TCA cycle-powered synthesis of fucosylated oligosaccharides

TCA cycle-powered synthesis of fucosylated oligosaccharides Abstract Microbial catalysis has recently emerged as one of the most promising approaches in oligosaccharide synthesis. However, despite significant progress, microbial synthesis still requires much improvement in efficiency and in reduction of process complexity. Additionally, given the stunning diversity and many varied applications of glycans, broadening the range of glycans accessible via microbial synthesis is of paramount importance. Major challenges in microbial synthesis include catabolite repression and high cellular energy requirement. Here we demonstrated a new approach to overcome these challenges by directly tapping into the cellular “power house,” the TCA cycle, to provide the cellular energy for synthesis. This approach not only circumvents catabolite repression but also eliminates acidic glycolysis by-products. As such, the whole-cell biocatalysis can be carried out without sophisticated fed-batch feeding and pH control in the synthesis stage. The system could achieve several grams per liter (3–4 g/L) within a 24-h period in shaker flask cultivation for two targets, fucosyllactose and fucosyllactulose, demonstrating efficiency of the biocatalyst developed and its applicability to both natural and non-natural targets. To the best of our knowledge, this is the first use of TCA cycle intermediates as the energy source for oligosaccharide synthesis and the first description of successful synthesis of fucosyllactulose with titers in several grams per liter. 2′ fucosyllactose, fucosylated oligosaccharide, lactulose, metabolic engineering, whole-cell biocatalysis Introduction Oligosaccharides, often conjugated with surface proteins or lipids, serve as molecular recognition elements in a wide variety of biological processes including many disease-causing events (Varki and Lowe 2009; Kobata 2010). In human milk, soluble oligosaccharides, with about 200 distinct structures (Davis et al. 2016), form the third most abundant component in human milk (after lactose and lipid). Increasing evidence shows that they are anti-infective (Kunz et al. 2000), prebiotic (Bode and Jantscher-Krenn 2012), and immune-stimulatory (Newburg 2013). Fucosylated oligosaccharides are a subset of oligosaccharides where fucose is alpha-linked to either galactose or N-acetyl-glucosamine (GlcNAc). The ability to synthesize these molecules and their derivatives is of interest to many scientific fields and for potential medical and diagnostic applications. Unfortunately, while important progress has been made recently, relative to the vast diversity in glycans existing in nature, significant innovation has to be made in order to make a large number of oligosaccharides available in quantities sufficient for structure–activity relationship studies and for various food, medical and other commercial interests. Microbial synthesis has recently emerged as one of the most effective approaches for the synthesis of oligosaccharides (Han et al. 2012). For several hotly pursued molecules such as 2-fucosyl-lactose (2′-FL) and 3′-sialyl-lactose (3′-SL), fermentation processes based on a single metabolically engineered strain are promising (Priem et al. 2002; Lee et al. 2012; Chin et al. 2015). However, in general, using microbial biocatalyst for oligosaccharide synthesis is still quite challenging (Ruffing and Chen 2006; Chen 2015). A major challenge in engineering single-strain biocatalyst stems from a universal metabolic regulation known as catabolite repression. In the synthesis stage, it is necessary for the whole-cell biocatalyst to engage multiple sugars, including both acceptor sugars and donor sugars. In addition, the cells need an energy source (such as glucose) to replenish cellular energy (in the form of adenosine triphosphate (ATP), guanosine triphosphate (GTP), etc.) for the synthesis to continue. While this conflict can be resolved by supplying glycerol, this strategy leads to a relatively low productivity and significant amount of acidic metabolic by-products. In this study, we describe an approach to directly tap into the cellular “power house,” the TCA cycle, to produce energy needed for the synthesis and to sustain biocatalysis. This approach not only circumvents catabolite repression but also eliminates acidic glycolysis by-products. As such, the whole-cell biocatalysis can be carried out without sophisticated fed-batch feeding and pH control in the synthesis stage. We first demonstrate the concept with 2′-FL synthesis in a shake flask process. Subsequently, we extend the biocatalysis process to a novel oligosaccharide molecule, fucosyllactulose, demonstrating the applicability of the biocatalyst developed to a non-natural target. To the best of our knowledge, this is the first use of TCA cycle intermediate as an alternative energy source to power the oligosaccharide synthesis in whole-cell biocatalysis. Results and discussion Construction of whole-cell biocatalyst for fucosylated oligosaccharides synthesis Our initial biocatalyst design features a salvage pathway for GDP-fucose synthesis through a plasmid-borne bifunctional enzyme fucosekinase/fucose-1-phosphate guanylyltransferase (FKP) from Bacteroides fragilis, which catalyzes the synthesis of GDP-fucose from exogenous fucose, ATP and GTP from cellular metabolism (Figure 1A). The recombinant plasmid (Figure 1B) was constructed to co-overexpress FKP and a Helicobacter pylori fucosyltransferase (FutC) from a T5 promoter. The futC gene was cloned in two separate pieces and modification of sequence was introduced to avoid difficulties in polymerase chain reaction (PCR) amplification and expression (Wang et al. 1999). The reconstructed gene was designated as an futCR. The resulting protein, however, is identical to GenBank ARB16053.1. As shown in Figure 2A, the resulting biocatalyst, Escherichia coli JM109/pQFKPfutCR was able to achieve a respectable titer of 0.74 g/L in a 24-h reaction from fucose and lactose using shake flask culture. Figure 1. View largeDownload slide (A) Engineered metabolic pathway for 2′-FL synthesis highlighting GDP-fucose synthesis via the salvage pathway, inactivation of fucose metabolism and TCA cycle for energy provision. (B) Recombinant plasmid for overexpression of FKP and FutCR. Figure 1. View largeDownload slide (A) Engineered metabolic pathway for 2′-FL synthesis highlighting GDP-fucose synthesis via the salvage pathway, inactivation of fucose metabolism and TCA cycle for energy provision. (B) Recombinant plasmid for overexpression of FKP and FutCR. Figure 2. View largeDownload slide Time course of 2′-FL synthesis. (A) Comparison between E. coli strains JCTFKP/pQFKPfutCR and JM109/pQFKPfutCR. (B) Comparison of three different TCA cycle intermediates as energy sources, α-KG (square), citrate (circle) and succinate (triangle). (C) Comparison of α-KG with glycerol and glucose as energy sources, α-KG (square), glycerol (circle) and glucose (triangle). (D) pH profiles during 2′FL synthesis with different energy sources. Results shown in this study are the mean of triplicated experiments and error bars indicate the standard deviation. Figure 2. View largeDownload slide Time course of 2′-FL synthesis. (A) Comparison between E. coli strains JCTFKP/pQFKPfutCR and JM109/pQFKPfutCR. (B) Comparison of three different TCA cycle intermediates as energy sources, α-KG (square), citrate (circle) and succinate (triangle). (C) Comparison of α-KG with glycerol and glucose as energy sources, α-KG (square), glycerol (circle) and glucose (triangle). (D) pH profiles during 2′FL synthesis with different energy sources. Results shown in this study are the mean of triplicated experiments and error bars indicate the standard deviation. To improve the synthesis and decrease the metabolism of fucose, we decided to knockout the genes associated with the first two steps of fucose metabolism. Specifically, the genes encoding fucose isomerase and fuculose kinase were knocked out (Figure 1A). As a result, the synthesis was significantly increased with the highest titer of 3.3 g/L reached at 24 h. This is a 4.7-fold increase compared to the control, JM109/pQFKPfutCR. Additionally, the molar yield from fucose was increased by 4-fold from 0.14 to 0.56 mol/mol. As shown in Figure 1A, knockout of fucI and fucK disrupts the metabolism of fucose via TCA cycle. This disruption is responsible for increased yield from fucose, suggesting that elimination of nonreaction-related metabolism of fucose increases the availability of fucose for GDP-fucose synthesis (Figure 1A). TCA cycle for energy provision for sustained synthesis Oligosaccharide synthesis is an energy-intensive process. Each glycosidic bond formation requires at least two high-energy compounds (two ATP equivalents) (Ruffing and Chen 2011). In the case of 2′-FL synthesis, the formation of donor sugar (GDP-fucose) requires one ATP and one GTP. There may be other energy-requiring steps associated with the reaction, such as substrate transport. Thus, to sustain the synthesis reaction over a long period, an energy source has to be provided. Due to the catabolite repression, glucose is not used. Instead, glycerol was almost exclusively used (Baumgärtner et al. 2013; Chin et al. 2016). However, glycerol metabolism is slow and to minimize acid production and accumulation, a fed-batch process is generally required, increasing the complexity of the process. We reasoned that a TCA cycle could be used to provide the necessary energy for the synthesis. Directly tapping the TCA cycle, without having to go through glycolysis, may have several important advantages. Besides avoiding the glucose repression, TCA cycle also produces GTP directly via substrate-level phosphorylation (Figure 1A), thus providing a direct link between energy production and target synthesis. In addition, no acidic by-products will be produced as only CO2 and H2O are the respiration end products. This eliminates the need for pH control and simplifies the downstream processing. We tested first with α-ketoglutarate (α-KG) for 2′-FL synthesis. As shown in Figure 2B, the respiring culture rapidly accumulated 2′-FL over the period of 20–24 h, reaching the highest 2′-FL concentration at 20 h, indicating α-KG is effective as an energy source to support the synthesis reaction (Figure 2B). We subsequently tested other TCA cycle substrates, citrate and succinate, for their ability to support synthesis. Both citrate and succinate are commercial fermentation products. As shown in Figure 2B, both could support the synthesis, but reaching lower final concentration of 2′-FL under the condition tested. Additional studies are needed to identify the best TCA intermediate and optimize the feeding regime to maximize the titer of synthesis targets. When using α-KG as an energy source, 2′-FL production reached a titer of 3.3 g/L, yield from cells of 0.21 g/g-dry weight cells, yield from fucose 0.56 mol-2′-FL/mol-fucose, and yield from lactose, 0.80 mol-2′-FL/mol-lactose. The productivity is 0.14 g/L/h (Table I). Fucose yield is particularly important as it impacts the cost most. Compared to the 0.36 molar yield, reported by Chin et al. (2016), the yield of 0.56 molar yield is a significant improvement (55.6% increase). The lower productivity, 0.14 g/L/h, compared to the highest reported value (0.57 g/L/h) reported by Baumgärtner et al. (2013), could be attributed to the much lower cell concentrations in our experiments. In our experiments, the cell dry weight was about 16 g/L, whereas the fed-batch processes in a bioreactor by Baumgärtner et al. (2013) was much higher, with the highest at 60 g/L. This relatively lower productivity does not represent the intrinsic limitation of our biocatalyst, rather this is a reflection of limitation of shaker flask methods used in this study, which can only support aerobic metabolism with limited cell concentrations. Similarly, the lower titer reported here could also be attributed to the much lower cell concentration used in our study and is a limitation of shaker flask method, not the biocatalyst. This limitation can be overcome by increasing cell density in a bioreactor, where adequate aeration supports full respiration as required by the synthesis. We envision that our biocatalyst could be first grown using fed-batch cultivation in a bioreactor, as commonly done today, and induced for recombinant enzyme production at a suitable optical density, and then continued to grow to a final cell density to 60 g/L or higher, in a manner similar to what was demonstrated in prior studies (Baumgärtner et al. 2013 or Chin et al. 2016). At this point, the synthesis is initiated by the addition of fucose, lactose, α-KG or another suitable TCA cycle intermediate. Since the rate of synthesis is directly proportional to the cell density, we should see a significant increase in space-time productivity as well as in product titer. In this way, the synthesis is separated from cell growth and recombinant enzyme production phase. Two different energy sources, glycerol and a TCA cycle intermediate, could be used in the growth phase and synthesis, respectively, to achieve better process performance metrics. Table I. Fucosyl-trisaccharide production by JM109/pQFKPfutCR and JCTFKP/pQFKPfutCRa Strain JM109/pQFKPfutCR JCTFKP/pQFKPfutCR Product Fucosyllactose Fucosyllactose Fucosyllactulose Trisaccharide concentration (g/L) 0.74 ± 0.036 3.28 ± 0.032 4.34 ± 0.183 Fucose consumed (g/L) 1.73 ± 0.110 1.98 ± 0.167 1.70 ± 0.110 Acceptor sugar consumed (g/L) 1.14 ± 0.047 2.87 ± 0.090 3.77 ± 0.008 YDCWb (g/g) 0.046 ± 0.002 0.205 ± 0.002 0.277 ± 0.011 Yfucosec (mol/mol) 0.144 ± 0.011 0.558 ± 0.050 0.860 ± 0.056 Yacceptor sugard (mol/mol) 0.455 ± 0.057 0.801 ± 0.034 0.807 ± 0.036 Productivity (g/L/h) 0.031 ± 0.002 0.137 ± 0.004 0.181 ± 0.008 Strain JM109/pQFKPfutCR JCTFKP/pQFKPfutCR Product Fucosyllactose Fucosyllactose Fucosyllactulose Trisaccharide concentration (g/L) 0.74 ± 0.036 3.28 ± 0.032 4.34 ± 0.183 Fucose consumed (g/L) 1.73 ± 0.110 1.98 ± 0.167 1.70 ± 0.110 Acceptor sugar consumed (g/L) 1.14 ± 0.047 2.87 ± 0.090 3.77 ± 0.008 YDCWb (g/g) 0.046 ± 0.002 0.205 ± 0.002 0.277 ± 0.011 Yfucosec (mol/mol) 0.144 ± 0.011 0.558 ± 0.050 0.860 ± 0.056 Yacceptor sugard (mol/mol) 0.455 ± 0.057 0.801 ± 0.034 0.807 ± 0.036 Productivity (g/L/h) 0.031 ± 0.002 0.137 ± 0.004 0.181 ± 0.008 aα-KG was used as an energy source for fucosyl-trisaccharide production. bYield to cells (YDCW) = concentration of product/DCW. cYield to fucose (Yfucose) = concentration of product/concentration of fucose consumed. dYield to acceptor sugar (Yacceptor sugar) = concentration of product/concentration of acceptor sugar (lactose/lactulose) consumed. Table I. Fucosyl-trisaccharide production by JM109/pQFKPfutCR and JCTFKP/pQFKPfutCRa Strain JM109/pQFKPfutCR JCTFKP/pQFKPfutCR Product Fucosyllactose Fucosyllactose Fucosyllactulose Trisaccharide concentration (g/L) 0.74 ± 0.036 3.28 ± 0.032 4.34 ± 0.183 Fucose consumed (g/L) 1.73 ± 0.110 1.98 ± 0.167 1.70 ± 0.110 Acceptor sugar consumed (g/L) 1.14 ± 0.047 2.87 ± 0.090 3.77 ± 0.008 YDCWb (g/g) 0.046 ± 0.002 0.205 ± 0.002 0.277 ± 0.011 Yfucosec (mol/mol) 0.144 ± 0.011 0.558 ± 0.050 0.860 ± 0.056 Yacceptor sugard (mol/mol) 0.455 ± 0.057 0.801 ± 0.034 0.807 ± 0.036 Productivity (g/L/h) 0.031 ± 0.002 0.137 ± 0.004 0.181 ± 0.008 Strain JM109/pQFKPfutCR JCTFKP/pQFKPfutCR Product Fucosyllactose Fucosyllactose Fucosyllactulose Trisaccharide concentration (g/L) 0.74 ± 0.036 3.28 ± 0.032 4.34 ± 0.183 Fucose consumed (g/L) 1.73 ± 0.110 1.98 ± 0.167 1.70 ± 0.110 Acceptor sugar consumed (g/L) 1.14 ± 0.047 2.87 ± 0.090 3.77 ± 0.008 YDCWb (g/g) 0.046 ± 0.002 0.205 ± 0.002 0.277 ± 0.011 Yfucosec (mol/mol) 0.144 ± 0.011 0.558 ± 0.050 0.860 ± 0.056 Yacceptor sugard (mol/mol) 0.455 ± 0.057 0.801 ± 0.034 0.807 ± 0.036 Productivity (g/L/h) 0.031 ± 0.002 0.137 ± 0.004 0.181 ± 0.008 aα-KG was used as an energy source for fucosyl-trisaccharide production. bYield to cells (YDCW) = concentration of product/DCW. cYield to fucose (Yfucose) = concentration of product/concentration of fucose consumed. dYield to acceptor sugar (Yacceptor sugar) = concentration of product/concentration of acceptor sugar (lactose/lactulose) consumed. To illustrate the advantage of using α-KG as an alternative energy source to support synthesis, additional experiments were conducted to compare three different energy sources (α-KG, glycerol and glucose) for 2′FL synthesis. As shown in Figure 2C, α-KG is far better than glycerol and the resulting titer is several folds higher than that of glycerol. Using glucose as an energy source resulted in no product synthesis. This can be explained by the catabolite repression, which prevents the uptake of either lactose or fucose. During the synthesis, the pH of the reaction medium was monitored. When α-KG is used as an energy source, the pH change was minimal with only a slight uptick to 7.7 at the end of a 24 -h reaction (Figure 2D). In contrast, for both glucose and glycerol, rapid acidification of reaction mixture was observed. The final pHs for glucose and glycerol were 3.9 and 4.4, respectively, which were far out of the range of pH supporting optimal carbon metabolism. The use of TCA cycle intermediate as an energy source is a unique metabolic state, where there is no glycolysis yet TCA cycle is fully functional. This contrasts with a metabolic state when glucose or glycerol is used, where both glycolysis and TCA cycle are active. Glycolysis intermediates, such as pyruvate, can feed into the TCA cycle or alternatively metabolize to generate mixed acids dominated by acetic acid and lactate. Pyruvate and other acids are possibly responsible for the rapid acidification of the medium as noted in Figure 2D. Therefore, besides being a better energy source, α-KG reduces the requirement of pH monitoring and adjustment during the synthesis phase. This advantage translates to less salts produced in a bioreactor and makes downstream processing much easier. Additionally, without the requirement of pH monitoring/adjustment, it makes whole-cell biocatalysts much more conducive to be used in shaker flask for various new targets that have not yet been mature enough for bioreactor operation. Thus, this approach could potentially accelerate laboratory research for the synthesis of new glycan targets. Synthesis of novel fucosylated trisaccharide To test whether our fucosylation biocatalyst could accept other substrates for novel structures, we carried out reactions under the otherwise identical conditions (as described above) except we supplied the respiring culture with lactulose as an acceptor sugar substrate. Lactulose is a disaccharide of galactose and fructose. It is similar to lactose but with fructose in the position of glucose. Figure 3 shows that the fucosylation biocatalyst fucosylated lactulose efficiently. In 24 h, the respiring culture resulted in 4.3 g/L fucosyllactulose. Fucosyllactulose synthesized by the whole-cell biocatalyst was adsorbed on an activated carbon column. After exhaustive washing with water, the trisaccharide was eluted with 40% of ethanol. The purified trisaccharide was analyzed by GC–MS at the Complex Carbohydrate Research Center and fucosylation was confirmed (Supplementary data). Lactulose is nondigestible and prebiotic. Its use has been associated with a number of beneficial effects, such as improved calcium absorption (Kuschel et al. 2017), inhibition of pathogenic bacteria while promoting growth of bifddobacteria and lactobacilli in the gastrointestinal tract (Wu et al. 2017). As the antiadhesion effect of 2′FL is associated with the terminal fucosyl linkage (Castanys-Munoz et al. 2013), it is speculated that successful fucosylation of lactulose adds a potential antiadhesive function to this already useful prebiotic molecule. To definitively prove this function, it requires significant amount of the new molecule. This study provided a viable approach for its synthesis. Figure 3. View largeDownload slide Synthesis of fucosylated lactulose by E. coli JCTFKP/pQFKPfutCR with α-KG as an energy source. Figure 3. View largeDownload slide Synthesis of fucosylated lactulose by E. coli JCTFKP/pQFKPfutCR with α-KG as an energy source. In summary, we have constructed an efficient E. coli whole-cell fucosylation biocatalyst and developed a versatile biocatalysis process for not only 2′-FL but also other related structures. We demonstrated that the developed system could achieve a titer of several grams per liter within a 24-h period by shaker flask cultivation without sophisticated control (such as energy source feeding, pH and oxygen level). It is conceivable that the system can be extended to other fucosyl oligosaccharides with only minor changes in the requisite fucosyltransfereases. Other Human Milk Oligosaccharides-fucosylated oligosaccharides such as 3′-fucosyllactose, lacto-N-fucopentaose I (FPLNT) are potential targets of the system. We anticipate that the concept of TCA-powered synthesis could also be applied to nonfucosylated oligosaccharides. Materials and methods Chemicals All chemicals were purchased from Sigma-Aldrich Chemical Co. (St. Louis, MO). DNA polymerase and restriction enzymes were purchased from New England Biolabs (Ipswich, MA). Bacterial strains and plasmids Escherichia coli JM109 was used as the host for gene cloning and expression. The primers used in this study were listed in Supplementary data, Table SI. The gene fkp was amplified from B. fragilis NCTC 9343 using primers FKP-F and FKP-R, digested by BamHI and SalI, and ligated to pQE80L to form pQFKP. To avoid mis- or no-amplification of the original futC gene caused by polyC sequence, two DNA fragments before and after the polyC repeats were first obtained from the H. pylori 25695 genomic DNA using primers futC1/futC2 and futC3/futC4. The complete polyC-modified futC gene (futCR) was obtained by overlapped PCR of the two fragments (Huang et al. 2017) using futCR-F and futCR-R, digested with SalI and PstI and cloned into pQFKP to constructed plasmid pQFKPfutCR coexpressing fkp and futCR by a T5 promoter (Figure 1B). To inactivate fucose-metabolizing pathway in E. coli JM109, T5-derived fkp was inserted between fucP (L-fucose transporter) and fucU (L-fucose mutarotase), while fucI (L-fucose isomerase) and fucK (L-fuculose kinase) were deleted at the same time through homologous recombination (Datsenko and Wanner 2000). Upstream DNA of fucI and downstream DNA of fucK were amplified by primers fucUP1/fucUP2 and fucDn1/fucDn2, respectively, and fused by overlapped PCR. The expression cassette of T5-fkp was amplified by T5-F and FKP-R from pQFKP. Chloramphenicol resistance gene (CmR DNA with FRT sequence at both ends) was amplified from pKD3 by CmR-F and CmR-R. The three DNA fragments were digested with BamHI and XhoI; BglII and SalI; and SalI and XhoI, respectively, and ligated to form upstream-T5FKP-CmR-downstream DNA fragment. The fragment was transformed to E. coli JM109 to replace fucI and fucK genes with T5FKP-CmR DNA (Figure 1A). pCP20 was then introduced to remove FRT-CmR gene, and E. coli JCTFKP strain with augmented FKP activity and inactivated fucose-metabolizing activity by the deletion of fucIK genes was obtained. Protein expression and fucosylated trisaccharide synthesis The expression conditions of exogenous proteins in engineered strains including IPTG concentration, induction temperature, and initial cell density and induction time were optimized. Overnight preinoculate was prepared in LB medium at 37°C and transferred to fresh LB at an inoculum concentration of 1% (v/v) and grown to OD600 = 0.6. Expression of the recombinant proteins was induced by the addition of 1 mM IPTG. The culture was incubated at 30°C for 12 h. Cells were harvested by centrifugation (4000 × g, 10 min, 4°C) and resuspended in the reaction mixture containing 80 mM Tris-HCl (pH 7.5), 11.9 mM PBS (pH 7.5), 20 mM MgCl2, 5 mM MnCl2, 5 g/L fucose, 5 g/L lactose and 5 g/L TCA cycle intermediate (α-KG or citrate or succinate). The reaction was carried out at 37°C and shaked for 24 h at 250 rpm, and additional dose of TCA cycle intermediate was added to a final concentration of 5 g/L α-KG at 4, 8 and 12 h after sampling. For the comparison of α-KG, glycerol and glucose, the above described synthesis procedure was followed, except glycerol and glucose were used in the respective experiments. pH was measured with pH indicator paper with accuracy to 0.1 unit. The initial and final pHs were additionally measured by a pH meter. Analytic methods Saccharides were detected by HPLC (Agilent 1100 (Santa Clara, CA)) equipped with an Aminex HPX-87H column (300 × 7.8 mm, Bio-Rad (Hercules, CA)) and a refractive index detector. The samples were boiled for 10 min in a water bath and centrifuged at 13,000 rpm for 5 min. The supernatant was filtered with a filtration membrane (0.22 μm pore size) and 10 μL was injected into the column. H2SO4 (5 mM) was used as the mobile phase at a flow rate of 0.3 mL/min. Optical cell density was measured at 600 nm absorbance using a spectrophotometer. The dry cell weight (DCW) was calculated as follows: DCW (g/L) = 0.32 × OD600. Supplementary data Supplementary data are available at GLYCOBIOLOGY online. Funding This research is supported by an NSF (grant CBET1509202 to R.C.). Conflict of interest statement The authors declare no conflict of interest. Abbreviations 2′-FL 2-fucosyl-lactose 3′-SL 3′-sialyl-lactose FKP fucosekinase/fucose-1-phosphate guanylyltransferase FutC fucosyltransferase α-KG α-ketoglutarate DCW dry cell weight ATP adenosine triphosphate GTP guanosine triphosphate. References Baumgärtner F , Seitz L , Sprenger GA , Albermann C . 2013 . Construction of Escherichia coli strains with chromosomally integrated expression cassettes for the synthesis of 2′-fucosyllactose . Microb Cell Fact . 12 ( 1 ): 40 . Google Scholar CrossRef Search ADS PubMed Bode L , Jantscher-Krenn E . 2012 . Structure-function relationships of human milk oligosaccharides . Adv Nutr . 3 ( 3 ): 383S – 391S . Google Scholar CrossRef Search ADS PubMed Castanys-Munoz E , Martin MJ , Prieto PA . 2013 . 2′-fucosyllactose: an abundant, genetically determined soluble glycan present in human milk . Nutr Rev . 71 ( 12 ): 773 – 789 . Google Scholar CrossRef Search ADS PubMed Chen R . 2015 . The sweet branch of metabolic engineering: cherry-picking the low-hanging sugary fruits . Microb Cell Fact . 14 : 197 . Google Scholar CrossRef Search ADS PubMed Chin YW , Kim JY , Lee WH , Seo J-H . 2015 . Enhanced production of 2′-fucosyllactose in engineered Escherichia coli BL21star (DE3) by modulation of lactose metabolism and fucosyltransferase . J Biotechnol . 210 : 107 – 115 . Google Scholar CrossRef Search ADS PubMed Chin YW , Seo N , Kim JH , Seo JH . 2016 . Metabolic engineering of Escherichia coli to produce 2′-fucosyllactose via salvage pathway of guanosine 5′-diphosphate (GDP)-l-fucose . Biotechnol Bioeng . 113 ( 11 ): 2443 – 2452 . Google Scholar CrossRef Search ADS PubMed Datsenko KA , Wanner BL . 2000 . One-step inactivation of chromosomal genes in Escherichia coli K-12 using PCR products . Proc Natl Acad Sci USA . 97 ( 12 ): 6640 – 6645 . Google Scholar CrossRef Search ADS PubMed Davis JC , Totten SM , Huang JO , Nagshbandi S , Kirmiz N , Garrido DA , Lewis ZT , Wu LD , Smilowitz JT , German JB . 2016 . Identification of oligosaccharides in feces of breast-fed infants and their correlation with the gut microbial community . Mol Cell Proteomics . 15 ( 9 ): 2987 – 3002 . Google Scholar CrossRef Search ADS PubMed Han NS , Kim TJ , Park YC , Kim J , Seo JH . 2012 . Biotechnological production of human milk oligosaccharides . Biotechnol Adv . 30 ( 6 ): 1268 – 1278 . Google Scholar CrossRef Search ADS PubMed Huang D , Yang K , Liu J , Xu Y , Wang Y , Wang R , Liu B , Feng L . 2017 . Metabolic engineering of Escherichia coli for the production of 2′-fucosyllactose and 3-fucosyllactose through modular pathway enhancement . Metab Eng . 41 : 23 – 38 . Google Scholar CrossRef Search ADS PubMed Kobata A . 2010 . Structures and application of oligosaccharides in human milk . Proc Jpn Acad Ser B . 86 ( 7 ): 731 – 747 . Google Scholar CrossRef Search ADS Kunz C , Rudloff S , Baier W , Klein N , Strobel S . 2000 . Oligosaccharides in human milk: structural, functional, and metabolic aspects . Ann Rev Nutr . 20 ( 1 ): 699 – 722 . Google Scholar CrossRef Search ADS Kuschel B , Seitl I , Gluck C , Mu W , Jiang B , Stressler T , Fischer L . 2017 . Hidden reaction: Mesophilic cellobiose 2-epimerases produce lactulose . J Agric Food Chem . 65 : 2530 – 2539 . Google Scholar CrossRef Search ADS PubMed Lee W-H , Pathanibul P , Quarterman J , Jo J-H , Han NS , Miller MJ , Jin J-S , Seo J-H . 2012 . Whole cell biosynthesis of a functional oligosaccharide, 2′-fucosyllactose, using engineered Escherichia coli . Microb Cell Fact . 11 : 48 . Google Scholar CrossRef Search ADS PubMed Newburg D . 2013 . Glycobiology of human milk . Biochemistry (Mosc) . 78 ( 7 ): 771 – 785 . Google Scholar CrossRef Search ADS PubMed Priem B , Gilbert M , Wakarchuk WW , Heyraud A , Samain E . 2002 . A new fermentation process allows large-scale production of human milk oligosaccharides by metabolically engineered bacteria . Glycobiology . 12 ( 4 ): 235 – 240 . Google Scholar CrossRef Search ADS PubMed Ruffing A , Chen RR . 2006 . Metabolic engineering of microbes for oligosaccharide and polysaccharide synthesis . Microb Cell Fact . 5 ( 1 ): 25 . Google Scholar CrossRef Search ADS PubMed Ruffing AM , Chen RR . 2011 . Citrate stimulates oligosaccharide synthesis in metabolically engineered Agrobacterium sp . Appl Biochem Biotech . 164 ( 6 ): 851 – 866 . Google Scholar CrossRef Search ADS Varki A , Lowe JB . 2009 . Biological roles of glycans . In: Varki A , Cummings RD , Esko JD , Freeze HH , Stanley P , Bertozzi CR , Hart GW , Etzler ME , editors . Essentials of Glycobiology . New York : Cold Spring Harbor . p. 75 – 88 . Wang G , Rasko DA , Sherburne R , Taylor DE . 1999 . Molecular genetic basis for the variable expression of Lewis Y antigen in Helicobacter pylori: Analysis of the α (1, 2) fucosyltransferase gene . Mol Microbiol . 31 ( 4 ): 1265 – 1274 . Google Scholar CrossRef Search ADS PubMed Wu L , Xu C , Li S , Liang J , Xu H , Xu Z . 2017 . Efficient production of lactulose from whey powder by cellobiose-2-epimerase in an enzymatic membrane reactor . Bioresour Technol . 233 : 305 – 312 . Google Scholar CrossRef Search ADS PubMed © The Author(s) 2018. Published by Oxford University Press. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Glycobiology Oxford University Press

TCA cycle-powered synthesis of fucosylated oligosaccharides

Loading next page...
 
/lp/ou_press/tca-cycle-powered-synthesis-of-fucosylated-oligosaccharides-TzXgoHgLhB
Publisher
Oxford University Press
Copyright
© The Author(s) 2018. Published by Oxford University Press. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com
ISSN
0959-6658
eISSN
1460-2423
D.O.I.
10.1093/glycob/cwy047
Publisher site
See Article on Publisher Site

Abstract

Abstract Microbial catalysis has recently emerged as one of the most promising approaches in oligosaccharide synthesis. However, despite significant progress, microbial synthesis still requires much improvement in efficiency and in reduction of process complexity. Additionally, given the stunning diversity and many varied applications of glycans, broadening the range of glycans accessible via microbial synthesis is of paramount importance. Major challenges in microbial synthesis include catabolite repression and high cellular energy requirement. Here we demonstrated a new approach to overcome these challenges by directly tapping into the cellular “power house,” the TCA cycle, to provide the cellular energy for synthesis. This approach not only circumvents catabolite repression but also eliminates acidic glycolysis by-products. As such, the whole-cell biocatalysis can be carried out without sophisticated fed-batch feeding and pH control in the synthesis stage. The system could achieve several grams per liter (3–4 g/L) within a 24-h period in shaker flask cultivation for two targets, fucosyllactose and fucosyllactulose, demonstrating efficiency of the biocatalyst developed and its applicability to both natural and non-natural targets. To the best of our knowledge, this is the first use of TCA cycle intermediates as the energy source for oligosaccharide synthesis and the first description of successful synthesis of fucosyllactulose with titers in several grams per liter. 2′ fucosyllactose, fucosylated oligosaccharide, lactulose, metabolic engineering, whole-cell biocatalysis Introduction Oligosaccharides, often conjugated with surface proteins or lipids, serve as molecular recognition elements in a wide variety of biological processes including many disease-causing events (Varki and Lowe 2009; Kobata 2010). In human milk, soluble oligosaccharides, with about 200 distinct structures (Davis et al. 2016), form the third most abundant component in human milk (after lactose and lipid). Increasing evidence shows that they are anti-infective (Kunz et al. 2000), prebiotic (Bode and Jantscher-Krenn 2012), and immune-stimulatory (Newburg 2013). Fucosylated oligosaccharides are a subset of oligosaccharides where fucose is alpha-linked to either galactose or N-acetyl-glucosamine (GlcNAc). The ability to synthesize these molecules and their derivatives is of interest to many scientific fields and for potential medical and diagnostic applications. Unfortunately, while important progress has been made recently, relative to the vast diversity in glycans existing in nature, significant innovation has to be made in order to make a large number of oligosaccharides available in quantities sufficient for structure–activity relationship studies and for various food, medical and other commercial interests. Microbial synthesis has recently emerged as one of the most effective approaches for the synthesis of oligosaccharides (Han et al. 2012). For several hotly pursued molecules such as 2-fucosyl-lactose (2′-FL) and 3′-sialyl-lactose (3′-SL), fermentation processes based on a single metabolically engineered strain are promising (Priem et al. 2002; Lee et al. 2012; Chin et al. 2015). However, in general, using microbial biocatalyst for oligosaccharide synthesis is still quite challenging (Ruffing and Chen 2006; Chen 2015). A major challenge in engineering single-strain biocatalyst stems from a universal metabolic regulation known as catabolite repression. In the synthesis stage, it is necessary for the whole-cell biocatalyst to engage multiple sugars, including both acceptor sugars and donor sugars. In addition, the cells need an energy source (such as glucose) to replenish cellular energy (in the form of adenosine triphosphate (ATP), guanosine triphosphate (GTP), etc.) for the synthesis to continue. While this conflict can be resolved by supplying glycerol, this strategy leads to a relatively low productivity and significant amount of acidic metabolic by-products. In this study, we describe an approach to directly tap into the cellular “power house,” the TCA cycle, to produce energy needed for the synthesis and to sustain biocatalysis. This approach not only circumvents catabolite repression but also eliminates acidic glycolysis by-products. As such, the whole-cell biocatalysis can be carried out without sophisticated fed-batch feeding and pH control in the synthesis stage. We first demonstrate the concept with 2′-FL synthesis in a shake flask process. Subsequently, we extend the biocatalysis process to a novel oligosaccharide molecule, fucosyllactulose, demonstrating the applicability of the biocatalyst developed to a non-natural target. To the best of our knowledge, this is the first use of TCA cycle intermediate as an alternative energy source to power the oligosaccharide synthesis in whole-cell biocatalysis. Results and discussion Construction of whole-cell biocatalyst for fucosylated oligosaccharides synthesis Our initial biocatalyst design features a salvage pathway for GDP-fucose synthesis through a plasmid-borne bifunctional enzyme fucosekinase/fucose-1-phosphate guanylyltransferase (FKP) from Bacteroides fragilis, which catalyzes the synthesis of GDP-fucose from exogenous fucose, ATP and GTP from cellular metabolism (Figure 1A). The recombinant plasmid (Figure 1B) was constructed to co-overexpress FKP and a Helicobacter pylori fucosyltransferase (FutC) from a T5 promoter. The futC gene was cloned in two separate pieces and modification of sequence was introduced to avoid difficulties in polymerase chain reaction (PCR) amplification and expression (Wang et al. 1999). The reconstructed gene was designated as an futCR. The resulting protein, however, is identical to GenBank ARB16053.1. As shown in Figure 2A, the resulting biocatalyst, Escherichia coli JM109/pQFKPfutCR was able to achieve a respectable titer of 0.74 g/L in a 24-h reaction from fucose and lactose using shake flask culture. Figure 1. View largeDownload slide (A) Engineered metabolic pathway for 2′-FL synthesis highlighting GDP-fucose synthesis via the salvage pathway, inactivation of fucose metabolism and TCA cycle for energy provision. (B) Recombinant plasmid for overexpression of FKP and FutCR. Figure 1. View largeDownload slide (A) Engineered metabolic pathway for 2′-FL synthesis highlighting GDP-fucose synthesis via the salvage pathway, inactivation of fucose metabolism and TCA cycle for energy provision. (B) Recombinant plasmid for overexpression of FKP and FutCR. Figure 2. View largeDownload slide Time course of 2′-FL synthesis. (A) Comparison between E. coli strains JCTFKP/pQFKPfutCR and JM109/pQFKPfutCR. (B) Comparison of three different TCA cycle intermediates as energy sources, α-KG (square), citrate (circle) and succinate (triangle). (C) Comparison of α-KG with glycerol and glucose as energy sources, α-KG (square), glycerol (circle) and glucose (triangle). (D) pH profiles during 2′FL synthesis with different energy sources. Results shown in this study are the mean of triplicated experiments and error bars indicate the standard deviation. Figure 2. View largeDownload slide Time course of 2′-FL synthesis. (A) Comparison between E. coli strains JCTFKP/pQFKPfutCR and JM109/pQFKPfutCR. (B) Comparison of three different TCA cycle intermediates as energy sources, α-KG (square), citrate (circle) and succinate (triangle). (C) Comparison of α-KG with glycerol and glucose as energy sources, α-KG (square), glycerol (circle) and glucose (triangle). (D) pH profiles during 2′FL synthesis with different energy sources. Results shown in this study are the mean of triplicated experiments and error bars indicate the standard deviation. To improve the synthesis and decrease the metabolism of fucose, we decided to knockout the genes associated with the first two steps of fucose metabolism. Specifically, the genes encoding fucose isomerase and fuculose kinase were knocked out (Figure 1A). As a result, the synthesis was significantly increased with the highest titer of 3.3 g/L reached at 24 h. This is a 4.7-fold increase compared to the control, JM109/pQFKPfutCR. Additionally, the molar yield from fucose was increased by 4-fold from 0.14 to 0.56 mol/mol. As shown in Figure 1A, knockout of fucI and fucK disrupts the metabolism of fucose via TCA cycle. This disruption is responsible for increased yield from fucose, suggesting that elimination of nonreaction-related metabolism of fucose increases the availability of fucose for GDP-fucose synthesis (Figure 1A). TCA cycle for energy provision for sustained synthesis Oligosaccharide synthesis is an energy-intensive process. Each glycosidic bond formation requires at least two high-energy compounds (two ATP equivalents) (Ruffing and Chen 2011). In the case of 2′-FL synthesis, the formation of donor sugar (GDP-fucose) requires one ATP and one GTP. There may be other energy-requiring steps associated with the reaction, such as substrate transport. Thus, to sustain the synthesis reaction over a long period, an energy source has to be provided. Due to the catabolite repression, glucose is not used. Instead, glycerol was almost exclusively used (Baumgärtner et al. 2013; Chin et al. 2016). However, glycerol metabolism is slow and to minimize acid production and accumulation, a fed-batch process is generally required, increasing the complexity of the process. We reasoned that a TCA cycle could be used to provide the necessary energy for the synthesis. Directly tapping the TCA cycle, without having to go through glycolysis, may have several important advantages. Besides avoiding the glucose repression, TCA cycle also produces GTP directly via substrate-level phosphorylation (Figure 1A), thus providing a direct link between energy production and target synthesis. In addition, no acidic by-products will be produced as only CO2 and H2O are the respiration end products. This eliminates the need for pH control and simplifies the downstream processing. We tested first with α-ketoglutarate (α-KG) for 2′-FL synthesis. As shown in Figure 2B, the respiring culture rapidly accumulated 2′-FL over the period of 20–24 h, reaching the highest 2′-FL concentration at 20 h, indicating α-KG is effective as an energy source to support the synthesis reaction (Figure 2B). We subsequently tested other TCA cycle substrates, citrate and succinate, for their ability to support synthesis. Both citrate and succinate are commercial fermentation products. As shown in Figure 2B, both could support the synthesis, but reaching lower final concentration of 2′-FL under the condition tested. Additional studies are needed to identify the best TCA intermediate and optimize the feeding regime to maximize the titer of synthesis targets. When using α-KG as an energy source, 2′-FL production reached a titer of 3.3 g/L, yield from cells of 0.21 g/g-dry weight cells, yield from fucose 0.56 mol-2′-FL/mol-fucose, and yield from lactose, 0.80 mol-2′-FL/mol-lactose. The productivity is 0.14 g/L/h (Table I). Fucose yield is particularly important as it impacts the cost most. Compared to the 0.36 molar yield, reported by Chin et al. (2016), the yield of 0.56 molar yield is a significant improvement (55.6% increase). The lower productivity, 0.14 g/L/h, compared to the highest reported value (0.57 g/L/h) reported by Baumgärtner et al. (2013), could be attributed to the much lower cell concentrations in our experiments. In our experiments, the cell dry weight was about 16 g/L, whereas the fed-batch processes in a bioreactor by Baumgärtner et al. (2013) was much higher, with the highest at 60 g/L. This relatively lower productivity does not represent the intrinsic limitation of our biocatalyst, rather this is a reflection of limitation of shaker flask methods used in this study, which can only support aerobic metabolism with limited cell concentrations. Similarly, the lower titer reported here could also be attributed to the much lower cell concentration used in our study and is a limitation of shaker flask method, not the biocatalyst. This limitation can be overcome by increasing cell density in a bioreactor, where adequate aeration supports full respiration as required by the synthesis. We envision that our biocatalyst could be first grown using fed-batch cultivation in a bioreactor, as commonly done today, and induced for recombinant enzyme production at a suitable optical density, and then continued to grow to a final cell density to 60 g/L or higher, in a manner similar to what was demonstrated in prior studies (Baumgärtner et al. 2013 or Chin et al. 2016). At this point, the synthesis is initiated by the addition of fucose, lactose, α-KG or another suitable TCA cycle intermediate. Since the rate of synthesis is directly proportional to the cell density, we should see a significant increase in space-time productivity as well as in product titer. In this way, the synthesis is separated from cell growth and recombinant enzyme production phase. Two different energy sources, glycerol and a TCA cycle intermediate, could be used in the growth phase and synthesis, respectively, to achieve better process performance metrics. Table I. Fucosyl-trisaccharide production by JM109/pQFKPfutCR and JCTFKP/pQFKPfutCRa Strain JM109/pQFKPfutCR JCTFKP/pQFKPfutCR Product Fucosyllactose Fucosyllactose Fucosyllactulose Trisaccharide concentration (g/L) 0.74 ± 0.036 3.28 ± 0.032 4.34 ± 0.183 Fucose consumed (g/L) 1.73 ± 0.110 1.98 ± 0.167 1.70 ± 0.110 Acceptor sugar consumed (g/L) 1.14 ± 0.047 2.87 ± 0.090 3.77 ± 0.008 YDCWb (g/g) 0.046 ± 0.002 0.205 ± 0.002 0.277 ± 0.011 Yfucosec (mol/mol) 0.144 ± 0.011 0.558 ± 0.050 0.860 ± 0.056 Yacceptor sugard (mol/mol) 0.455 ± 0.057 0.801 ± 0.034 0.807 ± 0.036 Productivity (g/L/h) 0.031 ± 0.002 0.137 ± 0.004 0.181 ± 0.008 Strain JM109/pQFKPfutCR JCTFKP/pQFKPfutCR Product Fucosyllactose Fucosyllactose Fucosyllactulose Trisaccharide concentration (g/L) 0.74 ± 0.036 3.28 ± 0.032 4.34 ± 0.183 Fucose consumed (g/L) 1.73 ± 0.110 1.98 ± 0.167 1.70 ± 0.110 Acceptor sugar consumed (g/L) 1.14 ± 0.047 2.87 ± 0.090 3.77 ± 0.008 YDCWb (g/g) 0.046 ± 0.002 0.205 ± 0.002 0.277 ± 0.011 Yfucosec (mol/mol) 0.144 ± 0.011 0.558 ± 0.050 0.860 ± 0.056 Yacceptor sugard (mol/mol) 0.455 ± 0.057 0.801 ± 0.034 0.807 ± 0.036 Productivity (g/L/h) 0.031 ± 0.002 0.137 ± 0.004 0.181 ± 0.008 aα-KG was used as an energy source for fucosyl-trisaccharide production. bYield to cells (YDCW) = concentration of product/DCW. cYield to fucose (Yfucose) = concentration of product/concentration of fucose consumed. dYield to acceptor sugar (Yacceptor sugar) = concentration of product/concentration of acceptor sugar (lactose/lactulose) consumed. Table I. Fucosyl-trisaccharide production by JM109/pQFKPfutCR and JCTFKP/pQFKPfutCRa Strain JM109/pQFKPfutCR JCTFKP/pQFKPfutCR Product Fucosyllactose Fucosyllactose Fucosyllactulose Trisaccharide concentration (g/L) 0.74 ± 0.036 3.28 ± 0.032 4.34 ± 0.183 Fucose consumed (g/L) 1.73 ± 0.110 1.98 ± 0.167 1.70 ± 0.110 Acceptor sugar consumed (g/L) 1.14 ± 0.047 2.87 ± 0.090 3.77 ± 0.008 YDCWb (g/g) 0.046 ± 0.002 0.205 ± 0.002 0.277 ± 0.011 Yfucosec (mol/mol) 0.144 ± 0.011 0.558 ± 0.050 0.860 ± 0.056 Yacceptor sugard (mol/mol) 0.455 ± 0.057 0.801 ± 0.034 0.807 ± 0.036 Productivity (g/L/h) 0.031 ± 0.002 0.137 ± 0.004 0.181 ± 0.008 Strain JM109/pQFKPfutCR JCTFKP/pQFKPfutCR Product Fucosyllactose Fucosyllactose Fucosyllactulose Trisaccharide concentration (g/L) 0.74 ± 0.036 3.28 ± 0.032 4.34 ± 0.183 Fucose consumed (g/L) 1.73 ± 0.110 1.98 ± 0.167 1.70 ± 0.110 Acceptor sugar consumed (g/L) 1.14 ± 0.047 2.87 ± 0.090 3.77 ± 0.008 YDCWb (g/g) 0.046 ± 0.002 0.205 ± 0.002 0.277 ± 0.011 Yfucosec (mol/mol) 0.144 ± 0.011 0.558 ± 0.050 0.860 ± 0.056 Yacceptor sugard (mol/mol) 0.455 ± 0.057 0.801 ± 0.034 0.807 ± 0.036 Productivity (g/L/h) 0.031 ± 0.002 0.137 ± 0.004 0.181 ± 0.008 aα-KG was used as an energy source for fucosyl-trisaccharide production. bYield to cells (YDCW) = concentration of product/DCW. cYield to fucose (Yfucose) = concentration of product/concentration of fucose consumed. dYield to acceptor sugar (Yacceptor sugar) = concentration of product/concentration of acceptor sugar (lactose/lactulose) consumed. To illustrate the advantage of using α-KG as an alternative energy source to support synthesis, additional experiments were conducted to compare three different energy sources (α-KG, glycerol and glucose) for 2′FL synthesis. As shown in Figure 2C, α-KG is far better than glycerol and the resulting titer is several folds higher than that of glycerol. Using glucose as an energy source resulted in no product synthesis. This can be explained by the catabolite repression, which prevents the uptake of either lactose or fucose. During the synthesis, the pH of the reaction medium was monitored. When α-KG is used as an energy source, the pH change was minimal with only a slight uptick to 7.7 at the end of a 24 -h reaction (Figure 2D). In contrast, for both glucose and glycerol, rapid acidification of reaction mixture was observed. The final pHs for glucose and glycerol were 3.9 and 4.4, respectively, which were far out of the range of pH supporting optimal carbon metabolism. The use of TCA cycle intermediate as an energy source is a unique metabolic state, where there is no glycolysis yet TCA cycle is fully functional. This contrasts with a metabolic state when glucose or glycerol is used, where both glycolysis and TCA cycle are active. Glycolysis intermediates, such as pyruvate, can feed into the TCA cycle or alternatively metabolize to generate mixed acids dominated by acetic acid and lactate. Pyruvate and other acids are possibly responsible for the rapid acidification of the medium as noted in Figure 2D. Therefore, besides being a better energy source, α-KG reduces the requirement of pH monitoring and adjustment during the synthesis phase. This advantage translates to less salts produced in a bioreactor and makes downstream processing much easier. Additionally, without the requirement of pH monitoring/adjustment, it makes whole-cell biocatalysts much more conducive to be used in shaker flask for various new targets that have not yet been mature enough for bioreactor operation. Thus, this approach could potentially accelerate laboratory research for the synthesis of new glycan targets. Synthesis of novel fucosylated trisaccharide To test whether our fucosylation biocatalyst could accept other substrates for novel structures, we carried out reactions under the otherwise identical conditions (as described above) except we supplied the respiring culture with lactulose as an acceptor sugar substrate. Lactulose is a disaccharide of galactose and fructose. It is similar to lactose but with fructose in the position of glucose. Figure 3 shows that the fucosylation biocatalyst fucosylated lactulose efficiently. In 24 h, the respiring culture resulted in 4.3 g/L fucosyllactulose. Fucosyllactulose synthesized by the whole-cell biocatalyst was adsorbed on an activated carbon column. After exhaustive washing with water, the trisaccharide was eluted with 40% of ethanol. The purified trisaccharide was analyzed by GC–MS at the Complex Carbohydrate Research Center and fucosylation was confirmed (Supplementary data). Lactulose is nondigestible and prebiotic. Its use has been associated with a number of beneficial effects, such as improved calcium absorption (Kuschel et al. 2017), inhibition of pathogenic bacteria while promoting growth of bifddobacteria and lactobacilli in the gastrointestinal tract (Wu et al. 2017). As the antiadhesion effect of 2′FL is associated with the terminal fucosyl linkage (Castanys-Munoz et al. 2013), it is speculated that successful fucosylation of lactulose adds a potential antiadhesive function to this already useful prebiotic molecule. To definitively prove this function, it requires significant amount of the new molecule. This study provided a viable approach for its synthesis. Figure 3. View largeDownload slide Synthesis of fucosylated lactulose by E. coli JCTFKP/pQFKPfutCR with α-KG as an energy source. Figure 3. View largeDownload slide Synthesis of fucosylated lactulose by E. coli JCTFKP/pQFKPfutCR with α-KG as an energy source. In summary, we have constructed an efficient E. coli whole-cell fucosylation biocatalyst and developed a versatile biocatalysis process for not only 2′-FL but also other related structures. We demonstrated that the developed system could achieve a titer of several grams per liter within a 24-h period by shaker flask cultivation without sophisticated control (such as energy source feeding, pH and oxygen level). It is conceivable that the system can be extended to other fucosyl oligosaccharides with only minor changes in the requisite fucosyltransfereases. Other Human Milk Oligosaccharides-fucosylated oligosaccharides such as 3′-fucosyllactose, lacto-N-fucopentaose I (FPLNT) are potential targets of the system. We anticipate that the concept of TCA-powered synthesis could also be applied to nonfucosylated oligosaccharides. Materials and methods Chemicals All chemicals were purchased from Sigma-Aldrich Chemical Co. (St. Louis, MO). DNA polymerase and restriction enzymes were purchased from New England Biolabs (Ipswich, MA). Bacterial strains and plasmids Escherichia coli JM109 was used as the host for gene cloning and expression. The primers used in this study were listed in Supplementary data, Table SI. The gene fkp was amplified from B. fragilis NCTC 9343 using primers FKP-F and FKP-R, digested by BamHI and SalI, and ligated to pQE80L to form pQFKP. To avoid mis- or no-amplification of the original futC gene caused by polyC sequence, two DNA fragments before and after the polyC repeats were first obtained from the H. pylori 25695 genomic DNA using primers futC1/futC2 and futC3/futC4. The complete polyC-modified futC gene (futCR) was obtained by overlapped PCR of the two fragments (Huang et al. 2017) using futCR-F and futCR-R, digested with SalI and PstI and cloned into pQFKP to constructed plasmid pQFKPfutCR coexpressing fkp and futCR by a T5 promoter (Figure 1B). To inactivate fucose-metabolizing pathway in E. coli JM109, T5-derived fkp was inserted between fucP (L-fucose transporter) and fucU (L-fucose mutarotase), while fucI (L-fucose isomerase) and fucK (L-fuculose kinase) were deleted at the same time through homologous recombination (Datsenko and Wanner 2000). Upstream DNA of fucI and downstream DNA of fucK were amplified by primers fucUP1/fucUP2 and fucDn1/fucDn2, respectively, and fused by overlapped PCR. The expression cassette of T5-fkp was amplified by T5-F and FKP-R from pQFKP. Chloramphenicol resistance gene (CmR DNA with FRT sequence at both ends) was amplified from pKD3 by CmR-F and CmR-R. The three DNA fragments were digested with BamHI and XhoI; BglII and SalI; and SalI and XhoI, respectively, and ligated to form upstream-T5FKP-CmR-downstream DNA fragment. The fragment was transformed to E. coli JM109 to replace fucI and fucK genes with T5FKP-CmR DNA (Figure 1A). pCP20 was then introduced to remove FRT-CmR gene, and E. coli JCTFKP strain with augmented FKP activity and inactivated fucose-metabolizing activity by the deletion of fucIK genes was obtained. Protein expression and fucosylated trisaccharide synthesis The expression conditions of exogenous proteins in engineered strains including IPTG concentration, induction temperature, and initial cell density and induction time were optimized. Overnight preinoculate was prepared in LB medium at 37°C and transferred to fresh LB at an inoculum concentration of 1% (v/v) and grown to OD600 = 0.6. Expression of the recombinant proteins was induced by the addition of 1 mM IPTG. The culture was incubated at 30°C for 12 h. Cells were harvested by centrifugation (4000 × g, 10 min, 4°C) and resuspended in the reaction mixture containing 80 mM Tris-HCl (pH 7.5), 11.9 mM PBS (pH 7.5), 20 mM MgCl2, 5 mM MnCl2, 5 g/L fucose, 5 g/L lactose and 5 g/L TCA cycle intermediate (α-KG or citrate or succinate). The reaction was carried out at 37°C and shaked for 24 h at 250 rpm, and additional dose of TCA cycle intermediate was added to a final concentration of 5 g/L α-KG at 4, 8 and 12 h after sampling. For the comparison of α-KG, glycerol and glucose, the above described synthesis procedure was followed, except glycerol and glucose were used in the respective experiments. pH was measured with pH indicator paper with accuracy to 0.1 unit. The initial and final pHs were additionally measured by a pH meter. Analytic methods Saccharides were detected by HPLC (Agilent 1100 (Santa Clara, CA)) equipped with an Aminex HPX-87H column (300 × 7.8 mm, Bio-Rad (Hercules, CA)) and a refractive index detector. The samples were boiled for 10 min in a water bath and centrifuged at 13,000 rpm for 5 min. The supernatant was filtered with a filtration membrane (0.22 μm pore size) and 10 μL was injected into the column. H2SO4 (5 mM) was used as the mobile phase at a flow rate of 0.3 mL/min. Optical cell density was measured at 600 nm absorbance using a spectrophotometer. The dry cell weight (DCW) was calculated as follows: DCW (g/L) = 0.32 × OD600. Supplementary data Supplementary data are available at GLYCOBIOLOGY online. Funding This research is supported by an NSF (grant CBET1509202 to R.C.). Conflict of interest statement The authors declare no conflict of interest. Abbreviations 2′-FL 2-fucosyl-lactose 3′-SL 3′-sialyl-lactose FKP fucosekinase/fucose-1-phosphate guanylyltransferase FutC fucosyltransferase α-KG α-ketoglutarate DCW dry cell weight ATP adenosine triphosphate GTP guanosine triphosphate. References Baumgärtner F , Seitz L , Sprenger GA , Albermann C . 2013 . Construction of Escherichia coli strains with chromosomally integrated expression cassettes for the synthesis of 2′-fucosyllactose . Microb Cell Fact . 12 ( 1 ): 40 . Google Scholar CrossRef Search ADS PubMed Bode L , Jantscher-Krenn E . 2012 . Structure-function relationships of human milk oligosaccharides . Adv Nutr . 3 ( 3 ): 383S – 391S . Google Scholar CrossRef Search ADS PubMed Castanys-Munoz E , Martin MJ , Prieto PA . 2013 . 2′-fucosyllactose: an abundant, genetically determined soluble glycan present in human milk . Nutr Rev . 71 ( 12 ): 773 – 789 . Google Scholar CrossRef Search ADS PubMed Chen R . 2015 . The sweet branch of metabolic engineering: cherry-picking the low-hanging sugary fruits . Microb Cell Fact . 14 : 197 . Google Scholar CrossRef Search ADS PubMed Chin YW , Kim JY , Lee WH , Seo J-H . 2015 . Enhanced production of 2′-fucosyllactose in engineered Escherichia coli BL21star (DE3) by modulation of lactose metabolism and fucosyltransferase . J Biotechnol . 210 : 107 – 115 . Google Scholar CrossRef Search ADS PubMed Chin YW , Seo N , Kim JH , Seo JH . 2016 . Metabolic engineering of Escherichia coli to produce 2′-fucosyllactose via salvage pathway of guanosine 5′-diphosphate (GDP)-l-fucose . Biotechnol Bioeng . 113 ( 11 ): 2443 – 2452 . Google Scholar CrossRef Search ADS PubMed Datsenko KA , Wanner BL . 2000 . One-step inactivation of chromosomal genes in Escherichia coli K-12 using PCR products . Proc Natl Acad Sci USA . 97 ( 12 ): 6640 – 6645 . Google Scholar CrossRef Search ADS PubMed Davis JC , Totten SM , Huang JO , Nagshbandi S , Kirmiz N , Garrido DA , Lewis ZT , Wu LD , Smilowitz JT , German JB . 2016 . Identification of oligosaccharides in feces of breast-fed infants and their correlation with the gut microbial community . Mol Cell Proteomics . 15 ( 9 ): 2987 – 3002 . Google Scholar CrossRef Search ADS PubMed Han NS , Kim TJ , Park YC , Kim J , Seo JH . 2012 . Biotechnological production of human milk oligosaccharides . Biotechnol Adv . 30 ( 6 ): 1268 – 1278 . Google Scholar CrossRef Search ADS PubMed Huang D , Yang K , Liu J , Xu Y , Wang Y , Wang R , Liu B , Feng L . 2017 . Metabolic engineering of Escherichia coli for the production of 2′-fucosyllactose and 3-fucosyllactose through modular pathway enhancement . Metab Eng . 41 : 23 – 38 . Google Scholar CrossRef Search ADS PubMed Kobata A . 2010 . Structures and application of oligosaccharides in human milk . Proc Jpn Acad Ser B . 86 ( 7 ): 731 – 747 . Google Scholar CrossRef Search ADS Kunz C , Rudloff S , Baier W , Klein N , Strobel S . 2000 . Oligosaccharides in human milk: structural, functional, and metabolic aspects . Ann Rev Nutr . 20 ( 1 ): 699 – 722 . Google Scholar CrossRef Search ADS Kuschel B , Seitl I , Gluck C , Mu W , Jiang B , Stressler T , Fischer L . 2017 . Hidden reaction: Mesophilic cellobiose 2-epimerases produce lactulose . J Agric Food Chem . 65 : 2530 – 2539 . Google Scholar CrossRef Search ADS PubMed Lee W-H , Pathanibul P , Quarterman J , Jo J-H , Han NS , Miller MJ , Jin J-S , Seo J-H . 2012 . Whole cell biosynthesis of a functional oligosaccharide, 2′-fucosyllactose, using engineered Escherichia coli . Microb Cell Fact . 11 : 48 . Google Scholar CrossRef Search ADS PubMed Newburg D . 2013 . Glycobiology of human milk . Biochemistry (Mosc) . 78 ( 7 ): 771 – 785 . Google Scholar CrossRef Search ADS PubMed Priem B , Gilbert M , Wakarchuk WW , Heyraud A , Samain E . 2002 . A new fermentation process allows large-scale production of human milk oligosaccharides by metabolically engineered bacteria . Glycobiology . 12 ( 4 ): 235 – 240 . Google Scholar CrossRef Search ADS PubMed Ruffing A , Chen RR . 2006 . Metabolic engineering of microbes for oligosaccharide and polysaccharide synthesis . Microb Cell Fact . 5 ( 1 ): 25 . Google Scholar CrossRef Search ADS PubMed Ruffing AM , Chen RR . 2011 . Citrate stimulates oligosaccharide synthesis in metabolically engineered Agrobacterium sp . Appl Biochem Biotech . 164 ( 6 ): 851 – 866 . Google Scholar CrossRef Search ADS Varki A , Lowe JB . 2009 . Biological roles of glycans . In: Varki A , Cummings RD , Esko JD , Freeze HH , Stanley P , Bertozzi CR , Hart GW , Etzler ME , editors . Essentials of Glycobiology . New York : Cold Spring Harbor . p. 75 – 88 . Wang G , Rasko DA , Sherburne R , Taylor DE . 1999 . Molecular genetic basis for the variable expression of Lewis Y antigen in Helicobacter pylori: Analysis of the α (1, 2) fucosyltransferase gene . Mol Microbiol . 31 ( 4 ): 1265 – 1274 . Google Scholar CrossRef Search ADS PubMed Wu L , Xu C , Li S , Liang J , Xu H , Xu Z . 2017 . Efficient production of lactulose from whey powder by cellobiose-2-epimerase in an enzymatic membrane reactor . Bioresour Technol . 233 : 305 – 312 . Google Scholar CrossRef Search ADS PubMed © The Author(s) 2018. Published by Oxford University Press. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)

Journal

GlycobiologyOxford University Press

Published: May 24, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off