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).