Research Paper 659

Beech Leaf Colonization by the Endophyte Apiognomonia errabunda

Dramatically Depends on Light Exposure and Climatic Conditions
G. Bahnweg1, W. Heller1, S. Stich1, C. Knappe1, G. Betz1, C. Heerdt2, R. D. Kehr3, D. Ernst1, C. Langebartels1, 6,
A. J. Nunn4, J. Rothenburger1, 7, R. Schubert5, 8, P. Wallis1, 9, G. Müller-Starck5, H. Werner2, R. Matyssek4,
and H. Sandermann, Jr.1
1
GSF – National Research Centre for Environment and Health, Institute of Biochemical Plant Pathology, Ingolstädter Landstraße 1,
85764 Neuherberg, Germany
2
Technische Universität München, Chair of Bioclimatology, Department of Ecology, Am Hochanger 13, 85354 Freising, Germany
3
HAWK Hochschule für Angewandte Wissenschaft und Kunst, Fachhochschule Hildesheim/Holzminden/Göttingen,
Fakultät Ressourcenmanagement, Büsgenweg 1A, 37077 Göttingen, Germany
4
Technische Universität München, Chair of Ecophysiology of Plants, Department of Ecology, Am Hochanger 13, 85354 Freising, Germany
5
Technische Universität München, Department of Plant Sciences, Section of Forest Genetics, Am Hochanger 13, 85354 Freising, Germany
6
Present address: GSF – National Research Centre for Environment and Health, Scientific-Technical Department, Ingolstädter Landstraße 1,
85764 Neuherberg, Germany
7
Present address: Sanochemia Pharmazeutika AG, Landegger Straße 7, 2491 Neufeld, Austria
8
Present address: Biotechnology Group, University of Applied Sciences, Kuelzufer 2, 02763 Zittau, Germany
9
Present address: Department of Chemistry, Christopher Ingold Laboratories, University College London, 20 Gordon Street, London WC1H 0AJ, UK

Received: May 9, 2005; Accepted: September 22, 2005

Downloaded by: Universität Bochum. Copyrighted material.

Abstract: Ozone and light effects on endophytic colonization by
Apiognomonia errabunda of adult beech trees (Fagus sylvatica)
and their putative mediation by internal defence compounds
were studied at the Kranzberg Forest free-air ozone fumigation
site. A. errabunda colonization was quantified by “real-time
PCR” (QPCR). A. errabunda-specific primers allowed detection
without interference by DNA from European beech and several
species of common genera of plant pathogenic fungi, such as
Mycosphaerella, Alternaria, Botrytis, and Fusarium. Colonization
levels of sun and shade leaves of European beech trees exposed
either to ambient or twice ambient ozone regimes were deter-
mined. Colonization was significantly higher in shade compared
to sun leaves. Ozone exhibited a marginally inhibitory effect on
fungal colonization only in young leaves in 2002. The hot and
dry summer of 2003 reduced fungal colonization dramatically,
being more pronounced than ozone treatment or sun exposure.
Levels of soluble and cell wall-bound phenolic compounds were
approximately twice as high in sun than in shade leaves. Acylat-
ed flavonol 3-O-glycosides with putatively high UV-B shielding
effect were very low in shade canopy leaves. Ozone had only a
minor influence on secondary metabolites in sun leaves. It
slightly increased kaempferol 3-O-glucoside levels exclusively
in shade leaves. The frequently prominent hydroxycinnamic acid
derivative, chlorogenic acid, was tested for its growth inhibiting
activity against Apiognomonia and showed an IC50 of approxi-
mately 8 mM. Appearance of Apiognomonia-related necroses
strongly correlated with the occurrence of the stress metabolite,
3,3′,4,4′-tetramethoxybiphenyl. Infection success of Apiognomo-
nia was highly dependent on light exposure, presumably affect-
ed by the endogenous levels of constitutive phenolic com-
pounds. Ozone exerted only minor modulating effects, whereas
Plant Biol. 7 (2005): 659 – 669
© Georg Thieme Verlag KG Stuttgart · New York
DOI 10.1055/s-2005-872943
ISSN 1435-8603
climatic factors, such as pronounced heat periods and drought,
were dramatically overriding.
Key words: European beech (Fagus sylvatica), Apiognomonia er-
rabunda, endophyte, “real-time PCR”, plant phenolics, sun and
shade leaves, light, ozone, climate, precipitation.
Introduction
The ascomycete Apiognomonia errabunda (Rob.) Höhn. (ana-
morph: Discula umbrinella [Berk. and Broome] Sutton) and
closely related species are causative agents of leaf anthracnose
and affect a wide variety of plants. These include species of a
variety of broad-leaved tree genera, e.g., Castanea, Fraxinus,
Acer, Platanus, Juglans or Quercus, and, in particular, Fagus syl-
vatica (European beech), which is of major importance in Cen-
tral Europe. Although morphological characteristics of the fun-
gal pathogens and typical disease symptoms of anthracnose
(shoot and leaf blight or blotch) are fairly similar in different
tree species, the various anthracnose pathogens are host-spe-
cific. Among these, different closely related Apiognomonia spe-
cies have been distinguished using randomly amplified poly-
morphic DNA markers (Haemmerli et al., 1992).
A. errabunda was detected in leaves of almost any European
beech tree within a growing season, frequently in the absence
of disease symptoms (Petrini, 1991; Sinclair and Cernauskas,
1996; Stone et al., 2000). The fungus has been isolated not only
from leaves but also from buds, twigs and from bark of the
branch base, where it lives endophytically (Sieber and Hugen-
tobler, 1987; Kowalski and Kehr, 1996). It was reported to be
among the five most widespread and frequent endophytes of
beech bark in the Apennines (Danti et al., 2002) and it should
therefore be considered a constant companion of beech, capa-
ble of causing leaf necroses under certain conditions. The life
660 Plant Biology 7 (2005)
cycle of A. errabunda has been extensively described by More-
let (1989). Infection studies (Viret and Petrini, 1994) demon-
strated conclusively that colonization of beech leaves is initiat-
ed mainly by conidial infection, as postulated before by More-
let (1989) and later confirmed indirectly by Haemmerli et al.
(1992). On the other hand, systemic colonization of leaves
from infected wood parts close to the buds cannot be ruled
out since the fungus was additionally isolated from bud scales
and twig sections adjacent to the buds (Toti et al., 1993). Infec-
tion strongly depends on weather conditions, particularly pre-
cipitation. Wilson and Carroll (1994) reported that infection
levels of Discula quercina in Oregon (USA) depended heavily
on rainfall. When spring rains ceased, infection levels did not
increase any further.
Little is known about the effect of light and chronic ozone ex-
posure on endophytic colonization of European beech leaves
(Pronos et al., 1999; Matyssek and Sandermann, 2003). It has
been shown that biochemical and molecular stress responses
of trees towards ozone impact are generally similar to fungal
attack (Sandermann, 2000). Predisposition by ozone may,
therefore, affect fungal infection. Reduced as well as increased
susceptibility towards fungal attack upon ozone exposure has
been observed (Manning and von Tiedemann, 1995; Sander-
mann and Matyssek, 2004). The Kranzberg Forest experimen-
tal site (Pretzsch et al., 1998; Nunn et al., 2002) provided a
unique opportunity to study chronic long-term ozone effects
on the interaction of biochemical factors and fungal coloniza-
tion in beech. The study was conducted throughout four grow-
ing seasons, including an experimentally enhanced, chronic
ozone regime to the sun-exposed and the shaded canopy of
adult beech trees (free-air ozone exposure system: Werner
and Fabian, 2002). We report here on A. errabunda coloniza-
tion levels at Kranzberg Forest. Variations in the contents and
composition of phenolic compounds were also determined
and both their dependence on light exposure and influence on
fungal infestation will be discussed.
Based on previous results of studies with juvenile beech trees
in exposure chambers and greenhouses, we expected to see:
(1) a massive induction of defence-related compounds follow-
ing sub-lethal ozone exposure, including phenolic metabolites,
leading to (2) strongly reduced colonization of endophytic
Apiognomonia.
Materials and Methods
Study site
The study was conducted at the Kranzberg Forest research site
(near Freising, Southern Germany: 48825′08′′N, 11839′41′′E,
485 m a.s.l.: cf. Pretzsch et al., 1998; Nunn et al., 2002) in a
mixed 60-year-old stand (closed canopy) with about 27 –
30 m high European beech (Fagus sylvatica) and Norway
spruce (Picea abies) trees. Scaffolding and a research crane pro-
vided access to sun and shade crowns. Since May 2000, ten
neighbouring individuals (5 beech and 5 spruce trees) have
been subjected to free-air ozone fumigation within a canopy
volume of 2000 m3 throughout the growing seasons (Werner
and Fabian, 2002). Ozone was generated from oxygen-en-
riched air; fumigated canopy concentrations were monitored
online and adjusted in real time to twice ambient conditions.
A cut-off at 150 nl l–1 was implemented in order to prevent
G. Bahnweg et al.
acute ozone injury (Nunn et al., 2002; cf. Reich, 1987). Control
samples were taken from adjacent similar beech trees exposed
to ambient air (1 × O3 = control). The enhanced ozone exposure
(2 × O3) was released from a tubing system suspended across
the canopy.
The mean annual air temperature at Kranzberg Forest was
7.0 – 7.5 8C, and 14.5 – 15.0 8C during the growing seasons of
150 – 155 days. Precipitation values amounted to 730 – 790
and 410 – 520 mm, respectively (Pretzsch et al., 1998). Annual
depositions of total N and SO4 sulphur were 9.8 and 6.3 kg–1
a–1, respectively, and were close to corresponding averages
across Bavarian beech forests (Pretzsch et al., 1998).
Sampling design
The study reported here was restricted to beech, i.e., focusing
on the five individual trees of the 2 × O3 regime, and another
five trees from the 1 × O3 regime. Leaf samples for the analysis
of secondary metabolites were taken during the growing sea-
sons of 2000 through 2004. Leaves for fungal quantification
were collected at three, four, five and four dates of consecutive
Downloaded by: Universität Bochum. Copyrighted material.
years throughout the growing seasons. Typical Apiognomonia-
related necroses were observed occasionally in summer and
autumn, exclusively on shade leaves. Quantitative DNA deter-
minations were performed on five (years 2001, 2003, 2004)
or 50 (year 2002) independent samples from each of the four
factorial treatments, i.e., (1) sun leaves/1 × O3, (2) sun leaves/
2 × O3, (3) shade leaves/1 × O3, and (4) shade leaves/2 × O3.
Fungal strains
A. errabunda isolates were maintained on 2 % malt extract and
on oatmeal agar (Stevens, 1974). Three strains (A282, D264,
A268) were obtained from the culture collection of the Federal
Biological Research Centre for Agriculture and Forestry, Insti-
tute for Plant Protection in Forests, Braunschweig, Germany.
Another nine cultures were isolated from beech leaves collect-
ed near the GSF National Research Centre (Neuherberg/Mu-
nich, Germany) and at the Kranzberg Forest research site. All
of these were used for the development of specific A. errabun-
da PCR markers.
Analysis of phenolic metabolites (soluble and cell wall bound)
Five sampling campaigns were conducted in 2000, 2001 and
2003, and four in 2004, and 20 leaves from each of the 5 beech
trees of the 1 × O3 and 2 × O3 regimes (10 leaves from sun and
shade crowns each) were collected. Ten sun and shade leaves
per tree were separately pooled, frozen in liquid nitrogen and
stored at – 80 8C prior to further analysis. Extraction and quan-
tification of phenolic metabolites was adapted from Turunen
et al. (1999). One hundred milligrams of crushed leaves and
100 mg diatomaceous earth (Sigma, catalogue no. 5384) were
homogenized together for 2 min in a Micro-Dismembrator II
(B. Braun, Melsungen, Germany) at liquid nitrogen tempera-
ture. Aliquots of 60 mg of the frozen powder were extracted in
Eppendorf tubes with 450 μl methanol at 4 8C overnight. Tubes
were then briefly vortexed and finally centrifuged for 10 min at
11000 rpm in a bench top centrifuge. The supernatant was
transferred to another Eppendorf tube and both extract and
residue were stored at – 80 8C until further use.
Light and Climate Effects on Beech Leaf Colonization by Apiognomonia
Soluble phenolics were analyzed by reversed-phase high per-
formance liquid chromatography (RP-HPLC) on a Spherisorb
ODS II column (Type NC, 4.6 × 250 mm, 5 μm furnished with a
pre-column, Bischoff, Leonberg, Germany) on a Beckman HPLC
System Gold (Beckman-Coulter, Krefeld, Germany). Samples
(methanol-water, 75 : 25, v/v, cleared by 5 min centrifugation
at 11000 rpm, injection volume 10 μl) were eluted from the
column by applying a gradient of solvent A [980 ml H2O +
20 ml 5 % ammonium formate in formic acid (98 %)] and sol-
vent B [882 ml methanol (HPLC-grade) + 96 ml H2O + 20 ml
5 % ammonium formate in formic acid (98 %)], column temper-
ature was 20 8C. The gradient was as follows: 0 – 5 min: 0% B;
5 – 55 min: 0 – 50 % B (linear); 55 – 85 min: 50 – 100% B; 85 –
90 min: 100% B; 90 – 93 min: 100 – 0% B; 93 – 98 min: 0% B. De-
tection was at 300 nm with a UV/visible diode array detector
(Beckman-Coulter, model 168).
The residues of methanol extraction were washed three times
with 1 ml methanol and once with 1 ml H2O (add solvent, mix
sample on a Vortex mixer and leave for 30 min at room tem-
perature in the dark, centrifuge 4 min at 11000 rpm and dis-
card supernatant). Freshly prepared 150 μl 1 N NaOH contain-
ing 100 mM ascorbic acid and 0.2 % NaBH4 was added, the sam-
ple well mixed and left overnight at room temperature in the
dark. Then 150 μl 1.5 M formic acid were added, the sample
mixed again, 300 μl methanol added, mixed again and centri-
fuged for 5 min at 11000 rpm. The supernatant was transferred
to a fresh tube and stored at – 80 8C until further analysis. Me-
tabolites were analyzed by RP-HPLC using a slightly different
protocol from that described above for soluble metabolites:
aliquots of 20 μl were eluted using the following solvent gra-
dient: 0 – 5 min: 0% B; 5 – 45 min: 0 – 100% B (linear); 45 –
50 min: 100% B; 50 – 53 min: 100 – 0% B; 53 – 58 min: 0% B.
Detection was at 280 nm (Turunen et al., 1999).
Metabolite identification, assignment of structural classes
and quantification
Metabolites were identified by co-chromatography with refer-
ence compounds either obtained commercially or isolated ear-
lier from a natural source and characterized by NMR spectros-
copy. Structural classes were assigned based on the typical di-
ode array spectra of model compounds as follows.
Soluble metabolites: hydroxycinnamic acid derivatives based
on 4-coumaric acid and caffeic acid esters, flavanone deriva-
tives based on naringenin 7-O-glucoside, flavonol glycosides
based on kaempferol and quercetin 3-O-glucosides, acylated
flavonol glycosides based on 3′′,6′′-dicoumaroylkaempferol 3-
O-glucoside, and flavans based on catechin.
Cell wall-bound metabolites: these metabolites were similar-
ly assigned except for cis- and trans-4-coumaric acid that oc-
curred in unconjugated form after hydrolysis. Other hydroxy-
cinnamic acids, such as ferulic acid, were present only in negli-
gible amounts.
Mass equivalents were determined assuming monohexoside
structures, except for the cell wall-bound 4-coumaric acid iso-
mers and for flavans, which mostly occur as their aglycone.
Calculation was based on the peak integrals of HPLC diagrams
routinely recorded at 300 nm for soluble and 280 nm for cell
wall-bound metabolites.
Plant Biology 7 (2005) 661
DNA extraction and PCR amplification of ITS DNA fragments
Fungal DNA from cultures covering two thirds of the surface of
a petri dish (containing 20 ml liquid 2 % malt extract) was iso-
lated according to Möller et al. (1992). Beech DNA from leaves
of sun and shade crowns was isolated as described by Bahn-
weg et al. (1998). DNA for sequencing the European beech ri-
bosomal RNA genes’ (rDNA) internal transcribed spacer region
(ITS) was obtained from a beech cell culture (FS-D1) free from
bacteria and fungi (C. Langebartels). Amplification of fungal
and tree ITS regions was performed with ITS1 and ITS4 primers
(White et al., 1990), as described previously (Lee and Taylor,
1992; Schulze et al., 1997). Cloning and sequencing of DNA
fragments purified by agarose gel electrophoresis was carried
out as described by Bahnweg et al. (2000).
Apiognomonia-specific PCR primers and real-time PCR
Amplification of A. errabunda DNA by conventional PCR was
achieved in 25 μl reactions as described above (Bahnweg et
al., 2000), except that the ITS primers were replaced by the
specific forward (APIO-f2: 5′ to 3′ ACT CTT GTT TTT GTA ATA
Downloaded by: Universität Bochum. Copyrighted material.
TCA TCT) and reverse primer pair (APIO-r1: 5′ to 3′ GAG GTA
AAA TTA CTA CGC TCA A). An amplicon of 311 bp was generated
from Apiognomonia target DNA. Quantitative PCR (QPCR, real-
time PCR) was performed in 50 μl reactions with Absolute™
QPCR SYBR Green® ROX Mix (ABgene, Hamburg, Germany).
Apiognomonia-specific primers (10 μM each) APIO-f1 (5′ to 3′
TTT GTG AAT ACT ACC TAA AAT G) and APIO-r3 (5′ to 3′ AGA
TAT TAC AAA AAC AAG AGT) were used to generate short am-
plicons of 122 bp. QPCR was performed with the ABI-PRISM
7700 Sequence Detection System (Applera, Weiterstadt, Ger-
many). Thermocycler conditions were 50 8C/2 min (elimina-
tion of uracil-DNA), 95 8C/10 min (activation of enzyme), 40 cy-
cles of 95 8C/15 s, 60 8C/1 min (denaturation, annealing, and
amplification, respectively). A standard curve was generated
for each run with a dilution series of purified and calibrated
A. errabunda genomic DNA (200 ng/20 ng/2 ng/200 pg/20 pg/
2 pg/200 fg prepared in a 10 ng μl–1 λ-DNA solution, the latter
stabilizing highly diluted genomic DNA). Newly generated
double stranded DNA amplicon was quantified by measuring
the SYBR Green® fluorescence intensity after each 60 8C step
of the cycler programme. Analysis of experiments, i.e., deter-
mination of Ct values (cycle threshold) to quantify initial
amounts of target DNA, was done using the ABI-PRISM 7700
Sequence Detection System Software version 1.6 (Böhm et al.,
1999).
In vitro inhibition of Apiognomonia by chlorogenic acid
and 3,3′,4,4′-tetramethoxybiphenyl
An inoculum of A. errabunda was prepared from mycelium
grown on a malt agar plate blended with 30 ml liquid malt ex-
tract (2 %) for 15 s at low speed in a Waring blender. The myce-
lial fragment suspension was adjusted to 200 ml with liquid
malt extract. Thirteen ml aliquots were then adjusted to 0,
0.2, 0.4, 0.8, 1.5, 2.0, 3.0, 3.5, 4.0, and 4.5 mg ml–1 chlorogenic
acid by adding suitable volumes of 150 mg ml–1 stock solution
of this compound in 50 % (v/v) ethanol. Ethanol adjusted to fi-
nal concentrations of 1.7 % in the growth medium was slightly
growth promoting. One ml aliquots of each of these mixtures
were then distributed into 12 2-ml wells of sterile 24-well re-
action plates and incubated at 20 8C in the dark. Four replicates
662 Plant Biology 7 (2005)
per chlorogenic acid concentration were harvested after 4, 5.5,
and 7 days, washed with distilled water and frozen until fur-
ther analysis. Biomass was estimated as glucose equivalents
by the method of Dubois et al. (1956).
The 3,3′,4,4′-tetramethoxybiphenyl experiment required the
following modification due to the low solubility of the com-
pound. A stock solution of 13 mg in 840 μl ethanol and 160 μl
DMSO was prepared at 50 8C by ultrasound treatment for
5 min. Suitable volumes of this solution to obtain final con-
centrations of 0, 5, 10, 20, 40, 60, 80, 100, 120, 150, 180, and
200 μg ml–1 were added to 10-ml aliquots of liquid medium in-
cubated at 35 8C and adjusted to final concentrations of 200 μl
solvent mixture. Three ml of mycelium preparation per metab-
olite concentration were finally added at incubation condition,
and 1-ml aliquots were then distributed into 12 2-ml wells of
sterile 24-well reaction plates and further incubated, as de-
scribed, for chlorogenic acid. Mycelial growth was followed
visibly.
Statistical analysis
Where indicated, the Student’s t-test was used to test for sta-
tistically significant differences between treatment means.
Results
Nucleotide sequencing of ITS regions of A. errabunda
and F. sylvatica
Development of species-specific primers utilizes differences in
base sequences of different organisms. ITS sequences between
ribosomal RNA genes (rRNA) are particularly useful as they
display sufficient sequence variation. Furthermore, rRNA oc-
curs in multiple copies per genome thus enhancing sensitivity
of detection.
Genomic DNA was extracted from different A. errabunda iso-
lates and from a beech cell culture. DNA fragments spanning
the internal transcribed spacer between the 18S and 26S ribo-
somal RNA genes (and including the 5.8S rDNA gene) were
amplified with ITS1 and ITS4 universal primers (White et al.,
1990), cloned and sequenced. Sequences were aligned and
found to be identical for three A. errabunda isolates, whereas
the beech sequence was quite different, as expected. The beech
ITS sequence was 149 bp longer than the corresponding Apio-
gnomonia sequence with a similarity of only 36 %.
Sequences of the A. errabunda and beech ITS region were sub-
mitted to the EMBL Library and are available under the acces-
sion numbers AJ888473 (F. sylvatica), AJ888475 (A. errabunda;
strain A282), AJ888476 (A. errabunda; strain D264), AJ888477
(A. errabunda; strain A268).
PCR-based detection assay for A. errabunda
The oligonucleotides 5′ to 3′: ACT CTT GTT TTT GTA ATA TCA
TCT (APIO-f2) and 5′ to 3′: GAG GTA AAA TTA CTA CGC TCA A
(APIO-r1) were derived from the Apiognomonia ITS nucleotide
sequence and selected to establish a diagnostic DNA marker.
The chosen primer pair yielded the predicted 311 bp amplicon
with all isolates of A. errabunda investigated in this study, but
not with target DNA of beech or any of a collection of sapro-
G. Bahnweg et al.
phytic and parasitic fungi tested, i.e., Penicillium sp., Aspergil-
lus sp., Alternaria spp., Fusarium spp., Botrytis cinerea, Mycos-
phaerella punctiformis, and Phytophthora spp.
The oligonucleotides APIO-f1 (5′ to 3′: TTT GTG AAT ACT ACC
TAA AAT G) and APIO-r3 (5′ to 3′: AGA TGA TAT TAC AAA AAC
AAG AGT) were also derived from the Apiognomonia ITS nu-
cleotide sequence and selected to establish diagnostic DNA
markers to be used in QPCR. This primer pair yielded the pre-
dicted 122 bp amplicon with target DNA of all isolates of A. er-
rabunda examined, but not of beech or any of the above-men-
tioned collection of saprophytic and parasitic fungi.
The sensitivity of the QPCR assay was determined with a dilu-
tion series (in triplicate) of pure A. errabunda DNA ranging
from 20 fg to 20 ng. A linear dynamic range was obtained over
five orders of magnitude, from 200 fg to 20 ng.
Climate and O3 regimes of the four-year study period
Precipitation at Kranzberg Forest in 2001 and 2002 was 1091
and 1015 mm, respectively, and was rather high compared to
Downloaded by: Universität Bochum. Copyrighted material.
the following two years (Fig. 1) as well as to the long-term
average of 730 – 790 mm (Pretzsch et al., 1998). The year
2003, however, was extraordinarily dry (558 mm), whereas
2004 was within the long-term range (786 mm) at this site.
Nonetheless, 2004 revealed outstandingly high precipitation
in January and rather evenly distributed precipitation at re-
duced levels throughout the rest of the year. Minimal precipi-
tation occurred at the onset of the growing season (Fig. 1).
With regard to the temperature regime, 2004 can be consid-
ered an average year, whereas the mean temperature in 2003
of 9.1 8C was well above the long-term average of approx 8 8C,
although January and February were very cold.
In summary, the year 2003 was considered extraordinarily hot
and dry. As a consequence of extremely hot and clear days in
late spring and summer, background ambient ozone levels
were rather high during the growing season (Fig. 2).
Levels of A. errabunda infestation
A. errabunda DNA was detectable in all leaf samples through-
out the four growing seasons studied. Overall amounts were
up to 1000 pg per mg fresh leaf mass. DNA preparations from
A. errabunda mycelium grown in liquid medium yielded, on
average, 10 μg DNA per 100 mg fresh weight. Thus, 100 pg fun-
gal DNA in 1 mg of fresh leaf mass of beech are equivalent to
1 μg mycelium or 0.1 % (leaf fresh mass related). Colonization
levels above 100 pg fungal DNA per 1 mg fresh leaf mass were
often accompanied by typical necroses during summer and au-
tumn.
In 2001 and 2002, striking differences in A. errabunda infesta-
tion were observed between sun and shade leaves, i.e., A. erra-
bunda DNA levels detected were almost 10 times higher in
shade leaves throughout the growing season (Figs. 3, 4). Differ-
ences between the two O3 regimes were not as pronounced,
and were clearly detectable only in spring and summer of
2002. Both sun and shade leaves at 2 × O3 gave lower infesta-
tion levels than the corresponding leaves of non-fumigated
trees in June (Figs. 3, 4). In late summer and autumn, however,
this putative “early protective” effect of enhanced O3 levels
Light and Climate Effects on Beech Leaf Colonization by Apiognomonia
disappeared, in particular in sun-exposed leaves (Fig. 4). A. er-
rabunda was hardly detectable in either sun or shade leaves
throughout the year 2003 (Figs. 3, 4). It can be assumed that
the extraordinarily hot and dry weather conditions prevented
successful proliferation of the fungus. Fungal populations had
indeed been reduced to such negligible levels that even the
moderate weather conditions of 2004 were not sufficient by
themselves for recovery (Figs. 3, 4). This was partly attributed
to lack of inoculum size (sporangia on leaf litter from the pre-
vious year) as a consequence of the poor fungal performance in
2003. Furthermore, it could also be due to the relatively dry
spring of 2004 which may have hampered spore germination.
Plant Biology 7 (2005) 663
Fig. 1 Monthly sums of precipitation and
means of temperatures at the Kranzberg For-
est site from 2001 to 2004.
Downloaded by: Universität Bochum. Copyrighted material.
Fig. 2 Course of ambient and twice ambient
ozone, based on daily means at the Kranzberg
Forest site from 2001 to 2004.
Quantitative fungal DNA data were corroborated by the occur-
rence of necrotic Apiognomonia-related lesions. Lesions were
rare in 2003 and 2004, whereas in 2001 and 2002 massive in-
festation had led to the formation of extensive, visible symp-
toms and early leaf loss of some branches of the shade canopy.
Light- and ozone-related effects on secondary metabolite levels
More than 15 soluble and 10 cell wall-bound phenolics were
detected in beech leaves at Kranzberg Forest. Distinct differ-
ences were found between sun and shade leaves. A typical ex-
ample of such an HPLC diagram is shown in Fig. 5 (August
664 Plant Biology 7 (2005) G. Bahnweg et al.

Downloaded by: Universität Bochum. Copyrighted material.

Fig. 3 A. errabunda DNA levels (pg mg–1 leaf fresh weight) in shade leaves of beech at Kranzberg Forest site from 2001 through 2004. Asterisks
indicate significantly different pairs of means at p ≤ 0.05.

Fig. 4 A. errabunda DNA levels (pg mg–1 leaf fresh weight) in sun leaves of beech at Kranzberg Forest site from 2001 through 2004. Asterisks
indicate significantly different pairs of means at p ≤ 0.05.

2002). Peak heights may differ to some extent among indi-
vidual trees and at different sampling dates. Compounds pro-
tective against UV-B radiation (flavonol and acylated flavonol
derivatives, F and A) were almost completely absent in shade
leaves. In contrast, no significant influence of ozone was de-
tectable in either sun or shade leaves (Fig. 6; years 2001,
2003, and 2004). Levels of soluble phenolics were highest in
spring but gradually decreased throughout the growing sea-
son. Only one newly formed metabolite, 3,3′,4,4′-tetrameth-
oxy-1,1′-biphenyl (Fig. 9), was observed, which was identified
by comparison with a chemically prepared sample character-
ized by UV and NMR spectroscopy, and mass spectrometry.
The compound appeared within a distance of 1 – 2 mm from
the margin of typical necrotic areas and occurred throughout
the necroses (Fig. 9). When necrotic leaf lesions were initiated
by application of the ethylene precursor ACC, the metabolite
was produced throughout the leaf (Nunn et al., 2005). Cell
wall-bound kaempferol derivatives were also more prominent
in sun than in shade leaves, while differences between the
ozone regimes were absent (Fig. 7; year 2000). On the other
hand, kaempferol glycoside levels concomitantly increased,
suggesting deposition of the soluble fraction into cell walls
during leaf ontogeny. This pattern was confirmed by data ob-
tained from the other years (not shown).
Light and Climate Effects on Beech Leaf Colonization by Apiognomonia
Fig. 6 Annual courses (three years) of total phenolics in sun and shade leaves of beech at Kranzberg Forest under the influence of ambient and
2 × ambient ozone.
In vitro inhibition experiments with A. errabunda and
chlorogenic acid, and 3,3′,4,4′-tetramethoxy-1,1′-biphenyl
Mycelial growth of A. errabunda was measured in liquid malt
medium in the presence of chlorogenic acid (Fig. 8), the com-
mercially available isomer of the caffeoylquinic acids observed
in beech leaves (see Fig. 5). Chlorogenic acid was tested at con-
centrations of up to 4.5 mg ml–1 (12.7 mM), mimicking values
measured in leaves of the sun canopy at Kranzberg Forest. Sig-
nificant inhibition (50 % reduction of fungal biomass) was ob-
served near the highest concentration of about 3.5 mg g–1 fresh
weight detected in the leaf tissue (Fig. 8). On the other hand,
the stress metabolite 3,3′,4,4′-tetramethoxy-1,1′-biphenyl had
no effect on hyphal growth of A. errabunda nor on two sapro-
phytic fungi, Aspergillus niger and Penicillium notatum at con-
Plant Biology 7 (2005) 665
Fig. 5 Representative HPLC diagrams of
methanol extracts from shade and sun leaves
of beech trees collected at Kranzberg Forest
in August 2002. Slight differences in peak
heights may occur with individual trees and
at different sampling times. H = hydroxycin-
namic acid derivative, Fl = flavanone deriva-
tive, F = flavonol glycoside, A = acylated flavo-
nol glycoside, H1= neo-chlorogenic acid, H2 =
chlorogenic acid, F1 = quercetin 3-O-galacto-
side + quercetin 3-O-glucoside, F2 = kaemp-
ferol 3-O-galactoside, F3 = kaempferol 3-O-
glucoside, F4 = kaempferol 3-O-rhamnoside,
A1 = 2′′,4′′-di-p-coumaroylkaempferol 3-O-
arabinopyranoside, A2= 2′′,4′′-di-p-coumaro-
ylkaempferol 3-O-rhamnopyranoside.
Downloaded by: Universität Bochum. Copyrighted material.
centrations up to 1 mM, exceeding that of 0.25 mM on a dry
weight basis present in vivo (average value for necroses). Since
the solubility of this compound was restricted in aqueous me-
dia owing to its highly lipophilic nature, ethanol and DMSO
were added as mediators at concentrations of 1.3 and 0.25 %
(v/v), which did not inhibit fungal growth.
Discussion
Apiognomonia errabunda is one of the most common fungal
companions of European beech (Butin, 1995; Danti et al.,
2002), occurring in leaves, bud scales, and the bark of twigs
and stems. Its infection and penetration strategies very closely
resemble those of certain biotrophic fungal pathogens (Viret
and Petrini, 1994). A. errabunda can indeed cause severe dis-
666 Plant Biology 7 (2005)
Fig. 7 Annual course (year 2000) of cell wall-bound kaempferol gly-
cosides in sun and shade leaves of beech at Kranzberg Forest under
the influence of ambient and 2 × ambient ozone.
ease symptoms (Fig. 9) occasionally leading to premature leaf
loss and die-back of young shoots of the tree (Butin, 1995). It is
the most prominent and widespread beech endophyte living a
secret life within beech tissues, mostly not causing disease
symptoms. Certain aspects of the ecological characteristics of
this fungus may be assumed to be a mutualistic interaction
with the tree, since infection-dependent necroses are appar-
ently also triggered by the presence of insect galls of Hartigiola
annulipes Htg. and Mikiola fagi Htg. and lead to premature gall
decline (Pehl and Butin, 1994). Similar interactions have been
described for the oak endophyte A. quercina and gall insect lar-
vae living in oak leaves (Butin, 1992). This view is in line with
the more general concept of endophytes bridging the gap be-
tween latent pathogens and mutualistic symbionts (Carrol,
1988).
Böhm et al. (1999) were the first to use real-time PCR (QPCR)
for high precision quantification of plant pathogens in plant
pathology with crops (Phytophthora infestans vs. Solanum tu-
G. Bahnweg et al.
berosum) as well as trees (Phytophthora citricola vs. Fagus syl-
vatica roots) using artificial inoculation techniques. In the
present study, QPCR was used for the first time to quantify
leaf endophytes of a major Central European forest tree spe-
cies in an ecological context. Infestation values between 1 and
1000 pg fungal DNA per mg tissue fresh weight were obtained
in both studies. Colonization of 800 single leaves was quanti-
fied in 2002 (summaries in Figs. 3, 4). A. errabunda was present
in virtually every leaf throughout the growing season.
The fungus obviously did not proliferate immediately upon
colonization, although a pathogen-like penetration of epider-
mal beech cells has been observed. Endophytic infections may
be limited to few or even a single cell (Stone, 1987), but it is dif-
ficult to determine the factors involved that keep fungi, among
them potential parasites, at bay. Preformed or elicited phenolic
compounds were considered good candidates as protectants
against fungal attack. Invading hyphae will presumably come
in contact with phenolics only on penetrating host cells by dis-
rupting cellular integrity. There is no information available on
the presence of phenolics in the apoplastic space of beech
leaves, where a direct interaction with invading fungi could oc-
Downloaded by: Universität Bochum. Copyrighted material.
cur. Challenging pure A. errabunda liquid cultures with chloro-
genic acid, a commercially available isomer of caffeoylquinic
acids present in beech leaves, indicated that these compounds
occurred in sufficient quantity, at least during spring and early
summer, and particularly in sun-exposed leaves (Fig. 8), to re-
duce fungal growth. Chlorogenic acid was chosen not only for
its availability but also because it was one of two phenolic me-
tabolites induced in leaves of beech saplings upon artificial in-
oculation with A. errabunda spores (Rothenburger and Bahn-
weg, unpublished). 3,3′,4,4′-Tetramethoxybiphenyl present in
tissues infected by A. errabunda was similarly tested but did
not exhibit any growth inhibition in this assay at concentra-
tions up to four times higher than those present in beech leaf
necroses. This is in agreement with the fact that high concen-
trations of A. errabunda DNA coincide with this metabolite,
which must then be considered a stress metabolite rather than
a phytoalexin. Zielke and Sonnenbichler (1990) suggested that
this compound may be a putative phytoalexin whose biologi-
cal activity should be further analyzed. No elicitation of the
Fig. 8 In vitro inhibition of Apiognomonia
mycelial growth by chlorogenic acid. Concen-
trations in the biotest ranged from 0 to
4.5 mg ml–1 (0 – 12.7 mM). Concentrations of
this compound in leaves ranged from 0.25
to 3.63 mg g–1 fresh weight, based on more
than 400 independent samples taken during
the growing seasons of the years 2000 and
2001. The arrow indicates the highest value
(in sun leaves) of chlorogenic acid observed
in this study.
Light and Climate Effects on Beech Leaf Colonization by Apiognomonia
Fig. 9 Shade leaf of European beech (August 2002) heavily infested
with Apiognomonia errabunda and showing typical necrotic lesions, ir-
regularly shaped and uneven in size and often delimited by leaf veins.
bulk phenolic compounds was observed even upon massive
fungal colonization, which is in good agreement with earlier
observations of Zielke and Sonnenbichler (1990).
Temporal and spatial patterns of infection showed that in-
oculum size, solar irradiation and cumulative rainfall promi-
nently influence the establishment of endophytic infections
(Wilson and Carroll, 1994; Wilson, 2000). Owing to adverse
conditions, individual infections may survive only for a short
period of time, leading to an overall reduction in infection fre-
quency. This was the case in 2003, when long periods of hot,
dry weather occurred. Apiognomonia beech leaf populations
at the Kranzberg Forest research site were then reduced to
negligible levels, and they did not recover during the following
year (2004) despite the moderate climate present, most likely
due to lack of an adequate inoculum size.
Plant Biology 7 (2005) 667
Downloaded by: Universität Bochum. Copyrighted material.
Parallel analyses of green leaf parts, borders of necroses (approx.
2 mm) and necroses indicated that 3,3′,4,4′-tetramethoxy-1,1′-biphe-
nyl occurred exclusively in heavily infected leaf parts.
Chronic ozone exposure has been shown to predispose plants
to subsequent fungal attack, either negatively by increasing
susceptibility or positively by making host plants more resist-
ant to fungal invasion (Manning and von Tiedemann, 1995). A
possible protecting effect was observed in the present study
only at the beginning of the growing season of 2002, where in-
creased ozone concentrations may have retarded beech leaf
colonization by A. errabunda, presumably through elicitation
of plant defence reactions. Weather conditions giving rise to
elevated tropospheric ozone concentrations (i.e., hot, dry, and
sunny days) usually do not coincide with conditions favour-
able for fungal propagation (i.e., cool and wet days). However,
a “memory effect” of ozone exposure, first described for Nor-
way spruce (Picea abies) and Scots pine (Pinus sylvestris) by
Langebartels et al. (1997, 1998) and Sandermann (2000), may
evoke persistent biochemical changes affecting the suscep-
tibility of host plants to fungal attack over a long period of
time. The putative ozone effect observed in 2002 was, at least
in spring and early summer, in accordance with one of our
668 Plant Biology 7 (2005)
expectations outlined in the introduction, i.e., hardening of
leaves against fungal invasion by sublethal chronic ozone fu-
migation. Considering (1) fungal infestation levels across the
four-year study period, (2) the dramatic influence of light
(sun-exposed versus shade leaves), and (3) hot and dry weath-
er conditions as observed in 2003, however, the ozone effect
observed was at best marginal with regard to retardation of
fungal colonization of leaves in the case of adult European
beech.
Although A. errabunda seems to be poorly equipped to over-
come the host tree’s defences, its latent but widespread occur-
rence in virtually all beech leaves investigated provides a cru-
cial advantage over soil-borne saprophytic fungi in litter deg-
radation. A. errabunda is present at the right place, inside the
beech leaves, ready to feed on the easily accessible substrates
of nutritional value, and the right time when these substrates
become available upon senescence, even prior to and upon leaf
shedding in late summer and autumn.
Acknowledgements
This study was supported by the Deutsche Forschungsgemein-
schaft (SFB 607).
References
Bahnweg, G., Schulze, S., Möller, E. M., Rosenbrock, H., Langebartels,
C., and Sandermann H. (1998) DNA isolation from recalcitrant ma-
terials such as tree roots, bark, and forest soil for the detection of
fungal pathogens by PCR. Analytical Biochemistry 262, 79 – 82.
Bahnweg, G., Schubert, R., Kehr, R. D., Müller-Starck, G., Heller, W.,
Langebartels, C., and Sandermann, H. (2000) Controlled inocula-
tion of Norway spruce (Picea abies) with Sirococcus conigenus:
PCR-based quantification of the pathogen in host tissue and infec-
tion-related increase of phenolic metabolites. Trees 14, 435 – 441.
Böhm, J., Hahn, A., Schubert, R., Bahnweg, G., Adler, N., Nechwatal, J.,
Oehlmann, R., and Oßwald, W. (1999) Real-time quantitative PCR:
DNA determination in isolated spores of the mycorrhizal fungus
Glomus mossae and monitoring of Phytophthora infestans and Phy-
tophthora citricola in their respective host plants. Journal of Phyto-
pathology 147, 409 – 416.
Butin, H. (1995) Tree Diseases and Disorders. Oxford: Oxford Univer-
sity Press.
Butin, H. (1992) Effect of endophytic fungi from oak (Quercus robur
L.) on mortality of leaf-inhabiting gall insects. European Journal
of Forest Pathology 22, 237 – 246.
Carrol, G. (1988) Fungal endophytes in stems and leaves: from latent
pathogen to mutualistic symbiont. Ecology 69, 2 – 9.
Danti, R., Sieber, T. N., and Sanguineti, G. (2002) Endophytic myco-
biota in bark of European beech (Fagus sylvatica) in the Apennines.
Mycological Research 106, 1343 – 1348.
Dubois, M., Gilles, K. A., Hamilton, J. K., Rebers, P. A., and Smith, F.
(1956) Colorimetric method for determination of sugars and relat-
ed substances. Analytical Chemistry 28, 350 – 356.
Haemmerli, U. A., Brändle, U. E., Petrini, O., and McDermott, J. M.
(1992) Differentiation of isolates of Discula umbrinella (teleo-
morph: Apiognomonia errabunda) from beech, chestnut, and oak
using randomly amplified polymorphic DNA markers. Molecular
Plant-Microbe Interactions 5, 479 – 483.
Kowalski, T. and Kehr, R. D. (1996) Fungal endophytes of living
branch bases in several European tree species. In Endophytic Fungi
in Grasses and Woody Plants – Systematics, Ecology and Evolution
(Redlin, S. C. and Carris, L. M., eds.), St. Paul, Minnesota: APS Press,
pp. 67 – 86.
G. Bahnweg et al.
Langebartels, C., Heller, W., Ernst, D., Lippert, M., Lütz, C., and Sander-
mann, H. (1997) Ozone responses of trees: results from controlled
chamber exposures in the GSF phytotron. In Forest Decline and
Ozone: A Comparison of Controlled Chamber and Field Experi-
ments, Ecological Studies No. 127 (Sandermann, H., Wellburn, A.
R., and Heath, R. L., eds.), Berlin: Springer Verlag, pp.163 – 200.
Langebartels, C., Heller, W., Führer, G., Lippert, M., Simons, S., and
Sandermann, H. (1998) “Memory effects” in the action of ozone
on conifers. Ecotoxicology and Environmental Safety 41, 62 – 72.
Lee, S. B. and Taylor, J. W. (1992) Phylogeny of five fungus-like proto-
ctisan Phytophthora species, inferred from the internal transcribed
spacer of ribosomal DNA. Molecular Biology and Evolution 9,
636 – 653.
Manning, W. J. and von Tiedemann, A. (1995) Climate change: Poten-
tial effects of increased atmospheric carbon dioxide (CO2), ozone
(O3), and ultraviolet-B (UV-B) radiation on plant diseases. Environ-
mental Pollution 88, 219 – 245.
Matyssek, R. and Sandermann, H. (2003) Impact of ozone on trees:
an ecophysiological perspective. Progress in Botany 64, 349 – 404.
Möller, E. M., Bahnweg, G., Sandermann, H., and Geiger, H. H. (1992)
A simple and efficient protocol for isolation of high molecular
weight DNA from filamentous fungi, fruit bodies, and infected
plant tissues. Nucleic Acids Research 20, 6115 – 6116.
Downloaded by: Universität Bochum. Copyrighted material.
Morelet, M. (1989) L’anthracnose des chênes et du hêtre en France.
Revue Forestière Française 41, 488 – 496.
Nunn, A. J., Reiter, I. M., Häberle, K.-H., Werner, H., Langebartels, C.,
Sandermann, H., Heerdt, C., Fabian, P., and Matyssek, R. (2002)
“Free-air” ozone canopy fumigation in an old-growth mixed for-
est: concept and observations in beech. Phyton 42, 105 – 119.
Nunn, A. J., Anegg, S., Betz, G., Simons, S., Kalisch, G., Seidlitz, H.,
Grams, T. E. E., Häberle, K.-H., Matyssek, R., Bahnweg, G., Sander-
mann, H., and Langebartels, C. (2005) Role of ethylene in the regu-
lation of cell death and leaf loss in ozone-exposed European beech.
Plant, Cell and Environment 28, 886 – 897.
Pehl, L. and Butin, H. (1994) Endophytische Pilze in Blättern von
Laubbäumen und ihre Beziehungen zu Blattgallen (Zoocecidien).
Mitteilungen der Biologischen Bundesanstalt für Land- und Forst-
wirtschaft, Berlin-Dahlem, Heft Nr. 297.
Petrini, O. (1991) Fungal endophytes of tree leaves. In Microbial Ecol-
ogy of Leaves (Andrews, J. H. and Hirano, S. S., eds.), New York:
Springer Verlag, pp.179 – 187.
Pretzsch, H., Kahn, M., and Grote, R. (1998) Die Fichten-Buchen-
Mischbestände des Sonderforschungsbereiches „Wachstum oder
Parasitenabwehr?“ im Kranzberger Forst. Forstwissenschaftliches
Centralblatt 117, 241 – 257.
Pronos, J., Merrill, L., and Dahlsten, D. (1999) Insects and pathogens
in a pollution-stressed forest. In Oxidant Air Pollution Impacts in
the Montane Forest of Southern California, Ecological Studies,
Vol. 134 (Miller, P. R. and McBride, J. M., eds.), Berlin, Heidelberg,
New York: Springer Verlag, pp. 317 – 336.
Reich, P. B. (1987) Quantifying plant response to ozone: a unifying
theory. Tree Physiology 3, 63 – 91.
Sandermann, H. (2000) Ozone/biotic disease interactions: molecular
biomarkers as a new experimental tool. Environmental Pollution
108, 327 – 332.
Sandermann, H. and Matyssek, R. (2004) Scaling up from molecular
to ecological processes. In Molecular Ecotoxicology of Plants, Eco-
logical Studies, Vol. 170 (Sandermann, H. ed.), Berlin: Springer Ver-
lag, pp. 207 – 226.
Schulze, S., Bahnweg, G., Möller, E. M., and Sandermann, H. (1997)
Identification of the genus Armillaria by specific amplification of
an rDNA-ITS fragment and evaluation of genetic variation within
A. ostoyae by rDNA-RFLP and RAPD analysis. European Journal of
Forest Pathology 27, 225 – 239.
Sieber, T. and Hugentobler, C. (1987) Endophytische Pilze in Blättern
und Ästen gesunder und geschädigter Buchen (Fagus sylvatica L.).
European Journal of Forest Pathology 17, 411 – 425.
Light and Climate Effects on Beech Leaf Colonization by Apiognomonia Plant Biology 7 (2005) 669

Sinclair, J. B. and Cernauskas, R. F. (1996) Latent infection vs. endo- G. Bahnweg

phytic colonization by fungi. In Endophytic Fungi in Grasses and GSF – National Research Centre for Environment and Health
Woody Plants – Systematics, Ecology and Evolution (Redlin, S. C. Institute of Biochemical Plant Pathology
and Carris, L. M., eds.), St. Paul, Minnesota: APS Press, pp. 3 – 29. Ingolstädter Landstraße 1
Stevens, R. B. (ed.) (1974) Mycology Guidebook. Seattle: University of 85764 Neuherberg
Washington Press. Germany
Stone, J. K. (1987) Initiation and development of latent infections by
Rhabdocline parkeri on Douglas-fir. Canadian Journal of Botany 65, E-mail: bahnweg@gsf.de
2614 – 2621.
Stone, J. K., Bacon, C. W., and White, J. F. (2000) An overview of endo- Editor: H. Rennenberg
phytic microbes: Endophytism defined. In Microbial Endophytes
(Bacon, C. W. and White, J. F., eds.), New York: Marcel Dekker,
pp. 3 – 29.
Toti, L., Viret, O., Horat, G., and Petrini, O. (1993) Detection of the en-
dophyte Discula umbrinella in buds and twigs of Fagus sylvatica.
European Journal of Forest Pathology 23, 147 – 152.
Turunen, M., Heller, W., Stich, S., Sandermann, H., Sutinen, M. L., and
Norokorpi, Y. (1999) The effects of UV exclusion on the soluble
phenolics of young Scots pine seedlings in the subarctic. Environ-
mental Pollution 106, 219 – 228.
Viret, O. and Petrini, O. (1994) Colonization of beech leaves (Fagus
sylvatica) by the endophyte Discula umbrinella (teleomorph: Apio-
gnomonia errabunda). Mycological Research 98, 423 – 432.

Downloaded by: Universität Bochum. Copyrighted material.

Werner, H. and Fabian, P. (2002) Free-air fumigation on mature trees:
a novel system for controlled ozone enrichment in grown-up
beech and spruce canopies. Environmental Science and Pollution
Research 9, 117 – 121.
White, T. J., Bruns, T., Lee, S., and Taylor, J. (1990) Amplification and
direct sequencing of fungal ribosomal RNA genes for phylogenet-
ics. In PCR Protocols: A Guide to Methods and Applications (Innis,
M. A., Gelfand, D. H., Sninsky, J. J., and White, T. J., eds.), San Diego:
Academic Press, pp. 315 – 322.
Wilson, D. (2000) Ecology of woody plant endophytes. In Microbial
Endophytes (Bacon, C. W. and White, J. F., eds.), New York: Marcel
Dekker, pp. 389 – 420.
Wilson, D. and Carroll, G. C. (1994) Infection studies of Discula quer-
cina, an endophyte of Quercus garryana. Mycologia 86, 635 – 647.
Zielke, H. and Sonnenbichler, J. (1990) Natural occurrence of 3,3′,4,4′-
tetramethoxy-1,1′-biphenyl in leaves of stressed European beech.
Naturwissenschaften 77, 384 – 385.