WO1987002619A1

WO1987002619A1 - Assay supports

Info

Publication number: WO1987002619A1
Application number: PCT/AU1986/000329
Authority: WO
Inventors: Marvin Alfred Akerman; Vincent Habib Atrache; Richard Peter Gregson; David Ira Klick; George Lange Paul
Original assignee: Biotechnology Australia Pty. Ltd.; The University Of New South Wales
Priority date: 1985-11-01
Filing date: 1986-10-31
Publication date: 1987-05-07
Also published as: NZ218128A; EP0245310A1

Abstract

Substrates such as microtitre plates, microtitre wells, dipsticks, heads, cuvettes, test tubes and the like which have been treated with abrasion, laser light or high voltage to improve binding thereto of proteins or other organic molecules.

Description

ASSAY SUPPORTS TECHNICAL FIELD

The present invention relates to improved assay supports, particularly microtitre plates or trays.

BACKGROUND ART

Enzyme linked immunosorbent assays, ELISAs, were introduced in the early 1970's. These assays are now firmly established as precise quantitative methods for the determination of antibodies and antigens. An increasing number of commercial diagnostic procedures are based on ELISAs and biological parameters traditionally measured by radioimmunoassays (RIA) are gradually being replaced by ELISAs. Applications for ELISAs have been reviewed (1), (2) and include detection of herpes simplex virus, rotavirus, reovirus and virus diseases of trees and plants. Perhaps the main impact of the ELISA has been in the quantitation of antibodies.

Prior to ELISA, antibody titres were measured by qualitative procedures such as haemaggluti nation, gel diffusion (Ochterlony), immunofluorescence and to a lesser extent, RIA. A commercial kit, Rubelisa, (Microbiological Associates) reliably measures antibodies to rubella (German measles). In Australia commercial ELISA' s have been developed for brucellosis eradication (Australian Monoclonal Antibody Developments), meat speciation (Victorian Department of Agriculture; differentiates between beef, horse, pig, sheep, kangaroo, buffalo, goat and donkey) heartworm (MAGCO) and a snake venom detection kit (CSL).

An initial and crucial step in the ELISA protocol is the hydrophobic adsorption of the antigen or antibody to a substrate such as microtitre trays, dipsticks or beads. The whole performance and especially the sensitivity and reproducibility of the ELISA are very dependent on this initial binding step.

The repeated washings during the assay with detergent cause leaching of the bound antigen or antibody. The most commonly adopted protocol for binding involves incubating a solution (1μg.ml^-1) of the protein at pH9.6, 37°C, 16 hours in wells of a polystyrene microtitre tray.

A variety of modifications to this binding procedure have been attempted with the aim of increasing the amount of bound antigen or antibody and, decreasing the non-specific binding and variability between trays or other substrates. These attempts have all been directed to derivatising the surface of the support such as a microtitre tray by chemical means. Amino groups hare been introduced onto the surface of polystyrene trays by nitration with nitric acid followed by reduction with sodium dithionite or iron salts. A 32-fold increase in sensitivity was claimed for amino-trays over normal polystyrene trays (3). Antigens have been cross-linked (4) and antibodies denatured (5) to enhance the hydrophobic binding to the polymer. Amino-Dylark, a new solid support, was used to bind antibodies to an ELISA (6) and this support showed some advantages. γ-irradiated polystyrene microtitre trays have a coefficient of variation which is half that of non-irradiated trays (7). The problem of reproducibility within a tray, between trays of the same production batch and between batches is so great that some practitioners advocate screening manufacturers' batches prior to purchase (8). The adsorptitive capacity is a major, limiting factor in ELISAs (9). Antigens often exhibit poor binding to the surface. Some of these attempts have been successful and some have numerous drawbacks including costs and storage capability of the tray.

DISCLOSURE OF THE INVENTION

The present invention results from the discovery that physical abrasion of the internal surface of the walls of the wells can lead to an increased sensitivity of the immunoassay as well as increased reproducibility.

The invention embodies the improvement of several performance characteristics of substrates when used in ELISAs after abrasion, exposure to light emanating from a laser or exposure to high voltage.

In the first form, the invention provides a substrate having improved binding capacity to protein or other organic molecules characterised in that the substrate has been abraded, exposed to light emanating from a laser or to high voltage. The improved binding capacity of substrates afforded by the invention finds application in improving the performance of ELISAs carried out in microtitre trays, microtitre wells, dipsticks, beads, cuvettes, test-tubes and the like. The substrates find application in other methodologies involving protein binding such as affinity chromotography and medical prostheses. Suitable materials include polystyrene, PVC and other substances to which proteins are known to bind.

In a second embodiment, the invention provides a method of increasing the protein or other organic molecules binding capacity of a substrate which method comprises abrading the substrate, exposing the substrate to light emanating from a laser or to high voltage. The wavelength of laser light chosen should be a wavelength capable of being absorbed by the substrate. It is most preferred to employ a wavelength which approximates the peak absorbance for the particular substrate. The material most frequently used in ELISAs is polystyrene. The far ultra-violet wavelengths have been found most suitable for improving the protein binding capacity of polystyrene. In particular, wavelengths of 193nm, 308nm and especially 248nm have been found satisfactory. When the high voltage alternative embodiment of the invention is employed, an electric field strength greater than 1000V.cm^-1 is preferred.

In a further form of the invention, the surface of the support may be treated further with an agent to increase the binding of biologically active molecules thereto, such as glutaraldehyde or other low molecular weight aldehydes and polymers thereof after abrasion or exposure to laser light or high voltage.

Abrasion, in the content of this invention includes roughening, etching and forming a roughened surface by moulding.

Whilst embodiments of the invention are described with reference to microtitre trays, the invention should not be construed as being limited thereto.

Polystyrene microtitre trays treated according to the invention are capable of binding at least twice as much, and commonly four times as much antibody or antigen as untreated trays. They bind as much antibody or antigen in five minutes as untreated trays do in three hours. The range for which the response in the ELISA is linear is extended compared to untreated trays. The reproducibility within a tray is improved compared to an untreated tray. ELISAs can be performed on samples which could not be performed on untreated trays due to the increased sensitivity of trays treated according to the invention.

The laser has been employed in two operating modes. Firstly, there is the ablative mode, which causes photochemical change to the surface of the microtitre tray by high energy irridiation of individual wells. The majority of the energy is deposited between about 5ps and about 1ms depending on wavelength. Secondly, the "low fluence mode" is used which involves irradiation of the wells of the ELISA plate with low fluence light i.e. low energy/surface area. Any wavelength of light absorbed by the well surface may be used. The optimum wavelength for this process approximates the peak absorbance of the material. Fluences are usually above about 100mJ.cm^-2 in the ablative mode while low fluence usually includes all intensities below about 200mJ.cm^-2. The fluence limit depends on wavelength. The low fluence light may be applied from a pulsed or continuous source. Antibodies and antigens will bind to ablated substrates fn tte presence of detergents. This is not possible with prior-art substrates.

The high voltage technique generally employs an electric field strength greater than 1000V.cm^-1 in the neighbourhood of a well which causes a permanent chemical modification to the well surface. A bluish-purple glow occurs near the surface when exposed to the high voltage which concomitantly exposes the surface to ozone and ultra-violet radiation. A metal plate under the tray or other substrate during high voltage treatment appears to be beneficial.

BRIEF DESCRIPTION OF THE DRAWINGS

Figures 1 to 9 illustate graphically the performance of the examples herein.

Figures 10a and 10b are a representation of treatment of microtitre trays according to the invention.

Figures 11 and 11a illustrate the ablation technique.

Figures 12a and 12b illustrate the low fluence technique.

Figures 13a and 13b illustrate the high voltage technique. BEST MODES FOR CARRYING OUT THE INVENTION

Notwithstanding other support materials which fall within the broad form of the present invention, the invention will now be described with specific reference to microtitre trays.

Four different techniques for assessing the treated trays have been employed. Method 1 : Analysis of Conjugate Binding

A microtitre tray was coated with a conjugate of donkey anti-rabbit IgG linked to horseradish peroxidase (Amersham), serially diluted (100μl/well) with phosphate buffer (0.1M, pH7). The tray was covered then stored at 37ºC for 1.5 hours.

The conjugate was emptied out of the tray, then it was washed five times with phosphate buffer (0.05M, pH7) containing sodium chloride (0.17M), and Tween (0.05%) (PBS/Tween).

A solution of 2,2'-azinodi-(3-ethylbenzthiazolinesulfonic acid) (Sigma) (1mg.ml^-1) in 0.1M citrate-phosphate buffer pH4 containing .001% hydrogen peroxide was then added to the tray (100μl/well) and colour allowed to develop for approximately 4 minutes.

Colour development was stopped by the addition of 1% sodium azide (10μl/well) then the absorbance of each well read using a Titertek Multiskan Spectrophotometer at 414nm and 492nm. Method 2: Half-Sandwich Elisa

A microtitre tray was coated with rabbit anti K99 IgG (100μl/well) diluted with carbonate buffer (0.1M, pH9.6) for normal, untreated wells, and phosphate buffer (0.1M, pH8) for laser-treated wells. K99 is a fimbrial antigen derived from pathogenic E . coli. Typically, a concentration range of 5ng.ml^-1 to 10μg.ml^-1 was used. The tray was covered and allowed to stand overnight at 37°C.

The IgG solutions were emptied out of the tray, then it was washed five times with PBS/Tween.

Conjugate (donkey anti-rabbit IgG linked to horseradish peroxidase [Amersham]) was diluted with PBS/Tween (1 in 1000) and then added to the plate 100μl/well). The plate was covered and allowed to stand at 37°C for 1.5 hours.

A solution of 2,2'-azinodi-(3-ethylbenzthiazol inesulfonic acid) (Sigma) (1mg.ml^-1) in 0.1M citrate-phosphate buffer pH4 containing .001% hydrogen peroxide was then added to the tray (100μl/well) and colour allowed to develop for approximately 4 minutes.

Colour development was stopped by the addition of 1% sodium azide (10μl/well) then the absorbance of each well read using a Titertek Multiskan Spectrophotometer at 414nm and 492nm. Method 3: Double Sandwich For K99

A microtitre tray was coated with rabbit anti-K99 IgG (0.5μg/ml, 100μl/well) in carbonate buffer (0.1M, pH9.6) for untreated wells, and in phosphate buffer (0.1M, pH8) for irradiated wells. The tray was covered, then allowed to stand overnight at 37°C.

The tray was coated with K99 antigen serially diluted with PBS/Tween (2ng/ml^-1 to 250ng/ml^-1; 100μ/well) then covered and allowed to stand at 37°C for 1.5 hours.

The antigen solutions were emptied out of the tray, then it was washed five times with PBS/Tween.

The tray was coated with conjugate (rabbit anti-K99 IgG linked to horseradish peroxidase) diluted with PBS/Tween (1 in 800; 100μl/well). The plate was covered then allowed to stand for 1 hour at 37°C.

The conjugate was emptied out of the tray, then it was washed five times with phosphate buffer (0.05M, pH7) containing sodium chloride (0-.17M), and Tween (0.05%) (PBS/Tween).

Colour development was stopped by the addition of 1% sodium azide (10μl/well) then the absorbance of each well read using a Titertek Multiskan Spectrophotometer at 414nm and 492nm. Method 4: Double Sandwich For LTB

A microtitre tray was coated with LTB (100μl/well) diluted with carbonate buffer (0.1M, pH9.6) for normal, untreated wells, and with phosphate buffer (0.2M, pH8) for irradiated wells. LTB is the B-subunit of the heat-labile toxin of an enterotoxigenic E. coli. The tray was covered then stored at 37°C overnight.

The LTB was emptied out of the tray, then it was washed five times with PBS/Tween.

The tray was coated with rabbit anti-LTB serum (100μl/well) serially diluted with PBS/Tween. The tray was covered then allowed to stand at 37°C for 1.5 hours.

The serum solutions were emptied out of the tray, then washed five times with PBS/Tween.

The tray was then coated with conjugate (donkey anti-rabbit IgG linked to horseradish peroxidase [Sigma]) (100μl/well) diluted with PBS/Tween (1 in 1000). The plate was covered then stored at 37°C for 1.5 hours.

Colour development was stopped by the addition of 1% sodium azide (10μl/well) then the absorbance of each well read using a Titertek Multiskan Spectrophotometer at 414nm and 492nm.

The following examples illustrate the improvements in bidding properties of proteins to microtitre trays and in performance characteristics of the ELISA achievable by preferred embodiments of the invention.

EXAMPLE 1

A typical 96 well polystyrene tray was used. The internal wall surfaces of each well were mechanically abraded by a wire brush rotated within the well to mechanically abrade the walls.

Varying concentrations of rabbit antibodies, IgG fraction, were incubated at 37º for 4 hours in

(a) untreated polystyrene wells (U),

(b) wells pretreated with IX glutaraldehyde solution (G),

(c) wells abraded then treated with IX glutaraldehyde solution, (S + G)

(d) well abraded by the above described method (A).

The amount of antibody adsorbed to the wells was detected using a peroxidase-labelled antirabbit IgG and 2,2'-azinodi-(3-ethy1benzthiazol inesulfonic acid) as substrate. Absorbance values quoted below are the average of triplicates.

The above results are graphed in Figure 1.

Whilst there was no significant increase in the control results, indicating no significant increase in non-specific binding with the wells of the present invention, a steep linear increase was observed in absorbance which correlated excellently with increased coating concentrations in the S+G wells. In the untreated wells (U), absorbance values levelled off at 5g.ml^-1 coating concentration. A further decrease in absorbance at 10g.ml^-1 in the untreated wells signifies the hook effect.

Whilst glutaraldehyde is used to treat the surface of the well, it is likely that the polyglutaraldehyde in the glutaraldehyde solution contributes to the binding capacity of the treated wells.

EXAMPLE 2 Columns 7 to 12 of microtitre trays were irradiated with 1000 pulses of laser light at a wavelength of 248nm, an energy of 83.1mJ. a spot size of

60cm², a fluence of 1.4mJ.cm^-2 and a repetition rate of 25Hz. The total energy delivered to each well was 4.60.

The trays were evaluated as in Method 1 above except that the conjugate was diluted with the buffers described in Table 2.

Figure 2 shows the effect of pH on the conjugate binding capacity of irradiated and untreated wells. It is clear that the binding to irradiated wells is sensitive to pH and that the optimum pH for binding is 8. Untreated wells are relatively insensitive but show a maximum at pH8 to pH9.6. Most protocols require coating at pH9.6.

EXAMPLE 3

Columns 1, 2, 5, 6, 9 and 10 of microtitre trays were irradiated with laser light at a wavelength of 248nm, an energy of 54mJ, a spot size of

0.12cm² and a fluence of 450mJ.cm². All irradiated wells showed some blackness. Columns 1 and 2 were irradiated at 10 pulses per location and repetition rate of 1hz. Columns 5 and 6 were irradiated at 10 pulses per location and a repetition rate of 25Hz. Columns 9 and 10 were irradiated at

20 pulses per location and a repetition rate of 25Hz. Columns 3, 4, 7, 8,

11, and 12 were untreated.

The microtitre trays were evaluated according to Method 1 above.

Figure 3 shows a comparison between the binding of conjugate to an ablated and an untreated well on the same plate. The conjugate was serially diluted froa 1 in 500 to 1 in 100000. At a dilution of 1 in 500 the ablated plate binds 400X more conjugate than the untreated wells. EXAMPLE 4

Microtitre trays were irradiated by the same method as Example 3. Columns 9, 10, 11 and 12 were then evaluated as follows.

Duplicate columns of irradiated and untreated wells were incubated with bovine serum albumin (BSA) in phosphate buffer which was serially diluted along the column. Then conjugate (1 in 500) was added according to ELISA Method 2. As in Example 3, Figure 4 shows that the ablated plate is binding approximately 400% more protein than the untreated plate at low concentrations of BSA but the effect decreases linearly with increasing BSA concentrations once an apparent threshold concentration of Iμg.ml^-1 is reached. The ablated wells bind more BSA as indicated by the steep, negative slope at BSA concentrations above Iμg.ml^-1.

EXAMPLE 5

Columns 1 to 6 of microtitre trays were irradiated with 1000 pulses of laser light at a wavelength of 248nm, an energy of 179mJ, a spot size of

12.6cm², a fluence of 14.2mJ.cm^-2 and a repetition rate of 25Hz.

Irradiated columns 1 to 5 appeared slightly yellow. The total energy delivered to each well was 4.69J. Columns 7 to 12 of microtitre trays were untreated. Columns 6 and 7 were evaluated according to Method 1 above.

A low fluence, irradiated half plate was compared with the untreated other half by ELISA Method 2. Figure 5 shows that the irradiated half binds 400X more conjugate at dilutions of 1 in 100 and 1 in 250 and that the response of the irradiated half is much more linear than the untreated half.

EXAMPLE 6

Columns 7 to 12 of microtitre trays were irradiated with 1000 pulses of laser light at a wavelength of 248nm, an energy of 336mJ, a spot size of

18cm², a fluence of 18.5mJ.cm^-2 and a repetition rate of 25Hz. The total energy delivered to each well was 6.1J. Columns 1 to 6 were untreated.

Low fluence irradiated wells were compared with untreated wells by a double sandwich assay according to Method 3 above, on two different commercially available trays. Figure 6 depicts results from a K-Line plate manufactured by Bunzyl and Figure 7 illustrates results from a

Linbro/Titertek plate supplied by Flow Laboratories. Figures 6 and 7 show that the coating antibody concentration of 0.1μg.ml^-1 was inadequate for an effective ELISA on untreated wells but in both trays the irradiated wells gave a measurable response. EXAMPLE 7 Columns 7 to 12 of microtitre trays were irradiated with 10000 pulses of laser light at a wavelength of 248nm, an energy of 53.6m0, a spot size of

37cm², a fluence of 1.45 mJ.cm^-2 and a repetition rate of 25Hz. The total energy delivered to each well was 4.79J. Columns 1 to 6 were untreated. The trays were evaluated according to Method 4 above.

Figure 8 shows a comparison between a low fluence half plate and the untreated half for a full, double sandwich ELISA as in method 4 above. In this case, antigen, not antibody, was adsorbed to the plate. The results in Figure 8 show that the treated plate showed a much higher binding capacity for the antigen at both concentrations and that the irradiated plate reflected the titration of the anti serum. The untreated plate was ineffective.

EXAMPLE 8

Columns 7 to 12 of microtitre trays were irradiated with 5000 pulses of laser light at a wavelength of 248nm, an energy of 46mJ, a spot size of

54cm² , a fluence of 0.85mJ.cm^-2 and a repetition rate of 25Hz. The total energy delivered to each well was 1.40. Columns 1 to 6 were untreated. Columns 1, 2, 11 and 12 were evaluated according to Method 2 above.

Figure 9 shows the comparison between a low fluence half plate and an untreated plate for a half sandwich according to method 2 above. Rabbit anti K99 IgG was adsorbed to the plate and then conjugate was added. Figure 8 shows that the laser treated wells exhibit a higher sensitivity.

Figures 10a and 10b schematically illustrate a processing plant for manufacture of microtitre trays having the improved characteristics of the invention. Microtitre trays 1, are supplied from a feed system, 2, to a conveyor belt, 3, travelling in the direction of arrows, A. The conveyor belt, 3, moves the microtitre trays, 1, passed a treatment zone, 4, and onto a stack, 5.

In the treatment zone, 4, a sensor, 6, detects the presence of microtitre trays, 1, at the approprite position and enables the conveyor belt, 3, to be stopped whilst a tray, 1, is aligned in a treatment area, 7, for treatment by laser or high voltage generator, 8.

The laser ablation technique is illustrated schematically in Figures 11a and 11b. The ablation technique requires a tightly focussed beam in each well of the microtitre tray. A broad beam, 9, of laser light, preferably of wavelength 248nm, is split by miirors, 10, to narrow beams, 11, which are focussed into the wells, 12, of a microtitre tray, 1, by cylindrical lens, 11. The conveyor belt, 3, is stopped by sensor, 6, (shown in Figure 10b) to permit exposure of successive rows of wells, 12.

The low fluence technique, illustrated in Figures 12a and 12b requires an expanded beam to cover an entire microtitre tray. A beam of laser light, 9, is reflected by convex mirror 14, to form a diverging beam, 15, which intercepts the whole surface area of microtitre tray 1 on conveyor belt 3. The sensor (6 in Figure 10b) enables the conveyor belt 3 to be stopped at the appropriate position such that divergent beam 15 is able to irradiate the whole surface of the microtitre tray 1.

One laser which has been found suitable for the purposes of the present invention is a Lambda Physik EMG 150 ETS. This laser was employed at a repetition rate of 25Hz, average power of 5W and a pulse energy of 0.20J. It is a rare gas halide exciter laser. The gases employed were krypton and fluorine.

Figure 13a shows a high voltage probe 16 placed in proximity to a well 12 of a microtitre tray 1 which is supported on a metal plate 17. Discharge 18 eminates from the probe 16 and modifies the well 12 so as to improve its protein binding capacity.

Figure 13b illustrates a Tesla coil which is a preferred method of generating the high voltage suitable for treating microtitre trays as illustrated in Figure 12a. A high voltage 19, usually in the order of 10 to 30kV is applied across a condenser 20. The circuit includes a spark gap 21 and transformer 22 having a very short coil 23 wound with a long coil 24 associated with high voltage output 25.

INDUSTRIAL APPLICABILITY

The substrates of the invention find use in any application where proteins or other organic molecules need to be bound to a solid surface.

REFERENCES

1. Methods in Enzymology, Ed. H. V. Vunakis and J. J. Langone, Vol. 70, 417, 439 (1980);

2. Voller, A., Bartlett, A. and Bidwell, D. E., J. Clinical Pathology, 31, 507-520 (1978).

3. Neurath, A. R. and Strick, N., J. Virological Methods, 3, 155-165 (1981)

4. Rotmans, J. P. and Delwel, H. R., J. Immun. Methods, 3, 57 87 (1983).

5. Conradie, J. D., Govender, M. and Visser, L., J. Immunol. Methods, 59, 289-299 (1983).

6. Tanimori, H. et al., J. Immunol. Methods, 62, 123-131 (1983). 7. McCullough, K. C. and Parkinson, D., J. Biological Standardisation, 12, 75-86 (1984).

8. Skekarchi, I. C, Sever, J. L., Lee, Y. J., Castellano, G. and Madden, D. L. J. Clin. Micro, 19, (2), 89-97 (1984).

9. Earle, J. A. P. and Wisdom, G. 8., J. Immunol. Methods. 76, 198-200 (1985).

Claims

1. A substrate having improved binding capacity to protein or other organic molecules characterised in that the substrate has been abraded, exposed to light emanating from a laser or to high voltage, the wavelength of said light being absorbable by said substrate.

2. The substrate as defined in claim 1, which is a microtitre tray, a microtitre well, a dipstick, a bead, a cuvette or a test tube.

3. The substrate as defined in claim 2, manufactured from polystyrene, polyvinylchloride or another substance to which proteins are known to bind.

4. A method of increasing the protein or other organic molecules binding capacity of a substrate which method comprises abrading the substrate, exposing the substrate to light emanating from a laser or to high voltage, the wavelength of said light being absorbable by said substrate.

5. The method as defined in claim 4, wherein the wavelength approximates the peak absorbance of the substrate.

6. The method as defined in claim 4, wherein th substrate is polystyrene and the wavelength is in the far ultra-violet range.

7. The method as defined in claim 6, wherein the wavelength is 193nm, 248nm or 308nm.

8. The method as defined in claim 4, wherein laser light is employed at a fluence of at least about 100mJ.cm^-2 .

9. A method as defined in claim 8 wherein laser light irradiates the substrate for a time between about 5ps and about 1ms.

10. The method as defined in claim 4, wherein laser light is employed at a fluence of less than about 200m.J. cm^-2 from a pulsed or continuous source.

11. The method as defined in claim 4, wherein high voltage of field strength greater than 1000V.cm^-1 is applied to the substrate.

12. The method as defined in claim 11, wherein a metal plate is placed under the substrate during exposure to high voltage.

13. The method as defined in claim 4, further comprising treating the substrate produced by the method of claim 4 with an agent known to increase the binding of biologically active molecules thereto.

14. The method as defined in claim 13, wherein the agent is glutaraldehyde or other low molecular weight aldehydes or polymers thereof.