1
568
N. Meitinger et al. / Tetrahedron Letters 57 (2016) 1567–1571
some conflicting data have been reported.8,10,11 For 3KSIs of plant
origin no sequence information is provided in the public databases.
We can only assume that plant 3KSI is not an ortholog of the bac-
terial 3KSI enzymes because homology searches using bacterial
the isoprogesterone:progesterone ratio is clearly shifted toward
progesterone after 3KSI addition. In comparison to the high iso-
merization rate promoted by 3KSI, the non-enzymatic production
of progesterone was not significant when tested under the same
conditions as the enzymatic (3KSI) assay. Bacterial 3KSIs belong
3
KSI sequences were not successful in the Arabidopsis thaliana
12
3
genome or the Digitalis purpurea transcriptome. Recently, the
first plant 3KSI has, however, been purified from Digitalis lanata
by our group and the characterization of the purified enzyme
revealed that the D. lanata 3KSI resembles the bacterial enzymes.
Dl3KSI was furthermore clearly separated from the Dl3bHSD by
size exclusion chromatography.12
to the most efficient enzymes. Dl3KSI also has a high catalytic effi-
ciency, but it could not be determined here because only partially
purified Dl3KSI preparations were used. As soon as a recombinant
form of Dl3KSI is available we will address this question in some
detail.
ESI-LC–MS analysis was used to determine whether deuterium
was incorporated from the medium into progesterone during the
isomerization reactions. Hence, the ratio of the parent m/z 315
We here intended to demonstrate that the isomerization
catalyzed by the D. lanata 3KSI involves an intramolecular proton
shift to promote isoprogesterone isomerization as had been
demonstrated previously for the Comamonas enzyme. Moreover,
we want to add more evidence to support the assumption that
+
+
[M+H] and m/z 316 [M+1+H] ions detected in the positive ioniza-
tion mode analysis was calculated for progesterone formed under
the different assay conditions (Table 1). The mass spectrum of
3
KSI is not associated with 3bHSD in D. lanata. Therefore, we
progesterone incubated for 30 min at 37 °C with 0.5 lL HCL in
intended to show that in enzyme-free controls and in assays
containing rDl3bHSD isoprogesterone is only isomerized non-
enzymatically and consequently, that the progesterone detected
after incubation is only formed by chemical isomerization.
deuterium oxide as well as the mass spectra of progesterone
formed in the 3KSI assay at 37 °C, the rDl3bHSD assay at 50 °C
and the assay containing both enzymes are shown in Figure 2 as
examples.
The 315/316 ion ratio calculated for progesterone dissolved in
water (1.82) corresponded well with the ion ratio calculated for
Results and discussion
The isomerization of isoprogesterone into progesterone was
investigated in deuterium oxide under the following conditions:
Isoprogesterone
Progesterone
(
a) enzyme catalysis with recombinant 3b-hydroxysteroid dehy-
1
00
drogenase from Digitalis lanata (rDl3bHSD) using pregnenolone
and NAD+ as the substrate and cosubstrate, respectively; (b)
enzyme catalysis with partially purified 3-ketosteroid isomerase
from D. lanata (Dl3KSI) using isoprogesterone as the substrate; (c)
enzyme catalysis with both rDl3bHSD and partially purified Dl3KSI
80
6
4
0
0
+
using pregnenolone and NAD as the substrate and cosubstrate,
respectively; (d) spontaneous isomerization of isoprogesterone in
deuterium oxide without enzyme and (e) acid catalyzed isoproges-
terone isomerization. Assays were then analyzed by HPLC/DAD/MS.
This analysis allows us to quantify how much progesterone was
formed in the different assays (UV/vis analysis) as well as to analyze
if deuterium was incorporated into progesterone during the iso-
merization reaction (ESI-LC–MS analysis). The isoprogesterone
20
0
1
2
3
4
5
6
7
8
9
used here was synthetized from pregnenolone (Sigma–Aldrich,
Munich, Germany) according to Djerassi et al.13 and purified by
Figure 1. Conversation rate of isoprogesterone into progesterone determined for
the different assay conditions by HPLC–UV/vis analysis. Mean ± SD of triplicates. 1—
rDl3bHSD enzyme assay at 22 °C; 2—rDl3bHSD enzyme assay at 37 °C; 3—rDl3bHSD
enzyme assay at 50 °C; 4—rDl3bHSD + Dl3KSI assay at 37 °C; 5—Dl3KSI enzyme
assay at 22 °C; 6—Dl3KSI enzyme assay at 37 °C; 7—Dl3KSI enzyme assay at 50 °C;
8—non-enzymatic isomerization; 9—acid catalyzed isomerization.
preparative TLC as specified in Ref. 12. The structure of isoproges-
terone was verified by NMR analysis employing the NMR data
14
reported for pregnenolone as reference for obvious assignments
1
3
1
of C and H signals. Because to our knowledge the NMR data of
isoprogesterone have not been published yet the chemical shifts
for all carbon atoms and protons of isoprogesterone are listed along
1
3
1
2
with the C, H and D NMR spectra in the Supplementary data.
It has already been demonstrated that the addition of a crude
Dl3KSI preparation to a rDl3bHSD activity test clearly shifted the
Table 1
Ratio of the parent m/z 315 [m+H] and m/z 316 [M+1+H]+ ions of progesterone
+
formed under different assay conditions
1
2
ratio of isoprogesterone to progesterone toward progesterone.
Sample
315/316 ion ratio
UV/vis detection of isoprogesterone and progesterone present in
the different assays described above confirmed these results
Progesterone, RTa
1.82
b
Progesterone, 37 °Cb
1.86 ± 0.045
1.87 ± 0.037
0.19 ± 0.016
0.27 ± 0.011
0.30 ± 0.010
1.43 ± 0.019
1.65 ± 0.073
2.26 ± 0.051
2.08 ± 0.040
Not detectable
0.44 ± 0.014
Progesterone, 50 °C
(Fig. 1). The results of rDl3bHSD activity tests which were per-
b
b
3
3
3
bHSD assay (d), 50 °C
formed at different temperatures showed that progesterone forma-
tion was significantly enhanced with increasing temperature. This
was explained with an enhanced chemical isomerization at higher
temperatures. When 3KSI alone was present in the assays, highest
progesterone formation was seen at 37 °C which matches the tem-
perature optimum of this enzyme.12 Progesterone formation was
slightly lower in the assays containing rDl3bHSD and Dl3KSI than
in the assays with Dl3KSI alone. This can be explained by the fact
that both enzymes are active and consequently both, isoproges-
terone and progesterone are formed until the assay is stopped.
But still, the combination of both enzymes again illustrates that
bHSD assay (d), 37 °Cb
bHSD assay (d), 22 °C
b
3bHSD + 3-KSI assay (e)
3-KSI assay (a), 50 °C
b
b
3
-KSI assay (a), 37 °Cb
3-KSI assay (a), 22 °C
b
Non-enzymatic isomerization (b)
Acid catalyzed isomerization (c)b
Samples were analyzed by ESI-LC–MS in the positive ion mode. Mean ± SD of
triplicate.
a
Sample in H
Sample in deuterium oxide.
2
O.
b