triene D4 (LTD4) and bovine albumin via the icosanoid carboxyl produced antibodies with comparable affinities for-leukotrienes. C4, D4, and E4 (LTC4, LTD4, ...
Proc. Nati Acad. Sci. USA Vol. 78, No. 12, pp. 7692-7696, December 1981 Immunology
Radioimmunoassay of the leukotrienes of slow reacting substance of anaphylaxis (leukotriene C/leukotriene D/leukotriene E/rat basophil leukemia cell line)
LAWRENCE LEVINE*, RICHARD A. MORGAN*, ROBERT A. LEWISt, K. FRANK AUSTENtt, DAVID A. CLARK§, ANTHONY MARFAT§, AND E. J. COREY§ *Department of Biochemistry, Brandeis University, Waltham, Massachusetts 02254; tDepartment of Medicine, Harvard Medical School, and Department of Rheumatology and Immunology, Brigham and Women's Hospital, Boston, Massachusetts 02115; and IDepartment of Chemistry, Harvard University, Cambridge,
Massachusetts 02138
Contributed by K. Frank Austen, August 21, 1981
ABSTRACT A rabbit immunized with a conjugate of leukotriene D4 (LTD4) and bovine albumin via the icosanoid carboxyl produced antibodies with comparable affinities for-leukotrienes C4, D4, and E4 (LTC4, LTD4, and LTE4) and their I1-trans stereoisomers. The antibodies bound 3H-labeled I1-trans-LTC4 and 11-trans-LTC4 with the same average association constant (K.) of 2.8 X l09 M ' at 370C and were present at a concentration of 0.32 ,ug/ml of the immune rabbit plasma. When 9.5 jl- of anti-LTD4 and 108 pmol of 11-trans-[3H]LTCi (40 Ci/mmol) were incubated in a volume of 300 ,ul with LTC4, LTD4, LTE4, or their 11-trans stereoisomers, 50% inhibition of 11-trans-[3H]LTC4 binding was achieved at levels varying between 0.3 and 0.7 ng. As assessed with synthetic analogs of the natural leukotrienes, the antibodies recognized neither those changes within the 6-sulfidopeptide unit of LTD4 produced by deamination or modest peptide lengthening nor the specific stereochemistry of the A14-cis double bond. However, the antibodies did recognize the triene lipid domain and the position and spatial orientation of the glutathione or cysteinylglycine function. Binding of l1-trans-[3H]LTC4 by anti-LTD4 was not inhibited by glutathione, cystinylbisglycine, arachidonic acid, or 5-hydroxy-6,8,11,14-icosatetraenoic acid, and leukotriene B4 (LTB4) was. about 1/1000th as active as LTC4, LTD4, or LTE4. Mouse lymphoma (WEHI-5) and rat basophil leukemia (RBL-1) cells, when stimulated with calcium ionophore A23187, each produced immunoreactive; leukotrienes; and LTC4, LTD4, and LTE4 from RBL-1 cells were individually quantitated by radioimmunoassay after resolution by high-performance liquid chromatography.
The biological significance of "slow reacting substance of anaphylaxis" (SRS-A) was initially associated with mast cell-dependent reactions because of its appearance with histamine in anaphylactic reactions in tissues (1, 2). The subsequent observations of its immunologic generation by either immune complex-, complement-, and neutrophil-dependent or IgE- and mast celldependent pathways (3) suggested a broader role for it as a mediator of host inflammatory responses. Chemical studies demonstrated that SRS-A was a polar lipid (4), that it. absorbed light in the ultraviolet spectrum, and that it possibly contained sulfur (5, 6). Its generation by immunologic reactions in lung tissue or with the calcium ionophore A23187 in mononuclear cells was stimulated in the presence of cysteine (7, 8). Upon the basis of evidence of chromatographic, physicochemical, and biologic identity of materials generated immunologically and with the ionophore (5, 9, 10), structural* identification was sought with material generated by ionophore activation of mouse mastocytoma cells. Radiolaheled arachidonic acid and [35S]cysteine were incorporated into a pharmacologically active
product, permitting its structural characterization as a 5-hydroxy-6-sulfidopeptidyl icosatetraenoic acid (11). Subsequently, several constituents of SRS-A generated with ionophore or immunologically were defined by comparison with totally synthetic structures as (5S,6R)-5-hydroxy-6-S-glutathionyl-7,9-trans-11,14-cis-icosatetraenoic acid (leukotriene C4, LTC4) (12) and its 6-sulfidocysteinylglycine (leukotriene D4, LTD4) (13-15) and 6-sulfidocysteine (leukotriene E4, LTE4) analogs (16). Immunization of a rabbit with a conjugate of LTD4 and bovine albumin elicited antibodies that bound radiolabeled 11trans-LTC4, and synthetic structural analogs and constituent portions of leukotrienes were used to identify the antigenic domains recognized by the anti-LTD4 rabbit plasma. The antiLTD4 was then used to measure the SRS-A leukotrienes after their cellular generation and resolution by reverse-phase highperformance liquid chromatography (RP-HPLC). MATERIALS AND METHODS Preparation of Leukotrienes and Leukotriene Analogs. The preparative routes to LTC4, LTD4, LTE4, and their respective 11-trans analogs have been described (12, 16, 17). Published methods of synthesis were also used to prepare A14-dihydroLTC4 (12); 6-epi-LTD4 (18); the 12-S-glutathionyl position isomer of LTC4 (19); the 11-S-glutathionyl-12-hydroxy analog of LTC4 (20, 21); the deamino, homocysteinyl, and D-penicillamyl analogs of LTD4 (22); leukotriene B4 (LTB4) (23); and 5-hydroxy6,8,11,14.icosatetraenoic acid (5-HETE) (24, 25). The 7-cisA9""1,14-hexahydro analogs of LTC4 and LTD4 were synthesized from the corresponding LTA4 analogs (12, 21). Production of Immunogen and Antibodies. The dimethyl ester of N-trifluoroacetyl LTD4 (12, 16) was treated with 2.5 eq .of lithium hydroxide in 4:1 (vol/vol) dimethoxyethane/water at 230C for 1.5 hr to yield the corresponding C-1 mono acid (icosanoid carboxyl-free, glycine carboxyl remaining as methyl ester), which was purified by silica gel thin-layer chromatography with 9:1 (vol/vol) methylene chloride/methanol and was obtained in a 56% yield. Reaction of this monoacid with 4 eq of triethylamine and 2 eq of isobutyl chloroformate in a minimum of dry dimethoxyethane at -250C to -30°C for 15 min produced the mixed anhydride of the (glycine)-monomethyl ester of N-trifluoroacetyl LTD4 and the isobutyl ester of carbonic acid, as analyzed by thin-layer chromatography (as above), Abbreviations: 5-HETE, 5-hydroxy-6,8,11,14-icosatetraenoic acid; LTA4, leukotriene A4; LTB4, leukotriene B4; LTC4, leukotriene C4; LTD4, leukotriene D4; LTE4, leukotriene E4; PG, prostaglandin; RIA, radioimmunoassay; RP-HPLC, reverse-phase high-performance liquid chromatography; SRS-A, slow reacting substance of anaphylaxis.
The publication costs ofthis article were defrayed in part by page charge payment. This article must therefore be hereby marked "advertisement" in accordance with 18 U. S. C. §1734 solely to indicate this fact.
f To whom reprint requests should be addressed.
7692
Immunology: Levine et aL
Proc. Natl. Acad. Sci. USA 78 (1981)
7693
with a 90% estimated yield. To this solution at -250C was added analyzed in a liquid scintillation counter (Beckman, LS7500, bovine albumin (Sigma; crystallized and lyophilized, globulinFullerton, CA) with an efficiency of 63%. free) in deionized distilled water (with a 50:1 molar ratio of leuTissue Culture. The sources and growth conditions for the kotriene to protein) as a 40 mg/ml solution. cell lines used in this study were as described (31). Bovine aorta The reaction mixture was gradually warmed to 00C over 1 hr smooth muscle cells were prepared in primary culture by puband stirred at 00C for an additional 1.5 hr. Methanol (3 vol) and lished methods (32). Exponentially growing cells were treated 0.15 M aqueous potassium carbonate (= 100 eq on the basis of with 1 ml of trypsin (0.5 mg)/EDTA (0.2 mg) solution, and 5 leukotriene) were added and the solution was stirred at 230C x 105cells were seeded into each 60 x 10 mm tissue culture for 1 hr to effect hydrolytic cleavage of N-trifluoroacetyl and dish with 4 ml of Eagle's minimal essential medium containing ester functions. After neutralization with glacial acetic acid the 10% fetal bovine serum, 2 mM L-glutamine, penicillin (250 mixture was chromatographed on Sephadex G-25 with 10% (vol/ units/ml), and streptomycin (250 Ag/ml). After a 24-hr incuvol) methanol in water used as eluant. The fractions containing bation, the cells were washed two times with 4 ml ofthe minimal the conjugate (by ultraviolet absorbance detection) were comessential medium. The cells were then incubated with 2 ml of bined and dialyzed against distilled water for 15 hr at 00C. The minimal essential medium containing penicillin and streptoresulting solution ofleukotriene-bovine albumin conjugate was mycin in the presence or absence of the calcium ionophore analyzed by ultraviolet absorbance and after correction of proA23187 (lpg/ml) for 20 min, 1 hr, and 16 hr in three replicates tein absorbance at 280 nm (0.5 mg/ml, A = 0.26) it was defor each time point. Cell culture media were collected, centritermined that the molar ratio of LTD4 to protein was 7. fuged to remove floating cells and cellular debris, and assayed Four 5-month-old New Zealand White rabbits each received by RIA for cyclooxygenase products and leukotrienes. an intramuscular injection of 500 pg of LTD4-bovine albumin RP-HPLC. The Waters model 600A pumps, model 660 solconjugate in complete Freund's adjuvant followed by a subvent programmer, model U6K injector, and a 3.9 X 300 mm cutaneous injection 3 weeks later with 250 Iug of LTD4-bovine C18 fatty acid analysis column (Waters Associates) were used for albumin in incomplete Freund's adjuvant. The rabbits were RP-HPLC' Each sample was run on a linear gradient program bled 10 days later. The single rabbit producing anti-leukotriene from 100% solvent A to 100% solvent B for 100 min at a flow antibodies was injected intramuscularly and subcutaneously at rate of 1 ml/min, and 1-ml fractions were collected. Solvent A 3 months with 250 pg of LTD4-bovine albumin in incomplete consisted of 93.4% 0.01 M phosphate buffer (pH 7.4), 6% methFreund's adjuvant and bled twice weekly until the antibody titer anol (HPLC grade, Fisher), and 0.6% t-amyl alcohol (Aldrich) fell by 50%. Blood was collected in acid/citrate/dextran and (% by volume); solvent B was 99.4% methanol with 0.6% t-amyl centrifuged at 400 x g, and the plasma was separated. alcohol. Preparation of 11-trane-PH]LTC4. 14,15-Ditritiated leukotriene A4 (LTA4) was prepared by the catalytic tritiation RESULTS (Lindlar Pd-Pb catalyst) of 14,15-dehydro-LTA4 [synthesized Coprecipitation of 11-trans-[3H]LTC4 with the anti-leukotriene in a manner analogous to LTA4 itself (12)]. The reaction of 14,15rabbit IgG-goat anti-rabbit IgG complex increased with increasditritiated LTA4 with glutathione yielded both LTC4 and 11ing amounts of rabbit anti-leukotriene plasma (Fig. 1). Of the trans-LTC4, which were resolved by C18 RP-HPLC (17). 5880 cpm of l-trans-[3H]LTC4 added, 1340 cpm (23%) was Radioimmunoassay (RIA). The serologic specificities of antibound by. 9.5 Al of immune rabbit plasma and goat anti-rabbit sera to prostaglandin (PG)E2, PGD2, PGF2,, thromboxane B2, IgG. After incubation of the ligand and rabbit nonimmune 6-keto-PGFia, and 12-HETE have been described (26-29). The plasma for 60 min at 37°C followed by addition of the second RIAs for all of the arachidonic acid metabolites were performed antibody, 200 cpm (3%) were bound; thus, specific binding to in 3.5-ml polypropylene test tubes (no. 535; Sarstedt, Princeton, NJ). The diluent for all reagents was Tris buffer (0.01 M Tris HCl, pH 7.4, containing 0.14 M NaCl and 0.1% gelatin). 3H-Labeled standards for each cyclooxygenase product and 12HETE were purchased from New England Nuclear. Appropriately diluted immune rabbit plasma to LTD4-bovine albumin, 11-trans-[3H]LTC4 (40 Ci/mmol, New England Nuclear; 1 Ci = 3.7 X 1010 becquerels), and either standard compounds or unknown samples were added to the test tubes in a total volume of 300 Al with mixing after each addition; the 10 reaction mixtures were incubated at 37C for 60 min. A 100-il portion of goat anti-rabbit IgG plasma (previously titrated to c. equivalence or antibody excess with respect to the rabbit IgG antigen in the immune plasma) was added, the mixture was shaken, and the rabbit IgG-goat anti-rabbit IgG complex was precipitated overnight at 40C. Carrier rabbit IgG was not added in the RIAs for the leukotrienes because the rabbit anti-leukotriene plasma contained enough IgG for adequate precipitation. In the other RIAs a 100-pl portion of normal rabbit plasma, diluted 1:25, was added as carrier just before addition 8 10 0 2 4 6 of the 100l1 of goat anti-rabbit IgG (30). The immune precipitates were centrifuged at 1500 X g for 60 min at 40C, and the Rabbit plasma, 1.d supernatant fluids were decanted. The precipitates were disFIG. 1. Coprecipitation of 11-trans-[3H]LTC4, as mean cpm, by solved in 200 p1 of 0.1 M NaOH and 2.5 ml of scintillation fluid increasing quantities of anti-LTD4 rabbit plasma (closed bars) or nonwas added to each tube. The tubes were stoppered with polyimmune rabbit plasma (open bars) with goat anti-rabbit IgG. Brackets show the SD for triplicate analyses. ethylene push-in stoppers (Sarstedt), thoroughly mixed, and 5-
Immunology: Levine et aL
7694
Proc. Nad Acad. Sci. USA 78 (1981)
antibodies was 20%. Specific binding was increased to 23% and 25% by incubating the 11-trans-[3H]LTC4-anti-LTD4 reactants at 40C for 24 and 48 hr, respectively, before addition of the goat anti-rabbit IgG. In all subsequent experiments the primary reactants, 1 1-trans-[3H]LTC4 and anti-LTD4, were incubated at 370C for 60 min. Binding of incremental additions of 11-trans-[3H]LTC4 by 9.5 A.l of the rabbit anti-LTD4 plasma is presented as a reciprocal plot in Fig. 2. The combining sites ofthe antibodies in 9.5 jul of immune plasma were saturated with 2220 cpm of 1 1-trans[3H]LTC4. At a counting efficiency of 63%, 9.5 ,1 of the immune plasma contains 0.02 pmol of 11-trans-[3H]LTC4-binding IgG molecules and each ml of undiluted rabbit immune plasma contains 2.1 pmol (0.32 Ag) of specific antibody. An average association constant, Ka, was calculated to be 2.8 x 109 M-1 at 370C. Serologic Specificity of the I1-trans-[3H]LTC4-Anti-LTD4 Reaction. Inhibition of 11-trans-[3H]LTC4-anti-LTD4 binding by several arachidonic acid metabolites and analogs was tested by use of 11-trans-[3H]LTC4 (5880 cpm), 9.5 p.l of rabbit immune plasma, and various amounts of inhibitors. The mixtures were incubated in a volume of 300 A.l for 60 min at 370C before addition of the goat (plasma) anti-rabbit IgG. LTC4, LTD4, and LTE4 in amounts of 0.33, 0.66, and 0.72 ng, respectively, inhibited binding of 11-trans-[3H]LTC4 by anti-LTD4 by 50%, indicating that the antibodies did not sharply discriminate among these 6-sulfidopeptide subunits (Fig. 3, Table 1). Furthermore, certain chemical modifications of the LTD4 6-sulfidopeptide subunit by deamination (deamino-LTD4) and modest chain lengthening (homocysteinyl-LTD4) did not appreciably affect their serologic activity. However, substitution of D-penicillamine for L-cysteine (D-penicillamyl-LTD4) decreased serologic activity to 1/ 10th, requiring 3.7 ng ofthe analog to inhibit binding by 50%. The spatial orientation and the position of the sulfidopeptide function were critical to binding by the antiLTD4, because 6-epi-LTD4 required 11 ng for 50% inhibition and thus reacted only 5% as effectively as LTD4, whereas the 5-hydroxy-12-S-glutathionyl-leukotriene analog reacted only 35
0
30
25
0
0
x
: 20
0
80 c
.0
60
._o
4
40 20 a
0.01
0.1
1
U
10
5
2
4
6
(1/cpm added)
8 x
10
12
104
FIG. 2. Coprecipitation of increasing amounts of 11-trans-[3H]LTC4, of cpm added (875-14,000 cpm), by 9.5 ALI of rabbit antiLTD4 plasma. Each point in the double reciprocal plot is the mean value for analyses performed in duplicate; the duplicate values differed by
