Total Synthesis of the Cyclic Dodecapeptides Wewakazole and Wewakazole B

Article abstract of DOI:10.1021/acs.orglett.7b01393

The cyclic dodecapeptides wewakazole and wewakazole B have been synthesized by a divergent strategy via a common tris-proline containing oxazole octapeptide and two separate bis-oxazole containing tetrapeptide units, followed by peptide coupling and macro

Full text of DOI:10.1021/acs.orglett.7b01393

Letter
pubs.acs.org/OrgLett
Total Synthesis of the Cyclic Dodecapeptides Wewakazole and
Wewakazole B
Martyn Inman, Hannah L. Dexter, and Christopher J. Moody*
School of Chemistry, University of Nottingham, University Park, Nottingham NG2 7RD, U.K.
S
* Supporting Information
ABSTRACT: The cyclic dodecapeptides wewakazole and wewa-
kazole B have been synthesized by a divergent strategy via a
common tris-proline containing oxazole octapeptide and two
separate bis-oxazole containing tetrapeptide units, followed by
peptide coupling and macrocyclization. The three oxazole amino
acid fragments are readily accessible by rhodium(II)-catalyzed
amide N−H insertion of diazocarbonyl compounds, or by the
cycloaddition of rhodium carbenoids with nitriles.
ewakazole 1 was isolated from the cyanobacterium
Lyngbya majuscula off the coast of Papua New Guinea
particularly those containing azoles,6−10 these two structures
W
caught our attention as attractive targets for total synthesis.
The two wewakazole structures differ only by the presence or
absence of a methyl group on one oxazole and the identity of two
of the peptide residues: the 5-unsubstituted oxazole, phenyl-
alanine, and valine present in wewakazole are replaced with a 5-
methyloxazole, alanine, and phenylalanine in wewakazole B. As
all these variations are found within one portion of the structures,
a divergent strategy for the synthesis of the two compounds
presented itself, whereby the common octapeptide 3 could be
synthesized and coupled to the two bis-oxazole containing
tetrapeptide units 4 and 5. An additional advantage of this
approach is that macrocyclization can be carried out by amide
coupling of the N-terminal to a C-terminal glycine, reducing the
steric bulk around the reactive center and minimizing the risk of
epimerization in what will inevitably be a highly dilute, and hence
slow, macrocyclization step. We have previously reported the use
of rhodium(II)-catalyzed amide N−H insertion of diazocarbonyl
compounds for the synthesis of 5-substituted oxazoles,11,12 as
well as the cycloaddition of metallocarbenoids with nitriles for
the synthesis of 5-unsubstituted oxazoles.13,14 Our retrosynthetic
plan employs both these reactions, allowing the wewakazoles to
be retrosynthesized to their component L-amino acids, and two
diazocarbonyl compounds 6 and 7 (Scheme 1). A protecting
group strategy using methyl and ethyl C-terminal esters and N-
terminal Boc groups was chosen, as the required intermediates
ought to be otherwise inert to the deprotection conditions for
these groups.
in 2003, and assigned a modified cyclododecapeptide structure
incorporating three prolines, and in which one threonine and two
serine residues had been cyclized to form oxazoles.1 Although no
biological activity was reported, subsequent studies showed that
it was active against nonsmall cell lung cancer H-460 cells.2
Thirteen years later, the structurally related wewakazole B 2
(Figure 1) was reported from a different genus of cyanobacteria,
Figure 1. Structures of the cyclododecapeptides wewakazole and
wewakazole B highlighting structural differences.
Moorea producens, in the Red Sea.3 Wewakazole B was of
particular interest, given its reported anticancer activity (IC50
0.58 μM against MCF7 breast cancer cells),3 and the scarcity of
material for further biological evaluation. Although two total
syntheses of wewakazole B have recently been reported,4,5its
mode of action remains unknown. Given our previous work in
the synthesis of post-translationally modified peptides, and
The synthesis began with the construction of the common
octapeptide fragment 3. Both coupling of Boc-alanine to proline
methyl ester and of Boc-proline to glycine methyl ester
Received: May 11, 2017
© XXXX American Chemical Society
A
DOI: 10.1021/acs.orglett.7b01393
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Scheme 1. Retrosynthesis of the Wewakazoles
Scheme 2. Synthesis of First Tetrapeptide Fragment
The coupling of the two tetrapeptide units 13 and 20 proved
challenging; EDCI gave only a 30% yield, while reagents such as
1-[bis(dimethylamino)methylene]-1H-1,2,3-triazolo[4,5-b] pyr-
idinium 3-oxide hexafluorophosphate (HATU), HBTU, benzo-
triazol-1-yl-oxytripyrrolidinophosphonium hexafluorophosphate
(PyBOP), and 2,4,6-tripropyl-1,3,5,2,4,6-trioxatriphos-
phorinane-2,4,6-trioxide (T3P) offered little improvement.
Eventually, it was found that the use of 4-(4,6-dimethoxy-1,3,5-
triazin-2-yl)-4-methylmorpholinium chloride (DMTMM)15 in
THF gave a satisfactory 77% yield, despite the poor solubility of
the coupling agent in that solvent. This reaction could be carried
out on gram scale in 67% yield, permitting access to large
quantities of the common octapeptide fragment 3 (Scheme 4).
Attention was then focused toward the synthesis of the two
bis-oxazole fragments 4 and 5. The wewakazole A fragment 4 was
synthesized via the phenylalanine-derived nitrile 22, which
underwent cycloaddition with diazo compound 6 to give oxazole
23 in 38% yield, again with significant recovery of the nitrile
(43%), followed by Boc deprotection. A solution of Boc-
valinamide and Rh2(OAc)4 in refluxing chloroform was subjected
to dropwise addition of diazo compound 7, giving the
intermediate N−H insertion product in moderate yield.6
Cyclodehydration under Wipf’s conditions16 using triphenyl-
phosphine, iodine, and triethylamine gave the oxazole 25 in good
yield, and treatment with lithium hydroxide gave the carboxylic
acid 26 (Scheme 5). Coupling amine 24 with acid 26 using
HATU gave the desired bis-oxazole 27 in excellent yield,
followed by hydrolysis to yield the wewakazole A fragment 4.
Toward the synthesis of wewakazole B, Boc-alaninamide and
Boc-phenylalaninamide were each subjected to the N−H
insertion−cyclodehydration sequence to give oxazoles 28 and
30 respectively. Boc-deprotection of 28 and ester deprotection of
30, followed by coupling of the resulting amine 29 and acid 31
proceeded without difficulty using N,N,N′,N′-tetramethyl-O-
(1H-benzotriazol-1-yl)uronium hexafluorophosphate (HBTU)
and N,N-diisopropylethylamine (DIPEA) to give dipeptides 8
and 10. Upon removal of both proline protecting groups, the
coupling of these acid 9 and amine 11 fragments proved more
difficult; HBTU gave very poor results, but the use of N-(3-
(dimethylamino)propyl)-N′-ethylcarbodiimide hydrochloride
(EDCI) and 1-hydroxybenzotriazole (HOBt) gave the desired
tetrapeptide 12 in 80% yield (Scheme 2). Deprotection of the
Boc-group in tetrapeptide 12 was only carried out immediately
prior to its further reaction, as the hydrochloride salt 13 was
hygroscopic. The presence of a Pro-Pro linkage generally caused
rotameric behavior to be observed in the NMR spectra of these
intermediates, but this could be minimized by the use of
deuterated DMSO as the solvent. The use of variable
temperature NMR further reduced the occurrence of rotamers,
although this was not performed routinely on all the rotameric
intermediates.
Concurrently, Boc-isoleucinamide was dehydrated with
diethyl chlorophosphate to give the nitrile 14. Slow addition of
diazo compound 6 to a solution of the nitrile and Rh2(pfm)4 in
chloroform at reflux gave only a 30% yield of the oxazole, but with
recovery of 65% of the nitrile giving an 86% yield based on
recovered starting material. Attempts to increase the conversion
beyond this led to much lower recovery of material. The oxazole
15 was treated with HCl in dioxane to give the hydrochloride salt
16. Finally, Boc-phenylalanine was coupled to proline methyl
ester with HBTU to give dipeptide 17, and the ester was
hydrolyzed to give the carboxylic acid 18. Coupling of amine 16
and acid 18 gave the tetrapeptide 19, hydrolysis of which gave the
carboxylic acid 20 (Scheme 3).
B
DOI: 10.1021/acs.orglett.7b01393
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Scheme 3. Synthesis of Second Tetrapeptide Fragment
Scheme 5. Synthesis of Bis-oxazole Fragment of Wewakazole
Scheme 6. Synthesis of Bis-oxazole Fragment of Wewakazole
B
Scheme 4. Synthesis of the Common Octapeptide 3
using HATU, gave the bis-oxazole 32. Final hydrolysis gave the
wewakazole B fragment 5 (Scheme 6).
With all the required fragments in hand, all that remained was
to couple them together to synthesize both natural products. The
bis-oxazole 4 was coupled to the octapeptide 3 using DMTMM
to give the cyclization precursor 33 in 69% yield. The termini
were deprotected under basic (97%) and acidic (quant)
conditions respectively, and cyclization of the 36-membered
ring, again using DMTMM as the coupling agent, delivered
wewakazole 1 in 68% yield after 4 days at room temperature, at
0.55 mM concentration (Scheme 7). The rotamerism observed
in the acyclic intermediates was entirely absent in the macrocycle,
and good NMR spectra could be obtained. A systematic 0.3 ppm
difference in the 13C NMR spectrum not-withstanding, the NMR
spectroscopic data obtained for this material matched closely
those reported for the natural product allowing us to confirm the
structure of wewakazole.
C
DOI: 10.1021/acs.orglett.7b01393
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Scheme 7. Completion of the Synthesis of the Wewakazoles
ASSOCIATED CONTENT
* Supporting Information
■
S
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.orglett.7b01393.
1
Experimental procedures; copies of H and 13C NMR
spectra (PDF)
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: c.j.moody@nottingham.ac.uk.
ORCID
Christopher J. Moody: 0000-0003-0487-041X
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
We thank the University of Nottingham for support.
■
REFERENCES
■
6.
(1) Nogle, L. M.; Marquez, B. L.; Gerwick, W. H. Org. Lett. 2003, 5, 3−
(2) Malloy, K. L.; Villa, F. A.; Engene, N.; Matainaho, T.; Gerwick, L.;
Gerwick, W. H. J. Nat. Prod. 2011, 74, 95−98.
(3) Lopez, J. A. V.; Al-Lihaibi, S. S.; Alarif, W. M.; Abdel-Lateff, A.;
Nogata, Y.; Washio, K.; Morikawa, M.; Okino, T. J. Nat. Prod. 2016, 79,
1213−1218.
(4) Long, B. H.; Zhang, J. Z.; Tang, X. D.; Wu, Z. Z. Org. Biomol. Chem.
2016, 14, 9712−9715.
(5) Nayani, K.; Hussaini, S. D. A. Tetrahedron Lett. 2017, 58, 1166−
1169.
(6) Bagley, M. C.; Bashford, K. E.; Hesketh, C. L.; Moody, C. J. J. Am.
Chem. Soc. 2000, 122, 3301−3313.
(7) Hughes, R. A.; Thompson, S. P.; Alcaraz, L.; Moody, C. J. J. Am.
Chem. Soc. 2005, 127, 15644−15651.
(8) Linder, J.; Blake, A. J.; Moody, C. J. Org. Biomol. Chem. 2008, 6,
3908−3916.
With wewakazole complete, attention turned to wewakazole B.
Coupling of the octapeptide 3 with the bis-oxazole fragment 5
proceeded in 76% yield using DMTMM in chloroform. The C-
terminal was deprotected first (85%), and then the Boc group
was removed from the N-terminal using HCl in dioxane (quant).
Cyclization was first attempted using the conditions reported by
Wu et al.,4 with HATU in dichloromethane. After 7 days, the
product could be isolated in 36% yield, with only a trace of
tetramethylurea proving difficult to remove. An attempt to use
PyBOP as the coupling agent, to circumvent the presence of the
urea side product, gave excellent conversion, but purification was
again very difficult due to the coelution of wewakazole B with the
tripyrrolidinophosphine oxide side product. As previously, the
best results were obtained using DMTMM in chloroform, which
gave a 78% yield of clean material after 5 days at room
temperature (Scheme 7). The spectroscopic data obtained for
synthetic wewakazole B matched the literature data for the
natural product.
(9) Wada, H.; Williams, H. E. L.; Moody, C. J. Angew. Chem., Int. Ed.
2015, 54, 15147−15151.
(10) Dexter, H. L.; Williams, H. E. L.; Lewis, W.; Moody, C. J. Angew.
Chem., Int. Ed. 2017, 56, 3069−3073.
(11) Bagley, M. C.; Buck, R. T.; Hind, S. L.; Moody, C. J. J. Chem. Soc.,
Perkin Trans. 1 1998, 591−600.
(12) Moody, C. J.; Bagley, M. C. J. Chem. Soc., Perkin Trans. 1 1998,
601−607.
(13) Doyle, K. J.; Moody, C. J. Tetrahedron 1994, 50, 3761−3772.
(14) Linder, J.; Garner, T. P.; Williams, H. E. L.; Searle, M. S.; Moody,
C. J. J. Am. Chem. Soc. 2011, 133, 1044−1051.
(15) Kunishima, M.; Kawachi, C.; Morita, J.; Terao, K.; Iwasaki, F.;
Tani, S. Tetrahedron 1999, 55, 13159−13170.
(16) Wipf, P.; Miller, C. P. J. Org. Chem. 1993, 58, 3604−3606.
In summary, the total syntheses of wewakazole and its
congener wewakazole B have been carried out using rhodium
catalyzed cycloadditions and N−H insertion reactions to provide
the oxazoles. The overall strategy is both convergent and
divergent in nature and can readily be applied to the synthesis of
analogues of the natural products.
D
DOI: 10.1021/acs.orglett.7b01393
Org. Lett. XXXX, XXX, XXX−XXX