White Light Spectroscopy For T-Cell Culture Growth Monitoring

INTRODUCTION

Advanced Therapy Medicinal Products (ATMPs) represent a new class of pharmaceuticals that provide innovative treatments for patients with limited therapeutic options. These therapies often involve complex processes like genetic modification and tissue engineering, making their production costly and limited to a small number of patients. A potential solution to reduce costs and increase accessibility is to automate the production process, particularly the monitoring of T-cell growth during the expansion phase. Traditional cell counting methods, although effective, often involve sampling, which can introduce contamination and increase production time. This application note explores the use of white light spectroscopy as a real-time, non-invasive method to monitor T-cell culture growth, offering a more efficient alternative for ATMP production.

METHODS

An absorbance system is used in this application. Historically in cell culture, cell concentration estimation relies on directly measuring cell number, by counting them one by one, because the cell size makes it possible to directly observe them with common microscopes. On the contrary, absorption-based methods like turbidimetry or Beer-Lambert law derived techniques (optical density measurements for example) are usually preferred when considering smaller biological entities such as bacteria.

The objective is to measure the absorption spectra of cell solutions at different concentrations and plot the cell concentration as a function of the spectral behavior. This is called “optical concentration modeling (OCM)”.

Figure 1a shows that the reproducibility and the accuracy of the spectroscopic measurements are extremely good. This is confirmed when calculating the number of cells in the flask using the OCM. The result is shown on figure 1b where spectroscopic measurements are plotted with red stars together with results obtained with the automatic cell counter already presented earlier. SDs measured during week 15 with the OCM show a median value less than 2% with extreme values at 0.4% and 3%. As highlighted on figure 1(b), the number of cells in the flask evolves exponentially with time as expected.

Figure 1: Monitoring the cell number using white light absorption spectroscopy. (a) Raw spectra recorded during week 15. Colors refer to days, spectra recorded the same day are drawn with the same color. (b) Total number of cells calculated using the OCM (red stars) compared to counts obtained from the flask and the cuvette with the automatic counter.

SDs corresponding to all experiments are reported on figure 2. On each box, the outliers are plotted individually using red crosses. It is observed that the spectroscopic measurement lead to the lowest median value (2%) while the median values obtained with the automatic counter are much larger (8.5% for the flask and 10% for the cuvette). Also, data dispersion appears to be larger when sampling the solution from the cuvette than directly from the flask.

Figure 2: Box and Whiskers plot representing the SDs measured during the 8 monitoring experiments (expressed in %). The outliers are plotted individually using red crosses.

Figure 3 shows data obtained during week 15 fitted exponentially with time applied to cell numbers measured from the flask, the cuvette and spectroscopically respectively. For this example, the R2 of the fittings were 0.97, 0.92 and 0.99 for flask, cuvette and spectroscopy respectively. The initial numbers of cells reported result from fitting data concerning the flask, the cuvette and the spectrometer. Flasks were initially seeded with a CEM concentration of 5×105 CEM.mL-1. The volume of the initial culture being 15 mL, this represents a theoretical initial number of cells equal to 7.5×106 cells. It can be noted that, whatever the counting method is, the observed initial number of cells is higher than the theoretical one.

Figure 3: Fitting experimental data obtained during week 15 with exponential functions.

Dividing times have been determined from the fitting of spectroscopic data. They range from 30 to 50 hours. Finally, a weak correlation was also noticed between the dividing times and the initial numbers of cells (figure 4) with a correlation coefficient R2=0.87. On figure 4, data were fitted with an exponential function for illustration purpose. The low value of the R2 makes it impossible to conclude that the evolution is actually exponential.

Figure 4: Evolution of the dividing times with the initial number of cells.

Figure 4 shows that the dividing times increase with the initial number of cells. This means that cells grow more slowly when they are numerous. This is in accordance with the cell growth kinetics which exhibits different phases. Among them, the exponential phase which continues until changes in the environment like depletion of nutrients and increase of waste metabolic products lead to a stationary phase. The evolution we observe on figure 4 is then logical because a larger number of cells places the culture closer to the stationary phase conditions.

BUILD SYSTEM

INTINS can provide a complete system for this application. The Ocean SR4 UV-VIS spectrometer is a high-performance spectrometer with high-speed spectral acquisition and excellent signal-to-noise ratio performance for diverse applications. This small-footprint instrument unlocks UV-VIS signature data from 190-1100 nm and entrance slit options in widths of 5 µm to 200 µm. The SR4 spectrometer is compact, versatile, and compatible with Ocean Insight light sources and accessories.  

The DH-2000 series is the world’s only balanced deuterium halogen source. It uses innovative filtering technology to produce a smooth spectrum across the entire range, balanced output from 215 to 2500 nm and eliminates problems associated with saturation. This same technology eliminates the alpha-deuterium line in the visible region. Using a combination of deuterium and halogen lamps, the DH-2000 is flexible and ideal for measuring a sample that has multiple features in different spectral regions or for analyzing a variety of different samples.

The Cuvette Holder for 1-cm pathlength cuvettes couples via SMA-terminated optical fibers to spectrometers and light sources to create small-footprint spectrophotometric systems for absorbance and fluorescence experiments. The Cuvette Holder has a fully integrated cover for eliminating ambient light and has 2 filter slots to enable filtering illumination light entering the cuvette holder and/or detected light leaving the cuvette holder. The unit is designed to snugly hold 1-cm square cuvettes without user adjustment, providing high data repeatability.

CONCLUSION

White light spectroscopy offers a robust and non-invasive method for real-time monitoring of T-cell culture growth, providing higher accuracy and reproducibility than traditional cell counting methods. This approach not only reduces the risk of contamination but also enhances the efficiency of ATMP production processes. Future developments will focus on integrating this spectroscopic monitoring system into a closed-loop setup for continuous quality control and automated production of cell-based therapies.