Review
Comparison of molecular and metabolomic methods as characterizationtools of
Debaryomyces hansenii
cheese isolates
Marzia Del Bove, Monia Lattanzi, Paolo Rellini, Cristina Pelliccia, Fabrizio Fatichenti, Gianluigi Cardinali
Department of Applied Biology – Microbiology, University of Perugia, Borgo 20 Giugno, 74, I-06121 Perugia, Italy
a r t i c l e i n f o
Article history:
Received 19 November 2008Received in revised form17 March 2009Accepted 21 March 2009Available online 2 April 2009
Keywords:Debaryomyces hansenii
Multidimensional analysisCheeseTypingMetabolomicsFTIR
a b s t r a c t
Debaryomyces hansenii
is one of the yeast species most frequently isolated from cheese and salty foods,however little is known about the phenotypic and molecular variability of its strains. In order to explorethe possibilities of a large study on its biodiversity, some
D. hansenii
strains were selectively isolated frompecorino cheese sampled in ten different Italian regions. All isolates were identified as
D. hansenii
on thebasis of conventional and molecular taxonomic analysis. The D1/D2 domain sequences of the 26S-rDNAdid not show any variation, confirming the extreme homogeneity of this species. PCR-duplex-RAPDbanding patterns analyzed with PCoA showed interesting clustering related to the geographic areas of isolation, although some overlapping between strains derived from different districts could be observed.A FTIR (Fourier Transform Infrared Spectroscopy) metabolomic fingerprint produced groupings weaklyrelated to those observed with RAPD and less associated with the isolation locales. The discriminatorypower of metabolomic fingerprint was able to discriminate strains otherwise considered identical. Thispreliminary study showed that, in spite of the homogeneity at the 26S-rDNA level, the
D. hansenii
strainsexhibit high molecular and metabolomic variability somehow linked to the places of isolation. Furtherstudies will be necessary to better investigate on the link between terroir and strain variability, as well ason the relation between genotypic and metabolomic fingerprints.
Ó
2009 Elsevier Ltd. All rights reserved.
1. Introduction
Debaryomyces hansenii
is one of the prevalent yeast speciesisolated from cheese (Fadda et al., 2004; Mounier et al., 2006b;Romano et al., 2001), exerting important metabolic activities incheese ripening (Bonaiti et al., 2004; Cholet et al., 2007; Kumuraet al., 2002; Lopez Del Castillo-Lozano et al., 2007) and in limitingthe growth of deleterious spoiling bacteria (Fatichenti et al.,1983).Its high resistance to sodium chloride (Butinar et al., 2005; Corteet al., 2006) makes
D. hansenii
ideally suited to many salty envi-ronments (Gadanho et al., 2003; Seiler and Busse, 1990), amongwhich cheese is the best (Kurtzman and Robnett, 1998) studied.This species is an important component in the production of Cheddar and Gouda cheeses, (Viljoen and Greyling, 1995; Weltha-genandViljoen,1998),inbluecheeses(Addisetal.,2001;Besanconet al.,1992; Roostita and Fleet,1996), smear cheese (Corsetti et al.,2001; Genteet al.,2007;Leclercq-Perlatet al., 2000; Mounieretal.,2006b; Rea et al., 2007), buffalo mozzarella (Romano et al., 2001)
andinseveralcheesetypesmadewithewemilk(Pereira-Diasetal.,2000), particularly the Italian ‘‘Pecorino cheese’’ (Cosentino et al.,2001; Fadda et al., 2004; Gardini et al., 2006). In spite of the greatamount of literature on the species
D. hansenii
, few studies havebeendedicatedtodifferentiationanddistributionatthestrainlevel(Petersen et al., 2001, 2002).The desire for quality food products related to specific ‘‘terroir’’(Barham,2003)orevenfoodsboughtattheirfarmoriginpoint(Gilg
andBattershill,1998)hasledtoaseriesofeffortstoproposereliableandrelativelyinexpensivemeasurestoestablishproceduresforfoodtraceability on a solid scientific ground. Among the strategiesexplored, special attention has been paid to sesquiterpens asmarkers of the mountain origin of some cheese types (Favaro et al.,2005), although the high compositional variability of cheeses hashindered the development of a general procedure applicable toevery cheese type based on these or similar compounds.Metabolomics is the last born of the ‘‘omics’’ sciences aftergenomics, transcriptomics and proteomics. Highly reproduciblemetabolomic fingerprints made with the Fourier TransformInfraredSpectroscopy(FTIR)havebeenusedsuccessfullytoidentifyyeast and bacteria at the species level (Helm et al.,1991; Kummerleet al., 1998; Rudol and Scherer, 2001; Wenning et al., 2006) andin some instances at the strain level (Naumann et al., 1991; Zhaoet al., 2004).
*
Corresponding author. Tel.:
þ
39 075 585 6478; fax:
þ
39 075 585 6470.
E-mail address:
gianlu@unipg.it(G. Cardinali).
Contents lists available atScienceDirect
Food Microbiology
journal homepage:www.elsevier.com/locate/fm
0740-0020/$ – see front matter
Ó
2009 Elsevier Ltd. All rights reserved.doi:10.1016/j.fm.2009.03.009
Food Microbiology 26 (2009) 453–459
The hypothesis explored in this paper is that
D. hansenii
strainsmay be natural and ubiquitous markers of cheese origin. Thispossibility is based on the presence of this yeast in all cheesesstudied and on the fact that a high number of cells survive afterripening (Mounier et al., 2006a). We took into considerationdifferent types of Pecorino cheese produced with diverse protocolsin various Italian regions to test the hypothesis that a correlationexists between strain characterizing descriptors and the areaof isolation. The variability within the species was studiedwith molecular and metabolomic tools in order to assess thediversity both at the DNA and at the phenotypic level, with rDNAsequencing, RAPD profiling, FTIR fingerprints and assimilativetraits.Thisstudyconsideredonly twenty-twostrainsfromdifferentlocales in order to assess the presence of some type of variabilityand its putative link with the area of isolation. On the basis of theresults obtained, further studies have been planned to investigatethe
D. hansenii
variability more systematically and with morecharacterization tools.
2. Materials and methods
2.1. Media, growth conditions and pre-identification procedure
Strains were grown and maintained in YEPDA plates (YeastExtract 1%, Peptone 1% Dextrose 2%, Agar 1.7%). For selective strainisolation, IM isolation medium was employed (YEPDA plates sup-plemented with Bengal Rose 0.025 g/l, NaCl 10% w/v and chlor-amphenicol 0.1 g/l). Pre-identification of strains was carried out byassimilation tests performed on plates according to the methods inYarrow (Yarrow, 1998), using the carbon sources listed inTable 1,
particularly those which do not show variability in the standardspecies description. Pre-identification was used as a dereplicationmethod in order to include the putative
D. hansenii
strains in thestudy and to exclude those belonging to other species. Identifica-tion was confirmed by sequencing the 26S D1/D2 domain asdescribed below. Biomass for FTIR analysis was obtained withmetabolomic medium (MM: YEPDA without peptone), using highpurity Yeast Extract (Difco Laboratories, MD, USA). Cells weregrown on a MM pre-culture for 24 h and then transferred to a freshMM plate for further 24h. This method with two successive 24 hgrowth ensures that cells are harvested in the same physiologicalcondition and that the differences between spectra are only due tothe different metabolomic arrangement typical of the strain.Growth, isolation and assimilation were carried out at 25
C;assimilations were read after 2 and 4 days.
2.2. Sampling
Pecorino cheese samples were collected directly from theproducers, stored aseptically in sterile bags and kept in an ice-boxfor transportation to the laboratory. In order to avoid crosscontamination, the external part of the sample was removed withflamedbladespriortosterileexcisionof thepieces employed intheisolation procedure.
2.3. Selective isolation
Approximatelyonegramofeachsamplewasmixedwith9 mlof sterile water and homogenized for 3 min with a lab-blender. Theslurry was diluted and then plated by spread plate technique ontoIM. After growth, colonies were subject to a second isolationprocedure to ensure microbial purity and maintained on YEPDAplates for short term periods at 4
C, whereas rapid freezing at
À
80
C in 17% glycerol was used for long-term storage.
2.4. RAPD-PCR and rDNA sequencing
DNA was extracted as previously described (Bolano et al., 2001;Cardinali et al., 2001). PCR amplification was conducted with PTC-100 Peltier Thermal Cycler (MJ Research Inc. Waltham, Massachu-setts USA). Two RAPD amplifications were done: one with primersM13m (5
0
- GAG GGT GGC GGT TC -3
0
) and Rp 11 (5
0
- GAA ACT CGCCAAG -3
0
), the other with primers RP8 (5
0
- AGATTC TTG GCAC -3
0
)and RP9 (5
0
- CCA AAA GAT CGA C -3
0
). DNA was amplified for 35cycles (denaturation, 94
C 1 min; annealing, 38
C 1 min; exten-sion, 72
C 1 min) followed by a single 15 min extension at 72
C.Amplicons were subject to submarine gel electrophoresis in 1%agarose, ethidium bromide pre-stained gels. Gel images weredigitalized with a Kappa camera (Kappa GmbH – Germany,www.kappa.de). The D1/D2 domain of the 26S rDNA was amplified andsequenced according to the procedure previously described byKurtzman (Kurtzman and Robnett,1998). Analysis of the sequenceswas carried out with BLAST (http://blast.ncbi.nlm.nih.gov/Blast.cgi)and Geneious (http://www.geneious.com).
2.5. FTIR analysis
FTIR measurements of each strain were carried out on threeindependent cell suspensions, using a TENSOR 27 FTIR spectrom-eter (Bruker Optics GmbH, Ettlingen, Germany) reading the rangebetween 4000 and 400cm
À
1
in transmission mode, with 4cm
À
1
spectral resolution. Each spectrum was the average of 64 scansamplings. Preliminary analyses (first derivative and baselinecorrection) were carried out with the software OPUS 6.5 (Bruker).
2.6. Statistical analysis
Statistical analyses were carried out with the free software ‘‘R’’(http://cran.r-project.org/). RAPD banding patterns were trans-formed into binary matrices reporting the presence of the band as‘‘1’’ and its absence as ‘‘0’’ with the ClassMaker software (Cardinaliet al., 2003). Distances between strains were calculated from theRAPD binary matrices with the
dist. binary
function using thecoefficient of Gower & Legendre
d
¼
{1
À
[2a/(2a
þ
b
þ
c)]}
0.5
,implemented in the ADE4 package.Distance matrices from averaged spectral (FTIR) data wereobtained with the basic
dist.
function (Euclidean Distance).Distance matrices from both RAPD and FTIR were processed withthe Principal Coordinate Analysis (PCoA) algorithm (Legendre andLegendre, 1998).
3. Results and discussion
3.1. Isolation and conventional identification
About 50 yeast strains were isolated from Pecorino cheese from10 regions of Italy (Fig. 1). Among all isolates, twenty-two strainsresulted compatible with the
D. hansenii
assimilation profile andwereusedinthispreliminarystudy.Theassimilationprofiles(Table1) indicated that all strains share nine positive and one negativenon-variable characters with the type strain (Nakase et al.,1998). Afewstrains,includingthetypestrain,grewonraffinose,
D
-mannitol,succinate and xylose less vigorously than the others (Table 1).Another five carbon sources (soluble starch, ribose, lactose, rham-nose and citrate), which were not necessarily assimilated by thestrains employed in the species description (Nakase et al., 1998),were also variable for the 22 strains analyzed here. In fact, the firstthree sources were used by the majority of the isolates, whereasmannose and citrate were metabolized only by seven and twostrains, respectively (Table 1). The seven strains isolated in the
M. Del Bove et al. / Food Microbiology 26 (2009) 453–459
454
Norcia and Sellano (30 km distance) surroundings displayed anidentical assimilation pattern to these five variable carbon sour-ces, with the only difference being that the two strains fromSellano assimilate rhamnose, whereas those from Norcia do not.Representatives of each geographic group were subject tomolecular identification by direct sequencing of the D1/D2domain of the DNA coding for the 26S RNA. The 570 bp longsequences showed an absolute identity to the type strain CBS 767(data not shown), confirming the preliminary identification thatassigned these isolates to the species
D. hansenii
. On the otherhand, theseresultsindicatethe lackof molecular variability in theD1/D2 domain of this species as previously found with a prelimi-nary analysis of the sequences deposited in GenBank.
3.2. Molecular analysis
Twenty-two cheese isolates were subject to molecular finger-print by using two double-primer RAPD procedures consisting of two independent amplificationseachwithtwoshort primers.Thefirst pair of primers (RP8 and RP9) produced only four bands of which the three largest were present in 3 strains, respectively LCF552,LCF561andLCF562.Interestingly,thesestrains(LCF552,LCF561 and LCF 562) were isolated in a restricted area about 70 kmwide. The other pair of primers (M13m and RP11) yielded 13different bands, indicating the particular efficiency of the pairM13m plus RP11. The usage of these two couples of primers wasdue to the observation that each double primed RAPD producedmore bands than the sum of the two RAPD amplification witha single primer (results not shown). Banding patterns were clas-sified with the ClassMaker software (Cardinali et al., 2003),obtaining a binary matrix inwhich the presence of the bands wasreported as ‘‘1’’ and the absence as ‘‘0’’. The binary matrix wasprocessed with the
dist. binary
function of the ‘‘ADE4’’ R packageobtaining a triangular distance matrix used as input of the Prin-cipal Coordinate Analysis (PCoA). The eigenvalues of the first twodimensions were plotted obtaining the point scattering reportedinFig. 3. The inspection of this picture indicates that all strains
T a b l e 1
S t r a i n s e m p l o y e d w i t h t h e i r o r i g i n o f i s o l a t i o n a n d a s s i m i l a t i o n p r o fi l e . S t r a i n L o c a l e o f i s o l a t i o n T y p e o f c h e e s e G l u c o s e G a l a c t o s e S u c r o s e M a l t o s e T r e h a l o s e C e l l o b i o s e R a f fi n o s e
D
- M a n n i t o l S u c c i n a t e
D
- X y l o s e I n o s i t o l S t a r c h
D
- R i b o s e
L
- R h a m n o s e L a c t o s e C i t r a t e L C F 3 7 3 P i e m o n t e P e c o r i n o o f B a g n o l o
þ þ þ þ þ þ þ þ þ þ À À þ þ À À
L C F 3 8 7 R o m a P e c o r i n o R o m a n o D O P
þ þ þ þ þ þ þ þ þ
w
À þ þ À þ À
L C F 4 0 9 M a c e r a t a P e c o r i n o D o l c e N e r o
þ þ þ þ þ þ þ þ þ þ À À þ À þ À
L C F 4 1 1 M a c e r a t a P e c o r i n o D o l c e N e r o
þ þ þ þ þ þ þ þ þ
w
À À þ À þ À
L C F 4 1 5 F e r m o P e c o r i n o D o l c e R o s s o
þ þ þ þ þ þ þ þ þ þ À þ þ À þ À
L C F 4 2 0 S e l l a n o P a r t i a l l y a g e d P e c o r i n o
þ þ þ þ þ þ þ þ þ þ À þ þ þ þ À
L C F 4 2 3 S e l l a n o P a r t i a l l y a g e d P e c o r i n o
þ þ þ þ þ þ þ þ þ þ À þ þ þ þ À
L C F 4 4 5 P i e n z a P e c o r i n o T o s c a n o D O P
þ þ þ þ þ þ þ þ
w w
À À þ À þ À
L C F 4 5 4 P i e n z a P e c o r i n o T o s c a n o D O P
þ þ þ þ þ þ þ þ þ þ À þ þ þ þ À
L C F 5 4 6 N o r c i a P e c o r i n o o f N o r c i a
þ þ þ þ þ þ þ þ þ þ À þ þ À þ À
L C F 5 4 7 N o r c i a P e c o r i n o o f N o r c i a
þ þ þ þ þ þ þ þ þ þ À þ þ À þ À
L C F 5 4 9 N o r c i a P e c o r i n o o f N o r c i a
þ þ þ þ þ þ þ þ þ þ À þ þ À þ À
L C F 5 5 0 N o r c i a P e c o r i n o o f N o r c i a
þ þ þ þ þ þ þ þ þ þ À þ þ À þ À
L C F 5 5 1 O r v i e t o P e c o r i n o o f O r v i e t o
þ þ þ þ þ þ þ þ þ
w
À À þ À þ À
L C F 5 5 2 O r v i e t o P e c o r i n o o f O r v i e t o
þ þ þ þ þ þ þ þ þ þ À þ þ À þ À
L C F 5 5 3 N o r c i a P e c o r i n o o f N o r c i a
þ þ þ þ þ þ þ þ þ þ À þ þ À þ À
L C F 5 5 4 L i v o r n o P e c o r i n o T o s c a n o D O P
þ þ þ þ þ þ þ þ þ þ À þ þ þ þ À
L C F 5 5 5 L i v o r n o P e c o r i n o T o s c a n o D O P
þ þ þ þ þ þ þ þ þ þ À þ þ À þ À
L C F 5 5 7 S a r d e g n a P e c o r i n o S a r d o D O P
þ þ þ þ þ þ þ þ þ þ À À þ þ þ þ
L C F 5 5 8 S a r d e g n a P e c o r i n o S a r d o D O P
þ þ þ þ þ þ þ þ þ þ À À þ À þ þ
L C F 5 6 1 P i e n z a P e c o r i n o T o s c a n o D O P
þ þ þ þ þ þ þ þ þ
w
À þ þ À À À
L C F 5 6 2 P i e n z a P e c o r i n o T o s c a n o D O P
þ þ þ þ þ þ þ þ þ þ À þ þ þ þ À
C B S 7 6 7 D e n m a r k S t r a i n T y p e
þ þ þ þ þ þ þ þ þ þ À þ þ
w
þ À þ
, p o s i t i v e g r o w t h ; w , w e a k g r o w t h ;
À
, n o g r o w t h . T h e l a s t fi v e c a r b o n s o u r c e s , r e p o r t e d i n i t a l i c , a r e v a r i a b l e ( v ) a c c o r d i n g t o t h e s p e c i e s d e s c r i p t i o n b y N a k a s e e t a l . ( 1 9 9 8 ) . L C F : i n t e r n a l s t r a i n d e s i g n a t i o n ; C B S : C e n t r a l b u r e a u v o r S c h i m m e l c u l t u r e , U t r e c h t , N L .
Fig. 1.
Yeast isolated from Pecorino cheese in 10 areas of Italy. Grey circles indicateregions of the Pecorino DOP (Designation of Protected Origin) (Pecorino of Bagnolo
Piemonte
, Pecorino Toscano DOP
Pienza
and
Livorno
, Pecorino Dolce Nero
Macerata
,Pecorino Dolce Rosso
Fermo
, Pecorino of
Norcia
, Pecorino semi-stagionato of
Sellano
,Pecorino of
Orvieto
, Pecorino Romano DOP –
Roma
, Pecorino Sardo DOP –
Sassari
)Boldface names indicate the origin of the cheese sampled according to the locationsreported in the figure.
M. Del Bove et al. / Food Microbiology 26 (2009) 453–459
455
of 00
Leave a Comment