Biotecnologia Aplicada Elfos Scientiae ISSN: 0684-4551 Vol. 15, Num. 4, 1998

Rolando Ochoa Azze

DACTA, Instituto Finlay, Ave. 27 No. 19805, AP 16017, La Lisa CP 11600, La Habana, Cuba.
Fax: (53-7) 33 6754; E-mail: inmunoquimica@finlay.edu.cu

Code Number: BA98041
Size of Files:
      Text: 39K
      Graphics: Line drawings (gif) - 22K

ABSTRACT

A new ELISA has been standardised, using two monoclonal antibodies directed against HBsAg. The sheep anti-HBsAg/alkaline phosphatase conjugate and the positive control were dehydrated at 37°C for 18 h and 35% relative humidity. Sucrose was used as stabiliser with an optimum concentration of 125 mg/mL. The one step ELISA is carried out by adding a washing solution into control wells and samples into the remainder. Sensitivity and specificity were 100%. Discrimination value was 0.22 U/mL. Stability was studied for coating antibodies, dehydrated conjugate and the dehydrated positive control. The stability during their use was over 80% at 150 days and storage stability at 2-8 °C was over 80% a year after.

Keywords: conjugate stability, HBsAg, HBsAg stability, new format ELISA, one step ELISA

Se estandarizó un ELISA en el que se usan dos anticuerpos monoclonales de captura anti-HBsAg de diferente especificidad. El conjugado anti-HBsAg (ovino)/fosfatasa alcalina y el control positivo se presentan desecados en los pocillos mediante incubación por 18 h a 37°C y humedad de 35%. La concentración óptima de sacarosa empleada como elemento estabilizador fue de 125 mg/mL. Para el ELISA en un paso de reacción se requiere tan sólo añadir solución de lavado en los pocillos correspondientes a los controles, las muestras se adicionan en los pocillos restantes. La sensibilidad y especificidad fueron del 100%. El valor de discriminación de 0,22 U/mL. La estabilidad de uso y en anaquel (2-8°C), fue superior al 80% a los 150 días y al año respectivamente, tanto para los anticuerpos de recubrimiento como para el conjugado y el control positivo desecados.

Palabras claves: ELISA de nuevo formato, ELISA en un paso, estabilidad de conjugados, estabilidad del HBsAg

In recent years, non-competitive solid-phase enzyme immunoassays (ELISA), based on a one-step sandwich principle, have been developed in order to simplify the technical procedures. The simultaneous addition of sample and conjugate is possible because of non-interference between immobilised molecules (antigens or antibodies) and conjugates (antibody/enzyme or antigen/enzyme) [1-6]. These assays achieve high sensitivity without a reduction in specificity [4, 5, 7].

Immunochromatographic assays have also been developed, with enzymes and depot substrates or coloured particles. These assays are read visually and require no laboratory equipment. Quality controls can also be included in an assay [8, 9].

We decided to develop a method that combines the advantages of new generation ELISA and some of the benefits of immunochromatographic assays. We have produced an ELISA for the detection of HBsAg with the conjugate and positive and negative controls dehydrated into the wells of the polystyrene plate.

A sheep anti-HBsAg/alkaline phosphatase (AP) conjugate made in the Immunoassay Centre (IAC), Havana, Cuba, was used at 0.28 mg/mL, and was diluted in a 0.05M tris buffer solution, 0.15 M NaCl, 0.5mg/mL NaN3, 0.05% tween 20 (v/v) pH8, containing 10mg/mL of bovine albumin (Merck, Germany) and 10% (v/v) of normal sheep serum IAC. Sucrose (Merck, Germany.) was added to a concentration of 500, 250, 125, 62.5 and 31.25 mg/mL. The selected concentration was that which achieved the best discrimination between HBsAg negative and HBsAg positive samples of 2 and 1 U/mL (from the Paul Ehrlich Institute, Germany (PEI)).

Liquid positive control (PC), was produced from heat inactivated human serum containing 44 mg/mL of the HBsAg subtype ad IAC. It was stabilised in a 0.02 M tris buffer solution, 0.15 M NaCl, 2 mg/mL NaN3 pH7.2, containing 80 mg/mL bovine albumin (Merck, Germany) and was prepared to a final concentration of 6U/mL. The dehydrated positive control (DPC) was made with the same reagents. It was mixed with the conjugate to a final concentration of 6U/mL.

The negative control (DNC) was the dehydrated conjugate added to the wells after the dilution with the buffer solution (0.015 M tris, 0.15 M NaCl, 0.2mg/mL NaN3, pH 7.8 (TBS) with 0.05% (v/v) tween 20). The liquid negative control was the HBsAg negative human serum (NHS) from the UMELISA HBsAgÒ kit IAC.

Liquid and dehydrated controls were compared by analysing 20 replicates with the Student's T test, previous determination of the normality of each distribution was carried out using the Kolmogorov Smirnov test [10].

a) Plates. MaxiSorpÒ 8 x 12 wells polystyrene (PS) plates (Nunc, Denmark).

b) Capture antibodies. Two different monoclonal antibodies anti-HBsAg to epitope "a"; Hep 1 (IgG2b) (Center for Genetic Engineering and Biotechnology, Havana, Cuba) [11, 12] and Hep 8 (IgG1) IAC [13], were diluted to 10 mg/mL and 5 mg/mL, respectively, in a buffer solution (0.05 M tris, 0.2 mg/mL NaN3, pH8.0). 100 mL of each solution per well was added and plates were incubated at 25 °C for 18 h in a humidity chamber. Unbound antibodies were removed by washing (1 x 30 seconds) with 300 mL per well of 0.15 M phosphate buffered saline (PBS), pH 7.3 with tween 20 0.05% (v/v).

c) Blocking and stabilisation. 300 mL of bovine albumin (Merck, Germany) 1 mg/mL and sucrose (Merck, Germany) 50mg/mL in PBS were added to each well. Plates were incubated at 25°C for 2 h.

d) Addition of controls and conjugate. After aspiration of the blocking solution, 50 mL of the conjugate was added to each well of columns 2-12. 50mL of the positive control plus conjugate (DPC) was added to each well of the first column. Plates were dried at 37°C for 18 h at a relative humidity of 35%.

e) Storage: Plates were sealed with silica gel in nylon bags and they were stored between 2-8 °C.

100 mL of the washing solution (TBS with 0.05% tween 20) was added to each well of the first column and to wells assigned as DNC. 100 mL of the sample was added to the remaining wells. Each well was mixed by pipetting. Plates were incubated for 2 h at 37 °C. Unbound samples and controls were removed after six washes. Inverted plates were tapped a few times on a paper towel and 100 mL of p-nitrophenylphosphate (Sigma) 1mg/mL was added to each well in 0.92 M diethanolamine buffer pH9.8. Plates were incubated for 30 min at 37 °C, and then the absorbances at 405 nm were read with a PR-521 reader IAC.

This was calculated using 100 serum samples from HBsAg negative blood donors. The mean (N) absorbance of DNC was subtracted from the absorbance of each sample. 100 standardised values were derived and the largest (k) was chosen as the cut-off value [14].

Sample values (S) were considered positive when S-N > k.

a) Diagnostic sensitivity and specificity. These were estimated using three panels of sera previously tested by UMELISA HBsAgÒ and HBsAg Confirmatory TestÒ IAC. The first panel consisted of 80 HBsAg negative and rheumatoid factor positive serum samples (from the Rheumatology Institute, Havana, Cuba), the second panel of 20 serum samples from HBsAg positive blood donors and the third one of 4 HBsAg positive serum samples (PEI) between 2 and 0.25 U/mL. The sensitivity and specificity was calculated according to the following formulae [15].

b) Discrimination value and width of the "grey zone". HBsAg positive samples of 2, 1, 0.5, 0.25, 0.125 and 0.0625 U/mL were produced by two fold dilutions of an HBsAg positive serum PEI with NHS. They were analysed daily for 20 days and the width of the "grey zone" (between 5 and 95% of the positive results) and the discrimination value (50% of positive results) were calculated [16].

c) Stability during use. The stability of the reaction plates (coating antibodies and dehydrated conjugate) was evaluated by repeated opening of the nylon bag, extraction of strips and re-closure. The stability was determined by comparing the mean absorbance of PC in these strips with the absorbance of PC in control strips. The stability of DPC was determined using the same methodology. The stability of the reaction plates and DPC was assessed at 30, 60, 90, 120 and 150 days. It was calculated as follows [13].

DPC stability % = (tested DPC/control DPC) x100
Plates stability % = (tested PC/control PC) x 100

Stability above 80% was considered good.

d) Storage stability. The stability of the DPC and reaction plates stored at 2-8 °C was determined by comparing the mean absorbance detected in testing at 6, 9 and 12 months with the absorbance detected on day one. Storage stability was calculated as follows [13].

DPC stability % = (tested DPC/initial reading) x 100
Plates stability % = (tested PC/initial reading) x 100

Stability above 80% was considered good.

Conjugates have previously been placed in the wells in some ELISA in order to simplify the technical procedure, e.g. Organon Teknika (Vironostika HIV Uni-formII) which uses lyophilised spheres of conjugate [6]. We decided to employ dehydration in order to facilitate the production process and to simplify the sealing of the plates. We also used dehydrated controls. The DPC was calculated for a final concentration of 6 U/mL, to produce an absorbance distinguishable from the DNC. The mean absorbance of the DPC was slightly lower than the liquid control, and the mean absorbance value of the dissolved conjugate in the washing solution (DNC) was slightly larger than NHS (Table 1).

Table 1. Statistical analysis between HBsAg positive and negative controls.

CV: Coefficient of variation.
* p: 0.018
** p: 0.011

The best concentration of sucrose was 125 mg/mL, this achieved the best discrimination between HBsAg negative and HBsAg positive samples of 2 and 1U/mL (PEI) (Figure 1). However, it may be necessary to use a different sucrose concentration if different conjugates are used.

It is interesting that low concentrations of sucrose can block the adsorption of the conjugate to the PS surface. On the other hand, an excess concentration of sucrose could be harmful (Figure 1), because it may interfere with the antibody/antigen interaction. For that reason, the determination of the correct sucrose concentration is essential.

Large concentrations of sucrose, even 600mg/mL, have been used in immunochromatographic assays as a support for the surface movement of the conjugate [8]. It should have had a similar role in our assay. Sucrose takes part in hydrogen bonding with chemical groups of the PS surface, especially on high binding surfaces, rich in hydrophilic groups, such as MaxiSorpÒ plates [17] and with hydrophilic macromolecules such as the conjugate. It hinders hydrophobic bonds and the stable binding of protein molecules. The binding of carbohydrates is not stable, but sucrose could be considered as a blocking agent. Also, its known stabilising role in the lyophilisation process, and the dehydration method (18-23) we use for the conjugate and the HBsAg positive controls, could allow its use in diverse immunoassays.

The dehydrated material remained firmly adherent to the bottom of the wells, which facilitated the manipulation of the reaction plates (strips), it also diluted quickly, which is important in the technical procedure.

The cut-off value was S-N > 0.05 (S > N + 0.05). We decided to use the method of limit value [14], calculated not with the absorbance values but standardised with the mean absorbance of DNC of each assay. We not only avoided false positive results, characteristic of this method [24], but also false negatives, which was demonstrated after correctly classifying HBsAg negative and positive panels (Figure 2). Therefore, the sensitivity and specificity was 100%.

The narrow "grey zone", between 0.1 and 0.5U/mL, gives a high precision [16]. The discrimination value was 0.22U/mL (Figure 3).

The use of the "strip by strip" method is important if small sample numbers are to be tested. We found excellent stability, higher than 80% at 150 days, for DPC and plates (Table 2). This stability is better than those previously reported [1-3, 6].

Table 2. Stability during use by means of repeated opening of the nylon bag, extraction
of strips and re-closing.

* It is referred to coating antibodies and the dehydrated conjugate.
Time 0 is the first reading.
DPC: dehydrated positive control.
PC: liquid positive control.
CV: coefficient of variation.
DPC stability (%): (tested DPC/control DPC) x 100.
Plate stability (%): (tested PC/control PC) x 100.

The storage stability (2-8°C) of the plates and DPC was good at 12 months (Table 3).

Table 3. Storage stability of the dehydrated positive control and reaction plates stored at 2-8 ° C.

* It is referred to coating antibodies and the dehydrated conjugate.
Time 0 is the first reading.
DPC: Dehydrated positive control.
PC: Liquid positive control.
CV: Coefficient of variation.
DPC stability (%): (tested DPC/initial reading) x 100.
Plate stability (%): (tested PC/ initial reading) x 100.

We believe sucrose to be very important for good stability. The activity of the proteins decreases with the lyophilisation and the dehydration [18-23]. In order to avoid this phenomenon many additives have been used, such as carbohydrates, which substitute water in order to stabilise proteins by means of hydrogen bonds [18-22]. Sucrose has demonstrated its role as a stabiliser, and this is reinforced in this investigation.

The stability should be similar for other dehydrated conjugates at least those from a sheep source. We have found similar behaviour in conjugates with horseradish peroxidase enzyme (unpublished data) and we believe that other markers should be studied. Controls should be carefully selected for each assay, positive control must be dehydration resistant like HBsAg.

This method not only simplifies the technical procedure but should also reduce the cost of a diagnostic kit because less equipment is required. An ELISA kit would only require 3 bottles; for the washing solution, the substrate and the substrate buffer. With coloured particles only one bottle would be sufficient. Therefore, this new method could be attractive for other immunoassays.

To the Immunochemistry Department of the Immunoassay Centre, specially Hepatitis Lab., supplier of resources used in our research.

1. Organon Teknika, Holland. Hepanostika HBsAg Uni-Form. Instructions. 1988.2. Sanofi Diagnostics Pasteur, France. Monolisa AgHBs 2nd generation. Instructions. 1992.

2. Sanofi Diagnostics Pasteur, France. Monolisa AgHBs 2nd generation. Instructions. 1992.

3. Labsystems, Finland. HBsAg EIA. Instructions for use. 1992.

4. Chernesky M, Crawford J, Castriciano S, Mahony J. The diagnosis of acute viral hepatitis A or B by microparticle enzyme immunoassay. Journal of Virological Methods 1991;34:291-296.

5. Fayol V, Ville G. Evaluation of automated enzyme immunoassays for several markers for hepatitis A and B using the Abbott IMxÒ analyser. Eur J Clin Chem Clin Biochem 1991;29:67-70.

6. Organon Teknika, Holland. Vironostika HIV Uni-form II. Instructions. 1995.

7. Ferguson M, Pipkin P, Heath A, Minor Ph. Working Standards for Hepatitis B Surface Antigen for Use in the UK Blood Transfusion Service: Results of a Collaborative Study. Vox Sang 1993; 65:303-308.

8. May K, Evans M, Richards I, inventors. UK Patent application GB 2204398 A. 1988 Nov 9.

9. Canales I, Asensio A, Benítez J. Rapid, one step immunoassay for HBsAg. Biotecnología Habana'97. Avances en Biotecnología Moderna; 1997 Dic 1-6; La Habana. Cuba: 1997;4:D9.

10. StatgraphicsÒ plus for windowsTM. Version 1,0. Statistical Graphics Corp. USA, 1994.

11. Valdés R, Leyva JL, González E, Mainet D, Costa L. Caracterización de anticuerpos monoclonales contra el antígeno de superficie del virus de la hepatitis B. Biotecnología Aplicada 1994;11:219-224.

12. Almagro I, Agraz A, Fernández de Cossio ME, Lugo V, Duarte C. Purificación del AgsHB-r por cromatografía de inmunoafinidad. Estudio de los AcMs CB-Hep.1 y CB-Hep.3. Biotecnología Aplicada 1994;11:175-179.

13. Normas de proceso para la producción del UMELISA HBsAg. Centro de Inmunoensayo, La Habana, 1997.

14. Voller A, Bidwell DE, Bartlett A. Enzyme immunoasays in diagnostic medicine. Bull World Health Organ 1976; 53:55-65.

15. Thielmann K. Principios de metodología en bioquímica clínica. ed. Organismos: Instituto cubano del libro, 1973.

16. Broughton PMG, Bergonzi C, Lindstedt G, Loeber JG, Malan PG. Guidelines for the evaluation of diagnostic kits. Part 2. General principles and outline procedures for the evaluation of kits for qualitative tests. European Commitee for Clinical Laboratory Standards, Vol 3, No. 3, 1986.

17. Esser P. Principles in adsorption to polystyrene. Nunc Bulletin 1988;6:1-5.

18. Back JF, Oaekenfull D, Smith MB. Increase thermal stability of protein in the presence of sugar and poliols. Biochemistry 1979;18:5191-5196.

19. Franks F, Hatley RH, Mathias Sh. Materials science and the production of shelf-stable biologicals. Biopharm 1991;4:38-42.

20. Moreira T, Gutiérrez A, Delgado H. Aspectos físicos relacionados con los aditivos en el proceso de liofilización. Papel relevante de los carbohidratos. Biotecnología Aplicada 1994;11:113-119.

21. Marcel PW. Evaluation of the physical stability of freeze-dried-sucrose-containing formulations by differential scanning calorimetry. Pharmaceutical Research 1992;9:224-226.

22. Pujol V, Sotolongo J, Vega M, Pineda J, Costa L. Influencia de; uso de aditivos en la liofilización del antígeno de superficie recombinante del virus de la hepatitis B. Biotecnologóa Aplicada 1996;13:12-15.

23. Vega M, Barbería D, Costa L, Currás T, Pineda J, Agraz A, et al. Estabilidad del antígeno de superficie de la bepatitis B recombinante en diferentes condiciones de almacenamiento. Biotecnología Aplicada 1994;11:255-261.

24. Kramer S, Jewell N, Cremer N. Discriminant Analysis of Data in Enzyme Immunoassay. Journal of Immunological Methods 1983;60:243-255.

Copyright 1998 Elfos Scientiae

The following images related to this document are available:

Line drawing images