Comparison of Direct and Indirect Enzyme Immunoassays for the


Comparison of Direct and Indirect Enzyme Immunoassays for the...

0 downloads 193 Views 658KB Size

Chapter 24

Comparison of Direct and Indirect Enzyme Immunoassays for the Detection of the Mycotoxin Citrinin 1

1

2

Ewald Usleber , Erwin Märtlbauer , and David Abramson 1

Institute for Hygiene and Technology of Food of Animal Origin, Veterinary Faculty, University of Munich, Schellingstrasse 10, 80799 Munich, Germany Agriculture and Agri-Food Canada, Winnigpeg Research Center, 195 Dafoe Road, Winnipeg, Manitoba R3T 2M9, Canada

Downloaded by UNIV LAVAL on July 11, 2016 | http://pubs.acs.org Publication Date: May 5, 1996 | doi: 10.1021/bk-1996-0621.ch024

2

Polyclonal antisera against the mycotoxin citrinin, produced in rabbits after immunization with a citrinin-keyhole limpet hemocyanin conjugate, were used to establish competitive direct and indirect enzyme immunoassays (EIA). A citrinin-horseradish peroxidase conjugate was used as the labelled antigen in a direct EIA, a citrinin-glucose oxidase conjugate was employed as the solid phase antigen in indirect EIA. The antibodies used in this study were highly specific for citrinin, having no measurable cross-reactivity with austdiol, alternariol, deoxynivalenol, or ochratoxin A. After optimization of test parameters, the detection limits for citrinin in buffer solutions by direct and indirect EIA were 5 — 10 ng/mL and 1—2 ng/mL, respectively. The mycotoxin citrinin (Figure 1) is produced by several Aspergillus and Pénicillium species (7, 2). It was originally isolated as an antibiotic in 1931, but its utility, as such, was negated due to its nephrotoxicity (3). Citrinin often occurs together with another common mycotoxin, ochratoxin A, as a natural contaminant in various cereals (4 — 9). Citrinin, like ochratoxin A, acts primarily as a nephrotoxin (10) and teratogen. Although a number of physico-chemical methods has been developed for the detection of citrinin in foods, feeds, and biological fluids, including thin-layer chromatography (6, 11-13) and liquid chromatography (14-16), so far no satisfactory routine analytical method for this toxin is available (77, 18). Immunochemical approaches to the detection of mycotoxins have been of increasing importance throughout the last decade. In particular, enzyme immunoassays have been established for many mycotoxins as a convenient alternative for detecting these substances in foods and feeds (19). Recently we have described the production of polyclonal antibodies against citrinin and their use in indirect EIA for citrinin in wheat flour (20). Here we describe the development of a direct EIA for citrinin, 0097-6156/96/0621-0322S15.00/0 © 1996 American Chemical Society

Beier and Stanker; Immunoassays for Residue Analysis ACS Symposium Series; American Chemical Society: Washington, DC, 1996.

Direct and Indirect Enzyme Immunoassays

Downloaded by UNIV LAVAL on July 11, 2016 | http://pubs.acs.org Publication Date: May 5, 1996 | doi: 10.1021/bk-1996-0621.ch024

24. USLEBER ET AL.

Beier and Stanker; Immunoassays for Residue Analysis ACS Symposium Series; American Chemical Society: Washington, DC, 1996.

323

324

IMMUNOASSAYS FOR RESIDUE ANALYSIS

using a citrinin-horseradish peroxidase conjugate. Important parameters which influence the test practicability both of the direct and indirect formats are compared. Materials and Methods

Citrinin, ochratoxin A, austdiol, alternariol, deoxynivalenol, formalde­ hyde 37% solution, 3,3\5,5'-tetramethylbenzidine (TMB), casein (sodium salt) and polyoxyethylenesorbitan monolaurate (Tween-20) were purchased from SigmaAldrich Vertriebs GmbH, Deisenhofen, Germany. Glucose oxidase (GOX) from Aspergillus niger van Tieghem (E.C. 1.1.3.4), molecular weight 186,000, and key­ hole limpet hemocyanin (KLH), molecular weight 3-7.5 million, from Megathura crenulata L. were obtained from Boehringer Mannheim GmbH Biochemicals, Mann­ heim, Germany. EIA grade horseradish peroxidase (HRP), molecular weight 40,000 (EC 1.11.1.7) also was obtained from Boehringer. Affinity chromatography purified sheep anti-rabbit immunoglobulin G (IgG) was used as described earlier (27). Goat anti-rabbit IgG-HRP conjugate was purchased from Sigma-Aldrich.

Downloaded by UNIV LAVAL on July 11, 2016 | http://pubs.acs.org Publication Date: May 5, 1996 | doi: 10.1021/bk-1996-0621.ch024

Materials.

Anti-Citrinin Antisera. Details of production of antisera against citrinin have been described earlier (20). In brief, three rabbits were intradermally immunized with a citrinin-keyhole limpet hemocyanin conjugate prepared by formaldehyde conden­ sation. Subcutaneous booster injections were given at week 15 after primary immuni­ zations. Blood samples were taken at 2—3 week intervals, and the serum was obtained by centrifugation. Sera of each rabbit, collected from week 22 to 27, were individually pooled (each 70-80 mL) and stored frozen at -18 °C without further treatment. Production of a Citrinin-Horseradish Peroxidase Conjugate. A citrinin-HRP conjugate was prepared for use in direct competitive EIA. HRP (2.0 mg) was dis­ solved in 0.8 mL sodium acetate buffer (0.1 M ; pH 4.2). Citrinin was dissolved in methanol to give a 5 mg/mL solution, and 0.2 mL added to the HRP-sodium acetate solution. A 100 μ\, aliquot of 37% formaldehyde solution was then added, and the mixture incubated for 24 h at 37 °C. Afterwards, the citrinin-HRP was dialyzed at 4 °C for three days against three changes of phosphate buffered saline (PBS; 0.01 M phosphate buffer, pH 7.2, containing 0.1 M NaCl) and stored at -18 °C.

Citrinin was conjugated to the glycoprotein GOX for use as the solid phase antigen in indirect competitive EIA (20). In brief, GOX (3.7 mg) was dissolved in 0.8 mL sodium acetate buffer (0.1 M ; pH 4.2) and mixed with citrinin solution (1 mg/mL methanol; 0.2 mL). A 324 aliquot of 37% formaldehyde solution was then added, and the mixture incu­ bated for 72 h at ambient temperature (22 °C). Afterwards, the citrinin-GOX was dialyzed at 4 °C for three days against three changes of PBS and stored at -18 °C. Production of a Citrinin-Glucose Oxidase Conjugate.

Optimum concentrations of antiserum and enzyme conjugate were determined by twofold checkerboard titration with and without addition of citrinin. A microtiter plate was coated (100 ^L/well) with goat

Titration of Antisera with Citrinin-HRP.

Beier and Stanker; Immunoassays for Residue Analysis ACS Symposium Series; American Chemical Society: Washington, DC, 1996.

24. USLEBER ET AL.

Direct and Indirect Enzyme Immunoassays

anti-rabbit IgG (10 μg/mL carbonate-bicarbonate buffer [0.05 M; pH 9.6]) overnight at ambient temperature in a humid chamber. The solution was removed, and free protein-binding sites of the wells were blocked with casein sodium salt, 2% in PBS, for 30 min at ambient temperature. The plate was washed three times with a 0.85% NaCl solution containing Tween-20 (250 μί/ί) and drained. Then 35 uL toxin standard buffer solution (PBS containing 10% methanol), toxin-free or containing 100 ng citrinin/mL was added to the wells of each half of the plate, followed by 35 /xL of a serial dilution (in PBS) of antiserum pool (rabbit #1, #2, #3), and 35 μΐ. of a serial dilution of citrinin-HRP (in PBS containing 1 % casein sodium salt), and incubated for 2 h at ambient temperature. Then each plate was washed as above, and 100 μL of enzyme substrate/chromogen solution (22) containing H 0 (3 mM) and TMB (1 mM) in potassium citrate buffer (0.2 M; pH 3.9) were added per well. After 15 min the enzyme reaction was stopped with 1 M HS0 (100 μί per well) and the absorbance measured at 450 nm. A combination giving the desired results, i.e., absolute absorbance values of the toxin-free wells of > 1.0 and absorbance reduction in the corresponding toxin-containing well of >80 %, was chosen for establishment of the direct EIA. 2

Downloaded by UNIV LAVAL on July 11, 2016 | http://pubs.acs.org Publication Date: May 5, 1996 | doi: 10.1021/bk-1996-0621.ch024

325

2

2

4

Competitive Direct EIA. A microtiter plate was coated (100 μΌν/οΙΙ) with antirabbit IgG (10 μg/mL carbonate-bicarbonate buffer) overnight at ambient tempera­ ture in a wet chamber. The solution was removed, and free protein-binding sites of the wells were blocked with casein sodium salt, 2% in PBS, for 30 min at ambien temperature. The plate was washed three times with a 0.85% NaCl solution con­ taining Tween-20 (250 μί/ί) and drained. To each well, 35 μΐ. citrinin standard solution (in PBS containing 10% methanol), 35 μ]~, antiserum against citrinin (pool rabbit #1, diluted 1:2000 with PBS), and 35 μΐ. citrinin-HRP (diluted 1:400 with PBS containing 1 % casein sodium salt) were added and incubated for 2 h at ambie temperature. Then each plate was washed as above, and further treated with enzyme substrate solution as described above. All standards solutions were analyzed in qua­ druplicate. After absorbance measurement, the test was evaluated using an on-line PC and an EIA calculation software developed by Martlbauer (23), which uses a cubic spline function for calculation of the standard curve. The program also determines the detection limit (students r, η=4; 95% confidence limit) and the 50% inhibition dose. The measuring range of the standard curve usually is from 20% to 80% relative binding (B/B χ 100). 0

Competitive Indirect EIA. The indirect EIA was performed essentially as de­ scribed earlier (20). In brief, the plates were coated with 100 μι/well of the citrininGOX conjugate (1:1000 in sodium carbonate buffer) overnight at ambient tempera­ ture. The citrinin-GOX solution was removed, free protein-binding sites of the wells were blocked with 2% casein sodium salt/PBS, then the plate was washed and drained. Fifty μ\^ each of citrinin standard solution (in 10 %methanol/PBS) and the anti-citrinin antiserum (pool rabbit #3; diluted 1:1000 with PBS) were added to each well and the plate was incubated for 1 h at room temperature. The plate was washed and goat anti-rabbit IgG-HRP conjugate (1:5000 in 1 % casein sodium salt in PBS;

Beier and Stanker; Immunoassays for Residue Analysis ACS Symposium Series; American Chemical Society: Washington, DC, 1996.

326

IMMUNOASSAYS FOR RESIDUE ANALYSIS

100 per well) was added. After 1 h at ambient temperature, the plate was washed and further treated with enzyme substrate/chromogen solution as described above.

Downloaded by UNIV LAVAL on July 11, 2016 | http://pubs.acs.org Publication Date: May 5, 1996 | doi: 10.1021/bk-1996-0621.ch024

Results and Discussion

By using formaldehyde, a common conjugation reagent reactive with amines, amides, guanidino and phenolic groups (24), citrinin was successfully bound to KLH, GOX, and HRP. Conjugation ratio (moles of citrinin per mole of carrier protein) in the citrinin-GOX were found to be approximately 3:1 by UV spectro­ scopy (20). Conjugation ratios in the citrinin-KLH and citrinin-HRP could not be de­ termined because of extensive precipitation of the former and ambiguous UV spectra obtained from the latter. However, in past experiments slightly turbid or even fully precipitated hapten-KLH conjugates have been found to be very powerful immunogens, which gave useful antibodies against a number of problem haptens (25-27). In all three rabbits dosed with the citrinin-KLH conjugate high specific serum titers could be detected as early as five weeks after immunization. All sera gave standard curves in indirect EIA which enabled specific detection of citrinin in the low ng/mL range. No cross-reactivity was found with the structurally similar myco­ toxins austdiol and alternariol, nor with the mycotoxins ochratoxin A and deoxyniva­ lenol which are likely to co-occur together with citrinin in high concentrations (20). For standardization of the indirect EIA, serum pool of rabbit #3 was chosen be­ cause it gave the most sensitive test system, with a detection limit for citrinin buffer solutions in the range of 1 -2 ng/ml. The dilutions of antiserum (1:1000) and solid phase citrinin-GOX (1:1000) were sufficiently high to enable their use in routine analysis, considering the serum pool volume of 75 mL and the simple conjugation protocol for citrinin-GOX. In contrast, for direct EIA the serum pool #3 and the citrinin-HRP had to be used at dilutions of approximately 1:100 to give absorbance values of — 1.0 units. Therefore, serum of rabbit #1, which had a higher serum titer (Figure 2) and showed stronger binding to the citrinin-HRP, was used for the direct EIA in a 1:2000 dilution. The amount of citrinin-HRP necessary for this EIA (dilution 1:400) was still comparatively high, corresponding to a peroxidase working concentration of approximately 2 — 3 μg/mL. The standard curve detection limit of this assay for citrinin (5 — 10 ng/ml) was slightly higher than that of the indirect test format (Figure 3). Further work will aim at improving the conjugation of citrinin to HRP, because the direct EIA format is in practice more convenient to perform than the indirect test, which requires one additional incubation step. However, both tests provided simple reliable procedures with low intra-plate coefficients (1.5-7%;n=4). Plates coated with citrinin-GOX (indirect EIA) or antirabbit IgG could be stored for at least 3-4 weeks at 4 °C, ready for use. The sensi tivity and specificity of the EIA described above would probably be sufficient to assay citrinin in agricultural commodities. Initial tests, using the indirect EIA for analysis of citrinin in artificially contaminated wheat flour, had shown that at toxin levels ranging from 200-2000 ng/g good recoveries of 89-104% could be achieved with coefficients of variation (n=4) of 6.9-13% (20). The immunoassays, with their advantages of speed and simplicity, could conveniently be used for screening samples followed by confirmation of positive results by physicochemical methods.

Beier and Stanker; Immunoassays for Residue Analysis ACS Symposium Series; American Chemical Society: Washington, DC, 1996.

Direct and Indirect Enzyme Immunoassays

Downloaded by UNIV LAVAL on July 11, 2016 | http://pubs.acs.org Publication Date: May 5, 1996 | doi: 10.1021/bk-1996-0621.ch024

24. USLEBER ET AL.

100

1,000

10,000

327

100,000

Antiserum Pool Dilution Figure 2. Comparison of antisera pools of rabbits #1, #2, and #3, using two­ fold checkerboard titration (Results are shown for one citrinin-HRP dilution [1:400] only) in direct EIA. Serial dilutions of antiserum were incubated with the citrinin-HRP and either toxin-free (CT-) or toxin-containing (CT+; 100 ng/mL) standard buffer solutions. 100-

direct E I A 20

indirect E I A

0.1

1

10

100

Citrinin (ng/mL) Figure 3. Comparison of standard curves for competitive EIA detection of citrinin, using direct EIA or indirect EIA format, respectively. The χ axis shows the logarithm of mycotoxin concentration. The y axis shows the corres­ ponding absorbance value Β relative to the absorbance of the negative control B , expressed as (B/B) χ 100. B values were 1.0 (direct EIA) and 1.1 (indirect EIA) absorbance units. Each point represents the mean of four deter­ minations. 0

0

0

Beier and Stanker; Immunoassays for Residue Analysis ACS Symposium Series; American Chemical Society: Washington, DC, 1996.

328

IMMUNOASSAYS FOR RESIDUE ANALYSIS

Acknowledgments We thank External Affairs and International Trade Canada for travel funding, and M. Straka, M. Lorber, and D. Smith for technical assistance. Literature Cited 1. 2.

Downloaded by UNIV LAVAL on July 11, 2016 | http://pubs.acs.org Publication Date: May 5, 1996 | doi: 10.1021/bk-1996-0621.ch024

3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19.

20. 21. 22. 23. 24.

Betina, V. 1989. In Mycotoxins: chemical, biological and environmental aspects; Betina V., ed.; Elsevier: Amsterdam, 1989; pp 174-191. Wilson, D. M. In Biotoxins, Biodegradation and Biodeterioation Research; O'Rear, C. E.; Llewelyn, G. C., eds.; Plenum Press: New York, NY, 1994, pp 65-73. Hetherington, A. C.; Raistrick, H. Thom. Phil. Trans. Roy. Soc. London Ser. Β 1931, 220, 269-295. Scott, P. M . ; van Walbeek, W.; Kennedy, B.; Anyeti, D. J. Agric. Food Chem. 1972, 20, 1103-1109. Krogh, P.; Hald, B.; Pedersen, E. J. Acta. Pathol. Microbiol. Scand. Sect. Β 1973, 81, 689-695. Chalam, R. V.; Stahr, Η. M. J. Assoc. Off. Anal. Chem. 1979, 62, 570-572. Hökby, E.; Hult, K.; Gatenbeck, S.; Rutqvist, L. Acta. Agric. Scand. 1979, 29, 174-178. Osborne, B. G. Food Cosmet. Toxicol 1980, 18, 615-617. Nelson, T. S., Kirby, L. K.; Beasley, J. N.; Johnson, Ζ. B; Ciegler, A. Poult. Sci. 1985, 64, 464-468. Phillips, R. D.; Berndt, W. O; Hayes, A. W. Toxicology 1979, 12, 285-298. Stubblefield, R. D. J. Assoc. Off. Anal. Chem. 1979, 62, 201-202. Gimeno, Α.; Martins, M. L. J. Assoc. Off. Anal. Chem. 1983, 66, 85-91. Gimeno, A. J. Assoc. Off. Anal. Chem. 1984, 67, 194-196. Lepom, P. J. Chromatogr. 1986, 355, 335-339. Marti, L. R., Wilson D. M.; Evans, B. D. J. Assoc. Off. Anal. Chem. 1978, 61, 1353-1358. Phillips, R. D.; Hayes, A. W.; Berndt, W. O. J. Chromatogr.1980, 190, 419-427. Scott, P. M. J. Assoc. Off. Anal Chem. 1991, 74, 120-128. Trucksess, M. W. J. AOAC Int. 1994, 77, 135-141. Chu, F. S. In Residue Analysis in Food Safety: Applications of Immunoassay Methods; Beier, R. C.; Stanker, L. H . , eds.; American Chemical Society: Washington, DC, 1996 (this ACS symposium volume). Abramson, D.; Usleber, E.; Märtlbauer, E. Appl. Environ. Microbiol. 1995, 61, 2007-2009. Märtlbauer, E.; Gareis, M.; Terplan, G. Appl. Environ. Microbiol. 1988, 54, 225-230. Gallati, H.; Pracht, J. J. Clin. Chem. Clin. Biochem. 1985, 23, 453-460. Märtlbauer, E. Enzymimmuntests für antimikrobiell wirksame Stoffe; Ferdinand Enke Verlag: Stuttgart, Germany, 1993; pp 206-222. Wong, S. S. Chemistry ofprotein conjugation and cross—linking; CRC Press: Boca Raton, FL, 1993; pp 101-103.

Beier and Stanker; Immunoassays for Residue Analysis ACS Symposium Series; American Chemical Society: Washington, DC, 1996.

24. USLEBER ET AL. 25. 26. 27.

Direct and Indirect Enzyme Immunoassays

329

Usleber, E . ; Straka, M . ; Terplan, G. J. Agric. Food Chem. 1994, 42, 1392-1396. Renz, V.; Terplan, G. Arch. Lebensmittelhyg. 1988, 39, 30-33. Schneider, E.; Usleber, E.; Märtlbauer, E. In Residue Analysis in Food Safety: Applications of Immunoassay Methods; Beier, R. C.; Stanker, L. H . , eds.; American Chemical Society: Washington, DC, 1996 (This ACS symposium volume).

Downloaded by UNIV LAVAL on July 11, 2016 | http://pubs.acs.org Publication Date: May 5, 1996 | doi: 10.1021/bk-1996-0621.ch024

RECEIVED August 16, 1995

Beier and Stanker; Immunoassays for Residue Analysis ACS Symposium Series; American Chemical Society: Washington, DC, 1996.