Cystobactamids 920-1 and 920-2: Assignment of the Constitution and Relative Configuration by Total Synthesis

Article abstract of DOI:10.1021/acs.orglett.9b00058

Total synthesis of cystobactamid 920-1 and its epimer has allowed an unambiguous assignment of the relative and absolute configuration of the natural product. A careful structural analysis of each isomer using both NMR and computational techniques also pr

Full text of DOI:10.1021/acs.orglett.9b00058

Letter
pubs.acs.org/OrgLett
Cite This: Org. Lett. XXXX, XXX, XXX−XXX
Cystobactamids 920‑1 and 920-2: Assignment of the Constitution
and Relative Configuration by Total Synthesis
‡
†
†
Therese Planke,†,§ María Moreno,††,§ Stephan Huttel, † Jorg Fohrer,‡ Franziska Gille,
̈
̈
†
,†
̈
Matthew D. Norris, Maik Siebke, Liangliang Wang, Rolf Muller, and Andreas Kirschning*
†Institut fu
1B, 30167 Hannover, Germany
‡Abteilung Mikrobielle Naturstoffe, Helmholtz Institut fu
̈
r Organische Chemie und Biomolekulares Wirkstoffzentrum (BMWZ) der Leibniz Universitat Hannover, Schneiderberg
̈
̈
r Pharmazeutische Forschung Saarland, Helmholtz Zentrum fu
̈
r
̈
̈
Infektionsforschung und Universitat des Saarlandes, Campus E8.1, 66123 Saarbrucken, Germany
S
* Supporting Information
ABSTRACT: Total synthesis of cystobactamid 920-1 and its epimer
has allowed an unambiguous assignment of the relative and absolute
configuration of the natural product. A careful structural analysis of
each isomer using both NMR and computational techniques also
prompted a constitutional revision of the structures originally
reported for cystobactamids 920-1 and 920-2, and has provided
further insight into the unique conformational preferences of the
cystobactamid family.
were determined unequivocally by Su
̈
ssmuth and co-workers
n 2014, a group of nonribosomal peptides, cystobactamids
through total synthesis.2 Only recently, nine new cystobacta-
mid derivatives were isolated from Cystobacter sp., including
cystobactamids 920-1 (3), 920-2 (4), and 861-2 (5), the latter
being the most potent antibiotic of all known congeners.3
Indeed, the cystobactamids and albicidin show strong
antibacterial activity, inhibiting several clinically relevant
Gram-positive and Gram-negative strains, such as Acinetobacter
baumannii (MIC = 0.5 μg/mL for 5), Citrobacter freundii (MIC
= 0.06 μg/mL for 5), carbapenem-resistant E. coli WT-III
(marRΔ74bp) (MIC = 0.5 μg/mL for 5), carbapenem-
resistant P. aeruginosa (CRE) (MIC = 1.0 μg/mL for 5), and
Proteus vulgaris (MIC = 0.25 μg/mL for 5).3 It was also found
that the cystobactamids inhibit bacterial type IIa top-
oisomerases, which are known to be targeted by the clinically
established group of quinolone antibiotics.
I919-1 (1) and 919-2 (2), were isolated in small amounts
(<100 μg/L) through cultivation of the myxobacterium
Cystobacter sp. (Figure 1).1 Compounds 1 and 2 are closely
related to albicidin, whose structure and absolute configuration
Total syntheses of the cystobactamids 861-2 (5) and 919-2
(4) were accomplished independently through our earlier
efforts3 and that of the Trauner group,4 respectively. However,
correct assignment of the two stereogenic centers in the central
methoxyaspartate “hinge” region has been a matter of debate
since the original isolation paper.1 The relative configuration of
1 and 2 was originally interpreted as anti (relative to the
orientation drawn in Figure 1; either 2S,3S or 2R,3R
configuration) through comparing the homonuclear vicinal
Figure 1. Cystobactamid natural products, synthetic epimer 3a and
the structural revision of cystobactamids 920-1 and 920-2 as
presented in this work (the related peptide albicidin is also shown).
Received: January 6, 2019
© XXXX American Chemical Society
A
DOI: 10.1021/acs.orglett.9b00058
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
coupling constants (3JH,H = 7−8 Hz) and observed ROESY-
NMR data to that of other β-oxyasparagine derivatives
reported in literature. The absolute configuration of each
stereocenter was similarly inferred through direct comparison
of the measured optical rotation of 1 and 2 with literature
compounds. Consequently, the configuration at C2 and C3
was assigned as 2S,3S.1
Scheme 1. Preparation of the Two Epimeric
Methoxyaspartates
Later, Kim and co-workers also isolated cystobactamid 919-2
(2) along with two other derivatives, which they named
coralmycins A and B, from cultures of Corallococcus coralloides
myxobacteria.5 In contrast to the original stereochemical
assignment, Kim proposed that the configuration of the
methoxyaspartate hinge region of 2 should be revised to 2S,3R.
This alternative structural assessment also resulted from
detailed NMR analysis of the natural compound, especially
3
n
concerning the values of JHH and JCH. Kim argued that the
nJC3,2‑H value of 6.7 Hz (in DMSO-d6/CD3OD = 4:1) is
consistent with a gauche conformation of 2-H and the methoxy
group at C3. Additionally, it was contested that 2-H and C4, as
well as 3-H and C1, are gauche oriented, implying that the
methoxyaspartate hinge is fixed with a syn configuration
(relative to the orientation drawn in Figure 1; either 2S,3R or
2R,3S configuration). Despite these differences in interpreta-
tion, it is notable that in both accounts the chemical shift
Synthesis of the 2S,3R-diastereomer 12 was accomplished
through extension of our existing route to azide 11 from
methyl cinnamate (6).3,9 Saponification and tert-butyl ester
formation paved the way for Kuhn−Roth oxidation, and the
resulting carboxylate was advanced to amine 12 through O-
methylation and Staudinger reduction (Scheme 1, part B).
With both epimers in hand, we proceeded to couple the
eastern and western polyaromatic fragments in a stepwise
approach. Best results for the first coupling reaction were
realized when acid 1310 was activated with ethyl chlorocar-
bonate and then added to freshly prepared methoxyaspartate
amines 10 or 12 in DMF (Schemes 2 and 3). Under these
3
values and coupling constant J2‑H,3‑H of compound 2 were
identical.
In view of this ongoing stereochemical uncertainty, we
endeavored to synthesize both possible epimers of the reported
structure of cystobactamid 920-2, 3a and 3b and to perform a
detailed conformational analysis of each isomer, using NMR
techniques and computer-aided molecular modeling, to
unambiguously assign the relative and absolute configuration
of the natural cystobactamid framework. To date, neither the
synthesis of a carboxylic acid bearing cystobactamid nor that of
any epimeric cystobactamid pair have been reported.
Following our previous work on the synthesis of
cystobactamid 861-2 (5),3,6 we sought to construct both
epimers of the reported structure of cystobactamid 920-2 using
a similar approach. Thus, the synthesis of both methoxyas-
partate epimers, 10 and 12, began with methyl cinnamate (6)
(Scheme 1). Similar to our earlier strategy, we envisaged that
the phenyl ring in 6 would ultimately act as an effective masked
carboxyl group, which may be unveiled at a later stage through
chemoselective Kuhn−Roth oxidation.7
Scheme 2. Preparation of 2S,3S-Isomer 3a
For synthesis of the 2S,3S-epimer 10, α-acetoxy-β-bromo
methylester 7 was first prepared from 6 following a known
procedure,8 which began with Sharpless asymmetric dihydrox-
ylation as the key step (Scheme 1, part A). Nucleophilic
substitution of the benzylic bromide with NaN3 yielded ester 8
with inversion of configuration, and saponification followed by
tert-butyl ester formation delivered azide 9. Notably, the latter
transformation was sensitive to azido elimination under acidic
conditions, necessitating use of the milder reagent Me2N−
CH(OtBu)2. Under optimized conditions, we were pleased
that this ester switch proceeded in 87% yield over two steps.
With tert-butyl ester 9 in hand, selective oxidation of the
pendent phenyl ring to a carboxylic acid with RuCl3 and NaIO4
allowed us to simultaneously introduce the second carboxyl
group of the methoxyaspartate linker and to maintain its
differentiation from the existing carboxyl moiety. This rarely
employed transformation proceeded with remarkable chemo-
selectivity and yielded amine 10 after O-methylation and
Staudinger reduction of the azido group.
conditions, amides 14 and 17 were accessed in good yield from
amines 10 and 12, respectively, after tert-butyl ester
elimination. Compound 14 was then coupled to the known
polyaromatic aniline fragment 153,6 after activation with
PPh3Cl2 to give amide 16. However, under these same
conditions, we were surprised to find that the epimeric acid 17
B
DOI: 10.1021/acs.orglett.9b00058
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Scheme 3. Preparation of 2S,3R-Isomer 3b
reacted exclusively at the phenol oxygen of 15. In order to
circumvent this issue, we had to employ the known allyl
protected aniline fragment 183,6 instead. Thus, activation of 17
with POCl3 in the presence of DIPEA and subsequent addition
of 18 afforded the desired amide 19 in 59% yield.
1
Figure 2. Comparison of the H NMR regions of (A) synthetic 3a
(blue) and (B) synthetic 3b (blue), with natural cystobactamids 920-
1 (red) and 920-2 (green). Left region: NH proton. Right region:
CHOMe proton (marked gray in the above structure).
Subsequent hydrolysis of the aspartate methyl ester proved
to be challenging. Under conditions typical for base hydrolysis,
the amino acid linker was highly sensitive to epimerization,
resulting in a mixture of isomers that were difficult to separate
on a preparative scale. We found that the exceptionally mild
reagent Me3SnOH was best suited to affect methyl ester
hydrolysis, although partial epimerization could not be
completely suppressed. After some optimization, methyl ester
16 was successfully hydrolyzed with Me3SnOH in 1,2-
dichloroethane at 80 °C, and tert-butyl ester elimination
afforded the 2S,3S-isomer of the originally assigned structure of
cystobactamid 920-2 (3a).
4.31* (d, J = 5.0 Hz, 1H−CH) (see atoms marked in gray in
Figure 2). From this we ultimately concluded that synthetic 3a
is a diastereomer of natural cystobactamid 920-1. This was
1
confirmed when comparing the H NMR spectra of synthetic
3b with natural samples of cystobactamids 920-1 and 920-2
(Figure 2, part B). Again, the amide proton and CHOMe
group in the methoxyaspartate hinge region were diagnostic
elements. Clearly, isomer 3b matches the authentic cystobac-
tamid 920-1 and not its constitutional isomer cystobactamid
920-2, as it was originally proposed. Our side-by-side
comparison confirms that (a) the relative configuration of
the methoxyaspartate unit in both cystobactamids 920-1 and
920-2 is 2S,3R and (b) the constitutional assignment of each
natural product must be reassigned to the structures depicted
in Figure 1.
Compound 19 was similarly advanced to the 2S,3R-isomer
of the originally assigned structure of cystobactamid 920-2
(3b), although for purification purposes, it was advantageous
to perform methyl ester hydrolysis in the final step. Applying
stoichiometric PhSiH3 as a scavenger in the palladium
catalyzed deallylation of 19 was troublesome, leading to partial
reduction of the aryl nitro group. Rather, aniline was found to
be superior, resulting in a clean transformation to the desired
product on treatment with Pd(PPh3)4. Curiously, hydrolysis of
the remaining ester was also complicated by inadvertent
alkylation of 1,2-dichloroethane (solvent) to the exposed
phenol group under our previously optimized conditions,
necessitating the use of toluene as solvent in this reaction.
Owing to its poor solubility, the final product 3b was isolated
in pure form using a combination of different chromatographic
techniques (see Supporting Information).
To better understand the difficulty of structural elucidation
in this family of compounds, we explored a conformational
̈
analysis of cystobactamid 920-1 (3) using the Schrodinger
suite12 maestro 9.8 computational model with the mixed
torsional/low-mode sampling conformational search (LMCS)
method and force field OPLS2005.13 Water was chosen as
solvent, and energy potentials were optimized for simulation in
solution (OPLS2005) to improve the quality of force field
parameters.14,15 We expected an out-of-plane orientation of the
aromatic rings of 3 that might also accommodate a helical
conformation. Indeed, our calculations showed that the
aromatic rings of cystobactamid 920-1 (3) are not likely to
preferentially be oriented in-plane. The calculations also
revealed that the helical shape is stabilized through intra-
molecular hydrogen bonds between an NH and the phenolic
group of the adjacent chain and another NH group of ring e
with the isopropoxy group of ring b (Figure 3; for numbering,
see Figure 1).
1
Comparison of the H NMR spectra of natural cystobacta-
mids 920-1 and 920-2 with that of synthetic 3a revealed
significant differences (Figure 2, part A). This included the
#
chemical shift values (δ; designated with for natural 920-1
and * for natural 920-2) of one N−H amide proton, 8.34 (d, J
= 8.9 Hz, 1H−NH) compared to 8.72# (d, J = 8.67 Hz, 1H−NH
)
and 8.47* (d, J = 8.1 Hz, 1H−NH), and one of the C−H
protons in the methoxyaspartate hinge region, 4.48 (d, J = 4.6
Hz, 1H−CH) compared to 4.13# (d, J = 8.05 Hz, 1H−CH) and
Calculating the ten lowest energy conformers of cystobacta-
mid 920-1 (3), we discovered that the terminal aminoarene
unit of the eastern fragment has greater conformational
C
DOI: 10.1021/acs.orglett.9b00058
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Accession Codes
CCDC 1553781 contains the supplementary crystallographic
data for this paper. These data can be obtained free of charge
via www.ccdc.cam.ac.uk/data_request/cif, or by emailing
data_request@ccdc.cam.ac.uk, or by contacting The Cam-
bridge Crystallographic Data Centre, 12 Union Road,
Cambridge CB2 1EZ, UK; fax: +44 1223 336033.
AUTHOR INFORMATION
Corresponding Author
■
*E-mail: andreas.kirschning@oci.uni-hannover.de.
ORCID
Andreas Kirschning: 0000-0001-5431-6930
Author Contributions
Figure 3. (A) Lowest energy conformation (lowest energy conformer
on the left and higher energy conformer on the right) of
cystobactamid 920-1 (3) in the initially published 2S,3S configuration.
The calculations show that both aromatic chains stack on top of each
other, forming a loop around the hinge region. This loop is observed
in calculations for all possible combinations of absolute configuration
of the two stereocenters. (B) Magnification of the hinge region with
the two stereogenic centers and (C) observed ROE of the natural
compounds.
§T.P. and M.M. contributed equally.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This work was supported by the German Center for Infection
Research (DZIF) and the Alexander von Humboldt
Foundation (scholarship for M.D.N.). We thank A. Kanakis
flexibility compared to the benzene rings of the western part.
This observation is consistent with the work of Wilson and co-
workers, who found similarly enhanced conformational
flexibility in the terminal rings of oligobenzamides.16
Furthermore, the western peptide unit of cystobactamid 920-
1 rotates backward to form a loop around the hinge region,
such that both aromatic hemispheres are stacked on top of
each other. This loop appears to be stabilized by intra-
molecular hydrogen bonds and hydrophobic interactions
between the two chains (Figure 3). It was also possible to
observe distant through-space interactions of the two
polyaromatic segments using ROE spectroscopy. It is clear
from the ROE spectrum of the natural compound that the
amide proton at 9.58 ppm interacts with phenyl protons of the
̈
(Leibniz Universitat Hannover) for helpful support in
computational calculations.
REFERENCES
■
(1) Baumann, S.; Herrmann, J.; Raju, R.; Steinmetz, H.; Mohr, K. I.;
ttel, S.; Harmrolfs, K.; Stadler, M.; Mu
Ed. 2014, 53, 14605.
Hu
̈
̈
ller, R. Angew. Chem., Int.
(2) (a) von Eckardstein, L.; Petras, D.; Dang, T.; Cociancich, S.;
̈
Sabri, S.; Gratz, S.; Kerwat, D.; Seidel, M.; Pesic, A.; Dorrestein, P. C.;
Royer, M.; Weston, J. B.; Sussmuth, R. D. Chem. - Eur. J. 2017, 23,
̈
̈
15316. (b) Kerwat, D.; Gratz, S.; Kretz, J.; von Eckardstein, L.; Seidel,
M.; Kunert, M.; Weston, J. B.; Sussmuth, R. D. ChemMedChem 2016,
11, 1899. (c) Gratz, S.; Kerwat, D.; Kretz, J.; von Eckardstein, L.;
Semsary, S.; Seidel, M.; Kunert, M.; Weston, J. B.; Sussmuth, R. D.
ChemMedChem 2016, 11, 1499. (d) Petras, D.; Kerwat, D.; Pesic, A.;
̈
̈
̈
̀
ring vis-a-vis that resonate at 7.89 ppm as shown in Figure 3,
́
Hempel, B.; von Eckardstein, L.; Semsary, S.; Araste, J.; Marguerettaz,
M.; Royer, M.; Cociancich, S.; Sussmuth, R. D. ACS Chem. Biol. 2016,
11, 1198. (e) Vieweg, L.; Kretz, J.; Pesic, A.; Kerwat, D.; Graetz, S.;
Royer, M.; Cociancich, S.; Mainz, A.; Sussmuth, R. D. J. Am. Chem.
Soc. 2015, 137, 7608. (f) Kretz, J.; Kerwat, D.; Schubert, V.; Gratz, S.;
Pesic, A.; Semsary, S.; Cociancich, S.; Royer, M.; Sussmuth, R. D.
Angew. Chem. 2015, 127, 1992. (g) Pieretti, I.; Royer, M.; Cociancich,
S.; Pesic, A.; Petras, D.; Sussmuth, R. D. Front. Plant Sci. 2015, 6, 289.
part C. It must be noted that for geometrical reasons a ROE
between this amide proton and the protons of the adjacent
aromatic is not possible.
̈
̈
In summary, total synthesis of cystobactamid 920-1 (3a) and
its epimer (3b) allowed an unambiguous assignment of the
relative and absolute configuration of the central methox-
yaspartate unit of cystobactamid 920-1 and prompted revision
of the original constitutional assignment of cystobactamids
920-1 and 920-2. A careful structural analysis using both NMR
and computational techniques provided insight into the unique
helical backflip conformation of the cystobactamids, which
exhibit hydrogen bond interactions between the two
polyaromatic arms. This strong out-of-plane preference results
in an altered conformation of the methoxyaspartate hinge
region and has previously led to misconception in the
structural assessment of this family of natural products.
̈
̈
̈
(h) Cociancich, S.; Pesic, D.; Petras, D.; Uhlmann, S.; Kretz, J.;
Schubert, V.; Vieweg, L.; Duplan, S.; Marguerettaz, S.; Noell, J.;
Pieretti, I.; Hu
Sussmuth, R. D. Nat. Chem. Biol. 2015, 11, 195.
(3) Huttel, S.; Testolin, G.; Herrmann, J.; Planke, T.; Gille, F.;
̈
gelland, M.; Kemper, S.; Mainz, A.; Rott, P.; Royer, M.;
̈
̈
̈
̈
Moreno, M.; Stadler, M.; Bronstrup, M.; Kirschning, A.; Muller, R.
Angew. Chem., Int. Ed. 2017, 56, 12760.
̈
(4) Cheng, B.; Muller, R.; Trauner, D. Angew. Chem., Int. Ed. 2017,
56, 12755.
(5) Kim, Y. J.; Kim, H. J.; Kim, G. W.; Cho, K.; Takahashi, S.;
Koshino, H.; Kim, W. G. J. Nat. Prod. 2016, 79, 2223.
(6) Moreno, M.; Elgaher, W. A. M.; Herrmann, J.; Schlager, N.;
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.or-
glett.9b00058.
■
̈
S
Hamed, M. M.; Baumann, S.; Mu
Kirschning, A. Synlett 2015, 26, 1175.
̈
ller, R.; Hartmann, R. W.;
(7) Mander, L. N.; Williams, C. M. Tetrahedron 2003, 59, 1105.
(8) (a) Deng, J.; Hamada, Y.; Shiori, T. Synthesis 1998, 1998, 627.
(b) Deng, J.; Hamada, Y.; Shioiri, T. J. Am. Chem. Soc. 1995, 117,
7824.
Detailed experimental procedures and spectral data
(PDF)
(9) Sayyed, I.; Sudalai, A. Tetrahedron: Asymmetry 2004, 15, 3111.
D
DOI: 10.1021/acs.orglett.9b00058
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
(10) Abbel, R.; Frey, H.; Schollmeyer, D.; Kilbinger, A. Chem. - Eur.
J. 2005, 11, 2170.
(11) Nicolaou, K. C.; Estrada, A. A.; Zak, M.; Lee, S. H.; Safina, B. S.
Angew. Chem., Int. Ed. 2005, 44, 1378.
̈
(12) MacroModel, version 9.8, Schrodinger, LLC: New York, NY,
2011.
(13) (a) Banks, J. L.; Beard, H. S.; Cao, Y.; Cho, A.; Damm, W.;
Farid, R.; Felts, A. K.; Halgren, T. A.; Mainz, D. T.; Maple, J. R.;
Murphy, R.; Philipp, D. M.; Repasky, M. P.; Zhang, L. Y.; Berne, B. J.;
Friesner, R. A.; Gallicchio, E.; Levy, R. M. J. Comput. Chem. 2005, 26,
1752. (b) Jorgensen, W. L.; Tirado-Rives, J. J. Am. Chem. Soc. 1988,
110, 1657.
(14) (a) DuBay, K. H.; Hall, M. L.; Hughes, T. F.; Wu, C.;
Reichman, D. R.; Friesner, R. A. J. Chem. Theory Comput. 2012, 8,
4556. (b) Tedesco, D.; Zanasi, R.; Guerrini, A.; Bertucci, C. Chirality
2012, 24, 741. (c) Stortz, C. A.; Johnson, G. P.; French, A. D.;
Csonka, G. I. Carbohydr. Res. 2009, 344, 2217.
(15) This force field is ideally suited for the conformational analysis
of peptides. See: (a) Hendricks, N. G.; Julian, R. R. Chem. Commun.
2015, 51, 12720. (b) Matthes, D.; de Groot, B. L. Biophys. J. 2009, 97,
599.
(16) (a) Plante, J. P.; Burnley, T.; Malkova, B.; Webb, M. E.;
Warriner, S. L.; Edwards, T. A.; Wilson, A. Chem. Commun. 2009,
5091. (b) Fuller, J. C.; Jackson, R. M.; Edwards, T. A.; Wilson, A. J.;
Shirts, M. A. PLoS One 2012, 7, No. e43253. (c) Plante, J.; Campbell,
F.; Malkova, B.; Kilner, C.; Warriner, S. L.; Wilson, A. Org. Biomol.
Chem. 2008, 6, 138.
E
DOI: 10.1021/acs.orglett.9b00058
Org. Lett. XXXX, XXX, XXX−XXX