The development and optimization of cell culture media for biotech applications is a fundamental step of process development. The composition of cell culture media requires an ideal blend of amino acids, vitamins, nucleosides, lipids, carbohydrates, trace elements and other components.
Journal of Chromatography A 1651 (2021) 462336 Contents lists available at ScienceDirect Journal of Chromatography A journal homepage: www.elsevier.com/locate/chroma Short Communication A rapid, simple and sensitive microfluidic chip electrophoresis mass spectrometry method for monitoring amino acids in cell culture media Meire Ribeiro da Silva a, Izabela Zaborowska a, Sara Carillo a, Jonathan Bones a,b,∗ a b NIBRT – National Institute for Bioprocessing Research and Training, Dublin, Ireland School of Chemical and Bioprocess Engineering, University College Dublin, Dublin 4, Ireland a r t i c l e i n f o Article history: Received 27 February 2021 Revised June 2021 Accepted June 2021 Available online June 2021 Keywords: Capillary electrophoresis mass spectrometry Amino acid analysis Spent media analysis Cell culture Monoclonal antibody Upstream processing a b s t r a c t The development and optimization of cell culture media for biotech applications is a fundamental step of process development The composition of cell culture media requires an ideal blend of amino acids, vitamins, nucleosides, lipids, carbohydrates, trace elements and other components The ability to monitor these constituents is required to ensure that cells receive sufficient nutrients to facilitate growth, viability and productivity Analysis of cell culture media is challenging due to the range and diversity of compounds contained in this matrix and normally requires time consuming methods A rapid, simple and sensitive microfluidic chip CE-MS method is described to monitor amino acids in chemically defined cell culture media from a Chinese hamster ovary cell line cultured over a period of 10 days The described platform enabled the separation of 16 amino acids in less than minutes and without the requirement for extensive sample preparation The analytical parameters evaluated were precision, linearity, limit of detection and limit of quantification The majority of essential amino acids were present in cell culture growth in high concentrations compared to non-essential amino acids Over the course of the 10 days cell culture the concentration of certain amino acids declined by up to 100% Microfluidic chip based CE-MS methods can be used effectively to obtain the consumption rates of amino acids in cell culture media during cell growth and to perform at-line monitoring and screening of cell culture status © 2021 The Author(s) Published by Elsevier B.V This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/) Introduction Cell culture media (CCM) is a key component of upstream bioprocessing and correct selection of an appropriate CCM is vital to ensure optimal performance for both cell growth and recombinant protein yield and quality CCM is composed of various amino acids, vitamins, inorganic salts, nucleosides and anti-shear agents The quantity of each component present may affect cell growth, specific productivity and other performance parameters Therefore, the optimization and customisation of CCM have been extensively studied to address nutrient demands of cell lines and to control toxic metabolite production [1–5] Among these nutrients, amino acids (AAs) are essential as they are the primary constituents of proteins and intermediates in a variety of cellular metabolic pathways During cell growth, AAs can be both consumed from and released into the CCM Monitoring AAs is crucial to understand the dynamic conditions and to adjust the concentration of each ∗ Corresponding author at: NIBRT – National Institute for Bioprocessing Research and Training, Dublin, Ireland E-mail address: Jonathan.bones@nibrt.ie (J Bones) AA in the CCM accordingly to optimize recombinant protein yield and quality [6] The analysis of AA in CCM can be challenging due to the potential presence of interfering compounds depending on CCM composition Additionally, AAs are typically amphoteric with significant differences in their chemical structures, different polarities from non-polar to highly polar and acidic to basic side chains, all these properties may affect detection and separation [7] Different approaches have been described to monitor AAs in spent media such as liquid chromatography with either optical (LC-UV or LC-fluorescence) or mass spectrometric detection (LCMS) and gas chromatography with either flame ionisation (GC-FID) or mass spectrometric detection (GC-MS) [8–13] The majority of these techniques include sample preparation steps such as derivatization, solid-phase microextraction (SPME), solid-phase extraction (SPE) and others, which increased the complexity of the method, the associated time required to perform the analysis and the risk of analyte loss during extraction and sample preparation [8,13] Rapid LC-MS methods have improved analysis time by avoiding derivatization steps, however, ion pairing reagents are often used that reduce sensitivity and increase the risk of potential problems with quantitation due to ion suppression [8,9] As the ideal analytical https://doi.org/10.1016/j.chroma.2021.462336 0021-9673/© 2021 The Author(s) Published by Elsevier B.V This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/) M Ribeiro da Silva, I Zaborowska, S Carillo et al Journal of Chromatography A 1651 (2021) 462336 approach to monitor AAs is to avoid complex sample preparation, miniaturization of analytical devices and the use of microfluidic techniques provides an interesting analytical option that allows fast analysis, high sensitivity and low consumption of reagents and sample [14–18] Microfluidic capillary electrophoresis (CE) is an attractive analytical option as it offers low consumption of background electrolyte and reagents, excellent performance, fast separation and is suited to a wide range of applications ranging from small molecule analysis to characterisation of intact proteins [19–24] As well, CE has been widely used for AA analysis in a variety of complex matrices [25] Microfluidic chip CE technology is versatile and can be coupled to different detection options including optical, electrochemical and mass spectrometric detectors [25] Microfluidic chip CE coupled to MS operates at low flow rates, similar to nanospray ionisation, which enables maximization of MS sensitivity However, CE and MS coupling is not without its challenges, due to the use of low pressure, ensuring stability of the low operational flow rates and compatibility of components of the background electrolyte with MS detection [26] Many applications utilizing microfluidic chip CE-MS using electrospray ionization have highlighted the reduction of analysis time and the increase in sensitivity [22,27,28] The present study focuses on the analysis of AAs in spent cell culture media using microfluidic chip CE-MS to generate a rapid, simple and sensitive method to monitor AAs in spent media from IgG1 monoclonal antibody expressing Chinese Hamster Ovary (CHO) cells over a 10 day batch culture, employing stable isotopically labelled AAs as internal standards The optimised method proved to be simple and rapid with minimal sample preparation required, while the use of high resolution Orbitrap mass spectrometry for the detection of the amino acids provided improved selectivity and quantitative capability for those AAs that were difficult to separate during the rapid CE analysis The method is advantageous and sufficiently fast to provide an at-line analysis for the identification and quantification of AAs consumed during cell growth to ensure that cells are maintained in an optimal environment for proliferation and production when the cell viability dropped below 70%, which generally occurred after day Samples were collected on a daily basis by removing mL aliquots from the cultures, which were centrifuged to pellet the cells and the collected supernatant was stored at 4°C prior to analysis Prior to analysis of analytical samples, the collected cell culture media samples were diluted 20 times with LCMS grade water and spiked with μL of the IS stock solution described below to yield a final concentration of μM of IS Data points for the concentration of each AA were calculated from triplicate injection of the samples obtained from each biological replicate (n = 9) 2.3 Preparation of calibration standards Stable heavy isotope-labelled AA were employed as internal standards (IS) and were prepared as standard 500 μM solutions and subsequently diluted to obtain 50 μM, 5.0 μM, 0.5 μM and 0.05 μM calibration solutions, using fresh cell culture media diluted 1:20 v/v with LC-MS water These stock solutions were diluted 1:10 v/v with metabolite sample diluent provided in the ZipChip metabolite kit to obtain a standard calibration curve These standards were analysed in triplicate and the peak area obtained from the base peak chromatograms were used to estimate a calibration factor that was used in sample concentration calculations 2.4 Microfluidic chip CE-MS settings All analyses were performed using a ZipChipTM device and autosampler from 908 Devices (Boston, MA), that was interfaced with a Q ExactiveTM Plus hybrid quadrupole Orbitrap mass spectrometer (Thermo, Bremen, Germany) The mass spectrometer settings included: mass resolution 17,500 at m/z 200, micro scans, an acquisition gain control (AGC) target × 106 , maximum inject time 100 ms, spray voltage kV, sheath gas flow rate arbitrary units (au), auxiliary and sweep gas flow rate au, capillary temperature at 200°C, S-lens radio frequency (RF) level 50, with data acquired over a mass range of 70–500 m/z Data acquisition was accomplished through the XcaliburTM tune page, which was triggered by the ZipChip software The microfluidic chip CE settings were: injection load time 30 s at Pa (4 nL), analysis run time minutes, pressure assist start time 0.5 minutes and field strength 10 0 V/cm Materials and methods 2.1 Chemicals and reagents A ZipChipTM HS chip (cat# 810-0013) and ZipChip Metabolites Kit (cat#850-0 033), including the ZipChip Metabolites BGE, sample diluent and acid were obtained from 908 Devices (Boston, MA, USA) Stable isotope labelled AAs and Metabolomics Amino Acid Mix Solution were obtained from Cambridge Isotope Laboratories, Inc (MSK-A2-1.2 P/N 17K-628, UK) LC-MS Optima grade water was purchased from Fisher Scientific (Dublin, Ireland) Results and discussion 3.1 Rapid separation of amino acids using microfluidic chip CE-MS Fig 1a depicts the base peak electropherogram following injection of the 50 μM of AAs standard mix solution In this instance, since the BGE is acidic, all AAs are positive charged and the resulting separation of the 16 AAs is based on the difference of electrophoretic mobility that is related to the charge and size of the analytes As expected, the amino acid migration order observed was highly charged AAs migrating first, followed by neutral AAs and then the acidic residues The 16 AAs could be separated using a three minute method, with a separation window spanning less than two minutes and with sample requirements of nL per injection The method resulted in symmetrical and narrow peaks with asymmetry values ranging between 0.96 and 1.32, average asymmetry 1.07, peak width at half height ranges between 0.006 and 0.036 minutes, average peak width 0.0133 minutes and good resolution Some of the AAs were not fully resolved within the separation window with comigrating pairs observed at migration times from 1.4 to 2.0 minutes correspond to methionine and threonine, phenylalanine and proline as showed in Fig 1A Comigration of certain amino acids such as methionine and threonine, asparagine 2.2 Cell culture and sample preparation CHO DP-12 cells [CHO DP12, clone#1934 aIL8.92 NB 28605/14] (ATCC® CRL12445TM ) were adapted to grow in suspension in animal-component free, chemically defined medium BalanCD CHO Growth A (Irvine Scientific, Wicklow, Ireland) with the addition of mM L-glutamine (Sigma Aldrich, Wicklow, Ireland) Cells were cultured using batch culture conditions in triplicate Cultures were initiated by seeding 0.3 × 106 cells/mL in 250 mL polycarbonate Erlenmeyer flasks (Corning, Amsterdam, The Netherlands) containing 100 mL of media supplemented with mM of L-glutamine on day Cells were cultivated in a shaking incubator at 37°C, 5% CO2 for 10 days Cell density and viability were determined daily by trypan blue exclusion assay using a Countess automated cell counter (Invitrogen, Carlsbad, CA, USA) Cell culture was stopped M Ribeiro da Silva, I Zaborowska, S Carillo et al Journal of Chromatography A 1651 (2021) 462336 Iso (A His 100 Val 90 Leu Lys Arg Relative Abundance 80 70 Met/Thr Phe/Pro Ala 60 50 40 Ser 30 Glu Gly 20 Tyr Asp 10 0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 2.0 Time (min) (B Asp Tyr Glu Pro Phe Thr Met Ser Iso/Leu Val Ala Gly Arg His Lys Fig (a) Electropherogram of the mixture of 16 heavy labelled AAs (50 μM) used as internal standard (b) Extracted ion chromatogram of the individual heavy labelled amino acids The figure shows the excellent resolution of the analytes in less than minutes of analysis time The microfluidic chip CE-MS approach was next applied to evaluate samples of conditioned cell culture media, however, prior to proceeding with the analysis of a real samples, calibration curves for each amino acid were prepared L-cysteine was observed to be unstable in the cell culture media and was easily oxidized to form cystine in aqueous solution [29,30] Accordingly, it was not included for further study The range of concentrations analysed in this study was overall well above the limit of detection and sensitivity of the instrument; however, internal standard (0.1 μM for all AAs) measurements were used to evaluate the precision of the method, which showed %RSD values