Endophytic and epiphytic phyllosphere fungi of ... - Mycologia

Mycologia, 100(3), 2008, pp. 387–391. DOI: 10.3852/07-110R1
# 2008 by The Mycological Society of America, Lawrence, KS 66044-8897
Endophytic and epiphytic phyllosphere fungi of Camellia japonica: seasonal and
leaf age-dependent variations
Takashi Osono1 Laboratory of Forest Ecology, Graduate School of
Agriculture, Kyoto University, Kyoto 606-8502, Japan
Abstract: Seasonal and leaf age-dependent variations
in the endophytic and epiphytic phyllosphere fungal
assemblages of Camellia japonica were examined and
compared. Live leaves of C. japonica were collected in
four seasons (May, Aug, Nov, Feb), and fungi were
isolated from healthy-looking leaves of 0, 1, 2 and 3 y
old. The infection rate and total number of endophytic
fungi increased May–Feb, and species richness
of endophytes increased as leaves aged. In contrast
the infection rate of epiphytic fungi was 100% for all
leaf ages at every sampling date. The total number of
epiphytic fungi isolated was greatest in May and
lowest in Aug. The species richness of epiphytes did
not differ significantly by season or leaf age. Eight
fungal species were recorded as major phyllosphere
fungi of C. japonica. Seasonal variations were detected
for the frequencies of Colletotrichum gloeosporioides, C.
acutatum, and epiphytes Pestalotiopsis sp.1, Aureobasidium
pullulans, Phoma sp.1 and Ramichloridium sp.,
whereas the frequency of the endophyte Geniculosporium
sp.1 varied with leaf age. The frequency of the
epiphyte Cladosporium cladosporioides varied with
both season and leaf age.
Key words: endophytes, epiphytes, fungal
assemblage, leaves, season
INTRODUCTION
The phyllosphere is the living leaf as a whole,
including the interior and surface (Carroll et al
1977), which provides habitats for a variety of
microorganisms. Phyllosphere fungi include endophytes
and epiphytes that colonize the interior or
surface of leaves, respectively (Petrini 1991). Seasonal
and leaf age-dependent variations in the endophytic
and epiphytic phyllosphere fungi are an important
aspect of phyllosphere ecology and have been
investigated on the leaves of evergreen (Ruscoe
1971, Cabral 1985, Hata et al 1998) and deciduous
trees (Breeze and Dix 1981, Wilson and Carroll 1994,
Sahashi et al 1999, 2000, Kaneko et al 2003). Some
studies have compared both endophytic and epiphyt-
Accepted for publication 20 March 2008.
1 Corresponding author. E-mail: fujijun@kais.kyoto-u.ac.jp
387
ic fungal assemblages on single leaves at the same
time (Wildman and Parkinson 1978, Legault et al
1989a, b, Osono 2002, 2007, Osono and Mori 2003,
2004, Santamariá and Bayman 2005), but only limited
information is available regarding the differences in
the seasonal and leaf age-dependent dynamics of
endophytes and epiphytes (Ruscoe 1971, Cabral 1985,
Osono and Mori 2005). Ruscoe (1971) reported that
both endophytes and epiphytes rarely varied with
season or leaf age. Conversely Cabral (1985) and
Osono and Mori (2005) found that both endophytes
and epiphytes varied with season and/or leaf age.
The purpose of this study was to examine and
compare the seasonal and leaf age-dependent variations
in the endophytic and epiphytic phyllosphere
fungal assemblages of Camellia japonica L. I investigated
(i) the infection rates, number of fungi
isolated, species richness of phyllosphere fungi and
(ii) the relative importance of season and leaf age to
the occurrence of major fungal species in the phyllosphere.
Camellia japonica is an evergreen broadleaf
tree and is one of the dominant trees in secondary
forests in western Japan.
MATERIALS AND METHODS
Study site.—The site was in the Oharano Forest Park of
Kyoto City (34u579N, 135u379E, 400 m a.s.l.), Japan. The
mean annual temperature is 15.3 C and the annual
precipitation is 1581.1 mm at the Kyoto Weather Station,
about 12 km from the study site. The site was in a
mountainous area of temperate forest, which was harvested
repeatedly for fuel production until the fuel revolution in
the 1950s. The site has been intact and undisturbed since
then (Ishikawa et al 2007). The plot, 20 3 10 m, was laid out
within a temperate secondary forest in which Quercus
serrata Murr., C. japonica and Pinus densiflora Sieb. et Zucc.
dominated in numbers and basal area. Most C. japonica do
not reach the forest canopy and constitute the subcanopy.
Sample collection.—Healthy-looking leaves of C. japonica
were collected from standing trees within the plot in May,
Aug and Nov 2004 and Feb 2005. On each sampling
occasion five branches that included leaves of the four age
classes (ages 0, 1, 2 and 3 y) were cut from five randomly
chosen trees, one branch per tree, at an approximate height
of 6 m. The ages of the leaves were determined by
examining the annual bud scars and counting the age from
the apex of the twigs. Two leaves were selected arbitrarily
from each age class within each branch, and a total of 40
leaves (five trees 3 one branch 3 four age classes 3 two
leaves) were collected on each sampling occasion. The

388 MYCOLOGIA
leaves were placed in paper bags and taken to the
laboratory.
Four leaf disks were punched from each sampled leaf, two
disks from the edge and another two from the central part
of the leaves, avoiding the primary vein, with a sterile cork
borer (5.5 mm diam). Two disks, one from the edge and
one from the central part of the leaf, were used to isolate
endophytic fungi after surface disinfection, and another two
disks were used to isolate epiphytic fungi after washing the
disks, as described below. Fungi were isolated from a total of
160 leaf disks on each sampling occasion within 24 h of
collection.
Fungal isolation and identification.—A surface disinfection
method (Kinkel and Andrews 1988, Hata et al 1998) and a
modified washing method (Harley and Waid 1955, Tokumasu
1996) were used according to Osono (2005). For
surface disinfection the leaf disks were submerged in 70%
ethanol (v/v) for 1 min to wet the surface, then surface
disinfected for 15 s in a solution of 15% hydrogen peroxide
(v/v) and then submerged again for 1 min in 70% ethanol.
The disks were rinsed with sterile, distilled water, transferred
to sterile filter paper in Petri dishes (9 cm diam) and
dried 24 h to suppress vigorous bacterial growth after
plating (Widden and Parkinson 1973). The disks were
placed on 9 cm Petri dishes containing LCA media (Miura
and Kudo 1970), two disks per plate. LCA contains glucose
0.1%, KH2PO4 0.1%, MgSO4.7H2O 0.02%, KCl 0.02%,
NaNO 3 0.2%, yeast extract 0.02% and agar 1.3% (w/v).
LCA was used because its low glucose content suppresses
the overgrowth of fast-growing species and because LCA
induces sporulation, which is useful for fungal identification
(Osono and Takeda 1999).
For modified washing disks were washed in a sterile test
tube and agitated in a vertical shaker 1.5 min to isolate
fungi growing on the surface. The disks were washed serially
in two changes of 0.005% aerosol-OT (di-2-ethylhexyl
sodium sulfosuccinate) solution (w/v) and rinsed four
times with sterile distilled water. The washed disks were
treated in the same manner as that used in the plating-out
procedure described for the surface disinfected leaves.
The plates were incubated at 20 C in the dark and
observed at 1, 4 and 8 wk after surface disinfection or
washing (Osono and Takeda 1999). Any hyphae or spores
on the plates were subcultured on fresh LCA, incubated and
identified to species or genus. Identification was based on
micromorphological observations, with reference to
Domsch et al (1980) and Ellis (1971, 1976).
Statistical analysis.—The infection rate was calculated as
the number of leaf disks with any fungus isolated divided by
the total number of disks (10) that were surface disinfected
or washed from each leaf position in each age class at each
sampling. The total number of fungi isolated and the total
number of species were recorded for the fungal assemblages
in each 10-disk set. The frequency of a single species was
calculated as the percentage of the number of disks
containing the species out of the 10 disks tested, which
had been either surface disinfected or washed, at each leaf
position in each age class at each sampling. Fungal species
were regarded as major species when they were isolated
from disks by either method at a frequency greater than
12.5% on any sampling occasion. Only the results for the
frequent species are shown in the present study.
Kruskal-Wallis test was used to evaluate differences in the
infection rate, total number of fungi isolated, number of
species of endophytes and epiphytes and the frequencies of
individual species among four leaf ages (0, 1, 2 or 3 y) and
four seasons (May, Aug, Nov or Feb). Mann-Whitney U test
was used to evaluate difference between two leaf positions
(edge or center). The analyses were performed on
Macintosh with Systat ver. 5.2 (Systat 1992).
Sørensen’s quotient of similarity (QS) was calculated to
examine the similarity of fungal assemblages in leaf
interiors and on leaf surfaces and to compare the similarity
with those previously reported in other host trees (Osono
and Mori 2004):
QS ~ 2a= ð2azbzcÞ where a is the number of common species and b and c
are the numbers of species specific to the interior and
the surface, respectively.
RESULTS
Phyllosphere fungal assemblages.—A total of 79 species
were isolated from C. japonica leaves—44 endophytic
species, 52 epiphytic species with 17 species common
to both habitats. Sørensen’s QS for the endophytic
and epiphytic fungal assemblages was 0.35.
Infection rate and total number of endophytic
fungi varied significantly with season, generally
increasing May—Feb (TABLE I). Species richness of
endophytes increased significantly as leaves aged
(TABLE I). In contrast the infection rates of epiphytic
fungi were 100% for all leaf ages at every sampling
date (TABLE I). Total number of epiphytic fungi
isolated varied significantly with season and was
greatest in May and lowest in Aug (TABLE I). Species
richness of epiphytes did not differ significantly by
season or leaf age (TABLE I). No significant differences
were found for the infection rate, the total number
of fungi isolated and the number of species of
endophytic and epiphytic fungi between leaf edge
and leaf center (TABLE I).
Variation in major species.—Eight fungal species were
regarded as major phyllosphere fungi (TABLE II). The
frequencies of Colletotrichum gloeosporioides (both
endophytic and epiphytic), Colletotrichum acutatum,
Pestalotiopsis sp.1, Cladosporium cladosporioides, Aureobasidium
pullulans, Phoma sp.1 and Ramichloridium
sp. varied significantly with season (TABLE II). Colletotrichum
gloeosporioides, Pestalotiopsis sp.1, Clad. cladosporioides
and Phoma sp.1 were frequent in May; A.
pullulans and Ramichloridium sp. were frequent in
Aug; C. gloeosporioides, C. acutatum, Pestalotiopsis sp.1,

TABLE I. Effects of season, leaf age, and leaf position on the infection rate, total number of fungi isolated, and species
richness of endophytic and epiphytic phyllosphere fungi in Camellia japonica. Data represent the mean values of leaf edge and
leaf center subsamples collected from 10 disks each
Season Leaf age
Clad. cladosporioides and A. pullulans were frequent
in Nov; and C. gloeosporioides, Pestalotiopsis sp.1, and
Clad. cladosporioides were frequent in Feb (TABLE II).
The frequencies of Geniculosporium sp.1 and Clad.
cladosporioides varied significantly with leaf age and
generally increased as leaves aged (TABLE II). No
significant differences were found for the frequencies
of major species between leaf edge and leaf center
(TABLE II).
DISCUSSION
Infection
rate (%)
This study investigated and compared the endophytic
and epiphytic phyllosphere fungi of C. japonica. The
finding that the total species number was greater on
the leaf surface (52) than in the interior (44) is
consistent with the results of previous studies (summarized
in Osono and Mori 2004). Sørensen’s QS of
0.35 for the endophytic and epiphytic fungal assemblages
is intermediate in the range (0.12–0.79) of
previous studies on forest tree leaves (summarized in
Osono and Mori 2004).
Koide et al (2005) found C. gloeosporioides and
Geniculosporium sp. 1 to be frequent endophytes of C.
japonica leaves at the same study site, supporting our
findings. Koide et al (2005) also isolated rhytismataceous
fungi as endophytes of C. japonica leaves, but in their
Endophytes Epiphytes
Total number
of fungi
Number
of species
Infection
rate (%)
Total number
of fungi
Number
of species
May 0 55 5.5 2.5 100 16.5 7.5
1 65 6.0 4.0 100 17.0 7.5
2 80 8.0 6.5 100 21.5 8.0
3 90 10.5 6.5 100 22.5 8.5
Aug 0 70 7.0 3.5 100 11.5 5.0
1 60 6.0 4.0 100 14.5 6.5
2 70 6.5 6.0 100 13.5 9.5
3 75 7.5 5.5 100 14.5 7.0
Nov 0 100 9.5 4.5 100 15.0 7.5
1 95 12.0 6.5 100 15.0 7.0
2 100 10.5 6.5 100 17.0 5.0
3 100 12.0 6.5 100 17.5 7.5
Feb 0 80 8.0 6.0 100 14.0 7.0
1 100 11.5 6.0 100 15.0 6.5
2 95 13.5 8.0 100 16.0 8.5
3 100 10.0 7.0 100 17.0 7.5
Probability Season ,0.001 0.002 0.09 na1 0.004 0.55
Age 0.37 0.29 0.01 na 0.11 0.37
Leaf position 0.34 0.73 0.59 na 0.83 0.86
1 Not applied.
OSONO: FUNGI OF CAMELLIA JAPONICA 389
study the frequency of occurrence of these fungi was
lower in live leaves than in newly shed leaves. The lack of
rhytismataceous fungi in the present study can be partly
due to the method of fungal isolation that might have
favored fast growing fungal species and to the relatively
large leaf disks we used to isolate fungi (5.5 mm diam).
Some species also might have been excluded in the
present study due to the 24 h delay in processing and
24 h period of drying disks before plating them.
Patterns of change in the infection rate, total
number of fungi isolated and species richness
differed between endophytes and epiphytes (TA-
BLE I), indicating that different patterns of fungal
succession occur in the two leaf habitats. Species
richness of fungal endophytes was low at leaf
emergence (i.e. on leaves of age 0 in May) and
increased as the leaves aged. A similar pattern of
colonization of the leaf interior by endophytic fungi
has been reported for Pseudotsuga menziesii (Mirb.)
Franco (Stone 1987) and Pinus spp. (Hata et al 1998).
In contrast colonization by epiphytic fungi began
shortly after leaf emergence because the infection
rate and species richness were already at or near their
maximum values in the youngest leaves sampled (age
0 in May). This colonization of leaf surfaces by
epiphytic fungi is consistent with the result of Osono
and Mori (2005).

390 MYCOLOGIA
TABLE II. Effects of season, leaf age, and leaf position on the frequency of occurrence of endophytic and epiphytic
phyllosphere fungi in Camellia japonica. Data represent the mean values of leaf edge and leaf center subsamples collected
from ten disks each
Season
Leaf
age
Colletotrichumgloeosporioides
The total number of endophytic fungi isolated was
lower in May and Aug than in Nov and Feb, whereas
that of epiphytic fungi was lower in Aug than in May
(TABLE I). This difference in seasonal pattern can be
partly related to the seasonal behavior of major
endophytes and epiphytes. The decrease of the
endophytes C. gloeosporioides and C. acutatum in
Aug (TABLE II) and the low infection by endophytes
in younger leaves in May (discussed above) can
account for the decrease of total number of endophytic
fungi in these seasons. The decrease of C.
gloeosporioides in Aug has been reported in deciduous
tree leaves in Japan (Terashita 1973). Similarly only
two of the common epiphytic species (A. pullulans
and Ramichloridium sp.) on C. japonica leaves
occurred more frequently in Aug, whereas the other
four common fungi occurred more often in May. The
higher incidence of A. pullulans and Ramichloridium
sp. in Aug might be related to the dark pigmentation
of hyphae in these species, thus conferring greater
competitive advantage over more lightly pigmented
species during hot, dry summer conditions (Butler
and Day 1998).
The occurrence of six of the eight major fungi of C.
japonica leaves varied with season, one endophyte
Endophytes Epiphytes
Geniculosporium
sp.1
ColletotriColletotrichumacuchumgloeostatumporioides Pestalotiopsis
sp.1
Cladosporiumcladosporioides
Aureobasidiumpullulans
Phoma
sp.1
Ramichloridium
sp.
May 0 40 0 0 10 0 25 25 35 0
1 25 5 5 10 15 55 0 50 0
2 20 10 0 25 40 70 25 20 0
3 20 15 0 25 50 60 15 30 0
Aug. 0 25 0 0 5 0 5 60 0 25
1 5 20 0 5 5 10 55 0 45
2 10 0 0 5 5 10 40 5 20
3 0 15 0 5 30 35 25 0 0
Nov. 0 30 5 30 40 25 10 35 5 5
1 50 5 25 10 50 15 40 0 0
2 35 15 10 30 50 45 35 0 5
3 55 5 0 20 80 25 10 0 0
Feb. 0 20 0 5 10 45 0 10 5 0
1 25 30 0 5 80 20 0 5 0
2 40 5 0 5 45 30 10 10 5
3 25 20 0 20 55 45 5 5 5
Probability Season 0.005 0.77 0.007 0.03 0.004 0.008 ,0.001 ,0.001 0.003
Age 0.92 0.04 0.20 0.44 0.06 0.006 0.31 0.95 0.48
Leaf
position
0.41 0.35 0.70 0.054 0.66 0.9 0.88 0.95 0.93
Kruskal-Wallis test was used to examine the effect of season and leaf age and Mann-Whitney test for the effect of leaf position.
varied with leaf age and one epiphyte with both
(TABLE II). Two studies have examined the effects of
season and leaf age on the occurrence of phyllosphere
fungi in evergreen broadleaf tree species.
Ruscoe (1971) recorded seven major phyllosphere
fungi of Nothofagus truncata, and one epiphyte
(Pestalotia funerea) showed seasonal variation, whereas
the other six fungi varied with neither season nor
leaf age. On Eucalyptus viminalis leaves (Cabral 1985)
two epiphytes (Clad. cladosporioides and Epicoccum
nigrum) and an endophyte (Zoellneria eucalypti)
showed seasonal variation, two endophytes (Alternaria
Alternaria alternata and Coccomyces maritinae) varied
with leaf age and one endophyte (Coniothyrium sp.)
varied with both season and leaf age. Comparisons
among previous studies and this study suggest that
seasonal variation might influence more frequently
than leaf age, but consistent patterns of phyllosphere
fungal colonization of evergreen leaves due to
seasonal and leaf age variation are difficult to predict.
For example Clad. cladosporioides and A. pullulans
showed different patterns of seasonal and leaf agedependent
variations on different hosts. More detailed
analyses of the seasonal and leaf age-dependent
changes in leaf environmental conditions might

provide further insights into the dynamics of endophytic
and epiphytic phyllosphere fungi on forest
trees.
ACKNOWLEDGMENTS
We thank Dr S. Tokumasu and Dr D. Hirose for their
helpful identification of fungi, Dr A. Mori for his comments
on statistical analysis and Ms K. Koide for her valuable
discussion. This study received partial financial support
from the Japanese Ministry of Education, Culture and
Sports (No. 14760099).
LITERATURE CITED
Breeze EM, Dix NJ. 1981. Seasonal analysis of the fungal
community on Acer platanoides leaves. Trans Br Mycol
Soc 77:321–328.
Butler MJ, Day AW. 1998. Fungal melanins: a review. Can J
Microbiol 44:1115–1136.
Cabral D. 1985. Phyllosphere of Eucalyptus viminalis:
dynamics of fungal populations. Trans Br Mycol Soc
85:501–511.
Carroll GC, Muller EM, Sutton BC. 1977. Preliminary
studies on the incidence of needle endophytes in some
European conifers. Sydowia 29:87–103.
Domsch KH, Gams W, Anderson TH. 1980. Compendium
of soil fungi. Volume I. Eching: IHW-Verlag. 860 p.
Ellis MB. 1971. Dematiaceous hyphomycetes. Oxon: CAB
International. 608 p.
———. 1976. More dematiaceous hyphomycetes. Oxon:
CAB International. 507 p.
Harley JL, Waid JS. 1955. A method of studying active
mycelia on living roots and other surfaces in the soil.
Trans Br Mycol Soc 38:104–118.
Hata K, Futai K, Tsuda M. 1998. Seasonal and needle agedependent
changes of the endophytic mycobiota in
Pinus thunbergii and Pinus densiflora needles. Can J
Bot 76:245–250.
Ishikawa H, Osono T, Takeda H. 2007. Effects of clearcutting
on decomposition processes in leaf litter and
the nitrogen and lignin dynamics in a temperate
secondary forest. J For Res 12:247–254.
Kaneko R, Kakishima M, Tokumasu S. 2003. The seasonal
occurrence of endophytic fungus, Mycosphaerella buna,
in Japanese beech, Fagus crenata. Mycoscience 44:277–
281.
Kinkel LL, Andrews JH. 1988. Disinfestation of living leaves
by hydrogen peroxide. Trans Br Mycol Soc 91:523–528.
Koide K, Osono T, Takeda H. 2005. Colonization and lignin
decomposition of Camellia japonica leaf litter by
endophytic fungi. Mycoscience 46:280–286.
Legault D, Dessureault M, Laflamme G. 1989a. Mycoflore
des aiguilles de Pinus banksiana et Pinus resinosa I.
Champignons endophytes. Can J Bot 67:2052–2060.
———, ———, ———. 1989b. Mycoflora of Pinus bank-
OSONO: FUNGI OF CAMELLIA JAPONICA 391
siana and Pinus resinosa needles II. Epiphytic fungi.
Can J Bot 67:2061–2065.
Miura K, Kudo M. 1970. An agar-medium for aquatic
hyphomycetes. Trans Mycol Soc Japan 11:116–118.
Osono T. 2002. Phyllosphere fungi on leaf litter of Fagus
crenata: occurrence, colonization, and succession. Can
J Bot 80:460–469.
———. 2005. Colonization and succession of fungi during
decomposition of Swida controversa leaf litter. Mycologia
97:589–597.
———. 2007. Endophytic and epiphytic phyllosphere fungi
of red-osier dogwood (Cornus stolonifera) in British
Columbia. Mycoscience 48:47–52.
———, Mori A. 2003. Colonization of Japanese beech leaves
by phyllosphere fungi. Mycoscience 44:437–441.
———, ———. 2004. Distribution of phyllosphere fungi
within the canopy of giant dogwood. Mycoscience 45:
161–168.
———, ———. 2005. Seasonal and leaf age-dependent
changes in occurrence of phyllosphere fungi in giant
dogwood. Mycoscience 46:273–279.
———, Takeda H. 1999. A methodological survey on
incubation of fungi on leaf litter of Fagus crenata.
Appl For Sci Kansai 8:103–108.
Petrini O. 1991. Fungal endophytes of tree leaves. In:
Andrews JH, Hirano SS., eds., eds. Microbial ecology of
leaves. New York: Springer Verlag. p 179–197.
Ruscoe QW. 1971. Mycoflora of living and dead leaves of
Nothofagus truncata. Trans Br Mycol Soc 56:463–474.
Sahashi N, Kubono T, Miyasawa Y, Ito S. 1999. Temporal
variations in isolation frequency of endophytic fungi of
Japanese beech. Can J Bot 77:197–202.
———, Miyasawa Y, Kubono T, Ito S. 2000. Colonization of
beech leaves by two endophytic fungi in northern
Japan. For Pathol 30:77–86.
Santamariá J, Bayman P. 2005. Fungal epiphytes and
endophytes of coffee leaves (Coffea arabica). Microb
Ecol 50:1–8.
Stone JK. 1987. Initiation and development of latent
infections by Rhabdocline parkeri on Douglas-fir. Can J
Bot 57:2800–2811.
Systat.. 1992. Statistics, version 5.2. Evanston: Systat Inc.
Terashita T. 1973. Studies of an anthracnose fungus on
broad-leaved trees in Japan, with special reference to
the latency of the fungus. Bull Gov For Exp Sta 252:1–
85. (in Japanese with English abstract)
Tokumasu S. 1996. Mycofloral succession on Pinus densiflora
needles on a moder site. Mycoscience 37:313–
321.
Widden P, Parkinson D. 1973. Fungi from Canadian
coniferous forest soils. Can J Bot 51:2275–2290.
Wildman HG, Parkinson D. 1978. Microfungal succession
on living leaves of Populus tremuloides. Can J Bot 57:
2800–2811.
Wilson D, Carroll GC. 1994. Infection studies of Discula
quercina, an endophyte of Quercus garryana. Mycologia
86:635–647.