Sandwich ELISA Protocol: Step-by-Step Guide for Antigen Detection

By Dr. Zubair Khalid, DVM, MS, PhD · 2026-07-15

Key Takeaways

• The sandwich ELISA is a highly specific immunoassay for antigen detection and quantification, utilizing two distinct antibodies to bind non-overlapping epitopes on the target antigen.
• This method is particularly effective for analyzing complex biological matrices such as serum, food homogenates, and culture supernatants, offering reduced background noise and enhanced sensitivity compared to direct or indirect ELISA formats.
• Critical to assay performance is the selection of compatible antibody pairs (monoclonal-monoclonal, monoclonal-polyclonal, or nanobody-based) and the optimization of reagent concentrations, blocking, and washing steps to minimize non-specific binding.
• Robust validation requires establishing intra- and inter-assay precision, spike recovery, linearity, and a defined Limit of Detection (LOD), with qualitative results interpreted against a calculated cut-off value and quantitative results against a standard curve.
• Biosafety considerations range from BSL-1 for non-pathogenic targets to BSL-2 for clinical or potentially infectious samples, necessitating appropriate containment and decontamination procedures.

The sandwich enzyme-linked immunosorbent assay (sandwich ELISA) is a highly specific immunoassay method that quantifies or detects an antigen by "sandwiching" it between two layers of antibodies: a capture antibody immobilized on a solid surface and a detection antibody conjugated to an enzyme or other reporter molecule. This format is particularly useful when the target antigen is present in complex mixtures (e.g., serum, food homogenates, or culture supernatants) and requires high specificity, as the antigen must be recognized by two distinct antibodies. The sandwich ELISA is widely applied in clinical diagnostics, food safety testing, and research laboratories for detecting proteins, hormones, pathogens, and biomarkers. Unlike direct or indirect ELISA formats, the sandwich configuration reduces background signal and increases sensitivity by requiring dual recognition of the target.

At a Glance

Scientific Principle of Sandwich ELISA

The sandwich ELISA relies on the sequential binding of two antibodies to distinct epitopes on the same antigen molecule. The capture antibody is first adsorbed onto a high-binding microtiter plate surface through passive hydrophobic interactions. After blocking unoccupied binding sites, the sample containing the target antigen is added, allowing the antigen to bind specifically to the capture antibody. Following washing steps to remove unbound material, a detection antibody (directly conjugated to an enzyme such as horseradish peroxidase or alkaline phosphatase) is added, which binds to a different epitope on the captured antigen. The final step involves adding a chromogenic, fluorogenic, or chemiluminescent substrate that produces a measurable signal proportional to the amount of antigen present.

This dual-antibody recognition provides exceptional specificity because both antibodies must bind their respective epitopes for signal generation. As demonstrated in a study developing a sandwich ELISA for Talaromyces marneffei antigen detection, the assay using the same monoclonal antibody for both capture and detection achieved 88.61% sensitivity and 96.06% specificity in clinical serum samples [<a href="#ref-1">1</a>]. Similarly, a sandwich ELISA targeting the p60 antigen of Listeria monocytogenes using monoclonal capture and polyclonal detection antibodies showed robust analytical performance [<a href="#ref-2">2</a>].

Materials and Instrumentation

Antibody Selection and Pairing

The most critical decision in sandwich ELISA development is selecting compatible antibody pairs. The capture and detection antibodies must recognize non-overlapping epitopes on the target antigen. Options include:

• Monoclonal-monoclonal pairs: Provide highest specificity but require careful epitope mapping to avoid competition
• Monoclonal-polyclonal pairs: Often more robust, as polyclonal detection antibodies recognize multiple epitopes, reducing the risk of epitope masking
• Nanobody-based systems: Offer advantages for conformation-sensitive targets, as shown in a study differentiating live from heat-inactivated Salmonella Typhimurium using nanobodies that recognize heat-induced conformational changes [<a href="#ref-3">3</a>]

For research-grade assays, commercial antibody pairs are available pre-validated. For novel targets, hybridoma technology or recombinant antibody production may be necessary. The study on Listeria monocytogenes detection generated monoclonal antibodies from BALB/c mice immunized with recombinant p60 antigen, screening hybridoma clones by indirect ELISA and Western blot [<a href="#ref-2">2</a>].

Plate Selection

High-binding polystyrene microtiter plates (96-well or 384-well) are standard. Consider:

• Maxisorp plates (Thermo Scientific Nunc or equivalent): For proteins with mixed hydrophobic/hydrophilic regions
• Medisorp plates: For highly hydrophobic proteins
• Streptavidin-coated plates: For biotinylated capture antibodies, enabling oriented immobilization

Buffers and Reagents

Detection Systems

• HRP/TMB system: Most common; TMB substrate yields blue color, stopped to yellow; read at 450 nm
• HRP/ABTS system: Green color; read at 405–410 nm
• Alkaline phosphatase/pNPP system: Yellow color; read at 405 nm
• Chemiluminescence: Higher sensitivity; requires luminometer; used in a magnetic bead-based CLEIA for Aspergillus galactomannan detection achieving 30-minute analysis [<a href="#ref-4">4</a>]
• Streptavidin-PolyHRP amplification: Enhances signal; used in nanobody-based sandwich ELISA for Salmonella detection with detection limit of 3.53 × 10³ CFU/mL [<a href="#ref-3">3</a>]

Controls and Standards

Essential Controls

1. Blank wells: No antigen, no detection antibody (only substrate and stop solution)
2. Negative control: Known negative sample matrix (e.g., pooled normal serum)
3. Positive control: Known positive sample or purified antigen at known concentration
4. Non-specific binding control: Wells coated with capture antibody, no antigen, then detection antibody
5. Standard curve: Serial dilutions of purified antigen (typically 7–8 points in duplicate)

Standard Curve Preparation

Prepare the standard antigen in the same matrix as test samples (e.g., pooled human serum for clinical assays, PBS with BSA for purified systems). The study on Talaromyces marneffei sandwich ELISA established an analytical limit of detection (LOD) of 19.398 μg/mL in pooled human serum using serial dilutions of recombinant antigen [<a href="#ref-1">1</a>]. For food matrices, spike known concentrations into the appropriate food homogenate to account for matrix effects.

Conceptual Workflow

Step 1: Plate Coating

Dilute capture antibody to optimal concentration (typically 1–10 μg/mL) in coating buffer. Add 50–100 μL per well. Incubate overnight at 4°C or 1–2 hours at 37°C. The coating concentration should be optimized by checkerboard titration against a fixed antigen concentration.

Decision point: Overnight coating at 4°C generally yields more uniform adsorption and higher signal-to-noise ratios than short incubations at 37°C.

Step 2: Blocking

Wash plate 3× with PBST. Add 200–300 μL blocking buffer per well. Incubate 1 hour at 37°C or overnight at 4°C. Blocking saturates remaining protein-binding sites on the plastic surface, reducing non-specific binding of subsequent reagents.

Edge case: For assays using biotinylated detection antibodies, avoid blocking buffers containing free biotin (e.g., some milk preparations), which can compete with the detection system.

Step 3: Sample Addition

Wash plate 3× with PBST. Add samples, standards, and controls in duplicate or triplicate. Incubate 1–2 hours at 37°C or room temperature. Sample volume is typically 50–100 μL.

Documentation detail: Record sample dilution factors, incubation times, and temperatures. For clinical specimens, note any visible hemolysis, lipemia, or icterus that may interfere.

Step 4: Detection Antibody Addition

Wash plate 3–5× with PBST. Add detection antibody diluted in blocking buffer or antibody diluent. Incubate 1 hour at 37°C. For unconjugated detection antibodies, an additional step with enzyme-labeled secondary antibody is required.

Quality check: The detection antibody concentration should be optimized to maximize signal while minimizing background. A typical starting range is 0.1–2 μg/mL.

Step 5: Enzyme Conjugate (if needed)

If using an unconjugated detection antibody, wash and add enzyme-labeled secondary antibody (e.g., goat anti-mouse IgG-HRP). Incubate 30–60 minutes at 37°C. Wash thoroughly (5–7×) to remove unbound conjugate.

Step 6: Substrate Addition and Signal Development

Wash plate 5–7× with PBST. Add substrate (e.g., TMB for HRP). Incubate in the dark for 15–30 minutes at room temperature. Monitor color development visually or by plate reader.

Stop point: Add stop solution when the highest standard reaches optimal color intensity (typically OD 1.5–2.0). Read absorbance within 30 minutes of stopping.

Step 7: Data Acquisition

Read absorbance at appropriate wavelength (450 nm with 570–650 nm reference for TMB). For chemiluminescence, read relative light units (RLU) immediately after substrate addition.

Quality Checks and Validation

Intra-assay Precision

Calculate coefficient of variation (CV) for replicates. Acceptable CV is typically <10% for high-concentration samples and <20% for low-concentration samples near the LOD.

Inter-assay Precision

Run the same control samples across multiple plates on different days. CV should be <15–20% for quantitative assays.

Spike Recovery

Add known amounts of purified antigen to test matrix and measure recovery. Acceptable recovery is 80–120%.

Linearity

Serially dilute a high-concentration sample and measure. The measured concentration should be proportional to the dilution factor (within 80–120% of expected).

Limit of Detection (LOD)

Calculate as mean blank signal + 3 standard deviations of blank. The study on Listeria monocytogenes sandwich ELISA reported a detection limit of 3.53 × 10³ CFU/mL using streptavidin-PolyHRP amplification [<a href="#ref-3">3</a>].

Result Interpretation

Qualitative Assays

Compare sample OD to a cut-off value. The cut-off is typically calculated as mean negative control OD + 2–3 standard deviations. For the Talaromyces marneffei sandwich ELISA, an optimized cut-off of OD 0.268 yielded optimal sensitivity and specificity [<a href="#ref-1">1</a>].

Quantitative Assays

Plot standard curve (log antigen concentration vs. OD) using 4-parameter logistic (4PL) or quadratic regression. Interpolate sample concentrations from the curve. Ensure sample OD falls within the linear range of the standard curve.

Signal Interpretation

• High signal in negative controls: Indicates non-specific binding or antibody cross-reactivity
• Low signal in positive controls: Indicates reagent degradation or insufficient incubation
• Uneven signal across replicates: Indicates pipetting errors or plate coating inconsistencies

Troubleshooting

Limitations

Known Constraints

1. Antibody pair requirement: Both antibodies must recognize distinct, non-overlapping epitopes, which may not be available for small antigens or highly conserved proteins
2. Matrix effects: Serum, plasma, food homogenates, and other complex samples can interfere with antibody binding or produce non-specific signals
3. Hook effect: At very high antigen concentrations, both capture and detection antibodies may be saturated, producing falsely low signals
4. Time requirement: Standard protocols require 4–6 hours; even rapid formats (e.g., 30-minute CLEIA) require specialized equipment [<a href="#ref-4">4</a>]
5. Cost: High-quality antibody pairs and enzyme conjugates can be expensive, especially for novel targets requiring custom antibody production

Comparison to Other Methods

• Direct ELISA: Faster but less specific (single antibody recognition)
• Indirect ELISA: More sensitive for antibody detection but not suitable for antigen quantification
• Competitive ELISA: Useful for small antigens but requires labeled antigen
• PCR-based methods: Higher sensitivity for nucleic acid targets but cannot detect protein antigens or post-translational modifications

Documentation Requirements

Essential Records

1. Reagent lot numbers and expiration dates: Capture antibody, detection antibody, enzyme conjugates, substrates, blocking reagents
2. Plate layout: Well positions for standards, controls, and samples
3. Incubation conditions: Temperature, time, and agitation settings for each step
4. Wash protocol: Number of washes, buffer composition, soak time
5. Raw data: Absorbance readings for all wells
6. Standard curve parameters: Regression type, R² value, equation
7. Quality control results: CV values, spike recovery, LOD determination

Data Management

Store raw plate reader files, analysis spreadsheets, and final reports in a laboratory information management system (LIMS) or organized folder structure. Include metadata such as operator name, date, and instrument calibration status.

Biosafety Considerations

BSL-1 Routine Scope

For sandwich ELISA using non-pathogenic antigens, purified recombinant proteins, or inactivated samples, standard BSL-1 practices apply as outlined in the CDC/NIH BMBL 6th Edition [<a href="#ref-5">5</a>]:

• Wear laboratory coats and gloves
• Perform work on benchtops with absorbent pads
• Decontaminate work surfaces with 70% ethanol or 10% bleach after each use
• Dispose of liquid waste containing enzyme substrates and stop solutions according to institutional hazardous waste guidelines

Handling Clinical or Food Samples

When testing clinical specimens (e.g., serum, bronchoalveolar lavage fluid) or food samples potentially containing pathogens, follow BSL-2 practices:

• Process samples in a biosafety cabinet (BSC)
• Use sealed rotors for centrifugation
• Inactivate samples with heat (56°C for 30 minutes) or chemical treatment (e.g., 0.5% formalin) before ELISA, if compatible with antigen detection
• The study on Aspergillus galactomannan detection used serum and BALF from immunocompromised patients, requiring BSL-2 containment during sample handling [<a href="#ref-4">4</a>]

Recombinant Antibody Work

If using recombinant antibodies or antigens produced in genetically modified organisms, consult the NIH Guidelines for Research Involving Recombinant or Synthetic Nucleic Acid Molecules [<a href="#ref-6">6</a>]. Ensure institutional biosafety committee (IBC) approval for any recombinant DNA work involved in antibody production.

Frequently Asked Questions

1. Can I use the same antibody for both capture and detection in sandwich ELISA? Yes, but only if the antigen has multiple copies of the same epitope (e.g., repetitive antigens on bacterial surfaces or polymeric proteins). The Talaromyces marneffei sandwich ELISA successfully used the same monoclonal antibody (4D1) for both capture and detection because the target cytoplasmic yeast antigen likely presents multiple identical epitopes [<a href="#ref-1">1</a>]. For most protein antigens with single epitopes, two different antibodies recognizing distinct epitopes are required.

2. How do I choose between HRP and alkaline phosphatase detection systems? HRP is more commonly used due to its lower cost, faster reaction kinetics, and compatibility with TMB substrate (which is non-carcinogenic). Alkaline phosphatase offers advantages when endogenous peroxidase activity is present in samples (e.g., some plant or blood samples) or when longer signal stability is needed. For maximum sensitivity, chemiluminescent substrates for either enzyme can improve detection limits 10–100 fold compared to chromogenic substrates.

3. What causes the "hook effect" and how can I prevent it? The hook effect occurs when antigen concentration is so high that both capture and detection antibodies become saturated, preventing the formation of the "sandwich" complex. This results in paradoxically low signal at high antigen concentrations. To prevent this, always test samples at multiple dilutions (e.g., neat, 1:10, 1:100). If the signal does not increase proportionally with concentration, suspect the hook effect and use the highest dilution that falls within the linear range of the standard curve.

4. How do I validate a sandwich ELISA for a new antigen? Validation requires: (1) confirming antibody pair specificity by Western blot or immunoprecipitation, (2) optimizing coating and detection antibody concentrations by checkerboard titration, (3) establishing LOD and limit of quantification (LOQ) using spiked matrix, (4) assessing precision (intra- and inter-assay CV <15–20%), (5) evaluating accuracy by spike recovery (80–120%), and (6) testing cross-reactivity against a panel of related antigens. For clinical applications, compare results against a gold standard method (e.g., culture or PCR) in a sufficiently powered cohort [<a href="#ref-1">1</a>, <a href="#ref-2">2</a>].

Related Clinical & Scientific Guides

• Loop-Mediated Isothermal Amplification (LAMP) Assay for Rapid Detection of African Swine Fever Virus in Oral Fluids
• How to Calculate the Concentration of Protein Using the Lowry Assay
• DNA Extraction from Saliva: Non-Invasive Sampling and Protocols

References and Further Reading

Related Articles

• Bradford Assay Protocol: Step-by-Step for Protein Quantification
• BCA Assay Protocol: Principles and Step-by-Step Instructions
• Biuret Assay Principle: Protein Quantification by Peptide Bond Detection
• ELISA Assay Principle: How Enzyme-Linked Immunosorbent Assay Works
• Protein Detection Methods: Western Blot, ELISA, and Immunofluorescence Compared
• Lowry Assay Protocol: Protein Quantification Method