Soft-enolization Baker-Venkataraman Rearrangement Enabled Total Synthesis of Dirchromones and Related 2-Substituted Chromones

Article abstract of DOI:10.1021/acs.orglett.8b03148

A seven-step total synthesis of the original scaffold of cytotoxic dirchromones involving an unprecedented soft-enolization Baker-Venkataraman rearrangement was designed. The methodology enabled access to naturally occurring dirchromone 1 (21% overall yield) at gram-scale, which was screened for cytotoxicity against 13 cancer cell lines. The scope of the soft-enolization Baker-Venkataraman rearrangement encompasses diversely substituted dirchromones, including flavonoids, 2-styrylchromones, and 2-phenylethylchromones.

Full text of DOI:10.1021/acs.orglett.8b03148

Letter
pubs.acs.org/OrgLett
Cite This: Org. Lett. XXXX, XXX, XXX−XXX
Soft-enolization Baker−Venkataraman Rearrangement Enabled
Total Synthesis of Dirchromones and Related 2‑Substituted
Chromones
†
,†
Alexis St-Gelais,† Jerome Alsarraf, Jean Legault,† Charles Gauthier,†,‡ and Andre Pichette*
́
̂
́
†
́
́
́
́
Chaire de recherche sur les agents anticancereux d’origine naturelle, Laboratoire d’analyse et de separation des essences vegetales
́
́
́
̀
́
(LASEVE), Departement des Sciences Fondamentales, Universite du Quebec a Chicoutimi, 555, boulevard de l’Universite,
́
Chicoutimi, Quebec, G7H 2B1, Canada
‡
́
́
́
INRS-Institut Armand-Frappier, Universite du Quebec, 531, boulevard des Prairies, Laval, Quebec H7V 1B7, Canada
S
* Supporting Information
ABSTRACT: A seven-step total synthesis of the original scaffold of cytotoxic dirchromones involving an unprecedented soft-
enolization Baker−Venkataraman rearrangement was designed. The methodology enabled access to naturally occurring
dirchromone 1 (21% overall yield) at gram-scale, which was screened for cytotoxicity against 13 cancer cell lines. The scope of
the soft-enolization Baker−Venkataraman rearrangement encompasses diversely substituted dirchromones, including flavonoids,
2-styrylchromones, and 2-phenylethylchromones.
irchromones are a series of sulfur-containing compounds
Dthat were isolated from Dirca palustris L. bark and
wood.1 Some of these compounds exhibited a moderately
selective cytotoxicity against human colorectal adenocarcinoma
(DLD-1) cells, with dirchromone 1 (Figure 1) being the most
active (IC50 1.0 μM). However, only a limited number of
derivatives were characterized, and they were isolated in
minute quantities. For instance, 12.5 kg of dried wood and
bark of D. palustris was required to obtain 36 mg of
dirchromone 1, which was isolated after an extensive
purification process. A synthetic route toward dirchromone 1
and analogues is thus needed for any further studies to be
conducted on the mechanism of action and/or structure−
activity relationships (SARs) of this peculiar class of
compounds. Dirchromones are structurally similar to two
other groups of biologically active molecules. The group of 2-
styrylchromones 2 is uncommonly found in nature, but within
a large diversity of organisms, and has been the focus of several
synthetic efforts, with a variety of reported biological activities
such as cytotoxic, anti-inflammatory, and neuroprotective
effects.2 As for the second group, several of the large number
of 2-phenylethylchromones 3 found in agarwood, the fungi-
infected wood of several Thymelaeaceae,3 were reported to
exhibit interesting in vitro anti-inflammatory properties.4,5
Among the reported synthetic approaches toward these
biologically active 2-substituted chromones, including the
ubiquitous flavone scaffold 4, the Claisen condensation of
activated cinnamates onto 2′-hydroxyacetophenone6 is the
most common route. More specifically, its intramolecular
counterpart, i.e., the Baker−Venkataraman (B−V) rearrange-
ment of cinnamoyl ester of 2′-hydroxyacetophenone,2,7 is
Figure 1. Parent structures of 2-substituted chromones: dirchromone
1, 2-styrylchromone 2, 2-phenylethylchromone 3, and flavone 4.
Received: October 2, 2018
© XXXX American Chemical Society
A
DOI: 10.1021/acs.orglett.8b03148
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Scheme 1. Failed Synthetic Routes Explored for the Preparation of (S)-Deoxydirchromone 15a
widely used. This reaction requires the formation of an
enolate8 that is typically produced by deprotonation of
acetophenone using a strong base.9 The scope of the reaction
is therefore restricted to substrates that tolerate strongly basic
conditions, which are referred to as “hard” enolization
conditions.10 However, “soft” enolization conditions, where a
Lewis acid is used to enable the formation of enolates in the
presence of a weak base,10,11 have never been applied to the
B−V rearrangement.12 Along these lines, we demonstrate that
hard enolization conditions proved inapplicable to the
synthesis of dirchromones, which prompted the development
of the first soft-enolization B−V rearrangement. This strategy
specifically enabled a facile and scalable synthesis of diversely
substituted dirchromone derivatives from readily available and
affordable buildings blocks. Illustrating the utility of the
proposed synthesis, it gave access to gram-scale preparation of
dirchromone 1. Its cytotoxicity was then screened against 13
cancer cell lines.
The preparation of (S)-deoxydirchromone 15a was the main
focus of the initial investigation, provided that it could then be
readily oxidized to racemic sulfoxide 1 using 3-chloroperben-
zoic acid (m-CPBA).13 The first explored strategy was modeled
based on classical syntheses of 2-substituted chromones
through a Claisen condensation6,14 of methyl 3-methylthio-
propionate 5 and 2′-hydroxyacetophenone 9a (Scheme 1,
route A). This would have been followed by acidic dehydrative
cyclization to compound 14 and oxidation to generate
vinylsulfide 15a. The Claisen condensation, under classical
hard conditions involving strong bases such as sodium
hydride6,9,15 or sodium alcoolate,16 however, failed to afford
the desired intermediate. A methodological change was thus
required.
We then envisioned an intramolecular version of this
transformation involving a B−V rearrangement of ester 8
followed by acidic cyclization (Scheme 1, route B). However,
ester 8 was found to readily degrade, preventing its isolation in
sufficient amount and purity to pursue this strategy.
Construction of a different sulfur-bearing chain via the easily
prepared ester 10 (Scheme 1, route C) was found to be an
acceptable workaround. The conversion of ester 10 to diketone
intermediate 11 under classical B−V conditions, however, gave
phenone 9a and starting ester as the main products.
The hard conditions involved in either the Claisen
condensation or classical B−V rearrangement were suspected
to interfere with the reaction. We thus envisioned applying the
soft-enolization Claisen reaction developed by Lim et al.10 to
the case of dirchromone synthesis. After screening various
Lewis acids, including ZnCl2, CuOTf2, and NiI2 in different
solvents such as THF or toluene, the authors showed that
room temperature Claisen condensations could be efficiently
conducted using acetophenone and a variety of activated
acylating agents. The optimized conditions were demonstrated
to involve N-acylbenzotriazoles or pentafluorophenyl esters as
acylating reagents in the presence of magnesium bromide
diethyl etherate in reagent-grade CH2Cl2 open to air and using
diisopropylethylamine (DIPEA) as a base.17 Yet, when 2′-
hydroxyacetophenone 9a was used as the substrate, the
reported yields were significantly reduced. Furthermore,
preparing activated acyl groups from sulfide bearing carboxylic
acid 7 to perform a Claisen condensation would involve
significant work, and our preliminary testing showed that the
B
DOI: 10.1021/acs.orglett.8b03148
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Scheme 2. Soft Baker−Venkataraman Enabled Synthesis of (S)-Deoxydirchromone and Preparation of Dirchromone 1
yields were very low (data not shown). For these reasons, the
soft-enolization intermolecular Claisen condensation seemed
not to be a relevant strategy for the preparation of
dirchromone precursors. However, an intramolecular soft-
enolization B−V rearrangement appeared as an attractive
alternative. Indeed, direct esterification of 2′-hydroxyaceto-
phenone 9a would yield a proper substrate for a B−V
rearrangement, leading to chromone precursors in an atom-
economical fashion. Applying Lim’s conditions to ester 10
(Scheme 1, route D), which could be described as the first
reported soft-enolization B−V reaction (soft B−V), followed
by acidic cyclization and cleavage of the crude thioester 12
with K2CO3, provided the thiol 13 in satisfactory yield. Yet,
subsequent methylation of thiol 13 to afford compound 14 did
not proceed under classical conditions.18,19
In search of a more efficient alternative, sulfide 7 was first
quantitatively oxidized to crude sulfoxide 16a20 (Scheme 2).
The latter was then reacted with oxalyl chloride in anhydrous
CH2Cl2, which had two purposes. Not only was the acyl
chloride generated but a Pummerer rearrangement followed by
an elimination led to an oxidation transfer from the sulfur atom
to the lateral chain, providing a vinyl sulfide with complete E-
selectivity. The formation of a vinylsulfide in such a way is, to
the best of our knowledge, quite uncommon,21,22 sometimes
observed as a side product in the course of other Pummerer-
type reactions.23 As such, our procedure proved to be a
straightforward route to useful vinylsulfide building blocks.24
The overall reaction to yield derivative 17a was completed
faster when a catalytic amount of DMF was added. Once gas
evolution had ceased, coupling the formed reactant with an
excess of the conjugated base of 2′-hydroxyacetophenone 9a
yielded stable vinylsulfide ester 18a in a convenient 40% yield,
given the relatively complex one-pot sequence. Ester 18a was
then submitted to the soft B−V conditions (2.5 equiv of
MgBr2 etherate, 3 equiv of DIPEA, CH2Cl2 open to air,
overnight at room temperature) to afford enol 19 (71%).
Mechanistically, the reaction is likely to proceed through the
coordination of Mg2+ to the ketone function of ester 18a that
would decrease the pKa of the α-proton and promote
deprotonation by the weak base DIPEA. The generated
enolate would then undergo the B−V rearrangement to afford
enol 19. The latter was then successfully cyclized into (S)-
deoxydirchromone 15a (73%). These two steps were found to
be applicable consecutively, omitting isolation of the enol, with
an identical overall yield (52% over two steps).
dirchromones to explore its scope (Scheme 3). The acyl
chloride formation/Pummerer rearrangement/esterification
step gave moderate overall yields. Pleasingly, the reaction
still proceeded well with ethylsulfoxide 18b even though the
elimination step could have been expected to occur on both
sides of the sulfide following the Pummerer rearrangement,
which could explain the decreased yield. However, the
sequence failed for isopropylsulfoxide 18c, indicating that
only unbranched alkylsulfides can be transformed by this
method. When a benzoate protecting group was present (18g),
generation of the phenolate had to be conducted with NaH
instead of KOH, resulting in decreased esterification efficiency.
For other dirchromones, at best, use of potassium phenolate of
ketone 9a in excess as both the substrate and the base was
found to slightly increase the yield by 5−10%, compared to an
alternative approach using 1.5 equiv of ketone 9a in the
presence of triethylamine. It thus seems that a stronger
conjugated base leads to more convenient yields. The excess
phenone could be almost quantitatively retrieved by
chromatography, so this procedure was preferred for
preparation of compounds 18a−b, 18d−f, and 18h. The
Pummerer rearrangement itself rather than the acyl chloride
generation or esterification steps is presumed to be the most
yield-limiting part of the sequence since transposition of the
same protocol starting with E-cinnamic acid 20 toward the
synthesis of chromone 3 gave a good 76% yield of ester 21.
When commercial acyl chlorides 21 and 22 were used, a milder
esterification with ketone 9a in the presence of triethylamine
was satisfactory, providing the expected ester with yields
ranging from 58 to 88%.
For the second step, it was found that the soft B−V
proceeded similarly well when generating a substituted enol,
starting from 2′-hydroxypropiophenone toward 15e. Substi-
tuted vinylsulfide 18d afforded the corresponding methylated
deoxydirchromone 15d in a very good 81% yield, showing that
steric hindrance did not interfere with the soft B−V
rearrangement. Also of interest was the possibility to conduct
the reaction while retaining a benzoyl group 15g, indicating the
compatibility of the soft conditions with base-sensitive
functionalities. The reaction was also readily scalable. The
gram-scale soft B−V, which enabled conversion of vinylsufide
18a into chromone 15a, proceeded even better than at the
mmol scale (52%), with a yield reaching 67%, without
changing the protocol. The conditions used for the soft B−V
rearrangement represent, to the best of our knowledge, a novel
approach to this classical reaction, enabling the formation of a
wider scope of compounds featuring sensitive functional
Since this overall strategy secured straightforward access to
the dirchromone scaffold, it was applied to several substituted
C
DOI: 10.1021/acs.orglett.8b03148
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
usefully applied in the future to a broad variety of syntheses,
especially if base-sensitive groups are present.
Scheme 3. Scope of the Soft B−V Rearrangement
Oxidation of sulfide 15a to sulfoxide 1 was completed in
good yields (69−87%) using m-CPBA, overall offering a
straightforward and efficient synthesis of dirchromone 1 whose
NMR data were in full agreement with the reported structure.1
With no significant modification to experimental protocols, the
seven-step sequence, i.e., oxidation, acyl chloride generation,
Pummerer rearrangement, esterification, soft B−V rearrange-
ment, condensation, and oxidation, was consecutively and
successfully applied to the synthesis of 1.5 g of dirchromone 1,
with an overall yield of 21%. With such quantity at hand,
cytotoxic activity screening of dirchromone 1 was broadened
to 13 cancer cell lines. As presented in Table 1, dirchromone 1
Table 1. Cytotoxic Activity of Synthesized Dirchromone 1
against a Panel of Cancer Cell Lines
a
IC50 (μM)
dirchromone
cell line
DLD-1
Caco-2
HT-29
PC-3
type
(1)
etoposide
colorectal adenocarcinoma
colorectal adenocarcinoma
colorectal adenocarcinoma
prostate adenocarcinoma
mammary adenocarcinoma
ovarian teratocarcinoma
1.5 0.2
6.5 0.5
13.7 0.8
1.7 0.1
3.0 0.3
3.1 0.8
3.6 0.3
3.7 0.5
3.8 0.2
4.1 0.7
4.3 0.1
11.9 0.7
14.1 0.6
5.1 1.0
>50
10
2
4.4 0.7
>50
<0.4
MCF-7
PA-1
PANC 05.04 pancreatic adenocarcinoma
>50
SK-Mel-2
SAOS2
HEP-G2
U-87
A-549
U251
skin melanoma
osteosarcoma
hepatocarcinoma
brain glioblastoma
lung carcinoma
glioma
3.5 0.4
3.5 0.4
0.6 0.1
>50
1.7 0.3
3.4 0.7
a
IC50 values
SD (n = 3) are representative of two different
experiments.
was found to be cytotoxic against all cancer cell lines tested,
with IC50 ranging from 1.5 to 14.1 μM. The range of activity
(μM) is similar to etoposide used as positive control. DLD-1
and PC-3 are the most sensitive cancer cell lines to
dirchromones 1 with IC50 of 1.5 and 1.7 μM, respectively.
The results confirm the previously reported activity for DLD-1
and A-549.1
In summary, we have reported the first soft B−V, which can
be applied to base-sensitive substrates. Combined to a
vinylsulfide-generating Pummerer reaction, this in turn enabled
the total synthesis of biologically active dirchromone 1, in an
amount that would have required over four tons of dried D.
palustris wood to be isolated. Synthetic preparation of
dirchromone 1 proceeded from two cheaply available
substrates (7 and 9a). It is expected that this work will pave
the way to a detailed study of mechanism of action of
dirchromone 1 and SAR. The vinylsulfide-generating Pum-
merer rearrangement described herein will also be further
studied to broaden its field of application. Investigations on the
mechanism of action are ongoing and will be reported in due
course.
groups that are not compatible with classical hard conditions.
Upon investigation, the reaction offered similarly useful
conditions for the preparation of 2-styrylchromone 2, 2-
phenylethylchromone 3, and flavone 4 in 79, 61, and 81%
yields, respectively. In comparison, 2-styrylchromone 2 was
obtained in 43% yield (from phenone 9) when the B−V
rearrangement was conducted with K2CO3 in refluxing
acetone;25 or in 73% yield (from cinnamoyl chloride)
following a KOH-mediated B−V rearrangement procedure.26
These backbones of a large number of biologically relevant
compounds are, as far as we are aware, always formed under
strongly basic conditions.12 The softer base/magnesium
system was here demonstrated to be applicable to the same
reactions in high yields, foretelling that soft B−V could be
D
DOI: 10.1021/acs.orglett.8b03148
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
(16) Gray, C. A.; Kaye, P. T.; Nchinda, A. T. J. Nat. Prod. 2003, 66
(8), 1144−1146.
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.or-
glett.8b03148.
■
S
(17) Lim, D.; Zhou, G.; Livanos, A. E.; Fang, F.; Coltart, D. M.
Synthesis 2008, 2008 (13), 2148−2152.
(18) Rioz-Martínez, A.; Kopacz, M.; de Gonzalo, G.; Torres
̃
Pazmino, D. E.; Gotor, V.; Fraaije, M. W. Org. Biomol. Chem. 2011, 9
(5), 1337.
1
Experimental procedures, spectroscopic data, and H
(19) Roe, R. M.; Kallapur, V.; Linderman, R. J.; Viviani, F. Pestic.
Biochem. Physiol. 2005, 83 (2−3), 140−154.
and 13C NMR spectra of new compounds (PDF)
(20) Sato, K.; Hyodo, M.; Aoki, M.; Zheng, X. Q.; Noyori, R.
Tetrahedron 2001, 57 (13), 2469−2476.
AUTHOR INFORMATION
Corresponding Author
■
(21) Bur, S. K.; Padwa, A. Chem. Rev. 2004, 104 (5), 2401−2432.
(22) Smith, L. H. S.; Coote, S. C.; Sneddon, H. E.; Procter, D. J.
Angew. Chem., Int. Ed. 2010, 49 (34), 5832−5844.
*E-mail: andre_pichette@uqac.ca. Tel: +1-418-545-5011, ext.
5081. Fax: +1-418-545-5012.
ORCID
(23) Horiguchi, Y.; Saitoh, T.; Terakado, S.; Honda, K.; Kimura, T.;
Toda, J.; Sano, T. Heterocycles 2000, 54 (2), 967−984.
(24) Bates, C. G.; Saejueng, P.; Doherty, M. Q.; Venkataraman, D.
Org. Lett. 2004, 6 (26), 5005−5008.
Alexis St-Gelais: 0000-0003-1091-8301
́
̂
Jerome Alsarraf: 0000-0003-4641-2599
Charles Gauthier: 0000-0002-2475-2050
Notes
(25) Desideri, N.; Conti, C.; Mastromarino, P.; Mastropaolo, F.
Antivir. Chem. Chemother. 2000, 11, 373−381.
(26) Charvin, D.; Pomel, V.; Ortiz, M.; Frauli, M.; Scheffler, S.;
Steinberg, E.; Baron, L.; Deshons, L.; Rudigier, R.; Thiarc, D.; Morice,
C.; Manteau, B.; Mayer, S.; Graham, D.; Giethlen, B.; Brugger, N.;
The authors declare no competing financial interest.
́
Hedou, G.; Conquet, F.; Schann, S. J. Med. Chem. 2017, 60 (20),
8515−8537.
ACKNOWLEDGMENTS
■
The authors acknowledge the Chaire de recherche sur les
́
agents anticancereux d’origine naturelle for funding. This work
was supported by the Canadian Institutes of Health Research
(CIHR, operating grants 311906 and 326083 to A.P. and J.L.)
and by the Natural Sciences and Engineering Research Council
of Canada (NSERC, RGPIN-2016-04950 to C.G.). A.S.G.
thanks NSERC for a Ph.D. scholarship. C.G. holds a Fonds de
́
́
recherche du Quebec−Sante (FRQS) Research Scholars Junior
2 Career Award. C. Dussault (UQAC) is gratefully acknowl-
edged for cytotoxicity assays.
REFERENCES
■
(1) St-Gelais, A.; Legault, J.; Mshvildadze, V.; Pichette, A. J. J. Nat.
Prod. 2015, 78 (8), 1904−1909.
(2) Santos, C. M. M.; Silva, A. M. S. A. Eur. J. Org. Chem. 2017, 2017
(22), 3115−3133.
(3) Kristanti, A. N.; Tanjung, M.; Aminah, N. S. Mini-Rev. Org.
Chem. 2018, 15 (1), 36−55.
(4) Huo, H. X.; Gu, Y. F.; Sun, H.; Zhang, Y. F.; Liu, W. J.; Zhu, Z.
X.; Shi, S. P.; Song, Y. L.; Jin, H. W.; Zhao, Y. F.; Tu, P.-F.; Li, J.
Fitoterapia 2017, 118 (2016), 49−55.
(5) Chen, D.; Xu, Z.; Chai, X.; Zeng, K.; Jia, Y.; Bi, D.; Ma, Z.; Tu,
P. Eur. J. Org. Chem. 2012, 2012 (27), 5389−5397.
(6) Williams, D. A.; Smith, C.; Zhang, Y. Tetrahedron Lett. 2013, 54
(32), 4292−4295.
(7) Jain, P. K.; Makrandi, J. K.; Grover, S. K. Synthesis 1982, 3, 221−
222.
(8) Evans, D. A.; Nelson, J. V.; Taber, T. R. Top. Stereochem. 2007,
13, 1−115.
(9) Zhao, D.; Beiring, B.; Glorius, F. Angew. Chem., Int. Ed. 2013, 52
(32), 8454−8458.
(10) Lim, D.; Fang; Zhou, G.; Coltart, D. M. Org. Lett. 2007, 9 (21),
4139−4142.
(11) Zhou, G.; Yost, J. M.; Coltart, D. M. Synthesis 2007, 3, 478−
482.
(12) Ameen, D.; Snape, T. J. Synthesis 2015, 47 (2), 141−158.
(13) Xing, H.; Chen, L.; Jia, Y.; Jiang, Z.; Yang, Z. Tetrahedron Lett.
2017, 58 (23), 2199−2202.
(14) Wang, S. J.; Bao, L.; Han, J. J.; Wang, Q. X.; Yang, X. L.; Wen,
H. A.; Guo, L. D.; Li, S. J.; Zhao, F.; Liu, H. W. J. Nat. Prod. 2013, 76,
45−50.
(15) Yu, D.; Chen, C. H.; Brossi, A.; Lee, K. H. J. Med. Chem. 2004,
47 (16), 4072−4082.
E
DOI: 10.1021/acs.orglett.8b03148
Org. Lett. XXXX, XXX, XXX−XXX