Joshee et al Podocarp endophytes

This article appeared in a journal published by Elsevier. The attached
copy is furnished to the author for internal non-commercial research
and education use, including for instruction at the authors institution
and sharing with colleagues.
Other uses, including reproduction and distribution, or selling or
licensing copies, or posting to personal, institutional or third party
websites are prohibited.
In most cases authors are permitted to post their version of the
article (e.g. in Word or Tex form) to their personal website or
institutional repository. Authors requiring further information
regarding Elsevier’s archiving and manuscript policies are
encouraged to visit:
http://www.elsevier.com/copyright

Author's personal copy
Diversity and distribution of fungal foliar endophytes
in New Zealand Podocarpaceae
Sucheta JOSHEEa,b
, Barbara C. PAULUSa
, Duckchul PARKa
, Peter R. JOHNSTONa,
*
a
Landcare Research, Private Bag 92170, Auckland 1142, New Zealand
b
BioDiscovery New Zealand, 24 Balfour Rd, Parnell, Auckland 1052, New Zealand
a r t i c l e i n f o
Article history:
Received 5 March 2008
Received in revised form
24 March 2009
Accepted 9 June 2009
Published online 17 June 2009
Corresponding Editor: Barbara Schulz
Keywords:
Dacrycarpus dacrydioides
Dacrydium cupressinum
Host specialisation
Kunzea ericoides
Multivariate analysis
Podocarpus totara
Prumnopitys ferruginea
a b s t r a c t
The diversity and distribution of fungal endophytes in the leaves of four podocarps (Dacry-
dium cupressinum, Prumnopitys ferruginea, Dacrycarpus dacrydioides, and Podocarpus totara, all
Podocarpaceae) and an angiosperm (Kunzea ericoides, Myrtaceae) occurring in close stands
were studied. The effects of host species, locality, and season on endophyte assemblages
were investigated. Host species was the major factor shaping endophyte assemblages.
The spatial separation of sites and seasonal differences played signiﬁcant but lesser roles.
The mycobiota of each host species included both generalist and largely host-specialised
fungi. The host-specialists were often observed at low frequencies on some of the other
hosts. There was no clear evidence for family-level specialisation across the Podocarpaceae.
Of the 17 species found at similar frequencies on several of the podocarp species, 15 were
found also on Kunzea. Many of the endophytes isolated appear to represent species of fungi
not previously recognised from New Zealand.
ª 2009 The British Mycological Society. Published by Elsevier Ltd. All rights reserved.
Introduction
Fungal endophytes live for all or a major part of their life cycle
within the healthy tissues of their host without causing any
symptoms of disease (Wilson 1995). Endophytes have been
isolated from almost all plants studied so far; bryophytes
(Davis et al. 2003), ferns (Petrini et al. 1992), monocotyledons
(Taylor et al. 1999), conifers (Carroll et al. 1977; Carroll & Carroll
1978; Petrini 1986; Petrini & Carroll 1981; Petrini & Mu¨ ller 1979),
and various dicotyledons (Arnold et al. 2000; Johnston 1998),
from a wide range of habitats, such as coastal mangroves
(Kumaresan & Suryanarayanan 2001), temperate evergreen
forests (Espinosa-Garcia & Langenheim 1990), xeric regions
(Suryanarayanan et al. 2003), and tropical forests (Arnold
et al. 2000), and even from lichens (Petrini et al. 1990). Endo-
phytic fungi colonise all plant parts such as roots, stems,
leaves, bark and ﬂoral organs and in some cases can affect
both ecological and physiological processes of their host (Pet-
rini 1991; Schulz et al. 1999). However, biologically they are
diverse, and may include saprophytic and pathogenic fungi
with an ‘endophytic’ phase (Carroll 1988; Rodriguez & Redman
1997; Wilson 2000).
A highly diverse endophytic mycota has been demon-
strated in leaves of numerous hosts (e.g. Arnold et al. 2000;
Fro¨hlich et al. 2000; Gamboa & Bayman 2001) although high en-
dophytic diversity may not be universal. For example, studies
in the dry forests of Tamil Nadu indicated a considerably
lower species richness (Suryanarayanan et al. 2003). The
* Corresponding author.
E-mail address: johnstonp@landcareresearch.co.nz
journal homepage: www.elsevier.com/locate/mycres
m y c o l o g i c a l r e s e a r c h 1 1 3 ( 2 0 0 9 ) 1 0 0 3 – 1 0 1 5
0953-7562/$ – see front matter ª 2009 The British Mycological Society. Published by Elsevier Ltd. All rights reserved.
doi:10.1016/j.mycres.2009.06.004

distribution of endophytes seems to be governed by a variety
of host and environmental factors. A degree of host prefer-
ence has been observed for endophytic communities both in
temperate (e.g. Petrini 1996; Petrini & Fisher 1988, 1990) and
tropical plant species (e.g. Arnold et al. 2001). This host prefer-
ence may be expressed at various levels of the taxonomic hi-
erarchy, including at the level of plant family (Petrini & Carroll
1981), subgenus (Carroll & Carroll 1978), and species (Petrini &
Fisher 1990).
In addition to host-related factors, spatial heterogeneity has
been reported for fungal endophytic assemblages at different
scales. While endophyte assemblages in a particular host may
be similar throughout its continuous range (e.g. in Sequoia sem-
pervirens; Rollinger & Langenheim 1993) major differences ap-
pear when hosts are from geographically disjunct areas (Fisher
et al. 1994; Taylor et al. 1999). Other studies have also demon-
strated differences in species composition at smaller spatial
scales, for example at the level of sites of the same region
(Arnold et al. 2001; Vujanovic & Brisson 2002), between trees of
the same stand (Petrini et al. 1992; Johnston 1998), between
branches of the same tree (Espinosa-Garcia & Langenheim
1990), and even withinareas of leaves (Johnston & Fletcher 1998).
The effect of seasonal changes on endophyte assemblages
remains unclear. Seasonal patterns have been detected
among taxa that were not host specific (Widler & Mu¨ ller
1984), but other studies could not confirm any seasonality
among endophytes (e.g. Sieber & Hugentobler 1987).
The Podocarpaceae are a notable group of ancient conifers
with a largely Southern Hemisphere distribution (Hill & Bro-
dribb 1999). They are widespread in New Zealand where
they are represented by eight genera and at least 18 species,
all of which are endemic (Connnor & Edgar 1987; Molloy
1995, 1996). The species sampled in this study include several
of New Zealand’s most important canopy trees, Dacrycarpus
dacrydioides, Dacrydium cupressinum, Prumnopitys ferruginea,
and Podocarpus totara.
The aim of the present study is to explore the diversity and
distribution of fungal endophytes associated with members of
the Podocarpaceae with special reference to the effects of host
species, geographic range and season. In particular, we ask
whether fungal endophytes are more similar in comparisons
between members of the Podocarpaceae than in comparisons
between podocarps and an angiosperm, Kunzea ericoides (Myr-
taceae). Host-specific endophytes can occur on morphological
similar trees growing in a close stand with the host (Kowalski
& Kehr 1996). Hence, K. ericoides was selected as it has leaves of
similar size and grows in the vicinity of these podocarps. Spa-
tial relationships are also explored across sites within the
same region and at a more distant site. In addition, sampling
was undertaken in summer and winter to provide a prelimi-
nary indication of seasonality among endophytes.
Materials and methods
Site and host characteristics
For our study, we selected four sites from the Waitakere
Ranges near Auckland and one close to the Urewera National
Park in the central North Island (Fig 1). The Waitakere Ranges
extend about 25 km north to south and 25 km west of central
Auckland in New Zealand. The topography is the result of
ancient volcanic eruptions and lava flows. The vegetation
typically comprises native forest in various stages of regener-
ation after extensive logging and farming in the last century.
The tea-tree species (Leptospermum scoparium and Kunzea eri-
coides) are the main early succession plants, being replaced
by kauri (Agathis australis), tanekaha (Phyllocladus trichoma-
noides), rimu (Dacrydium cupressinum), totara (Podocarpus
totara), miro (Prumnopitys ferruginea), kahikatea (Dacrycarpus
dacrydioides), hinau (Elaeocarpus dentatus), and rewarewa
(Knightia excelsa) after about 100 years (Esler & Astridge
1974). D. cupressinum is the most common emergent, while
shrubs like Myrsine australis, Geniostoma ligustrifolium, Cop-
rosma spp., Melicytus ramiflorus, and Pseudopanax arboreum,
along with tree ferns, form the understorey. At the fifth
site, further to the south, the podocarp–hardwood forest
has been highly modified by logging. The sampling site com-
prises a dense stand of regenerating D. cupressinum and
Beilschmiedia tawa, along with smaller shrubs and scattered
seedlings of the other podocarp species, with scattered
stands of K. ericoides around the margins. Seedlings and sap-
lings of the four podocarp species we sampled co-occurred in
close stands.
The four podocarps species we sampled demonstrate vari-
able leaf morphology. D. cupressinum has imbricate leaves
(4–7 Â 0.5–1 mm), D. dacrydioides has bilaterally flattened
leaves in juvenile stages (3–7 Â 0.5–1 mm) that become imbri-
cate with age, P. totara has larger bifacially flattened leaves
(15–30 Â 3–4 mm) that are helically arranged, and P. ferruginea
has bifacially flattened leaves (15–30 Â 2–3 mm) with petioles
twisted to arrange them in a two dimensional plane along
the shoot axis. The leaves of K. ericoides are also flattened,
about 4–12 Â 1–2 mm in size. Most of the leaves of K. ericoides
remain attached for two years, although a few are lost within
that time. All of the podocarp species sampled retain all of
their leaves for more than two years.
Sampling and isolation strategy
Sampling was carried out in the winter months (July–August
2004) and the following summer (January 2005). Two trees
per site were sampled for each of the five host species, and
the same trees were sampled in each season. For each sea-
son, a total of 2500 leaves (10 000 leaf segments) were sam-
pled; this represents 500 leaves (or 2000 leaf segments) for
each tree species. Five branches were arbitrarily selected
and cut from each tree, kept cool in polythene bags until
they were brought to the laboratory. Isolations were mostly
undertaken within 4–6 h of collection, always within 12 h.
From each branch, ten healthy leaves from the previous sea-
son’s growth were carefully excised and surface sterilised
(1 min in 95 % ethanol, 3 min in bleach solution (approxi-
mately 14 % free chlorine), 30 s in 95 % ethanol, then rinsed
in sterile distilled water). Leaves of Dacrydium cupressinum,
Dacrycarpus dacrydioides, and Kunzea ericoides were cut into
four pieces. For the larger leaves of Podocarpus totara and
Prumnopitys ferruginea, four separate slices approximately 1–
2 mm Â 3–4 mm were taken from along the length of the
leaf. The leaf pieces were plated on Petri dishes containing
1004 S. Joshee et al.

1.25 % malt extract (Difco) and 2 % agar. The inoculated
plates were incubated in daylight at room temperature. Petri
dishes were checked regularly for the growth of fungi for one
month. Fungi growing from the leaf pieces were subcultured
onto potato dextrose agar, oatmeal agar (Difco) and/or water
agar, sometimes amended with small pieces of sterile host
tissue in an attempt to stimulate sporulation. The endo-
phytes were grouped into morphospecies, based on cultural
appearance together with conidial and ascospore morphol-
ogy. Sequences of the internally transcribed spacer regions
of nuclear rDNA (ITS) were generated for representatives of
the ten most frequently isolated groups for each host species
that could not be identiﬁed to genus by morphological char-
acters. The methods used were the same as those of John-
ston & Park (2005). Sequences were compared with those
deposited in Genbank using a BLAST search, and directly
with sets of authentic sequences from published studies of
taxa such as Xylariaceae (Guo et al. 2003), Pezicula (Verkley
et al. 2003), and Mycosphaerella (Crous et al. 2001). Representa-
tive isolates of each group have been deposited in the Inter-
national Collection of Micro-organisms from Plants (ICMP,
maintained by Landcare Research, Auckland).
Fig 1 – Five collection sites at two geographical regions in the North Island, New Zealand. Adjacent Waitakere Ranges sites
are approximately 3–4 km apart; the Waitakere and Urewera sites are approximately 300 km apart. Site 1, Upper Nihotupu
Dam Walk, 36 56.221
S, 174 33.523
E, 300 m asl; Site 2, Home Track, 36 57.183
S, 174 30.924
E, 312 m asl; Site 3, Puke-
matekeo Summit Track, 36 53.008
S, 174 32.090
E, 336 m asl; Site 4, Fairy Falls Track, 36 54.940
S, 174 32.746
E, 300 m asl;
Site 5, Urewera, Mangapae Stream, 38 38.285
S, 176 52.048
E, 600 m asl. Mean average annual rainfall 1971–2000, Waitakere
sites approximately 1600–1800 mm, Urewera site approximately 1800–2000 mm (Anonymous 2004a). Mean average tem-
perature Waitakere sites approximately 13–14
C, Urewera site approximately 10–11
C (Anonymous 2004b).
Foliar endophytes in Podocarpaceae 1005

Statistical analysis of data
The percentageofleafpieceswithsingle andmultipleinfections
was computed in Microsoft Excel, and Shannon’s Diversity In-
dex (H0
) in PRIMER v.5 (Clarke Warwick 2001). In addition to
thetotaldataseta seconddatasetwascreatedthatexcludedsin-
gleton species. Expected species accumulation curves based on
the Mao Tau estimator were computed for both datasets in Esti-
mateS v.8.0 (Colwell 2006) and plotted for individual tree species
in each season separately using Microsoft Excel.
We used non-parametric approaches of data analysis be-
cause the endophyte assemblage data were skewed and con-
tained many zero counts, making parametric analysis
unsuitable (Anderson 2001). Data were square-root trans-
formed to reduce the influence of the most abundant species
(Clarke Warwick 2001). Bray–Curtis similarity indices were
calculated (Bray Curtis 1957), and patterns from the resulting
similarity matrix were examined using Nonmetric Multidi-
mensional Scaling (NMDS) ordination in PRIMER v.5. An anal-
ysis of similarities (ANOSIM) was carried out using PRIMER
v.5 to assess the statistical significance of differences between
and within hosts and sites, using the same similarity matrix
calculated for NMDS. Data were square-root transformed to re-
duce the influence of the most abundant species.
The R value in ANOSIM gives an absolute measure of how
separated the groups are, on a scale of 0 (indistinguishable)
to 1 (all similarities within groups are less than any
similarities between groups) (Clarke Gorley 2001). Permuta-
tion tests established the statistical significance (P) of the R
values.
Variability in the endophyte assemblages at different spatial
scales, i.e. among sites, trees within sites, and branches within
trees, was measured explicitly using a semi-parametric permu-
tational multivariate analysis of variance (PERMANOVA;
Anderson 2001; McArdle Anderson 2001). This approach al-
lows a direct additive partitioning of variation among individual
terms using the multivariate analogue of the ANOVA compo-
nent estimators (e.g. Searle et al. 1992). For each season a three
factornesteddesignwasapplied:sites(fivelevelscorresponding
to each site), trees (two trees per genus nested within sites), and
branches (five branches per tree nested within trees). Data were
square-root transformed, and Bray–Curtis dissimilarities were
used in all analyses. The mean squares derived from PERMA-
NOVA were used to calculate components of multivariate varia-
tion following the methods in Anderson et al. (2005).
Results
Diversity of endophytes
In winter, a total of 4922 endophyte strains representing 479
morphotypes and in summer, 4045 endophyte strains repre-
senting 495 morphotypes were isolated. Table 1 provides the
Table 1 – Number of isolates from each host in each season, and their relative diversity
Host D. cupressinum P. ferruginea P. totara D. dacrydioides K. ericoides
Season (w ¼ winter; s ¼ summer) w s w s w s w s w s
% leaf pieces infected (n ¼ 2000) 42.9 45.8 74.2 67.6 49.8 44.3 48.3 61.4 84.3 81.6
% leaf pieces with multiple infections 4.5 4.3 3.2 3.0 1.9 1.8 2 1.7 7.7 11.0
Total number of isolates 872 806 1197 1117 557 702 838 1140 1458 1280
Total species of endophytes 137 162 75 155 88 53 141 131 100 88
Shannon’s diversity index H0
3.96 3.54 2.66 3.09 2.74 2.97 3.18 3.39 2.40 2.57
Fig 2 – Expected endophyte species accumulation curves for each host species based on the Mao Tau estimator calculated
using EstimateS v.8.0 (Colwell 2006).(A) Winter sample. (B) Summer sample.

number of leaf pieces with single and multiple infections, the
number of species, and Shannon’s diversity indices for each
host from each season. All hosts contained a high number of
taxa that were observed only once (singletons); 360 (75.2 %)
in winter, 423 (85.5 %) in summer. Shannon’s diversity indices
reveal that diversity was highest in the endophyte assemblage
of Dacrydium cupressinum (3.52 and 3.96 in summer and winter
respectively) and low in Kunzea ericoides (2.57 and 2.40 in
Table 2 – Number of isolates of the ten most frequently isolated endophytes per host species. Fungi showing an apparent
host preference (more than 75 % of the isolates are from one host) are indicated in bold
Host D. cupressinum P. ferruginea P. totara D. dacryioides K. ericoides
Season (w ¼ winter; s ¼ summer) w s w s w s w s w s
Agaricomycotina
Agaricomycetes, Agaricomycetidae, Agaricales
Cylindrobasidium sp.a
25 0 0 0 0 0 0 0 0 0
Pezizomycotina
Dothideomycetes, Dothideomycetidae, Mycosphaerellales
Mycosphaerella sp. 1a
0 0 0 0 0 2 1 2 678 249
0 0 0 0 0 0 0 0 146 375
0 0 0 0 67 93 0 0 0 0
Phyllosticta sp. 1 3 30 38 125 9 3 0 27 8 5
Phyllosticta sp. 2 3 2 0 17 0 12 111 209 0 6
Phyllosticta sp. 3 0 0 0 0 0 0 143 0 0 0
Leotiomycetes, Helotiales
Helotiales sp. 1a
25 46 0 0 0 0 0 2 0 0
Helotiales sp. 2a
46 28 0 1 0 1 0 0 0 0
Pezicula sp. 1a
53 193 0 6 0 1 0 12 0 1
Pezicula sp. 2a
0 0 458 195 0 0 0 10 0 0
Cryptosporiopsis actinidiaea
4 5 0 27 6 27 0 19 6 13
Neofabraea sp. 1a
0 0 0 0 119 50 0 0 0 0
Torrendiella sp. 0 0 0 0 0 0 0 0 192 209
Sordariomycetes
Sordariomycetes sp. 1a
5 1 0 12 3 0 2 0 0
Sordariomycetes sp. 2a
36 3 0 0 0 0 0 0 0 0
Sordariomycetes, Hypocreomycetidae, Glomerellaceae
Colletotrichum sp. 1 (C. gloeosporioides group) 0 0 123 0 0 0 0 0 0 0
Colletotrichum sp. 4 (C. boninense group) 22 0 0 138 0 0 0 62 0 0
Sordariomycetes, Sordariomycetidae, Diaporthales
Ophiognomonia sp. 15 11 62 48 34 48 22 13 33 33
Phomopsis sp. 2 2 8 19 0 21 15 32 15 28
Sordariomycetes, Sordariomycetidae, Sordariales
Lasiosphaeria sp.a
24 63 60 68 19 88 0 28 6 9
Sordariomycetes, Xylariomycetidae, Xylariales
Amphisphaeriaceae sp. 1a
0 0 0 0 0 0 0 48 0 0
Amphisphaeriaceae sp. 2a
130 48 15 16 0 1 7 7 1 4
Rosellinia sp. 1a
0 0 0 0 0 29 0 28 0 19
Xylaria sp. 1a
30 35 10 178 4 45 51 139 0 35
Xylaria sp. 2a
1 23 0 9 0 16 0 12 0 9
Xylaria sp. 3a
0 0 0 5 4 3 54 16 0 1
Xylariaceae sp. 1a
14 25 81 56 10 78 0 56 12 0
Xylariaceae sp. 2a
4 39 0 0 0 0 0 0 0 0
Xylariaceae sp. 3a
2 0 8 0 4 7 79 33 0 3
Xylariaceae sp. 4 2 0 28 0 0 0 0 0 2 0
Xylariaceae sp. 5a
24 2 13 1 0 25 0 31 0 11
a Identiﬁcation based on comparison with ITS sequences in Genbank (vouchers listed in Table 3). This is a cumulative list, many species rep-
resented in the top ten of more than one host. Names for higher taxa follow Hibbett et al. (2007).

summer and winter) as compared with other hosts. Species
accumulation curves did not reach an asymptote when single-
ton species were included in the analysis (not shown). When
singletons were removed, all curves approximated an asymp-
tote (Fig 2).
The ten most frequently isolated species from each host
are listed in Table 2. Genbank accession numbers for ITS se-
quences generated for the species in this list which could
not be identified morphologically are given in Table 3.
Differences between host species
The endophyte assemblages of the podocarps and Kunzea eri-
coides are strongly shaped by the host species, shown by en-
dophyte assemblages of the same host species clustering
together in the NMDS graph (Fig 3). Stress levels of 0.18 and
0.19 indicate that the NMDS plots provide a satisfactory rep-
resentation of the data (Podani 1994). The statistical signifi-
cance of assemblage differences in host species was
confirmed by ANOSIM (Global R ¼ 0.825, P ¼ 0.001) (Table 4).
Endophyte assemblages in each host species also differed
significantly in all pair-wise comparisons. The average R
value of the podocarp–podocarp comparisons was slightly
lower than the podocarp–K. ericoides comparisons, 0.795 ver-
sus 0.862 respectively.
Several of the most frequently isolated fungal taxa were
common to all plant species, including Cryptosporiopsis actini-
diae, Ophiognomonia sp., Lasiosphaeria sp., Phomopsis sp., some
species of Colletotrichum, and some xylariaceous species (Table
2). Among these taxa there was limited evidence of endophyte
specificity at the family level; only two of the frequently
isolated species shared by different hosts (Sordariomycetes sp.
1 and Colletotrichum sp. 4) were restricted to members of the
Podocarpaceae.
Differences between sites
Endophyte assemblages from sites in the Waitakere Ranges
clustered together in NMDS plots and were often separated
from those of the Urewera site (Fig 3). Significant differences
in endophyte assemblages were detected in a global compari-
son of sites using ANOSIM (Global R ¼ 0.523, P ¼ 0.001) (Table 5).
Most sites also differed significantly in pair-wise comparisons,
except for comparisons between the Waitakere sites involving
site 3 (Table 5). In addition, pair-wise comparisons suggest
a greater similarity for sites within the Waitakere Ranges
(R ¼ 0.23–0.75) than for comparisons between the Waitakere
sites and the Urewera site (R ¼ 0.70–0.90).
Differences between seasons
The differences in endophyte assemblages between summer
and winter were less pronounced than those observed for
both host species and sites, indicated by a lower Global R
value, but these differences were still statistically significant
(Global R ¼ 0.367, P ¼ 0.001). Some of the endophytes were
dominant in summer (e.g. Amphisphaeriaceae sp. 1, Xylaria
sp. 2) and others in winter (e.g. Cylindrobasidium sp., several
of the Colletotrichum spp., Phyllosticta sp. 3 and Xylariaceae sp.
4). Most of the taxa observed at higher frequencies were found
at similar levels in both seasons (Table 2).
Table 3 – Representative isolates of taxa listed in Table 2 with ITS sequences generated as part of this study. Identifications
based on comparison with sequences deposited in GenBank
Endophyte species ICMP voucher number(s) Genbank accession number(s)
Amphisphaeriaceae sp. 1 16120, 16023 EU482201, EU482202, EU482203
Amphisphaeriaceae sp. 2 16130 EU482204, EU482205
Sordariomycete sp. 1 16009, 16011 EU482206, EU482207
Sordariomycete sp. 2 15976, 16021 EU482208, EU482209
Colletotrichum boninense group-species 17319, 17320, 17321, 17322 EU482210, EU482211, EU482212, EU482213
Colletotrichum gloeosporioides group-species 17323, 17324, 17325, 17326 EU482214, EU482215, EU482216, EU482217
Colletotrichum sp. 17327, 17328 EU482218, EU482219
Cryptosporiopsis actinidiae 15963, 15964, 15978, 15970 EU482220, EU482221, EU482222, EU482223
Cylindrobasidium sp. 16027 EU482224
Helotiales sp. 1 16015, 16121, 16016, 16013, 16014 EU482225, EU482226, EU482227, EU482228, EU482229
Helotiales sp. 2 16134, 16012 EU482230, EU482231, EU482232, EU482233
Lasiosphaeria sp. 16019 EU482234
Mycosphaerella sp. 1 16478 EU482235, EU482236, EU482237
Mycosphaerella sp. 2 16122, 16123 EU482238, EU482239, EU482240
Mycosphaerella sp. 3 16124, 16025 EU482241, EU482242
Neofabraea sp. 1 16020, 16127 EU482243, EU482244
Pezicula sp. 1 16024, 15983 EU482245, EU482246, EU482247, EU482248
Pezicula sp. 2 15980, 15981, 15971, 15982 EU482249, EU482250, EU482251, EU482252
Rosellinia sp. 1 15984, 16033, 16032 EU482253, EU482254, EU482255
Xylaria sp. 1 16132, 15985, 15990, 16129, 15972 EU482256, EU482257, EU482258, EU482259, EU482260
Xylaria sp. 2 15989, 16128 EU482261, EU482262
Xylaria sp. 3 15991 EU482263
Xylariaceae sp. 1 15987, 16131, 16030, 15988, 15986 EU482264, EU482265, EU482266, EU482267, EU482268
Xylariaceae sp. 2 16034 EU482269
Xylariaceae sp. 3 16031 EU482270
Xylariaceae sp. 5 16029 EU482271, EU482272

Differences within a host species
The results of a PERMANOVA analysis based on a Bray–Curtis
dissimilarity measuring differences in endophyte assem-
blages at the scale of site, tree, branch, and leaf (residual) for
each host/season combination are tabulated in Table 6. Endo-
phyte assemblages differed significantly at most spatial levels
for all hosts. The size of the component of multivariate varia-
tion between endophyte assemblages was only slightly lower
for branches within the same tree (12.8–22.6 %) compared with
conspecific trees at the same site (variation 14.6–27.4 %) (Fig 4).
Discussion
Diversity of endophytes
Members of the New Zealand Podocarpaceae and Kunzea
ericoides host a rich diversity of foliar fungal endophytes
(Table 1). Between 53 and 162 morphotypes were recovered
for individual tree species during the summer (a total of 479
morphotypes) and between 75 and 141 morphotypes during
winter (a total of 495 morphotypes). General species richness
patterns were similar to those of extensive studies of foliar
fungal endophytes in conifers from the Northern Hemisphere.
For example, 100 endophytic species were isolated from Nor-
way Spruce needles (Sieber 1988), 48 taxa from Pinus abies nee-
dles (Lorenzi et al. 2004), and up to 110 species (mean value 60)
for individual hosts among a number of conifers studied (Pet-
rini 1986). Fungal endophytes in tropical dicotyledonous trees
appear even more diverse. For example, Arnold et al. (2000)
detected 242 morphotypes in Heisteria concinna and 259 in Our-
atea lucens in a study that utilised approximately half the num-
ber of leaf segments per tree species compared to the present
study. A considerably lower diversity was reported from dry
tropical forests of Tamil Nadu, India (Suryanarayanan et al.
2003), which may suggest that endophyte diversity is influ-
enced to some extent by climatic factors.
The total diversity of endophytes in the tree species stud-
ied was not captured in the present study, as morphotypes
continued to accumulate with each additional sampling
unit. However, accumulation curves levelled off when single-
ton species were removed (Fig 3). This pattern agrees with that
observed for fungal endophytes in tropical trees (Arnold et al.
2001) and suggests that although additional sampling would
have yielded more morphotypes, these would have primarily
comprised ‘rarer’ taxa. In the present study, a high percentage
of morphotypes were observed as singletons (75.2 % in winter
and 85.5 % in summer). The reasons for the high number of
singleton species detected remain unclear, but may primarily
be the result of chance events rather than ecological con-
straints (Magurran Henderson 2003). Just as many of the
characteristic, dominant, host species-specialised podocarp
endophytes were occasionally isolated from the other podo-
carps sampled, it is likely that some of the rarely isolated spe-
cies could have been dominant on other unsampled plant
species from the same forests.
Distribution of endophytes – host-related factors
Endophyte assemblages in the four New Zealand members of
the Podocarpaceae studied and in Kunzea ericoides are strongly
shaped by host species. This is indicated by the clear separa-
tion of groups corresponding to host taxa in NMDS plots (Fig
3), and by significant differences in global (Global R ¼ 0.825,
P ¼ 0.001) and pair-wise comparisons of endophyte assem-
blages in different hosts (Table 4). Our results agree with those
of other studies, which also indicated a degree of host prefer-
ence among endophyte assemblages (e.g. Petrini 1991).
We observed between one and six endophyte taxa strongly
dominant in each host species. Many of these taxa were also
isolated in lower frequencies from other hosts, where they
may represent chance infections (Table 2). Some of these ap-
parently host-restricted taxa belong to genera such as Colleto-
trichum, Mycosphaerella, and Phyllosticta, which include known
host-specific plant pathogens as well as endophytes. Helotia-
ceous fungi were also strongly represented among the appar-
ently host-restricted taxa, including two species of Pezicula,
Neofabraea sp., Torrendiella sp., and two unidentified
Fig 3 – Nonmetric Multidimensional Scaling plot based on
Bray–Curtis similarities, comparing endophyte assem-
blages across host species and site. Each symbol represents
a single tree, 2 trees from each site, with site number indi-
cated on each data point. Sites 1–4 in the Waitakere Ranges
(each site approximately 3–4 km apart), Site 5 at Urewera
(about 300 km from the Waitakere sites). (A) Winter sample.
(B) Summer sample.

helotialean genera (Table 2). Within this group, at least two
genera (Pezicula and Neofabraea) include known pathogens,
and Pezicula has also been reported as a widespread endo-
phyte of shrubs and trees from the northern temperate zone
(Abeln et al. 2000). Species of Torrendiella are strongly host-spe-
cialised, and some have been reported as having an endo-
phytic stage to their life cycle (Cabral 1985; Johnston 1998).
Among the most frequently isolated xylariaceous taxa,
four appeared host-restricted while five infected three or
more hosts (Table 2). Xylariaceae are well-known endophytes
of a wide range of plants, from liverworts to angiosperms
(Davis et al. 2003). Data from Anonymous (2001–2007) shows
most of New Zealand’s xylariaceous species exhibit little
host preference, although there are exceptions. Examples of
putatively host-restricted species include several species of
Hypoxylon on Nothofagus wood, Rosellinia rhopalostilicola on
fronds of the palm Rhopalostylis, and Hypoxylon torrendii on
leaves of the epiphytic lily Astelia. The fruiting bodies of Xylar-
iaceae have never been found on leaves of Podocarpaceae or
K. ericoides in New Zealand (data from Anonymous 2001–
2007). Whether or not the species we isolated as endophytes
are the same that develop fruiting bodies on fallen wood
and dead leaves in the same forests is unknown (see Discus-
sion below).
In contrast to host-restricted species, twelve of the fre-
quently isolated fungal taxa were detected not only in all
four podocarp species, but also from K. ericoides (Table 2).
These included species of Xylariaceae, Colletotrichum, Phyllos-
ticta, and Pezicula (as the anamorphic state Cryptosporiopsis
actinidiae), all families or genera which included some host-re-
stricted species in our study. A small number belonged to
other ascomycetous genera, such as Lasiosphaeria and Ophiog-
nomonia. Some studies have shown endophyte assemblages to
reflect host relationships above the level of species. For exam-
ple, Carroll Carroll (1978) showed that taxonomic affinities
within subgenera of Abies were mirrored closely by the degree
of similarity among their endophytic assemblages. Petrini
Carroll (1981) suggested there may be host preference of foliar
endophytes at the level of host family in Cupressaceae, as sev-
eral of the observed fungal taxa had been previously reported
from other hosts within the family (Petrini Carroll 1981). In
contrast, the phylogenetic relationship among Podocarpaceae
was barely reflected in the degree of similarity among their
endophyte assemblages when compared to K. ericoides (Fig 2,
Table 4). Only two of the common fungal endophytes occur-
ring in three or more podocarp species were absent from K. eri-
coides, Colletotrichum sp. 4 and Sordariomycete sp. 1 (Table 2),
and the ANOSIM R values differed only slightly between podo-
carp–podocarp comparisons compared with podocarp–K. eri-
coides comparisons (Table 4).
Although our study has answered the question of host-
preference of endophytes at the level of host species and fam-
ily, more work is required to elucidate fungal-plant affiliations
at the generic level of the host plant. The inclusion of a second
species of Prumnopitys in our study, Prumnopitys taxifolia, was
hampered by the absence or the rare occurrence of this conge-
neric tree species at the study sites.
Distribution of endophytes – other factors
While host-related factors were the strongest determinant,
geographic separation and seasonal effects also modulated
fungal endophytic assemblages, albeit to a lesser extent. Over-
all, species assemblages differed significantly between the
sites studied (Table 5) and greater geographic distances, i.e.
between the Waitakere and Urewera sites separated by ap-
proximately 300 km, were reflected in lower similarities be-
tween endophyte assemblages (Fig 3, Table 5). These results
contrast with a study of endophytic assemblages in Sequoia
sempervirens, which were similar across the natural range of
the tree species (Rollinger Langenheim 1993). Although situ-
ated on the same island, the sites in the Waitakere Ranges and
Urewera are in disjunct stretches of native forest, separated
not only by distance but also by farmland. The observed differ-
ences might indicate either barriers to dispersal, or other fac-
tors, such as climate or site history.
Table 4 – Analysis of similarity (ANOSIM) pair-wise
comparisons of endophyte assemblages between each
host. Global R value [ 0.825 (P [ 0.001). The R value gives
an absolute measure of how separated the groups are, on
a scale of 0 (indistinguishable) to 1 (all similarities within
groups are less than any similarities between groups)
(Clarke Gorley 2001)
Pair-wise comparisons
between hosts
R value P
D. cupressinum, P. ferruginea 0.8 0.008
D. cupressinum, P. totara 0.8 0.004
D. cupressinum, D. dacrydioides 1 0.004
D. cupressinum, K. ericoides 1 0.004
P. ferruginea, P. totara 0.65 0.012
P. ferruginea, D. dacrydioides 0.9 0.004
P. ferruginea, K. ericoides 0.9 0.004
P. totara, D. dacrydioides 0.625 0.021
P. totara, K. ericoides 0.55 0.021
D. dacrydioides, K. ericoides 1 0.004
Table 5 – Analysis of similarity of (ANOSIM) pair-wise
comparisons of endophyte assemblages between each
site. Sites 1–4 in the Waitakere Ranges with adjacent sites
approximately 3–4 km apart, and site 5 in near the
Urewera National Park, about 300 km distant from
Waitakere. Global R value [ 0.523 (P [ 0.001). The R value
gives an absolute measure of how separated the groups
are, on a scale of 0 (indistinguishable) to 1 (all similarities
within groups are less than any similarities between
groups) (Clarke Gorley 2001)
Pair-wise comparisons
between sites
R value P
1, 2 0.5 0.037
1, 3 0.225 0.148
1, 4 0.5 0.012
1, 5 0.7 0.008
2, 3 0.2 0.255
2, 4 0.75 0.004
2, 5 0.9 0.004
3, 4 0.4 0.074
3, 5 0.65 0.008
4, 5 0.95 0.004

Endophyte assemblages were also influenced by season.
Seasonal differences, although statistically significant, were
less pronounced than those observed for host species and sites
across the total endophyte assemblage (R ¼ 0.367, P ¼ 0.001).
Seasonal patterns differed for individual taxa with a few spe-
cies being present only either in summer or winter, such as
Cylindrobasidium sp. More commonly, there were differences
in isolation frequency across seasons (Table 2). Because the
winter and summer samples were selected from the same co-
hort ofleaves, the variation observed suggeststhat formost en-
dophytes, individual infections are not persistent. Rather, over
time there may be a high and ongoing turnover of endophyte
populations within a single leaf. A preferential loss of infected
leavesmayalso explain thisresultforKunzea ericoides, butthere
was noleaf loss from the one seasonoldtwigsof the podocarps.
Phylogenetic diversity and biology of the endophytes isolated
Stone et al. (2004) listed taxa commonly isolated as endophytes
from woody plants, and most of the broad groups in their list
are represented amongst the fungi we found frequently (Table
2). Several of the ascomycetes we isolated were not represented
in Genbank with ITS sequences, and their identity remains un-
known. Basidiomycetes are generally recorded only rarely as
foliar endophytes, although Crozier et al. (2006) reported large
numbers from trunks of mature trees. The single basidiomy-
cete species in our survey was found only on Dacrydium cupres-
sinum in winter, but at that time the fungus was isolated from
all eight trees sampled from the Waitakere Ranges. The ITS se-
quence of this basidiomycete is only 4 bp different from a fun-
gus isolated from Podocarpus falcatus seeds in Ethiopia (Gure
et al. 2005). Although the Ethiopian fungus was identified as Pol-
yporus gayanus on the basis of cultural characters (Gure et al.
2005), comparison with other ITS sequences in Genbank sug-
gest it is more likely to be a Cylindrobasidium sp. The Ethiopian
fungus is pathogenic to germinating seeds of P. falcatus (Gure
et al. 2005). There is no information on podocarp seed patho-
gens in New Zealand, and the biology of our Dacrydium endo-
phyte remains unknown, although it is intriguing that these
two closely related fungi share a host in the Podocarpaceae.
Table 6 – Permutational multivariate analyses of variance (PERMANOVA) based on Bray–Curtis dissimilarity measure for
square-root transformed abundance data of all endophyte species for each host/season combination. MS values were used
to calculate the proportion of variance for each component (see Fig 4)
Source df D. dacrydioides summer D. dacrydioides winter
MS F P MS F P
Site 4 78 322.8 3.5523 0.0120 112 439.7 3.5530 0.0030
Tree (Si) 5 22 048.5 3.3291 0.0010 31 646.6 5.5436 0.0010
Branch (Tr(Si)) 40 6622.9 1.9502 0.0010 5708.7 2.0607 0.0010
Residual 450 3396.4 2770.2
Total 499
P. ferruginea summer P. ferruginea winter
MS F P MS F P
Site 4 102 596.2 3.1938 0.0070 120 983.4 4.4956 0.0060
Tree (Si) 5 32 123.1 5.8673 0.0010 26911.5 3.5095 0.0010
Branch (Tr(Si)) 40 5474.9 1.8310 0.0010 7668.2 3.0061 0.0010
Residual 450 2990.1 2550.9
Total 499
P. totara summer P. totara winter
MS F P MS F P
Site 4 40 989.7 2.3501 0.0330 70 390.2 1.6424 0.1130
Tree (Si) 5 17 441.4 2.5889 0.0010 42 858.3 8.1070 0.0010
Branch (Tr(Si)) 40 6736.9 1.8179 0.0010 5286.6 1.9011 0.0010
Residual 450 3705.9 2780.9
Total 499
D. cupressinum summer D. cupressinum winter
MS F P MS F P
Site 4 42 636.8 2.0145 0.0440 60 665.8 2.9169 0.0010
Tree (Si) 5 21 165.4 2.8616 0.0010 20 797.9 4.0431 0.0010
Branch (Tr(Si)) 40 7396.5 2.0603 0.0010 5144.1 1.4686 0.0010
Residual 450 3589.9 3502.8
Total 499
K. ericoides summer K. ericoides winter
MS F P MS F P
Site 4 60 805.5 2.3857 0.0260 53 919.0 1.5072 0.1160
Tree (Si) 5 25 487.4 5.2612 0.0010 35 774.9 6.5125 0.0010
Branch (Tr(Si)) 40 4844.4 1.8119 0.0010 5493.3 2.4209 0.0010
Residual 450 2673.7 2269.1
Total 499
F ¼ pseudo F statistic (McArdle Anderson 2001).

Xylariaceae were frequently isolated in our study, and these
fungi are also common on fallen wood in the forests we sam-
pled. However, the fruiting bodies of Xylariaceae are found
rarely on leaves, raising a question about the biological signif-
icance of the leaf endophyte infections. Despite their strong
lignolytic enzymatic capability (Pointing et al. 2003) and ability
to cause various types of disease primarily through extensive
tissue degradation, several studies have indicated that some
xylariaceous species isolated as endophytes may not be latent
saprophytes (Bayman et al. 1998; Griffith Boddy 1990; Laessøe
Lodge 1994), but rather exist solely as mutualistic endo-
phytes (Davis et al. 2003; Rogers 2000). There is insufficient mo-
lecular data for New Zealand Xylariaceae to confirm whether
the species we isolated as leaf endophytes are the same as
those fruiting on fallen wood in the same forests. However,
a comparison with taxa in the Xylariaceae tree published by
Guo et al. (2003) suggests that some at least some may not be.
Four of our Xylariaceae groups belong in the WMS15 clade of
Guo et al. (2003). All isolates in this clade are known only as
leaf endophytes and it could perhaps represent an ecologically
specialised group of Xylariaceae, biologically and genetically
distinct from the species which form ascomata on wood.
Pezicula and Neofabraea are two closely related genera of in-
operculate discomycetes that have been reported rarely from
New Zealand native forests (Anonymous 2001–2007), but were
surprisingly diverse as podocarp endophytes. Host-special-
ised species were isolated frequently from D. cupressinum
(two species), Prumnopitys ferruginea, and Podocarpus totara. As-
suming the leaf endophyte infections were initiated from as-
cospores, New Zealand has at least four species of Pezicula
awaiting discovery. In addition, the Pezicula anamorph Crypto-
sporiopsis actinidiae, first described as a pathogen of Actinidia in
orchards (Johnston et al. 2004), was commonly isolated in our
study, but it was not host specialised.
We used ITS sequences to compare several of the endo-
phytic Colletotrichum species with isolates from fruits of horti-
cultural crops, previously intensively studied in New Zealand
(e.g. Johnston Jones 1997; Johnston et al. 2005; Lardner et al.
1999). Most of the endophyte isolates were members of the
Colletotrichum gloeosporioides (e.g. Genbank deposit numbers
EU482214 – EU482217) and Colletotrichum boninense (e.g. Gen-
bank deposit numbers EU482210 – EU482213) clades (sensu
Johnston et al. 2005), and Colletotrichum acutatum was also iso-
lated. Colletotrichum isolates from fruits also fall commonly
into these three clades. However, one of the less frequently
isolated Colletotrichum species (e.g. Genbank deposit numbers
EU482218 and EU482219) was genetically distinct from all the
fruit-inhabiting Colletotrichum species known from New Zea-
land, was also distinct from all species included in the analy-
sis of Farr et al. (2006), and may represent an indigenous,
forest-inhabiting species. This species was isolated from sev-
eral different podocarps and was morphologically distinctive
in having colonies with restricted growth, the surface of the
colonies densely covered with orange conidial ooze and al-
most no aerial mycelium, setae lacking, conidia about 20–
25 Â 4–4.5 mm, gently curved, tapering toward each end.
Several Phyllosticta and Mycosphaerella species were fre-
quently isolated. Both genera are known to have many unde-
scribed species associated with plants in New Zealand’s
native forests (PRJ, unpubl. data).
It remains unknown whether the podocarp and Kunzea en-
dophytes form a true mutualistic association with their hosts
or whether they are latent pathogens, commensalistic or
have a putatively neutral relationship (Cabral et al. 1993; Rodri-
guez Redman 1997). In a concurrent study, fungal hyphae in
leaves of Kunzea ericoides were visualised using a fluorescent la-
belling method (Johnston et al. 2006), and three different host
reactions to fungal hyphae were observed: (1) no reaction
with fungal hyphae extending deep into the leaf within inter-
cellular spaces; (2) callose formation and restriction of fungus
to stomatal cavity and intercellular spaces, indicating a plant
defence reaction; and (3) a hypersensitive plant defence reac-
tion with intracellular penetration and death of a single plant
cell. These data suggest different fungal life styles for the fungi
present in K. ericoides, and a similar diversity is likely amongst
the podocarp-associated fungi. The hypersensitive plant
Fig 4 – Sizes of components of multivariate variation in endophyte assemblages between sites, trees, branches, and leaves
(residual), as multivariate analogues to the univariate variance components, obtained using mean squares from the PER-
MANOVA results in Table 6. The values plotted are the square root of the sizes of the components of variation, the values
thus matching the scale of the original Bray–Curtis dissimilarities (expressed as a percentage difference between assem-
blages). a–e are the values for hosts Dacrycarpus dacrydioides, Prumnopitys ferruginea, Podocarpus totara, Dacrydium cupressi-
num, and Kunzea ericoides respectively. (A) Winter sample. (B) Summer sample.

defence reaction is a common expression of disease resistance
in plants, controlled by interactions between pathogen aviru-
lence genes and plant resistance genes (Heath 2000). The other
reactions are difficult to interpret as responses to either path-
ogenic or endophytic invasions of the leaf, with Schulz et al.
(1999) describing a range of putatively defensive plant reac-
tions to hyphae of both pathogenic and endophytic fungi.
Practical sampling considerations
Observed fungal species composition in studies using cultur-
ing methods to assess endophyte diversity can be affected
by the size of the leaf segments sampled, as multiple infec-
tions per leaf segment may bias species composition in favour
of more competitive or fast growing strains (Carroll 1995;
Gamboa et al. 2002). Only 1.7–4.5 % of leaf segments from
podocarp leaves and 7.7–11.0 % of leaf segments from Kunzea
ericoides leaves had multiple infections, and on average about
45 % of leaf pieces from the podocarps and 17 % of leaf pieces
from K. ericoides remained uninfected (Table 1). Furthermore,
slow growing morphotypes were commonly isolated among
both the abundant and rarer taxa. Together, this suggests
the degree of bias introduced by species competition is likely
to have been low in the present study.
Sampling strategies for studies on endophyte diversity
must take into account natural patchiness in the distribution
of fungal species across the landscape. For example, Johnston
(1998) illustrated between-tree variation in the distribution of
some endophytes in even-aged stands of Leptospermum scopa-
rium. The hierarchical sampling structure used in this study
allows the proportion of the variances between branches,
trees, and sites to be estimated. We applied a PERMANOVA
(Anderson 2005; Anderson et al. 2005) and partitioned the var-
iation among three nested components: individual sites, trees,
and branches. For each host species/season combination the
variability in overall patterns of diversity between branches
was similar to, or less than, that between trees (Fig 4). Thus,
our within-tree sampling strategy of selecting ten leaves
from each of five branches, rather than the less practical se-
lection of 50 leaves across the whole tree, was unlikely to
have biased the results to a major extent.
All isolations were carried out within a few hours of the
branches being picked, minimising the impact that leaf senes-
cence following picking may have had on the diversity ob-
served (Johnston 1998; Millar Richards 1974).
An attempt was made to define taxa at about the level of
species. For those fungi that did not sporulate in culture, this
involved initial morpho-taxa groupings that were subse-
quently tested using ITS sequences. In most cases, the origi-
nal morpho-taxa were well supported, but initial groupings
were sometimes modified on the basis of the sequencing re-
sults. A few of the original morpho-taxa were combined, and
one xylariaceous taxon was removed from the analysis after
it was shown to be polyphyletic. When initially defining the
morpho-taxa, allowance was made for expected natural var-
iation in cultural appearance. This is a particular issue in
some taxa such as members of the Helotiales and Colletotri-
chum, and previous experience with these groups (e.g. John-
ston Gamundı´ 2000; Johnston Park 2005; Lardner et al.
1999; PRJ, unpubl.) proved valuable. Despite this, it is probable
that if all the singleton isolates had been sequenced, some
would have been found to match some of the morpho-taxa.
Naming of sterile morpho-taxa, and decisions on species
limits, were based on an initial BLAST search followed by in-
corporation of the sequence of the unknown endophyte into
the alignment of an appropriate published phylogeny. In only
one instance did the endophyte sequence match exactly an
existing sequence deposited in Genbank, this for Cryptospor-
iopsis actinidiae, a species originally described from New Zea-
land. For those that could be identified to the level of genus
and could be compared to taxa in a published phylogeny,
species-level variation was defined on the basis of the level
of between-species genetic variation generally accepted for
that taxon. Mechanical decisions on taxon limits based on
pre-determined levels of sequence similarity were avoided.
Several taxa could be identified only to the level family, order,
or class.
Acknowledgements
We wish to thank Drs Marti Anderson (Institute of Informa-
tion and Mathematical Sciences, Massey University), Margaret
Stanley, and Greg Arnold (Landcare Research) for helping with
statistical analyses. P.W. Wilkie, M. Fletcher, K. McDermott
and M. Sue are thanked for technical assistance, and R.E.
Beever for helpful discussions. Auckland Regional Council
are thanked for permission to collect the Waitakere samples
and the Tuhoe Tuawhenua Trust for allowing collections in
their native forests. Funds for this research were provided
by Landcare Research through its Investment Fund, and the
New Zealand Foundation for Research, Science and Technol-
ogy through the Defining New Zealand’s Land Biota OBI and
Agrochemicals From Microbes (FRST Contract number
BIDX0201).
r e f e r e n c e s
Abeln ECA, de Pagter MA, Verkley GJM, 2000. Phylogeny of Pezi-
cula, Dermea and Neofabraea inferred from partial sequences of
the nuclear ribosomal RNA gene cluster. Mycologia 92: 685–693.
Anderson MJ, 2001. A new method for non-parametric multivar-
iate analysis of variance. Austral Ecology 26: 32–46.
Anderson MJ, 2005. PERMANOVA: a FORTRAN computer program for
permutational multivariate analysis of variance. Department of
Statistics, University of Auckland, New Zealand.
Anderson MJ, Diebel CE, Blom WM, Landers TJ, 2005. Consistency
and variation in kelp holdfast assemblages: spatial patterns of
biodiversity for the major phyla at different taxonomic reso-
lutions. Journal of Experimental Marine Biology and Ecology 320:
35–56.
Anonymous, 2001–2007. NZFUNGI Database of New Zealand Fungi.
Landcare Research, New Zealand.http://nzfungi.
landcareresearch.co.nz.
Anonymous, 2004a. NIWA Poster No. 4. Mean Annual Temperature
1971–2000. National Institute of Water and Atmospheric Re-
search, Wellington.
Anonymous, 2004b. NIWA Poster No. 5. Mean Annual Rainfall
1971–2000. National Institute of Water and Atmospheric Re-
search, Wellington.

Arnold AE, Maynard Z, Gilbert GS, Coley PD, Kursar TA, 2000. Are
tropical fungal endophytes hyperdiverse? Ecology Letters 3:
267–274.
Arnold AE, Maynard Z, Gilbert GS, 2001. Fungal endophytes in
dicotyledonous trees: patterns of abundance and diversity.
Mycological Research 105: 1502–1507.
Bayman P, Anulo-Sandoval P, Baez-Ortiz Z, Lodge DJ, 1998. Dis-
tribution and dispersal of Xylaria endophytes in two tree
species in Puerto Rico. Mycological Research 102: 944–948.
Bray JR, Curtis JT, 1957. An ordination of upland forest communi-
ties of southern Wisconsin. Ecological Monographs 27: 325–349.
Cabral D, 1985. Phyllosphere of Eucalyptus viminalis: dynamics of
fungal populations. Transactions of the British Mycological Society
85: 501–511.
Cabral D, Stone JK, Carroll GC, 1993. The internal mycobiota of
Juncus spp.: microscopic and cultural observations of infection
patterns. Mycological Research 97: 367–376.
Carroll FE, Mu¨ ller E, Sutton BC, 1977. Preliminary studies on the
incidence of needle endophytes in some European conifers.
Sydowia 29: 87–103.
Carroll G, 1988. Fungal endophytes in stems and leaves: from la-
tent pathogen to mutualistic symbiosis. Ecology 69: 2–9.
Carroll G, 1995. Forest endophytes: patterns and process. Cana-
dian Journal of Botany 73: S1316–S1324.
Carroll GC, Carroll FE, 1978. Studies on the incidence of coniferous
needle endophytes in the Pacific Northwest. Canadian Journal of
Botany 56: 3034–3043.
Clarke KR, Gorley RN, 2001. PRIMER v.5: user manual/tutorial.
Primer-E, Plymouth, U.K.
Clarke KR, Warwick RM, 2001. Change in Marine Communities: an
approach to statistical analysis and interpretation, 2nd edn.
Primer-E, Plymouth, U.K.
Colwell RK, 2006. EstimateS: statistical estimation of species richness
and shared species from samples, version 8.0 Persistent
URL: http://viceroy.eeb.uconn.edu/estimates http:
//purl.oclc.org/estimates
Connnor HE, Edgar E, 1987. Name changes in the indigenous New
Zealand flora, 1960–1986 and Nomina Nova IV, 1983–1986. New
Zealand Journal of Botany 25: 115–170.
Crous PW, Hong LI, Wingfield BD, Wingfield MJ, 2001. ITS rDNA
phylogeny of selected Mycosphaerella spp. and their ana-
morphs occurring on Myrtaceae. Mycologia 105: 425–431.
Crozier J, Thomas SE, Aime MC, Evans HC, Holmes KA, 2006.
Molecular characterisation of fungal endophytic morphospe-
cies isolated from stems and pods of Theobroma cacao. Plant
Pathology 55: 783–791.
Davis CE, Franklin JB, Shaw AJ, Vilgalys R, 2003. Endophytic Xy-
laria (Xylariaceae) among liverworts and angiosperms: phylo-
genetics, distribution, and symbiosis. American Journal of
Botany 90: 1661–1667.
Espinosa-Garcia FJ, Langenheim JH, 1990. The endophytic fungal
community in leaves of a coastal redwood populationddi-
versity and spatial patterns. New Phytologist 116: 89–97.
Esler AE, Astridge SJ, 1974. Tea tree (Leptospermum) communities
of the Waitakere Range, Auckland, New Zealand. New Zealand
Journal of Botany 12: 485–501.
Farr DF, Aime MC, Rossman AY, Palm ME, 2006. Species of
Colletotrichum on Agavaceae. Mycological Research 110:
1395–1408.
Fisher PJ, Petrini O, Petrini LE, Sutton BC, 1994. Fungal endo-
phytes from the leaves and twigs of Quercus ilex L. from
England, Majorca, and Switzerland. New Phytologist 127:
133–137.
Fro¨hlich J, Hyde KD, Petrini LE, 2000. Endophytic fungi associated
with palms. Mycological Research 104: 1202–1212.
Gamboa MA, Bayman P, 2001. Communities of endophytic fungi
in leaves of a tropical timber tree (Guarea guidonia: Meliaceae).
Biotropica 33: 352–360.
Gamboa MA, Laureano S, Bayman P, 2002. Measuring diversity of
endophytic fungi in leaf fragments: does size matter? Myco-
pathologia 156: 41–45.
Guo LD, Huang GR, Wang Y, He WH, Zheng WH, Hyde KD, 2003.
Molecular identification of white morphotype strains of en-
dophytic fungi from Pinus tabulaeformis. Mycological Research
107: 680–688.
Gure A, Wahlstro¨m K, Stenlid J, 2005. Pathogenicity of seed-
associated fungi to Podocarpus falcatus. Forest Pathology 35:
23–35.
Griffith GS, Boddy L, 1990. Fungal decomposition of attached an-
giosperm twigs I: decay community in ash, beech and oak.
New Phytologist 116: 407–415.
Heath MC, 2000. Hypersensitive response-related death. Plant
Molecular Biology 44: 321–334.
Hibbett DS, et al., 2007. A higher-level classification of the fungi.
Hill RS, Brodribb TJ, 1999. Southern conifers in time and space.
Australian Journal of Botany 47: 639–696.
Johnston PR, 1998. Leaf endophytes of manuka (Leptospermum
scoparium). Mycological Research 102: 1009–1016.
Johnston PR, Fletcher M, 1998. Do fungi influence leaf palatability
to browsing possums? He Korero Paihama Possum Research News
10: 8–9.
Johnston PR, Gamundı´ IJ, 2000. Torrendiella (Ascomycota, Hel-
otiales) on Nothofagus. New Zealand Journal of Botany 38:
493–513.
Johnston PR, Jones D, 1997. Relationships amongst Colletotrichum
isolates from fruit-rots assessed using rDNA sequences.
Mycologia 89: 420–430.
Johnston PR, Manning MA, Meier X, Park D, Fullerton RA, 2004.
Cryptosporiopsis actinidiae sp. nov. Mycotaxon 89: 131–136.
Johnston PR, Park D, 2005. Chlorociboria (Fungi, Helotiales) in New
Zealand. New Zealand Journal of Botany 43: 679–719.
Johnston PR, Pennycook SR, Manning MA, 2005. Taxonomy of
fruit rotting fungal pathogens: what’s really out there? New
Zealand Plant Protection 58: 42–46.
Johnston PR, Sutherland PW, Joshee S, 2006. Visualising endo-
phytic fungi within leaves by detection of (1/3)-ß-D-glucans
in fungal cell walls. Mycologist 20: 159–162.
Kumaresan V, Suryanarayanan TS, 2001. Occurrence and distri-
bution of endophytic fungi in a mangrove community. Myco-
logical Research 105: 1388–1391.
Kowalski T, Kehr RD, 1996. Fungal endophytes of living branch
bases in several European tree species. In: Redlin SC,
Carris LM (eds), Endophytic Fungi in Grasses and Woody Plants.
Systematics, Ecology and Evolution. APS Press, St Paul, Minne-
sota, pp. 67–86.
Lardner R, Johnston PR, Plummer KM, Pearson M, 1999. Morpho-
logical and molecular analysis of Colletotrichum acutatum sensu
lato. Mycological Research 103: 275–285.
Laessøe T, Lodge DJ, 1994. Three host-specific Xylaria species.
Mycologia 86: 436–446.
Lorenzi E, Rodolfi M, Picco AM, 2004. Fungal endophytes and
pathogens in Picea abies in natural and urban sites. Journal of
Plant Pathology 86: 323.
McArdle BH, Anderson MJ, 2001. Fitting multivariate models to
community data: a comment on distance-based redundancy
analysis. Ecology 82: 290–297.
Magurran AE, Henderson PA, 2003. Explaining the excess of rare
species in natural species abundance distribution. Nature 422:
714–718.
Millar CS, Richards GM, 1974. A cautionary note on the collection
of plant specimens for mycological examination. Transactions
of the British Mycological Society 63: 607–610.
Molloy BPJ, 1995. Manoao (Podocarpaceae), a new monotypic conifer
genus endemic to New Zealand. New Zealand Journal of Botany
33: 183–201.

Molloy BPJ, 1996. A new species name in Phyllocladus (Phyllocla-
daceae) from New Zealand. New Zealand Journal of Botany 34:
287–297.
Petrini O, 1986. Taxonomy of endophytic fungi of aerial plant
tissues. In: Fokkema NJ, van den Heuvel J (eds), Microbiology
of the Phyllosphere. Cambridge University Press, Cambridge,
pp. 175–187.
Petrini O, 1991. Fungal endophytes of tree leaves. In: Andrews JH,
Hirano SS (eds), Microbial Ecology of Leaves. Springer-Verlag,
New York, pp. 179–197.
Petrini O, 1996. Ecological and physiological aspects of host speci-
ficity in endophytic fungi. In: Redlin SC, Carris LM (eds), Endo-
phytic Fungi in Grasses and Woody Plants. Systematics, Ecology and
Evolution. APS Press, St Paul, Minnesota, USA, pp. 87–100.
Petrini O, Carroll G, 1981. Endophytic fungi in foliage of some
Cupressaceae in Oregon. Canadian Journal of Botany 59: 629–636.
Petrini O, Fisher PJ, 1988. A comparative study of fungal endophytes
in xylem and whole stems of Pinus sylvestris and Fagus sylvatica.
Transactions of the British Mycological Society 91: 233–238.
Petrini O, Fisher PJ, 1990. Occurrence of fungal endophytes in
twigs of Salix fragilis and Quercus robur. Mycological Research 94:
1077–1080.
Petrini O, Hake U, Dreyfuss MM, 1990. An analysis of fungal com-
munities isolated from fruticose lichens. Mycologia 82: 444–451.
Petrini O, Mu¨ ller E, 1979. Pilzliche Endophyten am Beispiel von
Juniperus communis L. Sydowia 32: 224–251.
Petrini O, Sieber TN, Toti L, Viret O, 1992. Ecology, metabolite
production, and substrate utilisation in endophytic fungi.
Natural Toxins 1: 185–196.
Podani J, 1994. Multivariate Analysis in Ecology and Systematics. SPB
Academic Publishing, The Hague.
Pointing SB, Parungao MM, Hyde KD, 2003. Production of wood-
decay enzymes, mass loss and lignin solubilization in wood by
tropical Xylariaceae. Mycological Research 107: 231–235.
Rodriguez RJ, Redman RS, 1997. Fungal lifestyles and ecosystem
dynamics: biological aspects of plant pathogens, plant endo-
phytes and saprophytes. Advances in Botanical Research 24:
169–193.
Rogers JD, 2000. Thoughts and musings on tropical Xylariaceae.
Rollinger JL, Langenheim JH, 1993. Geographic survey of fungal
endophyte community composition in leaves of coastal red-
wood. Mycologia 85: 149–156.
Schulz B, Ro¨mmert A-K, Dammann U, Aust H-J, Strack D, 1999.
The endophyte–host interaction: a balanced antagonism?
Searle SR, Casella G, McCulloch CE, 1992. Variance Components.
John Wiley and Sons, New York.
Sieber TN, 1988. Endophytische Pilze in Bla¨ttern und A¨ sten
gesunder und gescha¨digten Fichten (Picea abies (L.) Karsten).
European Journal of Forest Pathology 18: 321–342.
Sieber TN, Hugentobler C, 1987. Endophytische Pilze in Bla¨ttern
und A¨ sten gesunder und gescha¨digten Buchen (Fagus sylvatica
L.). European Journal of Forest Pathology 17: 411–425.
Stone JK, Polishook JD, White JF, 2004. Endophytic fungi. In:
Mueller GM, Bills GF, Foster MS (eds), Biodiversity of Fungi:
Inventory and Monitoring Methods. Elsevier, Amsterdam, pp.
241–270.
Suryanarayanan TS, Venkatesan G, Murali TS, 2003. Endophytic
fungal communities in leaves of tropical forest trees: diversity
and distribution patterns. Current Science 85: 489–493.
Taylor JE, Hyde KD, Jones BG, 1999. Endophytic fungi associated
with the temperate palm, Trachycarpus fortunei, within and
outside its natural geographic range. New Phytologist 142:
335–346.
Verkley GJM, Zijlstra JD, Summerbell RC, Berendse F, 2003.
Phylogeny and taxonomy of root-inhabiting Cryptosporiopsis
species and C. rhizophila sp. nov., a fungus inhabiting roots of
several Ericaceae. Mycological Research 107: 689–698.
Vujanovic V, Brisson J, 2002. A comparative study of endophytic
mycobiota in leaves of Acer saccharum in eastern North
America. Mycological Progress 1: 147–154.
Widler B, Mu¨ ller E, 1984. Untersuchungen u¨ ber endophytische
Pilze von Arctostaphylos uva-ursi (L.) Sprengel (Ericaceae).
Botanica Helvetica 94: 307–337.
Wilson D, 1995. Endophytedthe evolution of a term, and clarifi-
cation of its use and definition. Oikos 73: 274–276.
Wilson D, 2000. Ecology of woody plant endophytes. In:
Bacon CW, White Jr JF (eds), Microbial Endophytes. Marcel Dek-
ker Inc, New York, pp. 389–420.

Joshee et al Podocarp endophytes

Recommended

Recommended

More Related Content

What's hot

What's hot (19)

Similar to Joshee et al Podocarp endophytes

Similar to Joshee et al Podocarp endophytes (20)

Joshee et al Podocarp endophytes