Improving the accuracy of sandwich ELISA experiments requires comprehensive optimization across experimental design, operational standards, reagent quality control, and other，with the core goal of reducing non-specific interference, minimizing errors,
and ensuring that detection results align with true values. Specific measures are as follows:
Select paired antibodies (capture and detection antibodies) targeting different epitopes of the target antigen to ensure no cross-reactivity (pre-experiments can verify cross-reactivity with homologous molecules to be <5%).
Prioritize monoclonal antibodies (high specificity) or affinity-purified polyclonal antibodies to avoid non-specific binding caused by low antibody purity.
Optimize antibody concentrations using checkerboard titration (optimal dilution ratio of capture to detection antibodies) to ensure signal values (OD) fall within the linear range of the standard curve and background values (blank well OD) are <0.15.
Use standards matched to the sample matrix (e.g., for serum samples, standards should be dissolved in a diluent containing the same serum matrix) to reduce matrix effect interference.
Standards must be calibrated (e.g., traceable to international standards) with inter-batch CV% <10%. They should be diluted, aliquoted, and frozen strictly according to instructions (avoid repeated freeze-thaw cycles).
Ensure enzyme conjugates (e.g., HRP-labeled secondary antibodies) retain activity (verified by chromogenic efficiency)
to prevent signal attenuation due to enzyme inactivation.
Chromogenic substrates (e.g., TMB) must be freshly prepared, stored protected from light, and matched to the enzyme conjugate (e.g., HRP with TMB, AP with PNPP). Stop solutions (e.g., H₂SO₄) must have accurate concentrations to avoid over- or under-termination.
Samples (e.g., serum, cell supernatants) should be freshly collected, avoiding hemolysis or lipemia (hemoglobin and lipids interfere with antigen-antibody binding). After centrifugation to remove precipitates, test immediately or aliquot and freeze
at -80°C (avoid repeated freeze-thaw cycles).
If antigen concentrations in samples are too high (exceeding the standard curve range), perform gradient dilution with a dedicated diluent (e.g., 1:10, 1:100) to ensure detection values fall within the linear range of the standard curve. Dilutions must be thoroughly mixed (using a vortex mixer).
Incubation temperature must be precise (e.g., 37°C water bath or constant-temperature incubator) to avoid variations in
antigen-antibody binding efficiency due to temperature fluctuations. Incubation time must strictly follow the protocol
(e.g., capture antibody coating requires overnight incubation at 4°C; avoid shortening time, which reduces coating efficiency).
Cover microplates during incubation to prevent evaporation-induced edge effects (higher OD values in edge wells); use anti-evaporation plate mats if necessary.
Washing buffers (e.g., PBS with Tween-20) must be freshly prepared with accurate concentrations (Tween-20 is typically 0.05%;
excessive concentrations disrupt antibody binding, while insufficient concentrations fail to remove non-specific binders).
Use an automated plate washer (to avoid inconsistencies in manual ) with ≥300μL washing volume per well and at least
3 washes.After each wash, blot dry on absorbent paper (to prevent residual liquid from diluting subsequent reagents).
Blank controls: Contain only diluent and detection reagents, used to correct background signals (OD values of all samples and standards must subtract blank control OD values).
Negative controls: Samples known to lack the target antigen (e.g., healthy human serum) to verify non-specific chromogenicity
(negative control OD values must be <1/2 of the OD value of the lowest standard concentration).
Positive controls: Samples known to contain the target antigen (at mid-range standard curve concentrations) to monitor experimental validity (if positive control OD values deviate from expectations by ±20%, repeat the experiment).
Matrix controls: Contain only sample matrix (e.g., serum without antigen) to assess matrix interference (matrix control OD values should approximate blank controls).
Both standards and samples should have 2-3 replicate wells, with the mean calculated as the final result.
If the coefficient of variation (CV%) of replicate well OD values >15%, retest.
Critical experiments require ≥3 independent repeats (inter-batch replicates) with inter-batch CV% <20% to ensure
reproducibility.
Standard concentration gradients must cover the expected sample concentration range (typically 5-8 points, including
0 concentration). Add extra points in regions with steep signal changes (e.g., EC30-EC70) to improve fitting accuracy.
Prefer the 4-parameter logistic regression (4PL) model for curve fitting (R² ≥0.99) to avoid errors in linear regression
at high/low concentrations. Define the lower limit of quantitation (LLOQ, lowest concentration with CV% <20%) and upper limit of quantitation (ULOQ); samples outside this range must be re-diluted and retested.
After blank correction of OD values, explicitly exclude outliers (e.g., wells with >±20% deviation from the mean of replicates) and document the reason for exclusion.
If samples exhibit significant matrix effects (spike recovery <80% or >120%), mitigate by diluting samples (reducing matrix concentration) or using matrix-matched standards.
Regularly calibrate microplate readers (wavelength accuracy, absorbance precision) to ensure detection wavelengths match chromogenic substrates (e.g., 450nm for TMB, with 630nm as a reference wavelength to correct for plate unevenness).
Preheat the instrument for 30 minutes before detection to avoid temperature fluctuations affecting readings. Ensure microplate bottoms are clean (no fingerprints or liquid residues) during reading to prevent light scattering interference.
Record detailed experimental parameters (e.g., antibody batches, incubation time/temperature, microplate reader model), raw data (replicate well OD values, standard curve equations), and anomalies (e.g., sample hemolysis, expired reagents) to facilitate result traceability and troubleshooting.
By implementing these measures, interference factors can be minimized from the source, ensuring the accuracy and reliability of sandwich ELISA experiments and providing a solid foundation for quantitative detection results.