New Variants of Enzyme Immunoassay of Antibodies to DNA

Anal. Chem. 1996, 68, 3827-3831

New Variants of Enzyme Immunoassay of Antibodies to DNA S. S. Babkina,* E. P. Medyantseva, H. C. Budnikov, and M. P. Tyshlek

Chemical Faculty, Kazan State University, Kazan 420008, Russia

A method of DNA immobilization on cellulose nitrate films has been developed. Modified films of uniform and stable surface have been used to devise two variants of solidphase enzyme immunoassays of antibodies. The coimmobilization of enzyme label (cholinesterase) and the DNA molecules makes it possible to carry out the procedure of solid-phase enzyme immunoassay without any separation of components. Thus, it takes only 15 min to diagnose an autoimmune disease (Aleutian disease of minks) with the immunoenzyme amperometric sensor, with a lower detection limit for antibodies of 0.5 × 10-10 M. For scaled diagnosing, solid-phase enzyme immunoassay on DNA-modified films with prior separation of components and spectrophotometric registration of peroxidase activity has been developed. The time for determination was 30 min, with a lower detection limit of 7.4 × 10-12 M. At present, the determination of biological macromolecules such as nucleic acids and proteins is a rapidly developing area of analytical chemistry. The continuous progress in studying their structure and their physical and chemical features contributes to our understanding of the principal vital processes and the effect of immunological reactions on the humans and animals.1 The so-called autoimmune diseases (including a number of severe animal and human diseases2), characterized by the response of an organism to its own components, e.g., DNA as an antigen, can be diagnosed with the study of immunological reactions involving DNA as a background.3 Aleutian mink disease (AMD) is characterized by a high level of autoantibodies to DNA produced by the destruction of lysosomes.4 A number of tests have been devised for lifetime diagnosis of AMD, i.e., iodine agglutination,5 glutaraldehyde-based test,6 and an immunoelectroosmophoresis reaction.7 All of them are insufficient in terms of specificity and do not involve the autoimmune nature of a disease. Recently, it has been more common to use enzyme immunoassay (EIA) for sensitive and specific diagnosing.8,9 The procedure for AMD, involving spectrophotometric registration of the analytical signal on polystyrene microplates, was described (1) Palecek, E. Bioelectrochem. Bioenerg. 1986, 15, 275-95. (2) Tan, E. M. Adv. Immunol. 1982, 33, 167. (3) Zuyev, V. A. Slow virus infections of humans and animals; Medicina: Moscow, 1988 (in Russian). (4) Porter, D.; Porter, H.; Suffin, S. Infect. Immun. 1984, 43, 463-6. (5) Henson, J. B. (Denver, CO) Natl. Fur News 1963, 34, 23. (6) Sandholm, M.; Kangas, J. Zbl. Vet. Med. 1973, 20, 206. (7) Slugin, V. S. Veterinariya 1981, 5, 29 (in Russian). (8) Blake, C.; Gould, B. J. Analyst 1984, 109, 533. (9) Ngo, T. T., Lenhoff, H. M., Eds. Enzyme-Mediated Immunoassay; Plenum: New York, 1985. S0003-2700(95)00645-7 CCC: $12.00

© 1996 American Chemical Society

elsewhere.10 At present, electrochemical methods have been involved in immunoassay in order to enhance its rate and sensitivity.11 Moreover, the use of electrochemistry makes it possible for us to draw some conclusions regarding the mechanism of macromolecule-involving processes, particularly regarding DNA behavior both in solution and at the surface of the electrode. The latter simulates the properties of DNA in a cell membrane.1,12 In ref 13, it was suggested to employ enzyme modulator-mediated immunoassay involving a cholinesterase (ChE)-based amperometric electrode for the determination of antibodies to DNA and AMD diagnosis, which takes the autoimmune origin of the disease into consideration, with organophosphorus inhibitors covalently bound to DNA used as modulators. Our consideration of literary data shows that assay of antibodies to DNA is a problem that requires new, more sensitive, easyto-use, and fast methods of analysis. This paper is devoted to the analytical possibilities for various types of EIA developed by us using DNA-modified cellulose nitrate (CN) membrane. The first one involves the new immunoenzyme amperometric sensor (IEAS) and the other one the spectrophotometric registration of enzymatic activity. EXPERIMENTAL SECTION Apparatus and Reagents. Voltammetric measurements were performed with a PO-5122/03 oscillopolarograph (Rostov Experimental Works, Rostov City, USSR). The IEAS constructed by us served as a working electrode. The reference electrode was a saturated calomel electrode (SCE). Optical densities of the solutions were measured with a Multiscan optical densitometer (by Flow), and the optical densities of colored spots on the cellulose nitrate film were measured with a two-beam chromatoscanner (CS-930, Shimadzu Corp., Kyoto, Japan). Medium-nitrogen cellulose nitrate films were employed. Electrochemical measurements were performed at 25 ( 0.2 °C in borate buffer solutions (pH 9.05) prepared from chemically pure substances. The solutions were deaerated with argon. Spectrophotometric measurements were performed in phosphate buffer saline (PBS, pH 7.3, chemically pure components). Tween20 solution (0.05%) in PBS (PBS+T) was used as a detergent. Horse serum butyryl cholinesterase (EC 3.1.1.8) with an initial activity of 110 units mg-1 was employed. Recrystallized butyryl thiocholine iodide (BuSChI) was dissolved in borate buffer (pH 8.9-9.1) and served as a ChE substrate. (10) Wright, P. Second International Congress on Fur Animals Production, Denmark, 1980. (11) Ngo, T. T., Ed. Electrochemical Sensors in Immunological Analysis; Plenum Press: New York, 1987; p 343. (12) Kim, S. D.; Vrana, O.; Klunwachter, V.; Niki, K.; Brabee, V. Anal. Lett. 1990, 23, 1505. (13) Babkina, S. S.; Medyantseva, E. P.; Budnikov, H. C.; Vinter, V. G. Anal. Chim. Acta 1993, 273, 419.

Analytical Chemistry, Vol. 68, No. 21, November 1, 1996 3827

Horseradish peroxidase (HRP) from Sigma, with an activity of 330 units mg-1 was used. A mixture of 3,3′-diaminobenzidine (Sigma) and 30% H2O2 was taken as a substrate of HRP. Highly purified organic solvents (toluene, butyl acetate, hexane) and 25% glutaraldehyde solution (Reanal) were used. The following biological preparations were also used: (1) bovine spleen DNA at a concentration of 0.01 mg mL-1 in SSC buffer (0.15 M NaCl + 0.015 M sodium citrate, pH 7.0); (2) immunoglobulin G (IgG) at a concentration of 22.2 mg mL-1 from the serum of diseased minks, which was salted-out twice with ammonium sulfate and isolated by ion-exchange chromatography in a DEAE cellulose packed column; (3) bovine serum albumin (BSA) from the Byelorussian Institute for Epidemiology and Microbiology (1% solution in 0.01 M PBS, pH 7.3); and (4) antimink HRP-labeled conjugate of IgG, obtained by the periodate method.14 METHODS Preparation of the Sensor Part of the IEAS. Cellulose nitrate was dissolved in toluene-butyl acetate, and 0.2 mL of a solution containing 0.018 g of ChE and denatured DNA, obtained by heating on a water bath for 20 min followed by fast cooling, was added. The solution was stirred, and 0.06 mL of 25% glutaraldehyde solution and hexane as a coagulator were added. The resulting mixture was cast on a glass plate and dried, and a membrane was obtained. The film was cut into small pieces (2.5 × 6.5 cm2) and immersed in 1% BSA in 0.01 M PBS (pH 7.3) to block the glutaraldehyde active sites and avoid nonspecific binding.15,16 After washing with borate buffer (pH 9.5), a crimped film was fixed with clamping rings on the surface of the body of a mercury film-covered stationary silver electrode.17 Determination of Antibodies to DNA by IEAS. To construct a calibration graph, 0.5 mL of 1.9 × 10-2 M BuSChI was placed in a 5 mL volumetric flask, and 1-100 µL of a 0.8 mg mL-1 γ-globulin fraction obtained from diseased mink blood serum was added. The solution was diluted to 5 mL with a borate buffer, stirred, and then injected into the cell equipped with SCE and IEAS, the preparation of which has been described in the previous paragraph. Oxygen was removed during 15 min by a stream of argon, and the voltammograms were recorded in the range of potentials from -0.1 to -0.8 V (scan rate, 1 V s-1; Estart ) -0.1 V; continuous polarization regime, triangular sweeping of polarization). The cathodic peak height at Ep ) -0.55 V was measured, and the dependence of peak height (Ip) on the concentration of antibodies was plotted. To determine antibodies in animal blood serum, 3 µL of the serum was added to 0.5 mL of 1.9 × 10-2 M BuSChI, and the mixture was diluted to 5 mL with borate buffer (pH 9.05). Further steps were performed as described above. The concentration of antibodies was calculated from Ip at Ep ) -0.55 V using the calibration graph. Preparation of Cellulose Nitrate Film Modified with DNA. A cellulose nitrate film was immersed in 8% aqueous glutaralde(14) Wilson, M. B.; Nakane, P. K. In Immunofluorescence and Related Staining Techniques; Knapp, W., Holuber, K., Wick, G., Eds.; Elsevier/North-Holland Biomedical Press: Amsterdam & New York, 1978; p 215. (15) Egorov, A. M.; Osipov, A. P.; Dzantiev, B. B.; Gavrilova, E. M. Theory and Practice of Enzyme Immunoassay; Vysshaya Shkola: Moscow, 1991; pp 21317 (in Russian). (16) Towbin, H.; Staehelin, T.; Gordon, J. Proc. Natl. Acad. Sci. U.S.A. 1979, 76, 4350-4. (17) Budnikov, G. K.; Medyantseva, E. P.; Ulakhovich, N. A.; Babkina, S. S. Invention Certificate No. 1562831 (USSR), Izobreteniya; Otkrytiya 1990, 17, 226 (in Russian).

3828

Analytical Chemistry, Vol. 68, No. 21, November 1, 1996

hyde for 1 h. The film was then taken out, excess of glutaraldehyde was shaken off, and then the film was dried at room temperature. The film was then allowed to remain in a 10 µg mL-1 DNA solution in SSC buffer for 30 min. This was followed by immersion in 10% BSA in 0.01 M PBS to block the active sites of glutaraldehyde for 10 min.15,16 The film was dried at room temperature after consequent washing with PBS+T and water.18 Spectrophotometric Determination of Antibodies to DNA on a Cellulose Nitrate Film. To obtain data for the calibration graph, the samples with known IgG concentrations in the range from 1 × 10-12 to 1.5 × 10-4 M were applied to a DNA-modified CN film by means of a capillary. After 5 min, the film was immersed in the solution of secondary HRP-labeled antibodies in PBS+T for 15 min. PBS+T was then used for twice washing the film out from unbound conjugate. This step was repeated with pure PBS. The film was immersed in a substrate mixture of 4 mg of 3,3-diaminobenzidine, 10 mL of PBS, and 20 µL of H2O2. The film was washed off with distilled water and blotted after the appearance of brown spots. The intensity of color is estimated by its optical density (A) at 370 nm. The ∆A ) A - Aback (where A is the optical density of a spot in the presence of antibodies, and Aback is that of the background, i.e., of DNA-modified CN membrane exposed to the substrate) dependence on -log CAb was plotted. To determine the unknown concentration of antibodies, the sample was applied to the film, and the procedure was performed as described above. The concentration was calculated from the calibration graph. RESULTS AND DISCUSSION DNA Immobilization on CN Membrane. When developing a method for sensitive and selective determination of antibodies to DNA, it is important to choose an immunosorbent and an approach to modify its surface in a way that provides strong fixation of a component of the immunological couple. In this case, CN films might be considered promising due to their considerable power of immobilizing compounds and obtaining films with predetermined properties. Much attention was paid to the choice of optimal conditions for DNA sorption on CN films. The following parameters were varied: 1, DNA concentration; 2, pH and composition of buffer system; 3, time of immobilization; 4, glutaraldehyde concentration; 5, quantity of immobilied ChE; 6, ChE:DNA:carrier ratio by weight; and 7, the order in which the components were mixed. CN-based membranes prepared as described in the Experimental Section possess very stable working and storage characteristics, reliable bonds with DNA or enzyme, and uniform distribution of components. It should be noted that the conditions chosen provide an optimal activity of ChE for three reasons: first, pH 9.5 is the value at which the maximum of a bell-shaped dependence of the analytical signal on pH occurs; second, the DNA and ChE optimal contents (in % by weight, 0.94-1.86 and 10.6-12.6, respectively) found by us make it possible to detect both steric shielding of the substrate during immunological reaction and ChE interaction with the substrate when antibodies are absent; and third, organic solvents used in CN film preparation provide a homogeneous solution and conserve initial ChE activity during immobilization. (18) Babkina, S. S.; Vinter, V. G.; Zaynullina, A. S. Zh. Anal. Khim. 1994, 49, 1313-1316 (in Russian).

It should be pointed out that, in the blood serum from humans suffering from autoimmune diseases, both antibodies to native (double-stranded) DNA and a large amount of those to denatured (single-stranded) DNA are detected. The latter emerge due to the masking of the native DNA antigenic determinants by its secondary structure.19,20 Taking the existence of the two types of antibodies into consideration, it was of interest to use both denatured DNA (for IEAS) and the native one (for spectrophotometric registration) in autoantibodies determination. The method of DNA immobilization on the CN film enjoys a number of advantages over one using the polystyrene microplate commonly used. Contrary to this, DNA inclusion in the CN matrix in parallel with the treatment with glutaraldehyde (cross-linking agent), and not physical sorption, is responsible for immobilization in our case. Probably, high-molecular-weight products of glutaraldehyde polymerization, usually present in the commercial reagent, play a definite part in DNA-glutaraldehyde interaction.21 As a result, DNA is not being washed out when used. This provides stability of the working surface and thus gives a good reproducibility of results. EIA of Antibodies to DNA with IEAS. CN film modified with ChE and DNA by the proposed method was used as the sensor part of IEAS based on a mercury film-covered silver electrode developed earlier.17,22 Such an electrode or IEAS serves as an immunological solid phase and contains ChE as an enzyme label. On the other hand, it provides electrochemical monitoring of the enzyme activity. One of the products of BuSChI enzymatic hydrolysis (eq 1), a thiol HSR, reacts with Hg, being the electrode material (eq 2), at potentials less negative than -0.50 V. That is why the product of the above interaction, mercury mercaptide (HgSR), has already been present on the electrode at -0.1 V initial potential. HgSR is reduced in the borate buffer medium at -0.55 V (eq 3). ChE

PrCOSCH2CH2N+(CH3)3I- + H2O 98 (BuSChI) + PrCOOH + HSCH2CH2N (CH3)3I (1) (HSR)

-

SR + Hg0 - e f HgISR

(2)

HgISR + e f Hg0 + -SR

(3)

The quantity of enzymatic hydrolysis products depends not only on the activity of the enzyme and the concentration of the substrate but also on the presence of autoantibodies. If IEAS is placed into an analyzing solution that does not contain antibodies, BuSChI is hydrolyzed enzymatically, and mercaptide reduction takes place at the electrode (Figure 1, curve 1). However, when specific antibodies are present in the solution, (19) Seligmann, M.; Arna, R. In The Various Types of DNA Antibodies in Lupus Sera. In Nucleic Acids in Immunology; Pescia, O., Braun, W., Eds.; SpringerVerlag: New York, 1968; p 98. (20) Goldfarb, D. M.; Zamchuk, L. A. Antibodies to Nucleic Acids; Nauka: Moscow, 1975; p 227 (in Russian). (21) Kulis, Yu. Yu. Analytical Systems Based on Immobilized Enzymes; Mokslas: Vilnius, 1981; pp 12-19. (22) Budnikov, H. C.; Medyantseva, E. P.; Babkina, S. S. J. Electroanal. Chem. 1991, 310, 49.

Figure 1. Voltammograms with fast potential scanning (IEAS). (1) 1.9 × 10-3 M BuSChI; (2) 1.9 × 10-3 M BuSChI and 7.0 × 10-9 M antibodies to AMD virus (pH 9.05, v ) 1 V s-1, Estart ) -0.1 V).

Figure 2. Scheme of enzyme immunoassay with IEAS.

they bind rapidly (if there is no transport limitation in the CN film) to give DNA-antibody (DNA-Ab) couples. The complexes so formed produce steric hindrances to approach of substrate molecules to the active sites of ChE, thus reducing the analytical signal (Figure 1, curve 2). Figure 2 shows the scheme of the enzyme immunoassay of antibodies to DNA with IEAS. The conformational distortions in the ChE structure caused by the formation of the immunological couple seem to be a possible reason for the change in the enzyme’s activity. When unbound DNA is introduced into the system, which competes for binding sites with bound molecules, the extent of inhibition decreases, and ChE activity is restored. This fact confirms the formation of immunological couples. IEAS has also been used for IgG determination in both simulated and real samples (IgG is produced by the organism of diseased minks as an antibody to DNA in the case of Aleutian mink disease). A new system for prompt diagnosis has been developed as a result of the present study. Peak current vs concentration of IgG in blood serum of the diseased animals was plotted according to the procedure described in the Experimental Section (Figure 3). The linear part of this plot, in the range from 0.5 × 10-9 to 9 × 10-9 M, is used as a calibration graph. A plato section for the concentrations exceeding 9 × 10-9 M may be ascribed to the complete inhibition of the ChE active sites in the film with immunological couples formed, and it is impossible for the substrate molecules to reach these sites. In this case, only the spontaneous hydrolysis of BuSChI seems to be responsible for the observed current. The results of the determination of specific antibodies are shown in Table 1. Analytical Chemistry, Vol. 68, No. 21, November 1, 1996

3829

Figure 3. Peak current (Ip) dependence on antibodies concentration in the blood serum of animals. Table 1. Results of Determination of Specific Autoantibodies by IEAS (n ) 5, P ) 0.95) inserted, 109 M

found, 109 M

102 RSD

1.5 2.5 5.0 7.0

1.59 ( 0.07 2.45 ( 0.08 5.20 ( 0.10 7.00 ( 0.50

3.9 2.6 1.5 1.2

The results of this determination can also be treated qualitatively, in a “green-red” mode.23 The procedure of diagnosing is based on the difference between the peak heights at -0.55 V in the presence and in the absence of the blood serum examined. This difference should be not less than 2-3s, where s is a standard deviation for the peak height at the potential in the presence of the blood serum of healthy animals.23 In our case, the standard deviation is 0.045 µA (n ) 5). We found 10 diseased animals among the 35 minks examined. The results of diagnosing with IEAS were found to be in good agreement with a routine EIA with spectrophotometric registration. It should be noted that the latter method, on polystyrene microplates, often used in tests for AMD,10 is much more difficult to perform and very time consuming. It involves a number of long-term stages, e.g., antigen sorption on a microtitration plate for 1.5 h (for IEAS, it takes only 0.5 h to prepare the membrane), 1 h incubation with a solution of assayed serum (∼15 min), 1.5 h incubation with a conjugate solution (this stage does not exist for IEAS), and 30 min interaction with an enzyme label (in our case, this process takes place upon exposure to the serum studied). Thus, the use of IEAS makes it possible for us to decrease the time of a single determination from 3 h to 15 min (not taking into account the preparation of the carrier). The procedure is simpler; there is no need to wash the carrier after each stage. Neither special sample preparation nor precise measuring of the sample volume is needed in this method. The lower limit of detection is 0.5 × 10-10 M. The electrode response appears practically instantaneously, making this technique applicable in a “red-green” regime (i.e., for indicating the presence or absence of antibodies). The lack of peak current drop at -0.55 V in the presence of human and bovine IgG proves the selectivity of IEAS. To be more precise, in these cases, peak current drops only by 0.05 and 0.095 µA, respectively, i.e., less than 3s. This again confirms the lack of nonspecific sorption of proteins on the modified CN membrane, after treatment with BSA with high affinity for the sorption centers of the carrier. The kinetic conditions of the experiment reached with stirring reduce the nonspecific sorption.15,16 (23) Verbov, V. N. Solid-Phase Immunoenzyme Analysis; Paster’s Institute Publishers: Leningrad, 1988; Vol. 64, p 3 (in Russian).

3830 Analytical Chemistry, Vol. 68, No. 21, November 1, 1996

The biosensitive part of the sensor developed was found to be stable enough to be used more than 20 times for diagnosing of healthy mink serum without any loss of activity. The reactivation of this part is, however, needed in the presence of the antibodies when immunocomplexes with DNA are formed, which causes a decrease of the analytical signal. Conditions for reactivating the biosensor part (i.e., CN film modified with immobilized ChE and DNA) have been worked out. The repeated use of the developed IEAS as well as automation of the runs was thus made possible. DNA-Ab immunological couples were found to be decomposed by treatment with MgCl2. After the diagnosing, the sensor part of IEAS is allowed to remain in 0.6 M aqueous MgCl2 for 15 min. As a result, the peak current reaches its initial value probably due to the complete decomposition of the DNA-Ab couples. It was established that, when analyzing the serum containing the autoantibodies, the biosensitive part can be regenerated 5-6 times before it should be replaced by a new one. EIA of Antibodies to DNA on CN Films with Spectrophotometric Registration. As it evidently follows from the above findings, the linearity range of IEAS is narrow enough [(0.5-9) × 10-9 M] but suits our practical needs. It is possible to extend this range by changing the concentration of immobilized DNA, the conditions for immobilization, and the fitting conditions for the steric inhibition of ChE active centers due to the formation of DNA-antibody couples. Thus, another approach was developed with DNA-modified CN membranes which overcomes the limitations of IEAS. This method can be assigned to heterogeneous EIA for the separation of bound components from the unbound ones. This is commonly performed with polystyrene microplates on which one of the immunological components has been physically adsorbed. As this appears difficult and time-consuming, finding new ways of heterogeneous EIA involving other solid phases proves to be of vital importance. The new method of CN membrane modification with DNA was found to be useful not only for preparing the biosensitive part of EIAS but also for obtaining the solid phase for EIA with spectrophotometric registration of analytical signal. Such an approach provides acceleration of the procedure, which is particularly important for scaled diagnosing of various diseases. The protocol of EIA of immunoglobulins on modified CN membranes is practically the same as that on microplates. DNA immobilization is followed by an immunological reaction with antibodies, the degree of advancement of the reaction being monitored using HRP-labeled secondary antibodies. HRP catalyzes the 3,3′-aminobenzydine interaction with H2O2 to give a water-insoluble colored product which is adsorbed on the membrane surface. It should be noted that it is much easier to position blood serum samples on the membrane than on a microtitration plate. One should only lightly touch the membrane with a capillary filled with a sample of serum. It consumes about 0.5 µL of the sample. There is no need to wash out unbound nonspecific antibodies. The membrane can be fully dipped into the washing solution, which makes it easier to remove unbound conjugate (secondary Ab-HRP). The membrane is also fully dipped into corresponding liquids when treated with the conjugate or substrate solution. This saves reagents. The conjugate solution can be used repeatedly without any loss of its immunological activity. The main advantage of the described method is its shortcut and easy-to-use approach in mass analyses. The procedure with

Table 3. Test for AMD by EIA on Microplatea and Modified CN Membraneb

Figure 4. ∆A as a function of the concentration of antibodies (cAb), where ∆A ) A - Aback; A is the optical density of a spot in the presence of antibodies to AMD, Aback is the optical density of the background. (1) Semilogarithmical plot; (2) linearized plot of logit ∆A vs (-log cAb), where logit ∆A ) ln[∆A/(1 - ∆A)]. Table 2. Results of Determination of Ab on CN Membrane Modified with DNA (n ) 5, P ) 0.95) inserted, M

found, M

102 RSD

3.7 × 7.4 × 10-9 14.8 × 10-10 5.0 × 10-11

3.7 × 7.5 × 10-9 14.6 × 10-10 5.0 × 10-11

1.4 1.3 2.4 3.5

10-8

10-8

stirring under kinetic conditions makes it possible to shorten the time of assay from several hours to 30 min (except for the time needed for membrane preparation and sampling). The membrane dried after a run can be used as a record of analysis. The developed method of the EIA of antibodies to DNA was used for diagnosing AMD. Sigma-like and linearized dependences of the change in optical density (∆A ) A - Aback, where A is the optical density of a spot in the presence of antibodies, and Aback is one of the background, i.e., of CN membrane exposed to the substrate) upon decadic logarithm of the concentration of autoantibodies are presented at Figure 4. The linear plot (y ) -0.70x + 2.06, r ) 0.999) is employed for quantitative determination of antibodies in a wide concentration range 7.4 × 10-12-1.5 × 10-4 M, which allows evaluation of the disease severity. The results of autoantibody determination are given in Table 2. Table 2 suggests that the error of determination does not exceed 4%. To compare the two approaches, the test for AMD was performed by heterogeneous EIA both on a microtitration plate and on a CN membrane (Table 3). ∆A (370 nm) values measured on the membrane, together with the values of the concentration of antibodies calculated on the basis of the correlation equation, are shown in the table. The results of determination are also presented in a “negative-positive” manner.23 An animal is regarded as diseased if ∆A exceeds 3s above the average ∆A0 ) A0 - Aback (where A0 is the optical density in the presence of the

∆A at 370 nm (membrane)

cAb, M

0.106 0.071 0.151 0.038 0.048 0.074 0.032 0.252 0.196 0.169

1.03 × 10-6 2.42 × 10-7 3.89 × 10-6 2.76 × 10-8 6.16 × 10-8 2.80 × 10-7 1.54 × 10-8 3.18 × 10-5 1.10 × 10-5 6.09 × 10-6

A at 492 nm diagnosing results (n ) 5) (microplate) microplate membrane 0.245 0.285 0.423 0.010 0.105 0.344 0.034 0.511 0.436 0.361

+ + + + + + +

+ + + + + + +

a A ) 0.035, 3s ) 0.105, where A is the optical density for healthy 0 0 animals, blood serum or for nonspecific one. b ∆A0 ) 0.038, 3s ) 0.018, where ∆A0 ) A0 - Aback; Aback is the optical density of the background, i.e., of CN membrane modified with DNA and after exposure to the substrate.

healthy animal blood serum or nonspecific serum). In our runs, the 3s value was 0.018 for n ) 5. The results obtained with the EIA on DNA-modified CN membrane are in good agreement with these of EIA on polystyrene microtitration plates, but the duration of the assay is decreased from more than 4 h to 30 min. The detection limit is 7.4 × 10-12 M. The specificity of the assay is confirmed by the lack of any response for immunoglobulins from bovine, rabbit, and pig. Namely, ∆A values are 0.038, 0.048, and 0.05, respectively, i.e., less than 3s. Hence, secondary HRP-labeled antibodies take no part in the immunological reaction and are not sorbed on the membrane. This evidence is in favor of the lack of nonspecific sorption of the proteins on modified CN membranes. It may be concluded that the developed method of DNA immobilization on a cellulose nitrate membrane makes it possible to devise a principally new type of sensor, i.e., and immunoenzyme amperometric sensor for autoantibodies determination. Such sensors make possible the automation of EIAs and the construction of immunoenzyme analyzers on their basis. On the other hand, this method of DNA immobilization is found to be most effective for the development of solid-phase EIA (on CN membrane), which is more simple and rapid when used in mass analyses but at the same time is highly sensitive and selective. It is also valuable that the suggested approach may be used for rapid determination of different types of autoantibodies and, hence, for diagnosing a number of autoimmune diseases. This approach can be applied to any antigen-antibody couple and is suitable for the case of the simultaneous presence of different antibodies in a sample, which is of particular importance for further improvement of clinical assays. Received for review October 4, 1995. Accepted July 10, 1996.X AC950645V X

Abstract published in Advance ACS Abstracts, August 15, 1996.

Analytical Chemistry, Vol. 68, No. 21, November 1, 1996

3831