Multiple methods exist to measure protein concentration: UV spectroscopy, BCA assay, Bradford assay,… They all have pros and cons. One of the main limitations is the influence of the presence of specific amino acids. UV spectroscopy requires for example the presence of aromatic amino acids. It can prevent the use of this technique for peptides that would not contain these amino acids. In addition, for the BCA and Bradford assay, the calibration is realized on common protein such as albumin and not on the protein of interest. This dependency on specific amino acids thus lead to inaccuracy of the measurement

FTIR spectroscopy turns out to be a technique of choice to measure the protein concentration. The two largest bands in protein FTIR spectra (amide I and amide II) are due to the vibrations of the amide bond present in proteins. The intensity of these absorption bands is proportional to protein concentration and only depends on the amide bond (no specificity due to amino acid side chain).

The figure below illustrates the variations of the intensity correlated to the protein concentration.

fig1 prot conc

Furthermore, using multivariate statistical tools, we built a predictive model that allows to measure protein concentration of unknown samples based on their FTIR spectra. The second figure shows the validation of this model, i.e., the application of this model on independent samples: it reports the predicted protein concentration as a function of the true concentration. The error bars indicate the standard deviation associated with this concentration. With a correlation coefficient of 0.999, the predicted values are highly correlated with the true values, underlining the performance of this model.

fig2 prot conc

Spectralys Biotech can thus develop accurate methods to measure protein concentration that do not depend on the presence of specific amino acids.