IVD Technology Magazine
IVDT Article Index

Originally published July, 1997

Blocking strategies for nylon membranes used in enzyme-linked immunosorbent assays

Andrew Dubitsky

A simple dot blot test can help ELISA developers optimize their use of nylon substrates and blocking agents.

In the development of membrane-based ELISA systems, sensitivity, background, and signal strength are primary concerns. These parameters are affected by the developers' choice of membrane, blocking strategy, concentration of antibodies, enzyme, and substrate. All these factors are interrelated. To determine the effects of membrane chemistry and blocking strategies, assay developers routinely turn to the manufacturers of materials and reagents.

Nylon membranes are often used for ELISA applications, and are available with a variety of surface chemistries. As when choosing surface chemistry, careful attention to blocking strategy is required to attain the best results. Dot blot experiments, in which multiple variables are manipulated in a simple matrix format, can quickly guide developers to the best combination of surface conditions, blocking concentrations, and reagents. This article describes a model assay system and experiments that can be used to optimize membrane choice and blocking.

Background

Membrane-based ELISA tests are commonly configured as flow-through devices in which a membrane is compressed against an absorbent pad and solutions are delivered to the top face of the membrane.¹ The primary antibody is bound to the surface of the membrane, which is then blocked to prevent nonspecific binding. The analyte is presented in a test solution, which can be either a buffer or a body fluid. The secondary antibody, conjugated to an enzyme, can be added either directly to the test solution or in a separate step after the analyte. To generate signal, membranes are washed and incubated with color-forming substrates.

Many ELISA tests use nylon membranes as support matrices.^2­4 In addition to their amphoteric (unmodified) form, nylon membranes are available in forms with a variety of surface chemistries, including those with amino and carboxyl groups, quaternary ammonium groups, carboxyl and hydroxyl groups, and active groups for covalent binding. A developer's choice of membrane should be determined empirically, according to how well it interacts with all the other test reagents.

To obtain good test results, it is critical to block any nonspecific binding sites available on the membrane.^5­8 Using the correct type and amount of blocking agent is important; insufficient blocking may result in a high background, while overblocking may decrease sensitivity. The blocking agents commonly used for nitrocellulose or polystyrene microwell plates, such as bovine serum albumin (BSA) and fish gelatin, are not effective for blocking nylon membranes. Although dry milk formulations are popular choices for use with nylon membranes, the most effective blocking formulation for all membrane types is a solution of Hammersten-grade casein, used at 0.05 to 0.5% in buffer, together with a low concentration of surfactant.

The balance of primary and secondary antibody concentrations can also affect the sensitivity and background of an assay.^9­12 For instance, increasing the concentration of the secondary antibody enzyme in order to attain higher sensitivity can also lead to higher background. System performance will also be affected by the developer's choice of an enzyme substrate.¹³ A substrate with a fast development time may result in higher background levels than one with a slow development time.

All these factors must be taken into account when developing a new assay, and any assay system always requires a balance between sensitivity and background (selectivity). To optimize the performance of an assay, variables can be matrixed in simple dot blot experiments. Only a few experiments are required to determine the best possible conditions for each step in the process.

Methods

Model ELISA System. A model sandwich ELISA system was designed using goat antirabbit IgG as the primary antibody (R2004, Sigma Chemical Co., St. Louis), rabbit antigoat IgG as the antigen (G4018, Sigma), and goat antirabbit IgG conjugated to peroxidase (A0545, Sigma) as the secondary antibody. The substrate solution consisted of 0.1 mg/ml diaminobenzidene, 1 mg/ ml imidazole, and 0.03% H^₂O^₂ (all from Sigma). Membranes were supplied by Pall Corp. (East Hills, NY). Blocking agents included BSA, gelatin (Sigma), and Hammersten-grade casein (Gallard Schlessinger, Westbury, NY).

Using a micropipettor, dilutions of the primary antibody from 1 to 100 ng were applied to membrane cards as 1-µl spots. Membranes were air dried for 5 minutes and blocked with one of the following:

• 5% BSA/PBS.

• 5% gelatin/PBS.

• 0.1% Hammersten-grade casein, 0.05% Tween 20.

• 0.5% Hammersten-grade casein, 0.05% Tween 20.

Membranes were soaked in blocking solution for 30 minutes at room temperature. After blocking, membranes were incubated with 100 ng/ml antigen for 10 minutes. Membranes were washed briefly with blocking solution and then incubated with a 1/500 dilution of conjugate for 10 minutes. Membranes were washed three times, 2 minutes per wash, with 0.1% Tween 20/water and then incubated with substrate solution for 2 minutes. Membranes were rinsed in water and air dried.

Immune Detection of Digoxigenin-Labeled DNA Probe. Dilutions of Lambda Hind III DNA (containing 1 ng to 1 pg/µl) were spotted onto nylon membranes. The spots were air dried for 15 minutes. For the Biodyne B membrane, DNA was fixed by baking; for all the other membranes, DNA was fixed through exposure to ultraviolet light (3 minutes with a handheld light source, Mineralight R52G, 6 in.). DNA on the membranes was detected by the use of a digoxigenin-labeled probe and antidigoxigenin antibody conjugated to alkaline phosphatase, with BCIP/NBT substrate (Genius kit, Boehringer Mannheim, Arlington, IL). Varied blocking agents (no block, BSA, Boehringer Mannheim blocking agent, or Hammersten-grade casein) were added at the prehybridization, hybridization, and blocking steps. Membranes were processed according to kit instructions.

Test Membranes

The nylon membranes used for these tests are representative of the alternatives available to the assay developer. Amphoteric (neutral at pH 7), positive-, and negative-charged membranes were included, along with a membrane that forms covalent bonds with proteins, and a low-protein binding membrane with hydroxyl surface chemistry. A description of each is shown in Table I.

Table I. Test membranes used to evaluate blocking strategies for ELISA.

Results

Model ELISA Experiments. Blocking agents commonly used for microplate assays (BSA and gelatin) are not effective on most nylon membranes. Hammersten-grade casein was the only blocking agent tested that resulted in clear signal and background on all membranes (see Figure 1). When the casein concentration was raised to 0.5%, good signal-to-noise ratios were observed on all membrane types (see Figure 2).

Figure 1. Sandwich ELISA test. Membranes were spotted with primary antibody at 0.1, 0.03, 0.001, 0.003, and 0.0001 µg per 1-µl spot. Membranes were blocked with either BSA, gelatin, or Hammersten-grade casein. Signal was developed with peroxidase conjugate and diaminobenzidene substrate.

Figure 2. ELISA membrane comparison. Dilutions of primary antibody were spotted onto cards made from different membrane types. Membranes were blocked with 0.5% casein, 0.1% Tween 20; 250-ng/ml analyte was used on all cards. Signal was developed with peroxidase conjugate and diaminobenzidene substrate.

DNA Dot Blot with ELISA Visualization. Figure 3 shows the differences in membrane performance that resulted as blocking agents were varied. As in the model ELISA test, the Biodyne B membrane required more stringent blocking to give a clear result than did the other membrane types. However, it also offered the highest sensitivity when blocking was optimized using 0.5% Hammersten-grade casein.

Figure 3: DNA dot blot with ELISA detection. Viral DNA immobilized on membranes was hybridized with digoxigenin-labeled probe. Hybridized probe was visualized with anti-DIG antibody conjugated to alkine phosphatase and BIC/NBT substrate. Varied blocking reagents listed at bottom were added to buffers for prehybridization, hybridization, and blocking steps. (Casein = Hammersten-grade casein; BM = Boehringer Mannheim blocking agent; BSA = bovine serum albumin.)

Discussion

The cumulative influence of all the reagents was shown most clearly in the ELISA detection of DNA dot blots. In this example, the positive surface charge on the Biodyne Plus membrane gave it an advantage in sensitivity over the amphoteric Biodyne A membrane, even though the blocking requirements for both were similar.

Hydroxylated nylon membrane (LoProdyne LP) binds very low amounts of protein. Low protein binding can be used to advantage in an ELISA: the drying of a protein onto the membrane forces the protein to bind more effectively (see Figure 2). However, compared to the other nylon membranes examined here, a higher concentration of the primary antibody was required to produce a positive result. A low protein-binding membrane can also be used with lower levels of blocking agents than other membranes (see Figure 1).

In these model tests, developed spots were larger on the hydroxylated nylon membrane (LoProdyne LP) than on the other substrates. Because the protein-binding properties of this material are low, the protein migrates with the solvent front. On membranes with higher protein-binding properties, spots are smaller because the protein is adsorbed on contact with the membrane, while the buffer wicks out to fill available pores (as in chromatography systems).

Spot size and definition can be controlled on all membrane types by varying the application method. For instance, homogeneous spots of equal size can be generated on all membrane types if the proteins are loaded in a volume greater than 10 µl using a 96-well vacuum manifold or similar device.

Immunodyne ABC membrane spontaneously forms covalent bonds with proteins. Since the membrane reacts primarily with amino groups, it is possible to obtain preferential orientation of molecules on the membrane surface. Covalent binding also allows the use of stringent wash conditions (such as high concentrations of surfactant) during the test.

Conclusion

The experiments conducted for this study demonstrate that Hammersten-grade casein (or equivalent) is the best blocking agent for all membrane types. Other casein brands may not work as well as Hammersten grade. To increase the effectiveness of the blocking solution, researchers should add 0.1% Triton X-100 or 0.05% Tween 20 to the solution. This technique will also speed rehydration of the membrane if it is to be dried at any time between blocking and signal development.

The test results also provide a predictor for membrane choice. The best starting point for an ELISA should be either a neutral or a covalently binding membrane. Both feature good combinations of sensitivity and background. Positively charged membranes may offer higher sensitivity, but they are more difficult to block effectively. Low protein-binding membranes can be used by forcing immobilization of antibody through drying in place (higher concentrations of antibodies may be required, but blocking requirements are minimal).

ELISA applications are complex systems in which all the steps and reagents interact. Obtaining best results requires a program to evaluate the following parameters:

• Membrane choice.

• Concentration of immobilized protein.

• Method for immobilization of protein.

• Drying conditions (if any) after immobilization.

• Blocking conditions (such as time, concentration).

• Conjugate concentration.

• Substrate choice (different substrates will also contribute their own effects to signal sensitivity and background).

Nylon membranes are widely used for dot ELISA tests. A model assay as described in this article can be used to optimize test conditions. Membrane choice and other assay parameters can be matrixed together in dot blot experiments to quickly determine the best combinations of membrane, blocking agent, antibodies, and substrate.

References

1. Christensen H, Thyssen HH, Schebye O, et al., "Three Highly Sensitive 'Bedside' Serum and Urine Tests for Pregnancy Compared," Clin Chem, 36(9):1686­1688, 1990.

2. Ali A, and Ali R, "Anti-DNA Antibodies in Autoimmune Disorders by ELISA Using Nylon as the Solid Phase," Clin Biochem, 19(4):205­208, 1986.

3. Jung R, and Terplan G, "Strip ELISA for Detection of Staphylococcal Enterotoxins in Culture Supernatants and Foods," Food Agric Immunol, 5(2):107­114, 1993.

4. Singh RP, Boucher A, Somervill TH, et al., "Selection of a Monoclonal Antibody to Detect PVY-N and Its Use in ELISA and DIBA Assays," Canadian J Plant Path, 15(4):293­300, 1993.

5. Kenna JG, Major GN, and Williams RS, "Methods for Reducing Non-Specific Antibody Binding in Enzyme Linked Immunosorbent Assays," J Immunol Methods, 85:409­419, 1985.

6. Pruslin FH, To SE, Winston R, et al., "Caveats and Suggestions for ELISA," J Immunol Methods, 137(1):27­35, 1991.

7. Spinola S, and Cannon J, "Different Blocking Agents Cause Variation in the Immunologic Detection of Proteins Transferred to Nitrocellulose Membranes," J Immunol Methods, 81:161­165, 1985.

8. Vogt RF, Phillips DL, Henderson LO, et al., "Quantitative Differences among Various Proteins as Blocking Agents for ELISA Microtiter Plates," J Immunol Methods, 101(1):43­50, 1987.

9. Doumit ME, Lonergan SM, Arbona JR, et al., "Development of an Enzyme Linked Immunosorbent Assay for Quantification of Skeletal Muscle Calpastatin," J Animal Science, 74(11):2679­ 2686, 1996.

10. Mohammed HO, Yamamoto R, Carpenter RE, et al., "A Statistical Model to Optimize Enzyme Linked Immunosorbent Assay Parameters for Detection of Mycoplasma gallisepticum and M. synoviae Antibodies in Egg Yolk," Avian Diseases, 30(2):389­397, 1986.

11. Shinagawa K, Suzuki M, Matsusaka N, et al., "Detection of Staphylococcal Enterotoxin A by Sandwich Enzyme Linked Immunosorbent Assay with Monoclonal Antibodies," J Vet Med Sci, 53(6):1093­1095, 1991.

12. Wang KC, and Leung BS, "Fluorometric ELISA Method for Rapid Screening of Anti-Estrogen Receptor Antibody Production in Hybridoma Cultures," J Immunol Methods, 84(1­2): 279­290, 1985.

13. Gabridge MG, Lundin DJ, and Gladd MF, "Detection and Speciation of Common Cell Culture Mycoplasmas by an Enzyme Linked Immunosorbent Assay with Biotin-Avidin Amplification and Microporous Membrane Solid Phase," In Vitro Cell Devel Bio, 22(8):491­498, 1986.

Andrew Dubitsky is senior staff scientist in the scientific and laboratory services department at Pall Corp. (Port Washington, NY).

Return to the IVDT home page