3.3 Verification of CYP5139G1 catalytic function by in
vitro enzymatic reaction
To verify the C-28 oxidase activity of CYP5139G1, we used in
vitro microsome reaction system as previously (Wang et al., 2018). The
microsomal fractions from yeast YL-T3 expressing CYP5139G1 (strain
CPR-CYP5139G1, Table S3) were incubated with HLDOA in the presence of
NADPH at 30°C for 2 h. LC-MS analysis of the reaction mixture revealed
that the CYP5139G1 containing microsome produced a new peak with the
same retention time as the purified compound A (DHLDOA). No
reaction was observed in boiled microsome sample or in that with void
plasmid (strain CPR-EV, Table S3) (Fig. 4a). The LC-MS fragmentation
pattern of the new peak included fragments with m/z ratios of 437
[M-2H2O+H]+ and 455
[M-H2O+H]+, which were consistent
with those of the purified DHLDOA (Fig. 4b). The results demonstrated
that CYP5139G1 catalyzed the conversion of HLDOA to DHLDOA.
We did not observe any hydroxylated products of HLDO and HLDA in the
strain CYP5150L8-P45 (Table S3). In order to understand whether
CYP5139G1 could catalyze C-28 hydroxylation of HLDOA derivatives - HLDO
and HLDA, we used CYP5139G1 containing microsome incubated with HLDO and
HLDA, respectively. As a result, no HLDO or HLDA oxidation was observed
compared with the boiled CYP5139G1 harboring microsome and the void
plasmid control (data not shown). Those evidences indicated CYP5139G1
had a strong substrate specificity to HLDOA.
Taking all the above facts together, we could conclude that CYP5139G1
was a HLDOA C-28 hydroxylase. By introducing CYP5139G1 into the yeast
chassis, the reconstructed GA biosynthetic pathway from glucose was
extended from HLDOA to DHLDOA. Here, both a novel GA and a new
hydroxylase were found (Fig. 5).