Occurrence and characterization of yeast isolated from artisanal Fiore Sardo cheese

Abstract

The occurrence of yeast microflora in artisanal Fiore Sardo cheese during ripening was studied. Mean yeast counts ranged from 2.64+/-1 log(10) cfu ml(-1) in milk to 0.65+/-1 log(10) cfu g(-1) in 9 months cheese, with the higher counts observed in 48-h-old cheese. Strains belonging to the prevalent species Debaryomyces hansenii, Kluyveromyces lactis, Geotrichum candidum, Candida zeylanoides and Candida lambica were selected for technological and genotypic characterization. All D. hansenii strains fermented glucose and assimilated lactate, a high percentage assimilated citrate and only a few showed proteolytic and lipolytic activity. All K. lactis strains were able to both assimilate and ferment lactose, to assimilate lactate and to exhibit proteolytic activity on casein. G. candidum assimilated lactate and some strains showed proteolytic and lipolytic activity. C. zeylanoides showed lipolytic activity on tweens and the majority of strains assimilated citrate. C. lambica fermented glucose and assimilated lactate. Considering their diffusion and technological characteristics, an important role for K. lactis and G. candidum in the early stages of the ripening process and for D. hansenii after the first month of ripening can be suggested. RAPD-PCR analysis with M13 primer grouped the isolates in well-separated clusters with their type strains and confirmed the previous phenotypic identification. The high intraspecific homogeneity observed in tested strains could be explained by their isolation from a common substrate and from neighbouring geographical areas. This preliminary study allowed us to isolate autochthon yeast strains showing particular properties which can contribute to the production of typical cheese taste and flavour.

Full text

Occurrence and characterization of yeasts isolated from artisanal
Fiore Sardo cheese
M.E. Fadda*, V. Mossa, M.B. Pisano, M. Deplano, S. Cosentino
Section of Hygiene, Department of Experimental Biology, University of Cagliari, Cittadella Universitaria,
S.S. 554, 09042, Monserrato (CA), Italy
Received 1 October 2003; received in revised form 21 January 2004; accepted 4 February 2004
Abstract
The occurrence of yeast microflora in artisanal Fiore Sardo cheese during ripening was studied. Mean yeast counts ranged
from 2.64 F1 log
10
cfu ml
1
in milk to 0.65 F1 log
10
cfu g
1
in 9 months cheese, with the higher counts observed in 48-h-old
cheese. Strains belonging to the prevalent species Debaryomyces hansenii,Kluyveromyces lactis,Geotrichum candidum,
Candida zeylanoides and Candida lambica were selected for technological and genotypic characterization. All D. hansenii
strains fermented glucose and assimilated lactate, a high percentage assimilated citrate and only a few showed proteolytic and
lipolytic activity. All K. lactis strains were able to both assimilate and ferment lactose, to assimilate lactate and to exhibit
proteolytic activity on casein. G. candidum assimilated lactate and some strains showed proteolytic and lipolytic activity. C.
zeylanoides showed lipolytic activity on tweens and the majority of strains assimilated citrate. C. lambica fermented glucose
and assimilated lactate. Considering their diffusion and technological characteristics, an important role for K. lactis and G.
candidum in the early stages of the ripening process and for D. hansenii after the first month of ripening can be suggested.
RAPD-PCR analysis with M13 primer grouped the isolates in well-separated clusters with their type strains and confirmed the
previous phenotypic identification. The high intraspecific homogeneity observed in tested strains could be explained by their
isolation from a common substrate and from neighbouring geographical areas. This preliminary study allowed us to isolate
autochthon yeast strains showing particular properties which can contribute to the production of typical cheese taste and flavour.
D2004 Elsevier B.V. All rights reserved.
Keywords: Fiore Sardo cheese; Ewe cheese; Yeasts; RAPD-PCR
1. Introduction
There are numerous references concerning to the
significance of the presence of yeasts in dairy prod-
ucts, where they may contribute positively to the
characteristic taste and flavour development during
the stage of maturation or, on the contrary, may lead to
product spoilage. Traditionally, yeasts are identified
by morphological and physiological criteria (Kreger
van Rij, 1984; Barnett et al., 1990; Kurtzman and Fell,
2000), but, in the last few years, molecular methods
have been developed in order to simplify identifica-
tion. Among such techniques, the random amplified
polymorphic DNA (RAPD) assay has been frequently
used for the identification of yeast species isolated
0168-1605/$ - see front matter D2004 Elsevier B.V. All rights reserved.
doi:10.1016/j.ijfoodmicro.2004.02.001
* Corresponding author. Tel.: +39-70-6754195; fax: +39-70-
6754197.
E-mail address: mefadda@unica.it (M.E. Fadda).
www.elsevier.com/locate/ijfoodmicro
International Journal of Food Microbiology 95 (2004) 51– 59

from dairy products (Romano et al., 1996; Prillinger et
al., 1999; Andrighetto et al., 2000).
Although the occurrence of yeasts in artisanal ewe’s
and goat’s milk cheeses has recently been investigated
(Tornadijo et al., 1998; Carreira et al., 1998; Pereira-
Dias et al., 2000), very little information on yeast
content in Sardinian ewe’s milk cheese is available.
There are old reports on the presence of yeasts in
Pecorino Romano cheese (Deiana et al., 1984, 1994),
and recently, we have studied the occurrence of yeasts
in different types of typical Sardinian ewe’s cheeses
(Cosentino et al., 2001; Fadda et al., 2001).
Fiore Sardo cheese is a hard variety made in
Sardinia from raw whole ewe’s milk at farm house
level using artisanal procedures. Milk is coagulated by
lamb rennet paste without addition of starter cultures,
which means that the ripening process is performed
only by the natural flora present in the milk and in the
environment. The first stage of maturation involves a
slight smoking for 10–15 days by natural means. In a
period of 3–9 months, maturation is complete. This
cheese is produced under EU regulations and has been
awarded PDO (Reg. CEE N. 1263/96).
The aim of this work was to identify the yeast
species present throughout the ripening process of this
artisanal ewe’s cheese and to characterize them both
technologically and genotypically.
2. Materials and methods
2.1. Cheese samples
Thirteen batches of Fiore Sardo cheese were
obtained from 12 farms located in different areas of
Sardinia. The cheeses were elaborated by experienced
cheesemakers according to the traditional artisanal
methods. From each batch, milk and 48-h, 1-, 3-, 6-
and 9-month-old cheese samples were taken. Each
cheese sample was represented by one whole cheese.
Samples were transported to the laboratory under
refrigeration and analyzed on arrival.
2.2. Isolation and identification of strains
Ten grams of product was taken from the interior
of the cheeses, diluted in 90 ml of sterile solution of
2% (w/v) sodium citrate (Sigma, St. Louis, MO, USA)
and homogenized in a Stomacher (PBI, Milan Italy)
for 30 s at normal speed. For all samples, decimal
dilutions were prepared in sterile solution of 2% (w/v)
sodium citrate and numbers of yeasts were determined
by surface plating on potato dextrose agar (PDA)
(Microbiol, Cagliari, Italy) with chloramphenicol
(0.01%) (Microbiol) after incubation at 25 jC for 5
days. All samples were prepared and analyzed in
duplicate. Yeast colonies grown on PDA were sorted
on the basis of their morphology (smoothness of
surface, regularity of border, consistency, colour,
etc.), streaked to single colonies on YPDA [1% yeast
extract (Microbiol), 2% dextrose (Sigma), 2% peptone
(Microbiol), 1.5% agar (Microbiol)], incubated for 5
days at 25 jC and checked for purity. Counts for each
individual type of colony were made in order to
estimate the relative occurrence of the various yeasts
present in the samples. Yeast species counts were
calculated as number of colony forming units per
gram of sample and are reported as log
10
cfu g
1
.
The isolates were identified using the tests sug-
gested and described by Kurtzman and Fell (2000).
When necessary, test results were checked using the
API ID 32C (bioMerieux, Rome, Italy).
The strains belonging to the prevalent yeast species
were analyzed both technologically and genotypically.
Type strains of Debaryomyces hansenii var. han-
senii DBVPG 6050, Kluyveromyces lactis DBVPG
6305, Pichia fermentans/Candida lambica DBVPG
4363, Candida zeylanoides DBVPG 6163 and Geo-
trichum candidum DBVPG 4350 were obtained from
the Industrial Yeast Collection (Dipartimento di Biol-
ogia Vegetale, Perugia, Italy).
2.3. Study of some properties of technological interest
The assimilation of lactic acid and citric acid was
carried out according to Disegna et al. (1997).A
modification of the method described by Koburger
(1972) for determination of lipolytic and proteolytic
activity was used. The basic medium was plate count
agar (PCA) (Microbiol). To test lipolytic activity,
0.01% calcium chloride (Sigma) and 0.5% of Tween
40 (Sigma) or Tween 80 (Sigma) were added. Lipo-
lytic yeast colonies form a halo of precipitated calci-
um palmitate or oleate and thus can be distinguished
from nonlipolytic yeasts. Tests for proteolytic activity
were carried out using the same basic medium with
M.E. Fadda et al. / International Journal of Food Microbiology 95 (2004) 51–5952

addition of 50% of skim milk (Oxoid, Basingstoke,
UK). A halo of clarity around the colony shows
proteolytic activity.
2.4. RAPD-PCR analysis
Genomic DNA was extracted with the method of
Sampaio et al. (2001) with some modifications, as
described below. A loopful of YPDA grown cultures
was suspended in 500-Al lysing buffer (50 mM Tris-
base l
1
(Sigma); 250 mM NaCl l
1
(Sigma); 50 mM
EDTA l
1
(Sigma); 0.3%, w/v, SDS (Sigma); pH 8)
and the equivalent to a volume of 200 Al of 425 – 600
Am glass beads (Sigma) was added. After vortexing
for 1 min, 45 Al of Proteinase K solution [10 mg ml
1
(Sigma)] was added and the tube was incubated for 1
hat65jC. The suspension was then centrifuged for
30 min at 4 jC. Nucleic acid was precipitated with 1/
10 volume of 5 M potassium acetate (Sigma) and two
volumes of absolute ethanol (Sigma) for 24 h at 20
jC. Finally, DNA was washed with ethanol (70% v/v),
dried and resuspended in 50 Al of TE [100 mM Tris/
HCl l
1
(Sigma), pH 8, 100 mM EDTA l
1
].
The core sequence of the phage M13 (5V-
GAGGGTGGCGGTTCT-3V)(Huey and Hall, 1989)
was used as a single primer. The RAPD-PCR anal-
ysis was performed in volumes of 50 Al containing
20 ng primer, 0.2 mM each of dATP, dGTP, dCTP
and dTTP (Eppendorf, Hamburg, Germany), 90 ng
of DNA, 1 U Taq-DNA polymerase (Eppendorf), 5 
Taq master (Eppendorf) and 10 Taq buffer (Eppen-
dorf). Amplification was performed in a Mastercy-
cler gradient 5331 (Eppendorf) at these conditions:
initial denaturing at 95 jC for 5 min; 40 cycles of
denaturation at 93 jC for 45 s, annealing at 50 jC
for 1 min and extension at 72 jC for 1 min; a final
extensionstepof6minfor72jC. A negative
control in which DNA was replaced by sterile
distilled water was also included.
Aliquots of 10 Al of amplified products were ana-
lyzed by electrophoresis on 2% agarose gel (Sigma) in
1TBE buffer [89 mM Tris-base, 89 mM boric acid
(Sigma), 2 mM EDTA] with 0.5 Agml
1
ethidium
bromide (Sigma) at 70 V for 85 min. Sizes were
estimated by comparison against a DNA length stan-
dard (PCR Ladder, Sigma). DNA fragments were
Fig. 1. Evolution of yeast counts (mean log
10
cfu g
1
FS.D.) in milk and in cheese during the ripening of Fiore Sardo.
M.E. Fadda et al. / International Journal of Food Microbiology 95 (2004) 51–59 53

visualized by transillumination and photographed by a
Polaroid system. Pictures were scanned using Jasc
Paint shop Pro 7 software. DNA banding patterns were
analyzed using the Gel Compar software package,
version 2.5 (Applied Maths, Kortrijk). Similarities
among isolates were estimated using the Pearson co-
efficient and clustering was based on the UPGMA
method.
The reproducibility of the RAPD fingerprints was
assessed by comparing the PCR products obtained
with primer M13 and DNA prepared from three
separate cultures of the same strain. Ten strains were
studied, and the patterns for the same strain were
>90% similar, indicating that the reproducibility of the
technique under the conditions used was high (data
not shown).
3. Results
Mean log
10
yeast counts in Fiore Sardo cheese
increased less than 1 unit during the first 48 h of
Table 1
Distribution of yeast species in 13 batches of Fiore Sardo cheese (numbers indicate the percentage of samples in which the species were present)
Milk Ripening time
48 h 1 month 3 months 6 months 9 months
Debaryomyces hansenii 23.1 38.4 46.2 30.8 15.4 –
Kluyveromyces lactis 30.8 53.8 23.1 15.4 – –
Geotrichum candidum 30.8 61.5 15.4 15.4 – –
Candida zeylanoides 7.7 38.4 15.4 7.7 – –
Candida lambica 15.4 15.4 23.1 – – –
Cryptococcus curvatus 38.4 15.4 – – – –
Trichosporon cutaneum 30.8 15.4 – – – –
Rhodotorula rubra 30.8 15.4 – – – –
Kluyveromyces marxianus – 23.1 – – – –
Torulaspora delbrueckii – 7.7 7.7 – – –
Saccharomyces cerevisiae – 7.7 7.7 – – –
Zygosaccharomyces bisporus – 7.7 7.7 – – –
Candida lipolytica – 7.7 – – – –
Candida rugosa – 7.7 7.7 – – –
Moniliella suaveolens – – 15.4 15.4 84.6 46.2
Pichia etchelsii – – – – 7.7 –
Candida magnoliae – – – – 7.7 7.7
Candida lusitaniae – – 7.7 – – –
Table 2
Biochemical characteristics of technological interest for the prevalent yeasts isolated from milk and cheese samples (% of positive strains)
Number of strains Debaryomyces hansenii Kluyveromyces lactis Geotrichum candidum Candida zeylanoides Candida lambica
19 14 10 10 9
Fermentation of:
Glucose 100 100 0 0 100
Galactose 0 100 0 0 0
Lactose 0 100 0 0 0
Assimilation of:
Lactose 52.6 100 0 0 0
Lactic acid 100 100 100 20 100
Citric acid 89.5 0 10 90 0
Hydrolysis of:
Casein 10.5 100 10 30 0
Tween 40 68.4 0 10 100 0
Tween 80 15.8 0 10 70 0
M.E. Fadda et al. / International Journal of Food Microbiology 95 (2004) 51–5954

Fig. 2. Cluster analysis of RAPD-PCR profiles generated with M13 primer of the strains belonging to the five prevalent species and the
respective type strains.
M.E. Fadda et al. / International Journal of Food Microbiology 95 (2004) 51–59 55

production with respect to the initial number in the
milk (from 2.64 F1 to 3.06 F0.9 log
10
cfu g
1
), then
they gradually decreased to a level of 0.65 F1 log
10
cfu g
1
at 9 months of ripening (Fig. 1).
Table 1 shows the distribution of the yeast species in
milk and during the ripening of Fiore Sardo cheese. The
presence of the species D. hansenii,K. lactis,G.
candidum,C. zeylanoides and C. lambica increased
from milk to 48-h-old cheese, whereas Cryptococcus
curvatus,Trichosporon cutaneum and Rhodotorula
rubra were not isolated in cheese after 48 h of ripening.
From a qualitative point of view, the most numer-
ous yeast species were detected in 48-h-old cheese. D.
hansenii was present until 6 months of ripening but
was prevalent at 1 month. K. lactis,G. candidum and
C. zeylanoides occurred until 3 months of ripening but
predominated at 48 h. Other species were only spo-
radically found, except Moniliella suaveolens that
occurred starting from 1 month to the end of ripening
and was present in 84.6% of samples at 6 months.
As can be see from Table 2, which reports the
biochemical characteristics of technological interest
for the five dominant yeast species, only the strains of
K.lactis were able to both assimilate and ferment
lactose. All the strains of D.hansenii,K.lactis and C.
lambica fermented glucose, while the other species
did not have fermentation activity. The assimilation of
lactate was observed in the majority of the species,
whereas the assimilation of citrate was a characteristic
of D.hansenii,C. zeylanoides and of a few strains of
G.candidum. Only K.lactis and few strains of C.
zeylanoides,D.hansenii and G.candidum exhibited
proteolytic activity on casein agar. C.zeylanoides and
some strains of D.hansenii and G. candidum showed
lipolytic reactions on both tweens agar.
The RAPD profiles generated with primer M13 for
the 62 strains belonging to the five prevalent species
and for the respective type strains produced the
UPGMA dendrogram shown in Fig. 2. As can be
seen, the isolates were grouped in well-separated
clusters with their type strains confirming the pheno-
typic identification for all the strains. At a similarity
level of 50%, five clusters were identified.
C. zeylanoides strains, included in cluster 1, were
divided into three subclusters, each containing strains
that were at least 80% similar. The majority of strains
were at least 67.8% similar to the reference strain
DBVPG 6163.
D. hansenii strains (cluster 2) had very similar
patterns and almost all of them were 73% similar to
the reference strain DBVPG 6050; only FS19DH2
strain had a lower similarity (67%).
The percentage similarity among K. lactis strains
(cluster 3) was 53%, but most of them were 73%
similar to the reference strain DBVPG 6350. At a
similarity level of 80%, four subclusters were
distinguished.
Cluster 4 contained C. lambica strains. In this
cluster, only two strains were 57% similar to the
reference strain DBVPG 4363 while all the others
were 73%. Three subclusters were observed at 80%
similarity.
Finally, three subclusters at least 80% similar were
distinguished in cluster 5, which grouped G. candi-
dum strains. The majority of the strains were differ-
entiated at a similarity level of 63% from the reference
strain DBVPG 4350.
4. Discussion
In our study, the mean yeast count of 2.64 log
10
cfu
ml
1
recorded in the raw milk was lower than that
reported by Tornadijo et al. (1998) in raw goat’s milk
for Armada cheese production. Even if few studies
have been carried out on the occurrence of yeasts in
raw ovine’s milk, it is well documented that yeasts
occur in raw milk at insignificant numbers (Fleet,
1990; Jakobsen and Narvhus, 1996; Roostita and
Fleet, 1996) probably due to competitive utilization
for the growth substrates by psychrotrophic bacteria
of milk or to inhibition by metabolites excreted by
bacteria (Viljoen, 2001).
Yeast counts recorded in the first steps of ripening
of Fiore Sardo were lower than those reported by
other authors who have studied the microbiology of
ovine cheeses made from raw milk without addition
of starter cultures. In fact, counts of 10
6
cfu g
1
in 1-
month-old cheese were reported by Tornadijo et al.
(1998) in Spanish Armada cheese; mean yeast counts
per gram of cheese ranging from 2.7 to 6.4 log
10
cfu
g
1
were observed in an artisanal Portuguese ewes’
cheese (Pereira-Dias et al., 2000); yeast counts from
10
3
in curd to 10
5
cfu g
1
in 1-month-old cheese
were shown by Hatzikamari et al. (1999) in Greek
Anevato cheese. The lower values of yeast counts
M.E. Fadda et al. / International Journal of Food Microbiology 95 (2004) 51–5956

recorded in Fiore Sardo cheese could be explained
not only by the low starting counts in milk, but also
by the faster decreasing of a
w
values due to the
smoking process and by the high NaCl values typical
of this product.
R. rubra,C. curvatus and T. cutaneum, three of the
dominating species in milk, were found until 48 h of
cheese ripening time. These species belong to the
family Basidiomycetaceae and show urease activity.
The presence of strain urease positive in milk and in
cheese at the beginning of the ripening period is
reported by many authors. Early studies, reviewed
by Walker and Ayres (1970), suggest the frequent
occurrence of pigmented yeasts of the genus Rhodo-
torula in raw milk and a recent study confirmed this
finding (Pereira-Dias et al., 2000).Van den Tempel
and Jakobsen (1998) found T. cutaneum in cow’s milk
for Danablu cheese manufacture and Corbo et al.
(2001) isolated this species from milk of different
animal origin. Moreover, T. cutaneum is a synonym of
T. beigelii (Barnett et al., 1990) which was found by
Tornadijo et al. (1998) in goat’s milk. Nahabieh and
Schmidt (1990) and Pereira-Dias et al. (2000) also
isolated the species Candida curvata, synonym of C.
curvatus and probably related to the genus Tricho-
sporon (Kurtzman and Fell, 2000), from French goat’s
cheese and Portuguese ewes’ cheese, respectively.
Since the importance of Trichosporon spp. in human
and animal infections is documented (Sugita et al.,
1999; Krukowski et al., 2001), an environmental,
human or animal contamination may be supposed
for these species.
In the present study, the predominance of K. lactis
and G. candidum in 48-h-old cheese samples suggests
an important role in the first steps of ripening of Fiore
Sardo for these species. Because the ability of K.
lactis to ferment and assimilate lactose is considered
to be one of the key properties contributing to its
growth in cheeses and dairy products (Fleet, 1990),
this species probably participates, together with the
lactic acid bacteria, to the initial acidification of the
curd which takes place in the early stages of the
ripening process.
Similarly, G. candidum has long been the subject
of biochemical and physiological studies due to its
biotechnological interest. It plays an important role in
competition with undesirable contaminants in the
cheese (Nielsen et al., 1998; Dieuleveux et al.,
1998). Moreover, its lipases and proteases release
fatty acids and peptides that contribute to the devel-
opment of distinctive flavour in cheese (Litthauer et
al., 1996; Holmquist, 1998; Tornadijo et al., 1998).
Recently, the contribution of K. lactis and G. candi-
dum strains to the formation of sulphur aroma com-
pounds in cheese (Arfietal.,2002)has been
indicated.
It is widely recognised that D. hansenii contributes
in cheese ripening by assimilation of lactic acid
causing an increase in the pH, which enables the
growth of microorganisms such as lactic acid bacteria.
In addition, this species has the capability to metab-
olise lactose, to assimilate citric acid and to generate
alcoholic, acidic and cheesy flavour (Welthagen and
Viljoen, 1998; Van den Tempel and Jakobsen, 2000).
The overall predominance of D. hansenii in the
present study confirmed that this species is an impor-
tant component of the microflora of Sardinian ewe’s
cheeses, in agreement with results reported by Deiana
et al. (1994) on ‘‘Pecorino Romano’’ and by Cosentino
et al. (2001) on samples of ‘‘Pecorino,’’ ‘‘Caciotta,’’
‘‘Feta’’ and ‘‘Ricotta’’ made in Sardinia. In this inves-
tigation, all D. hansenii strains assimilated lactate and
most (89.5%) assimilated citrate. Furthermore, a small
proportion of strains showed proteolytic and lipolytic
activity. Absence of these activities in this species has
been reported by Roostita and Fleet (1996),while
Welthagen and Viljoen (1998) and Van den Tempel
and Jakobsen (2000) observed extracellular lipolytic
activity on Tributyrin and no proteolytic activity.
Recently, intracellular proteolytic activity was found
in D. hansenii strains isolated from cheese by Klein
et al. (2002) and Bintsis et al. (2003). Considering the
diffusion of D. hansenii in Fiore Sardo cheese
together with its biochemical characteristics, we
may suppose an important role played by this species
in the ripening process, especially after the first
month of maturation.
C. zeylanoides has been isolated from spoiled
cottage cheese by Brocklehurst and Lund (1985) and
from milk of different animal origin by Corbo et al.
(2001). In an artisanal ewes’ cheese, Pereira-Dias et
al. (2000) referred that the decrement of C. zeyla-
noides levels after 30 days of maturation may be
related with their restricted fermentative ability that
limits the growth under the semianaerobic or anaero-
bic conditions in the cheese body. C. lambica, the
M.E. Fadda et al. / International Journal of Food Microbiology 95 (2004) 51–59 57

anamorph form of P. fermentans, was found in fer-
mented milk products (Rohm et al., 1992) and was
also the predominant species in milk and in 4- and 8-
week-old Armada cheese (Tornadijo et al., 1998).
Even if C. zeylanoides and C. lambica have been
observed in milk and in Fiore Sardo cheese until 3 and
1 month of ripening, respectively, these species prob-
ably represent casual contaminants, being present in a
quite negligible percentage of samples.
The remaining yeast species do not seem to take
part in the ripening process of Fiore Sardo cheese. Of
a certain interest appears the presence of M. suaveo-
lens in the majority of cheese samples at 6 and 9
months of ripening. Kurtzman and Fell (2000)
reported the presence of this fungus in cheese, but
further studies are needed to understand its role in
Fiore Sardo.
Random amplified polymorphic DNA (RAPD)
analysis is one of the more recent molecular typing
techniques. In a preliminary study, Lieckfeldt et al.
(1992) showed that RAPD could be used to charac-
terize fungal species. This method has been success-
fully used by Baleiras Couto et al. (1994) for
discriminating between Saccharomyces cerevisiae
and members of the Zygosaccharomyces genus. Re-
cently, this approach for the typing of moulds and
yeasts in dairy products has been described by many
authors (Prillinger et al., 1999; Andrighetto et al.,
2000; Psomas et al., 2001; Kure et al., 2002).
In our study, classification of the tested strains by
the use of RAPD agreed with the previous phenotypic
identification. This suggests this method as an alter-
native to the conventional approaches to yeast iden-
tification since it is a technique that is rapid, easy to
perform and rather inexpensive.
Identical profiles (>90% similarity) were found not
only for strains isolated from the same samples,
suggesting the presence of multiple isolates of the
same strain, but also from strains isolated from
different batches. The high intraspecific homogeneity
observed in tested strains could be due to their
isolation from a common substrate and from neigh-
bouring geographical areas. It can also be noted that
identical profiles were shown by some type strains
and isolates of the same species. These results suggest
that in order to establish if two isolates of the same
species have a common origin, other methods should
be used together with M13 fingerprinting.
In conclusion, this preliminary study allowed us to
isolate autochthon yeast strains showing particular
properties which can contribute to the production of
typical cheese taste and flavour. Some of this strains
are currently being used, together with lactic acid
bacteria, in experimental trials as natural starter for
Fiore Sardo cheesemaking. These trials will be useful
to evaluate the effective contribution of predominant
yeast strains to aroma and flavour of this typical
cheese.
Acknowledgements
This work was supported by a grant from MURST,
Plan ‘‘Agroalimentary products: dairy products’’,
Cluster 08B, Project no. 7.
References
Andrighetto, C., Psomas, E., Tzanetakis, N., Suzzi, G., Lombardi,
A., 2000. Randomly amplified polymorphic DNA (RAPD) PCR
for the identification of yeasts isolated from dairy products.
Letters in Applied Microbiology 30, 5– 9.
Arfi, K., Spinnler, H.E., Tache, R., Bonnarme, P., 2002. Production
of volatile compounds by cheese-ripening yeasts: requirement
for a methanethiol donor for S-methyl thioacetate synthesis by
Kluyveromyces lactis. Applied Microbiology and Biotechnology
58, 503– 510.
Baleiras Couto, M.M., Van der Vossen, J.M.B.M., Hofstra, H., Huis
in ’t Veld, J.H.J., 1994. RAPD analysis: a rapid technique for
differentiation of spoilage yeasts. International Journal of Food
Microbiology 24, 249 – 260.
Barnett, J.A., Payne, R.W., Yarrow, D., 1990. Yeasts: Character-
istics and Identification, 2nd ed. Cambridge Univ. Press,
Cambridge.
Bintsis, T., Vafopoulou-Mastrojiannaki, A., Litopoulou-Tzanetaki,
E., Robinson, R.K., 2003. Protease, peptidase and esterase ac-
tivities by lactobacilli and yeast isolates from Feta cheese brine.
Journal of Applied Microbiology 95, 68– 77.
Brocklehurst, F., Lund, B.M., 1985. Microbiological changes in
cottage cheese varieties during storage at + 7 jC. Food Micro-
biology 2, 207 – 233.
Carreira, A., Paloma, L., Loureiro, V., 1998. Pigment producing
yeasts involved in a brown surface discoloration of ewes’
cheese. International Journal of Food Microbiology 41,
223 – 230.
Corbo, M.R., Lanciotti, R., Albenzio, M., Sinigaglia, M., 2001.
Occurrence and characterization of yeasts isolated from milks
and dairy products of Apulia region. International Journal of
Food Microbiology 69, 147 – 152.
Cosentino, S., Fadda, M.E., Deplano, M., Mulargia, A.F., Palmas, F.,
M.E. Fadda et al. / International Journal of Food Microbiology 95 (2004) 51–5958

2001. Yeasts associated with Sardinian ewe’s dairy products. In-
ternational Journal of Food Microbiology 69, 53– 58.
Deiana, P., Fatichenti, F., Farris, G.A., Mocquot, G., Lodi, R.,
Todesco, R., Cecchi, L., 1984. Metabolization of lactic and
acetic acids in Pecorino Romano cheese made with a combined
starter of lactic acid bacteria and yeast. Lait 64, 380– 394.
Deiana, P., Caredda, M., Rossi, J., Gobbetti, M., 1994. Secondary
microflora in Pecorino Romano cheese. XXIV International
Dairy Congress, 23.
Dieuleveux, V., Van Der Pyl, D., Chataud, J., Gueguen, M., 1998.
Purification and characterization of anti-Listeria compounds
produced by Geotrichum candidum. Applied Environmental
Microbiology 64, 800 – 803.
Disegna, L., Rudello, F., Spolaor, D., 1997. Selezione di lieviti da
usare nella tecnologia dell’Asiago pressato. Latte 3, 142– 143.
Fadda, M.E., Cosentino, S., Deplano, M., Palmas, F., 2001. Yeast
populations in Sardinian Feta cheese. International Journal of
Food Microbiology 69, 153 – 156.
Fleet, G.H., 1990. Yeasts in dairy products. Journal of Applied
Bacteriology 68, 199 – 211.
Hatzikamari, M., Litopoulou-Tzanetaki, E., Tzanetakis, N., 1999.
Microbiological characteristics of Anevato: a traditional Greek
cheese. Journal of Applied Microbiology 87, 595 – 601.
Holmquist, M., 1998. Insights into the molecular basis for fatty acyl
specificities of lipases from Geotrichum candidum and Candida
rugosa. Chemistry and Physics of Lipids 93, 57– 65.
Huey, B., Hall, J., 1989. Hypervariable DNA fingerprinting in
Escherichia coli. Minisatellite probe from bacteriophage M13.
Journal of Bacteriology 171, 2528 – 2532.
Jakobsen, M., Narvhus, J., 1996. Yeasts and their possible benefi-
cial and negative effects on the quality of dairy products. Inter-
national Dairy Journal 6, 755 – 768.
Klein, N., Zourari, A., Lortal, S., 2002. Peptidase activity of four
yeast species frequently encountered in dairy products and com-
parison with several dairy bacteria. International Dairy Journal
12, 853– 861.
Koburger, J.A., 1972. Fungi in foods: III. The enumeration of lipo-
lytic and proteolytic organisms. Journal of Milk and Food Tech-
nology 35, 117 – 118.
Kreger van Rij, N.J.W., 1984. The Yeasts: A Taxonomic Study.
Elsevier, Amsterdam.
Krukowski, H., Tietze, M., Majewski, T., Rozanski, P., 2001. Sur-
vey of yeast mastitis in dairy herds of small-type farms in the
Lublin region, Poland. Mycopathologia 150, 5 – 7.
Kure, C.F., Abeln, E.C.A., Holst-Jensen, A., Skaar, I., 2002. Differ-
entiation of Penicillium commune and Penicillium palitans iso-
lates from cheese and indoor environments of cheese factories
using M13 fingerprinting. Food Microbiology 19, 151 –157.
Kurtzman, C.P., Fell, J.W., 2000. The Yeasts: A Taxonomic Study.
Elsevier, Amsterdam.
Lieckfeldt, E., Meyer, W., Bo¨rner, T., 1992. Rapid identification
and differentiation of yeasts by DNA and PCR fingerprinting.
Journal of Basic Microbiology 33, 413– 426.
Litthauer, D., Louw, C.H., Du Toit, P., 1996. Geotrichum candidum
P-5 produces an intracellular serine protease resembling chymo-
trypsin. The International Journal of Biochemistry and Cell Bi-
ology 28, 1123 – 1130.
Nahabieh, P., Schmidt, J., 1990. Contribution a l’etude de la flore
levure de quelques grands types de fromages de chevre. Lait 70,
325 – 343.
Nielsen, M.S., Frisvad, J.C., Nielsen, P.V., 1998. Colony interaction
and secondary metabolite production of cheese-related fungi in
dual culture. Journal of Food Protection 61, 1023– 1029.
Pereira-Dias, S., Potes, M.E., Marinho, A., Malfeito-Ferreira, M.,
Loureiro, V., 2000. Characterisation of yeast flora isolated from
an artisanal Portuguese ewes’ cheese. International Journal of
Food Microbiology 60, 55 – 63.
Prillinger, H., Molnar, O., Eliskases-Lechner, F., Lopandic, K.,
1999. Phenotypic and genotypic identification of yeasts from
cheese. Antonie van Leeuwenhoek 75, 267 – 283.
Psomas, E., Andrighetto, C., Litopoulou-Tzanetaki, E., Lombardi,
A., Tzanetakis, N., 2001. Some probiotic properties of yeast
isolates from infant faeces and Feta cheese. International Jour-
nal of Food Microbiology 69, 125– 133.
Rohm, H., Eliskases-Lechner, F., Bra¨ uer, M., 1992. Diversity of
yeasts in selected dairy products. Journal of Applied Bacteriol-
ogy 72, 370 – 376.
Romano, A., Casaregola, S., Torre, P., Gaillardin, C., 1996. Use
of RAPD and mitochondrial DNA RFLP for typing of Can-
dida zeylanoides and Debaryomyces hansin yeast strains iso-
lated from cheese. Systematic and Applied Microbiology 19,
255 – 264.
Roostita, R., Fleet, G.H., 1996. Growth of yeasts in milk and asso-
ciated changes to milk composition. International Journal of
Food Microbiology 31, 205 – 219.
Sampaio, J.P., Gadanho, M., Santos, S., Duarte, F.L., Pais, C.,
Fonseca, A., Fell, J.W., 2001. Polyphasic taxonomy of the ba-
sidiomycetous yeast genus Rhodosporidium:Rhodosporidium
kratochvilovae and related anamorphic species. International
Journal of Systematic and Evolutionary Microbiology 51,
687 – 697.
Sugita, T., Nishikawa, A., Ikeda, R., Shinoda, T., 1999. Identifica-
tion of medically relevant Trich osporo n species based on
sequences of Internal Transcribed Spacer Regions and construc-
tion of a database for Trichosporon identification. Journal of
Clinical Microbiology 37, 1985 – 1993.
Tornadijo, M.E., Fresno, J.M., Sarmiento, R.M., Carballo, J., 1998.
Study of the yeasts during the ripening process of Armada
cheeses from raw goat’s milk. Lait 78, 647 – 659.
Van den Tempel, T., Jakobsen, M., 1998. Yeasts associated with
Danablu. International Dairy Journal 8, 25 – 31.
Van den Tempel, T., Jakobsen, M., 2000. The technological char-
acteristics of Debaryomyces hansenii and Yarrowia lipolytica
and their potential as starters cultures for production of Danablu.
International Dairy Journal 10, 263 – 270.
Viljoen, B.C., 2001. The interaction between yeasts and bacteria in
dairy environments. International Journal of Food Microbiology
69, 37– 44.
Walker, H.W., Ayres, J.C., 1970. Yeasts as spoilage organisms. In:
Rose, A.H., Harrison, J.S. (Eds.), Yeast Technology. The Yeasts,
vol. 3. Academic Press, London, pp. 464 – 527.
Welthagen, J.J., Viljoen, B.C., 1998. Yeast profile in Gouda cheese
during processing and ripening. International Journal of Food
Microbiology 41, 185 – 194.
M.E. Fadda et al. / International Journal of Food Microbiology 95 (2004) 51–59 59