The stereochemical course of tricho-acorenol biosynthesis

Article abstract of DOI:10.1039/c3ob41755g

The biosynthesis of tricho-acorenol in Trichoderma harzianum was investigated by feeding stereospecifically deuterated mevalonolactone isotopomers, followed by a detailed GC/MS analysis of the incorporation of labelling. The results establish a highly ste

Full text of DOI:10.1039/c3ob41755g

Organic &
Biomolecular Chemistry
View Article Online
View Journal | View Issue
COMMUNICATION
The stereochemical course of tricho-acorenol
biosynthesis†
Cite this: Org. Biomol. Chem., 2013, 11,
7447
Christian A. Citron and Jeroen S. Dickschat*
Received 28th August 2013,
Accepted 17th September 2013
DOI: 10.1039/c3ob41755g
www.rsc.org/obc
The biosynthesis of tricho-acorenol in Trichoderma harzianum was
investigated by feeding stereospecifically deuterated mevalono-
lactone isotopomers, followed by a detailed GC/MS analysis of
the incorporation of labelling. The results establish a highly
stereospecific hydride migration and antarafacial attack of water
in the terminal step towards tricho-acorenol.
Fungi of the genus Trichoderma are well known to be plant
growth promoting symbionts. Due to their biological activity
against phytopathogenic bacteria, Trichoderma species are
used as biocontrol agents in agriculture.1,2 Several natural pro-
ducts are known from Trichoderma such as the cytotoxic tricho-
dermamides3 and harziphilone,4 the mycotoxin T-2 toxin,5
and a series of volatile pyrones6–9 and acorane sesquiter-
penes10,11 of unknown functions. One of the strongest produ-
cers of volatile terpenes, T. harzianum, showed the production
of large quantities of the sesquiterpenes tricho-acorenol 1 and
acorenone 2 (Fig. 1).11 The biosynthesis of these compounds
was investigated in feeding experiments with deuterated meva-
lonolactone isotopomers12 and proceeds via several cyclisation
steps in combination with three hydride shifts and water
capture to 1, while 2 requires one additional oxidation step
(Scheme 1).11
Fig. 1 Gas chromatogram of the headspace extract of T. harzianum.
In detail, the biosynthesis of 1 starts with the rearrange-
ment of farnesyl diphosphate 3 by extrusion and reattack of
the diphosphate moiety to nerolidyl diphosphate 4. Its reioni-
sation initiates the first cyclisation to the bisabolyl cation 5.
After a 1,2-hydride shift to the homobisabolyl cation 6 a
second cyclisation affords the acorenyl cation 7. As demon-
strated in previous experiments with deuterated mevalono-
lactones the following steps include a combination of a 1,4- and
a 1,2-hydride shift via cation 8 to 9 that is trapped by water to
yield 1.11 The third hydride shift in this biosynthetic scheme
Scheme 1 Biosynthesis of 1 and 2 in T. harzianum.
from 8 to 9 can occur suprafacially on either face of the six-
membered ring, followed by a final attack of water at C-3 that
may proceed supra- or antarafacially to the migrating hydride.
The stereochemical course of these late steps was investigated
by feeding of stereospecifically deuterated mevalonolactone
Technische Universität Braunschweig, Institut für Organische Chemie, Hagenring 30,
38106 Braunschweig, Germany. E-mail: j.dickschat@tu-bs.de; Fax: +49 531 3915272;
Tel: +49 531 5264
†Electronic supplementary information (ESI) available: Experimental procedures
and NMR spectra of synthetic compounds. See DOI: 10.1039/c3ob41755g
This journal is © The Royal Society of Chemistry 2013
Org. Biomol. Chem., 2013, 11, 7447–7450 | 7447

View Article Online
Communication
Organic & Biomolecular Chemistry
Scheme 3 Synthesis of deuterated mevalonolactone isotopomers. (a) NaH,
BnBr, 92%; (b) n-BuLi, (CH2O)n, 67%; (c) Red-Al (or LiAl2H4), NIS, 92%, (65%);
(d) Me2CuLi or [2H6]Me2CuLi, 82–86%; (e) Ti(OiPr)4, (−)-(D)-DIPT, 84–88%; (f)
LiAlH4 or LiAl2H4, 86–90%; (g) IBX, 86–87%; (h) 1. NaClO2, 2. H2, Pd/C, 45–83%
(2 steps).
Scheme 2 (A) Retro-Diels-Alder fragmentation of 1 in EI mass spectrometry;
(B) possible outcomes of feeding experiments with stereospecifically deuterated
mevalonolactone isotopomers.
between two alternative stereochemical courses (pathways a
and b in Scheme 2B).
For this purpose the mevalonolactones 22a–d were syn-
thesised via known 1715 (Scheme 3). Homopropargyl alcohol
13 was protected as benzyl ether 14 that was deprotonated
with n-BuLi followed by reaction with para-formaldehyde to
yield alcohol 15. Reduction of the alkyne with Red-Al and
quenching with N-iodosuccinimide gave the vinyl iodide 16a.
When Red-Al was replaced with LiAlD4 deuterated 16b was
obtained. The alkenyl iodides 16 were alkylated with Me2CuLi
resulting in 17a and 17b. Alternatively, deuterated [2H6]-
Me2CuLi was used to yield the isotopomers 17c and 17d.
Sharpless epoxidation resulted in the epoxides 18a–d with
95% ee based on Mosher analysis16 (reported ee: 94%).17
Reductive opening of the epoxide with LiAlH4 or LiAlD4 gave
the diols 19 with stereospecific labelling. The diols were oxi-
dised to the aldehydes 20 followed by Pinnick oxidation to the
acids 21 that were used in the next step without purification.
Finally, hydrogenolysis of the benzyl protecting group afforded
isotopomers, and the results of these investigations are reported
here.
The biosynthesis of 1 was investigated by feeding of deuter-
ated mevalonolactone isotopomers. Incorporation of labelling
was followed by capturing the volatiles on charcoal using a
closed-loop stripping apparatus (CLSA) and GC/MS analysis of
the obtained headspace extracts. The EI mass spectrum of 1
exhibits a major fragment ion at m/z = 84 that arises from
Retro-Diels Alder (RDA) fragmentation (Scheme 2A). This frag-
ment ion can be used as a probe for the stereochemical
problem under investigation. During FPP cyclisation to 1
either the original 2-pro-R or the 2-pro-S hydrogen of mevalono-
lactone, ending up in stereochemically defined positions of
313,14 and at C-3 of cation 8, is shifted to the adjacent C-2 in 9
(Scheme 2B). Deuterated mevalonolactones with specific label-
ling in the 2-pro-R and 2-pro-S positions were used, leading to
diagnostic fragment ions at m/z = 84 or m/z = 85, to distinguish
7448 | Org. Biomol. Chem., 2013, 11, 7447–7450
This journal is © The Royal Society of Chemistry 2013

View Article Online
Organic & Biomolecular Chemistry
Communication
Fig. 2 Total ion chromatograms of the headspace extracts from Trichoderma
harzianum. (A) Total ion chromatogram of a headspace extract obtained in the
feeding experiment with 22b; (B) feeding experiment with 22d. Comparable
results were obtained in the feeding experiments with 22a and 22c,
respectively.
the mevalonolactones 22 in high yields of 8–32% (9 steps,
based on the minimum and maximum yields for each step).
The synthetic compounds were fed to agar plate cultures of
T. harzianum. The emitted volatiles were collected using the
CLSA technique18 and the obtained headspace extracts were
analysed by GC/MS. It is well known that deuterated com-
pounds can be gas chromatographically separated from their
non-deuterated analogs,19 and this separation is better the
higher the deuterium content is. This is one of the major
advantages of the GC/MS technique in analysing the incorpor-
ation of deuterium labelled precursors into volatile natural
products, because a “clean” mass spectrum of the deuterated
compound that is not superimposed with the mass spectrum
of the unlabelled analog can be obtained, allowing for a
detailed analysis. Although feeding of 22a and 22b resulted in
the incorporation of up to three deuterated isoprene units into
1 (Fig. 2A), the low differences in the deuterium content of the
various isotopomers of 1 were insufficient for a gas chromato-
graphic separation which prohibited unambiguous data
interpretation. To circumvent this problem the mevalono-
lactones 22c and 22d with an additional deuterium labelling
in the methyl groups were synthesised and fed to the fungus.
The higher deuterium content of four deuterium atoms per
isoprene unit resulted in a clear separation of the isotopomers
of 1 in GC according to their deuterium content (Fig. 2B),
allowing for an accurate data interpretation.
Fig. 3 Mass spectra of 1. (A) Mass spectrum of authentic 1;20 (B) mass spec-
trum of [2H12]-1 obtained after feeding of 22d; (C) mass spectrum of [2H12]-1
obtained after feeding of 22c.
2
feeding of 22d (HS = H, R = C2H3) the molecular ion for the
isotopomer with the highest deuterium content of 1 is shifted
to m/z = 234, indicating that up to three isoprene units each
carrying four deuterium atoms were incorporated (Fig. 3B).
The diagnostic fragment ion at m/z = 84 is shifted to m/z = 87
due to the incorporation of three deuterium atoms. This
finding is in agreement with pathway a, but not with pathway
2
b in Scheme 2. In a complementary experiment, 22c (HR = H,
R = C2H3) was fed to T. harzianum, also resulting in the incor-
poration of up to twelve deuterium atoms (Fig. 3C). The diag-
nostic Retro-Diels-Alder fragment ion is observed at m/z = 88,
again in full agreement with pathway a and excluding pathway
b of Scheme 2. These observations clearly demonstrate that
the final of a series of three hydride shifts during the bio-
synthesis of 1 proceeds stereospecifically with migration of the
original 2-pro-S hydrogen (HS) of mevalonolactone. The sub-
sequent nucleophilic attack of water at the cationic centre
The resulting mass spectra from the feeding experiments
with 22c and 22d are presented in Fig. 3. The mass spectrum
of natural 1 shows a molecular ion at m/z = 222 (Fig. 3A). After
This journal is © The Royal Society of Chemistry 2013
Org. Biomol. Chem., 2013, 11, 7447–7450 | 7449

View Article Online
Communication
Organic & Biomolecular Chemistry
occurs antarafacially at C-3 with respect to the migrating
hydride.
3 E. Garo, C. M. Starks, P. R. Jensen, W. Fencial,
E. Lobkovsky and J. Clardy, J. Nat. Prod., 2003, 66, 423.
4 J. Qian-Cutrone, S. Huang, L. P. Chang, D. M. Pirnik,
S. E. Klohr, R. A. Dalterio, R. Hugill, S. Lowe, M. Alam and
K. F. Kadow, J. Antibiot., 1996, 49, 990.
5 J. R. Bamburg and F. M. Strong, Phytochemistry, 1969, 8,
2405.
6 R. P. Collins and A. F. Halim, J. Agric. Food Chem., 1972, 20,
437.
7 M. O. Moss, R. M. Jackson and D. Rogers, Phytochemistry,
1975, 14, 2706.
8 N. Claydon, M. Allan, J. R. Hanson and A. G. Avent, Trans.
Br. Mycol. Soc., 1987, 88, 503.
9 S. M. Wickel, C. A. Citron and J. S. Dickschat, Eur. J. Org.
Chem., 2013, 2906.
Conclusions
In summary the stereochemical course of the final hydride
migration in tricho-acorenol biosynthesis was followed by
feeding experiments with stereospecifically deuterated
mevalonolactone isotopomers that were obtained by chemical
synthesis. Careful analysis of the incorporation of isotope
labellings by GC/MS established that the final 1,2-hydride
migration proceeds specifically with shifting of the original
2-pro-S hydrogen (HS) from mevalonolactone, followed by a
nucleophilic attack of water at C-3 in an antarafacial manner
with regard to the migrating hydride. Further mechanistic 10 Q. Huang, Y. Tezuka, Y. Hatanaka, T. Kikuchi, A. Nishi and
studies on terpene biosynthetic pathways are now possible to K. Tubaki, Chem. Pharm. Bull., 1995, 43, 1035.
gain a deeper understanding of terpene cyclases as some of 11 C. A. Citron, R. Rîclea, N. L. Brock and J. S. Dickschat, RSC
the most fascinating enzymes in natural product biosynthesis.
Adv., 2011, 1, 290.
12 J. S. Dickschat, C. A. Citron, N. L. Brock, R. Riclea and
H. Kuhz, Eur. J. Org. Chem., 2011, 3339.
13 J. W. Cornforth, R. H. Cornforth, G. Popjak and
L. Yengoyan, J. Biol. Chem., 1966, 241, 3970–3987.
Acknowledgements
This work was funded by the Deutsche Forschungs- 14 J. S. Dickschat, Nat. Prod. Rep., 2011, 28, 1917.
gemeinschaft with an Emmy Noether grant (DI1536/1-3) and a 15 B. Ganem, Y. Dong, Y. F. Zheng and G. D. Prestwich, J. Org.
Heisenberg grant (DI1536/4-1). We thank Gabriele König
Chem., 1999, 64, 5441.
(Bonn) for Trichoderma harzianum 714 and Stefan Schulz 16 J. A. Dale and H. S. Mosher, J. Am. Chem. Soc., 1973, 95,
(Braunschweig) for his support.
512.
17 C. J. Hollowood, S. Yamanoi and S. V. Ley, Org. Biomol.
Chem., 2003, 1, 1664.
18 K. Grob and F. Zürcher, J. Chromatogr., 1976, 117, 285.
19 J. S. Dickschat, S. C. Wenzel, H. B. Bode, R. Müller and
S. Schulz, ChemBioChem, 2004, 5, 778.
20 N. L. Brock and J. S. Dickschat, Eur. J. Org. Chem., 2011,
5167.
Notes and references
1 G. E. Harman, C. R. Howell, A. Viterbo, I. Chet and
M. Lorito, Nat. Rev. Microbiol., 2004, 2, 43.
2 G. E. Harman, Phytopathology, 2006, 96, 190.
7450 | Org. Biomol. Chem., 2013, 11, 7447–7450
This journal is © The Royal Society of Chemistry 2013