Synthesis of the Natural Product Marthiapeptide A

Article abstract of DOI:10.1021/acs.orglett.5b02574

The first total synthesis of marthiapeptide A is reported. Two synthetic procedures are described: the first, which was unsuccessful, attempts to close the ring at position I, and the second, which was successful, closes the ring at position II. It appear

Full text of DOI:10.1021/acs.orglett.5b02574

Letter
pubs.acs.org/OrgLett
Synthesis of the Natural Product Marthiapeptide A
Yuqi Zhang,† Md. Amirul Islam,† and Shelli R. McAlpine*,†
†School of Chemistry, The University of New South Wales, Gate 2 High Street, Sydney 2052, Australia
S
* Supporting Information
ABSTRACT: The first total synthesis of marthiapeptide A is reported. Two synthetic procedures are described: the first, which
was unsuccessful, attempts to close the ring at position I, and the second, which was successful, closes the ring at position II. It
appears that the first route was unsuccessful because it required cyclization next to the rigid thiazole moiety, whereas the second
route closed next to the more flexible thiazoline ring.
inked azoles are unique substructures that are present in
many natural products. Over the past two decades, there
has been a surge in the discovery of natural product compounds
that contain linked azoles, where many of these compounds are
now candidates for drug development.1 Urukthapelstatin A
(Ustat A) (Figure 1) is a natural product isolated from marine
MTA was isolated by Zhou and co-workers in 2012.6 It is a
cyclic peptide that consists of three consecutive thiazoles and
thiazoline and was isolated from a south China sea derived
strain Marinactinospora thermotolerans. This natural product
exhibits significant cytotoxicity against a panel of human cancer
cell lines with GI50 value ranges from 0.38 to 0.52 μM. D-
Alanine, L-isoleucine, and D-phenylalanine connected to a
trithiazole thiazoline system makes up the complex structure
shown in Figure 1. Outlined is our unsuccessful initial route as
well as the successful final route for completing the synthesis of
MTA.
L
Our initial strategy (A) involved a macrocyclization of the
linear precursor via a peptide coupling reaction between the
amine on the alanine residue and the carboxylic acid end of
isoleucine (position I, Scheme 1). However, the cyclization was
not successful, which was attributed to the closing point being
too close to the rigid heterocyclic thiazole moiety. Our second
strategy (B) involved closing between the thiazoline and
peptide (position II Scheme 1). This approach worked, and
although it was not high yielding, we believe that this successful
cyclization can be attributed to the flexibility of the thiazoline,
which allows a connection between the molecule’s head and
tail.
Cyclization for the initial strategy involved the synthesis of 2
(dipeptide) and 3 (trithiazole ester) and coupling at position II
to give the linear molecule shown in route A. Cyclization then
occurred at position I. The successful strategy involved
coupling 2 and 3 at position I first and then cyclizing at
position II. Generation of 3 was accomplished by sequentially
Figure 1. Structures of Ustat A, HXDV, and MTA (1).
bacteria, and it has a bisoxazole and a bisthiazole moiety located
within the macrocycle. It shows potent anticancer activity
against a panel of human cancer cell lines, with an average IC50
value of 12 nM.2,3 HXDV (Figure 1) is a synthetic derivative of
telomestatin, and it has two linked trioxazoles within its
macrocyclic backbone. HXDV exhibits antiproliferative and
apoptotic activity by stabilizing the G-quadruplex of DNA that
is formed during replication.4,5 Marthiapeptide A (MTA)
(Figure 1) is another potent natural product that contains a
linked trithiazole−thiazoline system and cytotoxicity with an
IC50 = ∼380−520 nM against a panel of cancer cell lines.6 We
previously reported the first synthesis of Ustat A3 and the
synthesis of trioxazole fragments of HXDV.7 Herein we report
the synthesis of MTA.
Received: September 6, 2015
© XXXX American Chemical Society
A
DOI: 10.1021/acs.orglett.5b02574
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Scheme 1. Retrosynthetic Scheme for Synthesizing
Marthiapeptide
Scheme 2. Initial Synthetic Route (A) Was Unsuccessful at
Producing the Desired Natural Product 1
a
a
Ring closing was successful only at position II via route B.
synthesizing thiazoles using a modified Hantzsch synthesis,
which condensed ethyl 2-bromopyruvate and a thioamide. The
synthesis of dipeptide 2 was accomplished via standard peptide-
coupling conditions between Boc-NH-D-Phe-OH carboxylic
acid with H2N-Ile-OMe amine.
We hypothesized that the ring failed to close at position I
because the ring-closure site was too close to the rigid thiazole
moieties. The proximity of the four relatively rigid structures to
the ring-closing site likely made it impossible to connect the
head and tail end of the compound.
The synthesis of 3 was completed using our previously
reported procedure.7 Specifically, BocHN-D-Ala-OH underwent
12 steps, via sequential thiazole Hantzsch synthesis, to generate
3 in an overall yield of 20%. The first strategy started from 3,
where the trithiazole ester was sonicated with a solution of
ammonium hydroxide (Scheme 2) generating amide 4.
Dehydration of the amide to the nitrile using trifluoroacetic
anhydride (TFAA) in the presence of N,N-diisopropylethyl-
amine (DIPEA) and CH2Cl2 produced the nitrile 5. The crude
material underwent flash column chromatography to afford
pure 5 in an 80% yield. Condensation of the nitrile with L-
cysteine was then carried out using NaHCO3 in the presence of
phosphate buffer (pH = 5.95) at 70 °C to furnish thiazoline 6.
Amine deprotection of 2 was achieved using trifluoroacetic acid
(TFA) generating 7. Coupling crude 7 to free acid thiazoline 6
using coupling agent 1-bis(dimethylamino)methylene]-1H-
1,2,3-triazolo[4,5-b]pyridinium 3-oxide hexafluorophosphate
(HATU) in the presence of DIPEA generated 8. Ester
hydrolysis of 8 produced crude free acid 9, where the crude
was treated with TFA to remove the Boc protecting group
generating 10. Subsequent cyclization of the double depro-
tected linear precursor 10 using HATU and 4-(4,6-dimethoxy-
1,3,5-triazin-2-yl)-4-methylmorpholinium chloride (DMTMM)
in the presence of DIPEA failed to generate compound 1.
Similar to the initial route, the final route (B) was also
initiated starting with 3, whereupon Boc removal produced 11
(Scheme 3). Conversion of 2 to the free acid using LiOH·H2O
produced 12 in preparation for coupling to 11. Using HATU in
the presence of DIPEA, 11 was coupled to 12 generating 13 in
a 72% yield. Subsequent conversion to the amide 14 using
ammonium hydroxide was accomplished using sonication over
48 h (Scheme 3). The larger structure 13 converted more
slowly to the amide 14 than the smaller structure (3 to amide 4,
Scheme 2). Dehydration of the amide 14 to the nitrile 15 was
produced in a lower yield than the corresponding dehydration
of the smaller molecule used in the initial route (Scheme 2,
structure 4 to 5). Indeed, the conditions for nitrile 15
formation required modification, whereupon the optimal yield
was produced when using molecular sieves. Condensing L-
cysteine with 15 in the presence of sodium bicarbonate in a
phosphate buffer generated linear precursor 16. Amine
deprotection using TFA produced the linear precursor.
Cyclization was accomplished using a syringe pump, which
added the compound into a cocktail of three coupling agents:
HATU, DMTMM, and O-(benzotriazol-1-yl)-N,N,N′,N′-tetra-
methyluronium tetrafluoroborate (TBTU) in the presence of
DIPEA.
B
DOI: 10.1021/acs.orglett.5b02574
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
isoleucine residue because this is likely the most flexible
position around the macrocyclic backbone.
Scheme 3. Final Synthetic Route (B) Successfully Produced
the Desired Natural Product 1
ASSOCIATED CONTENT
* Supporting Information
■
S
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.or-
glett.5b02574.
Synthetic and biological procedures, characterization
data, and biological data (PDF)
AUTHOR INFORMATION
Corresponding Author
■
*E-mail: s.mcalpine@unsw.edu.au.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This research was funded by an NHMRC grant (No.
APP1043561) which provided a Ph.D. scholarship to A.I. We
also thank the University of New South Wales for a tuition fee
waiver for A.I. We acknowledge the kind donation of the
natural product from Dr. Jianhua Ju, Huang Hongbo, and co-
workers,6 South China Sea Institute of Oceanology, Chinese
Academy of Sciences.
REFERENCES
(1) (a) Riego, E.; Hernan
■
́
́
dez, D.; Albericio, F.; Alvarez, M. Synthesis
2005, 2005, 1907−1922. (b) Baumann, M.; Baxendale, I. R.; Ley, S.
V.; Nikbin, N. Beilstein J. Org. Chem. 2011, 7, 442−495. (c) Duchler,
M. J. Drug Target. 2012, 20, 389−400. (d) Boger, D. L.; Lee, J. K. J.
Org. Chem. 2000, 65, 5996−6000. (e) Dang, Q.; Liu, Y.; Cashion, D.
K.; Kasibhatla, S. R.; Jiang, T.; Taplin, F.; Jacintho, J. D.; Li, H.; Sun,
Z.; Fan, Y.; DaRe, J.; Tian, F.; Li, W.; Gibson, T.; Lemus, R.; van
Poelje, P. D.; Potter, S. C.; Erion, M. D. J. Med. Chem. 2011, 54, 153−
165. (f) Rzuczek, S. G.; Pilch, D. S.; Liu, A.; LaVoie, E. J.; Rice, J. E. J.
Med. Chem. 2010, 53, 3632−3644.
Upon disappearance of the starting material, as observed
when monitoring by LCMS, the reaction was washed with an
acid and then base in order to remove coupling agent side
products. The resulting crude material was purified via flash
column chromatography on silica gel. The relatively product
was isolated from the column, and a second purification was
done using reverse phase HPLC. Final purity was verified by
evaluating the compound using two different conditions on the
LCMS (Supporting Information) and co-injecting it with the
natural product kindly provided by the original isolation team6
using these two systems. The LCMS coinjection data show that
the synthetic product was identical in retention time and mass
to the natural product. The HRMS also indicated that our
sample was pure, where the calculated exact mass and observed
mass were 688.1269 and 688.1260 respectively, (Supporting
(2) (a) Matsuo, y.; Kanoh, K.; Imagawa, H.; Adachi, K.; Nishizawa,
M.; Shizuri, Y. J. Antibiot. 2007, 60, 256−260. (b) Kim, S.-J.; Lin, C.-
C.; Pan, C.-M.; Rananaware, D. P.; Ramsey, D. M.; McAlpine, S. R.
MedChemComm 2013, 4, 406−410.
(3) Lin, C.-C.; Tantisantisom, W.; McAlpine, S. R. Org. Lett. 2013,
15, 3574−3577.
(4) Tsai, Y.-C.; Lin, C.-P.; Lin, R.-K.; Kerrigan, J. E.; Rzuczek, S. G.;
LaVoie, E. J.; Rice, J. E.; Pilch, D. S.; Lyu, Y. L.; Liu, L. F. J. Biol. Chem.
2009, 284, 22535−22543.
(5) Satyanarayana, M.; LaVoie, E. J.; Pilch, D. S.; Liu, A.; Liu, L. F.;
Rice, J. E. Bioorg. Med. Chem. Lett. 2008, 18, 3802−3804.
(6) Zhou, X.; Huang, H.; Chen, Y.; Tan, J.; Song, Y.; Zou, J.; Tian,
X.; Hua, Y.; Ju, J. J. Nat. Prod. 2012, 75, 2251−2255.
(7) Islam, A.; Zhang, Y.; Wang, Y.; McAlpine, S. R. MedChemComm
2015, 6, 300−305.
1
Information). H and 2-D NMR were assigned (Supporting
Information), and by use of 2-D NMR a carbon spectral shift
difference was completed between the synthetic and the natural
product (Supporting Information). The carbon shifts were
essentially identical between the natural product and the
synthetic compound 1 (with a maximum of 1 ppm difference
between assigned carbons).
In conclusion, we have described two synthetic routes where
one produced the natural product marthiapeptide A. The
successful ring closing occurred next to the more flexible
thiazoline versus the unsuccessful ring closing attempted next
to the more rigid thiazole. Future analogues might be generated
most effectively by cyclizing between the phenylalanine and the
C
DOI: 10.1021/acs.orglett.5b02574
Org. Lett. XXXX, XXX, XXX−XXX