Decoupling Protein Concentration and Aggregate Content Using Diffusion and Water NMR

Abstract

Protein-based biopharmaceutical drugs, such as monoclonal antibodies, account for the majority of the best-selling drugs globally in recent years. For bioprocesses, key performance indicators are the concentration and aggregate level for the product being produced. In water NMR (wNMR), the use of the water transverse relaxation rate [R₂(¹H₂O)] has been previously used to determine protein concentration and aggregate level; however, it cannot be used to separate between them without using an additional technique. This work shows that it is possible to "decouple" these two key characteristics by recording the water diffusion coefficient [D(¹H₂O)] in conjunction with R₂(¹H₂O), even in the event of overlap in either D(¹H₂O) or R₂(¹H₂O). This method is demonstrated on three different systems, following appropriate D(¹H₂O) or R₂(¹H₂O) calibration data acquisition for a protein of interest. Our method highlights the potential use of benchtop NMR as an at-line process analytical technique.

Figure 1

Plots showing the effect of the concentration on R₂(¹H₂O) for BSA (A) and mAb (B) solutions. The colors represent individual concentrations studied, while different plot markers represent changing aggregate contents from lowest to highest within each concentration (for exact aggregate content values, see Tables S2 and S3). Sample errors were determined by taking the standard error of the arithmetic mean of three sample measurements; values can be found in Tables S2 and S3. Error bars are excluded for clarity where the errors are smaller than the symbols used.

Figure 2

Plots showing the effect of aggregate percentage on R₂(¹H₂O) for BSA (A) and mAb (B) solutions over a range of different concentrations. The solid lines represent linear regression fits to individual sample concentrations (as determined in MATLAB 2022a [MathWorks, US]), and the blue-shaded areas represent 95% confidence intervals calculated from average of the standard error of all data points for each protein. The 95% confidence intervals are multiplied by a factor of 5 for visibility. Sample errors were determined by taking the standard error of the arithmetic mean of three sample measurements; values can be found in the Tables S2 and S3. Error bars are excluded for clarity where the errors are smaller than the symbols used.

Figure 3

Section of a SEC chromatogram recorded for 9.81 mg mL^–1 mAb solutions, with varying stressed fractions (SF) between 0 and 100%.

Figure 4

Plots showing the effect of aggregate percentage on D(¹H₂O) for BSA (A) and mAb (B) solutions over a range of different concentrations. The solid lines represent linear regression fits to individual sample concentrations (as determined in MATLAB 2022a [MathWorks, US]), and the blue-shaded areas represent 95% confidence intervals calculated from average of the standard error of all data points for each protein. Individual data point errors were determined by taking the standard error of the arithmetic mean of three sample measurements; values can be found in Tables S2 and S3.

Figure 5

Plots showing the effect of concentration with R₂(¹H₂O) (A) and D(¹H₂O) (B) for BisAb solutions before (Pre-HT, circles) and after heat-stress (Post-HT, diamonds). The dashed line in (A) represents a constant R₂(¹H₂O) value of 0.920 s^–1. The dotted and dashed lines in (B) represent constant D(¹H₂O) values of 2.310 × 10^–9 and 2.410 × 10^–9 m² s^–1, respectively. Sample errors were determined by taking the standard error of the arithmetic mean of three sample measurements; values can be found in Table S5. Error bars are excluded for clarity, where the errors are smaller than the symbols used.