Multi-dimensional fluorescence (MDF) spectroscopies like excitation-emission matrix (EEM) and Total Synchronous Fluorescence Scan (TSFS) offer unique possibilities for the quantitative and qualitative analysis of proteins in complex biogenic liquids. Conventional EEM spectroscopy was used for the quantitative predictive analysis of glycoprotein production in a CHO cell fed-batch process. EEM spectra of complex solutions are very sensitive to compositional change and this can be monitored using Multivariate Curve Resolution (MCR). Despite the fact that the relationship between spectral change and fluorophore concentrations are highly non-linear in these biogenic liquids, MCR can extract useful informative data about the changes in some amino acids and the protein product. Here MCR clearly showed the increase of a unique dityrosine emission from the product glycoprotein with process time. Accurate quantitative predictive regression models for end glycoprotein yield using EEM were possible with relative errors of <5%. This approach can be used for process management at early stage by predicting final process outcome at the early, small-scale (100-200L) stage of fed-batch manufacturing processes.
The incorporation of extra dimensions to MDF measurements such as anisotropy, provides an alternative approach to macromolecular component and fluorophore resolution based on rotational diffusion and/or molecular size. We show how protein can be quantified in model cell culture media (yeast hydrolysate) and how the methodology can be extended to the structural analysis of multi-fluorophoric proteins. This Anisotropy Resolved Multi-dimensional Emission Spectroscopy (ARMES) method for example could clearly resolve different tyrosine fluorophore populations and room temperature phosphorescence from albumin protein emission.