Using ELISA/RIA Kits in Peptide Research: Standards, Range and Sensitivity

Most ELISA kits for peptide research are competitive formats rather than sandwich assays, because peptides are too small to bind two antibodies at once — and in a competitive assay signal falls as analyte concentration rises, which is the single most common source of misinterpreted results. A sandwich assay needs two antibodies binding non-overlapping epitopes simultaneously, which requires roughly 3 kDa or more of surface; below that, competition against a labeled tracer is the workable design. This guide covers format selection, how to build and read a standard curve, the validation checks that decide whether a number means anything, and worked interpolation and unit-conversion examples. Research use only.

ELISA kits for peptide research: three formats and when each applies

The trade is straightforward: sandwich formats are more sensitive and have a wider range, but they are unavailable for most peptide targets. RIA retains an edge in detection limit and suffers less from matrix interference, at the cost of licensing and a tracer whose specific activity falls continuously — see the decay arithmetic in labeled peptides.

Building the standard curve

The standard curve is the assay. Everything else is sample handling.

1. Reconstitute the supplied standard exactly as the kit insert specifies, and record the resulting concentration. Kit standards are usually lyophilized peptide with carrier protein, so the reconstitution volume is not negotiable.
2. Serially dilute in the assay diluent, not in buffer of your own choosing. A typical two-fold series from a 1,000 pg/mL top standard runs 1,000 / 500 / 250 / 125 / 62.5 / 31.25 / 15.6 / 0 pg/mL.
3. Run every point in duplicate at minimum, triplicate where material allows.
4. Include the zero standard. In a competitive assay this is B₀ — maximum binding, maximum signal — and every other point is expressed against it.
5. Fit a four-parameter logistic curve, not a straight line. Immunoassay concentration–response curves are sigmoidal, and a linear fit through the middle discards the shape that defines the working range.
6. Check the fit. R² above 0.99 on the 4PL, and back-calculated standards within 80–120% of nominal.

Worked example: interpolation and dilution factor

1. Sample reads within the curve. An unknown gives an optical density falling between the 125 and 250 pg/mL standards; the 4PL fit interpolates it at 170 pg/mL.
2. Apply the dilution. The sample was diluted 1:20 before loading, so the concentration in the original material is 170 × 20 = 3,400 pg/mL, or 3.4 ng/mL.
3. Convert to molar units. For a peptide of 1,419.55 Da, 1 ng/mL = 1 µg/L ÷ 1,419.55 g/mol = 0.704 nM. So 3.4 ng/mL = 3.4 × 0.704 = 2.39 nM. The general conversion is in molecular weight, moles and molarity.
4. Sanity-check the placement. 170 pg/mL sits comfortably inside the curve. A reading above the top standard or below the second-lowest is an extrapolation, not a measurement — re-run at a different dilution.

The validation checks that decide whether a number is real

• Spike recovery. Add a known amount of standard to your matrix, measure, and calculate recovery. 80–120% is the usual acceptance window. Recovery of 40% means the matrix is suppressing the signal and your unknowns are being under-reported by the same factor.
• Dilution linearity (parallelism). Measure a sample at 1:2, 1:4 and 1:8, multiply each by its dilution factor, and compare. If the back-calculated values diverge, something in the matrix is interfering and no single dilution is trustworthy.
• Intra-assay CV — replicate wells on one plate — should sit under about 10%. Inter-assay CV across plates and days should sit under about 15%.
• Cross-reactivity. The kit insert should list percentage cross-reactivity against related sequences. For peptide families this is decisive: an antibody raised against a full-length sequence commonly recognises its fragments and its close analogues, and a kit that cannot distinguish an analogue from the native peptide cannot answer a question that depends on the difference.
• Sensitivity terms. The limit of detection is the lowest concentration distinguishable from zero; the lower limit of quantification is the lowest concentration measurable with acceptable precision. Kits often advertise the first and are usable only from the second.

Matrix, handling and the errors that come from them

1. Sample collection matters more than the assay. Many peptides are degraded within minutes by plasma peptidases; protease-inhibitor tubes, immediate chilling and rapid separation are routine requirements, and the kit insert usually specifies them.
2. Serum, plasma and cell-culture supernatant are not interchangeable. A kit validated for one matrix has to be re-validated by spike recovery for another.
3. Repeated freezing degrades the analyte, not just the standard. Aliquot samples on collection.
4. Bring every reagent to room temperature and mix thoroughly. Temperature gradients across a plate are a classic source of edge effects.
5. Wash consistently. Incomplete washing raises background; over-vigorous washing strips the coating. Both flatten the curve.
6. Read within the specified window after stop solution. Colour development continues, and a plate read twenty minutes late is a different plate.

Common misinterpretations

• Reading a competitive assay as if it were direct. In competition, high signal means low analyte. Every conclusion inverts if this is missed.
• Extrapolating beyond the top or bottom standard. The 4PL asymptotes there; the numbers it returns are curve artefacts.
• Reporting "total" peptide from an assay that measures one epitope. Degraded fragments may or may not be counted, depending on where the antibody binds.
• Assuming immunoreactivity equals biological activity. An oxidised or misfolded peptide can be fully immunoreactive and inactive. Orthogonal confirmation — chromatographic or by mass spectrometry — is what closes that gap.
• Comparing absolute values across kits. Different antibodies and different standards give different absolute numbers for the same sample; only within-kit comparisons are safe.

Choosing a kit

Match four things in order: the species reactivity, the matrix you will actually run, the quantification range against the concentrations you expect, and the cross-reactivity profile against the sequences you need to exclude. A kit that fails on any of these cannot be rescued by careful technique. Format guidance and the catalogue structure are set out in ELISA, EIA and RIA kits: what they are and how to choose, with the kits themselves under assay kits and consumables under kit supplies in the research catalog. Where an assay needs to be built rather than bought, the antibody and labeled-peptide ranges supply the components.