Understanding host-selective phytotoxicity: Synthesis and biological discrimination of phomalide and its (Z)-isomer

Full text of DOI:10.1021/jo961661a

8008
J . Org. Chem. 1996, 61, 8008-8009
Sch em e 1
Un d er sta n d in g Host-Selective
P h ytotoxicity: Syn th esis a n d Biologica l
Discr im in a tion of P h om a lid e a n d Its
(Z)-Isom er
Dale E. Ward,* Alfredo Vazquez, and
M. Soledade C. Pedras*
Department of Chemistry, University of Saskatchewan,
110 Science Place, Saskatoon, SK, Canada S7N 5C9
Received August 28, 1996
Numerous fungal pathogens selectively damage host
plants via the synthesis and release of host-selective
phytotoxins.1 Although a large number of fungal diseases
appear to be mediated by host-selective toxins, the
molecular basis for the selectivity of this process is not
well understood in the majority of the cases.1 One such
disease, blackleg, affects many economically important
brassica crops and can be particularly devastating for the
oilseeds canola (Brassica napus and B. rapa) and rape-
seed (B. napus and B. rapa).2 Blackleg disease has
caused significant crop losses in Canada and worldwide,
as there are no disease resistant varieties commercially
available. Recently, a better understanding of the black-
leg causing fungus [Leptosphaeria maculans (Desm.) Ces.
et de Not., asexual stage Phoma lingam (Tode ex Fr.)
Desm.] was achieved with the isolation of phomalide (1),
a host-selective toxin, which appears to be involved in
disease development.3 Remarkably, phomalide is pro-
duced by P. lingam in liquid culture only for a short
period (24-60 h); older fungal cultures produce only
nonselective phytotoxins (epipolythiodioxopiperazines)
such as sirodesmin PL.4 This unusually short production
period was attributed to an inhibitory effect of sirodesmin
PL on the biosynthesis of phomalide. Because of the
small quantities obtained from fungal cultures and
difficulties in scaling up production, evaluation of the role
of 1 in the development of blackleg disease has been
seriously hampered. Therefore, the chemical synthesis
of phomalide was an essential prelude to biological
studies. We have now synthesized phomalide (1) and its
(Z)-isomer (isophomalide, 2) and assayed their phytotox-
icity to blackleg resistant and susceptible plants. Most
noteworthy, we established that blackleg susceptible
plants discriminate between 1 and 2.
noteworthy structural features within this class of natu-
ral products.5
The synthesis of cyclic depsipeptides (and peptides)
typically proceeds by coupling (linear or convergent) of
intact hydroxy acid and amino acid fragments followed
by cyclization.6 Such an approach focuses the strategic
decisions on the site of cyclization and the order of the
residue coupling. Application of this strategy to phoma-
lide leaves the introduction of the Aba residue and control
of its stereochemistry as a major concern because R,â-
unsaturated amino acids are not normally incorporated
into peptides as intact residues.7 A variety of methods7
for the synthesis of derivatives of unsaturated amino
acids including Aba8 have been reported but few have
been applied to stereocontrolled peptide (or depsipeptide)
synthesis.9 Our retrosynthetic analysis of phomalide is
presented in Scheme 1. We selected the Val-D-Leu
linkage as the most favorable cyclization site because it
is an amide bond between two amino acid residues of
opposite configuration; all of these features are prece-
dented to facilitate cyclization.6,10 The [2 + 3] fragment
coupling approach to the acyclic precursor was chosen
to provide flexibility in the choice of potential Aba
precursors [e.g. D- or L-threonine],11 an alternative site
for cyclization,12 and convergence. The three component
residues of the tridepsipeptide 4 are readily available.
We chose to incorporate an intact Aba residue (rather
than, for example, a threonine derivative)8 in the dipep-
tide fragment for increased efficiency and because this
would allow [2 + 3] fragment coupling to proceed without
risk of “racemization”. We initially sought to utilize an
N-blocked Val-(E)-Aba dipeptide; however, because the
stereochemical integrity of the thermodynamically less
stable (E)-configuration could not be preserved through
fragment coupling and cyclization (vide infra), the syn-
Phomalide (1) is an unusual cyclic depsipeptide com-
posed of three R-amino acid and two R-hydroxy acid
residues. The structure was assigned on the basis of
spectroscopic data; the (E)-configuration of the 2-amino-
2-butenoic (Aba) residue was determined from NOE
experiments and the absolute configurations of the four
stereogenic centers were determined by acid hydrolysis
and comparison of the resulting intact residues with
authentic samples.3 The 15-membered ring, the consecu-
tive ester linkages, and the unusual (E)-Aba residue are
(7) Reviews: (a) Schmidt, U.; Lieberknecht, A.; Wild, J . Synthesis
1988, 159-172. (b) Noda, K.; Shimohigashi, Y.; Izumiya, N. In The
Peptides; Gross, E.; Meienhofer, J ., Eds.; Academic Press: New York,
1983, Vol. 5, pp 285-339.
(1) For a recent multiauthor review on phytotoxins, see: Graniti,
A.; et al. Experientia 1991, 47, 751-826.
(2) (a) For a recent review on the history and control of blackleg
disease, see: Gugel, R. K.; Petrie, G. A. Can. J . Plant Pathol. 1992,
14, 36-45. (b) Canola refers to varieties of rapeseed containing very
low amounts of erucic adic and glucosinolates.
(3) Pedras, M. S. C.; Taylor, J . T.; Nakashima, T. T. J . Org. Chem.
1993, 58, 4778-4780.
(4) For a recent review on phytotoxins of the blackleg fungus, see:
Pedras, M. S. C. Rev. Latinoam. Quim. In press.
(5) (a) Turner, W. B.; Aldrige, D. C. Fungal Metabolites II; Academic
Press: New York, 1983; pp 436-442. (b) Kleinkauf, H.; von Doheren,
H. Eur. J . Biochem. 1990, 192, 1-15.
(6) (a) Bodanszky, M. Principles of Peptide Synthesis, 2nd ed.;
Springer-Verlag: Berlin, 1993. (b) Izumiya, N.; Kato, T.; Aoyagi, H.;
Waki, M.; Kondo, M. Synthetic Aspects of Biologically Acitve Cyclic
Peptides; Halsted Press: New York, 1979. (c) Kopple, K. D. J . Pharm.
Sci. 1972, 61, 1345-1356.
(8) Most methods give the (Z)-isomers selectively.7 The (E)-isomers
can be prepared by syn-elimination of certain threonine derivatives
[(a) Rich, D. H.; Tam, J . P. J . Org. Chem. 1977, 42, 3815-3820] or by
anti-elimination of allo-threonine derivatives: (b) Somekh, L.; Shanzer,
A. Ibid. 1983, 48, 907-908 and cited references.
(9) For a recent example involving (Z)-Aba, see: Li, K. W.; Wu, J .;
Xing, W.; Simon, J . A. J . Am. Chem. Soc. 1996, 118, 7237-7238.
(10) (a) Cavelier-Frontin, F.; Achmad, S.; Verducci, J .; J acquier, J .;
Pe`pe, G. J . Mol. Struct. (THEOCHEM) 1993, 286, 125-130. (b) Ueda,
K.; Waki, M.; Izumiya, N. Int. J . Pept. Protein Res. 1987, 30, 33-39
and cited references.
(11) Changing the configuration of one residue in the linear precur-
sor can have a dramatic effect on cyclization.6
(12) The Aba-Hpp linkage is also an attractive cyclization site
because it is not subject to “racemization”.
S0022-3263(96)01661-1 CCC: $12.00 © 1996 American Chemical Society

Communications
J . Org. Chem., Vol. 61, No. 23, 1996 8009
exclusively the (Z)-isomer 11.17 Hydrogenolysis of 11
gave the corresponding amino acid 12 which failed to
cyclize under the influence of EDC/HOBt but gave
isophomalide (2) in 20% yield after reaction with BOP-
Cl/iPr2EtN. Alternatively, cyclization of 13 via the pen-
tafluorophenyl ester gave 2 in excellent overall yield.18,19
Other than small differences in the chemical shifts of the
Sch em e 2
1
signals for the Aba residue, the H and 13C NMR spectra
of isophomalide were very similar to those of phomalide
(∆δH < 0.1; ∆δC < 0.5). Numerous attempts to isomerize
2 into 1 under acidic, basic, or photochemical conditions
failed. Finally, addition of PhSeH to 2 gave three stereo-
isomeric selenides 14 in a 10:2:1 ratio (65%).20 Oxidation
of the major selenide with H2O2 gave phomalide (1)
uncontaminated by isophomalide (2).20-23 The synthesis
of isophomalide (2) proceeds in five steps and 30% overall
yield from 5 and 6; the isomerization of 2 into 1 occurs
with complete stereoselectivity and represents a rare
example of a stereocontrolled synthesis of an (E)-dehy-
droamino acid containing depsipeptide (or peptide).7
To evaluate the selective phytotoxicity of phomalide
(1) and its (Z)-isomer (isophomalide, 2) to plants resistant
and susceptible to blackleg, both compounds were as-
sayed in planta on leaves of canola (susceptible), mustard
(B. juncea, resistant), and wasabi (Eutrema wasabi,
resistant),24 as previously reported for other toxins.25 The
naturally occurring phomalide (1) caused necrotic, chlo-
rotic, and red lesions on canola leaves (10-5 M), whereas
no damage was observed on mustard or wasabi leaves,
even at significantly higher concentrations (10-4 M).
Thus, the selective phytotoxicity of phomalide (1) appears
to mimic the pathogenicity range of the blackleg fungus.25
That is, phomalide causes lesions on plants that are
susceptible to blackleg infection (canola), but not on
resistant plants (mustard and wasabi). Most impor-
tantly, the (Z)-isomer 2 did not cause lesions on any of
the plant leaves tested, even at 5 × 10-4 M. These results
indicate that the configuration of the double bond in
phomalide (1) is important for phytotoxicity; however, it
remains to be determined whether (and how) resistant
plants enzymatically convert phomalide to less toxic
products. An interesting possibility would be the isomer-
ization of 1 into the nonphytotoxic 2, a process that would
not occur in susceptible plants.
thesis proceeded with the more stable (Z)-isomer followed
by a late-stage stereoselective isomerization.7,13
Condensation of Hmp-OH (5) with D-Leu-OBn (6)
mediated by DCC/HOBt gave the didepsipeptide 7 which,
in turn, was acylated with TBDMSO-Hpp-Cl (8; gener-
ated in situ)14 in the presence of DMAP to provide the
three-residue depsipeptide fragment Hpp-Hmp-D-Leu-
OBn (4) after removal of the silyl ether protecting group.
Reaction of Cbz-Val-NH2 (10)15 with 2-oxobutanoic acid
gave the dipeptide 3 as a 4:1 mixture of (Z)- and (E)-
isomers, respectively, in analogy to literature precedent
(Scheme 2).16 The adduct resulting from coupling 4 with
the mixture of 3 isomers using DCC/DMAP was almost
Ackn owledgm en t. Financial support from the Natu-
ral Sciences and Engineering Research Council (Canada)
in the form of a Research Grant (to D.E.W.) and a
Strategic Project Grant (to M.S.C.P. and D.E.W.) is
gratefully acknowledged. A.V. thanks the University
of Saskatchewan and the Universidad Nacional Au-
tonoma de Me´xico for scholarships.
Ab b r evia t ion s: N,N-bis(2-oxo-3-oxazolidinyl)phos-
phonic chloride (BOP-Cl); 1-(3-(dimethylamino)propyl)-
3-ethylcarbodiimide (EDC); 2-hydroxy-4-methylpentanoic
acid (Hmp); 1-hydroxybenzotriazole (HOBt); 2-hydroxy-
3-phenylpropanoic acid (Hpp); leucine (Leu); valine (Val).
(13) (a) Nitz, T. J .; Holt, E. M.; Rubin, B.; Stammer, C. H. J . Org.
Chem. 1981, 46, 2667-2673. (b) Arenal, I.; Bernabe, M.; Fernadez-
Alvarez, E. An. Quim., Ser. C 1981, 77, 56-62. (c) Makowski, M.;
Rzezotarska, B.; Kubica, Z.; Pietrzynski, G.; Hetper, J . Liebigs Ann.
Chem. 1986, 980-991.
(14) Wissner, A.; Grudzinskas, C. V. J . Org. Chem. 1978, 43, 3972-
3974.
(15) Chen, S.-T.; Wu, S.-H.; Wang, K.-T. Synthesis 1989, 37-38.
(16) (a) Makowski, M.; Rzeszotarska, B.; Kubica, Z.; Wieczorek, P.
Liebigs Ann. Chem. 1984, 920-928.(b) Smelka, L.; Rzeszotarska, B.;
Pietrznski, G.; Kubica, Z. Liebigs Ann. Chem. 1988, 485-486.
(17) A small amount of the corresponding (E)-isomer (ca. 5%) could
be isolated. A similar result was obtained starting from either the pure
(E)- or pure (Z)-isomer of 3.
(18) Schmidt, U.; Weinbrenner, S. J . Chem. Soc., Chem. Commun.
1994, 1003-1004.
(19) Cyclization of the (E)-isomer of 13 under the same conditions
gave a mixture of 1 and 2 in good yield.
Su p p or tin g In for m a tion Ava ila ble: Experimental pro-
cedures and characterization data for 1-14 (9 pages).
J O961661A
(20) (a) Mazur, R. H.; Pilipauskas, D. R. Pept.: Synth, Struct.,
Funct., Proc. Am. Pept. Symp., 7th; Rich, D. H., Gross, E., Eds.; Pierce
Chem. Co.: Rockford, IL, 1981; pp 81-84. (b) Mazur, R. H.; Pilipaus-
kas, D. R. Pept., Proc. Eur. Pept. Symp., 17th; 1982; Blaha, K., Malon,
P., Eds.; de Gruyter: Berlin, 1983; pp 319-322.
(21) Similar oxidation of the minor selenide isomer gave a 6:1
mixture of 1 and 2; oxidation of the other isomer gave only 2. Thus,
assuming a stereoselective syn elimination of the selenides (via the
corresponding selenoxides), the relative stereochemistry for major and
minor isomers should be lk (i.e., syn PhSe and NH groups) and ul for
the second major isomer. Phomalide and isophomalide are separable
(PTLC), albeit with considerable difficulty and with low efficiency (i.e.,
mainly mixed fractions are obtained).
(22) For a similar syn-elimination of sulfoxide derivatives of threo-
nine, see ref 8a. For other examples of dehydroamino acid syntheses
by selenoxide fragmentation, see: (a) Walter, R.; Roy, J . J . Org. Chem.
1971, 36, 2561-6563. (b) Reich, H. J .; J asperse, C. P.; Renga, J . M.
Ibid. 1986, 51, 2981-2988. (c) Hashimoto, K.; Sakai, M.; Okuno, T.;
Shirahama, H. Chem. Commun. 1996, 1139-1140.
(23) Synthetic phomalide was identical in all respects (1H and 13C
NMR, [R]D, TLC and HPLC retention times) with an authentic sample.
(24) Pedras, M. S. C.; Taylor, J . L.; Morales, V. M. Phytochemistry
1995, 38, 1215-1222.
(25) Pedras, M. S. C.; Se´guin-Swartz, G.; Abrams, S. R. Phytochem-
istry 1990, 29, 777-782.