Continuous artificial synthesis of glucose precursor using enzyme-immobilized microfluidic reactors

Food production in green crops is severely limited by low activity and poor specificity of D-ribulose-1,5-bisphosphate carboxylase/oxygenase (RuBisCO) in natural photosynthesis (NPS). This work presents a scientific solution to overcome this problem by immobilizing RuBisCO into a microfluidic reactor, which demonstrates a continuous production of glucose precursor at 13.8 μmol g−1 RuBisCO min−1 from CO2 and ribulose-1,5-bisphosphate. Experiments show that the RuBisCO immobilization significantly enhances enzyme stabilities (7.2 folds in storage stability, 6.7 folds in thermal stability), and also improves the reusability (90.4% activity retained after 5 cycles of reuse and 78.5% after 10 cycles). This work mimics the NPS pathway with scalable microreactors for continuous synthesis of glucose precursor using very small amount of RuBisCO. Although still far from industrial production, this work demonstrates artificial synthesis of basic food materials by replicating the light-independent reactions of NPS, which may hold the key to food crisis relief and future space colonization.

Fourier transform infrared spectroscopy (ATR-FTIR, BRUKER). New peaks (1368 cm -1 and 1564 cm -1 in Raman spectra and 1400-1800 cm -1 and 3000-3750 cm -1 in ATR-FTIR spectra) can be observed after the PDA modification and the RuBisCO/BSA immobilization when compared with the pristine PDMS microfluidic channels. Particularly, the FTIR spectrum of the RIMRs has a weak band at about 1600 cm -1 , which could be referred to the vibrational modes of the peptide bond (amide bands). This absorption band confirms the effective RuBisCO immobilization and gives information about conformational changes induced by the immobilization procedure. Source data are provided as a Source Data file.

Fluorescence experiment
The fluorescence experiment was also conducted to verify the RuBisCO immobilization. The RuBisCO-FITC (RuBisCO tagged by fluorescein isothiocyanate) solution was introduced into the PDA-PDMS microfluidic reactor and then kept at room temperature for 6 hours. The reactor was observed under a fluorescence microscope (Olympus BX41) to record the fluorescence images.

Amplification method for RuBisCO activity assay
For the immobilized RuBisCO activity assay, reactant mixture (66 mM HCO3and 0.5 mM RuBP in the reaction buffer) was passed through the RIMRs and the production solution I (containing RuBisCO, RuBP, HCO3and 3-PGA in the reaction buffer) was collected from the outlet of the reactors for further assay. In terms of the free RuBisCO activity assay, RuBisCO was incubated with reactant mixture for 1 min before the reaction was stopped with the same volume of 80% ethanol. The production solution II (containing RuBisCO, RuBP, HCO3 -, products and ethanol in the reaction buffer) was kept for further assay. The amount of RuBisCO used here was the same as that immobilized into the microfluidic reactors (21.875 μg). 20 μL of the production solutions (i.e., production solution I for immobilized RuBisCO assay and production solution II for free RuBisCO assay) were added with 80 μL of assay mixture (the final concentrations were 5 unit•mL -1 PGK, 0.5 unit•mL -1 GAPDH, 0.5 unit•mL -1 TPI, 0.5 unit•mL -1 G3PDH, 1 unit•mL -1 G3POX, 1000 unit•mL -1 catalase, 0.5 mM ATP, 2 mM NADH, 1.5 mM MgCl2 and 100 mM Tricine/KOH pH 8.0).
The reaction was immediately and continuously monitored by measuring the absorbance change at 340 nm by a UV-Visible spectrometer. During the reaction, the product 3-PGA was first converted to dihydroxyacetone-phosphate (DAP) with PGK, GAPDH, ATP and NADH. Catalase was also added here to prevent the inhibition of GAPDH. Then, DAP was transformed into the cycle of mutual conversion with glycerol-3 phosphate (G3P). It can be monitored as the cumulative oxidation of NADH, whose amount is much larger than the original amount of 3-PGA, therefore

Determination of the reaction time in RIMRs
The reaction time tr is regarded as the residence time of the reaction mixture flowing through the RIMRs, which is calculated by the equation of where Vr is the volume of the RIMRs and Q is the flow rate of the injected RuBP solution controlled by the syringe pump. In this work, the volume of RIMR is 7 μL, the corresponding reaction time is 1 min, 5 min, 7 min, 10 min for the flow rates of 7 μL•min -1 , 1.4 μL•min -1 , 1 μL•min -1 , 0.7 μL•min -1 , respectively. production with the reaction time of 10 min, two tests are conducted to explore the reason of no 3-PGA producing after 2 min-reaction. One test directly uses the production solution I after 10 min for analysis. As shown by dark squares, the production of 3-PGA still does not increase. The other one is to add fresh RuBP into the production solution I after 5 min (final RuBP concentration is 0.5 mM and HCO3is 66 mM) and then to collect the production solution I after another 5 min for analysis (red circles). By contrast, more 3-PGA is produced this time, which shows that the stopped increasing of 3-PGA after 2 min is due to the exhaustion of RuBP. The amount of RuBisCO used is 21.875 μg. Production solution I is collected for 21 μL. Source data are provided as a Source Data file.

HPLC-MS/MS analysis
The monitoring of the reaction was performed using a liquid chromatography-tandem mass