mycoses Diagnosis,Therapy and Prophylaxis of Fungal Diseases

Original article

Molecular identification and antifungal susceptibility of Curvularia

australiensis, C. hawaiiensis and C. spicifera isolated from human eye
infections

n,1 Eszter To
Krisztina Krizsa
th,1 La

szlo

G. Nagy,1 La

szlo

Galgo

czy,1 Palanisamy Manikandan,2,3
Muthusamy Chandrasekaran, Shine Kadaikunnan, Naiyf S. Alharbi,4 Csaba Va
4 4

gvo
€ lgyi1,4 and
Tama
s Papp 1

1
Faculty of Science and Informatics, Department of Microbiology, University of Szeged, Szeged, Hungary, 2Aravind Eye Hospital and Postgraduate Institute
of Ophthalmology, Coimbatore, India, 3Department of Medical Laboratory Sciences, College of Applied Medical Sciences, Majmaah University, Riyadh,
Saudi Arabia and 4Botany and Microbiology Department, King Saud University, Riyadh, Saudi Arabia

Summary A reliable identification method was developed for three closely related Curvularia
species, which are frequently isolated from human keratomycoses. Since the tradi-
tionally used morphological method and the increasingly used internal transcribed
spacer (ITS)-based molecular method proved to be insufficient to discern C. aus-
traliensis, C. hawaiiensis and C. spicifera, other molecular targets, such as b-tubulin,
translation elongation factor 1-a and the nuclear ribosomal intergenic spacer (IGS),
were tested. Among them, the use of the highly divergent IGS sequence is suggested
and the species-specific discriminating characters were determined in appropriate ref-
erence strains. It was also concluded that C. hawaiiensis and C. spicifera can be pre-
dominantly isolated from eye infections among the three species. The in vitro
antifungal susceptibility of 10 currently used antifungal agents against 32 Curvularia
isolates was also investigated. MICs were determined in each case. Isolates of C. spi-
cifera proved to be less susceptible to the tested antifungals than those of C. hawaiien-
sis, which underline the importance of the correct identification of these species.

Key words: Bipolaris, Cochliobolus, keratomycosis, nuclear ribosomal intergenic spacer, reference strain.

Curvularia based on molecular phylogenetic studies

Introduction
Three closely related species, Curvularia australiensis,
C. hawaiiensis and C. spicifera, are frequently isolated
from human phaeohyphomycoses, infections caused
by melanin-producing filamentous fungi. These three
species formerly were classified as members of the
genus Bipolaris. However, they were transferred to
Correspondence: Tam
as Papp, Faculty of Sciences and Informatics, Depart-
ment of Microbiology, University of Szeged, H-6726 Szeged, Ko € z

ep fasor
52., Hungary.
Tel.: +36 62 544516. Fax: +36 62 544823.
E-mail: pappt@bio.u-szeged.hu
Submitted for publication 24 February 2015
Revised 17 July 2015
Accepted for publication 17 July 2015
and to resolve the taxonomic and nomenclatural con-
flicts within the generic complex of the teleomorphic
Cochliobolus and the asexual Bipolaris and Curvular-
ia.1,2 These otherwise saprophytic species are the most
common agents of allergic fungal sinusitis and
together with other dematiaceous fungi are among the
third most important causes of fungal keratitis in
humans.3–6 Identification of C. australiensis, C. spicifera
and C. hawaiiensis at species level is rather difficult.1
Differentiation of these highly similar fungi is tradi-
tionally based on the conidial morphology. Until
today, this procedure has remained the primary
method for their identification.6–12 From this point of
view, the most important features are the septation,
the length and the width of the conidia.13 However,
the number of septa and conidial size may vary

© 2015 Blackwell Verlag GmbH

Mycoses, 2015, 58, 603–609 doi:10.1111/myc.12367
 an et al.
K. Krizs

depending on the isolate, age of the colony or culture
conditions, which makes species discrimination chal-
lenging with many misidentifications, especially in
cases of C. australiensis and C. spicifera. Because of the
lack of clear discriminating characteristics, clinical
microbiologists often report these fungi just as Bipolaris
or Curvularia sp. without further identification.12,14–16
Recently, sequencing of the internal transcribed spacer
(ITS) region of the nuclear ribosomal RNA coding
cluster and comparison of the sequence data with the
entries in GenBank have been used to confirm the spe-
cies-level diagnosis of Bipolaris/Curvularia isolates.16–19
Unfortunately, only a few data are available con-
cerning the antifungal susceptibility of these fungi20
and categorical breakpoints are not available for sus-
ceptibility testing. Bipolaris/Curvularia infections have
often been reported to be refractory to amphotericin B
treatment, whereas sinusitis and keratitis responded
well to itraconazole therapy in some studies.3,4
In the present study, Curvularia isolates obtained
from human keratomycoses were examined to improve
discrimination among C. australiensis, C. spicifera and
C. hawaiiensis. Therefore, their most important mor-
phological characters and several molecular markers,
such as the whole nuclear ribosomal transcribed
spacer (ITS) and two segments of the nuclear riboso-
mal intergenic spacer (IGS) region, as well as the pro-
tein coding genes, translation elongation factor 1-a
(tef) and b-tubulin (tub) were analysed. Antimicrobial
susceptibility of the involved fungal isolates against
the most important and currently used antifungal
agents was also evaluated.
Materials and methods
Strains
Fungal strains involved in the study are presented in
Table S1. The strains from human keratomycoses
were isolated in the years 2007 and 2008 from
immunocompetent patients 11–75 years old at the
Aravind Eye Hospital and Postgraduate Institute of
Ophthalmology, Coimbatore, Tamil Nadu, India.
Morphological study
Strains were grown on potato dextrose agar (PDA;
Difco, Becton Dickinson, Franklin Lakes, New Jersey,
USA) medium for 14 days at 28 °C. Size and number
of septa of 50 conidia per isolates were measured
using an ocular micrometre fitted on a light micro-
scope (Zeiss Axiolab, Carl Zeiss, Jena, Germany).
604
Statistical analysis was performed by Kruskal–Wallis
test in RSTUDIO (Version 0.97.336; Boston, MA.
Retrieved May 20, 2012. RStudio: Integrated develop-
ment environment for R. http://www.rstudio.org/).
DNA extraction and PCR
Strains were grown in potato dextrose broth (PDB;
Difco) for 5 days at 28 °C on a rotary shaker at
200 rpm. Genomic DNA was extracted from 10 mg of
mycelia ground to a fine powder in liquid nitrogen and
purified by using the MasterPure Yeast DNA purifica-
tion kit (Epicentre, Madison, Wisconsin, USA) accord-
ing to the instructions of the manufacturer. The
complete ITS (including ITS1 – 5.8S rRNA – ITS2) and
IGS regions and fragments of the tef and tub genes were
amplified by PCR, using the following primers: ITS1
and ITS4 for ITS,21 invSR1R and LR12R for IGS,22
983F and 2218R for tef23 and T12 and T22 for tub 24
(for primer sequences see Table S2). Each reaction mix-
ture contained 10 Taq polymerase buffer (Zenon,
Szeged, Hungary), 200 lmol l1 dNTPs, 1 lmol l1 of
each primer, 5 U Taq polymerase (Zenon) and 50 ng of
genomic DNA in a final volume of 25 ll (amplification
conditions are detailed in Table S3).
Sequence analysis
The ITS, IGS, tef and tub amplicons were sequenced at
LGC Genomics (Germany) with the primers used in the
PCR experiments. Individual sequence readings were
assembled to contigs by using the PREGAP and GAP4 pro-
grammes of the STADEN Package.25 In case of IGS,
instead of the whole region, two partial sequences were
determined, the 1920-bp partial 18S rDNA – 18S-28S
intergenic spacer and the 877-bp partial 28S rDNA –
18S-28S intergenic spacer named in this study as
IGS_1 and IGS_2, respectively. All sequences have been
deposited in NCBI GenBank (for the accession numbers
see Table S1). Sequences were aligned using the Proba-
bilistic Alignment Kit (PRANK version 091016).26 The
concatenated alignment of the ITS, tef, tub, IGS_1 and
IGS_2 was used in the phylogenetic analyses. For the
phylogenetic analysis, the best fit models were deter-
mined by using JMODELTEST 0.1.1.27,28 For the tub and
tef sequences the Hasegawa, Kishino and Yano (HKY)
model, for the ITS, IGS_1 and IGS_2 the generalised
time reversible (GTR + G + c model) were applied. Phy-
logenetic reconstruction was carried out by Bayesian
MCMC using MRBAYES v 3.1.2.29 Four Markov chains
were run for 3 9 107 generations, with a burn-in
value of 7.5 9 104. Trees were sampled every 100th
© 2015 Blackwell Verlag GmbH
Mycoses, 2015, 58, 603–609
 Molecular identification of Curvularia spp.

Figure 1 Fifty per cent majority rule consensus tree of 26 990 post-burn-in trees obtained from the Bayesian analysis of the
ITS + tef + tub + IGS_1 + IGS_2 concatenated alignment, which was generated with the MRBAYES v 3.1.2. The Bipolaris sorokiniana CBS
110.14 was used as out-group. On the phylogram, four well-defined clades can be discerned, which correspond to C. spicifera, C. hawai-
iensis, C. australiensis and C. geniculata and Curvularia sp. Neotype strains of Curvularia spicifera, C. hawaiiensis and C. australiensis are
indicated by *. IGS, intergenic spacer; ITS, internal transcribed spacer.

generations. Post-burn-in trees (altogether 26 990
trees) were used to compute a 50% Majority Rule con-
sensus phylogram. Posterior probabilities ≥0.95 were
considered as significant. In all phylogenetic analysis,
appropriate sequences of Bipolaris sorokiniana strain
SZMC 13012 were used as the outgroup (Table S1).
In vitro antifungal susceptibility tests
In vitro antifungal activity of amphotericin B (AMB),
clotrimazole (CLT), econazole (ECO), fluconazole (FLC),
itraconazole (ITC), ketoconazole (KTC), miconazole
(MCZ), voriconazole (VRC), natamycin (NAT) and ter-
binafine (TRB) against the isolates was tested using a
broth microdilution method according to the CLSI
guideline M38-A2.30 Minimal inhibitory concentration
(MIC) values were determined in 96-well microtiter
plates by both visual inspection and measuring the
optical density of the cultures at 620 nm after 48 and
72 h incubations at 28 °C in standard RPMI-1640
medium (Sigma, St. Louis, Missouri, USA). Inocula
were prepared from 14-day cultures grown on PDA;
the final conidial suspensions contained 1–
5 9 104 CFU ml1. Final concentrations of the drugs
ranged from 0.03 to 16 lg ml1, except those of NAT
and FLC, which ranged from 0.125 to 64 lg ml1.
Each plate contained a sterile control (containing med-
ium alone) and a growth control (containing inocu-
lated medium without the drugs). The uninoculated
medium was used as background for the spectrophoto-
metric calibration. Each experiment was repeated three
times; Paecilomyces variotii ATCC 22319 was used as
the quality control strain.

© 2015 Blackwell Verlag GmbH

Mycoses, 2015, 58, 603–609 605
 an et al.
K. Krizs

Table 1 Length and width of the conidia and number of the conidial septa measured in case of the studied Curvularia species.
C. australiensis (3)1
Length 21 (13.7–27.5)2
Width 7.3 (5–10)
Number of septa 3 (1–4)3
1
Number of the tested isolates per species are given in parentheses. Fifty conidia were measured in case of each tested isolate.
2
For length and width of the conidia, average values are presented in lm and ranges of the measured values (lm) are shown in
parentheses.
3
Average values are presented and ranges of the measured values are shown in parentheses.
Results
Twenty-one strains were isolated from human kerato-
mycoses in the Aravind Eye Hospital and Postgraduate
Institute of Ophthalmology (Coimbatore, Tamil Nadu,
India). The isolates from corneal scrapings were identi-
fied as Bipolaris spp. in the diagnostics laboratory of
the hospital and were deposited in the Szeged Microbi-
ological Collection (SZMC, Szeged, Hungary; http://
www2.sci.u-szeged.hu/microbiology/collection.htm).
Molecular identification
To identify the strains isolated from keratomycoses, the
complete ITS region was amplified and sequenced.
However, species level identification by BLAST searches
in the NCBI GenBank was not successful because the
determined sequences showed >99% identity with
sequences of strains representing C. australiensis,
C. hawaiiensis and C. spicifera and some other Bipolaris
and Curvularia species. Moreover, ITS sequences of
B. australiensis and B. spicifera deposited in GenBank
showed 99–100% identity with each other and 98–
100% identity with those of B. hawaiiensis.
For comparison, 12 other strains obtained from
international strain collections originating from
human infections, plants and soil were also involved
in the study and the ITS sequences of these strains
were determined (Table S1). A phylogenetic analysis
of these 33 sequences together with 60 further ITS
sequences downloaded from GenBank and representing
17 species of the genus Curvularia was then performed.
The analyses resulted in an unresolved tree, which is
abundant in polytomies (Fig. S1). This ITS phylogeny
reinforced the lack of sufficient informativeness in the
ITS region to resolve well-defined groups within Curvu-
laria and to discern these closely related species.
To obtain a more resolved phylogeny, besides the
ITS, two segments of the IGS region (IGS_1 and
IGS_2), and a fragment of the tef and tub genes were
606
C. hawaiiensis (7) C. spicifera (9) C. geniculata (3)
16.6 (10.6–17.5) 22.4 (15–33.8) 21.5 (20.9–22)
5.9 (5–7.5) 9.3 (6.2–12) 9.1 (8.8–9.3)
3 (1–6) 3 (1–4) 3 (1–3)
sequenced from the above mentioned 33 strains
(Table S1) and a multilocus analysis was performed.
On the resulting tree four well-defined clades can be
discerned (Fig. 1). Three of them correspond to C. spi-
cifera, C. hawaiiensis and C. australiensis, while the
fourth clade contain C. geniculata and other Curvularia
sp. isolates, which could not identified at species level
in the lack of a reference strain. Thus, eight C. spi-
cifera, seven C. hawaiiensis, three C. geniculata and
three other Curvularia sp. isolates could be identified
from the human keratomycoses samples. According to
this analysis, the B. australiensis strain CBS 705.71
was identified as C. spicifera. Thus, the C. australiensis
clade did not contain any isolates from keratomycoses.
Molecular markers for the proper identification of
C. autraliensis, C. hawaiiensis and C. spicifera
The ITS region, IGS_1 and IGS_2, as well as the tef
and tub sequences of the type strains of C. australiensis
(CBS 172.57), C. hawaiiensis (CBS 173.57) and C. spi-
cifera (CBS 274.52) were analysed to find species-
specific sequence motifs appropriate to discriminate
these three species. The potential differentiating motifs
were then tested for all of our 33 strains by aligning
the sequences separately (TreeBASE Submission ID
15951).
In the ITS and tub alignments, numerous one or
more nucleotide long differences were observed but
they were rather strain than species specific (Fig. S2).
Although tef sequences could not be used to discern
C. australiensis and C. spicifera, two nucleotide posi-
tions specific for C. hawaiiensis could be identified
(Fig. S3). In the two segments of the IGS region, sev-
eral distinctive species-specific sequence motifs were
identified. In the alignment of the IGS_1 sequences,
numerous 1–10 nucleotides long motifs were identi-
fied, which clearly differentiated C. australiensis,
C. hawaiiensis and C. spicifera as well as the other six
Curvularia isolates (Fig. S4). Among them, a
© 2015 Blackwell Verlag GmbH
Mycoses, 2015, 58, 603–609
 Molecular identification of Curvularia spp.

seven-nucleotide motif was identified (at the position Table 2 In vitro minimal inhibitory concentration (MIC) values
260–266 in Fig. S4), which was unique in each spe- (lg ml1) of selected antifungal agents against the Curvularia
isolates.
cies. Many species-specific single nucleotide differences
(19 for C. australiensis, six for C. hawaiiensis and two MIC (lg ml1)
Species (number
for C. spicifera) and several 2- to 43-nucleotide motifs
of isolates) Agents Range MIC MIC50 MIC90
were detected in the IGS_2 partial sequence (Fig. S5).
Among them, there is a 17-nucleotide motif at the C. australiensis (3) AMB 1 1 n.d.1
position 782–798 (Fig. S5), which contain unique NAT 4 4 n.d.
characters for each species. CLT 0.25–0.5 0.5 n.d.
ECO 0.25–≥16 0.5 n.d.
FLC 16–64 32 n.d.
Conidial morphology ITC 0.125–0.5 0.125 n.d.
KTC 0.5–1 0.5 n.d.
Length, width and the number of septa of 50 conidia MCZ 0.25–0.5 0.5 n.d.
were measured for each tested isolate (Table 1). Differ- VRC 0.5–1 0.5 n.d.
TRB 0.5 0.5 n.d.
ences in the length and width of the conidia of the C. hawaiiensis (12) AMB 0.125–1 0.5 1
C. australiensis, C. spicifera and C. hawaiiensis isolates NAT 2–4 1–2 2
proved to be not significant. Although conidia of the CLT 0.5–2 0.5–1 1
C. geniculata isolates also had similar dimensions, these ECO 0.25–≥16 2 16
isolates could be easily differentiated from those of the FLC 32–≥64 32 64
ITC 0.06–1 0.125 1
three other species based on their typical, curved, KTC 0.25–1 0.5–1 1
Curvularia-type conidia. MCZ 0.25–2 0.5–1 1
VRC 0.5–4 1 4
TRB 0.25–4 2 4
Antimicrobial susceptibility assay C. spicifera (12) AMB 0.125–2 0.5–1 1
NAT 1–4 2 4
Minimal inhibitory concentrations (MIC50 and MIC90 CLT 0.125–8 1 8
values) of the tested antifungal agents are shown in ECO 0.5–≥16 8–16 >16
Table 2. The MIC50 of ITC was found to be FLC 32–64 32–64 64
0.125 lg ml1 for C. autraliensis and C. hawaiiensis, ITC 0.25–4 0.5 2
0.25 lg ml1 for C. genicluata and 0.5 lg ml1 for KTC 0.125–8 2 4
MCZ 0.5–4 2 4
C. spicifera, while the MIC90 value was 1 lg ml1 for VRC 2–4 2 4
C. hawaiiensis and 2 lg ml1 for C. spicifera. FLC dis- TRB 0.25–8 2 4
played MIC50 values in the range of 8–64 lg ml1 and C. geniculata (3) AMB 0.5–1 1 n.d.
MIC90 at 64 lg ml1. The four groups of isolates NAT 2–4 2 n.d.
showed high variability in their susceptibility to ECO as CLT 0.25–2 0.5 n.d.
ECO 0.5–4 0.5 n.d.
the MIC50 values ranged from 0.25 to 16 lg ml1. FLC 8–64 8 n.d.
MIC50 and MIC90 values of the other tested azoles (CLT, ITC 0.125–2 0.25 n.d.
KTZ, MCZ and VRC) were found in the ranges of 0.25–4 KTC 0.25–2 0.5 n.d.
and 1–8 lg ml1, respectively. MIC50 values of AMB MCZ 0.125–2 0.5 n.d.
were found at 1 lg ml1 for C. australiensis, C. spicifera VRC 0.5–2 0.5 n.d.
TRB 0.5–2 1 n.d.
and C. geniculata and at 0.5 lg ml1 for C. hawaiiensis,
whereas the MIC90 value was determined to 1 lg ml1 AMB, amphotericin B; CLT, clotrimazole; ECO, econazole; FLC,
for C. hawaiiensis and C. spicifera. MIC50 of NAT for fluconazole; ITC, itraconazole; KTC, ketoconazole; MCZ, micona-
C. hawaiiensis, C. spicifera and C. geniculata was found at zole; VRC, voriconazole; NAT, natamycin; TRB, terbinafine.
1
2 lg ml1, while it was 4 lg ml1 for the C. australien- n.d., not defined. MIC50 and MIC90 were defined as the lowest
concentrations inhibiting the growth of 50 or 90% of the isolates
sis isolates. The MIC90 values were 2 lg ml1 for of a species, respectively. MIC90 could not be defined for species
C. hawaiiensis and 4 lg ml1 for C. spicifera. For TRB, represented by <10 isolates.
MIC values ranged from 0.25 to 8 lg ml1.

characters. As molecular identification revealed,

Discussion
C. hawaiiensis and C. spicifera were the predominant
Isolates examined in this study were formerly identi- species and no representatives of C. australiensis could
fied as Bipolaris species based on morphological be detected similarly to a previous study.20

© 2015 Blackwell Verlag GmbH

Mycoses, 2015, 58, 603–609 607
 mycoses Diagnosis,Therapy and Prophylaxis of Fungal Diseases

Original article

Molecular identification and antifungal susceptibility of Curvularia

australiensis, C. hawaiiensis and C. spicifera isolated from human eye
infections

n,1 Eszter To
Krisztina Krizsa
th,1 La

szlo

G. Nagy,1 La

szlo

Galgo

czy,1 Palanisamy Manikandan,2,3
Muthusamy Chandrasekaran, Shine Kadaikunnan, Naiyf S. Alharbi,4 Csaba Va
4 4

gvo
€ lgyi1,4 and
Tama
s Papp 1

1
Faculty of Science and Informatics, Department of Microbiology, University of Szeged, Szeged, Hungary, 2Aravind Eye Hospital and Postgraduate Institute
of Ophthalmology, Coimbatore, India, 3Department of Medical Laboratory Sciences, College of Applied Medical Sciences, Majmaah University, Riyadh,
Saudi Arabia and 4Botany and Microbiology Department, King Saud University, Riyadh, Saudi Arabia

Summary A reliable identification method was developed for three closely related Curvularia
species, which are frequently isolated from human keratomycoses. Since the tradi-
tionally used morphological method and the increasingly used internal transcribed
spacer (ITS)-based molecular method proved to be insufficient to discern C. aus-
traliensis, C. hawaiiensis and C. spicifera, other molecular targets, such as b-tubulin,
translation elongation factor 1-a and the nuclear ribosomal intergenic spacer (IGS),
were tested. Among them, the use of the highly divergent IGS sequence is suggested
and the species-specific discriminating characters were determined in appropriate ref-
erence strains. It was also concluded that C. hawaiiensis and C. spicifera can be pre-
dominantly isolated from eye infections among the three species. The in vitro
antifungal susceptibility of 10 currently used antifungal agents against 32 Curvularia
isolates was also investigated. MICs were determined in each case. Isolates of C. spi-
cifera proved to be less susceptible to the tested antifungals than those of C. hawaiien-
sis, which underline the importance of the correct identification of these species.

Key words: Bipolaris, Cochliobolus, keratomycosis, nuclear ribosomal intergenic spacer, reference strain.

Curvularia based on molecular phylogenetic studies

Introduction
Three closely related species, Curvularia australiensis,
C. hawaiiensis and C. spicifera, are frequently isolated
from human phaeohyphomycoses, infections caused
by melanin-producing filamentous fungi. These three
species formerly were classified as members of the
genus Bipolaris. However, they were transferred to
Correspondence: Tam
as Papp, Faculty of Sciences and Informatics, Depart-
ment of Microbiology, University of Szeged, H-6726 Szeged, Ko € z

ep fasor
52., Hungary.
Tel.: +36 62 544516. Fax: +36 62 544823.
E-mail: pappt@bio.u-szeged.hu
Submitted for publication 24 February 2015
Revised 17 July 2015
Accepted for publication 17 July 2015
and to resolve the taxonomic and nomenclatural con-
flicts within the generic complex of the teleomorphic
Cochliobolus and the asexual Bipolaris and Curvular-
ia.1,2 These otherwise saprophytic species are the most
common agents of allergic fungal sinusitis and
together with other dematiaceous fungi are among the
third most important causes of fungal keratitis in
humans.3–6 Identification of C. australiensis, C. spicifera
and C. hawaiiensis at species level is rather difficult.1
Differentiation of these highly similar fungi is tradi-
tionally based on the conidial morphology. Until
today, this procedure has remained the primary
method for their identification.6–12 From this point of
view, the most important features are the septation,
the length and the width of the conidia.13 However,
the number of septa and conidial size may vary

© 2015 Blackwell Verlag GmbH

Mycoses, 2015, 58, 603–609 doi:10.1111/myc.12367

Indian tertiary care hospital. Indian J Pathol Microbiol 2008; 51:
304–6.
13 McGinnis MR, Rinaldi MG, Winn RE. Emerging agents of phaeohy-
phomycosis: pathogenic species of Bipolaris and Exserohilum. J Clin
Microbiol 1986; 24: 250–9.
14 Filizzola MJ, Martinez F, Rauf SJ. Phaeohyphomycosis of the central
nervous system in immunocompetent hosts: report of a case and
review of the literature. Int J Infect Dis 2003; 7: 282–6.
15 Shin EJ, Guertler N, Kim E, Lalwani AK. Screening of middle ear effu-
sion for the common sinus pathogen Bipolaris. Eur Arch Otorhino-
laryngol 2003; 260: 78–80.
16 Toul P, Castillo L, Hofman V, Bouchara JP, Chanelet S, Gari-Tous-
saint M. A pseudo tumoral sinusitis caused by Bipolaris sp. J Infect
2006; 53: 235–7.
17 Buzina W, Braun H, Schimpl K, Stammberger H. Bipolaris spicifera
causes fungus balls of the sinuses and triggers polypoid chronic rhi-
nosinusitis in an immunocompetent patient. J Clin Microbiol 2003;
41: 4885–7.
18 Dyer ZA, Wright RS, Rong IH, Jacobs A. Back pain associated with
endobronchial mucus impaction due to Bipolaris australiensis colo-
nization representing atypical allergic bronchopulmonary mycosis.
Med Mycol 2008; 46: 589–94.
19 Bagyalakshmi R, Therese KL, Prasanna S, Madhavan HN. Newer
emerging pathogens of ocular non-sporulating molds (NSM) identi-
fied by polymerase chain reaction (PCR)-based DNA sequencing tech-
nique targeting internal transcribed spacer (ITS) region. Curr Eye Res
2008; 33: 139–47.
20 da Cunha KC, Sutton DA, Fothergill AW et al. Diversity of Bipolaris
species in clinical samples in the USA and their antifungal suscepti-
bility profiles. J Clin Microbiol 2012; 50: 4061–6.
21 White TJ, Bruns T, Lee S, Taylor JW. Amplification and direct
sequencing of fungal ribosomal RNA genes for phylogenetics. In: In-
nis MA, Gelfand DH, Sninsky JJ, White TJ, (eds), PCR Protocols: A
Guide to Methods and Applications. New York, NY, USA: Academic
Press, 1990: 315–22.
22 Vilgalys R, Hester M. Rapid genetic identification and mapping of
enzymatically amplified ribosomal DNA from several Cryptococcus
species. J Bacteriol 1990; 172: 4238–46.
23 Binder M, Hibbett D. (2003) Oligonucleotides. Available from:
http://www.clarku.edu/faculty/dhibbett/Protocols_Folder/Primers/
Primers.pdf [accessed on January 2014].
24 O’Donnell K, Cigelnik E. Two divergent intragenomic rDNA ITS2
types within a monophyletic lineage of the fungus Fusarium are
nonorthologous. Mol Phylogenet Evol 1997; 7: 103–16.
25 Staden R, Beal KF, Bonfield JK. The Staden package. Methods Mol
Biol 2000; 132: 115–30.
26 L€oytynoja A, Goldman N. An algorithm for progressive multiple
alignment of sequences with insertions. Proc Natl Acad Sci USA
2000; 102: 10557–62.
27 Guindon S, Gascuel O. A simple, fast and accurate method to esti-
mate large phylogenies by maximum-likelihood”. Syst Biol 2003; 52:
696–704.
28 Posada D. jModelTest: phylogenetic model averaging. Mol Biol Evol
2008; 25: 1253–6.
29 Huelsenbeck JP, Ronquist F. MRBAYES: Bayesian inference of phy-
logeny. Bioinformatics 2001; 17: 754–5.
30 Clinical and Laboratory Standards Institute. Reference method for
broth dilution antifungal susceptibility testing of filamentous fungi;
© 2015 Blackwell Verlag GmbH
Mycoses, 2015, 58, 603–609
Molecular identification of Curvularia spp.
Approved standard. 2nd edn. Place: CLSI document M38-A2. Clinical
and Laboratory Standards Institute 2008.
31 Kobayashi H, Sano A, Aragane N et al. Disseminated infection by
Bipolaris spicifera in an immunocompetent subject. Med Mycol 2008;
46: 361–5.
32 Viola GM, Sutton R. Allergic fungal sinusitis complicated by fungal
brain mass. Int J Infect Dis 2010; 14: 299–301.
33 Shukla PK, Kumar M, Keshava GBS. Mycotic keratitis: an overview
of diagnosis and therapy. Mycoses 2008; 51: 183–99.
34 Kumar M, Mishra NK, Shukla PK. Sensitive and rapid polymerase
chain reaction based diagnosis of mycotic keratitis through single
stranded conformation polymorphism. Am J Ophthalmol 2005; 140:
851–7.
35 McGinnis MR, Pasarell L. In vitro testing of susceptibilities of filamen-
tous Ascomycetes to voriconazole, itraconazole, and amphotericin B,
with consideration of phylogenetic implications. J Clin Microbiol
1998; 36: 2353–5.
36 Espinel-Ingroff A. In vitro activity of the new triazole voriconazole
(UK-109,496) against opportunistic filamentous and dimorphic fungi
and common and emerging yeast pathogens. J Clin Microbiol 1998;
36: 198–202.
37 Espinel-Ingroff A, Chaturvedi V, Fothergill A, Rinaldi MG. Optimal
testing conditions for determining MICs and minimum fungicidal
concentrations of new and established antifungal agents for uncom-
mon molds: NCCLS collaborative study. J Clin Microbiol 2002; 40:
3776–81.
Supporting information
Additional supporting information may be found in
the online version of this article.
Table S1. Curvularia isolates involved in the study:
taxon name, collection number, origin and GenBank
accession numbers of ITS, IGS_1, IGS_2, tef and tub
sequences.
Table S2. Sequences of the primers used in the
study.
Table S3. PCR conditions used to amplify the ITS
and IGS regions and the tef and tub gene fragments.
Figure S1. Fifty per cent majority rule consensus
tree of 15002 post-burn-in trees obtained from the
Bayesian analysis of 93 ITS sequences.
Figure S2. Alignment of ITS region sequences of
Curvularia species.
Figure S3. Alignment of tef sequences of Curvularia
species.
Figure S4. Alignment of IGS_1 sequences of Curvu-
laria species.
Figure S5. Alignment of IGS_2 sequences of Curvu-
laria species.
609