Determination of anti-acetylcholine receptor antibodies in myasthenic patients by use of time-resolved fluorescence.
High-sensitivity detection of lanthanides (comparable to the sensitivity for detection of [sup.125]I) is possible by the use of time-resolved fluorescence methods, and those labels are used for measurement of numerous compounds (36). However, lanthanides cannot be used for direct labeling, and the compound to be detected (e.g., a protein) must first be derivatized with a suitable chelate, which is then compexed with the lanthanide.
We describe the preparation of [Eu.sup.3+]-labeled a-cobratoxin ([alpha]-CTx; long neurotoxin from Naja naja kaouthia, which is highly similar to [alpha]-BuTx), which can be measured similar to the so-called dissociation-enhanced lanthanide fluorescence immunoassay (DELFIA), and demonstrate its use in an immunoassay for antibodies to nAChR. The assay procedure is analogous to that in routine use (2) and shows identical clinical specificity and an only slightly higher limit of detection (LOD).
Materials and Methods
MATERIAL AND REAGENTS
Venoms of Bungarus multicinctus and N. naja kaouthia were obtained from Latoxan; [sup.125]I-labeled [alpha]-BuTx (specific activity, 8.2 TBq/mmol or 74 TBq/mmol) was from Amersham Pharmacia Biotech; diethylenetriaminepentaacetic acid (DTPA) anhydride and Eu[(N[O.sub.3]).sub.3] were from Sigma-Aldrich; Sephadex G-25, CM-Sephadex C-50, and mono-Q 5 x 5 columns were from Pharmacia Biotech; octylSi300 (5 [micro]m) HPLC support was from Serva; antihuman IgG (QSwAHuIgG) was from Sevapharma; DELFIA enhancement solution was from Wallac Oy; TE671 cell culture was from Deutsche Sammlung von Mikroorganismen and Zellkulturen GmbH; and Torpedo marmorata electric organs were from the Station Biologique d'Arcachon. All other chemicals were of analytical grade and were purchased mostly from Sigma-Aldrich.
PURIFICATION OF TOXINS
[alpha]-CTx was purified from the venom of N. naja kaouthia by the procedure of Karlsson et al. (7). [alpha]-BuTx was purified from the venom of B. multicinctus by a procedure similar to that described by Lee et al. (8). Both toxins showed a single peak when tested by various HPLC analyses and were estimated to be at least 96% pure.
CTX LABELING AND PURIFICATION OF DERIVATIVES
To 2 mg (0.25 [micro]mol) of [alpha]-CTx in 0.6 mL of 150 mmol/L sodium carbonate (pH 9.3), we added 16 mg of DTPA anhydride dissolved in 0.3 mL of dimethyl sulfoxide (in 50-[micro]L portions, with stirring) and left the mixture to react overnight at room temperature. The reaction mixture was then passed through a Sephadex G-25 column (1 x 15 cm) preequilibrated with 50 mmol/L ammonium acetate. The toxin fraction (5 mL) was applied to a CM-Sephadex C-50 column (bed volume, 3 mL; preequilibrated with 50 mmol/L ammonium acetate). The wash-through fraction (12 mL) contained the derivatized toxin. The underivatized (native) toxin was retained on the resin and was washed out by 6 mL of 50 mmol/L ammonium acetate--0.4 mol/L NaCl. The derivatized fraction was concentrated by lyophilization, treated with 1 mmol/L Eu[(N[O.sub.3]).sub.3], and further purified by reversed-phase and ion-exchange chromatography.
The [Eu.sup.3+]-treated fraction was loaded onto a 8 x 80 mm column packed with octylSi300 (Serva) and eluted at 1 mL/min with the following gradient: 10% B for 5 min; 10-70% B over 120 min; 70% B for 15 min; 70-10% B over 10 min. The gradient system was as follows: solvent A, 1 g/L trifluoroacetic acid in water; solvent B, 1 g/L trifluoroacetic acid in 700 mL/L acetonitrile. The absorbance at 280 nm was monitored, and 4-mL fractions were collected.
Fractions from reversed-phase chromatography were further purified on a mono-Q column (5 x 50 mm; Pharmacia) eluted at 0.5 mL/min with the following gradient profile: 0% B for 10 min; 0-40% B over 90 min; 40-100% B over 20 min. The gradient system was as follows: solvent A, 20 mmol/L ammonium acetate in 100 mL/L methanol; solvent B, 500 mmol/L ammonium acetate in 100 mL/L methanol. The absorbance at 280 nm was monitored, and 2-mL fractions were collected and concentrated by lyophilization.
DETERMINATION OF THE AFFINITY OF DERIVATIVED [alpha]-CTx FOR NICOTINIC RECEPTORS
The relative affinity was determined by competition experiments in which different amounts of the derivatized [alpha]-CTx inhibited the binding of [sup.125]I-[alpha]-BuTx to AChR prepared from the electric organ of T. marmorata, according to the method of Fels et al. (9).
CULTURE OF TE671 CELLS
The rhabdomyosarcoma cell line of human origin, TE671, was the source of the nAChR. The use of this cell line was introduced for measurements of antibodies in myasthenia (10-12). Cells were grown, essentially as described by Somnier (13), in DMEM containing 2 mmol/L glutamine and 100 mL/L fetal bovine serum in glass bottles (~180 [cm.sup.2]) in a 5% C[O.sub.2] atmosphere at 37[degrees]C, with seeding of ~3 x 106 cells/bottle. On the 4th day, when cells approached confluence, the fetal bovine serum concentration was reduced to 10 mL/L, and the medium was supplemented with 2.5 [micro]mol/L dexamethasone and 100 [micro]mol/L nicotine. Cells were harvested on the 7th day of culture, washed, and centrifuged, and the pellet was kept at -65[degrees]C until further processing.
PREPARATION AND LABELING OF HUMAN NICOTINIC RECEPTORS
Triton X-100 extracts containing the receptors were prepared from human muscles (mostly from amputated legs from patients with complicated diabetes) by the procedure described by Lindstrom et al. (14) or from TE671 cells as described by Voltz et al. (15). nAChR complexed with [Eu.sup.3+]-labeled [alpha]-CTx was prepared so that ~0.5 nmol/L nAChR was saturated to 60-75% with the labeled toxin. The concentration required was first established by binding studies using different concentrations of the [Eu.sup.3+]-labeled [alpha]-CTx.
ISOTOPIC ASSAY OF MYASTHENIC ANTIBODIES
The procedure described by Lindstrom et al. (14) was used, with volumes scaled down fivefold. The detection limit of the assay was ~0.1 nmol/L. The final reaction volume of 0.2 mL contained 0.2 nmol/L nAChR, 0.3 nmol/L [sup.125]I-[alpha]-BuTx, and 2-5 [micro]L of the investigated human serum. If required for quantitative determination of antibody, the sera were first diluted with a negative control serum.
We incubated 0.2 mL of nAChR-[Eu.sup.3+]-[alpha]-CTx complex (nAChR concentration ~0.5 nmol/L) with 10 [micro]L of human serum for 2 h at room temperature and then overnight at 4[degrees]C. Human IgG was then precipitated with anti-human IgG (100 [micro]L of QSwAHuIgG) during a 5-h incubation at room temperature. The resulting precipitate was centrifuged (10 min at 10 000g and 4[degrees]C), washed four times with 1 mL of phosphate-buffered saline, centrifuged (10 min at 10 000g and 4[degrees]C), and dissolved in 0.1 mL of "enhancement solution' (100 mmol/L acetate-phthalate, pH 3.2, 15 [micro]mol/L 2-naphthoyltrifluoroacetone, 50 [micro]mol/L tri-n-octylphosphine oxide, 1 mL/L Triton X-100). After an overnight incubation at room temperature, the contents were transferred from the test tubes to a microtitration plate, and time-delayed fluorescence was measured with a Victor 1420 Multilabel Counter (Wallac Oy), using the factory-set mode for [Eu.sup.3+].
Sera from 43 MG patients (14 males, 29 females; age range, 22-85 years) with titers covering a wide range were used in this study; 8 of those patients were seronegative by the isotopic assay. The severity of their disease ranged from group I to group III as graded according to the scale proposed by Osserman (16). Diagnosis of MG was based on typical clinical, pharmacologic (17), electromyographic (18), immunologic, and neurophysiologic features, and the radiologic status of the thymus was checked by computerized tomography of the mediastinum. The Neurologic Clinic registered 1185 MG patients from the Czech Republic (457 males and 728 females). Sera from healthy blood donors (n = 16) and patients with nonmyasthenic diseases (5 with multiple sclerosis, 5 with lupus erythematosus) served as negative controls. Laboratory workers worked with numbered samples without knowledge of the diagnosis.
PREPARATION OF THE [EU.sup.3+]-LABELED [alpha]-CTx DERIVATIVE
Approximately 60% of [alpha]-CTx was derivatized by DTPA under the described conditions. The [Eu.sup.3+]-labeled toxin was analyzed by reversed-phase HPLC, and several positional isomers were detected (Fig. 1). Fractions isolated by HPLC were assayed for their relative affinity for nAChR by measuring their ability to inhibit the binding of [sup.125]I-[alpha]-BuTx to nAChR from electric organ membrane fragments. Fig. 2 shows the inhibition of binding of [sup.125]I-[alpha]-BuTx to AChRs by native and [Eu.sup.3+]-labeled [alpha]-CTx. The concentration of [Eu.sup.3+]-labeled [alpha]-CTx required for half-maximal saturation of nAChR was ~50 times higher than the respective concentration of the native [alpha]-CTx.
[FIGURE 1 OMITTED]
Derivatives eluted at 0-60 mL showed no measurable affinity for nAChR and contained ~30% of the applied toxin. Fractions eluting at ~110, 150, and 190 mL showed binding to nAChR, but their contribution was minor both in protein content and total binding capacity. The peak eluting between 60 and 90 mL contained ~60% of the derivatized [alpha]-CTx and the bulk of the total binding capacity for nAChR. Analysis by ion-exchange HPLC on a mono-Q column showed that this fraction contained three positional isomers, all of which were actively bound to nAChR. Therefore, a pool of the fractions eluting between 60 and 90 mL was used in the present work.
[FIGURE 2 OMITTED]
BINDING OF [EU.sup.3+]-LABELED [alpha]-CTx TO RECEPTORS
To measure the binding of [Eu.sup.3+]-labeled toxin to solubilized nAChRs from human muscle, we used immunoprecipitation with a highly positive MG serum. Binding of our [Eu.sup.3+]-labeled toxin preparation was specific and saturable (Fig. 3A), with two different apparent affinities: 3.7 x [10.sup.-9] mol/L (~40%)and 145 x 10-9 mol/L (~60%). For comparison, similar results with [sup.125]-[alpha]-BuTx are shown in Fig. 3B.
TITRATION OF MYASTHENIC SERA IN [EU.sup.3+] ASSAY
To detect antibodies in myasthenic sera, we used 100 fmol of nAChR saturated to 60-75% with [Eu.sup.3+]-[alpha]-CTx. The [Eu.sup.3+]-[alpha]-CTx-nAChR complexes were prepared "in bulk" and were stable for at least 1 year if stored in appropriate aliquots (to avoid repeated thawing and freezing) at -80[degrees]C (data not shown). Receptors extracted from both sources (TE 671 cells or human muscles) showed similar suitability for the assay.
[FIGURE 3 OMITTED]
Examples of the titration of individual myasthenic sera are shown in Fig. 4. Taking into account the differences in specific activity of the two toxins, the shapes of individual titration curves obtained were very similar. The specific activity of the [Eu.sup.3+]-labeled [alpha]-CTx preparation used in this study was 1500 fluorescence counts/fmol. Negative sera typically gave 11300 [+ or -] 1050 (mean [+ or -] SD) fluorescence counts, and the same background values were obtained when specific binding of the label was blocked by an excess of cold [alpha]-BuTx ([10.sup.-7] mol/L).
We used three 10-[micro]L serum aliquots for the initial estimation of the titer. If >20% of the available [Eu.sup.3+]-[alpha]-CTx-nAChR complex precipitated, the serum was diluted with negative control serum (to maintain the immunoglobulin concentration) so that at least two dilutions were in the range of 30% of the precipitated complex. The titration data were plotted and fitted with a hyperbolic curve, and the resulting titer was calculated from the initial asymptote.
LOD, PRECISION, AND DIAGNOSTIC ACCURACY OF [EU.sup.3+] ASSAY
We made an effort to adhere to the published guidelines for diagnostic accuracy for our test (19). The calculated LOD, determined as 3 SD of the blank with 10 [micro]L of serum and [Eu.sup.3+]-[alpha]-CTx-nAChR complex at 70% saturation, was 0.3 nmol/L. None of the negative sera tested gave a value exceeding that limit. The typical "within-run" CV for triplicates in the isotopic assay was ~5%. The imprecision of the fluorescence method was greater, with a within-run CV, for triplicates, of ~10% and a between-run CV ~16%. The usefulness of the fluorescence method, however, was demonstrated by comparison of titers obtained by the two methods (Fig. 5). We compared 43 myasthenic sera (with titer values between 0.3 and 1290 nmol/L as determined by the isotopic assay); 8 of those sera (not included in Fig. 5) were seronegative by the isotopic assay, with titers indistinguishable from those of healthy controls (titer [less than or equal to] 0.3 nmol/L), and were negative by the [Eu.sup.3+] assay as well. We included eight sera with titers of 0.4-1 nmol/L; three of those gave values [less than or equal to] 0.3 nmol/L in the [Eu.sup.3+] assay. Two of these three were also near the LOD for the isotopic assay: 0.4 nmol/L. The third serum that was negative in the [Eu.sup.3+] assay gave a titer of 0.7 nmol/L in the isotopic assay. Overall, the linear regression of data yielded the following equation: y = 1.035x - 0.013 nmol/L ([S.sub.y|X] = 0.172 nmol/L; [r.sup.2] = 0.977).
We established a nonradioactive immunoassay for antibodies to AChRs that used [Eu.sup.3+]-labeled [alpha]-CTx rather than [sup.125]I-[alpha]-BuTx. The reagent used to label the antigen is more stable than [sup.125]I-[alpha]-BuTx, the concentration of antibodies can be expressed in molar units as in conventional RIAs, and the results show similar detection limits and identical diagnostic specificity.
[FIGURE 4 OMITTED]
Several nonradioactive methods for the detection of anti-nAChR antibodies have been described previously. One approach is based on solid-phase attachment of nAChR and detection of bound anti-nAChR immunoglobulin by labeled anti-immunoglobulin antibody (20-23); another is on the substitution of horseradish peroxidase-labeled [sup.125]I-[alpha]-BuTx for [sup.125]I-[alpha]-BuTx in the immunoassay (24). None of these methods has been adopted for routine use, probably because they are less sensitive than the [sup.125]I-[alpha]-BuTx radioimmunoprecipitation assay.
We attempted to use [Eu.sup.3=]-labeled [alpha]-BuTx, but the derivatives of [alpha]-BuTx that we obtained lost much of their affinity for nAChRs. Labeling of [alpha]-CTx produced a useful derivative. Our preparation probably represents a mono-substituted toxin with DTPA conjugated with either lysine or the N-terminal amino acid (proline). It would be possible to isolate individual positional isomers, but our procedure has been worked out for a mixture of several isomers that are isolated together. Binding analysis of the HPLC-purified fraction used for the experiments suggested the presence of two main components with calculated binding affinities of 3.7 x [10.sup.-9] mol/L and 145 x [10.sup.-9] mol/L.
As was already noted, the results obtained with the [Eu.sup.3+] assay correlate closely to the results obtained with the isotopic assay. This was to be expected because the design of both methods is almost identical and the only important difference is in the label used for the [alpha]-toxin. The described fluorescence assay can measure concentrations of anti-nAChR antibodies >0.3 nmol/L, but reliable estimates can be expected for titers >0.5 nmol/L. This LOD is slightly higher than that of the isotopic assay, which has a LOD in the range of 0.1-0.2 nmol/L. In MG, titers >0.5 nmol/L are considered positive, whereas titers of 0.2-0.5 nmol/L are evaluated as equivocal [e.g., see Vincent and Newsom-Davis (25)]. Although the isotopic assay might be more advantageous for such "equivocal" cases, the advantages of the nonradioactive fluorescence assay that we describe may make it a useful tool in the diagnosis and follow-up of myasthenic patients.
[FIGURE 5 OMITTED]
This work was supported by Grant 4646-3 of the Grant Agency of the Ministry of Health of the Czech Republic. We thank Dr. Stanislav Tucek for help with preparing the manuscript, Vera Lisa for work with tissue cultures, and Dana Novotna for skillful technical assistance (all from the Institute of Physiology, Academy of Sciences of the Czech Republic, Prague, Czech Republic); Dr. Yuri N. Utkin (Shemyakin-Ovchinnikov Institute of Bioorganic Chemistry, Russian Academy of Sciences, Moscow, Russia) for discussions concerning the modification of [alpha]-CTx; Dr. Helena Tlaskalova (Institute of Microbilogy, Academy of Sciences of Czech Republic) for providing sera from healthy controls; and Dr. Katerina Nova (Institute of Nuclear Medicine, 1st Medical Faculty of Charles University, Prague, Czech Republic) for some of the myasthenic sera.
(1.) Kurtzke J. Epidemiology of myasthenia gravis. In: Schoenberg BS, ed. Advances in neurology. New York: Raven Press, 1978:54564.
(2.) Lindstrom JM, Seybold ME, Lennon VA, Shithingham S, Duane D. Antibody to acetylcholine receptor in myasthenia gravis, prevalence, clinical correlates and diagnostic value. Neurology 1976; 26:1054-9.
(3.) Hemmila I. Fluoroimmunoassays and immunofluorometric assays. Clin Chem 1985;31:359-70.
(4.) Diamandis EP. Immunoassays with time-resolved fluorescence spectroscopy, principles and applications. Clin Biochem 1988;21: 139-50.
(5.) Lovgren T, Pettersson K. Time-resolved fluoroimmunoassay: advantages and limitations. In: van Dyke K, van Dyke R, eds. Luminescence immunoassay and molecular applications. Boca Baton, FL: CRC Press, 1990:233-53.
(6.) Mitrunen K, Makinen M-L, Pettersson K, Lovgren T. The use of fluorescent lanthanide chelates as labels in time-resolved fluoroimmunoassay. Clin Chem 1995;41:1115-20.
(7.) Karlsson E, Arnberg H, Eaker D. Isolation of the principal neurotoxins of two Naja naja subspecies. Eur J Biochem 1971;21:1-16.
(8.) Lee CY, Chang SL, Kau ST, Luh S-H. Chromatographic separation of the venom of Bungarus multicinctus and characterization of its components. J Chromatogr 1972;72:71-82.
(9.) Fels G, Wolff EK, Maelicke A. Equilibrium binding of acetylcholine to membrane-bound acetylcholine receptor. Eur J Biochem 1982; 127:31-8.
(10.) Schoepfer R, Luther M, Lindstrom J. The human medulloblastoma cell line TE671 expresses a muscle-like acetylcholine receptor. Cloning of the [alpha]-subunit cDNA. FEBS Lett 1988;226:235-40.
(11.) Sine SM. Functional properties of human skeletal muscle acetylcholine receptors expressed by the TE671 cell line. J Biol Chem 1988;263:18052-62.
(12.) Stratton MR, Darling J, Pilkington GJ, Lantos PL, Reeves BR, Cooper CS. Characterization of the human cell line TE671. Carcinogenesis 1989;10:899-905.
(13.) Somnier FE. Anti-acetylcholine receptor (AChR) antibodies measurement in myasthenia gravis: the use of cell line TE671 as a source of AChR antigen. J Neuroimmunol 1994;51:63-8.
(14.) Lindstrom J, Einarson B, Tzartos S. Production and assay of antibodies to acetylcholine receptor. Methods Enzymol 1981;74: 432-60.
(15.) Voltz R, Hohlfeld R. Fateh Moghadam A, Witt TN, Wick M, Reimers C, et al. Myasthenia gravis: measurement of anti-AChR autoantibodies using cell line TE671. Neurology 1991;41:1836-8.
(16.) Osserman KE, Genkins G. Studies in myasthenia gravis: review of twenty-year experience in over 1200 patients. Mt Sinai J Med 1971;38:497-537.
(17.) Daroff RB, Seybold ME. The office Tensilon test for ocular myasthenia gravis. Arch Neurol 1986;43:842-4.
(18.) Schumm F, Stohr M. Accessory nerve stimulation in the assessment of myasthenia gravis. Muscle Nerve 1984;7:147-51.
(19.) Bruns DE, Huth EJ, Magid E, Young DS. Toward a checklist for reporting of studies of diagnostic accuracy of medical tests. Clin Chem 2000;46:893-5.
(20.) Norcross NL, Griffith IJ, Lettieri JA. Measurement of acetylcholine receptor and anti-receptor antibodies by ELISA. Muscle Nerve 1980;3:345-9.
(21.) Hinman CL, Hudson RA, Burek CL, Doodlow G, Rauch HC. An enzyme-linked immunosorbent assay for antibody against acetylcholine receptor. J Neurosci Methods 1983;9:141-55.
(22.) Jailkhani BL, Asthana D, Jaffery NF, Khajuria A, Subbalaxmi B, Ahuja GK. ELISA for detection of IgG and IgM antibodies to human nicotinic acetylcholine receptor in myasthenia gravis. Indian J Med Res 1986;83:187-95.
(23.) Franciotta D, Martino G, Brambilla E, Zardini E, Locatelli V, Bergami A, et al. TE671 cell-based ELISA for anti-acetylcholine receptor antibody determination in myasthenia gravis. Clin Chem 1999;45:400-5.
(24.) Furukawa S, Akazawa S, Furukawa Y, Kamo I, Satoyoshi E, Hayashi K. A practical enzyme immunoassay for anti-acetylcholine receptor antibodies in myasthenia gravis. J Neuroimmunol 1984; 6:397-409.
(25.) Vincent A, Newsom-Davis J. Acetylcholine receptor antibody as a diagnostic test for myasthenia gravis: results in 153 validated cases and 2967 diagnostic assays. J Neurol Neurosurg Psychiatry 1985;48:1246-52.
Jan Ricny,  * Libuse Simkova,  and Angela Vincent 
 Institute of Physiology, Academy of Sciences of Czech Republic, Videoska 1083, 142 20 Prague, Czech Republic.
 Neurologic Clinic, 1st Medical Faculty of Charles University, Kateoinska 30, 120 00 Prague, Czech Republic.
 Neurosciences Group, Department of Clinical Neurology, Institute of Molecular Medicine, John Radcliffe Hospital, Oxford OX3 9DU, United Kingdom.
 Nonstandard abbreviations: MG, myasthenia gravis; nAChR, nicotinic acetylcholine receptor; [alpha]-BuTx, [alpha]-bungarotoxin; [alpha]-CTx, [alpha]-cobratoxin; LOD, limit of detection; and DTPA, diethylenetriaminepentaacetic acid.
* Author for correspondence. Fax 420-2-4752488; e-mail ricny@biomed. cas.cz.
Received September 4, 2001; accepted December 17, 2001.
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||Clinical Immunology|
|Author:||Ricny, Jan; Simkova, Libuse; Vincent, Angela|
|Date:||Mar 1, 2002|
|Previous Article:||Evaluation of an automated preanalytical robotic workstation at two academic health centers.|
|Next Article:||Age-related reference values for serum selenium concentrations in infants and children.|