Total Synthesis of Alloviroidin

Article abstract of DOI:10.1021/acs.orglett.9b00567

Alloviroidin is a cyclic heptapeptide, produced by several species of Amanita mushrooms, that demonstrates high affinity for F-actin as is characteristic of virotoxins and phallotoxins. Alloviroidin was synthesized via a [3 + 4] fragment condensation of F

Full text of DOI:10.1021/acs.orglett.9b00567

Letter
Cite This: Org. Lett. XXXX, XXX, XXX−XXX
pubs.acs.org/OrgLett
Total Synthesis of Alloviroidin
Carol M. Taylor,* Samuel K. Kutty, and Benson J. Edagwa
Department of Chemistry, Louisiana State University, Baton Rouge, Louisiana 70803, United States
*
S Supporting Information
ABSTRACT: Alloviroidin is a cyclic heptapeptide, produced by several species of
Amanita mushrooms, that demonstrates high affinity for F-actin as is characteristic of
virotoxins and phallotoxins. Alloviroidin was synthesized via a [3 + 4] fragment
condensation of Fmoc-D-Thr(OTBS)-D-Ser(OTBS)-(2S,3R,4R)-DHPro(OTBS) -OH
2
and H-Ala-Trp(2-SO Me)-(2S,4S)-DHLeu(5-OTBS)-Val-OMe to form bond A. The
2
1
linear heptapeptide favored a turn conformation, facilitating cyclization between Val
and D-Thr (position B). Global deprotection and HPLC purification afforded
2
alloviroidin with NMR spectra in excellent agreement with the natural product.
9
b
oxins of the Amanita mushrooms have been the subject of
folklore and murder mysteries. Fascination with these
nmol/g when administered via intraperitoneal injection.
11
T
Virotoxins have the same rigid conformation as the
12
notorious compounds and their pharmacology was the life’s
phallotoxins and bind actin with comparable affinity. This is
quite remarkable given the monocyclic nature of the
compounds. The rigid conformation is likely due to the amino
1
work of Theodore Wieland. Ongoing interest and develop-
2
ments have recently been updated in Walton’s monograph.
13
Chemical synthesis, exemplified by an elegant approach to
acids in positions 3 and 4, a D-Ser and a (2S,3R,4R)-3,4-
3
14
amanatoxin by Perrin’s group last year, is challenging due to the
dihydroxy-L-proline (DHPro), respectively.
Virotoxin analogs have been synthesized by Wieland and co-
density of post-translational modifications and bridging
functionality. One of the best-known toxins is the bicyclic
heptapeptide, phalloidin (1), characterized by a thioindolyl
bridge, which adopts a rigid conformation. This protypical
phallotoxin binds tightly to F-actin and is the ongoing target of
1
5
workers, substituting cis-4-hydroxy-L-proline at position 4.
Further structure−activity studies have been limited by the
availability of the unusual constituent amino acids. Several years
ago we developed a reliable route to 3,4-dihydroxyprolines from
pentose sugars. In 2009, we reported the synthesis of a
dipeptide containing a (2S,4S)-4,5-dihydroxyleucine (DHLeu)
16
4
synthetic chemistry. Conjugated to fluorophores, phalloidin is
5
widely used to study the dynamics of actin filaments.
17
residue, as found in alloviroidin. We have also studied the
Modulators of actin polymerization are valuable in the study
1
8
6
impact of prolyl hydroxylation on peptide conformation.
of the cytoskeleton of cancer cells, a target for chemo-
7
Thus, we were uniquely positioned to contemplate the synthesis
therapeutic drugs. The virotoxins (e.g., 2 and 3, Figure 1) are
of virotoxins.
recognized members of the cast of amanotoxins but have been
In a landmark discovery, about 10 years ago, Hallen et al.
demonstrated that amanatoxins and phallacins, produced by A.
bisporangia (another destroying angel), were the first known
example of cyclic peptides to be produced by ribosomal peptide
less well-studied. They were first isolated from Amanita
8
virosa the so-called “destroying angel”and have also been
9
identified in extracts from A. suballiacea and A. subpallidoro-
1
0
sea. With respect to white mice, the LD of alloviroidin is 1−2
50
19
synthesis in a fungus. The variable toxin region of the linear
precursor peptides is converted to cycloamanides (cyclic
peptides from Amanita species) in two steps, catalyzed by a
20
specialized prolyl oligopeptidase (POPB).
While biosynthesis is often an inspiration for chemical
synthesis, that is rarely so in peptide chemistry. Fundamentally,
ribosomal peptide synthesis proceeds from N- to C-terminus,
whereas chemical synthesis is routinely conducted in the reverse
direction to avoid epimerization of Cα stereogenic centers. A C-
terminal Pro residue is advantageous in peptide fragment
condensations, because Pro is not susceptible to epimerization.
Received: February 13, 2019
Figure 1. Selected Amanita toxins.
©
XXXX American Chemical Society
A
DOI: 10.1021/acs.orglett.9b00567
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
However, to facilitate cyclization, we advocate for embedding
the Pro residue in the middle of the linear peptide, in the
expectation that there will be conformations through which
cyclization is entropically enabled. Indeed, others have invoked
Scheme 2. Synthesis of Tetrapeptide 4
“
pseudoprolines” to facilite the cyclization of recalcitrant
21
sequences. Thus, our retrosythetic analysis is presented in
Scheme 1A. Amalgamation of a “northern” tetrapeptide amine
Scheme 1. (A) Retrosynthetic Analysis of Alloviroidin (3)
and (B) Backbone Cleavage of Peptides Containing γ-
Hydroxy-amino acids under Acidic Conditions
(
5
derived from 4) and a “southern” tripeptide acid (derived from
) would generate a linear heptapeptide with the DHPro residue
(Dhl) (8) by porcine kidney acylase (PKA), Cbz-protection,
and incorporation into a dipeptide followed by a Sharpless
asymmetric dihydroxylation of the side chain double bond
in the third position. This was appealing, since the activated
tripeptide, with its C-terminal dihydroxyproline residue, would
not be susceptible to epimerization during coupling/cyclization
30
(Scheme 2). On the basis of our subsequent experience with
31
(
vide infra). This penultimate step should also be favored by the
the Corey−Lygo asymmetric synthesis of amino acids, we
recognized that the synthesis of Dhl was an ideal application of
this method, since the benzophenone imine of glycine tert-butyl
ester would be alkylated by an allylic electrophile. Known
fact that the two residues (L-Val and D-Thr) are of opposite
2
2
configuration, albeit somewhat hindered.
While modern peptide synthesis is dominated by Fmoc
chemistry, in conjunction with acid-labile side chain protecting
groups, this approach would have been incompatible with the
DHLeu residue. Under acidic conditions, γ-hydroxyamides form
31a
compound 10 was readily converted to Cbz-Dhl-OH (9) and
condensed with valine methyl ester (11). Elaboration, as
17
described previously, gave dipeptide 12. Activation of Fmoc-
Ala-OH (14) and condensation with tryptophan 7 gave
dipeptide acid 15; for the purposes of characterization this
was converted to methyl ester 16. Hydrogenolysis of the N-
terminal Cbz group of compound 12 gave amine 13 that was
1
2
7
3
γ-lactones with concomitant cleavage of the amide bond
Scheme 1B). We elected to employ fluoride-labile TBS ethers
(
as protecting groups for the heavily hydroxylated peptide. The
application of silyl ethers in peptide chemistry has gathered
3
,24,25
32
momentum in the past decade,
notably in the presence of
coupled with acid 15, according to Carpino, to give
functional groups that do not tolerate acidic conditions, e.g., the
recent synthesis of imine-containing scytonemide A by Wilson
tetrapeptide 4 in good yield, with no epimerization of the Trp
residue observed. The tetrapeptide fragment was stored in the
form of 4 and deprotected immediately prior to the fragment
condensation.
2
6
et al.
Construction of the 2-methylsulfonyl-tryptophan residue
[
Trp(SO Me)] is depicted in Scheme 2. Specifically, Boc-Trp-
The synthesis of tripeptide 5 is summarized in Scheme 3.
2
t
14
O Bu was subjected to methylsulfenyl chloride generated in
situ. While indoles typically undergo S Ar at C-3, it is well
(2S,3R,4R)-3,4-Dihydroxy-L-proline (DHPro) building block
2
7
16a
18 was prepared as described previously. Dipeptide acid 23
E
established that tryptophan derivatives react with alkylsulfenyl
was prepared in the standard manner (vide supra) and coupled
with dihydroxyprolylamine 20. As discussed earlier, we needed
to substitute for the acid-labile protecting groups. Removal of
these protecting groups, notably the MEM ethers, was not
straightforward. Cleavage with TFA in dichloromethane was
28
chlorides to afford the 2-thioether, presumably via initial attack
29
at C-3 followed by rearrangement. The resulting thioether was
not particularly stable, so it was oxidized immediately to the
corresponding sulfone. Liberation of the free amino acid 7 was
accomplished under acidic conditions.
33
sluggish; treatment with TiCl4 sometimes gave good results,
but was not reproducible. Ultimately, treatment with hydro-
Our original synthesis of (2S,4S)-4,5-dihydroxyleucine
DHLeu) included resolution of N-acetyl-dehydroleucine
34
(
chloric acid in TFA was the most effective and reliable.
B
DOI: 10.1021/acs.orglett.9b00567
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Scheme 3. Synthesis of Tripeptide 5
phosphate buffer with ethyl acetate and used without further
purification. Cyclization was achieved with HATU. Use of
collidine as base was not as effective as diisopropylethylamine.
Cyclizations were originally attempted in DMF, but found to be
faster and cleaner in dichloromethane, at a concentration of 1
mM, affording an excellent yield of cyclic heptapeptide 29.
On standing, partial cleavage of one or more TBS ethers from
2
9 was observed, making it impossible to obtain good NMR
data. Thus, we proceeded directly with a global deprotection
using buffered tetrabutylammonium fluoride. DiLauro et al.
36
recently demonstrated that it is not necessary to use
37
stoichiometric fluoride; deprotection of the five TBS groups,
n
to afford 3, was accomplished with 2.5 equiv of Bu NF
4
overnight. We experienced great difficulty separating the
resulting alloviroidin (3) from salts and found that the protocol
38
of Kishi and Kaburagi was beneficial. Reversed phase HPLC
then afforded pure alloviroidin in disappointing yield, but high
purity. Synthetic 3 was characterized by NMR in D O/H O at
2
2
500 MHz, with full assignment of signals (see Supporting
Information). Proton NMR spectra acquired in DMSO-d were
6
in excellent agreement with those of the natural product (see
Supporting Information).
Compounds 19, 25, 26, and 5 existed as mixtures of
conformations about the prolyl peptide bond. H NMR
Protection of the four alcohols as TBS ethers was uneventful.
The C-terminal benzyl ester in 5 was removed hydrogenolyti-
cally to afford acid 27. This reaction had to be monitored
carefully, as cleavage of the Fmoc group was observed on
prolonged exposure; triethylsilane was preferable to hydrogen
gas in this regard.
1
indicated ratios ranging from 1:1 (for 19) up to 2:1 (for 25),
presumably favoring the trans conformation. Since NMR spectra
were complicated by these equilibria, we demonstrated the
purity of key compounds by normal phase HPLC (see
Supporting Information). Our earlier studies on the con-
Coupling of tripeptide acid 27 and tetrapeptide amine 17 gave
linear heptapeptide 28 in good yield using HATU/collidine
18
formation of dipeptides containing hydroxylated prolines did
not include the (2S,3R,4R)-DHPro isomer. Crystallographic
(Scheme 4). The N-terminus was deprotected by treatment with
evidence is available for related monohydroxylated prolines with
piperidine; it was advantageous to purify the heptapeptide amine
by flash chromatography at this stage. The C-terminal methyl
ester was hydrolyzed with trimethyltin hydroxide, using the
39
40
(
2S,3R) [3-OH] and (2S,4R) [4-OH] configurations. Both
amino acids demonstrated a Cβ-exo conformation of the
pyrrolidine ring. In the latter case, the residue was incorporated
into collagen-like sequences and found to destabilize the triple
helix, implying a destabilization of the trans prolyl peptide bond
relative to its cis rotamer. We suspect that the DHPro residue in
the virotoxins also prefers a Cβ-exo conformation (Scheme 5)
35
conditions initially reported by Nicolaou et al. The linear
heptapeptide with free termini was extracted from pH 5.5
Scheme 4. Fragment Condensation, Cyclization,
Deprotection, and Purification To Afford Alloviroidin (3)
Scheme 5. DHPro-Induced Turn in Linear Peptide Facilitates
Cyclization Followed by “Relaxation” to Type II β-Turn
that is stabilized by hyperconjugative donation from σ(Cα-CO)
→
σ*(Cβ-O) and σ(Cδ-H ) → σ*(Cγ-O). The proton NMR
ax
signals for Hα, Hβ, and Hγ of peptides containing the DHPro
appear as broad singlets, consistent with small coupling
constants in predominantly the Cβ-exo conformation. Further,
we speculate that a turn induced by the proline residue brings
the two “arms” of the peptide close enough to enable cyclization.
Following ring formation and deprotection, alloviroidin adopts a
type II β-turn in this region of the molecule, with the prolyl
peptide bond in the trans conformation, as previously described
by Kobayashi et al., who noted shielding of the Ala methyl group
by the aromatic side chain of the adjacent Trp residue and a
11a
hydrogen bond between the D-Ser CO and the Trp NH.
C
DOI: 10.1021/acs.orglett.9b00567
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
In summary, we have prepared alloviroidin from the seven
constituent amino acids via a solution-based fragment
condensation approach. There are two longest linear sequences
for the peptide assembly, including deprotection steps and
purification: 11 steps from DHPro 18 (7.7% overall) and 11
steps from DHLeu 9 (5.4% overall). To the best of our
knowledge this represents the first total synthesis of a virotoxin
natural product.
(9) (a) Little, M. C.; Preston, J. F., III J. Nat. Prod. 1984, 47, 93−99.
(
b) Little, M. C.; Preston, J. F., III; Jackson, C.; Bonetti, S.; King, R. W.
Biochemistry 1986, 25, 2867−2872.
10) Wei, J.; Wu, J.; Chen, J.; Wu, B.; He, Z.; Zhang, P.; Li, H.; Sun, C.;
Liu, C.; Chen, Z.; Xie, J. Toxicon 2017, 133, 26−32.
11) (a) Kobayashi, N.; Endo, S.; Kobayashi, H.; Faulstich, H.;
Wieland, T.; Munekata, E. Eur. J. Biochem. 1995, 232, 726−736.
(
(
(
b) Bhaskaran, R.; Yu, C. Int. J. Pept. Protein Res. 1994, 43, 393−401.
(
(
́
12) Gicquaud, C.; Pare, M. Biochem. Cell Biol. 1992, 70, 719−723.
13) Zanotti, G.; Kobayashi, N.; Munekata, E.; Zobeley, S.; Faulstich,
ASSOCIATED CONTENT
H. Biochemistry 1999, 38, 10723−10729.
■
(14) Buku, A.; Faulstich, H.; Wieland, T.; Dabrowski, J. Proc. Natl.
*
S
Supporting Information
Acad. Sci. U. S. A. 1980, 77, 2370−2371.
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.orglett.9b00567.
(15) Kahl, J. U.; Vlasov, G. P.; Seeliger, A.; Wieland, T. Int. J. Pept.
Protein Res. 1984, 23, 543−550.
(16) (a) Taylor, C. M.; Jones, C. E.; Bopp, K. Tetrahedron 2005, 61,
Procedures for the synthesis of all compounds in Schemes
9611−9617. (b) Taylor, C. M.; Barker, W. D.; Weir, C. A.; Park, J. H. J.
Org. Chem. 2002, 67, 4466−4474. (c) Weir, C. A.; Taylor, C. M. Org.
Lett. 1999, 1, 787−789.
1
13
2
−4, along with their H and C NMR spectra and
HPLC traces for compounds 10, 4, 5, and 28; assigned
1
1
(
17) Edagwa, B. J.; Taylor, C. M. J. Org. Chem. 2009, 74, 4132−4136.
H− H COSY and HSQC spectra for alloviroidin (3),
along with H NMR comparisons of synthetic and natural
material (PDF)
1
(18) Taylor, C. M.; Hardre,
0, 1306−1315.
19) Hallen, H. E.; Luo, H.; Scott-Craig, J. S.; Walton, J. D. Proc. Natl.
Acad. Sci. U. S. A. 2007, 104, 19097−19101.
20) (a) Luo, H.; Hong, S. Y.; Sgambelluri, R. M.; Angelos, E.; Li, X.;
́
R.; Edwards, P. J. B. J. Org. Chem. 2005,
7
(
(
AUTHOR INFORMATION
■
*
Walton, J. D. Chem. Biol. 2014, 21, 1610−1617. (b) Sgambelluri, R. M.;
Smith, M. O.; Walton, J. D. ACS Synth. Biol. 2018, 7, 145−152.
Corresponding Author
E-mail: cmtaylor@lsu.edu.
(
21) (a) Ruckle, T.; de Lavallaz, P.; Keller, M.; Dumy, P.; Mutter, M.
̈
Tetrahedron 1999, 55, 11281−11288. (b) Skropeta, D.; Jolliffe, K. A.;
Turner, P. J. J. Org. Chem. 2004, 69, 8804−8809.
ORCID
(22) Kessler, H.; Kutscher, B. Liebigs Ann. Chem. 1986, 1986, 869−
Carol M. Taylor: 0000-0002-6262-0796
Samuel K. Kutty: 0000-0001-8137-9988
Notes
8
92.
23) Fischer, P. M. Tetrahedron Lett. 1992, 33, 7605−7608.
(24) Barbie, P.; Kazmaier, U. Org. Lett. 2016, 18, 204−207.
25) Katsuyama, A.; Paudel, A.; Panthee, S.; Hamamoto, H.;
Kawakami, T.; Hojo, H.; Yakushiji, F.; Ichikawa, S. Org. Lett. 2017,
9, 3771−3774.
26) Wilson, T. A.; Sullivan, P.; Demoret, R. M.; Orjala, J.;
Rakotondraibe, L. H.; Fuch, J. R. J. Nat. Prod. 2018, 81, 534−542.
27) Dillard, R. D.; Bach, N. J.; Draheim, S. E.; Berry, D. R.; Carlson,
(
(
The authors declare no competing financial interest.
1
(
ACKNOWLEDGMENTS
■
These studies were supported by the Department of Chemistry
at LSU. We thank the following staff in the Department of
Chemistry at LSU for their expertise: Connie David for mass
spectrometry and Thomas K. Weldeghiorghis and Fengli Zhang
for NMR. We thank Emeritus Professor James F. Preston,
University of Florida, for retrieving and sharing his NMR spectra
of natural alloviroidin.
(
D. G.; Chirgadze, N. Y.; Clawson, D. K.; Hartley, L. W.; Johnson, L. M.;
Jones, N. D.; McKinney, E. R.; Mihelich, E. D.; Olkowski, J. L.; Schevitz,
R. W.; Smith, A. C.; Snyder, D. W.; Sommers, C. D.; Wery, J.-P. J. Med.
Chem. 1996, 39, 5119−5136.
(
28) Crich, D.; Davies, J. W. Tetrahedron Lett. 1989, 30, 4307−4308.
(29) Jackson, A. H.; Lynch, P. P. J. Chem. Soc., Perkin Trans. 2 1987,
1
(
(
483−1488.
30) Yadav, S.; Taylor, C. M. J. Org. Chem. 2013, 78, 5401−5409.
REFERENCES
■
31) (a) Corey, E. J.; Xu, F.; Noe, M. C. J. Am. Chem. Soc. 1997, 119,
(
1) Wieland, T. Peptides of the poisonous Amanita mushrooms;
Springer-Verlag: New York, 1986.
1
3
(
(
2414−12415. (b) Lygo, B.; Wainwright, P. G. Tetrahedron Lett. 1997,
8, 8595−8598.
(
2) Walton, J. The cyclic peptide toxins of Amanita and other poisonous
mushrooms; Springer: International, 2018.
3) Matinkhoo, K.; Pryyma, A.; Todorovic, M.; Patrick, B. O.; Perrin,
D. M. J. Am. Chem. Soc. 2018, 140, 6513−6517.
4) (a) Schuresko, L. A.; Lokey, R. S. Angew. Chem., Int. Ed. 2007, 46,
547−3549. (b) Anderson, M. O.; Shelat, A. A.; Guy, R. K. J. Org.
Chem. 2005, 70, 4578−4584.
5) Wulf, E.; Deboben, A.; Bautz, F. A.; Faulstich, H.; Wieland, T.
32) Carpino, L. A.; El-Faham, A. J. Org. Chem. 1994, 59, 695−698.
33) (a) Corey, E. J.; Gras, J.-L.; Ulrich, P. Tetrahedron Lett. 1976, 17,
(
8
09−812. (b) Miyata, O.; Shinada, T.; Ninomiya, I.; Naito, T.
Tetrahedron Lett. 1991, 32, 3519−3522.
34) MEM protection for (4R)-4-hydroxy-l-proline was used
(
3
(
previously in the synthesis of collagen mimetics: Vadolas, D.;
Germann, H. P.; Thakur, S.; Keller, W.; Heidemann, E. Int. J. Pept.
Protein Res. 1985, 25, 554−559.
(
Proc. Natl. Acad. Sci. U. S. A. 1979, 76, 4498−4502.
(
35) Nicolaou, K. C.; Estrada, A. A.; Zak, M.; Lee, S. H.; Safina, B. S.
(
(
6) Olson, M. F.; Sahai, E. Clin. Exp. Metastasis 2009, 26, 273−287.
7) (a) Yoon, J.; Terman, J. R. Mol. Cell Oncol. 2018, 5, No. e1384881.
Angew. Chem., Int. Ed. 2005, 44, 1378−1382.
(
(
36) Wipf, P.; Uto, Y. J. Org. Chem. 2000, 65, 1037−1049.
37) DiLauro, A. M.; Seo, W.; Phillips, S. T. J. Org. Chem. 2011, 76,
(
b) Stehn, J. R.; Haass, N. K.; Bonello, T.; Desouza, M.; Kottyan, G.;
Treutlein, H.; Zeng, J.; Nascimento, P. R. B. B.; Sequeira, V. B.; Butler,
T. L.; Allanson, M.; Fath, T.; Hill, T. A.; McCluskey, A.; Schevzov, G.;
Palmer, S. J.; Hardeman, E. C.; Winlaw, D.; Reeve, V. E.; Dixon, I.;
Weninger, W.; Cripe, T. P.; Gunning, P. W. Cancer Res. 2013, 73,
7
(
352−7358.
38) Kaburagi, Y.; Kishi, Y. Org. Lett. 2007, 9, 723−726.
39) Jenkins, C. L.; Bretscher, L. E.; Guzei, I. A.; Raines, R. T. J. Am.
(
Chem. Soc. 2003, 125, 6422−6427.
40) Shamala, N.; Row, T.N. G.; Venkatesan, K. Acta Crystallogr., Sect.
B: Struct. Crystallogr. Cryst. Chem. 1976, B32, 3267−3270.
5
(
1
169−5182.
8) Faulstich, H.; Buku, A.; Bodenmu
980, 19, 3334−3343.
(
̈
ller, H.; Wieland, T. Biochemistry
D
DOI: 10.1021/acs.orglett.9b00567
Org. Lett. XXXX, XXX, XXX−XXX