Cell Culture Monitoring in Stirred-Tank Bioreactor with Optical pH Sensors

Novel pH probe and pH sensor spots used in long-term cell culture

Melissa Hill1, Greg Laslo1, Marcelo Kern1, Sayantan Bose1, Christian Krause2
GlaxoSmithKline, King of Prussia, PA, USA
2PreSens Precision Sensing GmbH, Regensburg, Germany

We used PreSens novel autoclavable optical sensor spots, as well as a novel pH probe, for online cell culture monitoring. Optical measurements were compared to reference data from Mettler-Toledo probes. In the first experimental phase, we evaluated the pH sensor spots attached to the bioreactor walls for signal stability, spot-to-spot variability, and drift. The spots trended with the reference measurements and showed only minimal drift over time. The second experimental phase focused on the evaluation of the pH probe, which also trended well with the Mettler-Toledo probe.

Monitoring and controlling important culture parameters such as DO and pH is a prerequisite for mammalian cell culture. Unfavorable conditions could impair growth or even cause culture failure. Therefore, precise and long-term stable sensors are needed, which can be implemented in bioreactors.
Optical sensors can be manufactured in many different sensor designs and sizes, therefore offering more flexibility when integrating these in the culture vessel. While standard 12 mm stainless steel probes must be inserted in the bioreactor through a designated port, optical sensor spots are comparatively much smaller and can even be integrated in bioreactors with no probe ports available. Furthermore, the spots allow non-invasive measurement through the vessel wall, significantly reducing the risk of contamination. While optical pH sensor spots are suitable for use in autoclavable glass bioreactors, the stainless-steel probe can also be applied in autoclavable steel fermenters. In this study, we evaluated two different optical pH sensor designs for cell culture monitoring in a stirred-tank bioreactor. Four pH sensor spots type LG1 were attached to the inside bioreactor wall. One of the spots was used for continuous measurements, while daily checks were made on the other three spots to determine spot-to-spot variability. The new 12 mm pH stainless steel probe by PreSens was inserted in the bioreactor through a standard port. At the same time, optical oxygen sensor spots and an O2 probe by PreSens were evaluated in these experiments, and the results are described in a separate application note. In future experiments, the applicability of these sensors together with a controller will also be tested.

Materials & Methods

The optical pH sensor spots (SP-LG1) were attached to the inner bioreactor wall of 3 L glass bioreactors. A hole was cut in the bioreactor heat jacket so the sensor spots could be read out via a polymer optical fiber from the outside. The reactor with the mounted optical sensors was autoclaved. The sensor spots were single-point calibrated immediately prior to inoculation with a CHO cell line. The pH of the bioreactor was controlled with the Applikon EZ Controller with a setpoint of 7.00 ± 0.05.
An adapter for round containers (ARC) was used to hold the optical fiber detector in place on the outside of the glass next to one spot for continuous read-out. The other sensors were used for periodic checks to determine spot-to-spot variability and the effects of photobleaching over a period of 15 days. The sensor spots were connected to pH-1 SMA LG1 pH meters and measurements were controlled with the PreSens Measurement Studio 2 software. The optical stainless-steel probe was inserted in the bioreactor via a standard port in the lid (Fig. 1). Continuous measurements were performed at 5 second intervals.


Figure 2 shows PreSens pH measurements and reference data for a 15-day monitoring period in cell culture. For this study, a single-point adjustment was made for the PreSens spots immediately prior to inoculation. For the remainder of the study, the Mettler-Toledo probe that controlled the system was sample corrected but the PreSens spots and the second Mettler-Toledo probe were not sample corrected again. The experiment should determine the optical sensor performance and possible drift. The optical pH readings trended well with the reference probes and showed only small drift from day 11 onwards (Fig. 3). Peaks seen on the PreSens measurements throughout the study occurred during the daily readings of the manually monitored spots.

Once a day during offline sampling, the fiber optic cable was manually moved between the three non-continuous spots to test for spot-to-spot variability and drift. Figure 3 shows that the 4 spots drifted to varying degrees. Spot 1 was the continuously monitored spot. Spot 2 was placed in the heating jacket window which allowed it to be exposed to light throughout the study. Spots 3 and 4 were behind the heating jacket and had limited light exposure. After 10 days, the continuously monitored spot drifted about - 0.1 pH units and drifted - 0.2 pH units by day 15. The light exposed spot (spot 2) drifted - 0.3 pH units which was more than all the other spots. The two spots that had limited light exposure and the second Mettler-Toledo probe drifted minimally. These data show that the pH spots are notably sensitive to constant light exposure.

In the following experiment, a pH sensor spot as well as a prototype pH probe by PreSens were used for monitoring and compared with the controlling Mettler-Toledo probe (Fig. 4). The PreSens pH spot and probe as well as the Mettler-Toledo probe were single-point calibrated once a day. Both the sensor spot and steel probe trended with the reference measurements. The pH probe showed more signal noise, the reason may be higher sensitivity to air bubbles.


Our evaluation of autoclavable optical pH sensor spots and a prototype stainless-steel probe showed that these sensors are suitable for pH monitoring in mammalian cell culture. The non-invasive sensor spots offer several advantages when the number of probe ports is limited. Multiple spots can be integrated at different positions in the bioreactor, which might give additional information on mixing effects. The sensors showed only minimal drift over time and trended with the reference probes. However, the PreSens sensors do have signal noise and light-sensitivity. In the next phase of our study, the optical pH probe will be connected to the controller and the signal used for controlling the culture pH.

The project leading to this application has received funding from the Innovative Medicines Initiative 2 Joint Undertaking under Grant Agreement n°777397. This Joint Undertaking receives the support from the European Union’s Horizon 2020 research and innovation program and EFPIA.

The communication reflects the author's view and neither IMI nor the European Union, EFPIA, or any Associated Partners are responsible for any use that may be made of the information contained therein.


Select by industry

Presens TV

Watch tutorials, webinars and informative videos about PreSens optical sensor systems.

All videos