Total Synthesis and Antibacterial Investigation of Plusbacin A3

Article abstract of DOI:10.1021/acs.orglett.7b01629

The total synthesis of plusbacin A₃ (1) has been accomplished using a solvent-dependent diastereodivergent Joullié-Ugi three-component reaction (JU-3CR) as a key step. Two trans-3-hydroxy-l-proline residues were constructed by combining the JU-3CR with a convertible isocyanide strategy. Subsequent peptide coupling and macrolactamization afforded plusbacin A₃. Investigating the antibacterial activity of 1 compared with that of its dideoxy analogue revealed that the threo-β-hydroxyaspartic acid residues are essential for antibacterial activity. Notably, there is a low potential for the development of resistance in S. aureus against plusbacin A₃.

Full text of DOI:10.1021/acs.orglett.7b01629

Letter
pubs.acs.org/OrgLett
Total Synthesis and Antibacterial Investigation of Plusbacin A3
Akira Katsuyama,† Atmika Paudel,‡ Suresh Panthee,‡ Hiroshi Hamamoto,‡ Toru Kawakami,§
Hironobu Hojo,§ Fumika Yakushiji,†,∥ and Satoshi Ichikawa*,†,∥
†Faculty of Pharmaceutical Sciences, Hokkaido University, Kita-12, Nishi-6, Kita-ku, Sapporo 060-0812, Japan
‡Institute of Medical Mycology, Teikyo University, 359 Otsuka, Hachioji, Tokyo 192-0395, Japan
§Institute for Protein Research, Osaka University, 3-2 Yamadaoka, Suita-shi, Osaka 565-0871, Japan
∥Center for Research and Education on Drug Discovery, Hokkaido University, Kita-12, Nishi-6, Kita-ku, Sapporo 060-0812, Japan
S
* Supporting Information
ABSTRACT: The total synthesis of plusbacin A3 (1) has been accomplished using a solvent-dependent diastereodivergent
Joullie−Ugi three-component reaction (JU-3CR) as a key step. Two trans-3-hydroxy-L-proline residues were constructed by
́
combining the JU-3CR with a convertible isocyanide strategy. Subsequent peptide coupling and macrolactamization afforded
plusbacin A3. Investigating the antibacterial activity of 1 compared with that of its dideoxy analogue revealed that the threo-β-
hydroxyaspartic acid residues are essential for antibacterial activity. Notably, there is a low potential for the development of
resistance in S. aureus against plusbacin A3.
he emergence of bacterial resistance to newly developed
of peptidoglycan, resulting in the cessation of polymerization.4
Tdrugs is a severe and ongoing global threat to current Considering the low bacterial resistance against vancomycin,
the mode of action targeting nonprotein precursors in
peptidoglycan biosynthesis promises characteristics of new
antibacterial agents that may be less susceptible to the
development of resistance, although cyclic lipodepsipeptides’
susceptibility to resistance, including 1, has not been
investigated yet. In addition, dual inhibition of teichoic acid
and peptidoglycan biosynthesis has also been reported for 1,5
which exhibits potent antimicrobial activity against drug-
resistant strains, such as methicillin-resistant S. aureus
(MRSA) and vancomycin-resistant Enterococci (VRE) (MIC
0.1−3.13 μg/mL).1,6 Therefore, 1 possesses a mode of action
different from those of existing antibacterial drugs and is
expected to be a promising lead in novel antibacterial drug
discovery. In its chemical structure, 1 contains (R)-3-hydroxy-
isohexadecanoic acid and five nonproteinogenic amino acids: D-
allo-threonine, two trans-3-hydroxy-L-prolines [Pro(3-OH)],
and L- and D-threo-β-hydroxyaspartic acid [Asp(β-OH)]. The
total synthesis of 1 has been reported by VanNieuwenhze et al.7
via a conventional peptide coupling using trans-Pro(3-OH) as a
starting material. Construction of the Pro(3-OH) residue,
which is a key element of 1, requires a number of steps,
therapies against infectious diseases. An important criterion for
selecting among potentially promising natural-product anti-
bacterial agents is their potential to be less susceptible to the
development of resistance. Plusbacin A3 (1), which was isolated
from Pseudomonas sp. PB-6250 contained in a soil sample
collected from Okinawa Island, Japan,1 is a representative cyclic
lipodepsipeptide antibiotic, other examples of which are
empedopeptin2 and tripropeptins3 (Figure 1). This class of
antibiotics exhibits potent antibacterial activity against a wide
range of Gram-positive bacteria. The mode of action of 1 is
anticipated to be inhibition of bacterial cell-wall biosynthesis via
Ca2+-dependent binding to lipid II, which is the final precursor
Received: May 30, 2017
Figure 1. Chemical structure of plusbacin A3 (1).
© XXXX American Chemical Society
A
DOI: 10.1021/acs.orglett.7b01629
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
although several synthetic methods have been developed.8 We
have developed a solvent-dependent diastereodivergent Joul-
over three steps. Then, the carboxylic acid was reacted with 3-
bromocyclohexene to give cyclohexenyl ester 12. Pd-fibroin
complex (Pd-Fib)14 was found to be an efficient hydrogenation
catalyst to suppress the hydrogenolysis of the cyclohexenyl
ester moiety, and the corresponding cyclohexyl ester was
provided in 85% yield. Removal of the isopropylidene group
and simultaneous oxidation of the liberated alcohol were
carried out under Jones’ oxidation conditions, providing the
protected L-Asp(β-OH) (13).7 The condensation between 10
and 13 afforded a depsipeptide in 77% yield, and then the Boc
group was cleaved under acidic conditions to afford 3. The
carboxylic acid of ent-13, which was prepared by the same
procedure as 13 from (S)-Garner’s aldehyde, was protected
with an allyl group. After the removal of the Boc group, the
liberated amine was condensed with Boc-Arg(Cbz)2-OH to
provide dipeptide 14. Then, cleavage of the Boc group
quantitatively afforded amine 4. Protecting allo-D-Thr (14)
with the Boc group followed by condensation with D-Ala-OBn
gave the corresponding dipeptide. The hydroxy group was
further protected with TIPS to give dipeptide 16 in 97% yield
over three steps, and hydrogenolysis of the benzyl group
furnished carboxylic acid 17.
The key strategy for constructing Pro(3-OH) residues 2 and
5 with the diastereoselective JU-3CR is shown in Scheme 4. As
an initial approach, the JU-3CR for the synthesis of 2 or 5 was
examined in HFIP using 2-isocyanophenyl acetate,11a which
was used as a conventional convertible isocyanide; however,
these conditions gave undesirable diastereoselectivity (trans/cis
= 32/68; details are described in the Supporting Information).
From our previous study on the JU-3CR,10 the observed cis
selectivity was presumed to be attributed to the high s character
of the sp2 carbon atom attached to the isocyano group in 2-
isocyanophenyl acetate, and an electron-rich isocyanide was
expected to preferentially provide trans selectivity. To improve
the trans selectivity, convertible isocyanide 18,11b which was
developed by Fukuyama and Kan, was used instead of 2-
isocyanophenyl acetate because the tertiary alkyl group in 18
has a stronger electron-donating inductive effect. As expected,
the JU-3CR between 6, 17, and 18 in HFIP favorably produced
the desired trans isomer trans-19 as a major product in 49%
yield and the cis isomer cis-19 in 29% yield (trans/cis = 62/38).
The secondary amide moiety in trans-19, which was derived
from isocyanide 18, was subsequently converted to the N-
acyloxazolidinone 20,11b and then hydrolysis provided carbox-
ylic acid 2. The other unit 5 was synthesized by a strategy
similar to the one used to synthesize 2, and the JU-3CR
between 6, 18, and Boc-D-Ser(OBn)-OH (21) gave the desired
trans-22 (51%) as well as cis-22 (37%). The sequential
formation of the corresponding oxazolidinone from trans-22
followed by hydrolysis provided the desired carboxylic acid 5.
The completion of the total synthesis is shown in Scheme 5.
Peptide coupling between 3 and 5 provided depsipeptide 23 in
88% yield, and subsequent cleavage of the allyl group gave
carboxylic acid 24 in quantitative yield. The condensation of 2
and 4 afforded the corresponding pentapeptide 25 in 90% yield,
and the Boc group was removed to afford amine 26. Carboxylic
acid 24 was condensed with amine 26 to obtain linear peptide
27 in 64% yield. Removal of the protecting groups on the N-
and C-termini of 27 gave the precursor amino acid, which was
successfully cyclized upon treatment with EDCI, HOAt, and
iPr2NEt in THF to provide fully protected plusbacin A3 in 60%
yield over three steps. Finally, global deprotection with
anhydrous HF in the presence of anisole at −78 to 0 °C
lie−
́
Ugi three-component reaction (JU-3CR)9 that can provide
the desired trans-Pro(3-OH) derivatives in a single step with
concurrent modifications at both the C- and N-termini
(Scheme 1).10 In this reaction, a trans isomer was obtained
Scheme 1. Solvent-Dependent Diastereodivergent JU-3CR
when 1,1,1,3,3,3-hexafluoroisopropanol (HFIP) was used as a
solvent, whereas a cis isomer was the predominant product in
toluene. Moreover, we found that the solvent effect of HFIP
became stronger when using more nucleophilic isocyanides.
Herein, we describe the total synthesis of 1 featuring our
diastereodivergent JU-3CR. The antibacterial activity of 1 and
its deoxy analogue and their susceptibility for drug resistance in
S. aureus were evaluated. Macrocycle 1 was retrosynthetically
disconnected into four units, namely, 2 and 5 containing the
Pro(3-OH) residue at the C-terminus, 3 containing a lipid
chain, and 4 (Scheme 2). Assembling these four units (2−5)
Scheme 2. Retrosynthetic Analysis of Plusbacin A3 (1)
could afford a precursor for macrolactamization. The synthesis
of 2 and 5 was planned to proceed through the JU-3CR using
five-membered cyclic imine 6 and a convertible isocyanide.11
First, the synthesis of 3, 4, and 17 is depicted in Scheme 3.
The lipid side chain of known β-lactone 712 was elongated via
olefin cross metathesis with 2-methylbut-3-en-2-ol to afford
allyl alcohol 8. After dehydrating 8, the resulting diene was
hydrogenated, and β-lactone 9 was hydrolyzed to provide
optically pure (R)-3-hydroxyisohexadecanoic acid. Then, the
carboxy group was protected using allyl bromide to afford
alcohol 10. Known alcohol 11,13 which was prepared from (R)-
Garner’s aldehyde, was protected with a TIPS group, and the
terminal olefin was converted to the aldehyde by ozonolysis.
Further oxidation produced the carboxylic acid in 84% yield
B
DOI: 10.1021/acs.orglett.7b01629
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Scheme 3. Synthesis of Unit Compounds 3, 4, and 17
Scheme 4. Synthesis of Compounds 2 and 5 via JU-3CR
Scheme 5. Completion of the Total Synthesis of Plusbacin
A3
afforded 1 in 66% yield after purification through reversed-
phase HPLC. The analytical data for synthetic 1 were in good
agreement with the previously reported data.7 A dideoxy
analogue 28, in which the L- and D-Asp(β-OH) groups in 1
were replaced with L- and D-Asp, respectively, was also prepared
in a manner similar to the synthesis of 1 (the preparation of 28
is described in the Supporting Information).
The antibacterial activities of 1 and 28 were evaluated (Table
1). Synthetic 1 showed antibacterial activity against methicillin-
sensitive S. aureus (MSSA, Smith and MSSA1) and MRSA
(MR-6) with MIC values of 1−2 μg/mL. However, 28, which
lacks the two hydroxy groups from the Asp(β-OH) residues,
did not exhibit any activity. The CD spectra of 1 and 28
indicated a three-dimensional conformational change that
might be caused by altered intramolecular hydrogen bond
networks (Figure S3). The lack of these two hydroxy groups in
1 induced a dramatic conformational change in the peptide
backbone, resulting in a loss of binding to its target and
ultimately causing a reduction in antibacterial activity. Next, we
assessed resistance acquisition by S. aureus against 1.
Methicillin-sensitive S. aureus (Smith) was treated with sub
MIC levels of 1 over 25 days, and an increase of the MIC values
was measured by a direct comparison with vancomycin and
rifampicin as controls (Figure 2). Under the conditions where a
32 000-fold increase in the MIC was observed after 2 days for
rifampicin, the resistance against 1 was only twice as much as
that of vancomycin, clearly indicating that 1 is much less
susceptible to drug resistance. The genome sequence analysis of
Table 1. Minimum Inhibitory Concentrations (μ/mL) of 1
and 28 against S. aureus
strain
compd
Smith
MR-6
MSSA1
1
2
1
2
28
−
>128
>128
C
DOI: 10.1021/acs.orglett.7b01629
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
25293026), Scientific Research on Innovative Areas “Chemical
Biology of Natural Products” (SI, Grant No. 24102502), The
Ministry of Education, Culture, Sports, Science and Technol-
ogy through Program for Leading Graduate Schools (Hokkaido
University “Ambitious Leader’s Program”), Takeda Science
Foundation, and the Platform Project for Supporting Drug
Discovery and Life Science Research (Platform for Drug
Discovery, Informatics and Structural Life Science).
REFERENCES
■
(1) (a) Shoji, J.; Hinoo, H.; Katayama, T.; Matsumoto, K.; Tanimoto,
T.; Hattori, T.; Higashiyama, I.; MIwa, H.; Motokawa, K.; Yoshida, T.
J. Antibiot. 1992, 45, 817−823. (b) Shoji, J.; Hinoo, H.; Katayama, T.;
Nakagawa, Y.; Ikenishi, Y.; Iwatani, K.; Yoshida, T. J. Antibiot. 1992,
45, 824−831.
Figure 2. Resistance acquisition during serial passaging in the presence
of sub-MIC levels of antimicrobials. The x axis is the number of days,
and the y axis is the number of folds of MIC during passaging.
(2) (a) Konishi, M.; Sugawara, K.; Hanada, M.; Tomita, K.; Tomatsu,
K.; Miyaki, T.; Kawaguchi, H.; Buck, R. E.; More, C.; Rossomano, V.
Z. J. Antibiot. 1984, 37, 949−957. (b) Sugawara, K.; Numata, K.;
Konishi, M.; Kawaguchi, H. J. Antibiot. 1984, 37, 958−964.
(3) (a) Hashizume, H.; Igarashi, M.; Hattori, S.; Hori, M.; Hamada,
M.; Takeuchi, T. J. Antibiot. 2001, 54, 1054−1059. (b) Hashizume, H.;
Hirosawa, S.; Sawa, R.; Muraoka, Y.; Ikeda, D.; Naganawa, H.; Igarashi,
M. J. Antibiot. 2004, 57, 52−58.
plusbacin A3-resistant strains obtained from the resistance
acquisition experiments (three strains) was performed, and
mutations in proteins involved in bacterial cell-wall biosynthesis
were found (Figure S4). In particular, a single mutation in
VraE, which is linked to bacitracin and nisin resistance,15 might
be involved in plusbacin A3 resistance. Nisin is a known binder
to lipid II; thus, these results support the proposed mechanism
of 1.
In conclusion, we have achieved the total synthesis of
plusbacin A3 (1) as well as its dideoxy analogue 28 via a JU-
3CR with a five-membered cyclic imine and convertible
isocyanide 18. In particular, the stereochemical outcome was
adjusted by the electron density of the isocyanide building
block. Evaluation of the antibacterial activity and analysis of the
structure via CD suggested that the hydroxyl groups of the
Asp(β-OH) residues in 1 are required to maintain its
antibacterial activity. Furthermore, it was revealed that
plusbacin A3 induces only low-level resistance. These results
revealed several promising properties of 1, corroborating its
potential as a lead compound for novel antibacterial agent
development and mechanistic studies.
(4) Muller, A.; Munch, D.; Schmidt, Y.; Reder-Christ, K.; Schiffer, G.;
̈
̈
Bendas, G.; Gross, H.; Sahl, H.-G.; Schneider, T.; Brotz-Oesterhelt, H.
̈
J. Biol. Chem. 2012, 287, 20270−20280.
(5) (a) Kim, S. J.; Singh, M.; Wohlrab, A.; Yu, T.-Y.; Patti, G. J.;
O’Connor, R. D.; Vannieuwenhze, M.; Schaefer, J. Biochemistry 2013,
52, 1973−1979. (b) O’Connor, R. D.; Singh, M.; Chang, J.; Kim, S. J.;
VanNieuwenhze, M.; Schaefer, J. J. Phys. Chem. B 2017, 121, 1499−
1505.
(6) Maki, H.; Miura, K.; Yamano, Y. Antimicrob. Agents Chemother.
2001, 45, 1823−1827.
(7) Wohlrab, A.; Lamer, R.; VanNieuwenhze, M. S. J. Am. Chem. Soc.
2007, 129, 4175−4177.
(8) (a) Herdeis, C.; Hubmann, H. P.; Lotter, H. Tetrahedron:
Asymmetry 1994, 5, 119−128. (b) Lee, J. H.; Kang, J. E.; Yang, M. S.;
Kang, K. Y.; Park, H. K. Tetrahedron 2001, 57, 10071−10076.
(c) Hughes, P.; Clardy, J. J. Org. Chem. 1989, 54, 3260−3264.
(d) Kumar, T. P.; Chandrasekhar, S. Synthesis 2012, 44, 2889−2894.
(e) Durand, J.-O.; Larcheveq
39, 5743−5746. (f) Ewing, W. R.; Joullie,
2843−2850.
(9) (a) Nutt, R. F.; Joullie,
̂
ue, M.; Petit, Y. Tetrahedron Lett. 1998,
́
M. M. Heterocycles 1988, 27,
ASSOCIATED CONTENT
* Supporting Information
■
S
́
M. M. J. Am. Chem. Soc. 1982, 104, 5852−
́
5853. (b) Bowers, M. M.; Carroll, P.; Joullie, M. M. J. Chem. Soc.,
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.or-
glett.7b01629.
Perkin Trans. 1 1989, 857−865.
(10) Katsuyama, A.; Matsuda, A.; Ichikawa, S. Org. Lett. 2016, 18,
2552−2555.
1
Experimental details; H, 13C NMR and mass spectra
(11) (a) Pirrung, M. C.; Ghorai, S. J. Am. Chem. Soc. 2006, 128,
11772−11773. (b) Rikimaru, K.; Yanagisawa, A.; Kan, T.; Fukuyama,
T. Heterocycles 2007, 73, 403−417. (c) Lindhorst, T.; Bock, H.; Ugi, I.
Tetrahedron 1999, 55, 7411−7420. (d) van der Heijden, G.; Jong, J. A.
W.; Ruijter, E.; Orru, R. V. A. Org. Lett. 2016, 18, 984−987.
(12) Nelson, S. G.; Peelen, T. J.; Wan, Z. J. Am. Chem. Soc. 1999, 121,
9742−9743.
(13) Srivastava, A. K.; Panda, G. Chem. - Eur. J. 2008, 14, 4675−4688.
(14) Ikawa, T.; Sajiki, H.; Hirota, K. Tetrahedron 2005, 61, 2217−
2231.
(PDF)
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: ichikawa@pharm.hokudai.ac.jp.
ORCID
Satoshi Ichikawa: 0000-0001-5345-5007
Notes
(15) (a) Song, Y.; Lunde, C. S.; Benton, B. M.; Wilkinson, B. J.
Antimicrob. Agents Chemother. 2012, 56, 3157−3164. (b) Hiron, A.;
The authors declare no competing financial interest.
Falord, M.; Valle, J.; Deb
2011, 81, 602−622.
́
arbouille,
́
M.; Msadek, T. Mol. Microbiol.
ACKNOWLEDGMENTS
■
We thank Prof. Kan, T. for the helpful suggestion to use the
convertible isocyanate 18. We also thank Ms. S. Oka (Center
for Instrumental Analysis, Hokkaido University) for measure-
ment of the mass spectra. This research was supported by JSPS
Grant-in-Aid for Scientific Research (B) (SI, Grant No.
D
DOI: 10.1021/acs.orglett.7b01629
Org. Lett. XXXX, XXX, XXX−XXX