8
84
S. Schmitz et al. / Tetrahedron Letters 42 (2001) 883–885
Figure 2. Squalene cyclization to (a) g-polypodatetraene (1) by mutant SHC Leu607Lys and (b) to 17-isodammara-20(21),24-diene
2) by mutant SHC Phe605Lys.
(
intermediary carbocations. Furthermore, the free elec-
tron pairs at the OH-group of Tyr420, 609 and 612 may
stabilize the intermediate carbocations. Yet, to date all
mutagenesis experiments resulted in the simple
enhancement of unspecific product formation: hopene
is always the main product besides varying amounts of
tonation. A reason for the low reaction velocity seems
to be the fact that the water bound to the g-amino
group of Lys607 will impede the intruding squalene.
Hence, our results together with modelling show that
the Leu607 in wild type SHC is located near the bicyclic
carbocationic intermediate and may have an important
steric contribution to chaperone the appropriate con-
formation of squalene during hopene formation.
6–10
di- to tetracyclic side-products.
In contrast, in this paper we describe mutant SHCs
which induce a premature termination of the polycy-
clization reaction leading to bicyclic g-polypodatetraene
The second mutant SHC, namely Phe605Lys, produced
90% 17-isodammara-20(21),24-diene (2) (Fig. 2b)
besides 10% hopene; the velocity of the reaction was
again 200 times lower than that of the wild type. The
identity of 2 was demonstrated by the retention time in
GC and by the GC–MS spectrum which are both
(
1) or tetracyclic 17-isodammara-20(21),24-diene (2),
11
respectively as the main products.
Leu607 is located in the central part of the catalytic
cavity of SHC pointing with the methyl groups to ring
5
identical to published values. The product spectrum of
3
B and C of the modelled-in hopene skeleton (Fig. 1).
Phe605Lys is in contrast to SHC mutants Tyr609Phe
By substituting Leu by Lys at this position a premature
quenching of the intermediate carbocations (bicyclic or
and Phe365Ala: These mutants form 2 concomitantly
9,10
with other products.
The 17-isodammara-20(21),24-
3
tricyclic) was expected. The catalysis by the mutant
diene is a particularly interesting product because it has
1
SHC Leu607Lys is 200 times slower as compared to the
wild type.
yet not been found to occur in nature.
It has been proposed that Phe605 stabilizes cation C17,
which facilitates the ring expansion to the final six-
The GC–MS spectrum of the main product of mutant
3
SHC Leu607Lys was identical to that of g-polypodatet-
membered ring D. In mutant SHC Phe605Lys this
8,14,15
raene which is found in ferns (1 in Fig. 2a).
The
stabilization is missing. Therefore, the polycyclization
stops prematurely. Lys605 may also induce a specific
deprotonation at C21 leading to 2. On the other hand,
mutant SHC Phe605Ala led to an unspecific production
NMR-data of the isolated product (purity 96% accord-
ing to GC measurements) are in accordance with the
structure of g-polypodatetraene (1) and also with pub-
1
3–15
16
lished data.
The g-polypodatetraene made up 80%
of ten tri- to pentacyclic skeletons including 2. This
of the products. Additionally, traces of hopene and of
other not yet identified components were detected.
indicates that in this mutant the conformational control
and deprotonation specificity are severely disturbed.
12
The presented data allow a conclusion for the function
of Lys607. In the modelled structure of the mutant
,
SHC Lys607 comes approximately 3 A close to the
Acknowledgements
tertiary carbocation of the bicyclic intermediate (per-
sonal communication by Tanja Schulz, Institut f u¨ r
Technische Biochemie, Universit a¨ t Stuttgart). Assum-
ing that during catalysis Lys is not protonated, this
distance between Lys and the tertiary carbocation of
the bicylic intermediate should be sufficient for depro-
This work has been supported by the Deutsche
Forschungsgemeinschaft
(SFB
323)
and
Graduiertenkolleg Mikrobiologie at the University of
T u¨ bingen. Thanks are due to Andrea Meinhardt and
Prof. R. Kr a¨ mer for support and contribution.