers of the different antisera ranged from 1 : 30000 (LTE4) , 1 : 40000 (LTC 4) to 1 : 50000 (LTB 4) and sensitivities (ICs0) were 5.5 pg, 4.3 pg and 2.4 pg, respectively. Cross reactivities were also examined against other LTs. Sensitivities and specificities of the different systems were dependent on the conditions of incubation (temperature). Validation of the technique was done (i) after spiking known a m o u n t s of LTC 4 in plasma and measuring the substance added after prior extraction and purification, (ii) by analyzing the s u p e r u a t a n t of h u m a n neutrophils suspended in buffer or in plasma, (iii) by measuring LTE 4 in urine. Due to the background provided by these complex matrixes, quantitation was performed after addition of [aH]LTs for recovery, protein precipitation, extraction by Sep-Pak R and purification by HPLC. Measurement of LTs can be dune in biological fluids with the same ease and advantages as other enzyme immunoassays that we have previously developed for eicosanoids analysis.
Introduction Leukotrienes (LTs) represent a class of biologically active lipids that derive from the oxidative metabolism of arachidonic acid [1]. Clarification of their function in allergic, inflammatory disorders or cardiovascular diseases require accurate and reliable quantitative analytical methods. Characterization and quantitation of LTs can be achieved by the use of conventional bioassay although it cannot monitor accurately the production of a mixture of the various LTs without prior chro-
Abbreviations:ACHE,acetylcholinesterase;EIA, enzyme immunoassay; LT, leukotriene; PG, prostaglandin. Correspondence: P. Pradelles, Service de Pharmacologic et d'lmmunologie, DR1PP, CEN/Saclay, 91191 Gif sur Yvette, Cede× France.
matographic separation [2]. Gas c h r o m a t o g r a p h y / m a s s spectrometry is specific and fairly sensitive for LTB 4, however, it requires catalytic desulfuration for peptidoLTs prior to analysis [3]. Radioimmunoassay analysis have turned out to be the most convenient method to measure all LTs [4-6]. However, the obtention of antisera against the various LTs with sensitivities adapted to biological problems has remained difficult because of the limitations in availability of synthetic material necessary to generate antibodies as well as the availability of 3H tracers with specific radioactivity lira;ted to approx. 2.22 T B q / m m o l allowing current ser : ivities (ICs0) of 0.3 pmol [4-6] which are insufficient for some applications. In addition, increasing costs of disposal of radioactivity have become a serious problem for the extensive use of radiolabeled ligands such as in RIAs techniques. We have previously used successfully acetylcholin-
163 esterase from Electrophorus electricus as a label for most eicosanoids providing solid phase enzyme immunoassays with sensitivities equal to or superior to those achieved with ~ZSl radioactive tracers [7,8]. We have undertaken a similar approach for LTs B4, C~ and E 4. However, because of the specific problems raised by these molecules, we have combined a dual strategy of coupling to prepare the protein-LT conjugates necessary for the generation of antibodies and corresponding enzyme tracers. In this work, we will describe the procedures followed for the development of such assays and their validation. Materials and Methods
Materials Glutaraldehyde, anhydrous dimethylformamide (DMF) and methanol were from Merck, Darmstadt. F.R.G. N,N'-Dicyclohexylcarbodiimide (DCC), N-hydroxysucciifimide ester (NHS), S-acetylmercaptosuccinic anhydride, hydroxylamine, HCI, bovine serum albumin {fraction V) were from Sigma Chemicals (St. Louis, MO). Succinimidyl 4-(N-maleimidomethyl)cyclohexane-l-carboxylate (SMCC) was purchased from Pierce Chemicals (Rockford, IL). N-Succinimidyl-Saeetylthioacetate (SATA) was from Calbiochem (San Diego, CA). Two buffers will be used for the couplings throughout the experiments and will be referred to as 'phosphate buffer' (0.1 M potassium phosphate buffer, pH 7.4) and 'borate buffer' (0.1 M borate buffer, pH 9). Acetyleholinesterase (ACHE) was purified by a onestep affinity chromatography as described by Massouli6 and Bon [9]. The tetrameric form of the enzyme (G4 form) was obtained from the crude purified preparation by incubation with trypsin and its maleimidated form was prepared as described in Ref. 10. LTC 4 and LTE 4 were synthesized as described in Refs. 11 and 12. LTB a was a generous gift from Dr. J.E. Gleason (Smith Kline and French, Swedeland, PA). Preparation of a LTB4-NH2 derivative was as follows. LTB 4 (1 tzmol) in 200 ILl acetonitrile was transformed into its lactone after addition of 3 g l of a 3% solution of iso-butyi chloroformate and 2 t~l N,N-diisopropylethylamine (both from Kodak, Rochester, NY). After 45 rain at 22°C, the completion of the reaction was assessed by HPLC and alkaline-induced reversibility into the free carboxyl followed an addition of 0.1 M N a O H to a small aliquot of the solution. After evaporation to dryness, 200 tzl of ethylene diamine (Kodak) were added and the reaction performed overnight at 22°C in the dark. Identity of the derivative was first assessed by (3C/MS; in further reactions, the presence of a ninhydrin-positive stain co-migrating with the derivative by T L C was used.
Coupling procedures Production of antisera. The haptenic nature of LTs requires their coupling to an antigenic molecule (i.e., carrier) in order to elicit antibodies. Because peptidoLTs possess both amino and earboxylic functions to which they can be attached to the carrier the final selection of the method will depend on twe important criteria, the uniqueness of the function carried on thc LT (there is only one amino group/molecule of LT and three or two carboxylie groups for either LTCa or LTE 4) and the final yield of the coupling. In addition, we have used throughout these experiments bovine serum albumin as the carrier. All preliminary tests were performed on LTE 4 and the general approach extrapolated to LTC 4 and LTB 4. Coupling of LTE 4 by its carboxylic groups. LTE 4 (25 nmol) was reacted overnight at 22°C in the dark with 250 nmol of NHS and 25 nmol of DCC in 60 ~.1 of DMF. The ester is reacted with 2 nmol bovine serum albumin in solution in 300/zl borate buffer. Coupling of LTE 4 by its am#~o group. LTE 4 (25 nmol) and bovine serum albumin (2 nmol) are reacted overnight at 22°C in the dark with 0.06% glutaraldehyde in 200 p.I phosphate buffer. Coupling of L TE4 to thiolated albumbt. Bovine serum albumin (0.83 ~mol) in 2 ml of borate buffer is reacted with 40 mmol of S-acetylmercaptosuccinic anhydride for 30 min at 22°C. The thio-ester is hydrolyzed using 3 ml of a ! M hydroxylamine solution (pH 7.2) overnight at 22°C and the modified 'SH'-albumin purified by gel filtration on a G25 (Pharmacia) column (1 × 30 cm). Colorimetric analysis using the method of EIIman [13] at 414 nm revealed that approx, seven thiols are present per molecule albumin. LTE 4 (25 nmol), in 100 tzl of phosphate buffer is mixed with SMCC (25 nmol) in 5 /~1 DMF. After 30 min at 22°C, thiolated albumin (3 nmol) is added in 500/zl borate buffer for overnight reaction. The three conjugates were purified by gel filtration (GF 0.5 column 20 x 1 cm, IBF France) that allows separation of free LTE 4 from LTE 4 covalently linked to albumin. All fractions were measured for their immunoreactivity using an antiserum and enzyme tracer previously prepared (P. Pradelles and J. Maclouf, unpublished data *). Calculation of the percentage of bound vs. free LTE4 showed that the yield of coupling was most efficient using glutaraldehyde method ( > 86%), contrasting with less than 0.1% reaction with the two other methods. Because of the limited amount of our LTs supply, we did not investigate further these
* In preliminaryexperiments, a small amount of an immunogenof LTB4 (kindly provided by Drs. Rokach and Young, Merck Frosst, Montreal, Canada) allowed us to generate enough antiserum to perform the initial tests of immunoreactivirydescribed in Fig. 1.
164 unsuccessful coupling approaches. The glutaraldehyde method was thus selected for LTs B4, C 4 and E 4 using 1, 3.2 and 4.5/~mol of LT, respectively, and respecting the ratio of the different reagents for the coupling. hnmunization. This followed the procedure described by Va'itukaitis [14]. The conjugate (1 mg in 1 ml water) is emulsified with 1 ml of Freund's complete adjuvant (Difco, Detroit, MI) and injected subcutaneously to three rabbits. 6 weeks later, the first booster is performed using the same dose of conjugate and rabbits are bled on a weekly basis and the serum analyzed for titer and sensitivity. When the titer drops, a new booster is performed and the same protocol of follow-up is used. Antisera are kept at 4°C after addition of sodium azide (0.02% final).
Preparation of tracers Different conditions of coupling LTs to AChE were studied by using LTE 4 as a model. Coupling of thiolated L TE 4 to maleimidated enzyme. LTE 4 (8.5 nmol in 100/zl of borate buffer) is added to 17 nmol of SATA in 10/xl DMF and incubated at 22°C for 30 min. The resulting LT-thioester is hydrolysed using 100/zl of 1 M hydroxylamine (pH 7) for 30 min and added to the maleimidated G4 form (0.28 nmol in 800/xl 0.1 M phosphate buffer (pH 6) containing 5 mM EDTA); the reaction proceeds overnight at 22°C. Coupling of LTE 4 to the enzyme using SMCC. LTE 4 (100 nmol in 100/~1 phosphate buffer) is added to 100 nmol of SMCC in 10/zl DMF during 30 min at 22°C in the dark. The (34 form (0.3 nmol in 500 /zl borate buffer) is subsequently added and reacted overnight at 22°C. Each of these conjugates is subsequently purified by gel filtration on a Biogel A15m column (90 × 1.5 cm, Bio-Rad, Richmond, CA) [7]. For each conjugate, the fractions corresponding to the enzyme activity are pooled and their capacity to bind antibodies is further tested as well as displacement of bound activity by unlabeled LTs. The coupling using SMCC was equivalent to the thiolated LTE 4 method. However, the SMCC procedure is much simpler to perform and was subsequently retained to prepare subsequent tracers. Enzyme immunoassay Briefly, this method is a solid phase immunoassay by competition which uses AChE coupled to different LTs as label. All assays are performed in EIA buffer (0.1 M phosphate buffer (pH 7.4) containing 0.4 M NaCI, 1 mM EDTA, 0.1% bovine serum albumin and 0.01% sodium azide). Standards or samples (50 ~1) are added to each well of a 96-well microtiter plate (Nunc 96F with certificate, Denmark) which was previously coated with mouse anti-rabbit IgG antibody solution (2 p.g/well) to separate bound from free [15]. Tracers and specific antisera, are added (50/xl each) at appropriate
dilutions. After incubation at 4°C, 18-48 h, the plates are washed using a 0.05 M phosphate buffer (pH 7.4) containing 0.05% Tween 20, and an automatic washer (Washer 120, Labsystems Oy, Helsinki, Finland). Each well is filled with 200 /xl of the Ellman's reagent [7] consisting of enzymatic substrate (acetylthiocholine, 7.5 | 0 -4 M) and chromogen (5,5'-dithiobis(2-nitrobenzoic acid), 5 - 10 -4 M) in 0.01 M phosphate buffer. After shaking 1-3 h in the dark at 22°C, the absorbance of a yellow colored product in each well is measured at 414 nm using a Multiskan MC spectrophotometer (Flow laboratories). Results are expressed in terms of B/Bo × 100 where B and Bo represent the absorbance measured on the bound enzyme fraction in the presence or in the absence of LTs competitors, respectively. Fitting of the standard curves and calculations are done with a microcomputer (Apple lie) using a linear Iog-logit transformation [16]. Non specific binding (less than 0.1% of the total enzyme activity) is determined, replacing specific antibody by 50 /~! of EIA buffer. •
Purification of the samples Human blood is collected from healthy volunteers who had not taken any drug for at least 1 week before donation. It is anticoagulated with 3.8% sodium citrate (1 vol./9 vol. blood) in polypropylene tubes• Platelet poor plasma is obtained after centrifuging blood for 15 rain at 200 x g. Neutrophils are isolated using the procedure described by Haslett et al. [17] using a plasma PercoU gradient. Neutrophils are suspended in platelet poor plasma at 5 • 106/ml. Platelet-rich plasma is obtained from the neutrophil preparation and neutrophil concentrates are suspended in this medium at the same concentration as above (platelet/neutrophil = 40). Incubations are performed at 37°C after a 5 min temperature equilibration period: calcium ionophore A23187 in ethanol is added to the cells at a final concentration of 15 /~M without exceeding 0.4% of ethanol in the suspension. The cells are gently mixed for a 10 min period and the reaction quenched by addition of ice-cold methanol (3 vol./1 vol. cells); the mixture is left overnight at - 2 0 ° C after addition of 22000 dpm [3H]LTB4 and LTE 4 (Du Pont, Wilmington, DE) (30 and 90 pg, respectively) for quantitation. The supernatant is evaporated to dryness, resuspended in 10% methanol and passed through a C,s Sep-Pak R cartridge (Waters/Millipore, Milford, MA) which has been preconditioned with 5 ml EDTA 0.5% (w/v), 8 ml methanol and finally 10 ml of water. The biological material retained is washed with 10 mi of water and LTs are eluted with 3 ml of methanol. The eluate is dried under reduced pressure and dissolved in the HPLC solvent together with 30 ng of prostaglandin (PG)B 2 (see below). Separation of the LTs is carried out using a LDC (Riviera Beach, FL) HPLC system.
165 The samples are injected onto an Ultrasphere R column (5 p,m, 25 cm × 4.6 mm, Beckman, San Ramon, CA). The sample is eluted using methanol/water/acetic acid (65:35:0.05%. v / v ) adjusted to pH 5.7 with ammonium hydroxide at 1 ml/min. Using this system. PGB2, LTC4, B4 and E 4 elute at 10.4, 11, 19 and 22 m/n, respectively. Between each sample, the system is run 5 min using 100% methanol and re-equilibrated in the initial solvent. PGB 2 serves to assess the reproducibility of the HPLC system. Fractions corresponding to the elution of radioactive LTs are dried in vacuo and assayed d!rectly by EIA. The values are then corrected for extraction recovery. Urine is obtained from a pool of normal volunteers and stored aliquoted at -70°C. Prior to assay, the urine (see volume in the figure legend) is thawed, centrifuged and [3H]LTE4 added as indicated above. All subsequent steps (extraction, HPLC, EIA, calculations) are performed as described in the previous section. Results
Sensitivity and optimization Fig. 1 represents dose-response curves of selected bleedings from antisera against different LTs. The sensitivity is excellent for all systems with concentrations ranging from low picogram values (between 0.7 and 100 pg). Interassay reproducibility was assessed by calculating the coefficient of variation at 50% displacement of tracer on 12 different assays. The mean + S.D. was 5.5 + 0.8 pg, n = 12 and the C.V. 14% for LTE~. Similar values were obtained for other LTs. Because sensitivity is a very important criteria for the quantitatoo 9o 8o 7o
6(3
~ ac 3¢
o,
........
;
.......
pg
'1'o
......
',;;o
Fig. 1. Dose-res~nse emwesof LTB4(m). LTC4 (*) and LTE4 (0) antisera using corresponding LT-AChE tracers. Incubations were carried out at 4"C. Values in parenthesis represent the IC50 (pg). The final dilutions used ,%r the antisera were respectively 1:50000, 1:40~.-~0 and 1:30060.
TABLE I
Effi'ct of temperature on the scnsitiri~, of the diffi,rent EIA of LE~ IC~o. pg / well D:~sc-responsecurves were done at the indicated temperatures by incuhating the plates at 4°C or 22°C(i.e.. room temperature)during 18 h. Separationof bound from free and subsequentstepswere done as described in Materials and Methods. IC5~}~r /well) LTB4
2.4
LTCz
4.3
2.5 5.1
LTE4
5.5
14.5
t/on of LTs, it is worth being optimized. Various procedures can maximize sensitivity [18], among these, we found that the temperature of incubation appears to be important. Table I shows sensitivities (ICs0) obtained using overnight incubation at 22°C and at 4°C. Lowering the temperature of incubation improves significantly the sensitivity for LTE~ and to a minor extent LTC~: however, temperature has no effect on the LTB4 assay.
Specificity Cross react/v/ties of the anti-peptido LTs with other LTs are shown in Table It. As can be seen, for both systems, there is a minimal cross reactivity with respect to LTB 4. Conversely, the cross reactivity of the anti LTB4 antiserum is very low when tested against pept/do LTs. However, the specificity of the peptido-LTs antisera were found to be dramatically affected by the temperature. Incubation at 4°C decreased the specificity of both LTC 4 and LTE 4 vis ~ vis the other peptido-LTs, suggesting that temperature induces a conformational change of the LTs a n d / o r of the antibodies binding sites. Measurement o f L Ts in biological samples Variable amounts of LTC 4 were added to aliquots c ~ a same plasma pool. The samples are analyzed as described in Materials and Methods after addition of [3H]LTC4. In preliminary attempts, we tried to measure directly the LTs but basal values (i.e., plasma alone) were in the same range as when LTC 4 was added to plasma (not shown). We subsequently analyzed the samples after addition of tracer amounts of [3H]LTs prior preci0itation of the proteins, Sep-Pak extraction and HPLC purification (see Materials and Methods). The regression analysis of measured concentrations as a function of added is satisfactory even at low concentrations (Fig. 2). In early experiments, we added gamma glutamyl transpeptidase and leucyl amino peptidase to drive the conversion of residual LTC 4
166 TABLE 11
Effect of temperature on the relatit'e cross.reaction of the different antisera Cross reactions were determined by using the heterologous ligand to displace bound ligand. The percentage corresponds to tile relative mass of beterologous that gives 50% displacement. Incubating conditions were similar to Table I. N.D., not determined. LTB 4 (%)
LTC4 ( % )
LTE 4 (%)
4°C
22°C
4°C
22°C
4oC
22°C
LTB 4
100
100
< 0.1
< 0.l
< 0.1
< 0.l
LTC 4
< 0.1
< 0.1
100
100
79
22
LTD4
< 0.1
< 0.1
82
76
52
20
LTE 4
< 0.1
< 0,1
27
100
100 N.D.
5,6-di-HETE GSH 20-OH-LTB4 20-COOH-LTB4
0.07
0.07
< 0.01
< 0.01
2.7 < 0.01
7.4
N.D.
N.D.
N.D.
< 0.01
< 0.0
< 0.01
< 0.01
0.6
N.D.
N.D.
N.D.
N.D.
< 0.0l
N.D.
N.D.
N.D.
N.D.
into L T E 4 [19]. H o w e v e r , w e o b s e r v e d c o n s i s t e n t l y t h a t p l a s m a c o n t a i n s e n o u g h p e p t i d a s c s to c o m p l e t e t h e conversion within a few minutes. LTs were measured after challenge of human neutrophils stimulated with the calcium ionophore A23187 e i t h e r a l o n e in p l a s m a o r in t h e p r e s e n c e o f p l a t e l e t s . Analysis of the HPLC fractions containing LTE4 and LTB 4 shows a marked increase of both LTs when the n e u t r o p h i l s a r e s t i m u l a t e d in t h e p r e s e n c e o f p l a t e l e t s as c o m p a r e d to n e u t r o p h i l s a l o n e (Fig. 3). I n all c a s e s , t h e v a l u e s o b t a i n e d w e r e well a b o v e t h e b a c k g r o u n d p r o v i d e d by p l a s m a a l o n e (i.e., less t h a n 200 p g / m l f o r LTB, and E,).
,
PJo
+
.
(~i
,_________.____........._ PMNs
+ A 23187
(20.M)
/ [" Blank ~
~
~
~
~
~
ng/lO 6 neutrophils
Y = 0.92 x • o.on4 r -- 0.99 2.
B
P,.,,4s + A 23187
1.5
( 20 ~M)
Blank I
0
LTC4 foun(~ ( n g l r n l )
Fig. 2. Spiking of LTC 4 in human plasma. Known amounts of LTC 4 were added to human plasma (1 ml) and treated as described in Materials and Methods. The samples were then analyzed as LTE 4 (n = 3-6).
0.5 ng/I0e neutrophlls
Fig. 3. Human neutrophils were suspended at 5.106/ml in plasma in the absence or in the presence of platelets (10'~/ml). The incubations were done in 5 ml and the cells were challenged with the ionophore A23187 (20 ~M) during 10 rain. 2 vol. of ice-cold methanol were then added to the incubate and the mixture left overnight at -2{PC, The samples were treated as described in Materials and Methods,
167 1oo 9C
8(
~ 6o 5O
i
4C
~Co
~'o
ld0
i~o
2~o
UmOl creotlnlne Fig. 4. Specificity threshold in relation to urine volume. Volumes of
urine of 2-40 ml correspondingrespectivelyto 9.7-194 ~.mol creatinine were analyzed followingthe procedure described in Materials and Methods. Finally, LTE 4 was measured from increasing volumes of a urine pool. We observed that the concentration of LTE 4 is independent of the volume above a minimal amount of urine corresponding to 50 /zmol creatinine (Fig. 4). This finding suggests that the use of volumes of urine corresponding to creatininuria greater than 0.1 mmol overcomes the problem of non specific values.
Discussion We have succeeded in obtaining antibodies as well as in developing enzyme immunoassays for peptido-LTs by combining separate approaches to attach these haptens to macromolecules. From previous experience, we can ascertain that the success of these assays is due to the use of distinct coupling reagents to generate antibodies and to prepare the enzyme label. Preparation of the immunogen was mainly concerned with optimizing the coupling reaction because of the limited amount of material. Labeling of LTs by the enzyme was rather a compromise between the efficiency of the coupling and preservation of the enzymatic activity. LTs B 4, C 4 and E4 can be coupled to the enzyme using the same chemical groups than for the immunogen preparation. However, caution should be taken so that the enzyme activity is not impaired during the coupling procedure (e.g., homo-bifunctional reagent such as glutaraidehyde polymerize the enzyme and may lead to a complete loss of activity). Additionally, it is our experience that the coupling reagent should be different for the immunogen and for the tracer in order to minimize any undesirable recognition of the coupling moiety (i.e., spacer molecule) on the label by the antibodies. Therefore, distinct methods of coupling have been tried for
the preparation of the immunogen and of the tracer. Finally, the stoichiometry of the reaction of the enzyme with the LT should be ideally 1 : 1 in order to achieve the highest specific activity of the tracer and hence maximize sensitivity. The selection of plasma and urine to validate the technique was guided by the fact these matrices represent the extreme situations for an analyte to be quantirated when it is present in minute amounts. Although HPLC step was required to obtain specificity as previously observed for such measurements [19-21], the values obtained corresponded either to predicted ones when LTC4 was spiked in plasma or to what was reported by others for basal urinary levels of normal subjects (i.e., 0.01-0.03 n g / m m o l creatinine) [21]. In the cellular incubate, the values obtained for LTE 4 are quite compatible with what is known of the nearly total absence of synthetic capacity of peptido-LTs by neutrophils contrasting with the overall production in coincubation with cells such as platelets known to serve as acceptor cells in the process of transceUular biosynthesis [22,23]. The accompanying increase of LTB 4 in the presence of platelets is intriguing although the relatively long incubation ( > l0 min) could also generate variable amounts of 20-OH-LTB4 in one condition of incubation as compared to the other (i.e., with or without platelets). Concluding remarks Although LTs B4, C 4 and E~ may not be the relevant metabolites to evaluate the in vivo production of peptido-LTs, there are a number of in vitro situations when such assays are needed. In most systems, isolated cells only generate LTC4, whereas in all other in vitro situations, the overall synthesis of peptido-LTs can easily be estimated after enzymatic transformation of all peptido-LTs into LTE 4 [19]. In addition, some reports suggest that measurement of urinary LTE 4 may be of interest to assess the production of peptidoLTs under some pathological states [20,21,24]. 1"his technique should provide valuable tools to measure these substances.
Acknowledgements These studies were made possible by financial support from Commissariat ~ I'Energie Atomique and from INSERM and CNRS and are part of an INS E R M / I t a l i a n C N R project.
References 1 Lewis, R.A., Austen, K.F. and Soberman R.J. (1990) N. Engl. J. Med. 323, 645-655.
168 2 Folco, G.C. and Sala, A. (1986) in Biology of lcosanoids (Lagarde, M., ed.), Vo]. 152, pp. 217-226, Editions I.N.S.E.R.M., Paris. 3 Balazy, M. and Murphy, R.C. (1986) Anal. Chem. 58,1098-1101. 4 Levine, L., Morgan, R.A., Lewis, R.A., Austen, K.F., Clark, D.A., Marfat, A. and Corey, E.J. (1981) Proc. Natl. Acad. Sci. USA 78, 7612-7616. 5 Hayes, E.C., Lombardo, D.L., Girard, Y., Maycock, A.L., Rokach, J., Rosenthal. A.L., Young, R.N., Egan, R.W. and Zweering, H.J. (1983) J. Immunol. 131,429-433. 6 Wynalda, W.A., Brashler, J.R., Bach, M.K. Morton, D.R. and Fitzpatrick, F.A. (1984) Anal. Chem. 56, 1862-1865. 7 Pradelles, P., Grassi, J. and Maclouf, J. (1985) Anal. Chem. 57, I170-1173. 8 Pradelles, P., Grassi, J. and Maclouf, J. (1990) Methods Enzymol. 187, 24-34. 9 Massouli~, J. and Bon, S. (1976) Eur. J. Biochcm. 68, 531-539. 10 Mc Laughlin, L., Wei, Y., Stockmann, P.T., Leahy, K.M., Needleman, P., Grassi, J. and Pradellcs, P. (1987) Biochem. Biophys. Res. Commun. 144, 469-476. I l Corey, E.J., Clark, D.A., Goto, G., Maffat, A., Miokowski, C., Samuelsson, B. and Hammarstr6m, S. (1980) J. Am. Chem. Soc. 102, 1436-1439. 12 Rokach, J., Girard, Y., Guindon, Y., Atkinson, J.G., Larue, M., Young, R.N., Masson, P. and Holme, G. (1980) Tetrahedron Lett. 21, 1485-1488.
13 Ellmann, G.L., Courney, K.D., Audres, V. and Featherstoue, R.M. (1961) Bioehem. Pharmacol. 7, 88-95. 14 Vaitukailis, J., Robbins, J.B., Nieschlag, E. and Ross, T. (1971) J. Clin. EnducrinoL 33, 988-990. 15 M6treau, E., Pleau, J.M., Dardenne, M., Bach, J.F. and Pradelles, P. (1987) J. Immu:iol. Methods 102, 233-242. 16 Rodbard, O.X',- ~lidson, W. and Rayford, P (!969)J. Lab. Clin. Mcd. 74, 770-776. 17 Haslet, C., Guthrie, L.A., Kopaniac, M.M., Johnston, R.B. and Henson, P.M. (1985) Am. J. Pathol. 119, 101-110. 18 Ciabattoni, G. (1987) in Radioimmunoassay in Basic and Clinical Pharmacology (Patrono, C. and Peskar, B.A., eds.), pp. 181-191, Springer-Verlag, New York. 19 Heavey, D.J., Soberman, R.J., Lewis, R.A., Spur, B. and Austen, K.F. (1987) Prostaglandins 33, 693-708. 20 Taylor, G.W., Black, P., Turner, N., Taylor, I., Maltby, N.H., Fuller, R,W. and DolleD', D.T. (1989) Lancet i, 584-587. 21 Huher, M., K~istner, S., Sch61merich, J., Gerok, W. and Keppler, D. (1989) Eur. J. Clin. Invest. 19, 53-60. 22 Maclouf, J. and Murphy, R.C. (1988) J. Biol. Chem. 263,174-181. 23 Edenius, C., Heidvall, K. and Lindgren, J.A (1988) Eur. J. Biochem. 178, 81-86. 24 Tagari, P., Rasmussen, J.B., Delorme, D., GiTard, Y., Eriksson, L.-O., Charleson, S. and Ford-Hutchinson, A.W. (1990) Eieosanoids 3, 75-80.
