Supplementary Material

Sample preparation, analytical method and data preprocessing

For the generation of experimental data used to validate the method, culture supernatant was obtained daily from duplicate fed-batch cultures of a Chinese hamster ovary cell line producing a recombinant antibody against the Rhesus D antigen (Chusainow et al., 2009). The cultures were grown in an in-house proprietary protein-free, chemically defined (PFCD) media and online sampling of glutamine/glutamate level was conducted every 1.5 hours to determine the amount of protein-free feed, formulated based on a fortified 10× DMEM/F12 (Hyclone, USA), required to maintain cultures at a pre-set glutamine level of 0.6 mM. The supernatant samples were filtered through a 10 kDa molecular weight cut-off device (Vivaspin 500 PES membrane, Sartorius AG, Germany) by centrifugation at 4oC for 30 min. The filtered samples were diluted 1:1 with sample buffer comprising of 20% (v/v) methanol (Optima grade, Fisher Scientific, USA) in water prior to analysis.

Each sample was analyzed in replicate using an ultra-performance liquid chromatography (UPLC) system (Acquity, Waters Corp., USA) coupled to a mass spectrometer (LTQ-Orbitrap, Thermo Scientific, USA). A reversed phase (C18) UPLC column with polar end-capping (Acquity UPLC HSS T3 column, 2.1 × 100 mm, 1.7 µm, Waters Corp.) was used with two solvents: ‘A’ being water with 0.1% formic acid (Merck, USA), and ‘B’ being methanol (Optima grade, Fisher Scientific) with 0.1% formic acid. The UPLC program was as follows: the column was first equilibrated for 0.5 min at 0.1% B. The gradient was then increased from 0.1% B to 50% B over 8 min before being held at 98% B for 3 min. The column was washed for a further 3 min with 98% acetonitrile (Optima grade, Fisher Scientific) with 0.1% formic acid and finally equilibrated with 0.1% B for 1.5 min. The solvent flow rate was set at 400 µlmin-1; a column temperature of 30oC was used. The eluent from the UPLC system was directed into the mass spectrometer (MS). Electrospray ionization (ESI) was conducted in both positive and negative modes in full scan with a mass range of 80 to 1000 m/z at a resolution of 15000. Sheath and auxiliary gas flow was set at 40.0 and 15.0 (arbitrary units) respectively, with a capillary temperature of 400oC. The ESI source and capillary voltages were 4.5 kV and 40 V respectively, for positive mode ionization, and 3.2 kV and -15 V, respectively, for negative mode ionization. Mass calibration was performed using standard LTQ-Orbitrap calibration solution (Thermo Scientific) prior to injection of the samples, and was found to be less than 3ppm.

The raw LC-MS data obtained was then converted to the generic mzXML format (Pedrioli et al., 2004). Peak detection was then performed using the preprocessing software XCMS (Smith et al., 2006), where the “matchedFilter” algorithm was used with parameters: snthresh = 2(low enough to prevent filtering of otherwise good quality peakgroups), step = 0.05 (step size for RT profiling, mzdiff = 0.1 (minimum m/z diff to distinguish peaks with similar RT), and fwhm = 3 (chromatographic width observed in raw data). XCMS peak-grouping was done using bw=8 (smoothing bandwidth for chromatographic intensity), mzwid=0.05 (m/z width of sliding window for RT-matching, minfrac=1 (consistent presence in this fraction of replicates for at least one sample) for the first round, and, bw=4, mzwid=0.05, minfrac=1 on the second round. The 3D plots of LC-MS data were generated using MSight (Palagi et al., 2005).

References

Chusainow,J. et al. (2009) A study of monoclonal antibody-producing CHO cell lines: what makes a stable high producer? Biotechnol. Bioeng., 102, 1182–1196.

Palagi,P.M. et al. (2005) MSight: an image analysis software for liquid chromatography-mass spectrometry. Proteomics, 5, 2381–2384.

Pedrioli,P.G.A. et al. (2004) A common open representation of mass spectrometry data and its application to proteomics research. Nat. Biotechnol, 22, 1459–1466.

Smith,C.A. et al. (2006) XCMS: processing mass spectrometry data for metabolite profiling using nonlinear peak alignment, matching, and identification. Anal. Chem, 78, 779–787.

Supplementary Table

Supplementary Table S1. List of 32 media metabolites studied.

Media metabolite / Theoretical monoisotopic mass (Da) / Positive modepredicted mass (Da) / Positive mode mass error (ppm) / Negative mode predicted mass (Da) / Negative mode mass error (ppm)
L-Alanine / 89.0477 / 89.0473 / -4.71 / - / -
L-Valine / 117.0790 / 117.0785 / -3.88 / - / -
Niacinamide / 122.0480 / 122.0477 / -2.61 / - / -
L-Asparagine / 132.0535 / 132.0532 / -2.61 / - / -
L-Cystine / 240.0238 / 240.0238 / 0.00 / - / -
Vitamin B12 / 1354.5674 / 1354.5660 / -1.00 / - / -
L-Proline / 115.0633 / 115.0628 / -4.01 / 115.0641 / 7.04
L-Isoleucine / 131.0946 / 131.0941 / -4.11 / 131.0956 / 7.71
L-Leucine / 131.0946 / 131.0941 / -4.11 / 131.0957 / 8.06
Hypoxanthine / 136.0385 / 136.0382 / -1.89 / 136.0395 / 6.99
L-Glutamine / 146.0691 / 146.0689 / -1.31 / 146.0701 / 6.91
L-Methionine / 149.0510 / 149.0508 / -1.61 / 149.0519 / 6.15
L-Phenylalanine / 165.0790 / 165.0783 / -3.99 / 165.0799 / 5.41
L-Arginine / 174.1117 / 174.1114 / -1.92 / 174.1126 / 5.25
L-Tyrosine / 181.0739 / 181.0733 / -3.07 / 181.0750 / 6.04
L-Tryptophan / 204.0899 / 204.0894 / -2.24 / 204.0911 / 5.97
Pantothenate / 219.1107 / 219.1106 / -0.42 / 219.1117 / 4.45
Riboflavin / 376.1383 / 376.1382 / -0.38 / 376.1395 / 3.29
Folic Acid / 441.1397 / 441.1396 / -0.12 / 441.1419 / 4.99
L-Serine / 105.0426 / - / - / 105.0436 / 9.86
L-Threonine / 119.0582 / - / - / 119.0592 / 8.81
L-Cysteine / 121.0197 / - / - / 121.0208 / 9.00
L-Aspartic Acid / 133.0375 / - / - / 133.0386 / 8.41
L-Lysine / 146.1055 / - / - / 146.1065 / 6.68
L-Glutamic Acid / 147.0532 / - / - / 147.0542 / 6.50
L-Histidine / 155.0695 / - / - / 155.0704 / 5.84
Myo-Inositol / 180.0634 / - / - / 180.0635 / 0.51
Citrate / 192.0270 / - / - / 192.0280 / 5.07
Choline / 104.1075 / Only [M]1+ ion detected in the positive mode
D-Glucose / 180.0634 / Only sodium adduct, [M+Na]1+,detected in the positive mode
Thymidine / 242.0903 / Only sodium adduct found in the positive mode raw data
Thiamine / 265.1123 / Only [M]1+ iondetected in the positive mode raw data

Footnote.

Supplementary Figures

Supplementary Fig. S1. Visualization of LC-MS data. (a) Chromatogram of base (most intense) peaks (top panel) and spectrum at a single retention time (RT) point (bottom panel). The mass spectrometer scans the eluting analyte repeatedly to give a spectrum at different RT. (b) 2D density map of the entire run on the left, and on the right, a 3D plot (RT vs m/z vs intensity) of a selected map region. The lines on the density map represent peaks and the darker the line, the greater the intensity of the peak signal.

Supplementary Fig. S2. Example of IP clustering. The figure shows the density maps for four different runs, along with 3D plots of the regions marked by the red dotted boxes. Three peak-groups are being considered for this example (PG1-PG3). The algorithm first clusters peak-groups in the RT domain by comparing the chromatographic peak profiles within individual runs. From the 3D plots, it appears that the peak shapes are similar in all four runs and are located at similar RT. The algorithm then examines the intensity ratios between pairs of peaks. The intensity ratio between peaks of PG1 and PG3 in run 4 appears to be very different from the rest of the runs, thus PG3 is separated from the cluster of PG1 and PG2.

Supplementary Fig. S3. Comparison with peak-groups without K-means clustering. (a) Example of m/z matching for two features where XCMS generated a single peak-group while we generated two peak-groups with different retention times (pg1 and pg2). The 3D plot of the data from one run clearly shows two distinct peaks with the same m/z but different retention times. The 2D plot of the region shows the locations of peaks from all runs, and how they were grouped in XCMS and our workflow. We separated the peaks into two peak-groups with 118 peaks each (peaks from all runs, except from the blank run, were present and represented only once in each of the peak-groups). These two peak-groups were later found to correspond to fragment ions of the isomers L-Leucine and L-Isoleucine. (b) Example of m/z matching for two features where XCMS generated a single peak-group with an m/z that is different from both of the peak-groups generated by us. The 3D plot shows two distinct peaks, with the more intense peak having a slightly larger m/z value. The inclusion of extra peaks from the same run in the XCMS peak-group skews its m/z (taken to be the median m/z of constituent peaks), such that the value is between the m/z of our two peak-groups.