Oxygen-to-Oxygen Silyl Migration of α-Siloxy Sulfoxides and Oxidation-Triggered Allicin Formation

Article abstract of DOI:10.1021/acs.orglett.1c01149

Oxidation of α-siloxy thioethers leads to the formation of the corresponding sulfoxides as unstable intermediates, which undergo an intramolecular oxygen-to-oxygen silyl migration to break the C-S linkage. This process produces silyl protected sulfenic acids and subsequently thiosulfinates. It was used to develop oxidation-triggered allicin donors.

Full text of DOI:10.1021/acs.orglett.1c01149

pubs.acs.org/OrgLett
Letter
Oxygen-to-Oxygen Silyl Migration of α‑Siloxy Sulfoxides and
Oxidation-Triggered Allicin Formation
Shane S. Kelly, Tun-Li Shen, and Ming Xian*
Cite This: Org. Lett. 2021, 23, 3741−3745
Read Online
sı
*
Metrics & More
Article Recommendations
Supporting Information
ACCESS
ABSTRACT: Oxidation of α-siloxy thioethers leads to the formation of the corresponding sulfoxides as unstable intermediates,
which undergo an intramolecular oxygen-to-oxygen silyl migration to break the C−S linkage. This process produces silyl protected
sulfenic acids and subsequently thiosulfinates. It was used to develop oxidation-triggered allicin donors.
Scheme 1. Design of RSS Donors
eactive sulfur species (RSS) are sulfur-containing
molecules that exhibit regulatory functions in biological
R
systems. These include small molecule species, such as
hydrogen sulfide (H2S), hydrogen polysulfide (H2Sn), and
cysteine- or glutathione-derived S-nitrosothiols and persulfides
(RSNO and RSSH), as well as protein-bound sulfur adducts
such as disulfides (PSSR), S-nitrosothiols (PSNO), sulfenic
acids (PSOH), and persulfides (PSSH). Many of these species
are highly reactive and short-lived, which makes their studies
quite challenging. To precisely generate specific RSS and
understand their associated biology, much effort has been
expended exploring RSS releasing scaffolds (i.e., donors) and
several donor molecules for H2S, RSSH, and H2Sn have been
reported.1
Our laboratory recently developed a general scaffold for the
design of RSS donors (Scheme 1).2 These donors share a
common α-hydroxyl protected sulfur structure. Upon depro-
tection, the resulting α-hydroxyl sulfur derivative degrades to
produce the desired sulfur-containing molecule. This O-to-S
relay deprotection strategy has been used to develop donors
for H2S, RSSH, H2S2, and HSNO (compounds 1−4).2 To
further extend this strategy to other RSS we decided to explore
sulfenic acid (RSOH) donors. RSOH are known to be critical
in maintaining cellular redox homeostasis. However, the full
extent of their biological significance remains unclear. This is
due in part to the inherent challenges with studying RSOH
whose average half-life is less than 1 min.3 Their short life span
stems from their high reactivity with both nucleophiles and
electrophiles. The most successful attempts to stabilize them
have employed a high degree of steric hindrance achieved
through embedment within a macrobicyclic cyclophane or
bowl-type cage, effectively preventing condensation.4,5 In situ
RSOH formation methods have also been explored. These
include base hydrolysis of a sulfenic acid ester or disulfide, acid
hydrolysis of oxidized mixed thioacetal, or thermolysis of
sulfoxides.6−9 These methods typically require harsh con-
ditions that are not biologically compatible. We envisioned that
compounds like 5 could serve as deprotection-triggered RSOH
donors. Herein we would like to report our attempts to prepare
Received: April 2, 2021
Published: April 19, 2021
https://doi.org/10.1021/acs.orglett.1c01149
© 2021 American Chemical Society
Org. Lett. 2021, 23, 3741−3745
3741

Organic Letters
pubs.acs.org/OrgLett
Letter
compounds 5 and the discovery of a unique O,O-silyl
migration that could be used to design oxidation-triggered
donors for thiosulfinates such as allicin.
We expected the desired RSOH donors (5) could be
obtained through oxidation of the corresponding α-siloxy
thioethers 6 (Scheme 2). Two methods were employed to
However, sulfoxide’s strong nucleophilicity and oxygen’s high
affinity to silyl are well documented. For example, during
dehydrosulfenylation of esters/ketones, the sulfoxide β to the
carbonyl undergoes an intramolecular proton transfer and β-
elimination to yield the unsaturated compound and RSOH
(Scheme 3a).9 Transfer of an alkylsilane to a sulfoxide oxygen
has also been reported in β-elimination reactions (Scheme
3b).11,12 Additionally, sulfoxides can attack a nearby silyl group
under heat via sila-Pummerer rearrangement to form
alkoxysilanes (Scheme 3c).13 These silyl transfers are
reminiscent of that observed presently in which a silyl moiety
undergoes O-to-O transfer, concurrent with cyclo-elimination,
producing a silyl protected sulfenic acid.
Scheme 2. Intended Synthesis of RSOH Donors
As shown in Scheme 3, the proposed silyl migration should
produce the parent aldehyde and silyl protected sulfenic acid.
The formation of the aldehyde product was demonstrated via
Brady’s test (Scheme 4). When 6c was subjected to oxidation
Scheme 4. Brady’s Test
prepare 6: (1) simple S-alkylation of our known α-siloxy thiol
1; (2) condensation of aldehyde and thiol in the presence of
silyl chloride and pyridine. With these methods, we quickly
assembled a small library of thioethers 6a−d. These
compounds were found to be stable in both organic and
aqueous (pH 5−9) solutions over 24 hours.
Next, we attempted to oxidize the thioether substrates.
Several established oxidation conditions were tried, which
included (1) Oxone in EtOH, (2) DMDO in acetone, (3)
mCPBA with Et3N or pyridine in DCM, and (4) mCPBA, aq.
phosphate buffer (pH 7.5), DCM. Unfortunately, this
oxidation was found to be more difficult than expected as we
could not obtain the desired sulfoxides. While it was likely the
oxidation on sulfur atom of these substrates should proceed as
expected, the extreme difficulty in obtaining the sulfoxide
products led us to speculate their inherent instability. A
previous report by Clennan et al. also found that α-siloxy
sulfoxides were very unstable.10 They can only exist at very low
temperature and rapidly decompose above −20 °C. It is known
that the oxygen atom of a sulfoxide possesses nucleophilicity.
Based on the unique structure of sulfoxide 5 we proposed an
O-to-O silyl migration as shown in Scheme 3. This process
should go through a favorable five-member ring transition state
to provide an aldehyde and a silyl-protected RSOH. To the
best of our knowledge, this reaction is unknown in literature.
by mCPBA under an argon atmosphere, Brady’s Reagent
(DNPH or 2,4-dinitrophenylhydrazine) was used to detect the
presence of aldehyde. As expected, the desired hydrazone
product was obtained in high yield. This demonstrated the
unmasking of the parent aldehyde.
Next, we attempted to identify other decomposition
products and understand the reaction mechanism. When 6c
was subjected to oxidation and then decomposition under
argon, subsequent MS analysis identified triethylsilanol 8,
benzyl disulfide 9, and sulfinic acid 10 as final products
(Scheme 5). The formation of 9 and 10 implicates dibenzyl
Scheme 5. Proposed Reaction Pathway of 6c under
Oxidation
Scheme 3. Oxygen-to-Oxygen Silyl Migration
thiosulfinate 13 as the intermediate. In fact, we were able to
isolate thiosulfinate 13 in good yield (80%) if the reaction was
purified immediately after the oxidation. With these products
identified, we proposed the mechanism as the following: The
oxidation of 6c and silyl migration should lead to compound
11. Silyl-protected sulfenic acids are known to be highly
susceptible to hydrolysis and solvolysis by protic reagents.14,15
Therefore, sulfenic acid 12 and silanol 8 should be readily
formed from 11. Sulfenic acids (11) are extremely unstable
and rapidly condense to form thiosulfinates (13). There are
3742
https://doi.org/10.1021/acs.orglett.1c01149
Org. Lett. 2021, 23, 3741−3745

Organic Letters
pubs.acs.org/OrgLett
Letter
many possible pathways for 13 to produce final products 9 and
10. For example, thiosulfinate hydrolysis has been reported to
ultimately produce a 1:1 molar ratio of disulfide and sulfinic
acid (RSO2H) at all pH values.16 Thiosulfinates are also
inherently unstable and can spontaneously disproportionate to
yield the corresponding disulfide and thiosulfonate,17 which
eventually react with thiols to form disulfides. We also carried
out a crossover experiment with 6c and 17b to bolster
evidence of the intermediate sulfenic acid 12 (see Supporting
Information). In addition to the two self-conjugated
thiosulfinates, we observed the formation of mixed thiosulfi-
nates. This further supports the proposed mechanism.
The reaction pathway shown in Scheme 5 explains the
instability of α-siloxy sulfoxides and the formation of silanol,
disulfide, and sulfinic acid as final products. To further
demonstrate the mechanism, we prepared three analogs with
similar structure and properties to α-siloxy sulfoxides, i.e. 14−
16 (Figure 1). These compounds were exposed to the same
conditions which degraded 6c to evaluate their stability and
thus gain mechanistic insight.
Stability studies on 14−16 further confirmed our under-
standing of the silyl migration-triggered degradation of α-siloxy
sulfoxides. This migration ultimately unmasks the parent
aldehyde, indicating the α-siloxy sulfide may be used in an
oxidant-triggered aldehyde deprotection strategy. Since this
process led to the formation of RSOH and subsequently the
thiosulfinate, we believed this reaction could be used as a
unique oxidation-triggered strategy for thiosulfinate produc-
tion. Although in general thiosulfinates are unstable, some
plant-derived thiosulfinates are biologically relevant. Allicin is
the main active component in garlic and possesses interesting
benefits for oxidative stress related diseases such as cancer and
drug withdrawal syndrome. Allicin (and thiosulfinates in
general) is typically prepared by treating allyl disulfide with a
strong peroxy acid, which is not a biocompatible process.
However, delivery to cells exhibiting oxidative stress could
prove beneficial as a therapeutic by increasing antioxidant
response. As such, oxidation-triggered allicin donors would be
interesting.
Based on the new silyl migration mechanism, it was expected
that allyl thioethers 17a−c (Scheme 6) would be the proposed
Scheme 6. Oxidation-Triggered Allicin Formation
Figure 1. Structures of analogs.
One carbon homologue 14 was used to investigate the
stability of the secondary siloxy group. With the siloxy group
moved to the β-position of sulfoxide, this compound was
found to be stable and no degradation was noted. This
indicated the instability of α-siloxy sulfoxides was not simply
derived from a trigger labile to the mCPBA oxidation.
Although the sulfoxide of compound 14 may form a six-
member ring with the siloxy group for the silyl migration, the
process should not be productive, as it will not lead to break
down of the molecule to drive the reaction. Compound 15 was
prepared to test the inherent stability of α-oxyl sulfoxides. This
α-acetoxy sulfoxide was found stable under the reaction
conditions. Apparently in this case sulfoxide could not promote
intramolecular acyl migration to remove the protection group
on the oxygen atom and cause degradation. As such the
stability of α-oxyl sulfoxides is not a problem. 14 and 15 did
not examine the stability of the O-silyl protecting group with
an α electron-withdrawing/-leaving group under the oxidation
conditions. It is plausible that the instability of the α-siloxy
sulfoxide was due to a combination of electronic and entropic
effects. These contributions were probed via analog 16 which
features cyanide as an electron-withdrawing group and
potential leaving group with similar electronic characteristics
to the sulfoxide. The α-proton’s pKa of nitrile and sulfoxide are
approximately 30 and 35 in DMSO, respectively. The nitrile is
slightly more electron-withdrawing but not significantly so.
When considered as leaving groups via their pKa they are quite
similar. The pKa of HCN is 9 while sulfenic acids possess
values of 5−10. The difference between them is their ability to
attack the siloxy group. Upon exposure to the oxidative and
acidic conditions used previously, the cyano analog 16 proved
stable. This indicates the instability is not simply due to the
electronic and entropic driving forces provided by an α leaving
group.
allicin donors. These are stable compounds that can be
administered as needed. They serve as allicin releasing
prodrugs only when appropriate oxidants are present, there-
fore, achieving release upon demand. 17a−c were then
prepared and tested under mCPBA oxidation. As expected,
the formation of allicin (29−71%) was observed. Interestingly,
the more stable protecting group led to higher conversion.
To further understand the scope of this reaction, other
oxidation conditions were examined (Figure 2). H2O2 is a
biologically relevant oxidant. It is known to be a central redox
a
Figure 2. Effects of oxidation conditions on allicin formation. Yields
determined by HPLC with external reference.
3743
https://doi.org/10.1021/acs.orglett.1c01149
Org. Lett. 2021, 23, 3741−3745

Organic Letters
pubs.acs.org/OrgLett
Letter
signaling molecule and a metabolite consequent of an oxidative
stress condition. H2O2 was found to produce allicin from 17a/
b very effectively. >90% Allicin was formed in 30 min, and the
yields dropped afterward due to the degradation of allicin.
Compounds like 17 may thus claim a role as biologically
relevant oxidant-triggered allicin donors. NaOCl is another
biologically relevant oxidant, although it did not produce
allicin. Two other oxidants, magnesium monoperoxyphthalate
(MMPP) and NaIO4, were also utilized to explore the reaction
scope. Of these, only MMPP produced allicin.
In summary, in our attempts to develop O to S relay
deprotection-based RSOH donors we discovered a unique O-
to-O silyl migration on sulfoxide substrates. This process led to
the formation of thiosulfinates, likely via the RSOH
intermediate (as shown in Scheme 5). Therefore, the sulfoxide
substrates may still be considered as RSOH precursors.
However, since we were unable to trap the RSOH intermediate
with dimedone and methyl iodide, more studies may be
needed to further confirm the mechanism. Nevertheless, this
reaction was applied to develop an oxidation-triggered allicin
formation strategy. Given the known medicinal benefits of
allicin, these oxidation-triggered allicin donors may be useful in
certain situations. It should be noted that besides allicin this
process also produces several side products such as aldehyde
and silanol. Their potential side-effects could be a limitation of
the method.
REFERENCES
■
(1) (a) Powell, C. R.; Dillon, K. M.; Matson, J. B. A review of
hydrogen sulfide (H2S) donors: chemistry and potential therapeutic
applications. Biochem. Pharmacol. 2018, 149, 110−123. (b) Zheng, Y.;
Yu, B.; De La Cruz, L. K.; Roy Choudhury, M.; Anifowose, A.; Wang,
B. Toward hydrogen sulfide based therapeutics: critical drug delivery
and developability issues. Med. Res. Rev. 2018, 38, 57−100. (c) Xu, S.;
Hamsath, A.; Neill, D. L.; Wang, Y.; Yang, C. T.; Xian, M. Chem. -
Eur. J. 2019, 25, 4005−4016.
(2) Kang, J.; Xu, S.; Radford, M. N.; Zhang, W.; Kelly, S. S.; Day, J.
J.; Xian, M. O to S relay deprotection, a general approach for the
design of controllable donors of reactive sulfur species. Angew. Chem.,
Int. Ed. 2018, 57, 5893−5897.
(3) Gupta, V.; Carroll, K. S. Sulfenic acid chemistry, detection and
cellular lifetime. Biochim. Biophys. Acta, Gen. Subj. 2014, 1840, 847−
875.
(4) Goto, K.; Tokitoh, N.; Okazaki, R. Synthesis of a stable
arenesulfenic acid bearing a bowl-shaped macrobicyclic cyclophane
skeleton. Angew. Chem., Int. Ed. Engl. 1995, 34, 1124−1126.
(5) Goto, K.; Holler, M.; Okazaki, R. Synthesis, structure, and
reactions of a sulfenic acid bearing a novel bowl-type substituent: the
first synthesis of a stable sulfenic acid by direct oxidation of a thiol. J.
Am. Chem. Soc. 1997, 119, 1460−1461.
(6) Fries, K. A anthrachioon-sulfur hydroxide. Ber. Dtsch. Chem. Ges.
1912, 45, 2965.
(7) (a) Pal, B. C.; Uziel, M.; Doherty, D. G.; Cohn, W. E. Isolation
and characterization of a pyrimidine sulfenic acid via scission of the
sulfur-sulfur bond in the methyl analog of bis(4-thiouridine) disulfide.
J. Am. Chem. Soc. 1969, 91, 3634−3638. (b) Okuyama, T.; Toyoda,
M.; Fueno, T. Acid-catalyzed hydrolysis of methoxymethyl phenyl
sulfoxide without concomitant racemization. Bull. Chem. Soc. Jpn.
1990, 63, 1316−1321.
ASSOCIATED CONTENT
■
sı
* Supporting Information
(8) (a) Aversa, M. C.; Barattucci, A.; Bonaccorsi, P.; Giannetto, P. L-
Cysteine, a versatile source of sulfenic acids. Synthesis of enantiopure
alliin analogues. J. Org. Chem. 2005, 70, 1986−1992. (b) Gupta, V.;
Carroll, K. S. Profiling the reactivity of cyclic C-nucleophiles towards
electrophilic sulfur in cysteine sulfenic acid. Chem. Sci. 2016, 7, 400−
415.
The Supporting Information is available free of charge at
https://pubs.acs.org/doi/10.1021/acs.orglett.1c01149.
Experimental and characterization of each compound
(PDF)
(9) Adams, H.; Anderson, J. C.; Bell, R.; Neville Jones, D.; Peel, M.
R.; Tomkinson, N. C. O. The synthesis and Diels−Alder reactions of
(E)- and (Z)-1-methoxy-3-(phenylsulfinyl)buta-1,3-dienes. J. Chem.
Soc., Perkin Trans. 1 1998, 23, 3967−3974.
AUTHOR INFORMATION
■
Corresponding Author
(10) Clennan, E. L.; Dillon, D. L. First experimental evidence for the
formation of a silicate anion by intramolecular addition of a
persulfoxide to a trimethylsiloxy group. Tetrahedron Lett. 1998, 39,
6827−6830.
Ming Xian − Department of Chemistry, Brown University,
Providence, Rhode Island 02912, United States;
orcid.org/0000-0002-7902-2987; Phone: 401-863-3588;
Email: ming_xian@brown.edu
(11) Fleming, I.; Perry, D. A. Cyclo-elimination of silyl and sulfoxide
groups in competition with the conventional cyclo-elimination of
sulfoxides. Tetrahedron Lett. 1981, 22, 5095−5096.
Authors
(12) Vedejs, E.; Mullins, M. Silicon-pummerer rearrangements;
Stereoselective sulfinylcarbanion silyation. Tetrahedron Lett. 1975, 16,
2017−2020.
(13) Angelaud, R.; Landais, Y. The phenylthiocyclopropylsilyl
group: a useful latent hydroxy group. Tetrahedron 2000, 56, 2025−
2036.
(14) Davis, F. A.; Friedman, A. J. Trimethylsilyl 2-nitrobenzene-
sulfenate (2-nitrobenzenesulfenic acid). J. Org. Chem. 1976, 41, 897−
898.
(15) Davis, F. A.; Rizvi, S. Q. A.; Ardecky, R.; Gosciniak, D. J.;
Friedman, A. J.; Yocklovich, S. G. Chemistry of sulfenic acids. 1.
Synthesis of trimethylsilyl arenesulfenates (arenesulfenic acids). J. Org.
Chem. 1980, 45, 1650−1653.
Shane S. Kelly − Department of Chemistry, Brown University,
Providence, Rhode Island 02912, United States; Department
of Chemistry, Washington State University, Pullman,
Washington 99164, United States
Tun-Li Shen − Department of Chemistry, Brown University,
Providence, Rhode Island 02912, United States
Complete contact information is available at:
https://pubs.acs.org/10.1021/acs.orglett.1c01149
Notes
The authors declare no competing financial interest.
(16) (a) Nagy, P.; Ashby, M. T. Reactive sulfur species: kinetics and
mechanism of the hydrolysis of cysteine thiosulfinate ester. Chem. Res.
Toxicol. 2007, 20, 1364−1372. (b) Nagy, P.; Ashby, M. T. Reactive
sulfur species: kinetics and mechanism of the oxidation of cystine by
hypochlorous acid to give N,N‘-dichlorocystine. Chem. Res. Toxicol.
2005, 18, 919−923.
ACKNOWLEDGMENTS
■
This work is supported by NIH (R21DA046386) and NSF
(CHE2100870). S.S.K. is supported by the WSU-PNNL
Distinguished Graduate Research Program.
3744
https://doi.org/10.1021/acs.orglett.1c01149
Org. Lett. 2021, 23, 3741−3745

Organic Letters
pubs.acs.org/OrgLett
Letter
(17) (a) Block, E.; Weidman, S. W. Chemistry of alkyl thiolsulfinate
esters. IV. Mechanistic study of the disproportionation reaction. J. Am.
Chem. Soc. 1973, 95, 5046−5048. (b) Koch, P.; Ciuffarin, E.; Fava, A.
Thermal disproportionation of aryl arenethiolsulfinates. Kinetics and
mechanism. J. Am. Chem. Soc. 1970, 92, 5971−5977.
3745
https://doi.org/10.1021/acs.orglett.1c01149
Org. Lett. 2021, 23, 3741−3745