+ Models

MYCMED-716; No. of Pages 8

Journal de Mycologie Médicale (2017) xxx, xxx—xxx

Available online at ScienceDirect

www.sciencedirect.com

ORIGINAL ARTICLE/ARTICLE ORIGINAL

The identification of Meyerozyma

guilliermondii from blood cultures and

surveillance samples in a university hospital

in Northeast Turkey: A ten-year survey

a, b b

N. Cebeci Güler *, ˙I. Tosun , F. Aydin

a

Department of Medical Microbiology, Faculty of Medicine, Giresun University, Giresun, Turkey

b

Department of Medical Microbiology, Faculty of Medicine, Karadeniz Technical University, Trabzon, Turkey

Received 8 March 2017; accepted 17 July 2017

KEYWORDS Summary Meyerozyma (Pichia) guilliermondii exists in human skin and mucosal surface

Meyerozyma microflora. It can cause severe fungal infections like candidemia, which is an opportunistic

guilliermondii; pathogen. One hundred and forty-one M. Guilliermondii isolates, consisting of 122 blood culture

Candida isolates, belonging to 126 patients; 13 total parenteral nutrition solution isolates; and two rectal

membranifaciens; swab isolates were identified according to carbohydrate assimilation reactions in a university

Blood culture; hospital in Turkey between January 2006 and December 2015. Following Candida albicans

ITS; (34.0%) and C. Parapsilosis (21.2%), the third yeast species most commonly isolated from blood

IGSAF; cultures in the Farabi Hospital was M. guilliermondii (20.6%). The patients were hospitalised in

Antifungal MIC 27 different departments. A total of 50% of the patients were in pediatric departments, 49.2%

were in intensive care units, and 17.2% were in haematology-oncology departments. Molecular

identification of the isolates was performed using DNA sequence analysis of ribosomal ITS gene

regions and IGS amplification-AluI fingerprinting (IGSAF). With molecular identification, 140

isolates were identified as M. guilliermondii and one isolate was identified as Candida mem-

branifaciens. It was observed that the ITS1 region specifically helps in identifying these species.

It was demonstrated that biochemical and molecular methods were 99.3% consistent in

identifying M. guilliermondii. The Wild-Type (WT) Minimum Inhibitory Concentrations (MICs)

distribution of fluconazole, voriconazole, itraconazole, and flucytosine were determined using

* Corresponding author.

E-mail address: [email protected] (N. Cebeci Güler).

http://dx.doi.org/10.1016/j.mycmed.2017.07.007

1156-5233/# 2017 Elsevier Masson SAS. All rights reserved.

Please cite this article in press as: Cebeci N, et al. The identification of meyerozyma guilliermondii from blood cultures and surveillance

samples in a university hospital in Northeast Turkey: A ten-year survey. Journal De Mycologie Médicale (2017), http://dx.doi.org/10.1016/ j.mycmed.2017.07.007

+ Models

MYCMED-716; No. of Pages 8

2 N. Cebeci Güler et al.

the Sensititre YeastOne YO2V system after 24 h of incubation. One M. guilliermondii strain was

determined to be non-WT for fluconazole, voriconazole, itraconazole and flucytosine. In total,

three M. guilliermondii strains, for fluconazole, were determined to be non-WT in this study.

# 2017 Elsevier Masson SAS. All rights reserved.

Introduction Nevertheless, antimycotic sensitivities of this fungus have

not been clarified.

Isolates identified as Candida guilliermondii (teleomorph In the Farabi Hospital, the M. guilliermondii has been the

Pichia guilliermondii) were included in the new Meyerozyma third yeast species most frequently isolated from blood

genus by Kurtzman and Suzuki in 2010 [1]. The cultures in the past 10 years. The objective of the present

M. guilliermondii complex is a genetically heterogeneous study was to assess whether the biochemical identification

complex comprising several phenotypically indistinguishable was correct using nucleotide sequencing and Restriction

taxa, including M. guilliermondii, Candida fermentati, Can- Fragment Length Polymorphism Analysis of PCR-Amplified

dida carpophila, and Candida xestobii [2—5]. However, the Fragments (PCR-RFLP) and to determine in vitro sensitivity of

species Candida famata (teleomorph Debaryomyces hanse- M. guilliermondii to fluconazole, voriconazole, itraconazole

nii) and M. guilliermondii are extremely difficult to diffe- and flucytosine.

rentiate phenotypically. They have a phylogenetically close

relationship [6,7].

Materials and Methods

M. guilliermondii is widely distributed in nature and is a

common constituent of the normal human microflora [8] and

is reported to be responsible for severe fungal infections Yeast strains

such as candidemia, a human opportunistic pathogen.

M. guilliermondii is defined as a newly emerging, rare This retrospective study evaluates a ten-year-period bet-

pathogen that is responsible for a small percentage of all ween January 2006 and December 2015 at Farabi Hospital,

candidemia. Moreover, in the past 20 years, it was observed which is a university hospital in Trabzon, Turkey. The

that this pathogen was responsible for 1—11.7% of all candi- M. guilliermondii isolate, which was first isolated from the

demia, with increasing incidence [9—11]. This species is a blood cultures of each patient, was included in the study.

more common cause of candidemia in cancer patients than it M. guilliermondii strains isolated at intervals of 4 weeks,

is in general hospital populations [11—14]. M. guilliermondii from successive specimens of the same patient, were consi-

fungemia may occur in children with underlying conditions dered as different candidemia episodes. One isolate repre-

other than cancer [15,16]. Besides sporadic candidemia senting each episode was included in the study. One hundred

cases, candidemia outbreaks (real or pseudo) caused by and forty-one M. guilliermondii isolates were identified from

the species were seen [17—19]. Candidemias are infections blood cultures (126 isolates) of 122 patients. Surveillance

with high mortality [20]. Furthermore, this species is espe- specimens consisted of total parenteral nutrition (TPN) solu-

cially notable for its greater propensity to express multidrug tions (13 isolates) and rectal swabs (two isolates). Most of the

resistance than other organisms of the genus Candida [21]. isolates (87.23%) belong to times of outbreak (Fig. 1).

Figure 1 Distribution of M. guilliermondii isolates isolated from blood cultures and surveillance specimens in the university hospital

in Northeast Turkey.

Please cite this article in press as: Cebeci N, et al. The identification of meyerozyma guilliermondii from blood cultures and surveillance

samples in a university hospital in Northeast Turkey: A ten-year survey. Journal De Mycologie Médicale (2017), http://dx.doi.org/10.1016/ j.mycmed.2017.07.007

+ Models

MYCMED-716; No. of Pages 8

Meyerozyma guilliermondii in the blood cultures and surveillance samples 3

Biochemical identification final extension step at 72 8C for 7 min. A negative control was

performed with each run by replacing the template DNA with

Blood cultures were processed in the Clinical Microbiology sterile water in the PCR mixture. All amplicons were purified

1

laboratory by an automated blood culture system (Bactec, using the NucleoSpin Extract II (Macherey-Nagel, Ger-

BD, Paramus, NJ, USA). Yeasts were initially observed on many). The DNA fragments were sequenced using a Genetic

Gram-stained preparations, subcultured on Sabouraud Dex- Analyzer 3130 automated DNA sequencer (Applied Biosys-

1

trose Agar (SDA; Oxoid, England). Surveillance samples were tems, USA) with a BigDye Terminator v3.1 Cycle Sequencing

directly inoculated to SDA. Germ tube-positive strains were Kit (Applied Biosystems). All amplicons were sequenced

identified as C. albicans and C. dubliniensis. The other using reverse primer ITS4 for the whole ITS region. ITS1,

strains were identified using carbohydrate assimilation reac- ITS2, ITS3 and ITS4 primers were placed on the whole ITS DNA

tions (API 20C AUX, bioMérieux, Marcy l’Etoile, France). sequence of the study isolates in silico. The sequence bet-

ween ITS1 and ITS4 primers was evaluated as the whole ITS

region sequence, the sequence between ITS1 and ITS2 pri-

DNA preparation

mers was evaluated as the ITS1 region sequence, and the

sequence between ITS3 and ITS4 primers was evaluated as

Meyerozyma guilliermondii isolates were subcultured on SDA

the ITS2 region sequence. Species were identified by search-

and incubated at 28 8C for 24 to 48 h. Colonies were suspen-

ing databases using the Basic Local Alignment Search Tool

ded in saline to obtain a turbidity of a 0.5 McFarland standard

(BLAST) (http://www.ncbi.nlm.nih.gov/BLAST/). A

at a 530 nm wavelength. One millilitre of the cell suspension

sequence identity of > 99% was used for species identifica-

was centrifuged at 5,000 g for 3 min in a microcentrifuge. The

tion. Pairwise comparisons and multiple sequence align-

genomic DNA was extracted by following isolation of nucleic

ments were also performed with CLUSTAL W2 (http://

acids from bacteria or yeast protocol, using a High Pure PCR

www.ebi.ac.uk/Tools/msa/clustalo/).

Template Preparation Kit (Roche, Germany) in accordance

with the manufacturer’s instructions. The extracted DNA was

stored at 20 8C until later use. Nucleotide sequence accession numbers

GenBank accession numbers of whole ITS sequences of study

Amplification and sequencing of the ITS region

isolates are given in Table 1.

Amplification of the whole internal transcribed spacer (ITS)

region of ribosomal DNA (rDNA) was performed with fungus- Amplification and AluI fingerprinting of the IGS

specific universal ITS1 (5"-TCCGTAGGTGAACCTGCGG-3") and region

ITS4 (5"-TCCTCCGCTTATTGATATGC-3") primers [22]. PCR was

performed in a total reaction volume of 50 mL consisting of Amplification of the whole rDNA intergenic spacer (IGS)

1.5 mM MgCl2, 0.8 mM deoxynucleoside triphosphates region was performed with LR13 (5"-CGATCTGCTGAGAT-

(0.2 mM each), 1.5 U of Taq DNA polymerase in 1X Taq buffer TAAG-3") and SR21 (5"-CTTAATCTTTGAGACAAGC-3") primers

(+KCl, MgCl2), 0.3 mM (each) of the ITS region primers designed by Nguyen et al. [23]. rRNA gene IGS amplification

(ITS1/ITS4), and 5 ng of DNA template. PCR was carried and AluI fingerprinting (IGSAF) was performed according to

out using the following conditions: initial denaturation at the method of Cornet et al. [24]. Differently, 1.5 U GoTaq

94 8C for 5 min, 35 cycles of denaturation (94 8C for 1 min), Flexi DNA polymerase was used and the restriction fragments

annealing (53 8C for 1 min), extension (72 8C for 1 min), and a were separated on 3% agarose gel.

Table 1 Blast outcomes of whole ITS, ITS1 and ITS2 sequences.

Isolate code GenBank Blast outcome of whole ITS, GenBank % Identity with

a

accession no ITS1 and ITS2 sequences accession sequence in

no GenBank of whole

ITS, ITS1and ITS2

sequences

C-569, C-610, C-650 KX580709 Meyerozyma guilliermondii KC237294 100 — 100 — 100

Candida carpophila DQ666191 99 — 99 — 99

Debaryomyces hansenii var. fabryi AF209874 99 — 99 — 99

Debaryomyces sp. JQ665430 99 — 99 — 100

C-840 KX580708 Candida membranifaciens EF197812 99 — 100 — 100

Candida friedrichii HQ283377 99 — 98 — 99

Other isolates KX580710 Meyerozyma guilliermondii AY939792 100 — 100 — 100

(137 isolates) Candida carpophila DQ666191 100 — 99 — 99

Debaryomyces hansenii var. fabryi AF209874 99 — 99 —99

a

The accession number were taken for one of the sequences completely identical to each other, and these isolates were shown on the

same row.

Please cite this article in press as: Cebeci N, et al. The identification of meyerozyma guilliermondii from blood cultures and surveillance

samples in a university hospital in Northeast Turkey: A ten-year survey. Journal De Mycologie Médicale (2017), http://dx.doi.org/10.1016/ j.mycmed.2017.07.007

+ Models

MYCMED-716; No. of Pages 8

4 N. Cebeci Güler et al.

Determination of WT MICs distributions of four 46.1% M. guilliermondii (identification code: 6576333) with

API 20 C AUX (bioMérieux, Marcy l’Etoile, France).

antifungal drugs

WT MICs values of fluconazole, voriconazole, itraconazole Sequencing of the whole ITS region

and flucytosine were obtained using Sensititre YeastOne

YO2V system panels (Trek Diagnostic Systems) in accordance PCR product size of the whole ITS region in all isolates was

with the manufacturer’s instructions. MIC values were deter- approximately 600 bp in 2% agarose gel. However, the pro-

mined visually for the first isolated 46 M. guilliermondii duct size of C-840 was larger than the others with a diffe-

strains (strains isolated between July 2006 and November rence can be distinguished visually. The blast outcomes of

2009) and 1 C. membranifaciens strain after 24 h incubation. sequences of whole ITS, ITS1 and ITS2 regions of this study’s

The M. guilliermondii epidemiological cutoff values (ECVs) isolates are seen in Table 1.

for fluconazole, voriconazole [25], itraconazole, and flucy-

tosine [26] were obtained as previously described.

AluI fingerprinting of the whole IGS region

Results The size of the PCR products, in the whole IGS region, was

approximately 3 kb in 0.7% agarose gel. When the intergenic

spacer rDNA amplification and AluI fingerprinting (IGSAF)

Distribution of the isolates according to years

method was applied, agarose gel band profiles of all study

and clinics

isolates (except for C-840) were the same, with

M. guilliermondii CBS 566 and M. guilliermondii CBS 2030

M. guilliermondii constituted 4.8% of all yeast strains isola-

reference strains (Fig. 2).

ted from blood cultures in the Farabi Hospital in 2006, 5.1% in

2007, and 3.8% in 2008. It was the fourth yeast species most

WT MIC distributions of four antifungal drugs

frequently isolated from blood cultures following

C. albicans, C. parapsilosis and C. tropicalis, in the said

WT MICs distributions of four antifungal drugs for 46

years. Interestingly, C. parapsilosis passed C. albicans in

M. guilliermondii isolates in Table 2 and one

2008 and became the most frequently isolated yeast species

C. membranifaciens isolate in Table 3 are shown.

from blood cultures. With a significant increase,

M. guilliermondii strain with C-962 code (MICs fluconazo-

M. guilliermondii constituted 71.3% of yeast isolated from

le > 256 mg/mL, voriconazole > 8 mg/mL, itraconazo-

blood cultures in 2009, 35.4% in 2011, and 29.4% in 2012

le > 16 mg/mL, flucytosine 2 mg/mL) was non-WT for

becoming the most frequently isolated yeast species from

fluconazole, voriconazole, itraconazole, and flucytosine.

blood cultures in these years. It was the third most fre-

In total, three M. guilliermondii strains was determined as

quently isolated yeast species from blood cultures in 2010,

non-WT for fluconazole in this study.

with a rate of 15.7%. This rate of M. guilliermondii isolated

from blood cultures continued to decline with rates of 5.1% in

2013, 0% in 2014, and 2.6% in 2015. In this study, according to

the data evaluated from the ten-year period between

January 2006 and December 2015, M. guilliermondii consti-

tuted 20.6% of all yeast isolated from blood cultures and was

the third most isolated yeast species following C. albicans

(34.0%) and C. parapsilosis (21.2%). C. tropicalis and other

yeast species constituted 7.0% and 17.2% of yeasts isolated

from blood cultures, respectively. M. guilliermondii was also

isolated from TPN solutions and rectal swabs of two patients.

2009 and 2011—2012 stand out as times of outbreak (Fig. 1).

The patients were hospitalised in 27 different depart-

ments in the Farabi Hospital. Fifty percent of the patients (61

patients) were hospitalised in paediatric departments. The

patients in the newborn intensive care unit (36 patients)

constituted 59.0% of paediatrics patients and 29.5% of all

patients. Sixty patients (49.2%) were hospitalised in inten-

sive care departments, 17.2% (21 patients) in haematology-

oncology departments, 4.9% (6 patients) in surgery depart-

ments, and 21.3% (26 patients) in other departments.

Biochemical identification

All study isolates, except for two (C-840 and C-987), were

identified as 84.3% M. guilliermondii -15.6% C. famata (iden- Figure 2 Band profiles of PCR products of whole IGS region

tification code: 6776373); C-840 was identified as 60.3% after cutting with AluI in 3% agarose gel. 1. It’s represents all

M. guilliermondii -39.6% C. famata (identification code: study isolates (except for C-840). 2. M. guilliermondii CBS 566. 3.

6756373) and C-987 was identified as 53.8% C. famata- M. guilliermondii CBS 2030. 4. 50-10000 bp HiLo DNA size marker.

Please cite this article in press as: Cebeci N, et al. The identification of meyerozyma guilliermondii from blood cultures and surveillance

samples in a university hospital in Northeast Turkey: A ten-year survey. Journal De Mycologie Médicale (2017), http://dx.doi.org/10.1016/ j.mycmed.2017.07.007

+ Models

MYCMED-716; No. of Pages 8

Meyerozyma guilliermondii in the blood cultures and surveillance samples 5 d

1 Discussion 256

>

Candida albicans is the most commonly isolated species

in candidemia. C. albicans is followed by C. glabrata,

C. tropicalis, and C. parapsilosis [27]. Although

M. guilliermondii has a low natural virulence, when compa-

red with C. albicans [28], it is regarded as a new, emerging

pathogen among non-albicans Candida species.

f

1 M. guilliermondii is responsible for approximately 2% of

candidemias [9,10]. However, M. guilliermondii was isolated

d

by Girmenia et al. [11] from 11.7% of 243 candidemia epi-

16)

sodes that occurred over the course of 22 years at the

>

e University La Sapienza in Rome, Italy; by Medeiros et al.

16 32 64 128

[18] from 43% of blood cultures in one year at a university

d hospital in Brazil; by Shah et al. [16] from 85.4% of the blood

8)

cultures in the paediatric population over the course of 22

>

c weeks in Ahmedabat, India; by Yagupsky et al. [17] from the

blood cultures of 14 infants in the neonatal intensive care

unit and 3 babies in the newborn unit over the course of three

weeks in Israel; and by Masala et al. [19] as the candidemia

d factor in 5 patients hospitalised in the surgery unit over the

391 2329 248

course of two weeks at the university hospital in Verona,

b Italy. However, in the studies of Medeiros, Shah and

Yagupsky, a candidemia pseudo-outbreak was mentioned

[16—18]. In literature, it is rarely observed that

0.5 1

M. guilliermondii is a species that is frequently isolated

from blood cultures and ranked first place among candidemia

a

factors. In this study, patients were hospitalised in 27 diffe-

7371 1( rent departments at the Farabi Hospital, most of them in the

newborn intensive care unit, haematology-oncology, and

surgery departments.

The receiving of TPN solution, intravenous catheters, and

hospitalization in intensive care units are reported as some

of the risk factors of M. guilliermondii candidemia

[11,15,19]. In this study, during the 1st and 2nd outbreak

strains using Sensititre YeastOne.

peaks (Fig. 1), M. guilliermondii was isolated from TPN

solutions of 13 patients, and blood cultures of 11 of these g/mL) of

m

13 patients on dates close to the isolates of TPN solution. TPN

0.032) 2 1

< solution ingredients were 3% NaCl (Polifarma Drug

Company), 30% dextrose (Eczac½bas¸½, Baxter), 6% TrophA-

mine (Eczac½bas¸½, Baxter), and 20% intralipid (Fresenius

M. guilliermondii Kabi). M. guilliermondii was isolated from the catheter

blood of 19.3% of patients (n = 24) and 49.2% of patients

(n = 60) were hospitalised in intensive care units.

M. guilliermondii outbreak is likely to be caused by TPN

solution. In this study, TPN solution was manually prepared

No. of isolates with MIC ( 0.008 0.016 0.032 0.064 0.12 0.25

by healthcare workers and given to patients. The outbreak

due to TPN solution is thought to have originated from

contamination from various sources during the preparation

phase. Following the outbreak, the manual methods used for

the preparation of TPN were abandoned and replaced by

Incubation time (h)

automated methods. Between October 2011 and March 2012,

318 patients in the Farabi Hospital received TPN solution, of

which 36 (11.32%) were infected. Eighteen of the 36 patients

(50%) died. No data was available for the other patients.

isolates tested

M. guilliermondii is a highly confusing species because of

the nomenclature having many synonyms and changes in the

WT MIC distribution of four antifungals for

taxonomy. M. guilliermondii cannot be accurately differen-

tiated with the routine biochemical identification method

ECV for voriconazole. ECV for itraconazole andECV flucytosine. for fluconazole. The C-962 strain. The C-787 strain. The C-876 strain.

used for C. famata (D. hansenii) [6,7]. Furthermore, f c a e b d

Fluconazole 46 24 Antifungal No. of Table 2 VoriconazoleItraconazole Flucytosine 1 43 ( 27 16 1 1 (

M. guilliermondii appears as a genetically heterogeneous

Please cite this article in press as: Cebeci N, et al. The identification of meyerozyma guilliermondii from blood cultures and surveillance

samples in a university hospital in Northeast Turkey: A ten-year survey. Journal De Mycologie Médicale (2017), http://dx.doi.org/10.1016/ j.mycmed.2017.07.007

+ Models

MYCMED-716; No. of Pages 8

6 N. Cebeci Güler et al.

Table 3 MIC (mg/mL) value of four antifungal drug for the C. membranifaciens (C-840) isolate using Sensititre YeastOne.

Species No. of isolates tested Incubation time (h) Fluconazole Voriconazole Itraconazole Flucytosine

C. membranifaciens 1 24 4 0.008 0.25 > 64

complex, including numerous phenotypically indistinguis- than other study isolates. Carbohydrate assimilation profiles

hable taxa that have been brought into synonymy, including and colony appearance can be guides in determining the

Candida fermantati (teleomorph Pichia caribbica), Candida isolates that need molecular identification during a routine

carpophila and Candida xestobii [2—5]. Differentiation of the diagnosis. C-840 isolate was isolated from the blood culture

M. guilliermondii complex (or clade) taxa and C. famata of a patient in the oncology department in 2008. Although

(D. hansenii) was attempted with methods sequencing rRNA C. membranifaciens is rarely isolated from clinical speci-

regions (ITS, D1/D2, IGS), electrophoretic karyotyping, spec- mens [34—36], it was reported as a factor for candidemia in a

trophotometric DNA/DNA reassociation, and PCR-RFLP [2— patient with AIDS [37].

7]. Previous studies have shown Matrix Assisted Laser Des- Only one nucleotide difference was observed between the

orption Ionization (MALDI) to be rapid, accurate, and cost- whole ITS sequences of all study isolates (except for C-840).

effective in the identification of uncommon species of Can- While ‘‘G (guanine)’’ nucleotide existed in C-569, C-610, and

dida [29]. However, M. guilliermondii was not identified by C-650 isolates in position 215 (in the ITS2 region, as nucle-

the MALDI Biotyper. A result of ‘‘no reliable identification’’ otides are counted starting from the ITS3 primer binding

was given rather than an incorrect organism identification region), ‘‘A (adenine)’’ nucleotide was observed in the same

[30]. position in the other isolates (137 isolates). The whole ITS

Compared to the rRNA genes, the rRNA ITS1 and ITS2 sequence of C-840 isolate was 75.94% similar to the whole ITS

regions tend to be more variable, offering extremely sequence of C-569, C-610 and C-650 isolates and 75.76%

valuable targets for fungal speciation and identification similar to the whole ITS sequence of the other isolates.

[31]. In this study, C-569, C-610, and C-650 isolates were It was observed that C-987 isolate grew slower than the

identified as M. guilliermondii using whole ITS and ITS1 other study isolates in SDA (approximately in 48 h) and the

sequences, but it was observed that whole ITS sequences carbohydrate assimilation profile (53.8% C. famata, 46.1%

did not differentiate C. carpophila and D. hansenii var. M. guilliermondii) was different than all other isolates.

fabryi. In addition, M. guilliermondii and Debaryomyces However, C-987 was identified as M. guilliermondii with

sp. could not be differentiated in the isolates when using ITS sequencing and PCR/RFLP (IGSAF). Sequencing of other

the ITS2 sequence. In the other isolates (except for C-840), rDNA gene regions (such as D1/D2, IGS) can be helpful for

M. guilliermondii and C. carpophila species could not be identification of this isolate.

differentiated when using the whole ITS sequence. However, In 2009, to differentiate D. hansenii and C. famata and

ITS1 and ITS2 sequences defined these isolates as reassess the phylogenetic relationship of Debaryomyces spe-

M. guilliermondii. It is observed that the ITS1 region is more cies, Nguyen et al. developed a PCR/FLP (IGSAF) method

distinctive than the other ITS regions regarding identifying based on cutting the IGS region amplification product with

M. guilliermondii. Moreover, there are very few ITS DNA AluI restriction endonuclease enzyme. In this study, LR13

sequences data regarding C. carpophila and C. caribbica in and SR21 primers were designed for the amplification of the

the GenBank. ITS DNA sequence results of the study isolates whole IGS region [23]. This IGS method could also easily

(141 isolates) were 99.3% consistent with the biochemical differentiate Debaryomyces hansenii (C. famata) from the

identification results (except for the C-840 isolate). M. guilliermondii complex. Moreover, the IGS fingerprints

While no differentiation could be made between Candida showed specific profiles for species belonging to the

(Pichia) membranifaciens and C. friedrichii using whole ITS M. guilliermondii complex (i.e., M. guilliermondii,

sequence, ITS1 and ITS2 sequences resulted in as C. carpophila, C. fermentati, and C. xestobii) [24], whereas

C. membranifaciens in the C-840 isolate. ITS1 and ITS2 these species cannot be distinguished using ITS or D1/D2

sequences of this isolate were 100% identical with sequencing [2]. The IGSAF method was applied to the study

C. membranifaciens. But, the ITS1 sequence was 98% iden- isolates, moving forward with the idea that ITS DNA sequenc-

tical and the ITS2 sequence 99% identical with C. friedrichii ing could not provide sufficient differentiation for identifica-

compared to GenBank data. Therefore, the ITS1 region tion of M. guilliermondii. In all isolates identified as

seems to be a more accurate preference for differentiation M. guilliermondii by ITS DNA sequencing, the same IGSAF

of these two species (see Table 1). It is demonstrated that band profile, with M. guilliermondii CBS 566 and

M. guilliermondii has a distant relationship with M. guilliermondii CBS 2030 strains, was observed. With

C. membranifaciens, the type species of genus Pichia [32]. molecular identification of the study isolates, the probability

For differentiation of C. membranifaciens and C. friedrichii of erroneous biochemical identification was eliminated.

species, which are included in genus Yamadazyma and have M. guilliermondii demonstrates cross-resistance to azoles

the closest phylogenetic relationship, it was previously (an isolate resistant to fluconazole and itraconazole [38]; an

reported that the ITS region was more distinctive than the isolate resistant to fluconazole, itraconazole and voricona-

D1/D2 region [33]. Moreover, C. membranifaciens cannot be zole [39]; and an isolate resistant to fluconazole, itracona-

identified by the biochemical identification kit used in this zole, voriconazole and posaconazole [40] were reported in

study. The colony appearance of C-840 isolate in SDA (media- the literature). Although mostly susceptible, the isolates

invasive, distinctly) and a carbohydrate assimilation profile resistant to flucytosine were reported [19].

(60.3% M. guilliermondii, 39.6% C. famata) was different M. guilliermondii appears as constitutively less susceptible

Please cite this article in press as: Cebeci N, et al. The identification of meyerozyma guilliermondii from blood cultures and surveillance

samples in a university hospital in Northeast Turkey: A ten-year survey. Journal De Mycologie Médicale (2017), http://dx.doi.org/10.1016/ j.mycmed.2017.07.007

+ Models

MYCMED-716; No. of Pages 8

Meyerozyma guilliermondii in the blood cultures and surveillance samples 7

to polyenes and echinocandins than other yeast-like fungi,

Disclosure of interest

and it has been labelled as one of the least echinocandin

susceptible yeast, along with C. parapsilosis [41]. Neverthe-

The authors declare that they have no competing interest.

less, polyenes and echinocandins were not analysed and

fluconazole, voriconazole, itraconazole and flucytosine sen- Acknowledgements

sitivity tests could only be performed for 46 isolates due to

financial constraints in this study. Even so, in this study, the

We thank Dr Huu-Vang Nguyen (Collection de Levures

number of M. guilliermondii isolates, for which antifungal

d"Intérêt Biotechnologique, Laboratoire de Microbiologie

susceptibility tests were performed, is higher than the num-

Génétique Moléculaire, Thiverval-Grignon, France) for

bers in previous studies and the identification of these

providing us with M. guilliermondii CBS 566 and

isolates were verified with molecular methods.

M. guilliermondii CBS 2030.

The C-962 strain of M. guilliermondii was non-WT for

fluconazole, voriconazole, itraconazole and flucytosine. In

total, three M. guilliermondii strains (C-962, C-787, and C- References

876) were determined as non-WT for fluconazole in this

study. The isolated C-962 strain belonged to the epidemic

[1] Kurtzman CP, Suzuki M. Phylogenetic analysis of ascomycete

period in November 2009. C-787 and C-876, determined as

yeasts that form coenzyme Q-9 and the proposal of the new

non-WT for fluconazole, were isolated in April 2008 and July

genera Babjeviella, Meyerozyma, Millerozyma, Priceomyces

2009, respectively. The patient with the C-876 isolate was a

and Scheffersomyces. Mycoscience 2010;51:2—14.

premature (25 weeks) male, lying in neonatal intensive care [2] Vaughan-Martini A, Kurtzman CP, Meyer SA, O’Neill EB. Two

unit. In a blood culture taken 25 days after hospitalization, new species in the Pichia guilliermondii clade: Pichia caribbica

M. guilliermondii was detected while patients with nosoco- sp. nov., the ascosporic state of Candida fermentati and

Candida carpophila comb. nov. FEMS Yeast Res 2005;5:463—9.

mial sepsis diagnoses were receiving antibiotic and liposomal

[3] Lan L, Xu J. Multiple gene genealogical analyses suggest diver-

amphotericin B therapy. The patient with the C-962 isolate

gence and recent clonal dispersal in the opportunistic human

was a 3-year-old boy. The patient taken diagnosis of mali-

pathogen Candida guilliermondii. Microbiology

gnant renal neoplasm 2.5 months ago was admitted to the

2006;152:1539—49.

hospital again due to recurrence. Following immunosuppres-

[4] Lockhart SR, Messer SA, Pfaller MA, Diekema DJ. Identification

sive treatment, the patient received antibiotic and caspo-

and susceptibility profile of Candida fermentati from a world-

fungin acetate with a diagnosis of nosocomial sepsis. In a

wide collection of Candida guilliermondii clinical isolates. J

blood culture taken on the twelfth day after hospitalization, Clin Microbiol 2009;47:242—4.

M. guilliermondii was detected. After the patient died, the [5] Kurtzman CP, Suzuki M. In: Kurtzman CP, Fell JW, Boekhout T,

blood culture was positive. There was no information on the editors. The yeasts: a taxonomic study. Elsevier; 2011. p. 622—

3 [2010].

patient with the C-787 isolate.

[6] Nishikawa A, Sugita T, Shinoda T. Differentiation between

The MIC value of flucytosine for C. membranifaciens (C-

Debaryomyces hansenii/Candida famata complex and Candida

840) isolate was determined to be high (> 64 mg/mL).

guilliermondii by polymerase chain reaction. FEMS Immunol

Although there are very few antifungal susceptibility results

Med Microbiol 1997;19:125—9.

of C. membranifaciens strains in the literature, azole resis-

[7] Desnos-Ollivier M, Ragon M, Robert V, Raoux D, Gantier JC,

tance and high MIC for flucytosine were reported [35,36]. It is

Dromer F. Debaryomyces hansenii (Candida famata), a rare

very valuable to understand the antifungal susceptibilities of

human fungal pathogen often misidentified as Pichia guillier-

this species that are rarely isolated from clinical specimens, mondii (Candida guilliermondii). J Clin Microbiol

to be able to successfully treat new cases. 2008;46:3237—42.

Based on the data shown in the present study, the pro- [8] Odds FC. Ecology of Candida and epidemiology of candidosis.

Candida and candidosis: a review and bibliography, . 2nd ed,

bability of an erroneous result using biochemical identifica-

London, England: Bailliere Tindall; 1988: 68—78.

tion was very remote for the M. guilliermondii. However,

[9] Tortorano AM, Peman J, Bernhardt H, et al. Epidemiology of

C. membranifaciens, a phylogenetically closely related cryp-

candidemia in Europe: results of 28-month European Confed-

tic species, can be identified from the M. guilliermondii

eration of Medical Mycology (ECMM) hospital-based surveillance

complex strains by ITS sequencing. Determining the clonal

study. Eur J Clin Microbiol Infect Dis 2004;23:317—22.

proximity of outbreak isolates by molecular typing methods

[10] Bedini A, Venturelli C, Mussini C, et al. Epidemiology of candi-

such as AP-PCR, Rep-PCR, and PFGE will contribute to the

demia and antifungal susceptibility patterns in an Italian ter-

confirmation of the outbreak source. tiary care hospital. Clin Microbiol Infect 2006;12:75—80.

[11] Girmenia C, Pizzarelli G, Cristini F, et al. Candida guilliermon-

dii fungemia in patients with hematologic malignancies. J Clin

Ethical statement

Microbiol 2006;44:2458—64.

[12] Mardani M, Hanna HA, Girgawy E, Raad I. Nosocomial Candida

The authors alone are responsible for the content and writing

guilliermondii fungemia in cancer patients. Infect Control Hosp

of this paper. Epidemiol 2000;21:336—7.

[13] Kabbara N, Lacroix C, Latour RP, Socie G, Ghannoum M, Ribaud

Funding/Support P, et al. parapsilosis and C. guilliermondii blood stream infec-

tions in allogeneic hematopoetic stem cell transplant recipients

receiving long-term caspofungin therapy. Haematologica

This study was partially supported by Giresun University 2008;93:639—40.

Scientific Research Projects Coordination Unit, with the [14] Chen CY, Huang SY, Tang JL, et al. Clinical features of patients

project no. SAG˘-BAP-A-140316-96. with infections caused by Candida guilliermondii and Candida

Please cite this article in press as: Cebeci N, et al. The identification of meyerozyma guilliermondii from blood cultures and surveillance

samples in a university hospital in Northeast Turkey: A ten-year survey. Journal De Mycologie Médicale (2017), http://dx.doi.org/10.1016/ j.mycmed.2017.07.007

+ Models

MYCMED-716; No. of Pages 8

8 N. Cebeci Güler et al.

fermantati and antifungal susceptibility of the isolates at a [28] Bramono K, Yamazaki M, Tsuboi R, Ogawa H. Comparison of

medical centre in Taiwan, 2001-10. J Antimicrob Chemother proteinase, lipase and alpha-glucosidase activities from the

2013;68:2632—5. clinical isolates of Candida species. Jpn J Infect Dis

[15] Peman J, Bosch M, Canton E, et al. Fungemia due to Candida 2006;59:73—6.

guilliermondii in a pediatric and adult population during a 12- [29] Dhiman N, Hall L, Wohlfiel SL, Buckwalter SP, Wengenack NL.

year period. Diagn Microbiol Infect Dis 2008;60:109—12. Performance and cost analysis of matrix-assisted laser desorp-

[16] Shah P, Bhatia P. Pseudooutbreak of Candida guilliermondii tion ionization — time of flight mass spectrometry for routine

fungemia in neonatal intensive care unit. IJSR 2012;1:112—3. identification of yeast. J Clin Microbiol 2011;49:1614—6.

[17] Yagupsky P, Dagan R, Chipman M, Goldschmied-Reouven A, [30] Castanheira M, Woosley LN, Diekema DJ, Jones RN, Pfaller MA.

Zmora E, Karplus M. Pseudooutbreak of Candida guilliermondii Candida guilliermondii and other species of Candida misiden-

fungemia in a neonatal intensive care unit. Pediatr Infect Dis J tified as Candida famata: Assessment by Vitek 2, DNA sequenc-

1991;10:928—32. ing analysis, and Matrix-Assisted Laser desorption Ionization-

[18] Medeiros EA, Lott TJ, Colombo AL, et al. Evidence for a pseudo- time of flight mass spectrometry in two global antifungal

outbreak of Candida guilliermondii fungemia in a university surveillance programs. J Clin Microbiol 2013;51:117—24.

hospital in Brazil. J Clin Microbiol 2007;45:942—7. [31] Liu D. Introductory Remarks. In: Liu D, editor. Molecular

[19] Masala L, Luzzati R, Maccacaro L, Antozzi L, Concia E, Fontana Detection of Human Fungal Pathogens. Boca Raton: CRC Press

R. Nosocomial cluster of Candida guilliermondii fungemia in Taylor & Francis Group; 2011. p. 15—20.

surgical patients. Eur J Clin Microbiol Infect Dis 2003;22: [32] Kurtzman CP, Robnett CJ. Identification of clinically important

0

686—8. ascomycetous yeasts based on nucleotide divergence in the 5

[20] Krcmery V, Barnes AJ. Non-albicans Candida spp. causing fun- end of the large-subunit (26S) ribosomal DNA gene. J Clin

gaemia: pathogenicity and antifungal resistance. J Hosp Infect Microbiol 1997;35:1216—23.

2002;50:243—60. [33] Groenewald M, Robert V. Smith MTh. The value of the D1/D2

[21] Pfaller MA, Diekema DJ, Mendez M, et al. Candida guilliermon- and internal transcribed spacers (ITS) domains for the identifi-

dii, an opportunistic fungal pathogen with decreased suscepti- cation of yeast species belonging to the genus Yamadazyma.

bility to fluconazole: geographic and temporal trends from the Persoonia 2011;26:40—6. http://dx.doi.org/10.3767/

ARTEMIS DISK antifungal surveillance program. J Clin Microbiol 003158511X559610 [Epub 2011 Feb 3].

2006;44:3551—6. [34] Mandras N, Tullio V, Allizond V, et al. In vitro activities of

[22] Leaw SN, Chang HC, Sun HF, Barton R, Bouchara JP, Chang TC. fluconazole and voriconazole against clinical isolates of Candi-

Identification of medically important yeast species by sequence da spp. determined by disk diffusion testing in Turin, Italy.

analysis of the internal transcribed spacer regions. J Clin Antimicrob Agents Chemother 2009;53:1657—9.

Microbiol 2006;44:693—9. [35] Cendejas-Bueno E, Gomez-Lopez A, Mellado E, Rodriguez-

[23] Nguyen HV, Gaillardin C, Neuveglise C. Differentiation of Deba- Tudela JL, Cuenca-Estrella M. Identification of pathogenic rare

ryomyces hansenii and Candida famata by rRNA gene intergenic yeast species in clinical samples: comparison between pheno-

spacer fingerprinting and reassessment of phylogenetic rela- typical and molecular methods. J Clin Microbiol 2010;48:1895—

tionships among D. hansenii, C. famata, D. fabryi, C. flareri 9.

(D. subglobosus) and D. prosopidis: description of [36] Cuenca-Estrella M, Gomez-Lopez A, Gutierrez MO, Buitrago MJ,

D. vietnamensis sp. nov. closely related to D. nepalensis. FEMS Rodriguez-Tudela JL. Reliability of the WIDERYST susceptibility

Yeast Res 2009;9:641—62. testing system for detection of in vitro antifungal resistance in

[24] Cornet M, Sendid B, Fradin C, Gaillardin C, Poulain D, Nguyen yeasts. Antimicrob Agents Chemother 2008;52:1062—5.

HV. Molecular identification of closely related Candida species [37] Couto FM, Macedo DP, Neves RP. Fungemia in a university

using two ribosomal intergenic spacer fingerprinting methods. J hospital: an epidemiological approach. Rev Soc Bras Med Trop

Mol Diagn 2011;13:12—21. 2011;44:745—8.

[25] Pfaller MA, Castanheira M, Diekema DJ, Messer SA, Jones RN. [38] Tietz HJ, Czaika V, Sterry W. Case report. Osteomyelitis caused

Triazole and echinocandin MIC distributions with epidemiologi- by high resistant Candida guilliermondii. Mycoses

cal cutoff values for differentiation of wild-type strains from 1999;42:577—80.

non-wild-type strains of six uncommon species of Candida. J [39] Chakrabarti A, Chatterjee SS, Rao KL, et al. Recent experience

Clin Microbiol 2011;49:3800—4. with fungaemia: change in species distribution and azole resis-

[26] Pfaller MA, Espinel-Ingroff A, Canton E, et al. Wild-Type MIC. tance. Scand J Infect Dis 2009;41:275—84.

distributions and epidemiological cutoff values for amphoteri- [40] Savini V, Catavitello C, Di Marzio I, et al. Pan-azole-resistant

cin B, flucytosine, and itraconazole and Candida spp. as deter- Candida guilliermondii from a leukemia patient silent funguria.

mined by CLSI broth microdilution. J Clin Microbiol Mycopathologia 2010;169:457—9.

2012;50(6):2040—6. [41] Savini V, Catavitello C, Onofrillo D, et al. What do we know

[27] Guinea J. Global trends in the distribution of Candida species about Candida guilliermondii? A voyage throughout past and

causing candidemia. CMI 2014;20:5—10. http://dx.doi.org/ current literature about this emerging yeast. Mycoses

10.1111/1469-0691.12539 [Epub 2014 Feb 10]. 2011;54:434—41.

Please cite this article in press as: Cebeci N, et al. The identification of meyerozyma guilliermondii from blood cultures and surveillance

samples in a university hospital in Northeast Turkey: A ten-year survey. Journal De Mycologie Médicale (2017), http://dx.doi.org/10.1016/ j.mycmed.2017.07.007