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
Fungal Genetics and Biology 46 (2009) 642–656
Contents lists available at ScienceDirect
Fungal Genetics and Biology
journal homepage: www.elsevier.com/locate/yfgbi
A polyphasic approach to the taxonomy of the Alternaria infectoria species–group
Birgitte Andersen a,*, Jens Laurids Sørensen a, Kristian Fog Nielsen a, Bert Gerrits van den Ende b,
Sybren de Hoog b
a
b
Center for Microbial Biotechnology (CMB), Department of Systems Biology, Building 221, Technical University of Denmark, DK-2800 Kgs. Lyngby, Denmark
Centraalbureau voor Schimmelcultures (CBS-KNAW), Fungal Biodiversity Centre, P.O. Box 85167, 3508 AD Utrecht, The Netherlands
a r t i c l e
i n f o
Article history:
Received 20 December 2008
Accepted 28 May 2009
Available online 6 June 2009
Keywords:
Chemical classification
Fungi
Haplotypes
Molecular cladification
Morphology
Multivariate statistics
Recombinants
a b s t r a c t
Different taxa in the species–group of Alternaria infectoria (teleomorph Lewia spp.) are often isolated from
various cereals including barley, maize and wheat grain, ornamental plants and skin lesions from animals
and humans. In the present study we made a polyphasic characterization of 39 strains morphologically
identifiable as belonging to the A. infectoria species–group together with 12 strains belonging to closely
related species: Alternaria malorum (syn. Cladosporium malorum), Chalastospora cetera (syn. Alternaria cetera) and Embellisia abundans. Morphological examination separated the 51 strains in three groups based
on conidial appearance and arrangement: the A. infectoria species–group, E. abundans and a group containing C. cetera and A. malorum. The metabolite analyses on three different media showed two clusters,
one containing all 39 A. infectoria species–group strains and one containing 10 strains of E. abundans, C.
cetera and A. malorum. One E. abundans strain and one A. malorum strain were not included due to insufficient metabolite production. The separation of the A. infectoria species–group from E. abundans, C. cetera
and A. malorum resulted mainly from the ability to produce altertoxins and novae-zelandins. The metabolite analyses also showed that all 51 strains were able to produce infectopyrones. The metabolite profiles
of C. cetera and A. malorum were very similar with several metabolites of unknown structure in common.
This is the first time that E. abundans, C. cetera and A. malorum have been reported as producers of infectopyrones. Sequence analyses of the internal transcribed spacer region (ITS), glyceraldehyde-3-phosphate
dehydrogenase (gpd) and translocation elongation factor 1a (tef-1a) showed two clades: one with the 39
strains from the A. infectoria species–group and one with the 12 strains of E. abundans, C. cetera and A.
malorum. The polyphasic approach in this study suggests that A. malorum var. polymorpha and the eight
A. malorum strains do not belong in Alternaria, but in Chalastospora, however, as several distinct species.
Splits Tree alignment of gpd sequences of 38 strains belonging to the A. infectoria species–group indicates
that only three strains showed signs of recombination, while the remaining strains appeared to be clonal.
Long term incubation at 7 °C in the dark showed that 12 out of 33 tested strains from the A. infectoria species–group were able to produce proascomata in axenic culture, but with no mature ascospores after
6 months. These findings suggest that Lewia/A. infectoria species–group must, at least in part, be homothallic. The results presented in this study show that ITS, tef-1a and gpd do not reflect ecology, secondary
metabolism or morphology of the A. infectoria species–group and that molecular cladification and phylogeny cannot predict pathogenicity, host specificity or mycotoxin production.
Ó 2009 Elsevier Inc. All rights reserved.
1. Introduction
The Alternaria infectoria EG Simmons species–group sensu Simmons (Simmons and Roberts, 1993; Simmons, 2007) comprises
more than 30 named anamorph taxa, among which Alternaria
arbusti EG Simmons, Alternaria ethzedia EG Simmons, A. infectoria,
Alternaria intercepta EG Simmons, Alternaria metachromatica EG
Simmons, Alternaria oregonensis EG Simmons, Alternaria photistica
EG Simmons, Alternaria triticimaculans EG Simmons, Alternaria tri* Corresponding author. Fax: +45 4588 4922.
E-mail address: ba@bio.dtu.dk (B. Andersen).
1087-1845/$ - see front matter Ó 2009 Elsevier Inc. All rights reserved.
doi:10.1016/j.fgb.2009.05.005
ticina, Alternaria viburni EG Simmons are some (Simmons, 2007).
It is the only group in Alternaria where some members have a teleomorph state, Lewia ME Barr and EG Simmons (Simmons, 1986).
Morphologically, the A. infectoria species–group differs from other
Alternaria species–groups in the three-dimensional sporulation
pattern (Simmons and Roberts, 1993). Characteristic for the A.
infectoria species–group is the production of small conidia (up to
70 lm in length) in branched chains with long, geniculate multilocus secondary conidiophores (up to 120 lm) between conidia
(Simmons, 2007).
Chemically, the A. infectoria species–group is very different from
other Alternaria species, producing metabolites that are not found
Author's personal copy
B. Andersen et al. / Fungal Genetics and Biology 46 (2009) 642–656
in other species–groups (Andersen and Thrane, 1996). None of the
taxa in the A. infectoria species–group has ever been shown to produce alternariols or tenuazonic acid, which are common in other
small-spored Alternaria species (Andersen et al., 2002) or altersolanols, common in some large-spored Alternaria (Andersen et al.,
2008). On the other hand, taxa in the A. infectoria species–group
produce infectopyrones and novae-zelandins (Christensen et al.,
2005), which have never been detected in other Alternaria species–groups. However, infectopyrones have been found in other
genera, such as Phoma Sacc., Stemphylium Wallr. and Ulocladium
Preuss (Pedras and Chumala, 2005; Christensen et al., 2005; Andersen and Hollensted, 2008, respectively).
Molecularly, taxa in the A. infectoria species–group have been
analyzed at the sub-genus level using the ribosomal internal transcribed spacer region (ITS), glyceraldehyde-3-phosphate dehydrogenase (gpd) and translocation elongation factor 1a (tef-1a)
sequences and the results showed that the A. infectoria species–
group constitutes a quite distinct clade (de Hoog and Horré,
2002; Pryor and Bigelow, 2003). Another ITS sequence analysis creating an unrooted radial tree based on maximum likelihood calculation showed that Alternaria malorum (Ruehle) U. Braun, Crous
and Dugan (Cladosporium malorum Ruehle), and the A. infectoria
species–group comprised a single clade (Braun et al., 2003). An inquiry on Alternaria in the CBS database of ITS sequences showed
that a strain of Chalastospora cetera (EG Simmons) EG Simmons
(Alternaria cetera EG Simmons) and three strains of Embellisia abundans EG Simmons also clustered in the same clade as A. malorum
and the A. infectoria species–group (unpublished results).
Many taxa in the A. infectoria species–group are associated with
various species in the grass family (Poaceae L.). They have been reported from stems, straw, leaves and grains of oat, barley, wheat
and rye (Simmons, 1986; Andersen et al., 2002; Dugan and Peever,
2002; Perelló et al., 2008) and are also known to occur on maize
(unpublished results). A. triticina Prasada and Prabhu, a known
plant pathogenic species in the A. infectoria species–group, was
first reported on wheat in India (Prasada and Prabhu, 1962) and later on the same host plant in Argentina (Perelló and Sisterna, 2006)
and in Iran (Simmons, 2007). Other species isolated from discrete
lesions of non-poaceae host plants, such as A. viburni and A. ethzedia, are presumed to have various degrees of pathogenicity (Simmons, personal commun.). In the last decade, taxa in the A.
infectoria species–group have increasingly been isolated from human cutaneous infections, especially from immuno-compromised
patients (de Hoog and Horré, 2002; Dubois et al., 2005; unpublished results).
The objective of this work was to prove the hypothesis that taxa
in Lewia/A. infectoria species–group are sexual fungi and that
molecular sequence analysis and metabolite profiling will yield a
number of clades and clusters that will correspond to the number
of morphological species in the group. Previous research on other
genera has shown that sequence analyses reveal cryptic molecular
species (Taylor et al., 2000; O’Donnell et al., 2004). Studies on Penicillium and large-spored Alternaria have shown that results from
metabolite profiling correlates with the morphological species concept (Andersen et al., 2008; Frisvad and Samson, 2004). Other research on Aspergillus and Stachybotrys has shown that molecular
sequence analyses correlated with the morphological species concept and metabolite profiling (Samson et al., 2007; Andersen et al.,
2003). Furthermore, controversy over whether the A. infectoria species–group comprises many species based on morphological differences or consists of only one species based on molecular sequence
analysis has arisen. Therefore, this polyphasic study of the A. infectoria species–group, including molecular sequence analysis,
metabolite profiling and mating tests, was set up. The study also
included A. malorum, since this species had been reported to belong to the A. infectoria species–group (Braun et al., 2003), together
643
with C. cetera and E. abundans. One aim was to compare chemical
and molecular findings with the current morphological classification and examine which factors correlate and which could resolve
and segregate Alternaria from Chalastospora and Embellisia. Another
aim was to test if other isolates than the human opportunists were
able to grow at 37 °C.
2. Materials and methods
2.1. Fungal strains
Fifty-one fungal strains belonging to the A. infectoria species–
group (39), A. malorum (9), C. cetera (1), and E. abundans (2) were
used. Identity, species–group affiliation, identification number,
host plant, and geographic origin of all strains are given in Table
1. Strains are available from CBS collection, Fungal Biodiversity
Centre, The Netherlands, and IBT collection, Department of Systems Biology, DTU, Denmark.
2.2. Morphological examination
For morphological examination and DNA analysis, each strain
was inoculated in three points onto a PCA plate (potato carrot agar;
Simmons, 2007) and incubated under standardized conditions
(Andersen et al., 2005). In brief: after inoculation the unsealed
plates were incubated in one layer for 7 days at 23 °C under an
alternating light/dark cycle consisting of 8 h of cool-white fluorescent daylight (tubes: TLD, 36 W/95, Philips, Denmark) and 16 h
darkness. Slides for microscopy were made after 7 days using
transparent tape preparations (Butler and Mann, 1959) mounted
in lactophenol. All unidentified strains were compared with type
cultures and descriptions according to Simmons (2007).
For ascoma production under laboratory conditions, strains
were transferred to another PCA plate that had been equipped with
autoclaved wooden toothpicks to encourage the production of
ascomata. In the first trial, each plate was divided into three sectors
with three toothpicks. Three different strains were inoculated on
the same plate, one in each sector. In the second trial, each plate
was divided into two sectors with one toothpick. The same strain
was inoculated in both sectors on the same plate. The plates were
first incubated as ordinary PCA plates under alternating light at
23 °C for 2 weeks. Then the plates were sealed with para filmTM,
placed up-side-down, and incubated at 7 °C in the dark for
6 months.
For the ability to grow at high temperature, strains were transferred (three points) to PDA plates (potato dextrose agar; Difco,
213400). Plates were allowed to stand at room temperature for
one day to ensure that all strains were viable and growing. The
edge of the colonies was marked before incubation. After 12 days
at 37 °C the colony edges were marked again and the plates were
allowed to stand for two days at room temperature. Strains that
had resumed their original growth characteristics were recorded
as positive.
2.3. Metabolite extraction
For metabolite analyses, each strain was transferred (three
points) onto a DRYES plate (dichloran Rose Bengal yeast extract sucrose agar; Frisvad, 1983), a DG18 plate (dichloran 18% glycerol
agar [31.5 g/l dichloran glycerol agar base (OXOID, CM0729);
220 ml/l anhydrous glycerol (JT Baker, 7044); 10 mg/l ZnSO47H2O
(Merck, 8883); 5 mg/l CuSO45H2O (Riedel-de Haën, 12849);
50 mg/l chloramphenicol (Sigma, C-0378); 50 mg/l chlortetracycline hydrochloride (Sigma, C-4881)]), and a PDA + DN plate (potato dextrose agar [39.0 g/l potato dextrose agar (Difco, 213400),
Author's personal copy
644
B. Andersen et al. / Fungal Genetics and Biology 46 (2009) 642–656
Table 1
Host, origin and species–group affiliation of the Alternaria, Chalastospora and Embellisia strains and type cultures (T) used in this study.
#a
Genus species
species–groupb
Host
Origin
CBS #
BA #
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
43
44
45
46
47
48
49
50
51
52
A. infectoria sp.–grp
A. triticina T
A. infectoria T
A. oregonensis T
A. photistica T
A. ethzedia T
A. metachromatica T
A. triticimaculans T
A. infectoria sp.–grp.
A. infectoria sp.–grp.
A. infectoria sp.–grp.
A. intercepta T
A. viburni T
A. arbusti T
A. infectoria sp.–grp.
A. infectoria sp.–grp.
A. infectoria sp.–grp.
A. infectoria sp.–grp.
A. infectoria sp.–grp.
C. cetera T
A. malorum
A. malorum
A. malorum
A. malorum
A. malorum
A. malorum var. polymorpha
A. malorum
A. malorum
A. malorum
E. abundans
E. abundans T
A. infectoria sp.–grp.
A. infectoria sp.–grp.
A. infectoria sp.–grp.
A. infectoria sp.–grp.
A. infectoria sp.–grp.
A. infectoria sp.–grp.
A. infectoria sp.–grp.
A. infectoria sp.–grp.
A. infectoria sp.–grp.
A. infectoria sp.–grp.
A. infectoria sp.–grp.
A. infectoria sp.–grp.
A. infectoria sp.–grp.
A. infectoria sp.–grp.
A. infectoria sp.–grp.
A. infectoria sp.–grp.
A. infectoria sp.–grp.
A. infectoria sp.–grp.
A. infectoria sp.–grp.
A. infectoria sp.–grp.
inf.
inf.
inf.
inf.
inf.
inf.
inf.
inf.
inf.
inf.
inf.
inf.
inf.
inf.
inf.
inf.
inf.
inf.
inf.
Hordeum, grain
Triticum
Triticum
Triticum
Digitalis
Brassica
Triticum
Triticum
Hordeum
Hordeum
Hordeum
Viburnum
Viburnum
Pyrus
Triticum, grain
Triticum, grain
Hordeum, grain
Hordeum, grain
Hordeum, grain
Elymus
Soil
–
–
–
Soil
Vitis
Triticum, straw
Triticum, grain
Gossypium, seed
Dianthus, seed
Fragaria
–
Zea, fresh silage
Zea, fresh silage
Zea, fresh silage
Zea, fresh silage
Zea, fresh silage
Zea, fresh silage
Zea, fresh silage
Zea, fresh silage
Zea, fresh silage
Zea, fresh silage
Zea, fresh silage
Homo, skin lesion
Hordeum
Paeonia
Avena, straw
Homo, skin lesion
Homo, skin lesion
Homo, skin lesion
Zea, fresh silage
DK, Jutland
India
UK
USA, OR
UK
Switzerland
Australia
Argentina
Italy
NZ
NZ
Europe
Europe
USA, CA
DK
DK
DK
DK
DK
Australia
Syria
USA
NZ
NZ
Lebanon
USA, WA
SA, Cape Prov.
USA, OR
Turkey
UK
NZ
UK
DK, Jutland
DK, Jutland
DK, N Jutland
DK, N Jutland
DK, NW Jutland
DK, Jutland
DK, S Jutland
DK, Jutland
DK, Jutland
DK, N Jutland
DK, E Jutland
Germany
UK
NL
–
Italy
NL
Austria
DK, N Jutland
120147
763.84
112250
542.94
212.86
197.86
553.94
578.94
704
1207
1209
1210
1211
1215
1218
1221
1228
1234
1239
1258
1259
1263
1286
1287
1294
1312
1315
1746
1747
1748
1749
1750
1751
1752
1753
1754
1755
1757
1758
1759
1760
1761
1762
1763
1764
1765
1766
JLS08
JLS09
JLS32
JLS33
a
b
inf.
inf.
inf.
inf.
inf.
inf.
inf.
inf.
inf.
inf.
inf.
inf.
inf.
inf.
inf.
inf.
inf.
inf.
inf.
inf.
sp.–grp.
sp.–grp.
sp.–grp.
sp.–grp.
sp.–grp.
sp.–grp.
sp.–grp.
sp.–grp.
sp.–grp.
sp.–grp.
sp.–grp.
sp.–grp.
sp.–grp.
sp.–grp.
sp.–grp.
sp.–grp.
sp.–grp.
sp.–grp.
sp.–grp.
sp.–grp.
sp.–grp.
sp.–grp.
sp.–grp.
sp.–grp.
sp.–grp.
sp.–grp.
sp.–grp.
sp.–grp.
sp.–grp.
sp.–grp.
sp.–grp.
sp.–grp.
sp.–grp.
sp.–grp.
sp.–grp.
sp.–grp.
sp.–grp.
sp.–grp.
sp.–grp.
119406
119407
596.93
120148
120149
120150
120151
120152
110898
173.80
148.66
114810
114809
900.87
112048
266.75
216.65
540.75
535.83
534.83
160.79
EGS #
17–061
27–193
29–194
35–172
37–143
38–132
41–050
42–086
43–070
43–160
49–137
49–147
91–136
41–072
34–063
29–159
102692
116001
106.52
308.53
109785
110803
110804
JLS38
Analysis number in this study. #42 is not included.
Member of the A. infectoria species–group based on morphology.
10 mg/l ZnSO47H2O (Merck, 8883) and 5 mg/l CuSO45H2O (Riedel-de Haën, 12849)] with additional 30 g/l dextrose (BHD,
10117) and additional 3 g/l NaNO3 (Merck, A855537 811). After
inoculation, the DRYES, DG18, PDA + DN plates were packed in perforated plastic bags and incubated in darkness at 25 °C prior to
extraction.
Metabolites were extracted separately from 14-day-old DRYES,
DG18 and PDA + DN cultures. The extractions were done using a
micro-scale extraction method modified for Alternaria metabolites
(Andersen et al., 2005). In brief: three 6-mm agar plugs were cut
from the center of the three colonies and the nine plugs were
placed in a 2-ml vial. Plugs were extracted with 1.0 ml ethyl acetate containing 1% formic acid (v/v) and ultrasonicated for
60 min. Extracts were transferred to clean 2-ml vials, evaporated
to dryness, re-dissolved ultrasonically in 400 ll methanol, and filtered through 0.45-lm PFTE filters (National Scientific Company,
Rockwood, TN, USA) into clean 2-ml vials prior to HPLC analysis.
Experiments were repeated once on DRYES a year after with all
strains and for strains with low metabolite production a third time
on DRYES a month later using 18 plugs.
2.4. HPLC-UV–VIS analyses
Metabolite profiling was performed on an Agilent 1100 HPLC
system (Agilent, Waldbronn, Germany) equipped a diode array
detector collecting two ultraviolet–visible (UV–VIS) spectra per
sec from 200 to 600 nm. Samples of 3.0 ll were separations on a
2 100 mm Luna 3 lm C18(2) (Phenomenex, Torrance, CA, USA)
at 40 °C using a linear water–acetonitrile gradient and a flow of
0.4 ml/min. The gradient started at 15% acetonitrile, reached
100% in 20 min and was held for 5 min. Both eluents contained
50 ppm trifluoro acetic acid. A homologous series of alkylphenones
Author's personal copy
B. Andersen et al. / Fungal Genetics and Biology 46 (2009) 642–656
was analyzed as external retention time references and used to calculate a bracketed retention index (RI) for each detected peak (Frisvad and Thrane, 1987). In this way each metabolite could be
identified by its RI value and its UV–VIS spectrum and be recognized in other extracts.
2.5. HPLC-TOF-MS analyses
Metabolite identification was done on the DRYES extracts using
on a similar HPLC-DAD system as described above with minor
changes: the column was 50 mm long, flow was 0.3 ml/min, and
the eluents were buffered with 20 mM formic acid (Nielsen et al.,
2005). High resolution MS detection was done on a Time Of Flight
mass spectrometer (Water-Micromass, Manchester, UK) scanning
m/z 60–900 and m/z 100–2000 in two separate scan functions at
different skimmer settings (Nielsen et al., 2005). All samples were
analyzed in both positive (ESI+) and negative electrospray (ESI–).
Extracts were also analyzed using a gradient system optimized
for A. infectoria metabolites, using a 2 mm 100 mm 3 lm Gemini
Phenyl column (Phenomenex, Torrance, CA, USA) and a gradient
starting at 10% acetonitrile, reaching 47% in 17 min and then
100% in 3 min and was held for 4 min. All samples were analyzed
in ESI+ and ESI–. Metabolite standards (Nielsen and Smedsgaard,
2003) of 4Z-infectopyrone, AAL-toxins TA and TB, AK toxin I, altenuene, altenusin, alternariol, alternariol monomethylether, altersolanol A, altertoxin I, macrosporin, maculosin, tentoxin and
tenuazonic acid were co-analyzed for verification.
2.6. Data treatment of metabolite profiles
Metabolite profile data from HPLC-UV–VIS were first treated
with the Chemical Image Analysis (CIA) program using an algorithm described by Hansen (2003) as stated by Andersen et al.
(2008). In brief: the raw HPLC data files, which are quantitative
2-D matrices (x-axis: time, y-axis: wave length, value in matrix:
UV–VIS absorbance), were transferred from the HPLC to a standard
PC and analyzed by an in-house written chemical image analysis
(CIA) program (Hansen, 2003). No manipulations or peak selections were made before processing. Each HPLC file was processed
first by a log10 scaling (to account for concentration differences
among extracts), then a baseline correction and finally an alignment (to account for drift in baseline and retention time among
identical metabolites in different runs) (Hansen, 2003). Each HPLC
file was then compared to the other 50 HPLC files, pair-wise, using
an algorithm described by Hansen (2003) giving a similarity value
for each pair, which was entered into a new matrix. The resulting
51 51 similarity matrix was then used to calculate a dendrogram
using WARD clustering method.
Based on the result of the CIA, a binary matrix was made manually by scoring each metabolite from four printed HPLC chromatograms as present or absent (137 metabolites for the 51 fungal
strains) and subjected to multivariate statistics using Unscrambler
version 9.2 (CAMO ASA, Oslo, Norway). The matrix was analyzed
using Partial Least Squares Regression (PLS-R) and Principal Component Analysis (PCA). A reduced matrix (124 metabolites and 49
strains) was subjected to cluster analysis using NTSYS-pc version
2.11 N (Exeter Software, Setauket, NY, USA) without standardization using Yule as correlation coefficient and Unweighted PairGroup Method using arithmetic Averages (UPGMA) as clustering
method. The matrix was also analyzed by simple matching and Jaccard similarity coefficients in NTSYS.
For metabolite identification and peak comparison, 12–20 of the
largest peaks in the data files from HPLC-TOF-MS were inspected in
UV–VIS, ESI+, ESI– modes were compared to peaks registered in the
Quanlynx 4.1 software (Water-Micromass) using a ±m/z 0.02 and a
retention time limit of ±0.3 min. If possible, a qualifier ion was
645
used for confirmation. Peaks of the metabolite standards mentioned above were also inserted along with predicted ions calculated from the masses of known Alternaria metabolites. In the
latter case a ±3 min window was used, with the retention time predicted using the log D of the compound (calculated using ACD v.10,
Advanced Chemical development Inc., Toronto, Canada), which
was correlated to retention time of 50 representative secondary
metabolites. All extracts were then analyzed by the Quanlynx software. For metabolite identification, each peak was matched against
an internal reference standard database (800 compounds) as well
as tentatively identified by searching the accurate mass in the
34,392 compounds in Antibase 2008 (Laatsch, 2008), comparing
UV–VIS data, fragmentations, ionization efficiency in ESI– versus
ESI+ and the retention time to information in the databases.
2.7. DNA extraction, PCR amplification and sequencing
For DNA analysis, each strain was inoculated in three points onto
a PCA plate and incubated as mentioned above. DNA was extracted
from 7-day-old PCA cultures with UltraClean Microbial DNA isolation kit (Mo Bio Laboratories, Solana Beach, CA, USA) according to
manufacturer’s protocol and stored at 20 °C. Amplifications of
the three target genes were performed with the following primer
combinations: ITS: V9G (de Hoog and Gerrits van den Ende, 1998)
and ITS4 (White et al., 1990): gpd: gpd1 and gpd2 (Berbee et al.,
1999), tef-1a: EF1–645F: (TCG TCG TYA TCG GMC ACG TCG A) and
EF1–1190R (TAC CAG TGA TCA TGT TCT TGA TGA). EF1–645F and
EF1–1190R were designed by aligning sequences of Gibberella
zeae (XM388987), N. crassa (D45837), Aspergillus fumigatus
(XM745295) and Ustilago maydis (XM751978) from GenBank. The
numbers of the primers refer to the nucleotide position in N. crassa
(D45837) at the 30 -end as done previously (Carbone and Kohn,
1999). ITS and gpd PCR reactions were performed in 12.5 ll volumes
containing 0.5 ll DNA, 1 NHþ
4 -buffer [160 mM (NH4)2SO4, 670 mM
Tris–HCI (pH 8.8 at 25 °C), 0.1% Tween-20] (Bioline, London, UK),
1 mM MgCl2, 0.04 mM dNTPs, 0.2 pmol of each primer and 0.5 U
BIOTAQTM DNA Polymerase (Bioline, London, UK). Tef-1a PCR reactions were performed with 0.8 pmol of each primer and 2.5 mM
MgCl2 and with the same concentrations of the remaining ingredients as for ITS and gpd. All PCR amplifications were performed after
the same scheme with an initial denaturation at 94 °C for 5 min followed by 40 amplification cycles of 94 °C for 30 s, annealing for 30 s
and 72 °C for 1:20 min, and a final extension at 72 °C for 7 min.
Annealing temperatures were: 48 °C (ITS), 52 °C (tef-1a) and 59 °C
(gpd). Amplicons were run on 1% agarose gels and visualised with
UV after ethidium bromide staining. Sequence reactions were performed with the BigDyeÒ Terminator v1.1 Cycle Sequencing Kit (Applied Biosystems, Foster City, CA). All sequence-PCR reactions were
performed with the same protocol: 94 °C for 1 min, followed by 30
cycles of 94 °C for 10 s, 50 °C for 5 s and 60 °C for 4 min. DNA was
purified with SephadexÒ G-50 (Pharmacia-Amersham) and sequenced. All sequences determined in this study have been submitted to GenBank and accession numbers are listed in Table 2.
2.8. Data treatment of molecular sequences
Sequence electrophorograms of forward and backward runs
were combined, analyzed, edited using DnaStar SeqMan II (LaserGene). Sequence data were saved and aligned with BioNumerics
(Applied Maths, Kortrijk, Belgium). The alignments of tef-1a, ITS
and gpd were concatenated into one alignment to construct a phylogenetic tree. The program RaxML (http://www.phylo.org/portal/
Home.do) was used to create the best tree using maximum likelihood and to calculate bootstrap values (Stamatakis et al., 2008).
The same program and conditions were used to create individual
trees of all three markers used in this study. Genetic diversity of
Author's personal copy
646
B. Andersen et al. / Fungal Genetics and Biology 46 (2009) 642–656
Table 2
Ascoma production on PCA and growth at 37 °C on PDA together with gpd haplotypes and GenBank accession numbers.
#a
Genus (species/group)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
43
44
45
46
47
48
49
50
51
52
Alternaria infectoria sp.–grp.
A. triticina T (inf–grp)
A. infectoria T (inf–grp)
A. oregonensis T (inf–grp)
A. photistica T (inf–grp)
A. ethzedia T (inf–grp)
A. metachromatica T (inf–grp)
A. triticimaculans T (inf–grp)
A. infectoria sp.–grp.
A. infectoria sp.–grp.
A. infectoria sp.–grp.
A. intercepta T (inf–grp)
A. viburni T (inf–grp)
A. arbusti T (inf–grp)
A. infectoria sp.–grp.
A. infectoria sp.–grp.
A. infectoria sp.–grp.
A. infectoria sp.–grp.
A. infectoria sp.–grp.
Chalastospora cetera T
A. malorum
A. malorum
A. malorum
A. malorum
A. malorum
A. malorum var. polymorpha
A. malorum
A. malorum
A. malorum
Embellisia abundans
E. abundans T
A. infectoria sp.–grp.
A. infectoria sp.–grp.
A. infectoria sp.–grp.
A. infectoria sp.–grp.
A. infectoria sp.–grp.
A. infectoria sp.–grp.
A. infectoria sp.–grp.
A. infectoria sp.–grp.
A. infectoria sp.–grp.
A. infectoria sp.–grp.
A. infectoria sp.–grp.
A. infectoria sp.–grp.
A. infectoria sp.–grp.
A. infectoria sp.–grp.
A. infectoria sp.–grp.
A. infectoria sp.–grp.
A. infectoria sp.–grp.
A. infectoria sp.–grp.
A. infectoria sp.–grp.
A. infectoria sp.–grp.
a
b
Ascoma production
1. Trial
2. Trial
+
+
+
+
+
+
+
+
+
+
+
+
+
+
ntb
nt
nt
nt
nt
nt
nt
nt
nt
nt
nt
nt
nt
nt
nt
nt
nt
nt
nt
nt
nt
nt
nt
nt
nt
nt
nt
nt
nt
nt
nt
nt
+
+
+
+
+
+
+
+
+
+
+
+
+
+
nt
+
nt
nt
nt
nt
nt
Growth at 37 °C
gpd
ITS
gpd
tef-1a
FJ214885
AY762948
FJ214897
AY762947
FJ214900
AY278833
AY762946
AY762949
FJ214890
FJ214886
FJ214859
FJ214882
FJ214876
FJ214857
FJ214887
FJ214863
FJ214865
FJ214866
FJ214868
FJ214864
FJ214888
FJ214870
FJ214894
FJ214861
FJ214860
FJ214883
FJ214884
FJ214895
FJ214896
FJ214898
AB120848
FJ214872
FJ214873
FJ214862
FJ214880
FJ214867
FJ214871
FJ214875
FJ214881
FJ214877
FJ214879
FJ214892
FJ214878
FJ214856
FJ214899
FJ214858
FJ214889
FJ214869
FJ214874
FJ214891
FJ214893
FJ214836
FJ214846
FJ214850
FJ214849
FJ214854
FJ214855
FJ214835
FJ214834
FJ214841
FJ214837
FJ214808
FJ214831
FJ214825
FJ214806
FJ214838
FJ214812
FJ214814
FJ214815
FJ214817
FJ214813
FJ214839
FJ214819
FJ214845
FJ214810
FJ214809
FJ214832
FJ214833
FJ214847
FJ214848
FJ214851
FJ214852
FJ214821
FJ214822
FJ214811
FJ214829
FJ214816
FJ214820
FJ214824
FJ214830
FJ214826
FJ214828
FJ214843
FJ214827
FJ214805
FJ214853
FJ214807
FJ214840
FJ214818
FJ214823
FJ214842
FJ214844
FJ214932
FJ214942
FJ214946
FJ214945
FJ214950
FJ214951
FJ214931
FJ214930
FJ214937
FJ214933
FJ214904
FJ214927
FJ214921
FJ214902
FJ214934
FJ214908
FJ214910
FJ214911
FJ214913
FJ214909
FJ214935
FJ214915
FJ214941
FJ214906
FJ214905
FJ214928
FJ214929
FJ214943
FJ214944
FJ214947
FJ214948
FJ214917
FJ214918
FJ214907
FJ214925
FJ214912
FJ214916
FJ214920
FJ214926
FJ214922
FJ214924
FJ214939
FJ214923
FJ214901
FJ214949
FJ214903
FJ214936
FJ214914
FJ214919
FJ214938
FJ214940
Haplotype
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
nt
+
+
nt
nt
nt
nt
nt
1
3
2
4
nt
5
6
7
21
22
22
8
9
10
11
12
2
13
9
nt
nt
nt
nt
nt
nt
nt
nt
nt
nt
nt
nt
6
23
24
15
15
15
9
15
19
9
2
20
14
15
6
16
17
15
18
15
Analysis number in this study. #42 is not included.
Not tested.
aligned sequences and the standardized Index of Association ðIsA Þ
was calculated with LIAN 3.5 (Haubold and Hudson, 2000) by combining clustering information of the individual trees. The phylogenetic network was created with SplitsTree v4.8. The network
structure was based on the neighbornet algorithm with a threshold
set to 104 and applying the LogDet transformation. LogDet is a
distance transformation that corrects for biases in the base composition (Wägele and Mayer, 2007). The phi-test incorporated in the
SplitsTree software (Huson and Bryant, 2006) was used to test signals of recombination (p < 0.05, significant evidence of recombination). The test is proven to be a robust calculation and no previous
knowledge about population history, recombination rate, mutation
rate and rate heterogeneity across sites (Bruen et al., 2006) is necessary. Although large splits in networks do not necessarily imply
recombination, split decomposition networks in conjunction with
the phi-test can easily detect which sequences in a given data set
contribute the most to the recombination signal (Salemi et al.,
2008). The phi-test is repeated after possible recombinants are deleted from the alignment until p > 0.05 (no evidence of recombination). DnaSP v4.5 (Rozas and Rozas, 1995) was used to find the
different haplotypes in the gpd alignment. Gaps and missing data
were not considered during calculation.
3. Results
3.1. Morphology
Examination of the sporulation patterns on PCA after seven days
of growth in alternating light showed three different and very distinct morphologies as shown in Fig. 1. A. malorum, A. malorum var.
Author's personal copy
B. Andersen et al. / Fungal Genetics and Biology 46 (2009) 642–656
647
Fig. 1. Morphology plates. (A): Chalastospora cetera (#20), (B): Alternaria malorum (#25), (C): Embellisia abundans (#31) and six strains belonging to the A. infectoria species–
group: (D): A. oregonensis (#04), (E): A. infectoria (#03), (F): A. arbusti (#14) and (G–I): three A. infectoria species–group strains (#32, #44 and #17, respectively). Bar in
G = 50 lm.
polymorpha and C. cetera all had cylindrical didymo- and phragmoconidia produced in very long branching chains (Fig. 1A and B).
Both strains of E. abundans had solitary, ovoid phragmoconidia
(Fig. 1C), while all strains of the A. infectoria species–group shared
the same morphology: ovoid, obpyriform or obclavate phragmoand dictyoconidia with secondary conidiophores of varying length
produced in branched chains (Fig. 1D–I). Type cultures fitted the
descriptions, except A. arbusti (#14), which, under the growth conditions in this study, showed a more branched three-dimensional
structure than the original description depicted (Simmons, 2007).
Conidial sizes, shapes, ornamentation, color etc. varied greatly between strains, whereas conidial appearance was quite consistent
within a strain: e.g. conidia of A. infectoria (#03) were smooth, narrow-obpyriform and sallow colored (Fig. 1E), while conidia of A.
infectoria species–group (#44) were narrow-obpyriform to broadovoid, finely rough and bronze colored (Fig. 1H). None of the
strains morphologically identified as belonging to the A. infectoria
species–group could be assigned to any of the type cultures and
no two strains, except A. infectoria species–group strains (#10
and #11), could be classified as belonging to the same taxon.
Table 2 shows the production of proascomata (ascomata without mature ascospores) after 6 months of incubation on PCA at
7 °C. In the first trial, where three strains were inoculated on the
same PCA plate, no mating between strains was observed. Proascomata (Fig. 2) were formed in the center of each colony and in areas
in the agar furthest away from the two other colonies and there
was a clear demarcation line between colonies. In the second trial,
where the same strain had been inoculated twice on the same
plate, proascomata were again formed at the center and furthest
away from the other colony. Fourteen out of nineteen A. infectoria
species–group strains produced proascomata in the first trial, and
only seven of these in the second trial. Additional 14 strains were
tested in the second trial, out of which five produced proascomata.
In the second trial, neither C. cetera nor E. abundans produced proascomata, but three out of the nine A. malorum did. In contrast to
the strains of the A. infectoria species–group, the three A. malorum
strains produced their proascomata on the toothpicks and not in
the agar. When the experiment was terminated after 6 months,
none of the proascomata had yielded any matured ascospores.
3.2. Growth on different media
The result of the experiment on PDA at 37 °C given in Table 2
showed that 28 out of the 45 tested strains were able to grow at
this temperature. Unfortunately, it was not possible to test all 51
strains in the set, since six strains were not viable after the first
chemical experiments. Table 2 shows that all nine A. malorum
strains were able to grow at 37 °C, but not C. cetera. During the
incubation period, the colonies of A. malorum became dark brown
and the mycelium thinner and more thread-like. Of the 33 tested
strains of A. infectoria species–group, 20 were able to grow at
37 °C. The colonies lost their aerial mycelium, became waxy in
their growth, and only produced filamentous mycelium after they
were taken out of the 37 °C incubator. The only viable strain of A.
infectoria species–group from a human infection (#45) was not
able to grow on PDA at 37 °C.
Examination of the colony appearance on DRYES incubated at
25 °C showed that most of the 51 strains produced hairy to granu-
Author's personal copy
648
B. Andersen et al. / Fungal Genetics and Biology 46 (2009) 642–656
Fig. 2. Proascomata in 3-week-old PCA cultures. (A): Alternaria infectoria species–group #19 (bar = 500 lm) and (B): A. infectoria species–group #16 (bar = 200 lm).
lar mycelium, whitish to grayish in color. Some of the A. infectoria
species–group strains produced only wet, yeast-like colonies on
DRYES, especially the strains isolated from human lesions, which,
however, did not affect their metabolite production. Filamentous
mycelium was produced when the strains were transferred to
PDA or PCA. Sporulation of the A. infectoria species–group was only
seen on PCA and sometimes scarification and prolonged incubation
were needed, while A. malorum, C. cetera and E. abundans sporulated well on PCA as well as on PDA + DN and PDA. DRYES and
DG18 did not accommodate any sporulation. As a curiosity, it
should be mentioned that A. metachromatica (#07) on PDA + DN
produced an extracellular pigment that turned the reverse of the
plate dark blue.
3.3. Chemical classification
The chemical HPLC-UV–VIS analyses of the three growth media,
DRYES, DG18 and PDA + DN, showed that the metabolite production, qualitative as well as quantitative, was greatest on DRYES
and that metabolite production seemed to be inhibited on
PDA + DN. Fig. 3 shows six selected HPLC chromatograms made
from the DRYES extracts.
Automated and unbiased chemical image analyses (CIA) of the
HPLC-UV–VIS files were made with extracts from all three growth
media to aid the selection of species-specific metabolites (dendrograms not shown). All three dendrograms gave the same grouping,
but the one made with extracts from DRYES gave the most detailed
dendrogram. It showed that the 51 strains grouped in four major
clusters. One cluster contained all eight A. malorum strains, the C.
cetera (#20) strain, and the type culture of E. abundans (#31). A second cluster contained 13 A. infectoria species–group strains, including the type cultures of A. infectoria (#03), A. ethzedia (#06), and A.
metachromatica (#07), together with E. abundans (#30) and A.
malorum var. polymorpha (#26). A third cluster contained 10 A.
infectoria species–group strains, including the type cultures of A.
intercepta (#12) and A. viburni (#13), while the last cluster contained 12 A. infectoria species–group strains, including type cultures of A. photistica (#05), A. arbusti (#14), A. oregonensis (#04),
A. triticina (#02) and A. triticimaculans (#08). Visual examination
of the HPLC chromatograms of E. abundans (#30) and A. malorum
var. polymorpha (#26) showed very few peaks, meaning a very
low metabolite production. The location of these two stains in
the dendrogram was therefore questionable and they where subsequently removed from further analyses.
A binary matrix was made by visual examination of each peak
and its associated UV–VIS spectrum in the HPLC chromatograms
and the CIA dendrograms. On no occasion were there metabolites
produced on DG18 or PDA + DN that were not found in DRYES,
however, some metabolites were easier to detect on DG18 due to
a qualitatively simpler metabolite profile. The matrix consisted of
137 recognizable metabolites for the 51 strains and was subjected
to a Partial Least Squares Regression (PLS-R) (result not shown),
which gave metabolites specific for the A. malorum strains, the C.
cetera strain, the A. infectoria species–group strains and the E. abundans strains, respectively (see Table 3). All 51 chromatograms were
analyzed for the production of known Alternaria metabolites and
tested negative for AAL-toxins, alternariols, altersolanols, altenuenes, tentoxin and tenuazonic acid. On the other hand, HPLC-UV–
VIS as well as HPLC-MS analyses showed that all 51 strains, including E. abundans (#30) and A. malorum var. polymorpha (#26), were
able to produce infectopyrone and 4Z-infectopyrone. Altertoxin derived metabolites were restricted to the A. infectoria species–group,
but not produced consistently throughout the species–group,
while macrosporin was found in three A. malorum strains. Novaezelandin production was shared by E. abundans and the A. infectoria
species–group, but again not produced consistently throughout the
species–group. A. malorum and C. cetera had a number of metabolites of unknown structure in common (e.g. RI value 694), but also
produced unknown metabolites specific to each species (e.g. RI values 715 and 894, respectively). E. abundans did not produce any
known metabolites, except for the infectopyrone, 4Z-infectopyrone, and novae-zelandin A, and only few species specific metabolites (e.g. RI values 691 and 779a) were detected.
Based on the combined results of the CIA and the PLS-R, a matrix that consisted of 49 strains [excluding E. abundans (#30) and A.
malorum var. polymorpha (#26)] and 124 known and unknown
metabolites (excluding infectopyrones and other consistently produced metabolites) was constructed. The matrix was subjected to
cluster analysis and the resulting dendrogram is shown in Fig. 4.
The dendrogram shows a division of the 49 strains into two clusters, A (in light grey) and B. Cluster A contains the type culture
of E. abundans (#31), the type culture of C. cetera (#20) together
with six A. malorum strains in sub-cluster A1 and two A. malorum
strains (#22 and #28) in sub-cluster A2. Cluster B, holding all the
39 strains of the A. infectoria species–group, could be divided into
two sub-clusters with type cultures of A. photistica (#05) and A. triticina (#02) as outliers. Sub-cluster B1 contained six A. infectoria
species–group strains and the type cultures of A. triticimaculans
(#08), A. intercepta (#12) and A. viburni (#12) (marked with ),
while the majority of the A. infectoria species–group strains clustered closely together in sub-cluster B2. Both B sub-clusters could
be further divided based on production of different metabolites.
The color-coding of strains in Fig. 4 is referring to different
haplotypes.
The result of a principal component analysis of the 39 A. infectoria species–group strains and 79 metabolites (given as their RI
Author's personal copy
649
B. Andersen et al. / Fungal Genetics and Biology 46 (2009) 642–656
A
Embellisia abundans (#31)
100
75
50
25
0
0
B
5
10
15
20
25
Chalastospora cetera (#20)
100
75
50
25
0
0
C
5
10
15
20
25
Alternaria infectoria sp-grp (#33)
100
75
50
25
0
0
D
5
10
15
20
25
Alternaria oregonensis (#04)
100
75
50
25
0
0
E
5
10
15
20
25
Alternaria infectoria sp-grp (#01)
100
75
50
25
0
0
F
5
10
15
20
25
Alternaria triticina (#02)
100
75
50
25
0
0
5
10
15
20
25
Fig. 3. HPLC chromatograms (wave length 210 nm). (A): Embellisia abundans (#31), (B): Chalastospora cetera (#20) and four strains belonging to the Alternaria infectoria
species group; (C): A. infectoria species–group (#33), (D): A. oregonensis (#04), (E): A. infectoria species–group (#01) and (F): A. triticina (#02).
values) is shown in Fig. 5. It shows the association between metabolites and the strains that produce them. Metabolites common to
most strains are located in the center, while metabolites specific
to a few strains are located around the perimeter together with
Author's personal copy
650
B. Andersen et al. / Fungal Genetics and Biology 46 (2009) 642–656
Table 3
Production of known metabolites and selected, unknown metabolites, given by their retention index (RI) value, from Alternaria infectoria species–group, A. malorum, Chalastospora
cetera and Embellisia abundans.
Metabolite (RI)a
b
Altertoxin derived (855)
Altertoxin derived (820)
Altertoxin derived (846)
Infectopyrone (839)c
4Z-Infectopyrone (824a)
Infectopyrone derived (706)
Infectopyrone derived (713b)
Macrosporin (1068)
Novae-zelandin B (892)
Novae-zelandin A (680a)
Novae-zelandin derived (726)
569
950d
691
779a
715
894
694
1010
642
1120
1076e
1047
752
a
b
c
d
e
f
A. infectoria sp–grp (39)
A. malorum (9)
C. cetera (1)
E. abundans (2)
31
9
4
39
39
30
23
f
29
19
6
19
13
39
35
26
9
9
2
1
3
5
7
4
7
6
7
5
5
1
1
1
1
1
1
1
1
1
2
2
2
1
1
1
1
2
1
1
RI: retention index value, calculated by the HPLC from retention time.
Same as metabolite 3 in Andersen and Thrane (1996).
Same as metabolite 2 in Andersen and Thrane (1996).
Same as metabolite 5 in Andersen and Thrane (1996).
Same as metabolite 6 in Andersen and Thrane (1996).
Not detected.
the strains producing them. From Fig. 5 it can be seen that strains
in sub-cluster B1 in Fig. 4 (marked with ) produced a large number of metabolites including the altertoxins and novae-zelandins,
whereas strains in sub-cluster B2 produce fewer metabolites and
not altertoxins or novae-zelandins. On the other hand, strains in
B2 produce metabolites of unknown structure (e.g. RI values 569,
706, 713b, 752, 813), which are not produced by strains in B1. In
general, many individual metabolites of unknown structure were
found to be specific to only one or a few strains in the A. infectoria
species–group, which hampered a clear grouping.
3.4. Molecular cladification
The obtained sequences of gpd were 444–446 bp for A. photistica
and the A. infectoria species–group and of 424 bp for A. malorum, E.
abundans, and C. cetera. The aligned gpd sequences contained one
intron of approximately 114 bp. The alignment dataset of all the
strains contained 456 bp with 131 variable sites of which 85 were
parsimony informative. Sequences of tef-1a were 437–440 bp containing two introns of approximately 250 bp in total. The tef-1a
alignment dataset consisted of 443 bp containing 113 variable sites
of which 65 were parsimony informative. The obtained ITS sequences were 490 bp for A. photistica, 519–525 bp for the A. infectoria species–group strains, 533–534 bp for the A. malorum strains
and 523 bp and 525 for the E. abundance and C. cetera strains,
respectively. The ITS sequence for A. photistica was smaller than
those of the remaining strains due to a major deletion in ITS1.
The ITS alignment dataset of all the strains contained 544 bp with
106 variable sites of which 60 were parsimony informative. The ITS
and tef-1a dendrograms gave the same major division as the gpd
dendrogram, but with lower resolution (data not shown). Strains
that were identical in one gene sequence were nearly always different in another. Molecularly, all the A. infectoria species–group
strains were mutually similar, but never identical, except for A.
infectoria species–group strains (#10 and #11), which have identical sequences in all three tested genes.
Fig. 6 shows an unrooted dendrogram for all 51 strains of the
concatenated ITS, tef-1a and gpd sequences using maximum likelihood in RaxML. It shows two major clades, one with 38 A. infectoria
species–group strains and with A. triticina (#02), three strains of A.
infectoria species–group (#09, #48 and #51) and A. photistica as
outliers and another clade with all A. malorum, C. cetera and E.
abundans strains (in light grey). Within the latter clade, A. malorum
var. polymorpha (#26) could not be distinguished from the remaining eight A. malorum strains. Nearest neighbors of A. malorum were
C. cetera and E. abundans. The degree of variability within the A.
infectoria species–group proved to be limited in all genes.
Fig. 7 shows the nucleotide differences between 38 strains in
the A. infectoria species–group [excluding A. photistica (#05)]. As
seen in Fig. 7, most nucleotide differences in the three genes were
observed in the spacers and introns, although there were some
mutations in the coding region of gpd, which all occurred on the
third codon position often with a silent C to T substitution. Table
4 shows alignment data for the three genes. Internal ITS alignment
of the 38 strains in the A. infectoria species–group resulted in 40
variable sites of which 35 were parsimony informative and located
in either the ITS1 or ITS2. The tef-1a alignment resulted in 113 variable sites of which 53 were parsimony informative. The gpd alignment showed 109 variable and 59 parsimony informative sites of
which 18 were located in the 114 bp intron.
Using DnaSP on the gpd alignment data of the 38 strains in the
A. infectoria species–group, haplotypic groups were defined and are
given in Table 2. Most haplotypic groups contained only one strain
except for haplotype 2 (#3, #17 and #43), haplotype 6 (#7, #32
and #47), haplotype 9 (#13, #19, #38 and #41), haplotype 15
(#35–37, #39, #46, #50 and #52) and haplotype 22 (#10–11),
resulting in 24 distinct haplotype groups. Fig. 8 shows the haplotype network. The standardized Index of Association ðIsA Þ of the
Author's personal copy
B. Andersen et al. / Fungal Genetics and Biology 46 (2009) 642–656
A1
A
A2
B1
B2
B
-0.20
0.10
0.40
0.70
651
E. abundans Fragaria 31 (T)
C. cetera Elymus 20 (T)
A. malorum Triticum 27
A. malorum Soil 25
A. malorum Gossypium 29
A. malorum Soil 21
A. malorum - 24
A. malorum - 23
A. malorum Triticum 28
A. malorum - 22
A. intercepta Viburnum 12 (T)
A. viburni Viburnum 13 (T)
A. infectoria Zea 52
A. infectoria Zea 36
A. infectoria Zea 34
A. infectoria Zea 33
A. infectoria Zea 37
A. infectoria Hordeum 46
A. triticimaculans Triticum 08 (T)
A. infectoria Zea 44
A. infectoria Hordeum 18
A. infectoria Avena 48
A. infectoria Triticum 03 (T)
A. infectoria Homo sapiens 51
A. infectoria Homo sapiens 49
A. infectoria Paeonia 47
A. arbusti Pyrrus 14 (T)
A. infectoria Hordeum 09
A. oregonensis Triticum 04 (T)
A. metachromatica Triticum 07 (T)
A. infectoria Zea 43
A. infectoria Straw 32
A. infectoria Zea 40
A. infectoria Hordeum 19
A. infectoria Zea 38
A. infectoria Homo sapiens 45
A. ethzedia Brassica 06 (T)
A. infectoria Zea 41
A. infectoria Zea 35
A. infectoria Hordeum 17
A. infectoria Triticum 15
A. infectoria Hordeum 11
A. infectoria Hordeum 10
A. infectoria Triticum 16
A. infectoria Homo sapiens 50
A. infectoria Zea 39
A. infectoria Hordeum 01
A. photistica Digitalis 05 (T)
A. triticina Triticum 02 (T)
1.00
Yule/UPGMA
Fig. 4. Dendrogram based on a cluster analysis of 49 metabolite profiles (1 Embellisia abundans, 1 Chalastospora cetera, 8 Alternaria malorum and 39 strains belonging to the A.
infectoria species–group). Color-coding in the B cluster corresponds to haplotype groups given in Table 2 and Fig. 8. Strain labels: strain ID/host/strain number/type culture.
Dendrogram calculated using the Yule correlation coefficient and UPGMA as the clustering method. Axis shows the correlation coefficient from 1 to 1.
same A. infectoria species–group strains showed a tendency
towards recombination events, ðIsA Þ = 0.1627. LIAN v3.5 was
used to calculate the standardized Index of Association with
1,000,000 Monte Carlo samplings. The neighbornet splitstree of
gpd alignment data (not shown) of the A. infectoria species–group
showed mostly a treelike structure. The network also showed conflicting phylogenetic trees (histories) that can not be shown with a
bifurcating tree. Conflicting phylogenetic signals can occur by
recombination or by convergent substitutions and can not be
distinguished by looking at the network alone (Salemi et al.,
2008). However, a phi-test was able to detect the presence of
recombination in aligned sequences. Repeated phi-test calculations
after removing single sequences from the alignment showed the
presence of recombinants. When the p-value increased till 0.05
or more, it was obvious that the recombinants were deleted from
the alignment. Table 4 shows the p-values for the three genes in
the phi-test of the 38 A. infectoria species–group strains [excluding
A. photistica]. ITS and tef-1a had p-values > 0.05, which indicated
Author's personal copy
652
1.0
B. Andersen et al. / Fungal Genetics and Biology 46 (2009) 642–656
PC2
833a
752
713b
706
A. infectoria 39
1047
869
910
A. infectoria 19
813
0.5
A. metachromatica 07 (T)
792b
A. infectoria 18
A. infectoria 17
A. infectoria
49 47
infectoria
A.A.
A.infectoria
infectoria4803 (T)569
z-Inf-824a1118
681
A. infectoria 51
937a
685
A. infectoria 01
950729
-0.5
A. infectoria 34
Alx-820
A. infectoria 38
758
878
777
723
781b
977
1076 750c 936 757b
1279 871
A. infectoria 44 A. ethzediaA.
765c
749a 713c
06infectoria
(T) 773605
1315
A.32
infectoria
15
602b
633 A. infectoria
16 A. infectoria 10
09
A. infectoria
753938b
712
Inp
766b
A. infectoria
45 43 A. infectoria
937b 793
938a 11
822
808
765b
847
A. infectoria
563
Alx-846 Alx-855750b636
721
757a
788
952
1024
1011
588 765a
NZ-B-892
979602a 1440713a
939
824b
738
749b
A. infectoria 40
816b
816a
NZ-A-680a
A. arbusti 14 (T)A. infectoria 750a
41
1549
A. oregonensis 04 (T)
726
2047
0
A. viburni 13 (T)
A. intercepta 12 (T)
A. infectoria 36
A. infectoria 52
A. infectoria 33
A. infectoria 50
A. infectoria 35
A. photistica 05 (T)
A. infectoria
37 46
A. infectoria
Sub-cluster B1
A. triticimaculans 08 (T)
A. triticina 02 (T)
-1.0
-0.6
-0.4
-0.2
0
0.2
PC1
0.4
0.6
0.8
1.0
Fig. 5. Loadings plot based on a principal component analysis of 79 individual metabolites (39 strains belonging to the A. infectoria species–group). Strains are in black and
metabolites in blue. Metabolites of known structure: Alx: altertoxin derivatives; NZ: novae-zelandin derivatives; z-Inf: infectopyrone derivate. Metabolites of unknown
structure are only given by their Retention Index (RI) values calculated from their retention time on HPLC. Strain labels: strain ID/strain number/type culture. Strains in the
grey box marked with correspond to sub-cluster B1 in Fig. 4. Axes are score values.
the absence of recombination. However, the p-value of gpd was
7.5 104 suggesting the presence of recombination events in this
gene. After deleting strains #08, #45 and #51 (CBS 578.94, CBS
102692 and CBS 110804, respectively), the phi-test showed no significant evidence of recombination in the gpd data (p > 0.05) and
therefore these three strains were considered to be recombinants.
4. Discussion
At genus level, the Alternaria infectoria species–group could
clearly be separated from the A. malorum/C. cetera/E. abundans
group based on morphology as well as chemical classification
and molecular cladification. The results show that strains morphologically identifiable as A. infectoria species–group produced altertoxins and novae-zelandins and yielded ITS, gdp and tef-1a
sequences that were different from those of the A. malorum/C. cetera/E. abundans group (Figs. 1, 4 and 6 and Table 3). E. abundans, on
the other hand, could only be segregated from the A. malorum/C.
cetera group by morphology (Fig. 1), but not with any confidence
by molecular or chemical means. Besides, the chemical similarity
turned out to be too large and the number of species/strains used
proved to be too few to speculate on the placement of Embellisia,
compared to A. malorum/C. cetera group. Other studies based on
ITS, SSU and gpd data, show species of Embellisia are scattered
among genus Alternaria as well as genus Ulocladium despite its distinct morphology (Pryor and Bigelow, 2003). In contrast, C. cetera
could not be segregated from the A. malorum/A. malorum var. polymorpha group by morphological, molecular, or chemical means
(Figs. 1, 4 and 6 and Table 3). Two A. malorum strains (#23 and
#24) were distinct and identical in all three methods. The other
A. malorum strains yielded metabolite profiles that were similar,
but not identical to each other and to that of C. cetera. Morphologically, A. malorum and C. cetera showed the same general sporulation pattern, but with some variation in conidial size and septation.
The polyphasic data in this study shows that A. malorum var. polymorpha and the eight A. malorum strains, do not belong in the A.
infectoria species–group as proposed by Braun et al. (2003), but
suggest they belong to the same genus as Chalastospora cetera,
however, as several distinct species.
The production of infectopyrones and a pair of compounds with
unknown structure (RI values 1047 and 1076) by all genera in this
study corroborates the close relationship found in the molecular
analyses, but on the other hand, some species of Ulocladium, which
are phylogenetically more related to small-spored Alternaria (Pryor
and Bigelow, 2003), also produce infectopyrones (Andersen and
Hollensted, 2008) and these metabolites may be more widespread
in Pleosporaceae. Furthermore, this is the first report on the production of infectopyrones by A. malorum, C. cetera, and E. abundans and
the production of macrosporin and novae-zelandin A by A. malorum and E. abundans, respectively.
At species level, the 10 Alternaria type cultures representing
morphological species within the A. infectoria species–group were
Author's personal copy
B. Andersen et al. / Fungal Genetics and Biology 46 (2009) 642–656
653
A. infectoria Triticum 03 (T)
A. infectoria Hordeum 17
A. infectoria Zea 43
A. infectoria Homo sapiens 45
A. intercepta Viburnum 12 (T)
A. infectoria Zea 41
81
A. viburni Viburnum 13 (T)
A. infectoria Hordeum 19
A. infectoria Zea 38
98
A. infectoria Hordeum 01
A. infectoria Hordeum 11
71
A. infectoria Hordeum 10
A. infectoria Zea 44
A. infectoria Triticum 16
A. oregonensis Triticum 04 (T)
A. arbusti Pyrus 14 (T)
A. infectoria Zea 33
A. infectoria Zea 34
A. infectoria Paeonia 47
A. metachromatica Triticum 07 (T)
A. infectoria Straw 32
A. triticimaculans Triticum 08 (T)
79
A. infectoria Zea 37
A. infectoria Hordeum 46
A. infectoria Zea 36
A. infectoria Homo sapiens 50
81
A. infectoria Triticum 15
A. infectoria Zea 52
A. infectoria Zea 39
A. infectoria Zea 35
A. infectoria Hordeum 18
A. ethzedia Brassica 06 (T)
A. infectoria Zea 40
A. infectoria Homo sapiens 49
A. infectoria Avena 48
A. triticina Triticum 02 (T)
99
A.
infectoria Hordeum 09
99
A. infectoria Homo sapiens 51
A. photistica Digitalis 05 (T)
A. malorum Soil 21
100
70
A. malorum - 23
A. malorum - 24
99
A. malorum var. polymorpha Vitis 26 (T)
A. malorum Triticum 28
87
A. malorum Gossypium 29
81 A. malorum Soil 25
A. malorum - 22
93
A. malorum Triticum 27
C. cetera Elymus 20 (T)
100 E. abundans Dianthus 30
E. abundans Fragaria 31 (T)
0.01
79
72
Fig. 6. Unrooted consensus dendrogram based on 51 strains (2 Embellisia abundans, 1 Chalastospora cetera, 39 Alternaria infectoria species–group, 8 A. malorum and 1 A.
malorum var. polymorpha). Maximum likelihood tree of 3 partial genes (ITS, gpd and tef-1a) constructed using RaxML (Cipres webserver). Bootstrap values > 70% are indicated.
Color-coding in the A. infectoria species-group clade corresponds to haplotype groups given in Table 2 and Fig. 8. Strains marked with correspond to cluster B1 in the
chemical analysis. Strain labels: strain ID/strain number/type culture.
located in different sub-clades depending on the molecular sequence examined, but with A. photistica (#05) and A. triticina
(#02) as outliers. With each individual gene, variability was largely
random, judging from low bootstrap values and from obtaining dif-
ferent groupings when different algorithms were used for tree
reconstruction. When genes were concatenated, A. viburni (#13)
clustered at 81% bootstrap support with three strains (#19, #38
and #41) identified as A. infectoria species–group sensu Simmons.
Author's personal copy
654
B. Andersen et al. / Fungal Genetics and Biology 46 (2009) 642–656
Fig. 7. Mutations at each position in the aligned ITS, tef-1a and gpd sequences of the 38 A. infectoria strains (except A. photistica #05) with the predicted exons and introns.
The single type cultures were not unambiguously separated from
the core of A. infectoria. The same pattern was seen in the chemical
data with A. photistica (#05) and A. triticina (#02) as outliers. No
distinct groups or clades were formed in the molecular analyses,
while there was a certain grouping in metabolite profiles and
metabolite families (Figs. 3–5 and Table 3). Some A. infectoria species–group strains were able to produce altertoxin derivatives,
while others produced metabolites of unknown structures, but
not altertoxins. No distinct morphological groups were seen either
among the strains in the A. infectoria species–group. Morphology
showed that basically each strain was a taxon in its own right.
Lastly, no groupings or correlations could be found between proascoma formation, ability to grow at 37 °C, host or geographic origin
and haplotypes, metabolite production or morphological identity.
Our original hypothesis was that taxa in Lewia/A. infectoria species–group were sexual fungi and that molecular sequence analyses and metabolite profiling would yield a number of groups
according to the genealogical concordance phylogenetic species
recognition (GCPSR) (Taylor et al., 2000). Our data, however, indicate that only three strains in the A. infectoria species–group show
evidence of recombination and that several isolates are able to produce proascomata in axenic culture. Since several taxa in the A.
infectoria species–group have been shown to produce ascomata
with viable ascospores in axenic cultures (Kwasna and Kosiak,
2003; Simmons, 2007; unpublished results), Lewia/A. infectoria
species–group must, at least in part, be homothallic and the purpose of ascoma formation in nature could be a survival strategy.
The high similarity in nucleotide sequence amongst the A. infectoria species–group strains (Fig. 6), suggests that most strains are
clonal and may have derived via mutations from one common
ancestor similar to the arbuscular mycorrhizal fungi (Rosendahl,
2008).
Several studies (reviewed in Taylor et al., 2000; O’Donnell et al.,
2004) show an increase in numbers of taxa, when going from morphological species recognition via biological recognition to GCPSR,
which corresponded with either geographic origin or hosts. In our
study we see the opposite: molecular cladification yields the lowest number of taxa in the A. infectoria species–group (A. photistica
and one phylogenetic taxon), while the chemical classification
gives more (A. photistica and A. triticina and two chemically different taxa) and with morphological appearance giving the highest
number (38 morphologically different taxa). Applying GCPSR to
the A. infectoria species–group would lead to synonymizing of all
morphological species in the A. infectoria species–group under
one name: A. infectoria Simmons. Alternatively, morphological species recognition could be applied and strains in the A. infectoria
species–group would represent new ‘‘emerging” species that require a name and a formal description. But according to Taylor
et al. (2000) and Rosendahl (2008), GCPSR can only be applied to
sexual/heterothallic fungi, not to homothallic/clonal strains, so neither of the two approaches (one species or 38 species) is workable.
In practice, however, there is a regular need for identification of
Alternaria isolates, because they have acquired different abilities in
nature, which affect us negatively. Some isolates have been
encountered as opportunistic human pathogens, others as plant
pathogens and others again are saprotrophic on cereals producing
biologically active metabolites. Artificial identification systems
based on any stable differentiation characters (e.g. PCR, AFLP,
metabolite profiles, sporulation patterns obtained under standardized conditions) still play an important role in taxonomy. Strains in
the A. infectoria species–group show characteristic phenotypical
traits, which can be detected, recognized, and used for identification. Depending on the users needs, identification of taxa in the
A. infectoria species–group can be done to different levels. In medical mycology, molecular identification using ITS is fast, wellknown and often the only method to obtain the correct diagnosis
for isolates in the A. infectoria species–group, since strains from human lesions rapidly loose their ability to sporulate in vitro. Strains
used in this study that originated from human skin lesions sporulated poorly, even under optimal conditions, and went sterile after
one or two transfers. However, they still maintained their ability to
produce all the characteristic metabolites in spite of their vegetative or yeast-like growth. Concerning alternarioses in humans or
animals, generally only identification to species–group level is
needed, since the same medical treatment (e.g. itraconazole) is
likely to be applicable regardless of taxon identification within
the A. infectoria species–group (Brasch et al., 2008; Dye et al.,
2009). In plant pathology, phytosanitary, and quarantine, on the
other hand, ITS sequencing is not enough to identify a known pathogen or discover a new pest that requires quarantine. With our current knowledge, described plant pathogens like A. triticina, A.
viburni, and A. intercepta can be distinguished from other taxa of
the A. infectoria species–group using morphology. In the cladistic
analyses, A. triticina (#02) grouped with different taxa in the A.
Table 4
Alignment data set for 38 strains in the Alternaria infectoria species–group, except A. photistica, of the three genes with number of mutations, parsimony informative mutations,
sites and p-value in phi-test.
gpd
tef-1a
ITS
a
b
c
Total number of sites
(gaps/missing)
Total number of
mutations
Number of parsimony
informative sites
Parsimony informative
mutations (%)a
Parsimony informative
sites (%)b
phi-Test
p-valuec
456 (40)
443 (17)
546 (71)
109
113
40
59
53
35
54.1
46.9
87.5
12.9
12.0
6.4
7.5104
0.15
0.41
Percentage is calculated using the total number of mutations.
Percentage is calculated using the total number of sites.
p-Value < 0.05 shows presence of recombination.
Author's personal copy
655
B. Andersen et al. / Fungal Genetics and Biology 46 (2009) 642–656
19
20
3
8
12
3
10
6
23
17
14
8
3
22
1
18
17
13
23
3
7
16
8
12
13
16
21
5
2
9
7
14
15
7
6
8
1
24
13
6
3
8
3
15
2
11
7
2
5
7
9
5
4
Fig. 8. Haplotype network of A. infectoria species–group strains (except A. photistica #05) based on maximum parsimony tree of gpd sequences. The circles represent the 24
different haplotype groups, which are given in Table 2. The lines between the groups connecting the haplotypes show the number of nucleotides differing. Circles with solid
colors (haplotypes 2, 6, 9, 15 and 22) contain more than one strain and circles with red lines (haplotypes 7, 14 and 18) show the position of the recombinant strains.
infectoria species–group depending on the chosen DNA sequence,
but was an outlier chemically, having a different metabolite profile.
Further research may yield A. triticina specific metabolites that can
be used to facilitate identification. In food safety, taxa in the A.
infectoria species–group regularly contaminate cereal grain
(Andersen et al., 1996; Pitt and Hocking, 1997; Kosiak et al.,
2004; Perelló et al., 2008). The most urgent need is to know what
secondary metabolites and other biologically active compounds
are produced in the cereals like wheat, barley, and maize. Since
current knowledge does not allow connections between metabolite profiles and morpho-species to be made, chemical analyses
are needed.
The results presented in this study show that these household
genes (ITS, tef-1a and gpd) do not reflect ecology, secondary metabolism or morphology of the A. infectoria species–group and that
molecular cladification and phylogeny cannot predict pathogenicity, host specificity or mycotoxin production. Concerning the classification and the systematic placement of the strains and morphospecies in the A. infectoria species–group, a polyphasic approach is
needed, but there are inconsistencies between the different taxonomic features and we therefore refrain from recommending any
taxonomic changes at this point in time.
Acknowledgments
The authors would like to thank EG Simmons for cultures and
for suggestions to the manuscript and Jens C. Frisvad and Ulf Thrane for fruitful discussions. This project was a collaboration between CMB, DTU, Denmark and CBS-KNAW, Fungal Biodiversity
Centre, The Netherlands and was supported in part by 1) a the
SYNTHESYS Project (NL-TAF-1843), http://www.synthesys.info/,
which is financed by European Community Research Infrastructure
Action under the FP6 ‘‘Structuring the European Research Area”
Programme”, 2) a grant from the Danish Directorate for Food, Fisheries and Agri Business (FFS05-3) and 3) the VILLUM KANN RASMUSSEN foundation.
References
Andersen, B., Hollensted, M., 2008. Metabolite production by different Ulocladium
species. Int. J. Food Microbiol. 126, 172–179.
Andersen, B., Thrane, U., 1996. Secondary metabolites produced by Alternaria
infectoria and their use as chemotaxonomic markers. Mycotoxin Res. 12,
54–60.
Andersen, B., Dongo, A., Pryor, B.M., 2008. Secondary metabolite profiling of
Alternaria dauci, A. porri, A. solani, and A. tomatophila. Mycol. Res. 112, 241–250.
Andersen, B., Hansen, M.E., Smedsgaard, J., 2005. Automated and unbiased image
analyses as tools in phenotypic classification of small-spored Alternaria spp.
Phytopathology 95, 1021–1029.
Andersen, B., Krøger, E., Roberts, R.G., 2002. Chemical and morphological
segregation of Alternaria arborescens, A. infectoria and A. tenuissima species–
groups. Mycol. Res. 106, 170–182.
Andersen, B., Nielsen, K.F., Thrane, U., Szara, T., Taylor, J.W., Jarvis, B.B., 2003.
Molecular and phenotypic descriptions of Stachybotrys chlorohalonata sp. nov.
and two chemotypes of Stachybotrys chartarum found in water-damaged
buildings. Mycologia 95, 1227–1238.
Andersen, B., Thrane, U., Svendsen, A., Rasmussen, I.A., 1996. Associated field
mycobiota on malt barley. Can. J. Bot. 74, 845–858.
Berbee, M.L., Pirseyedi, M., Hubbard, S., 1999. Cochliobolus phylogenetics and the
origin of known, highly virulent pathogens, inferred from ITS and glyceraldehyde3-phosphate dehydrogenase gene sequences. Mycologia 91, 964–977.
Brasch, J., Busch, J.-O., de Hoog, G.S., 2008. Cutaneous phaeohyphomycosis caused
by Alternaria infectoria. Acta Derm. Venereol. 88, 160–161.
Braun, U., Crous, P.W., Dugan, F., Groenewald, J.Z., de Hoog, G.S., 2003. Phylogeny
and taxonomy of Cladosporium-like hyphomycetes, including Davidiella gen.
nov., the teleomorph of Cladosporium s. str. Mycol. Prog. 2, 3–18.
Author's personal copy
656
B. Andersen et al. / Fungal Genetics and Biology 46 (2009) 642–656
Bruen, T., Philippe, H., Bryant, D., 2006. A simple and robust statistical test for
detecting the presence of recombination. Genetics 172, 2665–2681.
Butler, E.E., Mann, M.P., 1959. Use of cellophane tape for mounting and
photographing phytopathogenic fungi. Phytopathology 49, 231–232.
Carbone, I., Kohn, L.M., 1999. A method for designing primer sets for speciation
studies in filametous fungi. Mycologia 91, 553–556.
Christensen, K.B., van Klink, J.W., Weavers, R.T., Larsen, T.O., Andersen, B., Phipps,
R.K., 2005. Novel chemotaxonomic markers for the Alternaria infectoria species–
group. J. Agric. Food Chem. 53, 431–9435.
de Hoog, G.S., Gerrits van den Ende, A.H.G., 1998. Molecular diagnostics of clinical
strains of filamentous Basidiomycetes. Mycoses 41, 183–189.
de Hoog, G.S., Horré, R., 2002. Molecular taxonomy of the Alternaria and Ulocladium
species from humans and their identification in routine laboratory. Mycoses 45,
259–276.
Dubois, D., Pihet, M., Le Clec’h, C., Croué, A., Beguin, H., Bouchara, J.-P., Chabasse, D.,
2005. Cutaneous phaeohyphomycosis due to Alternaria infectoria.
Mycopathologia 160, 117–123.
Dugan, F.M., Peever, T.L., 2002. Morphological and cultural differentiation of
described species of Alternaria from Poaceae. Mycotaxon 83, 229–264.
Dye, C., Johnson, E.M., Gruffydd-Jones, T.J., 2009. Alternaria species infection in nine
cats. J. Feline Med. Surg. 11, 332–336.
Frisvad, J.C., 1983. A selective and indicative medium for groups of Penicillium
viridicatum producing different mycotoxins on cereals. J. Appl. Bacteriol. 54,
409–416.
Frisvad, J.C., Samson, R.A., 2004. Polyphasic taxonomy of Penicillium subgenus
Penicillium A guide to identification of food and air-borne terverticillate
Penicillia and their mycotoxins. Stud. Mycol. 49, 1–173.
Frisvad, J.C., Thrane, U., 1987. Standardized high-performance liquid
chromatography of 182 mycotoxins and other fungal metabolites based on
alkylphenone retention indices and UV–VIS spectra (diode array detection). J.
Chromatogr. 404, 195–214.
Hansen, M.E., 2003. Indexing and analysis of fungal phenotypes using morphology
and spectrometry. Ph.D. Thesis. IMM, DTU, Denmark. ISSN 0909-3192.
Haubold, B., Hudson, R.R., 2000. LIAN 3.0: detecting linkage disequilibrium in
multilocus data. Bioinformatics 16, 847–849.
Huson, D.H., Bryant, D., 2006. Application of phylogenetic networks in evolutionary
studies. Mol. Biol. Evol. 23, 254–267.
Kosiak, B., Torp, M., Skjerve, E., Andersen, B., 2004. Alternaria and Fusarium in
Norwegian grains of reduced quality – a matched pair sample study. Int. J. Food
Microbiol. 93, 51–62.
Kwasna, H., Kosiak, B., 2003. Lewia avenicola sp nov and its Alternaria anamorph
from oat grain, with a key to the species of Lewia. Mycol. Res. 107, 371–376.
Laatsch, H., 2008. AntiBase 2008. The Natural Compound Identifier. Wiley-VCH
GmbH & Co., Weinheim, Germany.
Nielsen, K.F., Smedsgaard, J., 2003. Fungal metabolite screening: database of 474
mycotoxins and fungal metabolites for dereplication by standardised liquid
chromatography-UV–mass spectrometry methodology. J. Chromatogr. A 1002,
111–136.
Nielsen, K.F., Gräfenhan, T., Zafari, D., Thrane, U., 2005. Trichothecene production by
Trichoderma brevicompactum. J. Agric. Food Chem. 53, 8190–8196.
O’Donnell, K., Ward, T.J., Geiser, D.M., Kistler, H.C., Aoki, T., 2004. Genealogical
concordance between the mating type locus and seven other nuclear genes
supports formal recognition of nine phylogenetically distinct species within the
Fusarium graminearum clade. Fungal Genet. Biol. 41, 600–623.
Pedras, M.S.C., Chumala, P.B., 2005. Phomapyrones from blackleg causing
phytopathogenic fungi: isolation, structure determination, biosyntheses and
biological activity. Photochemistry 66, 81–87.
Perelló, A.E., Sisterna, M.N., 2006. Leaf blight of wheat caused by Alternaria triticina
in Argentina. Plant Pathol. 55, 303.
Perelló, A.E., Moreno, M., Sisterna, M.N., 2008. Alternaria infectoria species–group
associated with black point of wheat in Argentina. Plant Pathol. 57, 379.
Pitt, J.I., Hocking, A.D., 1997. Fungi and Food Spoilage. Blackie Academic and
Professional, London, UK.
Prasada, R., Prabhu, A.S., 1962. Leaf blight of wheat caused by a new species of
Alternaria. Indian Phytopathology 15, 292–293.
Pryor, B.M., Bigelow, D.M., 2003. Molecular characterization of Embellisia and
Nimbya and their relationship to Alternaria, Ulocladium and Stemphylium.
Mycologia 95, 1141–1154.
Rosendahl, S., 2008. Communities, populations and individuals of arbuscular
mycorrhizal fungi. New Phytologist 178, 253–266.
Rozas, J., Rozas, R., 1995. DnaSP, DNA sequence polymorphism: an interactive
program for estimating Population Genetics parameters from DNA sequence
data. Comput. Applications Biosci. 11, 621–625.
Salemi, M., Gray, R.R., Goodenow, M.M., 2008. An exploratory algorithm to
identify intra-host recombinant viral sequences. Mol. Phylogenet. Evol. 49,
618–628.
Samson, R.A., Noonim, P., Meijer, M., Houbraken, J., Frisvad, J.C., Varga, J., 2007.
Diagnostic tools to identify black Aspergilli. Stud. Mycol. 59, 129–145.
Simmons, E.G., 1986. Alternaria themes and variations (22–26). Mycotaxon 25, 287–
308.
Simmons, E.G., 2007. Alternaria. An Identification Manual. CBS Fungal Biodiversity
Centre, Utrecht, The Netherlands.
Simmons, E.G., Roberts, R.G., 1993. Alternaria themes and variations (73).
Mycotaxon 48, 109–140.
Stamatakis, A., Hoover, P., Rougemont, J., 2008. A rapid bootstrap algorithm for the
RAxML web-servers. Syst. Biol. 75, 758–771.
Taylor, J.W., Jacobson, D.J., Kroken, S., Kasuga, T., Geiser, D.M., Hibbett, D.S., Fisher,
M.C., 2000. Phylogenetic species recognition and species concept in fungi.
Fungal Genet. Biol. 31, 21–32.
Wägele, J.W., Mayer, C., 2007. Visualizing differences in phylogenetic information
content of alignments and distinction of three classes of long-branch effects.
BMC Evol. Biol. 7, 147–170.
White, T.J., Bruns, T., Lee, S., Taylor, J., 1990. Amplification and direct sequencing of
fungal ribosomal RNA genes for phylogenetics. In: Innis, M.A., Gelfand, D.H.,
Sninsky, J.J., White, T.J. (Eds.), PCR Protocols: A Guide to Methods and
Applications. Academic Press, San Diego, pp. 315–322.