Novel Biosynthetic Pathway for Nicotinamide Mononucleotide Production from Cytidine in Escherichia coli

Abstract

Nicotinamide mononucleotide, known as NMN, is an important nicotinamide adenine dinucleotide (NAD⁺) precursor. It is integral in cellular metabolism, energy generation, and processes associated with aging. Since NMN provides healthy value, it becomes a major focus for the biotechnological industry. This study presents a new biosynthetic pathway for producing NMN without limits on intracellular PRPP (5′-phosphoribosyl pyrophosphate) metabolic flux. The route started by converting cytidine into 1-phosphoribose via pyrimidine-nucleoside phosphorylase (PyNP), after transforming into nicotinamide riboside (NR) through either purine-nucleoside phosphorylase (XapA) or nicotinate riboside kinase (NRK). NR was phosphorylated by NRK in the presence of nicotinamide (NAM) to produce NMN. We established an in vitro enzyme activity verification system for the feasibility check. The optimization of multienzyme cascade reactions was figured out for the NMN biosynthesis. Finally, the enzymes of PyNP and NRK were expressed in the cytidine-producing strain; we established a de novo biosynthesis pathway from glucose to NMN, achieving a production titer of 33.71 mg/L at a shake-flask scale.

1. Introduction

The biosynthesis of nicotinamide mononucleotide (NMN) is critical for maintaining cellular nicotinamide adenine dinucleotide (NAD⁺) levels, which in turn are essential for a variety of vital processes, including energy metabolism, DNA repair, and stress response [1,2,3]. As NAD⁺ is involved in numerous cellular functions, such as mitochondrial health, aging, and immune response, NMN has gained significant attention as a potential therapeutic agent to mitigate age-related diseases and metabolic disorders [4,5].

The natural biosynthesis of NMN primarily occurs through two major pathways: the Preiss–Handler pathway and the salvage pathway [6]. In the Preiss–Handler pathway, nicotinic acid (NA) is converted to NMN by nicotinic acid phosphoribosyltransferase (NAPRT) and is subsequently converted into NAD⁺ through additional enzymatic steps. On the other hand, the salvage pathway utilizes nicotinamide (NAM), derived from NAD⁺ degradation, and nicotinamide riboside (NR) [7,8] to synthesize NMN. The key enzymes involved in these processes include NAMPT (nicotinamide phosphoribosyltransferase) [9,10], NRK (nicotinamide riboside kinase) [11,12], and NAPRT. Both pathways are tightly regulated and play a critical role in maintaining NAD⁺ homeostasis, which is essential for proper cellular function [13,14].

Early industrial-scale production of NMN predominantly relied on chemical catalytic processes using AMP or tetraacetylribose as substrates [6]. However, these methods faced challenges, including demanding reaction conditions and the formation of isomeric byproducts, leading to persistently elevated production costs [14,15,16]. There is a growing interest in figuring out biosynthetic approaches that are less demanding and have superior stereoselectivity. To date, two principal synthetic routes for NMN production have been established. SHOJI et al., using the exogenous NAM and intracellular phosphoribosyl pyrophosphate (PRPP), were converted to NMN in a single-step enzymatic reaction catalyzed by nicotinamide phosphoribosyltransferase (Nampt) [4]. Wang et al. developed a four-step enzymatic cascade for NMN synthesis starting from D-ribose. D-ribose was first converted to ribose-5-phosphate, known as R-5-P, by utilizing ribokinase, also known as RK [17]; R-5-P turned to D-ribosyl-1-phosphate (R-1-P) by phosphopentomutase (PPM) [18,19,20]. The synthesis of NR from R-1-P and NAM was accomplished via transglycosylation catalyzed by purine-nucleoside phosphorylase, (XapA) or nicotinate riboside kinase (NRK), or purine-nucleoside phosphorylase (PNP) [3,21]. Finally, NR was subsequently phosphorylated to NMN by NRK [22]. The insufficient supply of PRPP [23] and low flux of R-1-P by the bidirectional catalytic activity of PPM restricted the efficiency of synthesizing NMN through two pathways. The further development of artificial NMN biosynthetic pathways will be beneficial.

This study proposes a novel NMN synthesis method that addresses the questions mentioned by employing an enzyme cascade reaction using cytidine as the precursor. Cytidine is an essential component of RNA in plants, animals, and humans, playing vital physiological roles [24]. It also serves as a crucial intermediate for synthesizing various antiviral and antitumor drugs [25]. Currently, cytidine production is primarily achieved through chemical synthesis and microbial fermentation, both of which have been implemented in industrial-scale manufacturing. Therefore, our designed pathway can leverage these established cytidine production processes to achieve NMN synthesis, representing a highly promising NMN manufacturing method with significant industrial potential.

Our engineered pathway involves the use of three key enzymes: pyrimidine-nucleoside phosphorylase (PyNP, EC 2.4.2.2) [26], nicotinate riboside kinase (NRK, EC 2.7.1.173), and purine-nucleoside phosphorylase (XapA, EC 2.4.2.1) [27]. Cytidine was first converted to R-1-P by PyNP. In the next step, D-ribosyl-1-phosphate combines with NAM under the catalysis of XapA or NRK to form NR. Finally, NMN is synthesized through the kinase activity of NRK, which catalyzes the transfer of a phosphate group from ATP to NR, resulting in NMN production. The new route has benefits compared to the usual practices, because it uses efficient enzyme-catalyzed reactions and cost-effective substrates. We fine-tuned the enzyme ratios and conditions, resulting in a production process for NMN that is both efficient and sustainable.

2. Results and Discussion

2.1. Design and Validation of a Novel In Vitro Enzymatic Pathway for NMN Biosynthesis

The synthesis of nicotinamide mononucleotide (NMN) is of considerable interest due to its pivotal role as a precursor in NAD⁺ biosynthesis, which is essential for cellular redox reactions, DNA repair, and signal transduction [28,29,30]. The study introduced a new in vitro biosynthetic technique to produce NMN, relying on cytidine as the principal substrate and nicotinamide (NAM) as a co-substrate. The cascade of enzymes comprised three key enzymes: pyrimidine-nucleoside phosphorylase (PyNP), purine-nucleoside phosphorylase (XapA), or nicotinate riboside kinase (NRK). NRK is also related to the phosphorylation step.

During the initial step, PyNP catalyzed the phosphorolysis of cytidine into D-ribose-1-phosphate. Following the interaction of XapA or NRK, the intermediate was merged with NAM to produce nicotinamide riboside (NR). After, NR was phosphorylated to NMN by NRK, while utilizing ATP as a phosphate donor. Figure 1 shows the synthetic approach in a schematic format. The heterologous expression of the enzymes is as follows: a 46.3 kDa PyNP, a 30.8 kDa XapA, and a 27.3 kDa NRK was achieved in E. coli BL21 (DE3) using pET28a plasmids. SDS-PAGE analysis confirmed soluble expression of all target proteins (Figure 2), establishing the foundational feasibility of the pathway.

2.2. Enzyme Activity Validation and Optimization of In Vitro Cascade Reaction

Each enzyme was purified individually and evaluated for its catalytic efficiency in the designed synthetic pathway. The structures of the three enzymes are shown in Supplementary Figure S2. Purified PyNP was able to convert 10 mM cytidine into 1-phosphoribose within 2 h at a 20 ng/μL enzyme concentration (Figure 3A). This indicated high enzymatic activity and stability, making it suitable for a cascade reaction setup. The subsequent combination of PyNP with XapA showed an effective conversion of cytidine into NR, particularly when the reaction pH was maintained at 7.5–8.0 (Figure 3B). Under these conditions, NR accumulation reached 0.38 g/L, highlighting the importance of pH optimization for maximizing yield. In the final step, 0.55 g/L of NR was phosphorylated into NMN by NRK with a 51% conversion efficiency, resulting in 0.28 g/L of NMN over 3 h (Figure 3C). This demonstrates the viability of the complete three-step enzyme cascade in vitro.

High-performance liquid chromatography (HPLC) and mass spectrometry (MS) confirmed the presence and identity of NMN (357.0450 [M + Na]⁺), matching the expected standard (357.0458 m/z), as shown in Supplementary Figure S1. As a control, a reaction lacking enzyme supplementation yielded negligible NMN, confirming the specificity of the enzymatic system. Interestingly, when comparing different cascade configurations, the PyNP–NRK (PN) pathway yielded 171 mg L⁻¹ of NMN, which was 94% higher than the PyNP–XapA–NRK (PXN) pathway (Figure 3D). This efficiency, combined with the reduced enzyme complexity, underscores the cost-effectiveness and operational simplicity of the PN pathway. The minimal enzyme requirement reduces both operational overhead and purification demands, making it an attractive option for industrial applications.

2.3. Intracellular Implementation of NMN Biosynthesis in Engineered E. coli

Building upon the in vitro success, the pathway was engineered into E. coli KQ_BG001 to enable intracellular NMN production. This strain was selected due to its ability to metabolize glucose and produce cytidine endogenously. To construct the intracellular cascade, PyNP was cloned into a pRSFduet-1 plasmid under a trc promoter, while NRK was inserted into a pTrc99a plasmid under an inducible promoter. Co-transformation into E. coli allowed for the simultaneous overexpression of both enzymes.

Initial shake-flask experiments using glucose as a carbon source showed that cytidine levels reached 0.73 g/L within 32 h, confirming sufficient precursor availability for NMN biosynthesis (Figure 4). Although initial NMN levels were low, the optimization of induction and fermentation conditions significantly improved yields. The optimal parameters were determined to be 25 °C, pH 7.0, and 0.5 mM IPTG (Figure 5). Under these conditions, NMN production reached 33.71 mg/L. This result demonstrates the functional implementation of the NMN biosynthetic pathway in vivo and validates the catalytic efficiency of the expressed enzymes under physiological conditions.

The study presents a pioneering strategy for NMN biosynthesis through both in vitro and in vivo enzyme cascade pathways. The key innovation lies in the use of cytidine as a substrate to bypass the cellular reliance on PRPP, an often limiting factor in NAD⁺ biosynthetic routes. Furthermore, the comparative analysis between the PXN and PN cascades illustrates the potential to streamline biosynthesis by reducing the number of enzymatic steps, thereby increasing yield and economic viability. The intracellular production system provides a valuable platform for further metabolic engineering. While current NMN titers are moderate compared to industrial standards, the results provide a strong foundation for subsequent enhancements. The upcoming research may focus on tuning the flow of cytidine biosynthesis, improving enzyme durability, or refining cofactors to a higher production level. Another possible approach to increase productivity might be optimizing the PyNP and NRK through adaptive laboratory evolution.

3. Materials and Methods

3.1. Plasmid Construction and Strains Used

For the cloning host, we chose E. coli DH5α, while E. coli BL21 (DE3) was selected for NMN synthesis and expressing proteins. E. coli KQ_BG001 was preserved in a tube containing 20% (v/v) glycerol at a temperature of −80 °C in our lab. For cloning experiments and the preparation of seed cultures, we kept E. coli cells at a temperature of 37 °C, using either the Luria–Bertani medium or Luria–Bertani two percent (w/v) agar plates.

Table 1. Strains and plasmids used in this work.

Strains or Plasmids	Description	Source
Strains
E. coli BL21 (DE3)	Used as host strain	Invitrogen
E. coli DH5α	Used as cloning strain	Invitrogen
E. coli KQ_BG001	Used as host strain for NMN biosynthesis	Laboratory preservation
Plasmids
pET-28a (+)	Cloning vector, KanR	General bio
pRSFDuet-1	With T7 promoter, KanR	General bio
pTrc99a	With trc promoter, AmpR	General bio
pRSFDuet-trc	With trc promoter, KanR	This study

provides information on each strain. The antibiotics, including kanamycin at 25 μg/mL, were added as required.

In Table 1, you can find a list of the plasmids, including pET-28a (+), pTrc99a, and pRSFduet-1, all used for constructing plasmids and NMN synthesis. In this study, the single-enzyme activity validation and protein purification were carried out using the pET-28a (+) plasmid vector. During the intracellular catalytic experiments, since the host cells only support the recognition of the trc promoter, the overexpressed genes PyNP and NRK were placed in the pRSFDuet-trc and pTrc99a vectors, respectively.

Table 2. The recombinant plasmids and specific primers used in this work.

Strains or Plasmids	Description	Source
Plasmids
pET28a-PyNP	pET-28a (+), with PyNP	This study
pET28a-XapA	pET-28a (+), with XapA	This study
pET28a-NRK	pET-28a (+), with NRK	This study
pRSFDuet-trc-PyNP	pRSFDuet-trc, with PyNP	This study
pTrc99a-NRK	pTrc99a, with NRK	This study
Primers
XapA-F	TCGCGGATCCGAATTCATGAGCCAGGTTCAGTTTAGCC	This study
XapA-R	GTGCGGCCGCAAGCTTTGCAATTTTACGCAGAAAACCGC	This study
NRK-F	TCGCGGATCCGAATTCATGACCACCACCAAAGTTAAACTGATTGCC	This study
NRK-R	GTGCGGCCGCAAGCTTATTTGCATCCAG	This study
PyNP-F	ATGAGAATGGTTGATATCATCACAAAAAAACAAAATGG	This study
PyNP-R	TTCCGTAATCACCGTATGCACAAGC	This study

shows the specific primers and recombinant plasmids. PyNP, XapA, and NRK were amplified by PCR using Bacillus licheniformis, E. coli, and Kluyveromyces marxianus as templates, respectively.

3.2. Optimizing Cultivation and Expressing Proteins for Recombinant E. coli

The cultivation of E.coli recombination took place in a five mL Luria–Bertani, known as LB medium. We combine the medium kanamycin, 50 μg mL⁻¹, while being shaken under 200 rpm at a temperature of 37 °C. Once the cells reached the constant phase, we moved the seed cultures to a 100 mL Luria–Bertani medium contained in several baffled Erlenmeyer flasks. The mixture remained at a consistent temperature and shaking levels until the density at 600 nm hit between 0.6 and 0.8. The expression was triggered by incorporating 0.5 mM IPTG, also known as “isopropyl-β-D-thiogalactopyranoside”. After the induction ended, we kept the mixture stirred under 200 rpm for 18 h at a temperature of 18 °C.

Subsequently, cell collection was achieved by centrifuging at 6000 rpm for 10 min at a temperature of 4 °C. In a 50 mM PBS buffer with pH 7.0, enriched with 20 mM imidazole and 300 mM NaCl, the pellet was suspended again. With a homogenizer from “Litu Mechanical Equipment Engineering Co., Ltd. Shanghai, China”, the resuspended cells were subjected to lysis at a temperature of 4 °C. We centrifuged the cell lysates at 12,000 rpm for 15 min under the same temperature to separate the supernatant. The mixture was conserved for further use.

3.3. Protein Purification and SDS-PAGE Assessment

We purified the soluble proteins through nickel-nitrilotriacetic acids, also known as Ni-NTA. The affinity chromatography column was a “His Trap™ FF 5 mL” setting from “Huiyan Biology Co., Ltd. Wuhan, Hubei, China”. The protein purification equipment uses an FPLC system from AKTA pure, General Electric Co., Ltd., Boston, MA, USA. For elution, we used a PBS buffer 50 mM, while the pH was 7.0, which included an imidazole gradient (20 mM to 500 mM) and 300 mM NaCl. After concentrating the target protein, we switched it to a storage buffer of 50 mM PBS while the pH was 7.0, utilizing a Millipore Amicon Ultra Centrifugal Filter Units from “Merck KGaA, Co., Ltd. Darmstadt, Germany”. The protein solution was added with 10 percent glycerol (v/v) and kept at −80 °C until ready for use. The protein’s purity was evaluated by SDS-PAGE, also known as “sodium dodecyl sulfate-polyacrylamide gel electrophoresis”, performed with a Mini-Protean Tetra system from “Bio-Rad Laboratories, Inc., Hercules, CA, USA”. A BCA Protein Assay Kit from “Tiangen Biotech Co., Ltd. Beijing, China” was used to calculate the protein concentrations.

3.4. Verification of Enzyme Activities in the Artificial NMN Synthesis Pathway

To ensure the synthetic pathway for artificial NMN was feasible, the performance of each enzyme was examined. The activity of the PyNP enzyme was evaluated in a reaction mixture that included 20 ng/μL of purified PyNP, 5 mM MgCl₂, 10 mM cytidine, 2 g/L polyphosphate, and 50 mM Tris-HCl buffer at a pH of 8.0. The experiment lasted at 30 °C for 4 h while the pH stayed at 7.5–8.0. The concentration of cytidine was monitored at time intervals.

To validate the effect of nicotinamide nucleoside synthetases, which were reacted with XapA and NRK, the catalytic system included 20 ng/μL of purified PyNP, 20 ng/μL of purified XapA or NRK, 5 mM MgCl₂, 10 mM cytidine, 2 g/L polyphosphate, and 10 mM NAM in 50 mM Tris-HCl buffer adjusted to pH 8.0. The process was executed at 30 °C, while the pH remained between 7.5 and 8.0 over 5 h. The consumption of NAM was recorded while the NR levels were added at specific intervals.

The nicotinamide riboside kinase enzyme activity was executed in a mixture of 20 ng/μL of purified NRK, 5 mM MgCl₂, 10 mM NR, 10 mM ATP, and 50 mM Tris-HCl buffer with a pH of 8.0. The procedure occurred at 30 °C, while the pH was maintained between 7.5 and 8.0 for 4 h. We measured the concentrations of NR and NMN alternately. Each experiment was conducted in triplicate. The data were analyzed to assess the practicality and efficiency of the artificial NMN biosynthesis pathway.

3.5. Multienzyme Cascade Catalysis Method for NMN Production

Since the nicotinamide nucleoside synthetase from the enzymes XapA and NRK were different, we formulated two pathways for NMN production within a multienzyme cascade catalysis system: the PyNP-XapA-NRK pathway (PXN pathway) and the PyNP-NRK pathway (PN pathway).

In the PXN pathway, the system contained 20 ng/μL of each purified enzyme (PyNP, XapA, and NRK), along with 5 mM MgCl₂, 10 mM cytidine, 2 g/L sodium polyphosphate, 10 mM NAM, 10 mM ATP, and 50 mM Tris-HCl buffer at pH 8.0. The reaction was conducted at 30 °C, maintaining a pH level consistently between 7.5 and 8.0. The reaction lasted for 5 h.

In the PN pathway, the reaction parameters, which include enzyme concentrations, buffer types, substrate concentrations, and pH ranges, matched those of the PXN pathway, but NRK replaced XapA to catalyze the nicotinamide nucleoside synthetase activity. In each case, the consumption of NAM and NR concentrations was recorded at specific intervals to monitor the catalytic activity.

3.6. Recombinant E. coli via Shake Flask Fermentation

The recombinant E. coli was grown in a Luria–Bertani medium, which included five g/L NaCl, five g/L yeast extract, and ten g/L peptone. Plates were prepared by joining 2% agar, and colonies were developed overnight. Each colony was added to 25 mL of Luria–Bertani broth, letting it grow at a temperature of 37 °C while being shaken at 220 rpm for 10 to 12 h. We moved the pre-cultured plates with 25 mL fresh seed medium at 2% (v/v) and cultivated at 37 °C while shaking under 200 rpm for 7 h. The seed culture was merged in 100 mL of fermentation medium at a 10–15% inoculation rate. The incubated condition had the same temperature and shaking speed but lasted for 5 h. IPTG was added at a concentration of 100 mg/L to induce expression. After 15 h, we added IPTG again when the concentration was 50 mg/L. We collected samples every 3 to 5 h to measure the glucose concentration and pH levels. Glucose was supplemented as needed, and pH was maintained at 7.0–7.2 by adding KOH.

The seed culture medium contained 30 g/L glucose, 6 g/L yeast extract, 1.2 g/L citric acid, 1.2 g/L peptone, 1.4 g/L dipotassium hydrogen phosphate, 2.2 g/L ammonium sulfate, 0.5 g/L magnesium sulfate heptahydrate, 10 mL/L corn steep liquor, 15 mg/L ferrous sulfate, 2 mg/L manganese sulfate, 2 mg/L cobalt chloride, 2 mg/L zinc sulfate, 1.2 mg/L Vitamin H, and 1.6 mg/L Vitamin B1. The fermentation medium contained 30 g/L glucose, 2 g/L yeast extract, 1.5 g/L citric acid, 2 g/L peptone, 1.4 g/L dipotassium hydrogen phosphate, 3 g/L ammonium sulfate, 10 g/L magnesium sulfate heptahydrate, 20 mg/L ferrous sulfate, 5 mg/L manganese sulfate, 2 mg/L cobalt chloride, 2 mg/L zinc sulfate, 4 mg/L calcium chloride, 0.6 mg/L Vitamin H, 1.5 mg/L Vitamin B1, and 1 g/L calcium carbonate.

3.7. Analysis Methods

We figured out how much the concentrations of NAM, cytidine, NR, ATP, and NMN were using a high-performance liquid chromatography system, or HPLC for short. The Agilent 1290 system conducted the test on each sample. The system had a Hypersil GOLD™ aQ Dim column UV detector. The column has a 5 μm particle size, 250 mm length, and 4.6 mm inner diameter from “Thermo Fisher Scientific, Inc. Waltham, MA, USA”. A 0.05 M dipotassium phosphate solution served as the mobile phase, combined with acetonitrile, 2% v/v. The flux rate was 0.6 mL per minute. Samples were taken at 260 nm while keeping the column at 30 °C. We used a Bruker autoflex maX MALDI-TOF from “Thermo Fisher Scientific, Inc. Waltham, MA, USA” for calculating mass spectrometry, applying the same HPLC settings for NMN identification. Glucose concentration was monitored by an SBA-40c biosensor analyzer (Shandong Province Academy of Sciences, Jinan, Shandong, China).

We achieved fast NMN quantity detection with a fluorescence method to oversee both the cascade reaction and shake-flask fermenting trails. The approach added 70 μL of twenty percent acetophenone in dimethyl sulfoxide and 70 μL of two M KOH together. We combined them with 175 μL of the mix that had NMN in it. After a two-minute ice bath, we stirred in 315 μL of 88% formic acid. The reaction was subjected to incubation at a temperature of 37 °C for ten minutes. We put the 200 μL culture into a 96-well, flat-bottomed black plate. Fluorescence detection was executed with a Fluoroskan Ascent FL plate reader from “Thermo Fisher Scientific, Inc. Waltham, MA, USA”. The excitation wavelength was 390 nm, and the emission wavelength was 460 nm.

4. Conclusions

In this study, we created and refined a synthetic biological pathway that enhanced the production of nicotinamide mononucleotide (NMN) via microbial fermentation. The approach featured an engineered enzymatic cascade that changes the conversion of cytidine into NMN. The key enzymes, such as pyrimidine-nucleoside phosphorylase (PyNP) and nicotinate riboside kinase (NRK), served integrally in the pathway. The effective selection and enzyme evaluation, with NRK for its nicotinamide riboside synthetase activity, were important for building a reliable in vitro synthesis route. Cytidine was first converted into 1-phosphoribose under the action of PyNP and NRK, which turned into nicotinamide riboside (NR) and was phosphorylated to create NMN.

By adjusting the PH, changing the reaction time for enzyme expression in E. coli BL21 (DE3) and optimizing single-enzyme reactions secured the feasibility of the pathway. We also calculated the effectiveness of in vitro NMN synthesis through the enzymatic system. NR and NMN were achieved in a regulated and gradual process. Mass spectrometry and high-performance liquid chromatography (HPLC) further validated the identity and purity of the final product. The NMN biosynthesis pathway in a high-cytidine-producing E. coli strain was even supplied for intracellular NMN production. The initial yields were modest, but optimizing fermentation factors, particularly IPTG induction and PH adjustment, increased NMN titers.

In conclusion, this work underscores the importance of enzyme optimization and pathway engineering in microbial fermentation for NMN production. NMN demonstrates efficacy in supporting cellular energy production and delaying cellular aging, establishing itself as a next-generation star product in the field of nutritional supplements. By integrating systems biology and metabolic engineering techniques, this study provides a comprehensive and scalable strategy for the efficient biosynthesis of NMN, which could be adopted for future industrial-scale applications.