Gold(I)-Catalyzed Domino Cyclizations of Diynes for the Synthesis of Functionalized Cyclohexenone Derivatives. Total Synthesis of (-)-Gabosine H and (-)-6-epi-Gabosine H

Article abstract of DOI:10.1021/acs.orglett.6b01898

1,6-Diynes with a t-butylcarbonate group in the propargylic position undergo gold(I)-catalyzed domino-cyclization which affords α-hydroxycyclohexenones. The described sequence can be applied on functionalized, highly oxygenated substrates, as examplified

Full text of DOI:10.1021/acs.orglett.6b01898

Letter
pubs.acs.org/OrgLett
Gold(I)-Catalyzed Domino Cyclizations of Diynes for the Synthesis of
Functionalized Cyclohexenone Derivatives. Total Synthesis of
(−)-Gabosine H and (−)-6-epi-Gabosine H
Bojan Vulovic,*,† Dusan Kolarski,† Filip Bihelovic,*,† Radomir Matovic,‡ Maja Gruden,†
and Radomir N. Saicic*,†
†Faculty of Chemistry, University of Belgrade, Studentski trg 16, P.O. Box 51, 11158 Belgrade, Serbia
‡ICTM Center for Chemistry, Njegoseva 12, 11000 Belgrade, Serbia
S
* Supporting Information
ABSTRACT: 1,6-Diynes with a t-butylcarbonate group in the
propargylic position undergo gold(I)-catalyzed domino-
cyclization which affords α-hydroxycyclohexenones. The
described sequence can be applied on functionalized, highly
oxygenated substrates, as examplified in the synthesis of
(−)-gabosine H and its epimer.
Scheme 1. Principle of Au(I)-Catalyzed Domino Cyclization
of 1,6-Diynes
educing the number of synthetic steps, as well as the time
and effort associated with workup procedures, minimizing
R
the use of solvents, and maximizing the atom economy of the
process remain constant challenges in organic synthesis. These
issues can be simultaneously addressed by use of domino
reactions,1 which allow for rapid increase of molecular
complexity in a single synthetic step. Reactions catalyzed by
gold complexes are particularly well suited for sequencing, as the
rich organogold chemistry involves a number of reactive
intermediates (carbocations, carbenes, or organometallics) and
the possibility of a mechanistic crossover.2 Gold complexes are
best known for their ability to trigger addition reactions to
alkynes.3 This property was exploited in the design of domino
sequences, which convert substrates with suitably positioned
unsaturated moieties into cyclic structures.
of the reaction conditions included screening of the solvent
(THF, DCM, DCE, MeCN, PhMe, DMF), ligand (PPh3,
P(C6F5)3, XPhos, JohnPhos, tri-o-furylphosphine, tris(2,6-di-t-
Bu-phenyl) phosphite), and additives (AgBF4, AgSbF6, AgPF6,
MeSO3H, MS 4 Å). We found that the catalytic cocktail involving
5 mol % of JohnPhosAuCl procatalyst with 20 mol % of AgSbF6
in DCE allows the domino cyclization to be accomplished at rt.
A series of 1,6-diynes 1a−i were then prepared and submitted
to the optimized reaction conditions. The results of these
experiments are presented in Table 1. In most examples, the
desired cyclohexenone derivatives were obtained in moderate/
good yields. Both secondary (entries 1, 5, 6, and 7) and tertiary
(entries 4, 8, and 9) propargylic carbonates can be used.
Unfortunately, the reaction did not work with nonterminal
alkynes (entries 2 and 3). Notably, no isomerization/
aromatization of hydroxycyclohexenones into the corresponding
phenols (their thermodynamically more stable redox equiv-
Some time ago, Gagosz and Buzas reported the gold(I)-
catalyzed cyclization of propargyl tert-butyl carbonates to cyclic
carbonates.4 We envisaged an extension of this reaction into a
tandem cyclization that would eventually provide α-hydrox-
ycyclohexenone derivatives.5 The blueprint of the trans-
formation is represented in Scheme 1. The starting compound
would be 1,6-diyne of type 1, and the sequence would start as
previously described to give a cyclic enol carbonate 2 which
should then undergo the second 6-exo-cyclization with the
formation of a carbon−carbon bond. The expected product of
the overall reaction would be bicyclic diene 3, a latent form of a
thermodynamically more stable α-hydroxycyclohexenone 4.
The feasibility of this concept was tested on diyne 1a as a
conformationally flexible, Thorpe−Ingold effect-free substrate.
In the first experiment, diyne 1a was treated with 5 mol % of
Ph3PAuNTf2 catalyst, at rt, in dichloromethane solution, which
resulted in quantitative conversion into cyclic carbonate of type
2a (within 30 min). However, increasing the temperature to 50
°C promoted a second reaction that was complete within 4 h and
afforded cyclohexenone derivative 4a in 55% yield. Optimization
Received: June 30, 2016
© XXXX American Chemical Society
A
DOI: 10.1021/acs.orglett.6b01898
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
number of family members has steadily increased. Although they
have shown modest biological activity so far, these carbasugar-
type molecules have attracted considerable interest from
synthetic organic chemists, and the subject has been recently
reviewed.10 Gabosine H (5) was previously synthesized by
Prasad and Kumar.11 Our approach to the synthesis of this
compound is outlined in Scheme 2. According to this plan, the
Table 1. Gold(I)-Catalyzed Domino Cyclizations of 1,6-
Diynes
Scheme 2. Retrosynthetic Analysis of Gabosine H (5)
trihydroxycyclohexenone framework should be directly as-
sembled from the corresponding diyne 6; this chiral synthon
should be derived from D-(−)-tartaric acid.
The synthesis of the cyclization precursors was accomplished
as represented in Scheme 3. The known aldehyde 8 was obtained
Scheme 3. Synthesis of the Cyclization Precursors 11a and
11b
from D-(−)-diethyl tartrate (38% over seven steps).12 The
conversion of aldehyde 8 into propargylic alcohol 9 was effected
by acetylide addition. This reaction was performed according to
several experimental protocols13 but, unfortunately, with an
unsatisfactory stereochemical outcome: either an equimolar
mixture of diastereoisomers was obtained or the undesired
isomer (leading to gabosine H epimer) predominated. Unstable
diynes 9 were not isolated but converted into the corresponding
Boc derivatives 10. These latter two compounds were
inseparable; therefore, they were converted into diastereoiso-
meric enol carbonates 11a and 11b and separated by column
chromatography.
The cyclization was first attempted with diastereoisomer 11a,
leading to epi-gabosine H (epi-5; Scheme 4). Upon exposure to
the previously established conditions, 11a gave a mixture of three
compounds, 12, 13, and 14, in 44% combined yield. All three
compounds were the products of 6-exo-cyclization, with
differences in the functionalization pattern. Whereas enone 14
was the expected product of the cyclization/isomerization/
hydrolysis sequence, the other two (i.e., 12 and 13) were acetals:
apparently, the reaction proceeds via a cabocationic intermediate
15, which can be intercepted by the product 14 (to give acetal
13) but which is reactive enough to break the etheral C−O bond
and pluck the oxygen nucleophile from the product 14 (or the
substrate 11a) to give acetal 12. To suppress this side reaction,
we performed the cyclization in the presence of water as an
external nucleophile: to our pleasure, a mixture of cyclo-
hexenones 16 and 14 was obtained in 70% combined yield.
Remarkably, the cyclization was faster than the gold-catalyzed
hydration of alkyne. Compound 16 could be converted into
conjugated enone 14 on treatment with a hot aqueous solution of
a
b
Ph3PAuNTf2 was used as a catalyst. JohnPhosAu(MeCN)SbF6 was
used as a catalyst at 55 °C.
alents) was observed under the reaction conditions. The
corresponding 1,5- and 1,7-diynes did not yield any cyclo-
alkenone derivatives, indicating that the 5-exo- and 7-exo-
cyclization as a second step are not feasible. The reactions of
diastereoisomeric diynes 1h and 1i are worthy of note from a
stereochemical viewpoint: whereas compound 1h gave stereo-
chemically pure cis product 4h (entry 8), the diastereoisomeric
substrate 1i produced a mixture of cis/trans decalones in 4h
(cis):4i (trans) = 1.8:1 ratio (entry 9). This could be explained by
a reversible Meyer−Schuster rearrangement of compound 1i,
where the equilibration of the tertiary stereogenic center in 4i is
effected via the corresponding allene.
Gold complexes are both carbo- and oxophilic. Therefore, we
wondered whether the described reaction sequence would
tolerate a relatively high level of functionalization of precursors
and thus be suitable for syntheses of highly oxygenated
cyclohexenone derivatives. We decided to test the synthetic
applicability of this domino reaction in the total synthesis of
gabosines. When this research was underway,6 Li, Luo, Yang, and
collaborators reported the sequential cyclization of 1,7-diynes
leading to exo-methylene cyclohexene derivatives and applied it
in an elegant synthesis of a series of drimane sesquiterpenoids.7
Later on, this principle was exploited in a very efficient
enantioselective synthesis of the cladiellin family of natural
products.8
Gabosines are secondary metabolites of Streptomyces strains.
Since the initial isolation of the first compound of the class,9 the
B
DOI: 10.1021/acs.orglett.6b01898
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Scheme 4. Cyclization of 11a and Synthesis of (−)-epi-Gabosine H (epi-5)
Scheme 5. Cyclization of 11b and Synthesis of (−)-Gabosine H (5)
p-TsOH in i-PrOH. An even higher yield of synthetically useful
intermediates was obtained when the reaction was performed in
the presence of 2-propanol: in this case, the mixture of cyclic
products 17, 16, and 14 was obtained in 91% combined yield.
This mixture was not separated, or hydrolyzed, but directly
treated with BCl3 to give (−)-epi-gabosine H (epi-5) in 67%
yield. It should be noted that acetal 17 could be selectively
converted into either cyclohexenone 16 or 14.
protecting group, instead of benzyl ethers) the product of
reaction was furan 23 (for the proposed mechanism of this
transformation, see the SI). This example shows that gold(I)
complexes can induce deprotection, or anchimeric assistance, of
suitably positioned protecting groups via transient carbocationic
species.
We believe that, in the cyclization reaction, gold(I) complex
acts as a carbophilic π-acid and that the reaction proceeds via a
carbocationic intermediate of type 15, as confirmed by the
interception of this intermediate by external nucleophile, such as
2-propanol. However, reactions on structurally related systems
have been shown in the literature to involve carbene
intermediates and gold acetylides.4,5d,e,14 In order to test the
possible intermediacy of these species, DFT calculations were
performed on the model systems. Our results clearly indicate the
dominant role of mononuclear gold species, with more than
double reaction barriers in comparison to the dinuclear
complexes (Figure S1 and Table S1). A detailed study of the
gold-catalyzed enyne cycloisomerization reaction on mono-
nuclear complexes (Figure 1) shows that cycloisomerization
from a chairlike transition state (TS 1) leads to a structure that
can be described as a gold-coordinated carbocation intermediate
(Int 1), while boatlike and cyclopropanoid TSs (TS 2 and TS 3)
form almost identical gold-stabilized singlet carbene intermedi-
ates (Int 2 and Int 3). The energy difference between both
activation energies and intermediate products of model systems
is sufficient to assert that the gold-carbocation intermediate is
predominant and governs the proposed reaction mechanism.
To summarize, a gold(I)-catalyzed domino cyclization of 1,6-
diynes offers an efficient access to 6-hydroxycyclohexenones.
The reaction sequence can be applied on highly oxygenated
systems, as shown in the syntheses of (−)-gabosine H and 6-epi-
(−)- gabosine H.
The cyclization of the second diastereoisomer (11b), leading
to gabosine H (5), was found to be more difficult. Whereas the
cyclization of 11a was complete within 9 h, the conversion of 11b
under the similar conditions required 3 days and produced a
mixture of four products (Scheme 5). By analogy with the former
protocol, the reaction mixture was not separated but directly
submitted to deprotection with BCl3. In this case, the required
product, (−)-gabosine H (5), was obtained in 25% overall yield.
A more detailed analysis of the reaction mixture showed that the
desired products (i.e., 18 and 19) were accompanied by ketone
20 and enal 21 (in quantities of up to 30% each). Apparently,
slow cyclization and the long reaction time gave rise to the side
reaction, hydration of the triple bond of the substrate 11b. Thus,
whereas this synthetic study showed that the gold-catalyzed
domino cyclization of 1,6-diyne can be successfully applied to
highly functionalized, oxygenated substrates, it also indicated
that the stereochemical aspects can have an important effect on
the efficiency of the cyclization. The proper choice of the
protecting groups is also important, as shown in Scheme 6: when
the cyclization was attempted with enol carbonate 22 (dioxolane
Scheme 6. Cyclization of a Dioxolane-Protected Substrate 22
into Furan 23
C
DOI: 10.1021/acs.orglett.6b01898
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Figure 1. Calculated geometries and transition states for the enyne cycloisomerization step. Gibbs free energies (in kcal/mol), obtained using M06/
TZ2P level of theory, are given relative to the React 2 (products and reactants), while the barriers are given relative to the appropriate reactant structure.
(3) (a) Dorel, R.; Echavarren, A. Chem. Rev. 2015, 115, 9028−9072.
(b) Fensterbank, L.; Malacria, M. Acc. Chem. Res. 2014, 47, 953−965.
(4) (a) Buzas, A. K.; Istrate, F. M.; Gagosz, F. Tetrahedron 2009, 65,
1889−1901. (b) Buzas, A. K.; Gagosz, F. Org. Lett. 2006, 8, 515−518.
(5) For selected examples of reactions of structurally related substrates
that give another type of products, see: (a) Lim, C.; Kang, J.-E.; Lee, J.-
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.orglett.6b01898.
■
S
Experimental procedures; spectral data and NMR spectra
for all compounds; mechanistic explanation of the reaction
shown in Scheme 6 (PDF)
́
E.; Shin, S. Org. Lett. 2007, 9, 3539−3542. (b) Barabe, F.; Levesque, P.;
Korobkov, I.; Barriault, L. Org. Lett. 2011, 13, 5580−5583. (c) Chen, Y.;
Chen, M.; Liu, Y. Angew. Chem., Int. Ed. 2012, 51, 6181−6186.
(d) Chen, Y.; Chen, M.; Liu, Y. Angew. Chem., Int. Ed. 2012, 51, 6493−
6497. (e) Rao, W.; Koh, M. J.; Li, D.; Hirao, H.; Chan, P. W. H. J. Am.
Chem. Soc. 2013, 135, 7926−7932. (f) Lauterbach, T.; Gatzweiler, S.;
AUTHOR INFORMATION
Corresponding Authors
■
Nosel, P.; Rudolph, M.; Rominger, F.; Hashmi, A. S. K. Adv. Synth. Catal.
̈
2013, 355, 2481−2487. (g) Yan, J.; Tay, G. L.; Neo, C.; Lee, B. R.; Chan,
P. W. H. Org. Lett. 2015, 17, 4176−4179. (h) Wildermuth, R.; Speck, K.;
Magauer, T. Synthesis 2016, 48, 1814−1824.
*E-mail: bvulovic@chem.bg.ac.rs.
*E-mail: filip@chem.bg.ac.rs.
*E-mail: rsaicic@chem.bg.ac.rs.
(6) Kolarski, D. Diploma Work, Faculty of Chemistry, University of
Belgrade, 2011.
Notes
(7) Shi, H.; Fang, L.; Tan, C.; Shi, L.; Zhang, W.; Li, C.-c.; Luo, T.;
Yang, Z. J. Am. Chem. Soc. 2011, 133, 14944−14947.
The authors declare no competing financial interest.
(8) Yue, G.; Zhang, Y.; Fang, L.; Li, C.-c.; Luo, T.; Yang, Z. Angew.
Chem., Int. Ed. 2014, 53, 1837−1840.
ACKNOWLEDGMENTS
■
This work was supported by the Serbian Ministry of Education,
Science and Technological Development (Project No. 172027)
and by the Serbian Academy of Sciences and Arts (Project No.
F193). We acknowledge the support of the FP7 RegPot project
FCUB ERA GA No. 256716. The EC does not share
responsibility for the content of the article.
(9) Tsuchiya, T.; Mikami, M.; Umezawa, H.; Naganawa, H. J. Antibiot.
1974, 27, 579−586. (b) Takeuchi, T.; Chimura, H.; Hamada, M.;
Umezawa, H.; Yoshka, H.; Oguchi, N.; Takahashi, Y.; Matsuda, A. J.
Antibiot. 1975, 28, 737−742. (c) Muller, A.; Keller-Schierlein, W.;
̈
Bielicki, J.; Rak, G.; Stumpfel, J.; Zahner, H. Helv. Chim. Acta 1986, 69,
1829−1832.
̈
̈
(10) Mac, D. H.; Chandrasekhar, S.; Gree, R. Eur. J. Org. Chem. 2012,
2012, 5881−5895.
REFERENCES
■
(11) Prasad, K. R.; Kumar, S. M. Synlett 2011, 2011, 1602−1604.
(12) Fourriere, G.; Leclerc, E.; Quirion, J.-C.; Pannecoucke, X. J.
Fluorine Chem. 2012, 134, 172−179.
(1) For the most recent book and review articles, see: (a) Domino
Reactions: Concepts for Efficient Organic Synthesis; Tietze, L. F., Ed.;
Wiley-VCH: Singapore, 2014. (b) Ardkhean, R.; Caputo, D. F. J.;
Morrow, S. M.; Shi, H.; Xiong, Y.; Anderson, E. A. Chem. Soc. Rev. 2016,
45, 1557−1569. (c) Volla, C. M. R.; Atodiresei, I.; Rueping, M. Chem.
Rev. 2014, 114, 2390−2431. (d) Pellissier, H. Chem. Rev. 2013, 113,
442−524. (e) Clavier, H.; Pellissier, H. Adv. Synth. Catal. 2012, 354,
3347−3403.
(13) (a) Midland, M. M.; Gallou, F. Lithium acetylide; e-EROS
Encyclopedia of Reagents for Organic Synthesis; John Wiley and Sons,
2006. (b) Midland, M. M. Propynylmagnesium bromide; e-EROS
Encyclopedia of Reagents for Organic Synthesis; John Wiley and Sons,
2001. (c) Paquette, L. A. Cerium(III) chloride; e-EROS Encyclopedia of
Reagents for Organic Synthesis; John Wiley and Sons, 2006. (d) Anand, N.
K.; Carreira, E. M. J. Am. Chem. Soc. 2001, 123, 9687−9688.
(14) (a) Marion, N.; Nolan, P. Angew. Chem., Int. Ed. 2007, 46, 2750−
2752. (b) Odabachian, Y.; Le Goff, X. F.; Gagosz, F. Chem. - Eur. J. 2009,
15, 8966−8970. (c) Simonneau, A.; Jaroschik, F.; Lesage, D.; Karanik,
M.; Guillot, R.; Malacria, M.; Tabet, J.-C.; Goddard, J.-P.; Fensterbank,
L.; Gandon, V.; Gimbert, Y. Chem. Sci. 2011, 2, 2417−2422. (d) Cheong,
P. H.-Y.; Morganelli, P.; Luzung, M. R.; Houk, K. N.; Toste, F. D. J. Am.
Chem. Soc. 2008, 130, 4517−4526. (e) Hashmi, A. S. K. Acc. Chem. Res.
2014, 47, 864−876.
(2) For the most recent books and selected reviews, see: (a) Gold
Catalysis. An Homogeneous Approach; Toste, F. D., Michelet, V., Eds.;
Imperial College Press: Singapore, 2014; Catalytic Science Series, Vol.
13. (b) Gagosz, F. Golden Opportunities in the Synthesis of Natural
Products and Biologically Active Compounds. In Modern Tools for the
Synthesis of Complex Bioactive Molecules; Cossy, J., Arseniyadis, S., Eds.; J.
Wiley & Sons Inc., 2012; pp 111−154. (c) Pflasterer, D.; Hashmi, S. K.
̈
Chem. Soc. Rev. 2016, 45, 1331−1367. (d) Furstner, A. Acc. Chem. Res.
̈
2014, 47, 925−938. (e) Ranieri, B.; Escofet, I.; Echavarren, A. E. Org.
Biomol. Chem. 2015, 13, 7103−7118. (f) Wang, Y.-M.; Lackner, A. D.;
Toste, F. D. Acc. Chem. Res. 2014, 47, 889−901.
D
DOI: 10.1021/acs.orglett.6b01898
Org. Lett. XXXX, XXX, XXX−XXX