Efficient derivatization-free monitoring of glycosyltransferase reactions via flow injection analysis-mass spectrometry for rapid sugar analytics

The widespread application of enzymes in industrial chemical synthesis requires efficient process control to maintain high yields and purity. Flow injection analysis-electrospray ionization-mass spectrometry (FIA-ESI–MS) offers a promising solution for real-time monitoring of these enzymatic processes, particularly when handling challenging compounds like sugars and glycans, which are difficult to quickly analyze using liquid chromatography-mass spectrometry due to their physical properties or the requirement for a derivatization step beforehand. This study compares the performance of FIA-MS with traditional hydrophilic interaction liquid chromatography (HILIC)-ultra high-performance liquid chromatography (UHPLC)-mass spectrometry (MS) setups for the monitoring of the enzymatic synthesis of N-acetyllactosamine (LacNAc) using beta-1,4-galactosyltransferase. Our results show that FIA-MS, without prior chromatographic separation or derivatization, can quickly generate accurate mass spectrometric data within minutes, contrasting with the lengthy separations required by LC–MS methods. The rapid data acquisition of FIA-MS enables effective real-time monitoring and adjustment of the enzymatic reactions. Furthermore, by eliminating the derivatization step, this method offers the possibility of being directly coupled to a continuously operated reactor, thus providing a rapid on-line methodology for glycan synthesis as well. Graphical Abstract Supplementary Information The online version contains supplementary material available at 10.1007/s00216-024-05457-9.


FIA-MS experiments
For the FIA-MS experiments, a SCIEX M5 Micro LC-TE with Autosampler (PAL 3 CTC) was used.It offers two binary gradient pumping systems (G1 or G2), G1 for low flow rates between 1 µL/min to 10 µL/min and G2 for high-flow rates between 20 µL/min to 200 µL/min.FIA-ESI-MS measurements were performed with 50 µL/min flow rates.For this, G2 was connected to the injection valve instead of G1 to secure a constant isocratic flow at this rate (see Scheme 1).The other connections on the injection valve remain the same as described by SCIEX for FIA-MS measurements.It should be noted that when using only G2, the temperature sensor of the M5 must be moved from G1 to G2.A 5 µL loop was used in full-loop injection mode.Before injection the needle was dipped twice in organic and aqueous solvent and washed thrice.After injection the syringe was washed twice with organic and aqueous solvent respectively.
Scheme S1: Illustration of the configuration of the valves for an FIA-MS experiment on the M5.Since higher flow rates of 50 µL/min were used in the experiment, G2 was connected to port 6 of the injection valve, not G1.

Statistical Anayses
As described in the Methods section, a statistical analysis of the data between FIA-MS and HILIC-UPLC-MS was performed.The calculated values are presented in Table 1-3.The formulas used are listed in the Methods section.

Calibration Curves
Calibrations between 0 -50 ng/mL were prepared for FIA-MS and HILIC Experiments for galactose, lactose and raffinose.
For the kinetic measurement of the beta-1,4-galactosyltransferase, a calibration curve between 0 -400 ng/mL was prepared.Ion signal intensities of fragments of uridine-diphosphate were not reproducible between 0-100 ng/mL.Accordingly, it is reasonable to assume that the limit of detection was reached in this area.
All measurements were carried out trice except for UDP-Gal 350 ng, here the experiment was only carried out twice.Blank measurements were conducted but could not be included in the calibration curve, as only background was measured and the software consequently could not divide the area of the analyte by the area of the standard, thus obtaining no value (see Figure S9) Table S4: Stock solutions were prepared by dissolving the respective analytes in 50:50 ACN/H2O to a concentration of 1 mg/mL.These were further diluted to 1000 ng/mL in the case of the analytes and 10000 ng/mL in the case of the internal standard.
To each calibration mix, 10 µL of internal Standard was added, resulting in 100 ng/mL 13 C-D-Glucose.The key parameters of the calibration curves are listed in the table below.

Figure S2 :Figure S3 :Figure S4 :
Figure S2: EICs for A: 1-13 C glucose, B: lactose, C: raffinose, D: galactose measured with the HILIC-setup.Sample concentrations were at 50 ng/mL (note: Although the LC program used lasted 10 minutes, MS data was collected for 15 minutes.This is an artifact of the software used to control the 6600 MS and the micro-LC M5.).

179.05 503.16 Raffinose counts m/z 179.05 161.04 341.1
TableS2Measurements of lactose using the two methods, FIA-MS and HILIC-UHPLC-MS, in various buffers and dilutions.The measurements were normalized to the expected concentration.The values from the statistical analysis are also displayed.
TableS3Measurements of galactose using the two methods, FIA-MS and HILIC-UHPLC-MS, in various buffers and dilutions.The measurements were normalized to the expected concentration.The values from the statistical analysis are also displayed.

Table S5 :
Summary of the key statistical parameters of the calibration curves.