FIGURE 6 NMN titers produced by different Nampt (a), Prs (b), and Rbk (c) homologs. Error bars represent the standard deviation of three biological replicates. See Table S3 for more detailed information on NMN synthesis reactions.
With the most productive enzyme homologs in hand, we next stepped through a series of optimizations to activate their full potential for NMN biosynthesis. To obtain the optimal reaction conditions, the effects of temperature, pH, enzyme ratios, Mg2+, and ATP concentrations on NMN production were estimated. The influence of different temperatures on NMN production was investigated at 25−45 °C for 3 h. The highest titer of NMN was obtained at 40 °C (Figure 7a). The influence of pH (6.0−9.0) on NMN production was determined at 40 °C for 3 h. As shown in Figure 7b, the highest titer of NMN was obtained at pH 8.0. Then the effect of enzyme ratios was explored. The optimal enzyme ratios of SuNampt, MjPrs, and OkRbk were found to be 0.5:1:1 (Figure 7c). Then, response surface methodology was carried out to optimize the concentrations of Mg2+and ATP. 10 mM Mg2+ and 16 mM ATP were determined to be optimal (Table S4). After the above optimizations, the NMN titer was improved to 433 mg/L (Figure 7d, Table S4), with a yield of 64.8% (Figure S5a).
As the yield of NMN was satisfactory, we stopped optimizing the reaction conditions and then sought to determine if the NMN production could be further improved when doubling the concentrations of substrates and enzymes simultaneously. Unfortunately, it was found that increasing substrates and enzymes concentrations led to a notable decrease in yield of NMN, while the NMN titer increased slightly (Figure S5a). Combined with the results in Figure 7c, which indicated that increasing the ratio of SuNampt excessively was deleterious to NMN production, we therefore speculated that the decreased yield was possibly due to the feedback inhibition caused by PPi, which was the byproduct of Nampt-catalyzed reaction and thus would be rapidly accumulated when the amount of SuNampt was increased. The openness of the cell-free systems allowed us to examine this hypothesis and adjust the NMN biosynthetic reaction in a fast and facile manner. Frist, an extra 0.5 mM PPi was directly added to the NMN synthesis reaction. As expected, a further decrease in the NMN yield was observed (Figure S5b), indicating that PPi indeed has the inhibitory effects on NMN production. This result was well consistent with a recent study that suggested that PPi has feedback inhibition on Nampt (Ngivprom et al., 2022). Next, we tested whether the NMN production could be improved by removing the PPi inhibition. To do this, we took advantage of the convenient nature of CFPS to express the pyrophosphate-hydrolyzing enzyme, pyrophosphatase (EC 3.6.1.1) from E. coli (EcPPase) and then added it to the NMN synthesis reaction. In comparison to the control reaction, the addition of EcPPase produced by CFPS significantly improved the NMN production; the yield of NMN was increased from 41.3% to 64.1% (Figure S5b). Taken together, these results suggested that there is an inhibition effect of PPi on NMN production and decreasing this feedback inhibition is beneficial for NMN biosynthesis. Finally, EcPPase was expressed in E. coli and purified, and then its concentration was finely tuned for the highest NMN production. The results showed that the optimal concentration of EcPPase was 0.5 µM (Figure S6), and under this condition, the final titer of NMN was increased to 1213 mg/L, with a yield of 90.8% (Figure 7d, Figure S6), which was a more than 12-fold improvement of NMN titer over the initial setup (Figure 2c).