Asymmetric Total Synthesis Enables Discovery of Antibacterial Activity of Siladenoserinols A and H

Article abstract of DOI:10.1021/acs.orglett.9b03857

Siladenoserinols A and H were found to show moderate inhibitory activity toward p53-Hdm2 interactions. Our total synthesis allowed us to further examine their bioactivities, which revealed that (i) siladenoserinols A and H were not cytotoxic against cancer cell lines and (ii) siladenoserinol A and its desulfamate analogue exhibited significant antibacterial activity against Gram-positive bacteria including MRSA. Our studies demonstrate that siladenoserinols are a promising new class of bactericidal Gram-positive antibiotics without hemolytic activity.

Full text of DOI:10.1021/acs.orglett.9b03857

Letter
pubs.acs.org/OrgLett
Cite This: Org. Lett. XXXX, XXX, XXX−XXX
Asymmetric Total Synthesis Enables Discovery of Antibacterial
Activity of Siladenoserinols A and H
†,∥
Miguel Adrian Marquez-Cadena, Jingyun Ren,†,∥,§ Wenkang Ye,‡ Peiyuan Qian,*,‡
́
́
and Rongbiao Tong*,†
†Department of Chemistry, The Hong Kong University of Science and Technology, Hong Kong, China
‡Department of Ocean Science, Division of Life Science and Hong Kong Branch of the Southern Marine Science and Engineering
Guangdong Laboratory, The Hong Kong University of Science and Technology, Hong Kong, China
S
* Supporting Information
ABSTRACT: Siladenoserinols A and H were found to show moderate inhibitory activity toward p53-Hdm2 interactions. Our
total synthesis allowed us to further examine their bioactivities, which revealed that (i) siladenoserinols A and H were not
cytotoxic against cancer cell lines and (ii) siladenoserinol A and its desulfamate analogue exhibited significant antibacterial
activity against Gram-positive bacteria including MRSA. Our studies demonstrate that siladenoserinols are a promising new
class of bactericidal Gram-positive antibiotics without hemolytic activity.
iladenoserinols A (1) and H (2) along with 10 congeners
were isolated in 2013 from a tunicate of the Didemnidae
S
family and found to show inhibitory activity against p53-Hdm2
interactions (IC50 values of 2.0 and 2.5 μM).1 They are
structurally very unique natural products because they contain
a medicinally privileged 6,8-dioxabicyclo[3.2.1]octane (6,8-
DOBCO) core,2 serinol sulfamate, and a betaine α-
glycerophosphocholine (α-GPC) through 8- and 14-carbons
linkages, respectively. In 2018, Doi3 and co-workers
documented the first total synthesis of siladenoserinol A and
found that the inhibitory activity of 1 against p53-Hdm2
interaction was much lower (IC50 = 17 μM) than originally
reported. Our interest in siladenoserinols was prompted by
possible construction of the 6,8-DOBCO core with our
previously established strategy4 and, most importantly, by
our hypothesis that siladenoserinol A might exhibit anti-
bacterial activity due to the likely interaction of the betaine α-
GPC and hydrophilic serinol sulfamate with the polar,
negatively charged cell wall of bacteria.5 Herein, we describe
the asymmetric total synthesis and verification of antibacterial
activity of siladenoserinol A and its analogues (Figure 1).
Siladenoserinol H was also synthesized for the first time
through a rather unexpected late-stage ester 1,2-migration,
which greatly expanded the utility of our strategy for the
synthesis of other members of siladenoserinols.
Figure 1. Molecular structures of siladenoserinols A and H.
olefination (5 + 6). The 6,8-DOBCO 5 could be synthesized
by our previously established Achmatowicz rearrangement/
bicycloketalization7 of furfuryl diol 8. The preparation of α-
GPC 6a could be achieved by direct regioselective silylation
(triisopropylsilyl, TIPS) of commercially available α-GPC (7)
followed by acylation.
As depicted in Scheme 2, our synthetic study began with the
preparation of furfuryl diol 8 in 85% yield through HWE
olefination of furfuryl aldehyde 9 with phenyltetrazole sulfone
and Sharpless asymmetric dihydroxylation.8 Using our green
As illustrated in Scheme 1, our synthetic strategy featured a
convergent assembling of three functional sectors: serinol
derivative 3, 6,8-DOBCO 5, and α-GPC 6a, through
Williamson etherification (3 + 4), Julia−Kocienski olefination
(4 + 5),6 and Horner−Wadsworth−Emmons (HWE)
Received: October 29, 2019
© XXXX American Chemical Society
A
DOI: 10.1021/acs.orglett.9b03857
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
the double bond on the side chain was simultaneously
achieved by palladium-catalyzed hydrogenation. Dess−Martin
periodinane (DMP) oxidation of the resulting alcohol
delivered the key aldehyde 14 for HWE olefination with α-
GPC 6a (14 → 15, Scheme 3), which was identified to be
Scheme 1. Retrosynthetic Analysis of Siladenoserinol A
Scheme 3. Completion of Total Syntheses of
Siladenoserinolipids A and H
Scheme 2. Synthesis of 6,8-DOBCO Framework 14
feasible after extensive experimentation.10 The practical
challenge to install this betaine α-GPC warranted some
comments on other attempted methods. First of all, the
hygroscopic and water-soluble properties of betaine α-GPC
(7) added an enormous challenge in the preparation of α-GPC
derivatives (6a−6d).11 Second, various classical and well-
established esterification methods including Yamaguchi,
Steglich, Mitsunobu, SOCl2, and carbonyldiimidazole (CDI)
to introduce the known α-GPC 6b (or 6c) did not work.10
Third, HWE olefination of aldehyde 14 with 6d under
Masamune−Roush conditions (note that Doi used it with
good yield), resulted in an inferior yield of 15a.
As depicted in Scheme 3, we found that α-GPC 6a
underwent efficient HWE olefination under Masamune−
Roush conditions (LiCl/DBU)12 to provide the desired
product 15 in 77% yield.13 Next, we removed both TBS and
TIPS protecting groups and acetylated the resulting alcohols as
diacetate 16. Removal of acetonide and Boc with 2.0 M HCl in
dioxane followed by chemoselective sulfamate formation with
SO3-py furnished siladenoserinol A (21 mg). Notably, our total
synthesis required only 16 steps (longest linear sequence),
which was considerably shorter than Doi’s synthesis (LLS 24
steps). Interestingly, we found that subjection of 16 to 2.0 M
HCl in Et2O (freshly prepared) and then SO3-py delivered
both siladenoserinols A (1, 5.7 mg) and H (2, 5.8 mg) as a 1:1
mixture, which could be separated by reversed-phase HPLC.
The unexpected production of siladenoserinol H was
attributed to unanticipated HCl-mediated deacetylation of α-
GPC sector of 16 followed by 1,2-acyl migration14 to the
primary position (17b → 17c). The spectroscopic data and
optical rotation of our synthetic samples were well consistent
with those reported, which suggested that the natural
Oxone−KBr protocol,9 Achmatowicz rearrangement of 8
proceeded efficiently and the crude product cyclized in the
presence of CSA to provide 6,8-DOBCO 10 in 64% yield.
Hydrogenation of the double bond of 10 under palladium
catalysis followed by K-Selectride reduction produced a 1:1
diastereomeric mixture of alcohol 11.
The undesired one (11b) could be partially recovered by
DMP oxidation and then reduction with K-Selectride to deliver
a 1:1 mixture of 11a and 11b. It was noted that our six-step
synthesis of 6,8-DOBCO 11b was highly efficient as compared
to the 15-step preparation of related 6,8-DOBCO through Au-
catalyzed double hydrohydroxylation by Doi.3 TBS protection
of the hydroxyl group of 11a and DIBAL-H reduction of the
ester afforded the required aldehyde for Julia−Kocienski
olefination of serinol-derived phenyltetrazole sulfone 13.
Removal of the benzyl protecting group and saturation of
B
DOI: 10.1021/acs.orglett.9b03857
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Table 1. Cytotoxicity and Antibacterial Activity
IC50 (μM)
MIC (μM)
cancer cell lines
Gram-negative bacteria
Gram-positive bacteria
b
compd
A549
MGC-803
E. coli
mutant E. coli
P. aeruginosa
M. luteus
B. subtilis
MRSA
1
2
16
17a
18
19
C1*
C2*
C3*
100
>100
>100
65
>100
>100
>100
78
>100
>100
35
>100
>100
8
>50
>50
>50
>50
>50
>50
8
6
>50
10
>50
16
>40
>40
>40
>40
>40
>40
8
20
>40
8
>40
20
6
20
>40
8
>40
16
6
20
>200
10
>200
10
a
a
6
>40
6
0.8
>60
>40
40
1
3
0.8
>60
0.5
a
a
C1*: cisplatin (control). C2*: kanamycin (control). C3*: Vancomycin (control). All bioactivity experiments were performed for three times.
Mutant E. coli is outer-membrane permeable and Low MIC value suggests the outer-membrane is not target.
b
siladenoserinols A and H were successfully synthesized in an
enantioselective way.
Cytotoxicity and Antibacterial Activity
Since siladenoserinol A was reported to be cytotoxic against
cancer cell line A549 (80% cell viability at 10 μM), we first re-
examined the cytotoxicity of siladenoserinols A and H, 16 and
17a, and simplified analogues 18 and 19 (Table 1).15 To our
surprise, none of these compounds were cytotoxic against
A549 and MGC-803. Next, we tested our hypothesis that these
surfactant-like compounds might possess antibacterial activity
as many antibacterial quaternary ammonium compounds.16 To
our excitement, siladenoserinol A and desulfamate analogue
17a exhibited potent and selective antibacterial activity against
Gram-positive bacteria including Micrococcus luteus (MIC = 8
μM), Bacillus subtilis (MIC = 6 μM) and methicillin-resistant
Staphylococcus aureus (MRSA, MIC = 6 μM). Intriguingly,
siladenoserinol H (2) was much less potent, which suggests the
ester linkage of at C2′ of the α-GPC sector was beneficial to
enhance antibacterial activity. No antibacterial activity was
observed for 18, which suggested that the absence of 6,8-
DOBCO (18) as compared to siladenoserinol A resulted in a
loss of antibacterial activity. In contrast to 17a and 1, their
precursor 16 did not show any antibacterial activity. These
results indicated that the hydrophilic serinol sector and its
sulfamate analogue were important to antibacterial activity. We
also evaluated the antibacterial activity of these compounds
against other Gram-negative bacteria including Enterobacter
cloacae, Klebsiella pneumoniae, Acinetobacter baumannii, and
Pseudomonas aeruginosa,17 and no obvious activity (MIC > 50
μM) was observed (see the Supporting Information). The
highly specific activity against Gram-positive bacteria was
attributed to their amphiphilic property: these compounds
containing betaine α-GPC sector belong to quaternary
ammonium compounds (QACs) which have been known as
cationic surfactants with potent antimicrobial activity because
the cationic ammonium can interact with the negatively
charged cell membrane of bacteria, disrupt its integrity, and
lead to the leakage of cytoplasmic matters.18
Figure 2. Time-kill experiment. MRSA were grown to early phase
(2h) and added ten times the MIC of antibiotics to the culture.
vancomycin, a quick bactericidal effect was observed after
bacteria exposure to our synthetic compounds. These results
further consolidated the observation of quick bactericidal
activity against MRSA biofilm.
Hemolytic Activity and Analysis of Lysis
Finally, we evaluated hemolytic activity and lysis of
siladenoserinols and its analogues with fresh red blood cells
isolated from healthy rabbits (Figure 3). Siladenoserinols A
and H and 17a did not show significant hemolytic activity
(<5%) at a concentration of >50 μg/mL. Therefore, these
promising antibacterial compounds were in the safety range
(5%), which holds a great promise for further pharmaceutical
development. Our lysis experiments (Figure 3d) indicated that
these compounds did not break down the mammalian cell
membrane and thus can be considered safe. We should not,
however, rule out bacterial membrane cell as the target of these
compounds. It would be necessary to perform SYTOX or PI
assays in live bacteria in order to determine the membrane
targeting.20 It should be noted that penicillin and related β-
lactam antibiotics killed bacteria of defective cell wall through
enzymatic lysis.21 Although many more efforts needed to
elucidate the target and the mode of action, we believe
siladenoserinols are a new promising class of Gram-positive
antibiotics that might combat antibiotic-resistant bacteria
(MRSA).
Time-Kill Experiments
To further determine whether they were bactericidal or
bacteriostatic,19 we performed time-kill kinetics experiments
(Figure 2). We found that all four examined compounds:
siladenoserinols A(1)/H(2) and analogues 17a/19, exerted
rapid in vitro bactericidal activity (the colony-forming unit has
a 3-log deduction in one hour). Notably, compared with
In summary, we have achieved the asymmetric total
syntheses of siladenoserinols A and H through a very
convergent strategy, which features (i) highly efficient
C
DOI: 10.1021/acs.orglett.9b03857
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Chemistry & Materials Science, Northwest University, Xi’an,
Shaanxi, 710127, P.R. China.
Author Contributions
∥M.A.M.-C. and J.R. contributed equally.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This research was financially supported by Research Grant
Council of Hong Kong (GRF 16303617, GRF 16304618, and
GRF 16311716) and Hong Kong Branch of Southern Marine
Science and Engineering Guangdong Laboratory (Guangzhou)
(SMSEGL20Sc01). We are very grateful to Prof. Guang Zhu
(HKUST) for providing assistance with NMR data collection.
REFERENCES
■
(1) Nakamura, Y.; Kato, H.; Nishikawa, T.; Iwasaki, N.; Suwa, Y.;
Rotinsulu, H.; Losung, F.; Maarisit, W.; Mangindaan, R. E. P.;
Morioka, H.; Yokosawa, H.; Tsukamoto, S. Org. Lett. 2013, 15, 322.
(2) Examples of the biologically active natural products containing
the 6,8-DOBCO core: (a) Uemura, D.; Chou, T.; Haino, T.; Nagatsu,
A.; Fukuzawa, S.; Zheng, S.; Chen, H. J. Am. Chem. Soc. 1995, 117,
1155. (b) McCauley, J. A.; Nagasawa, K.; Lander, P. A.; Mischke, S.
G.; Semones, M. A.; Kishi, Y. J. Am. Chem. Soc. 1998, 120, 7647.
(c) Tanaka, N.; Suenaga, K.; Yamada, K.; Zheng, S.; Chen, H.;
Uemura, D. Chem. Lett. 1999, 28, 1025. (d) Mitchell, S. S.; Rhodes,
D.; Bushman, F. D.; Faulkner, D. Org. Lett. 2000, 2, 1605. (e) Yin, S.;
Fan, C.; Dong, L.; Yue, J. Tetrahedron 2006, 62, 2569.
Figure 3. Hemolytic activity in different concentration. (a) Hemolytic
activity of siladenoserinol A (1), (b) hemolytic activity of
siladenoserinol H (2), (c) hemolytic activity of compound 17a, and
(d) lysis experiments of 1, 2, 19, and vancomycin (positive control)
against MRSA (200 μL of MRSA culture liquid with OD600 value
equal to 1.0, then treated with compound at 10 times of the MIC for
24 h. Hemolytic activity was determined with fresh red blood cells
isolated from healthy rabbits and used 10% Triton X-100 as a positive
control.
enantioselective construction of a 6,8-DOBCO core (six steps)
with Achmatowicz rearrangement−bicyclization and (ii)
efficient unification of betaine α-glycerylphosphorylcholine
derivative and 6,8-DOBCO using robust Horner−Wads-
worth−Emmons olefination. Our total synthesis enables us
to examine their unknown biological activity, which reveals
that siladenoserinol A and its desulfamate analogue 17a are not
cytotoxic against two cancer cell lines but exhibited potent and
selective antibacterial activity against Gram-positive bacteria.
Time-kill kinetics, hemolysis, and lysis experiments suggest
that siladenoserinols and analogues are bactericidal but not
lysing the cell membrane. A new mode of action was suspected
to operate, and more experiments are needed to fully elucidate
the mechanism and target for this new class of antibacterial and
antibiofilm compounds.
(3) Yoshida, M.; Saito, K.; Kato, H.; Tsukamoto, S.; Doi, T. Angew.
Chem., Int. Ed. 2018, 57, 5147−5150.
(4) Ren, R.; Tong, R. J. Org. Chem. 2014, 79, 6987.
(5) For selected reviews, see: (a) Winum, J.; Scozzafava, A.;
Montero, J.; Supuran, C. T. Med. Res. Rev. 2005, 25, 186.
(b) Petkowski, J. J.; Bains, W.; Seager, S. J. Nat. Prod. 2018, 81,
423. (c) Tischer, M.; Pradel, G.; Ohlsen, K.; Holzgrabe, U.
ChemMedChem 2012, 7, 22.
(6) (a) Blakemore, P. R.; Cole, W. J.; Kocienski, P. J.; Morley, A.
Synlett 1998, 26. For a recent review, see: (b) Chatterjee, B.; Bera, S.;
Mondal, D. Tetrahedron: Asymmetry 2014, 25, 1.
(7) For a review, see: (a) Zhang, W.; Tong, R. J. Org. Chem. 2016,
81, 2203. For selected examples see: (b) Ren, J.; Wang, J.; Tong, R.
Org. Lett. 2015, 17, 744. (c) Ren, J.; Liu, Y.; Song, L.; Tong, R. Org.
Lett. 2014, 16, 2986.
(8) Morikawa, K.; Park, J.; Andersson, P. G.; Hashiyama, T.;
Sharpless, K. B. J. Am. Chem. Soc. 1993, 115, 8463.
ASSOCIATED CONTENT
* Supporting Information
(9) Li, Z.; Tong, R. J. Org. Chem. 2016, 81, 4847.
■
(10) Ren, J. Studies on Diastereoselective Dichlorination and Total
Syntheses of Natural Products with 6,8-Dioxabicyclo[3.2.1]octane
Framework. Ph.D. Thesis; Hong Kong University of Science and
Technology, 2015.
S
The Supporting Information is available free of charge at
https://pubs.acs.org/doi/10.1021/acs.orglett.9b03857.
(11) (a) Wang, P.; Blank, D. H.; Spencer, T. A. J. Org. Chem. 2004,
69, 2693. (b) Jorgensen, M. R.; Bhurruth-Alcor, Y.; Røst, T.; Bohov,
̈
P.; Muller, M.; Guisado, C.; Kostarelos, K.; Dyrøy, E.; Berge, R. K.;
Experimental procedures and analytical data for all new
compounds (PDF)
Miller, A. D.; Skorve, J. J. Med. Chem. 2009, 52, 1172. (c) Huang, Z.;
Szoka, F. C., Jr. J. Am. Chem. Soc. 2008, 130, 15702. (d) Gu, X.; Sun,
M.; Gugiu, B.; Hazen, S.; Crabb, J. W.; Salomon, R. G. J. Org. Chem.
2003, 68, 3749.
(12) Blanchette, M. A.; Choy, W.; Davis, J. T.; Essenfeld, A. P.;
Masamune, S.; Roush, W. R.; Sakai, T. Tetrahedron Lett. 1984, 25,
2183.
AUTHOR INFORMATION
Corresponding Authors
■
*E-mail: rtong@ust.hk.
*E-mail: boqianpy@ust.hk.
ORCID
(13) In a Ph.D. thesis (2015) we documented the synthesis of
compound 15 using HWE olefination under Roush−Masamune
conditions; see ref 10.
(14) (a) Otera, J. Chem. Rev. 1993, 93, 1449. (b) Ren, B.; Rahm, M.;
Zhang, Z.; Zhou, Y.; Dong, H. J. Org. Chem. 2014, 79, 8134.
Rongbiao Tong: 0000-0002-2740-5222
Present Address
§(J.R.) Key Laboratory of Synthetic and Natural Functional
Molecule Chemistry of Ministry of Education, College of
D
DOI: 10.1021/acs.orglett.9b03857
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
(c) Laszlo, J. A.; Compton, D. L.; Vermillion, K. E. J. Am. Oil Chem.
Soc. 2008, 85, 307.
(15) The synthesis of simplified analogues 18 and 19 was
successfully achieved in a similar manner to the total synthesis of
17a and 1, respectively.
(16) For a leading review, see: Jennings, M. C.; Minbiole, K. P.;
Wuest, W. M. ACS Infect. Dis. 2015, 1, 288.
(17) For more information about ESKAPE pathogens, see: Boucher,
H. W.; Talbot, G. H.; Bradley, J. S.; Edwards, J. E.; Gilbert, D.; Rice,
L. B.; Scheld, M.; Spellberg, B.; Bartlett, J. Clin. Infect. Dis. 2009, 48, 1.
(18) Gerba, C. P. Appl. Environ. Microbiol. 2015, 81, 464−469.
(19) Li, Y.; Zhong, Z.; Zhang, W.; Qian, P. Nat. Commun. 2018, 9,
3273.
(20) Two references that support bacterial membrane targeting:
(a) Kim, W.; Zou, G.; Hari, T. P. A.; Wilt, I. K.; Zhu, W.; Galle, N.;
Faizi, H. A.; Hendricks, G. L.; Tori, K.; Pan, W.; Huang, X.; Steele, A.
D.; Csatary, E. E.; Dekarske, M. M.; Rosen, J. L.; Queiroz-Ribeiro, N.;
Lee, K.; Port, J.; Fuchs, B. B.; Vlahovska, P. M.; Wuest, W. M.; Gao,
H.; Ausubel, F. M.; Mylonakis, E. Proc. Natl. Acad. Sci. U. S. A. 2019,
116, 16529. (b) Kim, W.; Zhu, W.; Hendricks, G. L.; Van Tyne, D.;
Steele, A. D.; Keohane, C. E.; Fricke, N.; Conery, A. L.; Shen, S.; Pan,
W.; Lee, K.; Rajamuthiah, R.; Fuchs, B. B.; Vlahovska, P. M.; Wuest,
W. M.; Gilmore, M. S.; Gao, H.; Ausubel, F. M.; Mylonakis, E. Nature
2018, 556, 103.
(21) For selected reviews, see: (a) Waxman, D. J.; Strominger, J. L.
Annu. Rev. Biochem. 1983, 52, 825. (b) Tomasz, A. Annu. Rev.
Microbiol. 1979, 33, 113.
E
DOI: 10.1021/acs.orglett.9b03857
Org. Lett. XXXX, XXX, XXX−XXX