A Synthesis Strategy for the Production of a Macrolactone of Gulmirecin A via a Ni(0)-Mediated Reductive Cyclization Reaction

Article abstract of DOI:10.1021/acs.orglett.0c00665

A synthesis strategy for the production of a key synthetic intermediate of gulmirecin A was described. The key reaction in the preparation of the 12-membered macrolactone is the Ni(0)-mediated reductive cyclization reaction of ynal using an N-heterocyclic carbene ligand and silane reductant. In addition, the α-selective glycosylation reaction of the macrolactone was performed to demonstrate the synthesis of gulmirecin and disciformycin precursors.

Full text of DOI:10.1021/acs.orglett.0c00665

pubs.acs.org/OrgLett
Letter
A Synthesis Strategy for the Production of a Macrolactone of
Gulmirecin A via a Ni(0)-Mediated Reductive Cyclization Reaction
Shun Kitahata, Akira Katsuyama, and Satoshi Ichikawa*
Cite This: https://dx.doi.org/10.1021/acs.orglett.0c00665
Read Online
sı
*
Metrics & More
Article Recommendations
Supporting Information
ACCESS
ABSTRACT: A synthesis strategy for the production of a key synthetic intermediate of gulmirecin A was described. The key
reaction in the preparation of the 12-membered macrolactone is the Ni(0)-mediated reductive cyclization reaction of ynal using an
N-heterocyclic carbene ligand and silane reductant. In addition, the α-selective glycosylation reaction of the macrolactone was
performed to demonstrate the synthesis of gulmirecin and disciformycin precursors.
ince the discovery of antibiotics and vaccines, the number
These natural products 1−4 had no cytotoxic effects on human
Sof deaths due to infectious diseases has drastically cells at concentrations of up to 10 μM. Because these natural
products have excellent properties for avoiding cross-resistance
with methicillin and vancomycin, they have the potential to
become effective antibacterial leads with activity against
multidrug-resistant bacteria although their targets have not
yet been clarified. The total synthesis of disciformycins has
decreased. However, the frequent prescription and unregulated
usage of antibiotics against infectious diseases have permitted
drug-resistant bacterial pathogens to spread rapidly.1 Accord-
ing to the antibiotic resistance threats reported by the U.S.
Centers for Disease Control and Prevention in 2019, more
than 2.8 million antibiotic-resistant infections occur in the U.S.
each year, causing 35,000 deaths.2 Hence, the development of
new antibacterial drugs with novel mechanisms of action is
urgently needed to avoid cross-resistance with approved drugs.
In 2014, gulmirecin A (1) and B (2) were isolated from
Pyxidicoccus fallax HKI 727 as potential antimicrobial agents
against Gram-positive bacteria, including methicillin-resistant
Staphylococcus aureus (MRSA) and vancomycin-resistant
Enterococci (VRE) (Figure 1).3 In the same year, disciformycin
A (3) and B (4) were isolated from Pyxidicoccus fallax
AndGT8 for similar use as potential antimicrobial agents.4
been accomplished by the groups of Mu
ller5 and Furstner,6
̈ ̈
and synthesis studies of disciformycin B and gulmirecin B have
also been reported by the groups of Kirschning7 and Maier,8
respectively. Despite considerable effort to synthesize gulmir-
ecins and disciformycins, no studies of the synthesis of their
analogues of gulmirecin A have been reported, and their
structure−activity relationships must still be elucidated.
Structurally, gulmirecins and disciformycins are closely
related but differ in the region of the C2−C4 positions and
in the presence or absence of an isovaleryl group attached to
the hydroxy group at the C6 position. The group of Kirschning
proposed a biosynthetic relationship between gulmirecins and
disciformycins (Scheme 1).7 First, the thioesterase on the
polyketide synthase GulF, which binds substrate 5, produces
the macrolactone and subsequent oxidation at the C6 position
Received: February 20, 2020
Figure 1. Structures of gulmirecins and disciformycins.
https://dx.doi.org/10.1021/acs.orglett.0c00665
© XXXX American Chemical Society
Org. Lett. XXXX, XXX, XXX−XXX
A

Organic Letters
pubs.acs.org/OrgLett
Letter
Scheme 1. Proposed Biosynthetic Relationship between
Gulmirecins and Disciformycins
Scheme 3. Ni(0)-Mediated Reductive Coupling Reaction
a
a
KS = β-ketoacyl synthase, AT = acyl transferase, DH = dehydratase,
ER = enoyl reductase, KR = ketoreductase, ACP = acyl carrier
protein, TE = thioesterase.
catalyst. According to previous research, the regioselectivity of
the oxidative cyclization can be controlled by the size of the
ligand used.9 This reaction is useful for the synthesis of natural
products because aldehydes and alkynes are easily synthesized,
and it can be applied to intramolecular reactions when ynal is
the substrate.10 It was assumed that the reductive coupling
with a Ni(0) catalyst described above would be useful for the
synthesis of intermediate 6. Ynal 8 was removed from
carboxylic acid 9 and alcohol 10 via esterification. Carboxylic
acid 9 would be produced by an Evans aldol reaction
establishing the stereochemistry of the methyl group at the
C2 position and the hydroxy group at the C3 position. Alcohol
10 would be synthesized by an epoxide-opening reaction with
1-propyne.
by Cyp450 gives gulmirecin B (2). Second, oxidation at the C3
position and the esterification of the hydroxy group at the C6
position of 2 result in gulmirecin A (1). Finally, the elimination
of the hydroxy group at the C3 position can produce
disciformycin A (3) and B (4). This proposed biosynthetic
pathway motivated the development of a standardized
synthesis route applicable to this class of natural products
using a gulmirecin A-type macrolactone as a key synthetic
intermediate (6). Here, the synthesis of a gulmirecin A-type
macrolactone through a Ni(0)-mediated reductive cyclization
reaction is reported as an example of a novel method for
preparing macrolides.
The synthesis commenced with the preparation of 9
(Scheme 4). The diol of commercially available diethyl L-
A retrosynthetic analysis of 1 is shown in Scheme 2. The
final stage of the synthesis was planned to include the
Scheme 4. Synthesis of Carboxylic Acid 9
Scheme 2. Retrosynthetic Analysis of Gulmirecin A (1)
introduction of a D-arabinose moiety (using trichloroacetimi-
date 7) and an isovaleryl group into the macrolactone and the
oxidation of the hydroxy group at the C5 position. The 12-
membered macrolactone 6 was to be synthesized via a Ni(0)-
catalyzed reductive coupling reaction9 involving ynal 8. In the
Ni(0)-catalyzed reductive coupling reaction reported by the
group of Montgomery, an alkyne and an aldehyde react over a
Ni(0) catalyst with phosphine or an N-heterocyclic carbene as
a ligand to give an allyl alcohol. The proposed reaction
mechanism is shown in Scheme 3. Initially, the ligand and
Ni(0) form a metal complex (13), which forms a π-bonded
complex with the aldehyde and alkyne. The oxidative
cyclization of this complex produces five-membered metalla-
cycle intermediate 14. Subsequently, σ bond metathesis of the
nickel and oxygen bonds proceeds with the use of a reducing
agent such as silane or borane to give 15. Finally, the reductive
elimination of nickel gives the allyl alcohol or the
corresponding silyl ether with the regeneration of the Ni(0)
(+)-tartrate (17) was protected by forming an acetonide,
which was followed by the reduction of the diester of 18 to
give the corresponding diol 19 in 86% yield in two steps. The
monoprotection of the resulting hydroxy groups by a TBS
group gave alcohol 20. The Appel reaction of 20 was
conducted to afford the corresponding iodide, which was
reacted with vinylmagnesium bromide in the presence of CuI
to give the alkene 21 in 83% yield in two steps. Although the
olefin of 21 was oxidized by treatment with O3, the desired
aldehyde 22 was not obtained because of the decomposition of
the substrate. The oxidative cleavage of the olefin by OsO4−
NaIO4 with 2,6-lutidine11 produced aldehyde 22 in quantita-
tive yield. In the case of the Evans aldol reaction between 22
B
https://dx.doi.org/10.1021/acs.orglett.0c00665
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
pubs.acs.org/OrgLett
Letter
n
and oxazolidinone 23 with Bu2BOTf as a Lewis acid, the
desired alcohol 24 was not obtained. On the other hand, the
TiCl4-mediated aldol reaction12 proceeded smoothly to give
the desired alcohol 24 stereoselectively in 96% yield in two
steps. The protection of the resulting alcohol of 24 by a MOM
group was followed by the hydrolysis of the oxazolidinone,
affording carboxylic acid 9.
Scheme 6. Synthesis of the Key Synthetic Intermediate 6
and Glycoside 35
The synthesis of alcohol 10 is shown in Scheme 5. (S)-
Glycidol (26) was protected with a PMB group in 92% yield.
Scheme 5. Synthesis of Alcohol 10
The epoxide-opening reaction of 27 with lithium methyl-
acetylide was promoted by BF3·OEt2 to give 28 in 70% yield,
and the resulting hydroxy group of 28 was protected by a TBS
group to obtain alkyne 29. The removal of the PMB group
gave alcohol 30. The Dess-Martin oxidation of 30 was followed
by the addition of methylmagnesium bromide to the aldehyde
to give alcohol 31. The resulting hydroxy group was oxidized
by the Dess-Martin periodinane to afford a ketone. Then, the
desired Z-olefin was produced by the Wittig reaction with ethyl
triphenylphosphonium bromide to give 32. Finally, the TBS
group was deprotected by TBAF to afford the desired alcohol
10. The stereochemistry of 10 was determined by comparing
the specific rotation with a literature value.6
Scheme 7. Scope and Limitation of Ni(0)-Mediated
Reductive Cyclization Reaction
The condensation of 9 with 10 was conducted with 1-(3-
(dimethylamino) propyl)-3-ethylcarbodiimide hydrochloride
(EDCI) and DMAP in CH2Cl2 to afford the desired ester, and
the subsequent deprotection of the TBS group of the ester by
3HF·Et3N gave alcohol 33 in 66% yield in three steps starting
from 25 (Scheme 6). The oxidation of the hydroxy group of 33
was conducted under various conditions to obtain ynal 8,
which is a precursor of the cyclization (see SI). However, the
desired aldehyde was not obtained because of an intermo-
lecular aldol reaction or polymerization via the hydration of the
aldehyde. After extensive optimization, a procedure for
obtaining 8 by Dess-Martin oxidation without an aqueous
workup was discovered. The synthesis route for 8 was
established and applied to the synthesis of various cyclization
precursors, which are described later in Scheme 7.
With the cyclization precursor 8 in hand, the Ni(0)-
mediated reductive cyclization reaction was investigated using
the conditions developed by the group of Montgomery to
promote the endo-selective reductive cyclization.9 Namely, 8
was reacted with bis(1,5-cyclooctadiene)nickel (Ni(cod)2) in
the presence of 1,3-dimesitylimidazolium chloride (IMes·
HCl), potassium tert-butoxide (tBuOK), and triethylsilane to
successfully produce the desired macrolactone 34 in 30% yield
in three steps starting from 33. The macrolactone was obtained
as a single diastereomer, and no regioisomers were obtained.
Therefore, the reaction proceeded stereoselectively as ex-
pected. When other ligands (1,3-bis(2,4,6-trimethylphenyl)-
4,5-dihydroimidazol-2-ylidene (SIMes), 1,3-bis(2,6-
diisopropylphenyl)imidazol-2-ylidene (IPr), and Me3P) and
reductants (iPr3SiH, Ph3SiH and Et3B) were employed no
C
https://dx.doi.org/10.1021/acs.orglett.0c00665
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
pubs.acs.org/OrgLett
Letter
cyclization products were obtained (see Supporting Informa-
tion (SI)).
commercially available simple materials in an overall chemical
yield of 8.1% via the longest linear sequence of 16 steps.
Furthermore, the α-selective glycosylation of 6 and 7
proceeded with high stereoselectivity in efficient yields. Studies
of the total syntheses of natural products and their analogues
using this strategy are ongoing.
To examine the effects of the structure of the cyclization
precursor on the Ni(0)-mediated reductive cyclization
reaction, 36 and 38 with a simplified side chain were
synthesized (Scheme 7). The reductive cyclization reaction
of the aldehyde, which was obtained by the Dess-Martin
oxidation of 36, in the presence of Ni(cod)2, IMes·HCl,
tBuOK, and iPr3SiH in THF was followed by the deprotection
of the TIPS group to afford macrolactone 37 in 14% yield in
three steps. Interestingly, cyclization reaction with the ynal
derived from 38, which has a p-methoxybenzylidene acetal
group, proceeded smoothly to afford 39 in 61% yield as a
single diastereomer.13 Given the importance of the protecting
group, the cyclization reaction of 40, which has the same
protection mode as 38, was examined. Unexpectedly, the
cyclization of the aldehyde derived from 40 gave no products.
The possibility of cyclization is assumed to depend on the
conformation of the precursor ynal, which is affected by the
side chain and the mode of protection. To gain further insight
into the cyclization, 42, which has a different stereochemistry
than 33 at the C5 position; 43, which has a flexible
conformation due to the absence of a cyclic protecting
group; and its stereoisomer 44 were synthesized. The
cyclization reactions of the aldehydes prepared from 42−44
gave no products. These results suggest that the reductive
coupling reaction in the presence of Ni(0) proceeds only in a
specific conformation.
With macrolactone 34 in hand, the stereoinversion of the
hydroxy group at the C7 position of 34 was examined. Initially,
the stereoinversion was conducted by an oxidation−reduction
sequence at the C7 position or an SN2 reaction with various
nucleophiles to mesylate 34; however, the desired inversion
product was not obtained. Fortunately, the Mitsunobu reaction
with p-nitrobenzoic acid, bis(2-methoxyethyl) azodicarboxylate
(DMEAD), and PhOPPh214 in toluene proceeded smoothly to
give a 67% yield. This reaction did not proceed when PPh3 or
Bu3P was used instead of PhOPPh2. The hydrolyzation of the
corresponding p-nitrobenzoate by LiOH in MeOH proceeded
smoothly to afford the desired macrolactone 6. The relative
configuration at the C7 position, which was obtained by the
cyclization and Mitsunobu reactions, was determined from the
NOE correlation between H-7 and H-9 in Scheme 6. To
demonstrate the possibility of the total synthesis, the
glycosylation reaction of 6 was performed. Macrolactone 6
was reacted with trichloroacetimidate 7, which was synthesized
from D-arabinose in four steps,15 in the presence of TMSOTf
and MS4A in CH2Cl2 to produce 35 in 98% yield with high α-
stereoselectivity (α/β > 19:1). Removal of the MOM and/or
isopropyridene groups of 35 was briefly investigated. However,
these protecting groups were resistant to the conditions using
camphorsulfonic acid (MeOH, rt to reflux) or trifluoroacetic
acid (CH2Cl2, rt. to reflux). The use of more harsh conditions
(i.e., TMSOTf, CH2Cl2, 0 °C to rt) gave a complex mixture of
products.
ASSOCIATED CONTENT
* Supporting Information
■
sı
The Supporting Information is available free of charge at
https://pubs.acs.org/doi/10.1021/acs.orglett.0c00665.
Detailed experimental procedures and characterization
of new compounds (PDF)
AUTHOR INFORMATION
Corresponding Author
■
Satoshi Ichikawa − Faculty of Pharmaceutical Science and
Center for Research and Education on Drug Discovery,
Hokkaido University, Sapporo 060-0812, Japan; orcid.org/
0000-0001-5345-5007; Email: ichikawa@
pharm.hokudai.ac.jp
Authors
Shun Kitahata − Faculty of Pharmaceutical Science, Hokkaido
University, Sapporo 060-0812, Japan; orcid.org/0000-
0001-8997-3877
Akira Katsuyama − Faculty of Pharmaceutical Science and
Center for Research and Education on Drug Discovery,
Hokkaido University, Sapporo 060-0812, Japan
Complete contact information is available at:
https://pubs.acs.org/10.1021/acs.orglett.0c00665
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This research was supported in part by JSPS Grant-in-Aid for
Scientific Research (B) (Grant Number 16H05097 to S.I.),
Grant-in Aid for Scientific Research on Innovative Areas
“Frontier Research on Chemical Communications” (No.
18H04599 to S.I.), partly supported by Hokkaido University,
Global Facility Center (GFC), Pharma Science Open Unit
(PSOU), and funded by MEXT under “Support Program for
Implementation of New Equipment Sharing System”, the
Platform Project for Supporting Drug Discovery and Life
Science Research (Basis for Supporting Innovative Drug
Discovery and Life Science Research; BINDS) from the
Japan Agency for Medical Research and Development
(AMED).
REFERENCES
■
(1) (a) Patridge, E.; Gareiss, P.; Kinch, M. S.; Hoyer, D. Drug
Discovery Today 2016, 21, 204−207. (b) Spellberg, B.; Gilbert, D. N.
Clin. Infect. Dis. 2014, 59, 71−75. (c) Ventola, C. L. J. Formul. Manag.
2015, 40, 277−283.
In conclusion, to develop a new synthesis route for the novel
macrolide glycosides 1−4, synthesis studies of the key
synthetic intermediate 6 were performed. The key to the
preparation of the 12-membered macrolactone was a Ni(0)-
mediated reductive coupling reaction using easily accessible
fragments. In addition, various cyclization precursors were
synthesized to study the scope and limitation of the substrate.
The key synthetic intermediate could be obtained from
(2) CDC’s Antibiotic Resistance Threats in the United States, 2019
(2019 AR Threats Report), Available online: The full 2019 AR
Threats Report, including methods and appendices, is available online
at www.cdc.gov/DrugResistance/Biggest-Threats.html.
̈
(3) Schieferdecker, S.; Konig, S.; Weigel, C.; Dahse, H. M.; Werz,
O.; Nett, M. Chem. - Eur. J. 2014, 20, 15933−15940.
D
https://dx.doi.org/10.1021/acs.orglett.0c00665
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
pubs.acs.org/OrgLett
Letter
(4) Surup, F.; Viehrig, K.; Mohr, K. I.; Herrmann, J.; Jansen, R.;
Muller, R. Angew. Chem., Int. Ed. 2014, 53, 13588−13591.
(5) Surup, F.; Steinmetz, H.; Mohr, K.; Viehrig, K.; Muller, R.; Nett,
̈
̈
M.; Schieferdecker, S.; Dahse, H. M.; Wolling, M.; Kirchning, A. WO
2016/005049 A1.
(6) Wolling, M.; Kirschning, A. Eur. J. Org. Chem. 2018, 2018, 648−
656.
(7) Kwon, Y.; Schulthoff, S.; Dao, Q. M.; Wirtz, C.; Fu
Chem. - Eur. J. 2018, 24, 109−114.
̈
rstner, A.
(8) Rengarasu, R.; Maier, M. E. Synlett 2019, 30, 1346−1350.
(9) (a) Mahandru, G. M.; Liu, G.; Montgomery, J. J. Am. Chem. Soc.
2004, 126, 3698−3699. (b) Sa-ei, K.; Montgomery, J. Org. Lett. 2006,
8, 4441−4443. (c) Chaulagain, M. R.; Sormunen, G.; Montgomery, J.
J. Am. Chem. Soc. 2007, 129, 9568−9569. (d) Malik, H. A.;
Chaulagain, M.; Montgomery, J. Org. Lett. 2009, 11, 5734−5737.
(e) Malik, H. A.; Sormunen, G. J.; Montgomery, J. J. Am. Chem. Soc.
2010, 132, 6304−6305. (f) Jackson, E. P.; Montgomery, J. J. Am.
Chem. Soc. 2015, 137, 958−963. (g) Jackson, E. P.; Malik, H. A.;
Sormunen, G. J.; Baxter, R. D.; Liu, P.; Wang, H.; Shareef, A. R.;
Montgomery, J. Acc. Chem. Res. 2015, 48, 1736−1745. (h) Wang, H.;
Lu, G.; Sormunen, G. J.; Malik, H. A.; Liu, P.; Montgomery, J. J. Am.
Chem. Soc. 2017, 139, 9317−9324.
(10) (a) Knapp-Reed, B.; Mahandru, M. G.; Montgomery, J. J. Am.
Chem. Soc. 2005, 127, 13156−13157. (b) Shareef, A. R.; Sherman, D.
H.; Montgomery, J. Chem. Sci. 2012, 3, 892−895. (c) Wang, H.;
Negretti, S. N.; Knauff, A. R.; Montgomery, J. Org. Lett. 2015, 17,
1493−1496. (d) Chrovian, C. C.; Knapp-Reed, B. K.; Montgomery, J.
Org. Lett. 2008, 10, 811−814. (e) Chan, J.; Jamison, T. F. J. Am.
Chem. Soc. 2004, 126, 10682−10691. (f) Gilbert, M. M.; DeMars, M.
D., II; Yang, S.; Grandner, J. M.; Wang, S.; Wang, H.; Narayan, A. R.;
Sherman, D. H.; Houk, K. N.; Montgomery, J. ACS Cent. Sci. 2017, 3,
1304−1310.
(11) Yu, W.; Mei, Y.; Kang, Y.; Hua, Z.; Jin, Z. Org. Lett. 2004, 6,
3217−3219.
(12) Crimmins, M. T.; She, J. Synlett 2004, 8, 1371−1374.
(13) Compound 39 was observed as a mixture of two conformers by
1
variable-temperature H NMR spectra (see SI).
(14) Mukaiyama, T.; Kuroda, K.; Maruyama, Y. Heterocycles 2010,
80, 63−82.
̀
(15) (a) Dureau, R.; Legentil, L.; Daniellou, R.; Ferrieres, V. J. Org.
Chem. 2012, 77, 1301−1307. (b) Gurjar, M. K.; Reddy, L. K.; Hotha,
S. J. Org. Chem. 2001, 66, 4657−4660. (c) Gurjar, M. K.; Reddy, L.
K.; Hotha, S. J. Org. Chem. 2001, 66, 4657−4660.
E
https://dx.doi.org/10.1021/acs.orglett.0c00665
Org. Lett. XXXX, XXX, XXX−XXX