mycoses Diagnosis,Therapy and Prophylaxis of Fungal Diseases

Original article

Molecular epidemiology and in vitro antifungal susceptibility testing

of 108 clinical Cryptococcus neoformans sensu lato and Cryptococcus
gattii sensu lato isolates from Denmark

Ferry Hagen,1 Rasmus Hare Jensen,2 Jacques F. Meis1,3 and Maiken Cavling Arendrup2
1
Department of Medical Microbiology and Infectious Diseases, Canisius-Wilhelmina Hospital, Nijmegen, The Netherlands, 2Department of Microbiological
Surveillance and Research, Unit of Mycology, Statens Serum Institut, Copenhagen, Denmark and 3Department of Medical Microbiology, Radboudumc,
Nijmegen, The Netherlands

Summary Cryptococcosis is mainly caused by members of the Cryptococcus gattii/Cryptococcus

neoformans species complexes. Here, we report the molecular characterisation and
in vitro antifungal susceptibility of Danish clinical cryptococcal isolates. Species, geno-
type, serotype and mating type were determined by amplified fragment length poly-
morphism (AFLP) fingerprinting and qPCR. EUCAST E.Def 7.2 MICs were determined
for amphotericin B, flucytosine, fluconazole, voriconazole and isavuconazole. Most iso-
lates were C. neoformans (serotype A; n = 66) and belonged to genotype AFLP1/VNI
(n = 61) or AFLP1B/VNII (n = 5) followed by Cryptococcus deneoformans (serotype D;
genotype AFLP2, n = 20), C. neoformans 9 C. deneoformans hybrids (serotype AD;
genotype AFLP3, n = 13) and Cryptococcus curvatus (n = 2). Six isolates were C. gattii
sensu lato, and one isolate was a C. deneoformans 9 C. gattii hybrid (genotype AFLP8).
All isolates were amphotericin B susceptible. Flucytosine susceptibility was uniform
MIC50 of 48 mg l 1 except for C. curvatus (MICs >32 mg l 1). Cryptococcus gattii
sensu lato isolates were somewhat less susceptible to the azoles. MICs of fluconazole
(>32 mg l 1), voriconazole (0.5 mg l 1) and isavuconazole (0.06 and 0.25 mg l 1
respectively) were elevated compared to the wild-type population for 1/19 C. deneofor-
mans and 1/2 C. curvatus isolates. Flucytosine MIC was elevated for 1/61 C. neofor-
mans (>32 mg l 1). Antifungal susceptibility revealed species-specific differential
susceptibility, but suggested acquired resistance was an infrequent phenomenon.

Key words: Cryptococcus neoformans, Cryptococcus gattii, Cryptococcus deneoformans, Cryptococcus bacillisporus,
Cryptococcus deuterogattii, antifungal susceptibility testing, amplified fragment length polymorphism fingerprinting,
multilocus sequence typing.

thoroughly taxonomically revised.1 The majority of

Introduction
Cryptococcosis is the encompassing description of
disease caused by species within the basidiomycetous
yeast genus Cryptococcus that recently has been
Correspondence: Dr Maiken Cavling Arendrup, Department of Microbio-
logical Surveillance and Research, Unit of Mycology, Statens Serum Institut,
Artillerivej 5, 43/317, DK-2300 Copenhagen S, Denmark.
Tel.: +45 3268 3223. Fax: +45 3268 3868.
E-mail: maca@ssi.dk
Submitted for publication 11 February 2016
Revised 29 February 2016
Accepted for publication 13 March 2016
cryptococcosis cases are caused by members of the
Cryptococcus gattii and Cryptococcus neoformans species
complexes, while infections with other species have
rarely been reported till date.24 Cryptococcosis is
mainly diagnosed among individuals who have an
impaired immune system, and it has been estimated
that annually nearly a million HIV-infected patients
develop cryptococcal meningitis with 625 000 deaths.5
Cryptococcus cells can remain dormant for decades after
they have been inhaled as basidiospores or desiccated
yeast cells, and can be activated when the host immune
status is attenuated.68

2016 Blackwell Verlag GmbH doi:10.1111/myc.12507

F. Hagen et al.
Molecular techniques, such as PCR fingerprinting by
M13 and (GACA)4 primers, restriction fragment length
polymorphism (RFLP) analysis of the PLB1 and URA5
gene, amplified fragment length polymorphism (AFLP)
fingerprinting, microsatellite typing and multilocus
sequence typing (MLST), have been regarded as help-
ful tools in unravelling the genotypic diversity.912
The taxonomy of the C. gattii and C. neoformans spe-
cies complexes has been revised due to the differences
between both species in terms of genetics, physiology,
pathogenicity and ecology (Fig. 1).6,9,10,12 The two
varieties of C. neoformans were raised to the species
level as C. neoformans (formerly var. grubii) represent-
ing serotype A isolates with genotypes AFLP1/VNI,
AFLP1A/VNB/VNII and AFLP1B/VNII, and Cryptococ-
cus deneoformans (formerly var. neoformans) represent-
ing serotype D and genotype AFLP2/VNIV isolates
(Fig. 1).10 Cryptococcus gattii was separated into five
species, Cryptococcus gattii sensu stricto (serotype B,
genotype AFLP4/VGI), Cryptococcus bacillisporus (sero-
type C, genotype AFLP5/VGIII), Cryptococcus deutero-
gattii (serotype B, genotype AFLP6/VGII), Cryptococcus
tetragattii (serotype C, genotype AFLP7/VGIV) and
Cryptococcus decagattii (serotype B, genotype AFLP10)
(Fig. 1).10 Hybrids of C. neoformans 9 C. deneoformans
(serotype AD, genotype AFLP3/VNIII), and although
Figure 1 Cryptococcal species identification, sero- and genotypes according to former and the current Cryptococcus taxonomy.1,10
2 2016 Blackwell Verlag GmbH
extremely rare C. gattii s.s. with either C. deneoformans
(serotype BD, genotype AFLP8) or C. neoformans (sero-
type AB, genotype AFLP9) were firstly described nearly
a decade ago (Fig. 1).13,14
The epidemiology of Cryptococcus infections in Eur-
ope has been studied, as is the case for many other
geographical regions.15,16 However, the majority of
recent European epidemiological studies that made
use of molecular typing techniques came from Wes-
tern and Mediterranean Europe.15 For the Scandina-
vian countries of Denmark, Finland, Iceland, Norway
and Sweden, recent epidemiological data are
sparse.1719 To update the current knowledge about
cryptococcal infections in Scandinavia, in particular
for Denmark, a set of Cryptococcus isolates was molec-
ularly investigated by mating and serotyping, AFLP
and MLST genotyping, and by antifungal susceptibility
testing according to the EUCAST E.Def 7.2 reference
method.
Materials and methods
Isolates and media
The mycological culture collection of the Staten Serum
Institute contained 114 cryptococcal isolates, which

were preserved during the period 19732013 and
stored in glycerol at 80 C. One hundred and eight
isolates were viable and included in the current study.
All isolates were previously identified by conventional
mycology procedures, including API ID32C (Biomer-
ieux, Marcy lEtoile, France), as being members of the
C. gattii species complex (n = 7; 6.5%), C. neoformans
species complex (n = 99; 91.7%) and Cryptococcus
curvatus (n = 2; 1.9%). Isolates were cultured onto
Sabouraud dextrose agar (Oxoid, Basingstoke, UK) for
48 h at 30 C.
Antifungal susceptibility testing
The determination of broth dilution minimum inhibi-
tory concentrations of antifungal agents was carried
out according to EUCAST guidelines (E.Def 7.2) for
amphotericin B, flucytosine, fluconazole, voriconazole
and isavuconazole.2 Manufacturers and stock solutions
(5000 mg l 1) in water (flucytosine) or DMSO (the
remaining compounds, dimethyl sulfoxide (DMSO,
D8779, Sigma-Aldrich, Vallensbk Strand, Denmark)
were the following: amphotericin B, flucytosine and
fluconazole (Sigma-Aldrich), isavuconazole (Astellas
Pharma Europe, Middlesex, UK) and voriconazole (Pfi-
zer, Ballerup, Denmark). The drug concentration range
studied was 0.034 mg l 1 for amphotericin B, isavu-
conazole and voriconazole, and 0.2532 mg l 1 for
5-flucytosine and fluconazole. Plates were incubated at
30 C to promote growth as suggested in EUCAST
E.Def 7.2 and read when the growth was 0.200 OD
above the background level. Candida parapsilosis ATCC
22019 and Candida krusei ATCC 6258 were included
as quality controls.
Genomic DNA extraction
A loop full of cryptococcal cells from 48 h old cul-
tures were suspended in 400 ll Bacterial Lysis Buffer
(Roche Diagnostics, Almere, The Netherlands) fol-
lowed by bead beating the cell suspension by using
green cap tubes prefilled with 1.4 mm zirconium
beads in the MagNA Lyser (both Roche Diagnostics).
The lysed cell suspension was subsequently inacti-
vated for 15 min at 95 C. Genomic DNA extraction
was automatically carried out with the Pathogen
Detection 200 protocol with a MagNA Pure 96 robot
(Roche Diagnostics) for which an input of 200 ll
lysed cell suspension was used, purified DNA was
finally eluted in 100 ll elution buffer (Roche Diagnos-
tics). The DNA samples were stored at 20 C until
further use.
2016 Blackwell Verlag GmbH
Cryptococcosis in Denmark during 19732013
Mating-type and serotype determination
The mating-type and serotype of C. neoformans was
determined using a recently established real-time
PCR method that targets the mating-type a and a
allele of the RUM1 gene for either the serotype A or
D background.20 Genomic DNA of the reference
strains CBS10512 (=125.91; aA; genotype AFLP1/
VNI), CBS10511 (=JEC20; aD; genotype AFLP2/VNIV),
CBS8710 (=H99 = CBS10515; aA; genotype AFLP1/
VNI) and CBS10513 (=JEC21; aD; genotype AFLP2/
VNIV) were included as controls.
A third recently described real-time PCR assay that
targets the multicopy intergenic spacer region (IGS1)
was applied to detect the presence of C. gattii sensu lato
among the samples.21 The reference strain CBS6956
(aB; genotype AFLP6/VGII) was included as a control.
The mating-type of C. gattii sensu lato isolates were
determined, as described before, by a conventional
PCR method that targets either the mating-type a or a
allele of the STE12 gene.23 As a control, the reference
isolates CBS6955 (aB; genotype AFLP5/VGIII) and
CBS6956 (aB; genotype AFLP6/VGII) were included.
Agarose gel electrophoresis was carried out with 5 ll
PCR product to check the presence of a fragment.
Amplified fragment length polymorphism
fingerprinting
All isolates were subjected to amplified fragment
length polymorphism (AFLP) genotyping following the
method as described before.22,23 Fragments were anal-
ysed on an ABI3500xL Genetic Analyzer (Applied
Biosystems, Foster City, CA, USA); for this purpose,
the amplicons were 209 diluted with ddH2O, and 1 ll
of this dilution was added to 0.1 ll LIZ600 internal
size marker (Promega, Leiden, The Netherlands) and
8.9 ll ddH2O. Raw data was further processed by
using Bionumerics v6.6 (Applied Maths, Sint-Martens-
Latem, Belgium) and a dendrogram was generated
based on the UPGMA algorithm.
Multilocus sequence typing of C. gattii sensu lato
isolates
To link C. gattii sensu lato isolates from the current
study with those from a large collection of genotyped
environmental, veterinary and clinical isolates, multi-
locus sequence typing (MLST) was performed.11 The
nuclear loci CAP59, GPD1, IGS1, LAC1, PLB1, SOD1
and URA5 were amplified and sequenced as described
before.6,11 Obtained sequences were concatenated and
3
F. Hagen et al.
compared to available sequences.6,2430 Phylogenetic
analysis on the concatenated dataset was carried out
using the software package MEGA v6.6.31 A maximum
likelihood analysis was performed on all aforemen-
tioned available MLST data, based on that a subtree
was generated that included isolates genetically closely
related to the Danish C. gattii sensu lato. Subsequently,
a 10009 bootstrap maximum likelihood analysis was
initiated with the best fitting nucleotide substitution
model for the Cryptococcus MLST dataset that was pre-
viously observed to be HKY with models incorporating
invariant sites (+I) along with discrete gamma rate
categories (+G).6,31 Obtained sequences were deposited
in Genbank under accession numbers KM230242
KM230318.
Results
Mating-, sero- and genotyping of C. gattii/
C. neoformans species complex isolates
The majority of the isolates were C. neoformans repre-
sented by the genotypes AFLP1/VNI (n = 61; 56.5%)
and AFLP1B (n = 5; 4.6%) (Fig. 1). C. deneoformans
genotype AFLP2/VNIV (n = 20; 18.5%) was the second
most common followed by the C. neoformans 9 C. dene-
oformans serotype AD hybrid genotype AFLP3/VNIII
(n = 13; 12.0%) (Fig. 1). Cryptococcus gattii sensu lato
was found to be represented by seven isolates dis-
tributed over the three species C. gattii sensu stricto
AFLP4/VGI (n = 2; 1.9%), C. bacillisporus AFLP5/VGIII
(n = 3; 2.8%) and C. deuterogattii AFLP6/VGII (n = 1;
0.9%) (Fig. 1). One isolate (0.9%) was an interspecies
hybrid between C. gattii sensu stricto and C. deneofor-
mans, known as genotype AFLP8 with serotype BD.
All 66 C. neoformans isolates were found to be aA. Of
the 20 C. deneoformans isolates, eight were found to be
aD and 12 aD. Of the 13 C. neoformans 9 C. deneofor-
mans hybrids, five were found to be aD-aA, one was aA-
aD. For seven isolates, only the serotype D background
could be detected with one isolate being aD and the
remaining six being aD. All the six C. gattii sensu lato
isolates were found to be mating-type a, and the mating-
types of the interspecies hybrid C. gattii sensu
stricto 9 C. deneoformans isolate could not be determined.
Identification of Cryptococcus species other than
C. gattii/C. neoformans species complex
The two isolates previously identified as C. curvatus by
conventional phenotypic identification were confirmed
by sequencing of the ITS region.32 Comparing the
4 2016 Blackwell Verlag GmbH
obtained ITS sequences with that of the recently estab-
lished ISHAM-ITS database showed that both Danish
isolates were 100% identical to that of the reference
sequence for C. curvatus. The taxonomy of the poly-
phyletic genus Cryptococcus has recently been thor-
oughly revised; as a consequence, Cryptococcus
curvatus was transferred to the novel genus Cutaneotri-
chosporon and retained is epithet curvatus.1
Multilocus sequence typing of C. gattii sensu lato
isolates
The six isolates that were identified by AFLP genotyp-
ing as being C. gattii sensu lato were subjected to
MLST to determine their genetic relatedness to a glo-
bal set of isolates (Fig. 2). The two C. gattii sensu
stricto isolates DK0821 and DK0842 were found to
be genetically indistinguishable from isolates
ATCC64062, CBS1622, CBS5757 and CBS6992
(Fig. 2). Two of the three C. bacillisporus isolates were
found to be genetically closely related (DK0841 and
DK0894) to each other as well as to a cluster of
CN043, CBS7819, CBS6993, CBS5758, 7597057,
384C, IHEM10079, B9422 and B8260 (Fig. 2). Cryp-
tococcus bacillisporus isolate DK0893 was found to be
similar to CBS6996 and ATCC24065, an environmen-
tal and clinical isolate, respectively, from the USA
(Fig. 2). The single C. deuterogattii isolate DK0022
clustered together with isolates that had a South
American origin (Fig. 2).
Antifungal susceptibility testing according the EUCAST
method
All isolates were susceptible to amphotericin B defined
as MICs being 1 mg l 1 with only discrete differ-
ences among the species/varieties (Table 1, Fig. 3).
Thus, C. neoformans was slightly less susceptible than
C. deneoformans (MIC50 0.125 vs. 0.06 mg l 1, and
GM 0.1490.018 vs. 0.096 mg l 1, respectively,
Table 1). Similarly, the in vitro activity of flucytosine
was uniform with an MIC50 of 48 mg l 1 for all spe-
cies with the exception of the two C. curvatus isolates
that were both resistant (MICs of >32 mg l 1).
However, flucytosine MIC was clearly elevated for
1/61 C. neoformans AFLP1/VNI isolate (>32 mg l 1,
Fig. 3) and a tail of C. deneoformans and C. neofor-
mans 9 C. deneoformans hybrid isolates with MICs in
the 3264 mg l 1 range were observed.
The C. gattii sensu lato isolates were less susceptible
to the three azoles and to fluconazole in particular com-
pared to the other species (Fig. 3). Thus, fluconazole,
 Cryptococcosis in Denmark during 19732013

Figure 2 Maximum likelihood 10009

bootstrapped phylogenetic analysis based
on the ISHAM consensus MLST scheme11
of Danish C. gattii sensu lato isolates
(in bold) compared to a set of globally
collected isolates.6,2729 Numbers next to
branches represent bootstrap values, only
those >80 are show.

voriconazole and isavuconazole MIC50s were 16, 0.25 (Table 1, Fig. 2). Finally, fluconazole (>32 mg l 1),
and 0.06 mg l 1 for C. gattii sensu lato in contrast to voriconazole (0.5 mg l 1) and isavuconazole (0.06
24, 0.06 and <0.030.06 mg l 1 for the others and and 0.25 mg l 1 respectively) were elevated for 1/19
geometric mean MICs followed the same pattern C. deneoformans and 1/2 C. curvatus isolates (Fig. 2).

2016 Blackwell Verlag GmbH 5

F. Hagen et al.

Discussion

2 to >32

1 to >32

1 to >32
Range

116
>32
Amplified fragment length polymorphism fingerprint-
Table 1 Amphotericin B, voriconazole, isavuconazole, fluconazole and flucytosine MICs against 108 Danish Cryptococcus isolates. Susceptibility testing was done following the

4
ing showed that the majority of the Danish cryptococ-

8.80
4.00

6.00
12.62

14.25
GM cal isolates (n = 61; 56.5%) represented C. neoformans,
Flucytosine

56 isolates (51.9%) belonged to the major genotype

AFLP1/VNI and a small proportion (n = 5; 4.6%)
MIC50

belonged to the genotype AFLP1B/VNII. The large pro-

4

8
4
4
4
portion of C. neoformans isolates mirrors the observa-
0.5 to >32

0.5 to >32

2 to >32
tions from other European epidemiological studies.

832
Range

0.58

48
Isolates collected during the period 19772007 in the

4
Netherlands were found to be represented by 79.3%
(n = 238) of genotype AFLP1/VNI and 2.3% (n = 7)
2.96

8.96
6.40
4.93
21.33
GM

of genotype AFLP1B/VNII.33 A retrospective study

Fluconazole

from Germany identified 78 C. neoformans isolates

MIC50

(77.2%) during the period 20042010.34 A prospec-

4

4
8
2
16

tive multicentre study from France reported that the

<0.030.06

<0.030.25
<0.030.06
<0.030.06

0.060.25

majority of the isolates during the period 19972001

<0.032

was C. neoformans (n = 161; 72.5%); of note, only

Range

<0.03

C. neoformans and C. deneoformans infections were

identified in this study.35 An earlier report from
0.039

0.065
0.042
0.041
0.400
0.155
Isavuconazole

France conducted from 1985 to 1993 included 1057

GM

cases of cryptococcosis, but only 413 isolates were

MIC50

<0.03

0.06
<0.03
<0.03
0.06

available for serotyping and the majority of them were

C. neoformans (n = 326; 78.9%).36 However, in
Mediterranean Europe, the number of C. neoformans
<0.030.25

<0.030.5

<0.030.5
0.060.5
0.1251

isolates in epidemiological surveys is often lower. For

Range

example, in Serbia, 58.8% (n = 20) of the studied iso-

0.06

0.06

lates were genotype AFLP1/VNI and 2.9% (n = 1)

genotype AFLP1B/VNII,20, while a report from a single
0.102

0.104
0.060
0.081
0.281
0.563
Voriconazole

GM

institute in Spain reported 44.8% (n = 26) C. neofor-

mans infections.37 The Spanish study reported 19%
MIC50

(n = 11) and 36.2% (n = 21) of the isolates being

0.06

0.06
0.06
0.06
0.25

C. deneoformans (genotype AFLP2/VNIV) and hybrids

between C. neoformans 9 C. deneoformans (genotype
EUCAST E.Def 7.2 method using incubation at 30 C for 48 h.

<0.030.25

0.060.25
<0.030.25
0.060.25
0.060.25
<0.031

AFLP3/VNIII) respectively. In Serbia, 10 isolates

0.125
Range

(29.4%) were found to be C. deneoformans and three

(8.8%) were hybrid serotype AD isolates.20 The per-
Amphotericin B

0.112

0.180
0.149
0.096
0.123
0.155

centage of C. deneoformans isolates in the current Dan-

GM

ish epidemiological study equals that of the Spanish

study (n = 20; 18.5% vs. n = 11; 19%).37 The neigh-
MIC50

0.125
0.125
0.06

0.06
0.06

bouring countries Germany and the Netherlands

reported a lower incidence of C. deneoformans infec-
C. deneoformans, AFLP2/VNIV (20)

tions (n = 15; 14.9% vs. n = 36; 12.0% respectively)

Cryptococcus gattii sensu lato (6)
C. neoformans, AFLP1B/VNII (5)
C. neoformans, AFLP1/VNI (61)

when compared to Denmark.33,34 The number of clini-

neoformans 9 Cryptococcus

cal C. neoformans 9 C. deneoformans hybrids in Den-

Cryptococcus curvatus (2)
Cryptococcus interspecies
deneoformans hybrid,

mark is low (n = 13; 12.0%), but, again, a lower

(number of isolates)

incidence was reported from Germany (n = 8; 7.9%)

Species, AFLP type

hybrid, AFLP8 (1)

AFLP3/VNIII (13)

and the Netherlands (n = 14; 4.7%).33,34

Cryptococcus

Seven Danish isolates were found to be C. gattii

sensu lato by phenotypic characterisation, and subse-
quent AFLP genotyping confirmed that six belonged

6 2016 Blackwell Verlag GmbH


Figure 3 MIC data for the Danish
Cryptococcus isolates by species. Species
isolates with intrinsic reduced susceptibil-
ity compared to that for C. neoformans
are highlighted with arrows. Isolates with
potential acquired resistance are
indicated by circles.
indeed to this species complex, while one isolate was
found to be a rarely reported interspecies hybrid
between C. gattii sensu stricto (genotype AFLP4/VGI)
and C. deneoformans (genotype AFLP2). So far, world-
wide, only a few C. gattii sensu stricto 9 C. neoformans
interspecies hybrids with genotype AFLP8 have been
described and in-depth genetic analyses showed that
these interspecies hybrids, next to those that exhibit
genotype AFLP9, had mixed phenotypic characteristics
from both parental species and that the hybrid gen-
ome had loss part of its genetic material.10,13,14
2016 Blackwell Verlag GmbH
Cryptococcosis in Denmark during 19732013
The six C. gattii sensu lato isolates were further
characterised by MLST analysis, and the two C. gattii
sensu stricto isolates DK0821 and DK0842 (genotype
AFLP4/VGI) were genetically indistinguishable from
two North American isolates, as well as from the type-
strain CBS1622 which had been isolated in France
during the early 20th century (Fig. 2).6,10 Interest-
ingly, the closest relatives of this cluster originated
from the South African environment, suggesting that
the closely related clinical isolates from Denmark,
France and the USA somehow had a link with the
7
F. Hagen et al.
African continent (Fig. 2). The three isolates DK0841,
DK0893 and DK0894 were molecularly characterised
as being C. bacillisporus (genotype AFLP5/VGIII), but
could be distinguished from each other by MLST
(Fig. 2). Nevertheless, all Danish C. bacillisporus iso-
lates were strongly related to clinical, environmental
and veterinary isolates from the west coast of the
USA, which suggests that these patients acquired the
infection while travelling outside Denmark (Fig. 2).
Isolate DK0022 appeared to be C. deuterogattii (geno-
type AFLP6/VGII) and genetically closely related to
clinical and environmental isolates from mainly South
America (Fig. 2). As for the C. bacillisporus isolates,
this implies that the patient probably acquired the
C. deuterogattii infection outside Denmark. This has
been reported before in a Danish traveller who shortly
visited Vancouver Island, British Columbia, Canada,
and developed a cryptococcal infection weeks after
returning to Denmark. The infecting isolate CBS10485
was genetically indistinguishable from the so-called
Vancouver Island outbreak isolates (Fig. 2).27 Similar
imported cases of C. deuterogattii infections due to tra-
vel to the endemic area in Canada have been reported
before.3840
Antifungal susceptibility testing data for Cryptococ-
cus by the EUCAST methodology is scarce. Zaragoza
and colleagues tested 10 C. neoformans s.l. and five
C. gattii s.l. following EUCAST methodology and found
amphotericin B MIC ranges of 0.030.125 mg l 1,
flucytosine MICs 216 mg l 1, fluconazole MICs
164 mg l 1 and voriconazole 0.0150.25 mg l 1
and as presented here with higher fluconazole MICs
for C. gattii s.l.41 Epidemiological cut-off values and
clinical breakpoints have not yet been established by
neither EUCAST nor CLSI, and hence categorisation
into susceptible, intermediate and resistant is not pos-
sible. However, the MICs for a few isolates clearly sep-
arated from the wild-type MIC range of their species,
suggesting acquisition of resistance mechanisms. This
was seen for flucytosine and one C. neoformans, and
for fluconazole and two isolates, one each of C. deneo-
formans and C. curvatus suggesting that susceptibility
testing is valuable for clinical isolates and that estab-
lishment of EUCAST epidemiological cut-off values and
clinical breakpoints are warranted. CLSI has deter-
mined modal MICs and proposed epidemiological cut-
off values (ECVs) for C. neoformans s.l. and ampho-
tericin B (0.25 and 1 mg l 1 respectively), flucytosine
(4 and 16 mg l 1 respectively), fluconazole (4 and
8 mg l 1 respectively), voriconazole (0.06 and
0.125 mg l 1 respectively) and isavuconazole (0.03
and 0.125 mg l 1 respectively) and for C. gattii s.l.
8 2016 Blackwell Verlag GmbH
and amphotericin B (0.5 and 1 mg l 1 respectively),
flucytosine (1 and 8 mg l 1 respectively), fluconazole
(4 and 8 mg l 1 respectively), voriconazole (0.12 and
0.5 mg l 1 respectively) and isavuconazole (0.03 and
0.25 mg l 1 respectively).4244 Whereas the modal
MICs mirror those found for C. neoformans, those for C.
gattii s.l. did not. In particular for C. gattii s.l., adop-
tion of the CLSI ECVs for our data set would bisect the
distributions and lead to a random number of isolates
with MICs above the ECV.
In conclusion, the epidemiology of cryptococcal
infections in Denmark reflects largely the Western
European situation with being C. neoformans genotype
AFLP1/VNI as the major cause of disease, followed by
C. deneoformans and C. gattii s.l. The latter were found
to be genetically closely related with those originating
from endemic regions, suggesting that the Danish
patients acquired their infection while travelling
outside of Europe. Antifungal susceptibility testing
showed species-specific differential susceptibility, but
suggested that acquired resistance was an infrequent
phenomenon.
Acknowledgments
J.F.M. received grants form Astellas, Merck and Basi-
lea. He has been a consultant to Basilea and Merck
and received speaker fees from Merck, Pfizer, United
Medical and Gilead. M.C.A. received speaker honoraria
over the past 5 years from Astellas, Basilea, Gilead,
MSD, Pfizer and T2Biosystems and received research
grants/support (company/investigator initiated) paid to
the institution from Astellas, Basilea, Gilead and
T2Biosystems.
References
1 Liu XZ, Wang QM, G oker M et al. Towards an integrated phylogenetic
classification of the Tremellomycetes. Stud Mycol 2015; 81: 85147.
2 Arendrup MC, Boekhout T, Akova M, Meis JF, Cornely OA, Lorthol-
ary O. ESCMID and ECMM joint clinical guidelines for the diagnosis
and management of rare invasive yeast infections. Clin Microbiol
Infect 2014; 20(Suppl. 3): 7698.
3 Khawcharoenporn T, Apisarnthanarak A, Mundy LM. Non-neofor-
mans cryptococcal infections: a systematic review. Infection 2007;
35: 5158.
4 Pan W, Liao W, Hagen F et al. Meningitis caused by Filobasidium
uniguttulatum: case report and overview of the literature. Mycoses
2012; 55: 1059.
5 Park BJ, Wannemuehler KA, Marston BJ, Govender N, Pappas PG,
Chiller TM. Estimation of the current global burden of cryptococcal
meningitis among persons living with HIV/AIDS. AIDS 2009; 23:
52530.
6 Hagen F, Colom MF, Swinne D et al. Autochthonous and dormant
Cryptococcus gattii infections in Europe. Emerg Infect Dis 2012; 18:
161824.

7 Saha DC, Goldman DL, Shao X et al. Serologic evidence for reactiva-
tion of cryptococcosis in solid-organ transplant recipients. Clin Vac-
cine Immunol 2007; 14: 15504.
8 Velagapudi R, Hsueh YP, Geunes-Boyer S, Wright JR, Heitman J.
Spores as infectious propagules of Cryptococcus neoformans. Infect
Immun 2009; 77: 434555.
9 Bovers M, Hagen F, Kuramae EE, Boekhout T. Six monophyletic lin-
eages identified within Cryptococcus neoformans and Cryptococcus gat-
tii by multi-locus sequence typing. Fungal Genet Biol 2008; 45:
40021.
10 Hagen F, Khayhan K, Theelen B et al. Recognition of seven species
in the Cryptococcus gattii/Cryptococcus neoformans species complex.
Fungal Genet Biol 2015; 78: 1648.
11 Meyer W, Aanensen DM, Boekhout T et al. Consensus multi-locus
sequence typing scheme for Cryptococcus neoformans and Cryptococcus
gattii. Med Mycol 2009; 47: 56170.
12 Ngamskulrungroj P, Gilgado F, Faganello J et al. Genetic diversity of
the Cryptococcus species complex suggests that Cryptococcus gattii
deserves to have varieties. PLoS ONE 2009; 4: e5862.
13 Bovers M, Hagen F, Kuramae EE et al. Unique hybrids between the
fungal pathogens Cryptococcus neoformans and Cryptococcus gattii.
FEMS Yeast Res 2006; 6: 599607.
14 Bovers M, Hagen F, Kuramae EE et al. AIDS patient death caused by
novel Cryptococcus neoformans 9 C. gattii hybrid. Emerg Infect Dis
2008; 14: 11058.
15 Cogliati M. Global molecular epidemiology of Cryptococcus neoformans
and Cryptococcus gattii: an atlas of the molecular types. Scientifica
2013; 2013: 675213.
16 Viviani MA, Cogliati M, Esposto MC et al. Molecular analysis of 311
Cryptococcus neoformans isolates from a 30-month ECMM survey of
cryptococcosis in Europe. FEMS Yeast Res 2006; 6: 6149.
17 Bergman F. Occurrence of Cryptococcus neoformans in Sweden. Acta
Med Scand 1963; 174: 6515.
18 Knudsen JD, Jensen L, Srensen TL et al. Cryptococcosis in Denmark:
an analysis of 28 cases in 19881993. Scand J Infect Dis 1997; 29:
5155.
19 Sonck CE. On the incidence of yeast species from human sources in
Finland. V. Yeasts occurring in mixed cultures with dermatophytes,
or with other yeast species. Mykosen 1980; 23: 34350.
20 Arsic Arsenijevic V, Pekmezovic MG, Meis JF, Hagen F. Molecular
epidemiology and antifungal susceptibility of Serbian Cryptococcus
neoformans isolates. Mycoses 2014; 57: 3807.
21 Hagen F, Vreuls W, Wouters M, Hendriks MJ, Meis JF. Cryptococcus
gattii infections in the Netherlands a retrospective molecular diag-
nostic study using formalin fixed paraffin embedded pulmonary coin
lesions. Mycoses 2014; 57(Suppl. 1): 103.
22 Boekhout T, Theelen B, Diaz M et al. Hybrid genotypes in the
pathogenic yeast Cryptococcus neoformans. Microbiology 2001; 147:
891907.
23 Hagen F, Illnait-Zaragozi MT, Bartlett KH et al. In vitro antifungal
susceptibilities and amplified fragment length polymorphism genotyp-
ing of a worldwide collection of 350 clinical, veterinary, and envi-
ronmental Cryptococcus gattii isolates. Antimicrob Agents Chemother
2010; 54: 513945.
24 Chowdhary A, Prakash A, Randhawa HS et al. First environmental
isolation of Cryptococcus gattii, genotype AFLP5, from India and a
global review. Mycoses 2013; 56: 2228.
25 Hagen F, Chowdhary A, Prakash A, Yntema JB, Meis JF. Molecular
characterization of Cryptococcus gattii genotype AFLP6/VGII isolated
from woody debris of divi-divi (Caesalpinia coriaria), Bonaire, Dutch
Caribbean. Rev Iberoam Micol 2014; 31: 1936.
26 Illnait-Zaragoz MT, Ortega-Gonzalez LM, Hagen F, Martnez-Machin
GF, Meis JF. Fatal Cryptococcus gattii genotype AFLP5 infection in
an immunocompetent Cuban patient. Med Mycol Case Rep 2013; 2:
4851.
27 Lindberg J, Hagen F, Laursen A, Stenderup J, Boekhout T. Cryptococ-
cus gattii risk for tourists visiting Vancouver Island, Canada. Emerg
Infect Dis 2007; 13: 1789.
2016 Blackwell Verlag GmbH
Cryptococcosis in Denmark during 19732013
28 Lockhart SR, Iqbal N, Harris JR et al. Cryptococcus gattii in the United
States: genotypic diversity of human and veterinary isolates. PLoS
ONE 2013; 8: e74737.
29 Springer DJ, Billmyre RB, Filler EE et al. Cryptococcus gattii VGIII iso-
lates causing infections in HIV/AIDS patients in Southern California:
identification of the local environmental source as arboreal. PLoS
Pathog 2014; 10: e1004285.
30 Trilles L, Wang B, Firacative C, Lazera Mdos S, Wanke B, Meyer W.
Identification of the major molecular types of Cryptococcus neoformans
and C. gattii by hyperbranched rolling circle amplification. PLoS ONE
2014; 9: e94648.
31 Tamura K, Stecher G, Peterson D, Filipski A, Kumar S. MEGA6:
molecular evolutionary genetics analysis version 6.0. Mol Biol Evol
2013; 30: 27259.
32 Irinyi L, Serena C, Garcia-Hermoso D et al. International Society of
Human and Animal Mycology (ISHAM)-ITS reference DNA barcod-
ing databasethe quality controlled standard tool for routine identifi-
cation of human and animal pathogenic fungi. Med Mycol 2015; 53:
3137.
33 Hagen F, Illnait-Zaragoz MT, Meis JF et al. Extensive genetic diver-
sity within the Dutch clinical Cryptococcus neoformans population. J
Clin Microbiol 2012; 50: 191826.
34 Sanchini A, Smith IM, Sedlacek L, Schwarz R, Tintelnot K, Rickerts
V. Molecular typing of clinical Cryptococcus neoformans isolates col-
lected in Germany from 2004 to 2010. Med Microbiol Immunol
2014; 203: 33340.
35 Dromer F, Mathoulin-Pelissier S, Launay O, Lortholary O; French
Cryptococcosis Study Group. Determinants of disease presentation
and outcome during cryptococcosis: the CryptoA/D study. PLoS Med
2007; 4: e21.
36 Dromer F, Mathoulin S, Dupont B, Laporte A. Epidemiology of cryp-
tococcosis in France: a 9-year survey (19851993). French Crypto-
coccosis Study Group. Clin Infect Dis 1996; 23: 8290.
37 Guinea J, Hagen F, Pel aez T et al. Antifungal susceptibility, serotyp-
ing, and genotyping of clinical Cryptococcus neoformans isolates col-
lected during 18 years in a single institution in Madrid, Spain. Med
Mycol 2010; 48: 9428.
38 Georgi A, Schneemann M, Tintelnot K et al. Cryptococcus gattii
meningoencephalitis in an immunocompetent person 13 months
after exposure. Infection 2009; 37: 3703.
39 Hagen F, van Assen S, Luijckx GJ, Boekhout T, Kampinga GA. Acti-
vated dormant Cryptococcus gattii infection in a Dutch tourist who
visited Vancouver Island (Canada): a molecular epidemiological
approach. Med Mycol 2010; 48: 52831.
40 Levy R, Pitout J, Long P, Gill MJ. Late presentation of Cryptococcus
gattii meningitis in a traveller to Vancouver Island: a case report.
Can J Infect Dis Med Microbiol 2007; 18: 1979.
41 Zaragoza O, Mesa-Arango AC, G omez-L opez A, Bernal-Martnez L,
Rodrguez-Tudela JL, Cuenca-Estrella M. Process analysis of vari-
ables for standardization of antifungal susceptibility testing of
nonfermentative yeasts. Antimicrob Agents Chemother 2011; 55:
156370.
42 Espinel-Ingroff A, Chowdhary A, Gonzalez GM et al. Multicenter
study of isavuconazole MIC distributions and epidemiological cutoff
values for the Cryptococcus neoformans-Cryptococcus gattii species com-
plex using the CLSI M27-A3 broth microdilution method. Antimicrob
Agents Chemother 2015; 59: 6668.
43 Espinel-Ingroff A, Aller AI, Canton E et al. Cryptococcus neofor-
mans-Cryptococcus gattii species complex: an international study
of wild-type susceptibility endpoint distributions and epidemio-
logical cutoff values for fluconazole, itraconazole, posacona-
zole, and voriconazole. Antimicrob Agents Chemother 2012; 56:
5898906.
44 Espinel-Ingroff A, Chowdhary A, Cuenca-Estrella M et al. Cryptococcus
neoformans-Cryptococcus gattii species complex: an international study
of wild-type susceptibility endpoint distributions and epidemiological
cutoff values for amphotericin B and flucytosine. Antimicrob Agents
Chemother 2012; 56: 310713.
9