Decentralized Mass Spectrometry for Point-of-Need Bioprocessing

For a monoclonal antibody to be safe and effective, the molecule must maintain stability from when it is produced to when it is used.

Although freezing has been the method of choice for storing proteins in bioprocessing industries, it is known to cause ice-water surface denaturation, cryoconcentration, and cold denaturation of proteins over time.^1,2,3

Occasionally, the drug substances must be stored for prolonged periods before conversion into drug products. At approximately 200 K, proteins experience glass transition through which most of the functions of the protein are altered.⁴

Glass transition occurs at lower temperatures due to the change in the dynamic behavior of individual proteins and viscosity during glass formation.

Mass spectrometry (MS) has many applications in the pharmaceutical industry. The new demands of modern bioprocessing and the implementation of PAT and QbD methods necessitate a better understanding of these processes. MS is ideal for providing the information-rich data required for this.

However, traditional MS instruments are large, costly, and must be operated by specialists in centralized laboratories. Integrated real-time MS, decentralized and at the point of need, offers essential information about a process quickly and at a lower cost.

This article discusses the utilization of an innovative point-of-need microscale electrospray ionization mass spectrometer (ESI-MS) for the analysis of the effect of storage time and conditions, medium type, and feeding strategies the lgG quality derived from Fed-Batch (Fresh) and Perfusion (Frozen) CHO cell cultures.

Figure 1. MiD^®ProteinID mass spectrometer and MiDas^TM sampling interface unit. Image Credit: Microsaic Systems plc

Focus

Analyzing the impact of different storage conditions on lgG samples and assessing the MiD^® ProteinlD for inclusion in bioprocessing operations.

Experimental

Table 1. Experimental samples. Source: Microsaic Systems plc

All samples were clarified using a 0.2-micron filter and purified using Pure Proteome Protein A magnetic beads (Sigma-Millipore). The elution was carried out with 1% (v/v) formic acid. Samples were injected with a 2 µL loop into the MS, with all injections conducted in triplicates.

The cell media were spiked with standard lgG to produce spiked recovery controls. The mobile phase employed was Water:MeCN in a ratio of 60:40, as well as with 0.5% (v/ v) formic acid.

Figure 2. Overview of sample preparation and workflow. Image Credit: Microsaic Systems plc

Results

The spiked recovery control sample of standard lgG displayed a consistent mass of 148.2 kDa over the injection, as shown in Figure 3. The Fresh Fed-Batch lgG sample also displayed a consistent mass of 147.2 kDa.

The injection of Frozen Perfusion lgG sample exhibited substantial variation of masses across the injection, having lgG of mass 146.5 kDa eluting early in the injection and lgG of mass 148 kDa eluting near the end of the injection.

Figure 3. Spiked recovery control IgG, Fresh Fed Batch IgG and Frozen Perfusion IgG mass distribution across the injection. Image Credit: Microsaic Systems plc

The lgG produced in Frozen Perfusion and Fresh Fed-Batch samples varied in mass distribution. Figure 4 presents the injection of the Fresh Fed-Batch and Frozen Perfusion samples, where different injection profiles can be observed for the Total Ion Chromatogram (TIC).

Figure 4. Total Ion Chromatogram (TIC) of Fresh Fed Batch and Frozen Perfusion Samples (top). Corresponding mass-to-charge spectra (bottom) for the TIC at 300 to 650 seconds. Image Credit: Microsaic Systems plc

There are differences in mass-to-charge spectra at different time points between the samples. At the peak maximum (300 seconds), clear lgG spectra can be observed for both samples.

In the peak tail (650 seconds), little lgG can be seen in the Frozen perfusion samples, which are dominated by other multiply charged species.

Figure 5 shows the Frozen perfusion sample with a heterogeneous distribution of the masses across the TIC. Different lgG masses (146.6 kDa and 147.4 kDa) were observed at the peak front and middle of the injection.

An increase in signal in the peak tail between 750 and 1750 m/z and extra multiply charged species were also seen.

The uneven masses of the lgG sample from the Frozen Perfusion injection and the uniform distribution of the lgG mass across the Fresh Fed-Batch injection imply that the Fresh lgG from the Fed-Batch was of higher quality. The prolonged storage of Perfusion samples is thought to be the leading cause of lgG deterioration.

This heterogeneous population of lgGs may be further examined through the glycan profiling of these samples.

Figure 5. IgG mass distribution in the Frozen Perfusion sample. Top shows the Extracted Ion Chromatogram (EIC) during the injection of the Frozen sample. Blue represents the intensity of IgG species with mass-146.6 kDa over the injection and green represents the intensity of IgG species with mass of 147.4 kDa over the injection. Red represents the intensity of signal between 750-1,750 m/z for other observed multiply charged species. Image Credit: Microsaic Systems plc

Conclusion

The study described in this article demonstrates that the novel miniaturized point-of-need mass spectrometer can monitor lgG and protein fragments and provide the opportunity to establish product quality.

There were significant differences in the mass distributions of Fresh Fed-Batch lgG and Frozen Perfusion samples following storage for one week and two years, respectively. The considerable difference in the masses indicates that deterioration of the proteins over time may have occurred.

The ice-water surface denaturation and the cryoconcentration (which are both outcomes of storage conditions over time) may have added to the observed heterogeneous mass distribution.

However, further research and experimentation are required to identify which feeding strategy or medium allows better optimization of lgG Production.

Acknowledgments

Produced from materials originally authored by Kala Tamang and Dr. Jean François Hamel from Microsaic Systems. The original authors wish to thank Taehong Kwon and lronie Nagasena from MIT.

References and Further Reading

1. Singh S, et al., (2009) Bioprocess Intl. 9, 32-44
2. Escalante-Minakata, P. et al (2014) J Int Food. 4, 1-11 4
3. Rayfield W. J, et al., (2017) J Pharm Sci. 8, 1944-1951
4. Ringe D., et al., (2003) Biophys Chem, 2-3, 667-680

This information has been sourced, reviewed and adapted from materials provided by Microsaic Systems plc.

For more information on this source, please visit Microsaic Systems plc.