Synthesis of Novel Europium-Labeled Estradiol Derivatives for Time

The O-(5-carboxypentyl)-, O-(4-aminobutyl)-, O-(6-aminohexyl)oximes of 2- and 4-formylestradiol as well as the 4-carboxyethylthioether derivative of e...
0 downloads 0 Views 125KB Size
Download PDF
Bioconjugate Chem. 1999, 10, 325−331

325

ARTICLES Synthesis of Novel Europium-Labeled Estradiol Derivatives for Time-Resolved Fluoroimmunoassays Niko Meltola,*,† Piitu Jauria,†,‡ Petri Saviranta,† and Heikki Mikola§ University of Turku, Medical Physics and Chemistry and Department of Biotechnology, P.O. Box 123, FIN-20521 Turku, Finland, and Wallac Oy, P.O. Box 10, FIN-20101 Turku, Finland . Received January 26, 1998; Revised Manuscript Received November 30, 1998

The O-(5-carboxypentyl)-, O-(4-aminobutyl)-, O-(6-aminohexyl)oximes of 2- and 4-formylestradiol as well as the 4-carboxyethylthioether derivative of estradiol were synthesized. These estradiol derivatives were characterized using IR-, 1H-, and 13C NMR spectroscopy. All synthesized estradiol derivatives were labeled with europium chelates. These “tracers” were purified and tested in a competitive timeresolved fluoroimmunoassay with monoclonal antibody (SSI 57-2) raised against the 6-O-(carboxymethyl)oxime-bovine serum albumin derivative of estradiol. All the tested europium-labeled estradiol4-derivatives were bound by the antibody, whereas tracers linked via position 2 were not recognized by this antibody. It was observed that tracers conjugated via C-4 gave more sensitive standard curves than tracers conjugated via C-6. Especially, the estradiol-4-thioether derivative was found to be highly useful in time-resolved fluoroimmunoassays of estradiol while using this antibody.

INTRODUCTION

In many biochemical applications, haptens coupled covalently to other molecules are needed (1). For example, in affinity chromatography and in immunoassays, haptens coupled covalently onto either a solid matrix, a carrier molecule, or a label molecule are used. The coupling site, the chemical structure of the linkage, and the length of the spacer arm between the hapten and the label or the carrier molecule play important roles in the biochemical recognition reactions (1-3). Usually before coupling, a spacer arm and/or a suitable reactive group has to be introduced to the hapten molecule. In steroid immunoassay procedures and in antibody rising O(carboxyalkyl)oximes, hemisuccinates, carboxyalkyl ethers, and thioethers are frequently used to couple the steroids to proteins or to the label molecule (1, 4). For the raising of anti-steroid antibodies, steroids conjugated to macromolecular carriers have to be used as immunogens. The specificities of the antibodies obtained partly depend on the nature of the linkage and the coupling site of the steroid on the carrier. We have studied the binding properties of a monoclonal antiestradiol antibody (SSI 57-2) using time-resolved fluoroimmunoassay. In these assays, we used estradiol derivatives labeled with europium chelates, which give a strong, long decay-time fluorescence, creating high specific activity in the labeled immunoreagents (5-7). The monoclonal antibody used here has been raised * To whom correspondence should be addressed, Medical Physics and Chemistry. E-mail: [email protected]. Phone: +358 2 333 7057. Fax: +358 2 333 7060. † Department of Biotechnology. ‡ Present address: InnoTrac Diagnostics Oy, Tykisto ¨ katu 6 A7, FIN-20520 Turku, Finland. § Wallac Oy.

against 6-oxoestradiol 6-(O-carboxymethyl)oxime-bovine serum albumin immunogen. This immunogen has a distorted steroid core compared to free estradiol because of the double bonding of the nitrogen to position C-6 (8). The used anti-estradiol antibody has a much higher affinity to derivatives linked via C-6 than to free estradiol (9) mainly due to the bridge effect also described elsewhere (8, 10). It was our aim to design new estradiol derivatives that behave as similarly to free estradiol as possible while using this antibody. Because of the aromatic A ring of estradiol, conjugations to the A ring do not cause any distortions in the core structure. In addition, our previous studies (9) show that this antiestradiol antibody has 90% cross-reactivity to estradiol4-derivative, which means that the antibody binds estradiol and estradiol-4-derivatives by almost the same selectivity. On the basis of the reasons above, we describe here the synthesis of some estradiol 4-alkyloxime and 4-thioether derivatives and their labeling with a europium chelate. These europium-labeled estradiol derivatives, called tracers, were briefly tested in a competitive time-resolved fluoroimmunoassay with the monoclonal anti-estradiol antibody in order to study their binding to the antibody and, thus, to find a useful estradiol derivative to be used as a panning antigen. EXPERIMENTAL PROCEDURES

Materials. The reagents for syntheses were purchased from either Aldrich-Chemie, Fluka, Sigma, or E. Merck and used without further purification. Europium-labeled 6-oxoestradiol 6-(O-carboxymethyl)oxime (abbreviated as E2-6-CMO-Eu), the europium chelates used for labeling (11), and aminooxyalkylamines (12) and aminooxyalkanoic acids (13) were obtained from Wallac. The solvents

10.1021/bc980013q CCC: $18.00 © 1999 American Chemical Society Published on Web 03/13/1999

326 Bioconjugate Chem., Vol. 10, No. 3, 1999

were p.a. grade from either E. Merck, Aldrich-Chemie, or Lab-Scan and were used as received unless otherwise stated. NMR spectra were recorded on either JEOL JNMLA400 or JEOL JNM-A500 spectrometers using deuterated chloroform as a solvent and tetramethylsilane as internal standard. Mass spectra (MS) were recorded on a VG7070E spectrometer and IR spectra on a PerkinElmer 1600 FTIR spectrophotometer. The UV spectrum was recorded on a Shimadzu UV-2100 spectrophotometer. Thin-layer chromatography (TLC) plates and silica gel for short-column chromatography were obtained from E. Merck. Purification of the europium-labeled estradiol derivatives was performed using a Superdex Peptide column (Pharmacia) with UV detection at 276 nm (Uvicord UV Monitor from Pharmacia). The murine antiestradiol antibody; SSI 57-2 (IgG1, κ) was obtained from Statens Seruminstitut (Denmark). Rabbit anti-mouse IgG microtitration plates, 1234 DELFIA fluorometer, DELFIA assay buffer, DELFIA wash solution, and DELFIA enhancement solution were from Wallac. 4- and 2-Formylestra-1,3,5(10)-trien-3,17β-diol (2 and 3). A solution of ethylmagnesium bromide was prepared from magnesium turnings (410 mg, 16.9 mmol) and ethyl bromide (1.27 mL, 16.9 mmol) in sodium-dried diethyl ether (115 mL). 17β-Estradiol (1) (1.00 g, 3.67 mmol) in anhydrous toluene (270 mL) was added dropwise to this solution. Paraformaldehyde (0.88 g, 29.4 mmol) and triethyl phosphate were then added to the mixture. The mixture was heated at 80 °C for 20 h. Hydrochloric acid (5%) was added, and layers were separated. The aqueous layer was extracted with toluene/ diethyl ether (1/1 v/v). Combined organic extracts were washed with water, dried over magnesium sulfate, filtered, and evaporated to dryness. The residue was chromatographed on a short silica gel column. Elution with chloroform (technical grade) gave a mixture of 4and 2-formyl-17β-estradiol (2 and 3) 688 mg (62%). 4-Formyl-17β-estradiol (2). 1H NMR (δ ppm): 12.00 (1H, s, ArOH), 10.38 (1H, s, CHO), 7.49 (1H, d, J ) 8.9 Hz, H-1), 6.79 (1H, d, J ) 8.9 Hz H-2), 3.74 (1H, m, H-17), 0.79 (3H, s, CH3-18). 2-Formyl-17β-estradiol (3). 1H NMR (δ ppm): 10.79 (1H, s, ArOH), 9.81 (1H, s, CHO), 7.43 (1H, s, H-1), 6.70 (1H, s, H-4), 3.74 (1H, m, H-17), 0.80 (3H, s, CH3-18). IR (film): 3413, 2931, 2869, 1653, 1571, 1488, 1239 cm-1. 4-Formylestra-1,3,5(10)-trien-3,17β-diol 4-[O-(5carboxypentyl)]oxime (4) and 2-formylestra-1,3,5(10)-trien-3,17β-diol 2-[O-(5-carboxypentyl)]oxime (5). A mixture of 4- and 2-formyl-17β-estradiol (2 and 3) (52.8 mg, 0.176 mmol) was dissolved in pyridine (0.5 mL), and 6-(aminooxy)hexanoic acid hydrobromide (40.0 mg, 0.176 mmol) was added. According to TLC (dichloromethane/methanol 10/1, v/v) the reaction occurred immediately. The reaction mixture was stirred overnight at room temperature and hydrochloric acid was added. The solution was extracted with ethyl acetate. The organic layer was washed with water and dried over sodium sulfate. After filtration, the solution was evaporated to dryness and purified by preparative TLC (2 mm silica gel plate) using dichloromethane and methanol (10/ 1, v/v) as an eluent. Extraction with dichloromethane/ methanol (5/1, v/v) gave a mixture of desired 4- and 2-carboxypentyloxime derivatives (4 and 5) 48.8 mg (64%). 4-Carboxypentyloxime Derivative (4). 1H NMR (δ ppm): 10.34 (1H, s, ArOH), 8.54 (1H, s, -CHdN-), 7.25 (1H, d, J ) 8.8 Hz, H-1), 6.80 (1H, d, J ) 8.8 Hz, H-2), 4.16 (2H, m, dN-O-CH2-), 3.74 (1H, t, H-17), 0.78 (3H, s, CH3-18).

Meltola et al.

2-Carboxypentyloxime Derivative (5). 1H NMR (δ ppm): 9.66 (1H, s, ArOH), 8.11 (1H, s, -CHdN-), 7.02 (1H, s, H-1), 6.69 (1H, s, H-4), 4.16 (2H, m, dN-OCH2-), 3.74 (1H, t, H-17), 0.77 (3H, s, CH3-18). IR (film): 3430, 2933, 2870, 1728, 1393, 1260 cm-1. 4-Formylestra-1,3,5(10)-trien-3,17β-diol 4-[O-(6Aminohexyl)]oxime (8) and 2-formylestra-1,3,5(10)trien-3,17β-diol 2-[O-(6-aminohexyl)]oxime (9). A mixture of 4- and 2-formyl-17β-estradiol (2 and 3) (100 mg, 0.33 mmol) was dissolved in pyridine (1.5 mL), and 6-(aminooxy)hexylamine hydrochloride (379 mg, 1.66 mmol) was added. The reaction was complete in 1 h. Pyridine was evaporated and the residue was purified by preparative TLC using chloroform/methanol (10/3, v/v) as eluent. Extraction with the same solvent gave a mixture of 4- and 2-aminohexyloxime derivatives (8 and 9) 75 mg (54%). 4-Aminohexyloxime Derivative (8). 1H NMR (δ ppm): 8.55 (1H, s, -CHdN-), 7.25 (1H, d, J ) 8.8 Hz, H-1), 6.80 (1H, d, J ) 8.8 Hz, H-2), 4.15 (2H, m, dN-OCH2-), 3.70 (1H, t, H-17), 0.77 (3H, s, CH3-18). 2-Aminohexyloxime Derivative (9). 1H NMR (δ ppm): 8.12 (1H, s, -CHdN-), 7.04 (1H, s, H-1), 6.69 (1H, s, H-4), 4.15 (2H, m, dN-O-CH2-), 3.70 (1H, t, H-17), 0.77 (3H, s, CH3-18). IR (film): 3376, 2928, 2863, 1628, 1385, 1265 cm-1. 4-Formylestra-1,3,5(10)-trien-3,17β-diol 4-[O-(4aminobutyl)]oxime (10) and 2-formylestra-1,3,5(10)trien-3,17β-diol 2-[O-(4-aminobutyl)]oxime (11). A mixture of 4- and 2-formyl-17β-estradiol (2 and 3) (30 mg, 0.1 mmol) was dissolved in pyridine (1.0 mL), and 4-(aminooxy)butylamine hydrochloride (28 mg, 0.2 mmol) was added. The reaction mixture was stirred overnight at room temperature. Pyridine was evaporated and the residue was purified by preparative TLC using chloroform/ methanol (10/3, v/v) as eluent. Extraction with the same solvent gave a mixture of 4- and 2-aminobutyloxime derivatives (10 and 11) 18 mg (47%). 4-Aminobutyloxime Derivative (10). 1H NMR (δ ppm): 8.55 (1H, s, -CHdN-), 7.25 (1H, d, J ) 8.8 Hz, H-1), 6.81 (1H, d, J ) 8.8 Hz, H-2), 4.17 (2H, m, dN-OCH2-), 3.73 (1H, t, H-17), 0.78 (3H, s, CH3-18). 2-Aminobutyloxime Derivative (11). 1H NMR (δ ppm): 8.12 (1H, s, -CHdN-), 7.03 (1H, s, H-1), 6.70 (1H, s, H-4), 4.17 (2H, m, dN-O-CH2-), 3.73 (1H, t, H-17), 0.78 (3H, s, CH3-18). IR (film): 3332, 2923, 2871, 1627, 1385, 1265 cm-1. 4,5-Epoxy-17β-hydroxyestran-3-one (17). 19-Nortestosterone (16) (980 mg, 3.57 mmol) was dissolved in methanol (100 mL) and the solution was cooled to 5 °C. Aqueous NaOH (5.4 mL, 4 M) and H2O2 (0.54 mL, 30%) were added, and the mixture was allowed to stand at 5 °C for 3.5 h. The solution was acidified with glacial acetic acid (2 mL), concentrated to 30 mL, diluted with water, and extracted with ethyl acetate. The organic layer was washed with NaHCO3 (5% aq) and water, dried over sodium sulfate, and evaporated to dryness. The crude product (970 mg, 93%) was used for the next synthesis step without further purification. 1H NMR (δ ppm): 3.66 (1H, t, J ) 8.5 Hz, H-17), 3.04 (1H, s, H-4), 0.78 (3H, s, CH3-18). 13C NMR (δ ppm): 206.82 (C-3), 81.56 (C-17), 67.50 (C-5), 61.75 (C-4). IR (film): 1705 cm-1. 4,5-Epoxy-17β-hydroxyestr-1-en-3-one (18). The crude 4,5-epoxy-17β-hydroxyestran-3-one (17) (900 mg, 3.10 mmol) was dissolved in 2-methyl-2-propanol (200 mL). Acetic acid (3 mL, glacial) and selenium dioxide (688 mg, 6.20 mmol) were added to the solution. The mixture was refluxed under nitrogen atmosphere for 42 h, filtered

Eu-Labeled Estradiol Derivatives for Immunoassays

through a Celite layer, and the filtrate was evaporated to dryness. The residue was dissolved in ethyl acetate, washed 3 times with NaHCO3 (5% aq), washed once with water, and dried over sodium sulfate. After removal of the solvent, the product was purified by short-column chromatography on silica gel (petroleum ether/ethyl acetate, 5/2, v/v). Yield: 382 mg (43%). 1 H NMR (δ ppm): 6.77 (1H, dd, J ) 10.5 and 5.4 Hz, H-1), 5.95 (1H, dd, J ) 10.5 and 1.2 Hz, H-2), 3.65 (1H, t, J ) 8.5 Hz, H-17), 3.24 (1H, s, H-4), 2.61 (1H, dd, J ) 10.7 and 5.4 Hz, H-10), 0.82 (3H, s, CH3-18). 13C NMR (δ ppm): 196.03 (C-3); 148.32 (C-1); 124.95 (C-2); 81.52 (C17); 63.86 (C-5); 61.91 (C-4). IR (film): 1682 cm-1. 3,17β-Dihydroxyestra-1,3,5(10)-trien-4-(2-carboxyethyl)thioether (19). 4,5-Epoxy-17β-hydroxyestr-1-en3-one (18) (415 mg, 1.44 mmol) was dissolved in methanol (4 mL), and Tris-HCl-buffer (4 mL, 0.2 M, pH 7.5) and mercaptopropionic acid (250 µL, 3.5 mmol) were added. The pH of the reaction mixture was adjusted to 9.0 with sodium hydroxide, and the mixture was stirred at 62 °C for 20 h. The mixture was washed with ethyl acetate, and the product was precipitated from aqueous layer by adjusting pH to 2.5 with hydrochloric acid. The product was filtered off, washed with water and dried in a vacuum desiccator. Yield: 411 mg (76%). 1H NMR (δ ppm): 7.42 (1H, d, J ) 8.5 Hz, H-1), 6.81 (1H, d, J ) 8.5 Hz, H-2), 3.70 (1H, t, J ) 8.5 Hz, H-17), 2.88 (2H, t, J ) 7.1 Hz, Ar-S-CH2-), 2.54 (2H, t, J ) 7.1 Hz, -CH2COOH), 0.77 (3H, s, CH3-18). 13C NMR (δ ppm): 174.19 (-COOH); 155.31 (C-3); 141.45 (C-5); 132.95 (C-10); 127.80 (C-1); 116.93 (C-4); 111.82 (C-2); 81.22 (C-17). IR (film): 1700 cm-1. MS: 376 (M+); 304 (M - CH2CH2COOH); 272 (M - SCH2CH2COOH). UV (EtOH): λmax ) 293 nm. Labeling of Carboxy and Amino Derivatives of Estradiol with Europium Chelate. The carboxy derivatives of estradiol were labeled with 4-aminobenzyldiethylenetriaminetetraacetic acid europium chelate in a mixture of dioxane and 4-morpholineethanesulfonic acid buffer (0.5 M, pH 5.5) using 1-(3-dimethylaminopropyl)3-ethylcarbodiimide as an activator. The reaction mixture was stirred for 2 h at room temperature, evaporated to dryness and the product was purified by preparative TLC using acetonitrile/water (2/1, v/v) as an eluent (3). The amino derivatives of estradiol were labeled with 4-isothiocyanatobenzyldiethylenetriaminetetraacetic acid europium chelate in pyridine/water/triethylamine (90/15/ 1, v/v/v). After reaction (2 h), the solution was evaporated to dryness and the product was purified by preparative TLC using acetonitrile/water (2/1, v/v) as an eluent (4). All the labeled estradiol derivatives were further purified on a Superdex Peptide column using 0.05 M TrisHCl buffer [pH 7.5, containing 0.05 M NaCl and 20% (v/ v) acetonitrile] as eluent. The fractions were detected by UV absorption at 276 nm and checked for europium content by DELFIA enhancement solution and timeresolved fluorometer as described later. The chromatogram of europium-labeled estradiol 4- and 2-carboxypentyloxime derivatives is shown by way of an example in Figure 1. The retention times of all europium-labeled estradiol derivatives obtained using a flow rate of 1 mL/ min are given in Table 1. Time-Resolved Fluoroimmunoassays. The synthesized and purified europium-labeled estradiol derivatives were tested in a time-resolved fluoroimmunoassay (4). Assays were carried out by adding 200 µL of antiestradiol antibody (5 ng/mL) to prewashed rabbit antimouse IgG microtitration wells and, after a 2 h incubation, washing the wells four times with wash solution

Bioconjugate Chem., Vol. 10, No. 3, 1999 327

Figure 1. Purification of europium-labeled estradiol-4- and 2-carboxypentyloximes (E2-4-CPO-Eu and E2-2-CPO-Eu) (6, 7). Table 1. Retention Times of Europium-Labeled Estradiol Derivatives at Purification with Superdex Peptide Column

name

abbreviation

retention time (min)

4-formylestra-1,3,5(10)-trien-3,17βdiol 4-[O-(5-carboxypentyl)] oxime europium chelate 2-formylestra-1,3,5(10)-trien-3,17βdiol 2-[O-(5-carboxypentyl)] oxime europium chelate 4-formylestra-1,3,5(10)-trien-3,17βdiol 4-[O-(6-aminohexyl)] oxime europium chelate 2-formylestra-1,3,5(10)-trien-3,17βdiol 2-[O-(6-aminohexyl)] oxime europium chelate 4-formylestra-1,3,5(10)-trien-3,17βdiol 4-[O-(4-aminobutyl)] oxime europium chelate 2-formylestra-1,3,5(10)-trien-3,17βdiol 2-[O-(4-aminobutyl)] oxime europium chelate estra-1,3,5(10)-trien-3,17β-diol 4-(2carboxyethyl)thioether europium chelate

E2-4-CPO-Eu

63

E2-2-CPO-Eu

75

E2-4-AHO-Eu

91

E2-2-AHO-Eu

106

E2-4-ABO-Eu

55

E2-2-ABO-Eu

61

E2-4-CET-Eu

30

(DELFIA Platewash). The tracer solution (200 µL) was added, and after 2 h incubation, the unbound tracer was washed away. Enhancement solution (200 µL) was dispensed into the wells and incubated for 30 min in order to release the europium ion from the chelate into the solution and enhance the fluorescence. The enhanced fluorescence signal was then measured with the 1234 DELFIA research fluorometer. All the dilutions of antibody, estradiol and tracers were made in assay buffer and all the measurements were made in duplicate. The affinity constants were determined by Scatchard analysis (14) of binding data obtained from the immunoassays described above. The amount of binding sites was titrated with increasing concentrations (0.05-20 nM) of labeled estradiol derivatives.

328 Bioconjugate Chem., Vol. 10, No. 3, 1999

Meltola et al.

Scheme 1. Synthesis of Europium-Labeled Estradiol-4- and 2-Carboxypentyloximes (6, 7)

The standard curves for estradiol were obtained with competitive fluoroimmunoassay using double antibody technique (4), in which the anti-estradiol antibody was captured on to the solid phase by secondary antibody (anti-mouse IgG), whereafter estradiol standards were added together with the tracer to the wells. Estradiol inhibits the binding of the labeled derivative, resulting in a descending standard curve. RESULTS AND DISCUSSION

Syntheses. On the basis of our previous good experiences with oxime-type spacers (4, 12, 15), we synthesized estradiol-4 derivatives starting with the introduction of a formyl moiety into position 4. Selective formylation of estradiol into position 2 has been published by Brueggemeier et al. (16); however the regioisomeric 4-formylestradiol could not be obtained by this route. Pert et al. (17) have published the synthetic route for 2- and also for 4-formylestradiol. The synthesis of 4-formylestradiol seemed laborious, and the overall yield was quite low. The procedure that Xie et al. (18) have used for the formylation of 17R-ethynylestradiol seemed attractive, because in this procedure there is no need of protective groups, and both 2- and 4-formyl-17R-ethynylestradiol were obtained in a single step in relative good yield. We used the same procedure for the synthesis of 2- and 4-formylestradiol (Scheme 1). The reaction occurred smoothly and the product obtained was as expected, a mixture of 4-formylestradiol (2) and 2-formylestradiol (3) with a total yield of 62%. Total separation of these two isomers was laborious and unnecessary at this stage. For partial purification, we used a short silica gel column where the faster moving 4-formylestradiol (2) could be enriched from 1/5 to 3/2 for further reactions.

Introduction of the oxime linkers (Schemes 1 and 2) was performed by dissolving a mixture of formylestradiol into pyridine and adding aminooxy compounds into the solution (12). The reaction was complete in a few minutes due to the relatively high reactivity of the salicylaldehyde-type formylestradiol. The reaction mixture was stirred at room temperature for longer period to make sure there was no starting material left. The reaction conditions were not optimized and better results may be obtained by using another solvent or temperature. All estradiol-oxime derivatives, except for europium-labeled ones, were characterized as a mixture of two isomers using 1H NMR and IR spectroscopy. The thioether derivative (19) was synthesized (Scheme 3) starting from 19-nortestosterone (16) using previously published methods (19, 20). The 19-nortestosterone (16) was converted to its epoxy derivative (17) by oxidation with alkalic hydrogen peroxide. The 4,5-epoxy-17βhydroxyestran-3-one was obtained in 93% yield and was used for the next synthesis step without further purification. Oxidation with selenium oxide gave the desired 4,5epoxy-17β-hydroxyestr-1-en-3-one (18) with 43% yield after purification by short-column chromatography. The previous two reactions were scaled up and compounds 17 and 18 were obtained in approximately the same yields (90 and 40%) as before. The mercaptopropionic acid was introduced into the 4,5-epoxy-17β-hydroxyestr-1-en3-one (18) by modifying the method of Ghaffari et al. (21). The product (19) was obtained in 76% yield. The estra1,3,5(10)-trien-3,17β-diol 4-(2-carboxyethyl)thioether (19) was characterized using 1H NMR, 13C NMR, IR, and UV spectroscopy. Mass spectrometry was also used to confirm the desired structure. All the synthesized estradiol amino and carboxy de-

Eu-Labeled Estradiol Derivatives for Immunoassays

Bioconjugate Chem., Vol. 10, No. 3, 1999 329

Scheme 2. Synthesis of Europium-Labeled Estradiol-4- and 2-aminoalkyloximes (12-15)

Scheme 3. Synthesis of Europium-Labeled Estradiol-4-(2-carboxyethyl)thioether (20)

rivatives were labeled with europium chelate and purified using first preparative TLC and then HPLC on a Superdex Peptide column (Table 1 and Figure 1). Tracer Testing. For testing of the tracers, we used a buffer-based model system. The results from the affinity measurements, analyzed according to the Scatchard equation (14), are shown in Figure 2. As expected, the affinity of the anti-estradiol antibody to europium-labeled 6-oxoestradiol-6-CMO used as an internal standard was the highest because it was the immunogen used to raise the antibody. Estradiol-4-derivatives have considerably

lower affinity as shown in Figure 2 and Table 2. In general, antibodies are expected to have lower affinity to steroids, which differ from the immunogen with regard to positions where direct contact is made with antibody side chains (22). The chemical structure of the linkage between the label and steroid also affects the process of the antibody recognition (2). This has also been observed in our studies. The antibody had a higher affinity to the thioether derivative than to the oxime derivatives when coupled to position C-4 of estradiol (Figure 2). This can be explained by hydrogen bond between oxime-nitrogen

330 Bioconjugate Chem., Vol. 10, No. 3, 1999

Meltola et al.

Figure 2. Scatchard analysis of equilibrium binding data for an anti-estradiol antibody. The plot includes only one oximederivative by the way of an example. Each point is the mean of two determinations. Table 2. IC50 Values, Signal Levels, and Affinities of the Tested Estradiol Tracers tracer

IC50a (nM)

corr Bob (cps)

bgc (cps)

Ka (M-1)

E2-6-CMO-Eu E2-4-CPO-Eu E2-4-AHO-Eu E2-4-CET-Eu E2-4-ABO-Eu

8.1 4.0 4.9 6.1 7.8

37 068 36 182 43 288 35 773 34 366

477 5148 12 222 729 10 978

1.3 × 1010 1.7 × 108 1.9 × 108 5.5 × 108 1.6 × 108

a

IC50 values are determined as estradiol concentration for 50% displacement of maximum binding in the absence of unlabeled estradiol. b Signal in the absence of unlabeled estradiol. c Instrumental background and nonspecific binding.

and 3-hydroxyl group, which obviously has a negative effect on antibody recognition. In summary, all the tested europium-labeled estradiol4-derivatives were bound by the antibody (SSI 57-2), whereas tracers linked via position 2 were not recognized by this antibody. This was anticipated because in the previous studies (9), the same antibody had only slight cross-reactivity with estradiol-2-derivatives. Figure 3 shows estradiol standard curves with three different tracers. The curves show a trend in sensitivity depending on the position of the linkage. Tracers conjugated via C-4 give more sensitive standard curves than tracers conjugated via C-6. This can also be seen in Table 2, where IC50 values are lower when using tracers conjugated via C-4. The reason for higher background signal with oxime derivatives is unspecific binding, resulting partly from the higher tracer concentration needed because of the lower affinity of these derivatives. Our results demonstrate that the new estradiol derivatives synthesized here, especially the thioether derivative, are useful tracers in the immunochemical characterization of the anti-estradiol Mab SSI 57-2. Because this antibody is a representative of a large group of monoclonal antibodies raised with estradiol-6-CMO conjugates, we foresee that our tracers will be potentially suitable for the other Mabs, too. Furthermore, as these new conjugates do not exhibit considerable bridge effect (unlike the 6-CMO derivatives), they could be used as panning antigens in the search of improved anti-E2-6CMO antibody variants from large mutant libraries using

Figure 3. Estradiol displacement curves with different europium-labeled estradiol derivatives in the time-resolved fluoroimmunoassay. B is binding at the estradiol concentration indicated, Bo is maximum binding without unlabeled estradiol.

the phage display technology (23). When used as a panning antigen, these estradiol derivatives can be conjugated with biotin instead of a europium chelate. We are currently using such biotinylated estradiol-4-thioether derivative for the panning of a mutant phage library derived from Mab SSI 57-2 (to be published elsewhere). ACKNOWLEDGMENT

We are grateful to Dr. V.-M. Mukkala for his valuable advise in the field of synthetic chemistry. This work was supported by Academy of Finland and Technology Development Centre Finland. LITERATURE CITED (1) Pratt, J. J. (1978) Steroid immunoassay in clinical chemistry. Clin. Chem. 24, 1869-1890. (2) Tiefenauer, L. X., and Andres, R. Y. (1990) Biotinyl-estradiol derivatives in enzyme immunoassays: Structural requirements for optimal antibody binding. J. Steroid Biochem. 35, 633-639. (3) Mikola, H., and Miettinen, P. (1991) Preparation of europium labeled derivatives of cortisol for time-resolved fluoroimmunoassays. Steroids 56, 17-21. (4) Mikola H., Sundell, A.-C., and Ha¨nninen, E. (1993) Labeling of estradiol and testosterone alkyloxime derivatives with a europium chelate for time-resolved fluoroimmunoassays. Steroids 58, 330-334. (5) Soini, E., and Lo¨vgren, T. (1987) Time-resolved fluorescence of lanthanide probes and applications in biotechnology. CRC Crit. Rev. Anal. Chem. 18, 105-154. (6) Lo¨vgren, T., Hemmila¨, I., Pettersson, K., and Halonen, P. (1985) Time-resolved fluorometry in immunoassays. In Alternative immunoassays (W. P. Collins, Ed.) Wiley, Chichester, NY. (7) Lo¨vgren, T. (1987) Time-resolved fluoroimmunoassays of steroid hormones. J. Steroid Biochem. 27, 47-51. (8) Tiefenauer, L. X., Bodmer, D. M., Frei, W., and Andres, R. Y. (1989) Prevention of bridge binding in immunoassays: a general estradiol tracer structure. J. Steroid Biochem. 32, 251-257. (9) Lamminma¨ki, U., Villoutreix, B. O., Jauria, P., Vihinen, M., Nilsson, L., Teleman, O., Saviranta, P., and Lo¨vgren, T. (1997) Structural analysis of an anti-estradiol antibody. Mol. Immunol. 34, 1215-1226.

Eu-Labeled Estradiol Derivatives for Immunoassays (10) Allen, R. M., and Redshaw, M. R. (1978) The use of homologous and heterologous 125I-radioligands in the radioimmunoassay of progesterone. Steroids 32, 467-486. (11) Mukkala, V.-M., Mikola, H., and Hemmila¨, I. (1989) The synthesis and use of activated N-benzyl derivatives of diethylenetriaminetetraacetic acids: Alternative reagents for labeling of antibodies with metal ions. Anal. Biochem. 176, 319-325. (12) Mikola, H., and Ha¨nninen, E. (1992) Introduction of aliphatic amino and hydroxy groups to ketosteroids using O-substituted hydroxylamines. Bioconjugate Chem. 3, 182186. (13) Mikola, H. (Unpublished results). (14) Scatchard, G. (1949) The attractions of proteins for small molecules and ions. Annals. N. Y. Acad. Sci. 51, 660-672. (15) Meltola, N., and Mikola, H. (unpublished results). (16) Lovely, C. J., and Brueggemeier, R. W. (1994) Synthesis of 2-substituted hydroxyalkyl and aminoalkyl estradiols. Tetrahedron Lett. 35, 8735-8738. (17) Pert, D. J., and Ridley, D. D. (1989) Formylation of oestrogens. Aust. J. Chem. 42, 405-419. (18) Xie, R., Chen, Q., Xie, J., and Zhao, H. (1990) A new efficient synthetic method for 2- and 4-hydroxy-17R-ethynylestradiol. Steroids 55, 488-490.

Bioconjugate Chem., Vol. 10, No. 3, 1999 331 (19) Mihailovic, M. Lj., Forsek, J., and Lorenc, Lj. (1977) A novel procedure for the aromatization of ring A in 19-nortestosterone. Tetrahedron 33, 235-237. (20) Le Quesne, P. W., Abdel-Baky, S., Durga, A. V., and Purdy, R. H. (1986) Active nonaromatic intermediates in conversion of the steroidal estrogens into catechol estrogens. Biochemistry 25, 2065-2072. (21) Ghaffari, M. A., and Abul-Hajj, Y. J. (1990) Reaction of thiol nucleophiles with 1,2-epoxy-and 4,5-epoxy-estrene-3-one-17βols. J. Steroid Biochem. Mol. Biol. 37, 237-244. (22) Gani, M., Coley, J., Piron, J., Humphrey, A. S., Arevalo, J., Wilson, I. A., and Taussig, M. J. (1994) Monoclonal antibodies against progesterone: effect of steroid-carrier coupling position on antibody specificity. J. Steroid Biochem. Mol. Biol. 48, 277-282. (23) Hoogenboom, H. R., Griffiths, A. D., Johnson, K. S., Chiswell, D. J., Hudson, P., and Winter, G. (1991) Multisubunit proteins on the surface of filamentous phage: Methodologies for displaying antibody (Fab) heavy and light chains. Nucleic Acids Res. 19, 4133-4137.

BC980013Q