I performed an indirect ELISA to check the binding strength of nanobodies against an antigen. The purified ‘Fc-conjugated nanobodies’ and the ‘antigen without Fc’ were first confirmed on SDS-PAGE, then the binding strength of the antigen was further confirmed with ELISA using a positive control (a binding nanobody different from the one tested in the given assay). When testing the four nanobodies using a standard ELISA protocol, instead of getting the OD450 in order of nanobody dilution (highest to lowest: column 1 to 11), the signal gradient is reversed. I am encountering this problem repeatedly with different nanobodies against the same antigen.
Protocol: (i) Ag coating with overnight incubation at 4 degrees. (ii) One-time wash with PBS (iii) Blocking with MPBS (non-fat dry milk in PBS) for one hour at room temperature (iv) Primary antibody (nanobody) two-fold dilution with 10,000M initial concentration, starting from column 1 to 11 in replicates (A and B same nanobody and so on), NO
primary Ab in column 12. C12 is a negative control (only coated with Ag and treated with MPBS). Incubation for two hours at room temperature with moderate shaking. (v) Three-times wash with PBST (1:10000) (vi) Secondary Ab/detection Ab (Horseradish peroxidase enzyme-conjugated anti-human Fc secondary Ab) incubation for one hour at room temperature with moderate shaking. (vii) Four-times wash with PBST (1:10000) (viii) Treated with TMB for 5-10 minutes (ix) Substrate (H2SO4) to stop the reaction

P.S. Multiple optimizations in the protocol were made, such as increasing the Nanobody and Ag concentrations and incubation time, and increasing or decreasing the secondary antibody volume, etc. There are no handling issues because the same treatment with the positive control gave results in the right order (OD from higher to lower).

Has anybody encountered a similar issue, or does anyone understand this problem?