Biosci. Biotechnol. Biochem.

, 69 (8), 1515–1519, 2005

Production of Phleichrome by Cladosporium phlei as Stimulated

by Diketopiperadines of Epichloe typhina
Yoshiya S ETO,1 Yasunori K OGAMI,1 Tadayuki S HIMANUKI,2 Kosaku T AKAHASHI,1
Hideyuki M ATSUURA,1; y and Teruhiko Y OSHIHARA1
1
Laboratory of Bioorganic Chemistry, Division of Applied Bioscience, Graduate School of Agriculture,
Hokkaido University, Sapporo 060-8589, Japan
2
National Agricultural Research Center for Hokkaido Region, Sapporo 062-8555, Japan

Received February 23, 2005; Accepted May 6, 2005

Epichloe typhina is an endophytic fungus, while
Cladosporium phlei is a pathogenic fungus of the timothy
plant (Phleum pretense L.). We found two activities in
the culture filtrate of E. typhina: one stimulated the
pathogenic fungus, C. phlei, to produce phleichrome
and the other inhibited its growth. The active ingre-
dients that stimulated the production of phleichrome
and inhibited the growth of C. phlei were isolated and
characterized. The isolated compounds were identified
as cyclo-(L-Pro-L-Leu) and cyclo-(L-Pro-L-Phe), which
were stimulatory compounds, and p-hydroxybenzalde-
hyde, which was the growth inhibitory compound, based
on an analysis of their spectral data. Of the two
stimulatory compounds, cyclo-(L-Pro-L-Phe) showed
higher activity. However, when 500 g of cyclo-(L-Pro-
L-Phe) was spotted on the TLC plate for bio-autogra-
phy, a growth inhibitory zone was identified in the
central red region, which contained phleichrome. On
the other hand, phleichrome showed antifungal activity
against E. typhina in the light, so it is assumed that there
might be antagonism between the endophytic fungus,
E. typhina, and the pathogenic fungus, C. phlei.
Epichloe thyphina produces a peculiar cat-tail-shaped
stroma in the upper part of the stalk which is referred to
as a choke. We have found many kinds of antifungal
substances from chokes, although these compounds
were not found in the aerial parts of healthy timothy
plants.5–7)
In order to explain this beneficial mutual relationship
between the fungus and the host plant, we next
examined the metabolites of E. thyphina and found
two different phenomena in the culture broth of
E. thyphina: one stimulated C. phlei to produce phlei-
chrome (Fig. 1), while the other inhibited the growth of
C. phlei. Phleichrome was first isolated from the culture
broth of C. phlei as a phytotoxic compound,8) and has
also been reported as an antimicrobial compound.9) We
report here the structural elucidation of these active
compounds and describe their biological activities in

Key words: endophyte; Epichloe typhina; Cladosporium

phlei; phleichrome; diketopiperazine

A mutual relationship exists between Acremonium

spp. and its host pasture grass.1) For instance, the mutual
relationship between Acremonium lolie and perennial
ryegrass (Lolium perenne L.) is associated with the
build-up of resistance to insects.1) This phenomenon is
considered to be due to the presence of feeding
deterrents2) whose biosynthesis is accelerated by the
infection of ryegrass with Acremonium lolie. In the case
of timothy (Phleum pretense L.), plants infected by the
choke disease fungus, Epichloe thyphina (Pers. ex Fr.)
Tul., whose imperfect state is Acremonium typhinum,3)
are resistant to leaf spot disease caused by the pathogen,
C. phlei (Gregory) de Vries.4) When infecting timothy, Fig. 1. Structure of Phleichrome.

y
To whom correspondence should be addressed. Tel: +81-11-706-2495; Fax: +81-11-706-2505; E-mail: matsuura@chem.agr.hokudai.ac.jp
1516 Y. S ETO et al.
order to explain an antagonistic relationship between
the endophytic fungus, E. typhina, and the pathogenic
fungus, C. phlei.
Material and Methods
General. FDMS and EI-HR-MS data were recorded
by a JMS-O1SG-2 instrument, and NMR spectra were
recorded by a JNM-EX 270 FT-NMR system (1 H at
270 MHz; 13 C at 67.5 MHz) (Jeol, Osaka, Japan).
Column chromatography was conducted with silica gel
60 (Kanto Chemical), Capcell Pak C18,
20 mm

250 mm (Shiseido) and Wakosil 5C18,
4:0
250 mm
(Wako).
Incubation of E. typhina. E. typhina was grown in
500 ml Erlenmeyer flasks, which contained 200 ml of a
potato dextrose medium (200 g/l of potato, 10 g/l of
peptone, 10 g/l of malt extract, 10 g/l of yeast extract).
The fungus was grown in the dark for 50 days at 25  C.
Isolation of compounds 1, 2 and 3. The isolation of
each active compound was monitored by TLC bio-
autography.10) The culture broth of E. thyphina (3.6-
liter) was filtered, and then the filtrate was partitioned
with ethyl acetate. (3.6-liter
3). After evaporating, the
residue of the ethyl acetate layer (1.67 g) was subjected
to silica gel column chromatography (200 g). The
column was eluted with 50% EtOAc/n-hexane (frs. E1
and E2, 500 ml each), EtOAc (fr. E4, 1000 ml), 20%
MeOH/CHCl3 (frs. E5 and E6, 500 ml each) and 50%
MeOH/CHCl3 (fr. E7, 1000 ml). Fraction E5 (406.8 mg)
was subjected to silica gel column chromatography
(50 g). The column was eluted with 3% MeOH/CHCl3
(frs. E5-1–E5-9, each 35 ml), 10% MeOH/CHCl3 (frs.
E5-10–E5-12, each 100 ml), and 20% MeOH/CHCl3 (fr.
E5-13, 300 ml). Fractions E5-5 and E5-6 were mixed,
concentrated, and purified by HPLC (Capcell Pak C18
column, Shiseido; MeOH:H2 O = 1:1; flow rate, 5 ml/
min; monitoring at A254 ) to give a crude fraction and
compound 2 (tR ¼ 18:8 min, 38.1 mg, ½25 D 93:9

(EtOH, c 0.10)). The crude fraction was purified by
HPLC (CAPCELL PAK C18 column, Shiseido;
MeOH:H2 O = 4:6; flow rate, 5 ml/min; monitoring at
A220 ) to give compound 1 (tR ¼ 24:6 min, 11.7 mg, ½25 D
124:206 (EtOH, c 0.13)). Fraction E1 was subjected
to silica gel column chromatography. The column was
eluted with CHCl3 (fr. E1-1, 600 ml), 5% MeOH/CHCl3
(frs. E1-2–41, 15 ml each) and MeOH (fr. E42, 600 ml).
Fractions E1-2–E1-22 were concentrated, subjected to
PTLC (CHCl3 ), and then purified by HPLC (Capcell Pak
C18 column, Shiseido; MeOH:H2 O = 3:1; flow rate,
5 ml/min; monitoring at A220 ) to give 3 (tR ¼ 11:2 min,
2.6 mg).
Spectral data for compounds 1, 2 and 3.
Cyclo-(L-Pro-L-Leu) (1). EI-HR-MS m=z (Mþ ): calcd.
for C11 H18 N2 O2 , 210.1369; found, 210.1387, ½25
D
124:2 (EtOH, c 0.13). 1 H-NMR (CDCl3 ) H : 4.12
(1H, t, J ¼ 8:1 Hz), 4.02 (1H, dd, J ¼ 3:7, 9.5 Hz), 3.57
(2H, m), 2.35 (1H, m), 1.7–2.2 (7H, m), 1.52 (1H, ddq,
J ¼ 4:9, 9.5, 14.3 Hz), 1.00 (3H, d, J ¼ 6:6 Hz), 0.96
(3H, d, J ¼ 6:6 Hz).
Cyclo-(L-Pro-L-Phe) (2). EI-HR-MS m=z (Mþ ): calcd.
for C14 H16 N2 O2 , 244.1213; found, 244.1219. ½25 D
93:9 (EtOH, c 0.10). 1 H-NMR (CDCl3 ) H : 7.29
(5H, m), 5.69 (1H, s), 4.28 (1H, dd, J ¼ 3:8, 10.7 Hz),
4.08 (1H, t, J ¼ 7:1 Hz), 3.60 (3H, m), 2.79 (1H, dd,
J ¼ 10:7, 16.2 Hz), 2.33 (1H, m), 1.98 (3H, m).
p-Hydroxybenzaldehyde (3). EI-HR-MS m=z (Mþ ):
calcd. for C7 H6 O2 , 122.0368; found, 122.0371. 1 H-
NMR (CDCl3 ) H : 9.88 (1H, s), 7.81 (2H, d, J ¼
8:7 Hz), 6.94 (2H, d, J ¼ 8:7 Hz).
Identification of phleichrome. TLC bio-autography
with C. phlei yield a red spot which was scraped off and
extracted with acetone. The extract was subjected to
HPLC (Wakosil 5C18, Wako; MeOH:H2 O = 8:2; flow
rate, 0.7 ml/min; monitoring at A474 ), and the tR value
was compared with that of authentic phleichrome
isolated in our laboratory.7) The tR value for phlei-
chrome in the extract matched that of the authentic
compound. FD-HR-MS for phleichrome derived from
the extract gave m=z 550.1827 (phleichrome, calcd. for
C30 H30 O10 , 550.1839), and the UV spectrum had the
characteristic absorbance of phleichrome at 474 and
584 nm.7)
Preperation of the spores of C. phlei. C. phlei was
incubated on a V-8 agar medium (V-8 juice, 200 ml/l;
CaCO3 , 3 g/l; agar, 20 g/l) for 2–4 weeks at 20  C. The
fungal mycelia were scraped from the slant and
suspended in a culture medium (A:B = 2:1:A = V-8
juice, 163 ml; CaCO3 , 1.53 g; H2 O, 245 ml; B = 30%
glucose in H2 O). The suspension was filtered through
gauze, and the filtrate was used for TLC bio-auto-
graphy.
Bio-autography. A portion of the tested fraction was
developed on a TLC plate, and then the solution
containing the spores of C. phlei was sprayed on to
the plate. The TLC plate was incubated at 20  C for 3
days.
Fungicide disc assay. Phleichrome (1 mg) in DMSO
(100 ml) was added to 20 ml of PDA. A PDA plate
containing only DMSO (100 ml) was used as the control.
The test fungus, E. typhina, grown on PDA plate
(
¼ 10 mm) was used as an inoculum and put on to
the control and test plates. After 10 days of incubation at
20  C under the either darkness or light (white fluores-
cent light, 8W, NEC, 15 cm distance), the mycelial
growth diameter was measured (average, n ¼ 2), and the
percentage of inhibition with respect to the control was
calculated according to the following formula:
 Production of Phleichrome by Cladosporium phlei as Stimulated by Diketopiperadines
Inhibition (%)
¼ ½ðgrowth diameter of untreated control
 growth diameter with treatmentÞ
100
=½growth diameter of untreated control
Result and Disscussion
The extraction and isolation of the active compounds
was monitored by TLC bio-autography. The culture
broth of E. typhina (3.6 l) was purified by a series of
chromatographic techniques to afford 1 and 2 as the
stimulatory compounds, and 3 as the growth inhibitory
compound. The structures of compounds 1, 2 and 3,
were determined to be (3S-cis)-hexahydro-3-isobutyl-
pyrrolo[1,2-a]pyrazine-1,4-dione, cyclo-(L-Pro-L-Leu),
(3S-cis)-hexahydro-3-benzyl-pyrrolo[1,2-a]pyrazine-1,4-
dione, cyclo-(L-Pro-L-Phe), and p-hydroxybenzalde-
hyde, respectively, by comparing their data with those
reported for authentic specimens (Fig. 2).11,12) There
was a possibility that these diketopiperazines could have
originated from the culture medium, so we checked the
activity of the culture medium which showed no
Fig. 2. Structures of Compounds 1, 2 and 3.
1517
activity. This clearly demonstrated that these diketopi-
perazines were derived from E. typhina. Compound 3 is
well known to have antifunal acivity,13,14) and in our
case, this compound showed antifungal activity at a dose
of 20 mg on a TLC plate.
In order to check the biological activities of the
diketopiperazines (1 and 2), we carried out a TLC bio-
autography using C. phlei at doses of 20, 50, 100, 200,
300, and 500 mg of each isolated compound. The color of
the spots changed to red after incubation (Fig. 3). Each
red spot was scraped off and extracted with acetone, and
the quantity of phleichrome in each spot was determined
by measuring the UV absorbance at 474 nm. The amount
of phleichrome of each spot was evaluated as shown in
Fig. 4. A dose of 500 mg of 2 on the TLC plate resulted
in inhibitory zones in the central area of the red spot
(Fig. 3 a), which indicating that C. phlei could not
survive on the silica gel containing diketopiperazine at
the highest concentration. This fungicidic activity of
diketopiperazine is consistent with that of previously
reported data.15,16) This result indicates that these
compounds stimulated C. phlei to produce a red spot,
but that an excess of compound 2 inhibited the growth of
C. phlei.
1518 Y. S ETO et al.

Fig. 3. TLC Bio-Autography Result for Compounds 1, 2.

Lane 1 is for 20, 50, 100, 200, 300 and 500 mg of compound 1, and lane 2 is for same doses of compound 2. A red spot was identified on the
TLC plate, and the inhibitory zone (a) is shown on the spot of 500 mg of 2 (a).

Fig. 4. Production of Phleichrome Stimulated by Compounds 1 and 2.

The amount of phleichrome was quantified after TLC bio-autography with 20, 50, 100, 200, 300 and 500 mg of compounds 1 ( ) and 2 ( ).
Each value is expressed as the mean  standard deviation from three results.

The red spot which was observed on the bio- Table 1. Growth Inhibitory Effect of Phleichrome against E. typhina
autography TLC plate (Fig. 3) was identified as phlei-
chrome by the tR value from HPLC, FD-HR-MS, and Spot diameter (mm) Inhibition
UV absorbance. Control 50 mg/ml of phleichrome (%)
It has already been reported that phleichrome re- Dark 33.5 26.5 20.9
pressed the growth of microorganisms living on timothy
Light 31.0 11.5 62.9
leaves,9) and the mechanism of inhibition was deduced
to involve a photodynamic effect.17,18) However, E. ty-
phina was not used as the test fungus, so we checked the
antifungal activity of phleichrome against E. thyphina diketopiperazine (2) induced the death of C. phlei, but,
under darkness and light. We carried out a fungicide disc in contrast, that a lower dose of diketopiperazines
assay19) using phleichrome at the concentration of stimulated C. phlei to produce phleichrome which had a
50 mg/ml. Pleichrome inhibited the growth of E. typhina growth inhibitory effect against E. thyphina in the light.
by 20.9% in the dark and by 62.9% in the light Based on these results, it is suggested that diketopiper-
(Table 1), indicating that phleichrome had antifungal azines induced stress in C. phlei, and the production of
activity against E. typhina, and that light was important phleichrome by C. phlei is thought to be a self-defense
for this activity. system of this fungus. It is assumed that there might be
We verified in this study that an overdose of an antagonistic relationship between the endophytic
 Production of Phleichrome by Cladosporium phlei as Stimulated by Diketopiperadines
fungus, Epichloe thyphina, and the pathogenic fungus,
C. phlei. However, more than 20 mg of diketopiperazines
on the leaves is thought to be impractical situation. In
order to establish the antagonistic mechanism in detail,
we need further experiments to confirm the mechanism;
for example, an analysis of the real contents of diketo-
piperazines in a timothy plant infected by E. thyphina,
and the signaling pathway of E. thyphina and the
timothy plant when infected by E. typhina and attacked
by C. phlei.
Acknowledgments
We thank Mr. K. Watanabe and Dr. E. Fukushi in our
faculty for measuring the MS data.
References
1) Siegel, M. R., Latch, G. C. M., and Johnson, M. C.,
Acremonium fungal endophytes of tall fescue and
perennial ryegrass: Significance and control. Plant
Disease, 69, 179 (1985).
2) Roman, D. D., Hunt, M. B., and Gaynor, D. L.,
Peramine, a novel insect feeding deterrent from ryegrass
infected with the endophyte Acremonium loliae. J.
Chem. Soc., Chem. Commun., 12, 935–936 (1986).
3) Zhi-Qiang, A. N., Siegel, M. R., Hollin, W., Huei-Fung,
T., Schmidt, D., and Schardl, C. L., Relationships among
non-Acremonium sp. fungal endophytes in five grass
species. Appl. Environ. Microbiol., 59, 1540–1548
(1993).
4) Shimanuki, T., Studies on the mechanisms of the
infection of timothy with purple spot disease caused by
Cladosporium phlei. Res. Bull. Hokkaido Natl. Agric.
Exp. Stn., 148, 1–56 (1987).
5) Yoshihara, T., Togiya, S., Koshino, H., Sakamura, S.,
Shimanuki, T., Sato, T., and Tajimi, A., Three fungitoxic
cyclopentanoid sesquiterpenes from stroma of E. typhi-
na. Tetrahedron Lett., 26, 5551–5554 (1985).
6) Koshino, H., Togiya, S., Yoshihara, T., Sakamura, S.,
Shimanuki, T., Sato T., and Tajimi, A., Four fungitoxic
C-18 hydroxy unsaturated fatty acids from stroma of
E. typhina. Tetrahedron Lett., 28, 73–76 (1987).
7) Koshino, H., Togiya, S., Yoshihara, T., Sakamura, S.,
Shimanuki, T., Sato, T., and Tajimi, A., New fungitoxic
sesquiterpenoids, Chokol A–G, from stroma of E. typhi-
1519
na and the absolute configuration of Chokol E. Agric.
Biol. Chem., 53, 789–796 (1989).
8) Yoshihara, T., Shimanuki, T., Araki, T., and Sakamura,
S., Phleichrome; A new phytotoxic compound produced
by C. phlei. Agric. Biol. Chem., 39, 1683–1684 (1975).
9) Araki, T., and Shimanuki, T., Phleichrome, a nonhost
specific toxin produced by the causal organism of
timothy purple spot, C. phlei, and its toxic effect on
timothy leaf blades and leaf surface microorganisms.
Nippon Sochi Gakkaishi, 28, 427–431 (1983).
10) Homans, A. L., and Fuchs, A., Direct chromatography in
thin-layer chromatography as a method for detecting
fungitoxic substances. J. Chromatogr., 51, 311 (1982).
11) Jayatilake, G. S., Thornton, M. P., Leonard, A. C.,
Grimwade, J. E., and Baker, B. J., Metabolites from
Antarctic sponge-associated bacterium, Pseudomonas
aeruginose. J. Nat. Prod., 59, 293–296 (1996).
12) Bodylev, M. M., Bodyleva, L. I., and Strobel, G. A.,
Synthesis and bioactivity of analogs of Maculosin, a
host-specific phytotoxin produced by Alternaria alter-
nate on spotted knapweed (Centaurea maculosa). J.
Agric. Food Chem., 44, 3960–3964 (1996).
13) Broneman, W. S., Akin, D. E., and VanEseltine, W. P.,
Effect of phenolic monomers on ruminal bacteria. Appl.
Environ. Microbiol., 52, 1331–1339 (1986).
14) Bell, A. A., 4-Hydroxybenzaldehyde and vanillin as
toxins formed in leaf wound sap of Phaseolus lunatus.
Phytopathology, 60, 161–165 (1970).
15) Wang, Y., Mueller, U. G., and Glardy, J., Antifungal
diketopiperazines from symbiotic fungus of fungus-
growing ant Cyphomyrmex minutus. J. Chem. Ecol., 25,
935–939 (1999).
16) Magnusson, J., Ström, K., Roos, S., Sjögren, J., and
Shnürer, J., Broad and complex antifungal activity
among environmental isolates of lactic acid bacteria.
FEMS Microbiol. Lett., 219, 129–135 (2003).
17) Macri, F., and Vianello, A., Photodynamic activity of
compounds structurally related to cercosporin. Agric.
Biol. Chem., 44, 2967–2970 (1980).
18) Rnone, A., Assante, G., Caronna, T., Modugno, V. D.,
and Nasini, G., Comparative evaluation of photodynamic
efficiency of some natural quinonoid fungal toxins.
Phytochemistry, 27, 1669–1674 (1988).
19) He, G., Matsuura, H., Takushi, T., Kawano, S., and
Yoshihara, T., A new antifungal metabolite from
Penicillium expansum. J. Nat. Prod., 67, 1084–1087
(2004).