Synthesis of 3,5-Disubstituted 1,2-Dioxolanes through the Use of Acetoxy Peroxyacetals

Article abstract of DOI:10.1021/acs.orglett.9b01616

The synthesis of acetoxyendoperoxyacetal derivatives allowed the formation of functionalized 3,5-disubstituted 1,2-dioxolanes through the formation of reactive peroxycarbenium species under Lewis acid mediation. The introduction of a neutral nucleophile s

Full text of DOI:10.1021/acs.orglett.9b01616

Letter
pubs.acs.org/OrgLett
Cite This: Org. Lett. XXXX, XXX, XXX−XXX
Synthesis of 3,5-Disubstituted 1,2-Dioxolanes through the Use of
Acetoxy Peroxyacetals
†
,†
Alexis Pinet,† Thuy Linh Nguyen,†,‡ Guillaume Bernadat,† Bruno Figadere, and Laurent Ferrie*
̀
́
†
́
́
BioCIS, Universite Paris-Sud, CNRS, Universite Paris-Saclay, Chatenay-Malabry 92290, France
‡Institute of Marine Biochemistry, Vietnam Academy of Science and Technology (VAST), Cau Giay, Hanoi, Vietnam
S
* Supporting Information
ABSTRACT: The synthesis of acetoxyendoperoxyacetal de-
rivatives allowed the formation of functionalized 3,5-disub-
stituted 1,2-dioxolanes through the formation of reactive
peroxycarbenium species under Lewis acid mediation. The
introduction of a neutral nucleophile such as allylsilane, silane,
or silyl enol ether was accomplished with moderate to good
yields. The two studied Lewis acids, TiCl4 and SnCl4, gave
contrasting results. The higher diastereoselectivity toward the
trans diastereomer in experiments with TiCl4 as Lewis acid was
explained by a faster degradation of the cis isomer product,
leading generally to lower yields. A rationalization of this result
was supported by calculations.
rganic peroxides are uncommon moieties for natural
1
Oproducts. Nevertheless, this pharmacophore is often a
source of potent biological properties due to the relative
reactivity of peroxides. In addition to the famous antimalarial
agent, artemisinin, isolated from Artemesia annua plant,2 many
other organisms produce peroxide-containing natural products.
In particular, plakinic acids, isolated from sponges Plakortis sp.,
exhibit a fatty acid structure with a 1,2-dioxolane ring between
positions 3 and 5 (Figure 1).3 Another characteristic of
plakinic acids is that the 1,2-dioxolane ring is 3,3,5,5-
tetrasubstituted. More recently, we investigated the synthesis
of simpler analogues of mycangimycin,4 a new peroxide
isolated from Streptomyces sp. living in symbiosis with the
insect Dendroctonus frontalis.5 Mycangimycin has a very similar
structure compared to plakinic acids, but besides its highly
conjugated fatty chain, the 1,2-dioxolane ring is only 3,5-
disubstituted (Figure 1). This difference in the structure is
detrimental for the synthesis of this compound. Indeed, the
substitution pattern present in the dioxolane ring of plakinic
acids facilitates the preparation of such compounds due to the
enhanced stabilization of intermediate carbenium or radical
species necessary to provide the expected structure.6
Figure 1. Artemisinin, plakinic acids, and mycangimycin.
Precursors of 3,5-tetrasubstituted dioxolane rings can also be
double bonds though peroxymercuration,7 and here again, the
the peroxycarbenium ion as intermediate. Moreover, a
disfavored withdrawing effect by induction of one oxygen
atom to the other at the peroxide moiety reduces the ability to
stabilize the peroxycarbenium compared to a classical acetal.
Indeed, we planned to use in early experiments some reported
substitution of starting olefins for mycangimycin derivatives is
detrimental to the reactivity of the required olefin. For these
reasons, relatively few examples of the preparation of 3,5-
disubstituted 1,2-dioxolanes are found in the literature.6
Peroxycarbenium ions have emerged as a powerful tool for
the synthesis of organic peroxides.8 Nevertheless, the synthesis
of 3,5-disubstituted 1,2-dioxolanes from the corresponding
peroxyacetals is still an issue due to a reduced stabilization of
Received: May 7, 2019
© XXXX American Chemical Society
A
DOI: 10.1021/acs.orglett.9b01616
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
methodologies from Dussault et al.8b to perform the
nucleophilic addition to peroxyacetal such as methoxyperox-
yketal 6, which has been prepared from hemiperoxyketal 4 by
acid-mediated metathesis with MeOH.9 However, to our
disappointment, we were unable to prepare entity 7 from
hemiperoxyacetal 5 (prepared from cyclopropanol by oxidative
ring opening with triplet oxygen, see the Supporting
Information) and any simple alcohol. By analogy with
tetrahydrofurans, we decided to activate the cyclic peroxyacetal
as acetate due to its known excellent leaving group ability
(Scheme 1).
Table 1. Screening of Lewis acid, Stoichiometry, And
Temperature on the Sakurai Reaction between Acetoxy
Peroxyketal 8 and AllylTMS 9
a
b
entry Lewis acid
equiv
temp (°C) yield (%) dr (trans/cis)
c
1
2
3
4
5
6
7
8
9
TMSOTf
BF3·OEt2
SnCl4
SnCl4
SnCl4
SnCl4
TiCl4
TiCl4
TiCl4
0.1 or 1
1.0
0.9
0.9
0.9
1.0
0.9
0.9
1.1
−78 to rt
−78 to rt
−78
<10
19
13
nd
nd
c
Scheme 1. Activation of Endoperoxy Ketals and Acetals
d
50:50
65:35
78:22
e
75:25
85:15
91:9
−40
0
rt
−78
−40
−40
−40
80
31
e
d
7
47
30
60
10
TiCl4
0.7
70:30
a
b
Isolated yield. Diastereomeric ratios were determined by 1H NMR.
Not determined. 50% of recovered starting material. Decomposi-
c
d
e
tion.
endure a selective degradation of the cis isomer, therefore
enriching the proportion of trans isomer but decreasing overall
yield. Indeed, this hypothesis was confirmed by adjusting the
number of equivalents of TiCl4 to the reaction mixture.
Increasing the stoichiometry of TiCl4 to 1.1 equiv decreased
the yield but increased the diastereoselectivity to 91:9 (Table
1, entry 9), while decreasing the amount of TiCl4 to 0.7 equiv
improved the yield but decreased the diastereoselectivity of 10
to 70:30 (Table 1, entry 10).
An additional experiment allowed us to confirm our
hypothesis by applying Sakurai reaction conditions to
dioxolane 10 with a reduced amount of TiCl4 (0.4 equiv).
After 30 min, the reaction mixture was quenched and purified
to afford starting material in 56% recovered yield but with an
improved diastereoselectivity in favor of the trans compound
(from 70:30 to 84:16, Scheme 2).
In the first attempts, we experienced issues in obtaining
acetate 8 under classical conditions involving pyridine and
Ac2O, but only decomposition was observed, which can
explain why a similar structure was scarcely reported.10
Fortunately, the preparation of such a compound was possible
by using Lewis acid catalyzed acetylation conditions. With rare
earth triflates such as Yb(OTf)3, we obtained expected acetate
8 in high yield. It is worth noting that stability of this
compound is pretty high compared to THF acetate derivatives
due to a reduced stabilization of the peroxycarbenium
compared to THF oxycarbeniums, and thus, compound 8
can then be generally purified by flash chromatography on
silica gel without notable decomposition (Scheme 1).
A Sakurai reaction between peroxyacetal 8 and allyltrime-
thylsilane 9 was then investigated in the search for optimized
reaction conditions. Dichloromethane was found to be the
most suited solvent for this reaction. TMSOTf or BF3·OEt2
was an ineffective promoter for this transformation, converting
only small amounts of starting material (Table 1, entries 1 and
2). In contrast, SnCl4 and TiCl4 were far more promising.
Indeed, SnCl4 gave a high isolated yield of dioxolane 10 (Table
1, entry 4) at −40 °C. At lower temperature (entry 3), the
reaction hardly proceeded with recovery of starting material
after 1 h. The extended reaction time did not significantly
improve the yield and/or the diastereoselectivity. At higher
temperature, the yields also decreased due to some degradation
(Table 1, entries 5 and 6). The observed diastereoselectivity is
moderated, in contrast to those observed on 2,4-disubstituted
THF derivatives.11 This limited selectivity could be attributed
to a competition between two conformers of the peroxycarbe-
nium species. We also turned our attention to TiCl4; similar to
SnCl4, the reaction must be conducted at −40 °C (Table 1,
entry 8) rather than −78 °C (Table 1, entry 7). However, in
contrast to SnCl4, diastereoselectivity for 1,2-dioxolane 10 was
largely improved up to 85:15, albeit with a lower yield. To
explain this enhancement, we suspected reaction product 10 to
Scheme 2. Partial Degradation of Dioxolane 10 Using a
Reduced Amount of TiCl4 under the Reaction Conditions of
a Sakurai Reaction
The mechanism of this TiCl4-catalyzed degradation was
further investigated, and some major degradation products
were identified. A TiCl4-catalyzed Kornblum−DeLaMare
rearrangement12 of dioxolane 10 afforded compounds 12
and 13 on one hand (Scheme 3, top route). On the other
hand, TiCl4-mediated fragmentation of dioxolane 10 also
afforded aldehyde 14 and ketone 15 from intermediate 11a.
Under Sakurai reaction conditions, a reactive species such as
aldehyde 14 was converted into allyl adduct 18 (Scheme 3,
bottom route). Presumably, a similar process occurred from
intermediate 11b as well, but the expected degradation
B
DOI: 10.1021/acs.orglett.9b01616
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Scheme 3. Identification of TiCl4-Mediated Degradation
Products of Dioxolane 10
Figure 2. Energy profiles of the TiCl4-mediated Kornblum−
DeLaMare rearrangements of model compound cis- and trans-22 at
the B3LYP/(6-31+G(d,p), LanL2DZ) level of theory. Electronic
energy along the coordinates is represented in kcal/mol relative to
[trans-22 + TiCl4 + Cl−]. A chloride anion was used as a base to
scavenge the proton freed in the reaction.
of an additional chloride anion, which is likely to be found in
the reaction medium, allowed the rearrangement to occur. The
difference of activation energy ΔΔE⧧ between the two
diastereomers was found to be 2.2 kcal/mol, in favor of
isomer cis-22. This result is consistent with our observation of
a faster degradation of cis-10 compared to trans-10 with TiCl4
and subsequent enrichment of this latter. The energy for [cis-
22 + TiCl4 + Cl−] is higher by 3.7 kcal/mol compared to the
trans isomer, probably due to steric interactions between the
two alkyl groups at the backside of the 1,2-dioxolane ring,
squeezed by TiCl4 and chlorine placed on the front face. In
contrast, the transition state for cis-22 is only 0.7 kcal/mol
higher due to the free field for proton trapping by chlorine
anion at the front face of the dioxolane ring. In the meantime, a
steric congestion exists between chlorine and one alkyl group
in the transition state for trans-22, demanding more energy
(2.9 kcal/mol) to perform the reaction (Figure 2).
We then turned our attention to the nature of the
nucleophile being able to react on endoperoxyacetal 8. The
use of various allylsilanes proved to be successful (Table 2,
entries 1−6). The selective degradation of cis diastereoisomer
with TiCl4 gave always better trans selectivity. Silyl enol ethers
were also screened as potent nucleophiles (Table 2, entries 7−
10). The cis diastereoisomer was obtained as major product,
although TiCl4 mitigated this result due to a selective
degradation of cis diastereoisomer. Cyanide 29 was a very
effective nucleophile with SnCl4 as well as azide 30, giving the
expected products 37 and 38 with excellent yields but no
control of the diastereoselectivity. TiCl4 was not able to
improve this result, resulting only in decreased yield (Table 2,
entries 11−13). Reduction of the endoperoxyacetal is also
possible with triethylsilane 31, producing 3-monosubstituted
dioxolane 39 (entries 14 and 15).
a
Not isolated but observed by GC/MS.
products were not produced in sufficient amounts, and also
due to the relative volatility for some of them, side products
14, 16, 19, 20, and 21 were only observed from GC/MS
analysis of the crude material and were not isolated. The
mechanism of this fragmentation is not completely clear on our
system, since the possibility of a retro-aldol reaction from 1,3-
hydroxyketones 12 and 13 cannot be completely excluded.
However, a similar fragmentation was studied on 3,3,4-
trisubstituted 1,2-dioxolanes with TiCl4, affording ketones
and aldehydes with migration of an alkyl group, an hydrogen,
or a deuterium.13 An analogous fragmentation was also
observed in one of our previous works, where a bicyclic
peroxyketal afforded a ketoester under triflic acid activation.4
These two examples show that a retro-aldol process might not
occur since it cannot explain the formed product. Here, the
migration of the proton from position 3 or 5 to position 4 of
the 1,2-dioxolane ring is probably concomitant with the
peroxide bond and carbon−carbon bond cleavage (Scheme 3).
The origin of the enhancement of the diastereoselectivity by
degradation in the presence of TiCl4 in favor of a trans
compound was then investigated at the B3LYP/(6-31+G(d,p),
LanL2DZ)14 level of theory (Pople’s basis set for organoele-
ments and Los Alamos effective core potential for metallic
centers). We focused our study on one of the two possible
TiCl4-mediated Kornblum−DeLaMare rearrangements, lead-
ing to intermediate 23 from dioxolane 22, a simplified model
of dioxolane 10 (Figure 2). Transition states starting from
diastereoisomers cis-22 and trans-22 were localized on the
potential energy surface. First attempts to search a pathway for
the transition state from cis/trans-22 and TiCl4 only were
unsuccessful. Introduction of a proton acceptor under the form
Next, we explored substrates 40−46 (easily obtained as
compound 8, see the Supporting Information) in the Sakurai
reaction with allylsilane 9. In order to improve the
C
DOI: 10.1021/acs.orglett.9b01616
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Table 2. Addition of Various Nucleophiles to Peroxyacetal
8
Table 3. Use of Various Peroxyacetals 40−45 and
a
Peroxyketal 46 in the Sakurai Reaction with
a
Allyltrimethylsilane 9
a
Reaction conditions: 40−46 (1 equiv), TiCl4 or SnCl4 (0.9 equiv),
b
a
allyltrimethylsilane (3 equiv) in CH2Cl2 at −40 °C. Isolated yield.
Reaction conditions: 8 (1 equiv), TiCl4, or SnCl4 (0.9 equiv),
c
d
Diastereoselectivities were determined by 1H NMR. Yield
b
c
nucleophile 24−31 (3 equiv). Isolated yield. Diastereomeric ratios
d
e
were determined by H NMR. E/Z ratio = 1:4 (1H NMR). E/Z
1
calculated over two steps from hemiperoxyketal precursor of acetate
46.
ratio = 1:2 (1H NMR).
formation of acetoxy peroxyacetal by using a Lewis acid
catalysis proved to be essential for their formation and allowed
sufficient reactivity for further functionalization. SnCl4 and
TiCl4 were found to be the best Lewis acids screened so far. A
wide variety of substrates and nucleophiles were compatible
with the reaction conditions. A variety of results were observed
between these two reagents: SnCl4 generally gave good yields
but moderate selectivities, while TiCl4 furnished higher
selectivity toward the trans products but generally lower
yields. In this case, a TiCl4-mediated decomposition of the
newly formed products takes place, where the cis-disubstituted
dioxolane seems to decompose faster than the trans isomer,
explaining the improved selectivities and lower yields thus
observed. Calculation about toward one degradation pathway
could rationalize why cis-1,2-dioxolanes decompose more easily
under treatment with TiCl4.
diastereoselectivity of the reaction, bulky dioxolanyl acetates
40−42 were studied (Table 3, entries 1−6). Unfortunately,
these substrates did not improve the diastereoselectivity
compared to acetate 8 (Table 1), and a selective degradation
of cis product was still observed when TiCl4 was used in most
of the cases. Nevertheless, high steric congestion of the tert-
butyl group for compound 42 seemed to suppress degradation
of product 49, affording a high yield of it using both Lewis
acids. Benzyl-substituted and protected alcohols are compat-
ible in this reaction, leading to good yield with SnCl4 and
improved trans diastereoselectivity with TiCl4, as expected
(Table 3, entries 7−11). Ketal derivative 46 was also examined
with our protocol (Table 3, entries 13 and 14). Due to the
improved stabilization of the peroxonium species with the
methyl group, compound 46 proved to be unstable and could
not be isolated cleanly. The crude material was used directly in
the Sakurai reaction, meaning that the yield was calculated over
two steps. Product 53 was obtained with good selectivity with
both SnCl4 and TiCl4. This result is comparable in yield (over
two steps) with Dussault’s procedure, meaning our method-
ology is not only limited to endoperoxyacetals.7,15
ASSOCIATED CONTENT
* Supporting Information
■
S
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.or-
glett.9b01616.
In conclusion, we developed a new method for the
formation of 3,5-disubstituted 1,2-dioxolanes through the use
of acetoxy peroxyacetals and the formation of peroxycabenium
intermediate species under SnCl4 or TiCl4 activation. The
1
Experimental procedures, characterization data, and H
and 13C NMR spectra of new compounds (including
precursors such as cyclopropanols, 1,2-dioxolanols, and
D
DOI: 10.1021/acs.orglett.9b01616
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
dioxolanyl acetates); Cartesian coordinates of com-
Letter
(14) (a) Hohenberg, P.; Kohn, W. Phys. Rev. 1964, 136, B864−
B871. (b) Kohn, W.; Sham, L. J. Phys. Rev. 1965, 140, A1133−A1138.
(c) Becke, A. D. J. Chem. Phys. 1993, 98, 5648−5652. (d) Lee, C.;
Yang, W.; Parr, R. G. Phys. Rev. B: Condens. Matter Mater. Phys. 1988,
37, 785−789. (e) Hehre, W. J.; Radom, L.; Schleyer, P. V. R.; Pople, J.
A. Ab Initio Molecular Orbital Theory; Wiley: New York, 1986.
(f) Hay, P. J.; Wadt, W. R. J. J. Chem. Phys. 1985, 82, 270−283.
(g) Hay, P. J.; Wadt, W. R. J. Chem. Phys. 1985, 82, 299−310.
(15) Lu, X.; Liu, Y.; Sun, B.; Cindric, B.; Deng, L. J. Am. Chem. Soc.
2008, 130, 8134−8135.
pounds cis-22, trans-22, 23 and their transition states
(PDF)
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: laurent.ferrie@u-psud.fr.
ORCID
Guillaume Bernadat: 0000-0001-8955-0364
̀
Bruno Figadere: 0000-0003-4226-8489
́
Laurent Ferrie: 0000-0002-1171-205X
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
A.P. thanks the MESRI for a PhD fellowship, and T.L.N.
thanks the USTH for financial support. We thank Karine
■
̂
Leblanc and Jean-Christophe Jullian (BioCIS, Chatenay-
Malabry) for HRMS analysis and NMR services, respectively.
Leo Goehrs from Alionis is gratefully thanked for the donation
of computing hardware.
REFERENCES
■
(1) (a) Casteel, D. A. Nat. Prod. Rep. 1999, 16, 55−73. (b) Liu, D.-
Z.; Liu, J.-K. Nat. Prod. Bioprospect. 2013, 3, 161−206. (c) Vil, V. A.;
Gloriozova, T. A.; Terent’ev, A. O.; Savidov, N.; Dembitsky, V. M.
Appl. Microbiol. Biotechnol. 2019, 103, 1627−1642. (d) Vil, V. A.;
Terent’ev, A. O.; Savidov, N.; Gloriozova, T. A.; Poroikov, V. V.;
Pounina, T. A.; Dembitsky, V. M. J. Steroid Biochem. Mol. Biol. 2019,
190, 76−87.
(2) (a) Klayman, D. L. Science 1985, 228, 1049−1055. (b) Tu, Y.
Nat. Med. 2011, 17, 1217−1220. (c) White, N. J. Science 2008, 320,
330−334. (d) O’Neill, P. M.; Posner, G. H. J. Med. Chem. 2004, 47,
2945−2964. (e) Haynes, R. K. Curr. Top. Med. Chem. 2006, 6, 509−
537.
(3) Rahm, F.; Hayes, P. Y.; Kitching, W. Heterocycles 2004, 64, 623−
676.
́
̀
(4) Nguyen, T. L.; Ferrie, L.; Figadere, B. Tetrahedron Lett. 2016, 57,
5286−5289.
(5) (a) Scott, J. J.; Oh, D.-C.; Yuceer, M. C.; Klepzig, K. D.; Clardy,
J.; Currie, C. R. Science 2008, 322, 63. (b) Oh, D.-C.; Scott, J. J.;
Currie, C. R.; Clardy, J. Org. Lett. 2009, 11, 633−636.
(6) Terent’ev, A. O.; Borisov, D. A.; Vil’, V. A.; Dembitsky, V. M.
Beilstein J. Org. Chem. 2014, 10, 34−114.
(7) Bloodworth, A. J.; Bothwell, B. D.; Collins, A. N.; Maidwell, N.
L. Tetrahedron Lett. 1996, 37, 1885−1888.
(8) (a) Dussault, P. H.; Lee, H.-J.; Liu, X. J. Chem. Soc., Perkin Trans.
1 2000, 3006. (b) Dussault, P. H.; Lee, I. Q.; Lee, H.-J.; Lee, R. J.;
Niu, Q. J.; Schultz, J. A.; Zope, U. R. J. Org. Chem. 2000, 65, 8407.
(c) Ramirez, A.; Woerpel, K. A. Org. Lett. 2005, 7, 4617. (d) Dai, P.;
Trullinger, T. K.; Liu, X.; Dussault, P. H. J. Org. Chem. 2006, 71,
2283. (e) Dussault, P. H.; Liu, X. Org. Lett. 1999, 1, 1391−1393.
(9) Lifchits, O.; Mahlau, M.; Reisinger, C. M.; Lee, A.; Fares, C.;
Polyak, I.; Gopakumar, G.; Thiel, W.; List, B. J. Am. Chem. Soc. 2013,
135, 6677−6693.
(10) (a) Witkowski, B.; Gierczak, T. J. Mass Spectrom. 2013, 48, 79−
88. (b) Baumstark, A. L.; Vasquez, P. C.; Chen, Y.-X. J. Org. Chem.
1994, 59, 6692−6696.
(11) Schmitt, A.; Reißig, H.-U. Eur. J. Org. Chem. 2000, 2000, 3893−
3901.
(12) Yaremenko, I. A.; Vil’, V. A.; Demchuk, D. V.; Terent’ev, A. O.
Beilstein J. Org. Chem. 2016, 12, 1647−1748.
(13) Yoshida, M.; Miura, M.; Nojima, M.; Kusabayashi, S. J. Am.
Chem. Soc. 1983, 105, 6279−6285.
E
DOI: 10.1021/acs.orglett.9b01616
Org. Lett. XXXX, XXX, XXX−XXX