651

Surface microbial consortia from Livarot, a French

smear-ripened cheese
Sandra Larpin-Laborde, Muhammad Imran, Catherine Bonaïti, Nagamani Bora,
Roberto Gelsomino, Stefanie Goerges, Françoise Irlinger, Michael Goodfellow,
Alan C. Ward, Marc Vancanneyt, Jean Swings, Siegfried Scherer,
Can. J. Microbiol. Downloaded from www.nrcresearchpress.com by YALE MEDICAL LIBRARY on 12/26/14

Micheline Guéguen, and Nathalie Desmasures

Abstract: The surface microflora (902 isolates) of Livarot cheeses from three dairies was investigated during ripening.
Yeasts were mainly identified by Fourier transform infrared spectroscopy. Geotrichum candidum was the dominating yeast
among 10 species. Bacteria were identified using Biotype 100 strips, dereplicated by repetitive extragenic palindromic PCR
(rep-PCR); 156 representative strains were identified by either BOX-PCR or (GTG)5-PCR, and when appropriate by 16S
rDNA sequencing and SDS–PAGE analysis. Gram-positive bacteria accounted for 65% of the isolates and were mainly as-
signed to the genera Arthrobacter, Brevibacterium, Corynebacterium, and Staphylococcus. New taxa related to the genera
Agrococcus and Leucobacter were found. Yeast and Gram-positive bacteria strains deliberately added as smearing agents
were sometimes undetected during ripening. Thirty-two percent of the isolates were Gram-negative bacteria, which showed
a high level of diversity and mainly included members of the genera Alcaligenes, Hafnia, Proteus, Pseudomonas, and Psy-
chrobacter. Whatever the milk used (pasteurized or unpasteurized), similar levels of biodiversity were observed in the three
dairies, all of which had efficient cleaning procedures and good manufacturing practices. It appears that some of the Gram-
negative bacteria identified should now be regarded as potentially useful in some cheese technologies. The assessment of
their positive versus negative role should be objectively examined.
For personal use only.

Key words: smear cheese, bacteria, yeast, diversity, dynamics.

Résumé : La microflore de surface (902 isolats) de Livarots provenant de trois fromageries a été examinée au cours de l’af-
finage. Les levures ont été identifiées principalement par spectroscopie infrarouge à transformée de Fourier. Geotrichum
candidum était la levure dominante parmi 10 espèces. Les bactéries ont été identifiées par Biotype 100, dérépliquées par
rep-PCR et 156 souches représentatives ont été identifiées par BOX-PCR ou (GTG)5-PCR, et éventuellement par séquençage
de l’ADNr 16S et SDS–PAGE. Les bactéries à Gram positif représentaient 65 % des isolats et appartenaient principalement
aux genres Arthrobacter, Brevibacterium, Corynebacterium et Staphylococcus. De nouveaux taxons associés aux genres
Agrococcus et Leucobacter ont été mis en évidence. Les souches de levures et de bactéries à Gram positif ajoutées en tant
que levains de maturation n’étaient pas toujours détectables au cours de l’affinage. Trente deux pour cent des isolats bacté-
riens étaient à Gram négatif. Ils révélaient un haut niveau de diversité et incluaient principalement des membres des genres
Alcaligenes, Hafnia, Proteus, Pseudomonas et Psychrobacter. Quel que soit le lait utilisé (pasteurisé ou non), des niveaux
similaires de diversité ont été observés dans les trois ateliers, qui avaient tous des procédures de nettoyage efficaces ainsi
que de bonnes pratiques de fabrication. Certaines bactéries à Gram négatif pourraient avoir un rôle positif dans certaines
technologies fromagères. L’examen de leurs bénéfices et risques devrait être réalisé objectivement.
Mots‐clés : fromage à croûte lavée, bactéries, levures, diversité, dynamiques.

Received 27 July 2010. Revision received 15 April 2011. Accepted 19 April 2011. Published at www.nrcresearchpress.com/cjm on
4 August 2011.
S. Larpin-Laborde,* M. Imran, M. Guéguen, and N. Desmasures. Université de Caen Basse-Normandie, Unité des microorganismes
d’intérêt laitier et alimentaire, E.A. 3213, IFR 146 ICORE, 14032 Caen CEDEX, France.
C. Bonaïti and F. Irlinger. Laboratoire de génie et de microbiologie des procédés alimentaires, INRA, 78850 Thiverval-Grignon, France.
N. Bora†, M. Goodfellow, and A.C. Ward. School of Biology, University of Newcastle, Newcastle upon Tyne NE1 7RU, UK.
R. Gelsomino, M. Vancanneyt, and J. Swings. BCCM/LMG Bacteria Collection, Universiteit Gent, 9000 Gent, Belgium.
S. Goerges‡ and S. Scherer. Abteilung Mikrobiologie, Zentralinstitut für Ernhrungs und Lebensmittelforshung, 85350 Freising, Germany.
Corresponding author: Nathalie Desmasures (e-mail: [email protected]).
*Present address: Bioprocess division, Millipore Corporation, 67120 Molsheim, France.
†Present address: Life and Health Sciences, Aston University, Aston Triangle, Birmingham, B4 7ET, UK.
‡Present address: Naturkost Ernst Weber GmbH, 81371 München, Germany.

Can. J. Microbiol. 57: 651–660 (2011) doi:10.1139/W11-050 Published by NRC Research Press
 652 Can. J. Microbiol. Vol. 57, 2011

Introduction cies, to our knowledge no data focus on the “whole” surface

microflora of Livarot cheese and its evolution during ripen-
Smear cheeses are so called because their surfaces are ing.
washed during ripening, giving rise to a glistening rind.
In this study, which was part of the European project
Among them, red smear cheeses are characterized by the de-
QLK1-CT-2001-02228, we made an inventory of the bacte-
velopment of an orangish red colour on the surface. Livarot
rial and yeast floras on the surface of Livarot cheese during
cheese is a traditional French red smear cheese with a Pro-
ripening. The objective was to determine the microbial
tected Designation of Origin. Milk production and cheese
groups dominating in Livarot at various stages of ripening,
making and ripening (minimum 3 weeks) are exclusively
and hence, having a putative role in cheese technology.
done in a restricted area in Normandy. After standardization
Can. J. Microbiol. Downloaded from www.nrcresearchpress.com by YALE MEDICAL LIBRARY on 12/26/14

(to a 25–30 g/L concentration of fat) and addition of starter,

cow’s milk is ripened at 8–10 °C over 12–15 h and then Materials and methods
heated to 36 °C to be rennet-coagulated. After it is cut into Cheese samples
small pieces and slowly mixed, the curd is transferred in cy-
Batches of Livarot cheese manufactured from either pas-
lindrical moulds (commonly 12 cm in diameter). Twenty-four
teurized or raw milk in three different dairies were studied.
hours later, the curd is removed from the moulds. The salting
The cheese samples were taken 1 day before the salting oper-
operation is done after a 24–48 h period. During ripening, at
ation (∼2 days of ripening), 1 day before packaging
least three rind-washing operations are done. Just before
(∼21 days of ripening), and at the end of ripening
packaging (17–18 ripening days), Livarot cheese is sur-
(∼60 days). Dairy A used thermized milk and Brevibacte-
rounded by three to five characteristic bands and thereafter
rium aurantiacum (sold as Brevibacterium linens), Staphylo-
kept at 4 °C for a minimal period of 35 days. Dry matter is
coccus sp., D. hansenii, G. candidum, and K. lactis as
at least 40 g per 100 g of cheese after complete desiccation.
commercial starter cultures; the ripening room temperature
Some of the production is farmhouse-made but most is done
was between 12 and 14 °C and humidity between 94% and
at the (semi)industrial scale. Although the same basic tech-
96%. Dairy B used raw milk and Brevibacterium aurantia-
nology is used by all the cheese producers, variations occur
cum (sold as B. linens), G. candidum, and K. lactis as com-
from one dairy to another. For example, (i) the milk used for
mercial starter cultures; the ripening room temperature was
Livarot manufacturing can be either pasteurized or unpasteur-
For personal use only.

ized, (ii) cheese is either salted in brine or with dry salt, and between 13 and 15 °C and humidity between 93% and 95%.
(iii) the surfaces of the Livarot cheeses are either washed Dairy C used pasteurized milk and Brevibacterium aurantia-
with brine or smeared three to five times during the first cum (sold as B. linens), Staphylococcus sp., D. hansenii,
3 weeks of ripening, using one or more combinations of cul- G. candidum, and K. lactis as commercial starter cultures;
tures of microorganisms, which may include Brevibacterium the ripening room temperature was between 12 and 15 °C
linens, Debaryomyces hansenii, Geotrichum candidum, or and humidity 97%.
Kluyveromyces lactis. These variations, and probably many
others, result in diversity (in colour, taste, and aroma) among Determination of viable counts
the Livarot cheeses produced (http://www.fromage-normandie. The surface microflora of each cheese was enumerated, as
com/livarot/). described by Brennan et al. (2002), on yeast extract glucose
Investigating the surface microbiota of smear cheeses is chloramphenicol agar (YGCA) for the yeasts and plate count
still scientifically relevant, as microbial populations are not agar + 3% NaCl (PCA+) for bacteria. For each sample, start-
always well known; and consequently, several new microbial ing from a single dilution, about 50 bacteria and 50 yeast
species also need to be described from such substrates (Bora colonies were isolated from PCA+ and YGCA, respectively,
et al. 2007, 2008; Schmidt et al. 2009). The dynamics of mi- with care taken to cultivate members of each observed mor-
crobial communities during ripening are also important to de- photype while keeping its relative abundance. The resulting
scribe in order to determine their specific role regarding 450 bacteria and 452 yeasts were purified, subcultured three
organoleptic properties of cheeses. It is well known that dur- times, and maintained at –80 °C.
ing ripening, yeasts dominate initially and D. hansenii is the
major species found on the surface. Geotrichum candidum or Characterization of bacteria
Yarrowia lipolytica have also been described on the surface Gram staining, motility, and catalase and oxidase activity
of Limburger, Münster, Tête de Moine (Wyder and Puhan were performed on all of the bacterial isolates.
1999), and Livarot (Larpin et al. 2006) cheeses. The physico- Genomic DNA was extracted from isolates as described in
chemical conditions created by the yeasts allow the growth of Versalovic et al. (1994), and amplification of template DNA
salt-tolerant, acid-sensitive bacteria. Diverse populations of was performed by repetitive extragenic palindromic (rep-
coryneform bacteria, micrococci, and staphylococci PCR), using primers RepIRI and Rep2I. The fingerprint pro-
(Eliskases-Lechner and Ginzinger 1995a; Abliz et al. 2004; files were analysed using BioNumerics software (Applied
Rademaker et al. 2005; Rea et al. 2007) as well as “marine” Maths, Sint-Martins-Latem, Belgium). Cluster analysis was
lactic acid bacteria (Roth et al. 2010) have been found on the performed using the Jacquard similarity coefficient and
surface of smear cheeses. Recently, the presence of Gram- Ward’s clustering method. In all clusters, isolates exhibiting
negative bacteria has also been highlighted (Rea et al. 2007; distinct profiles were chosen in conjunction with morphology
Fontana et al. 2010). While some information is available re- to represent the observed diversity. They are cited below as
garding the microbial diversity found on the surface of one dereplicated isolates.
commercial Livarot cheese sold on the retail market (Mou- Apart from Gram-positive and catalase-negative isolates,
nier et al. 2009) and on the dynamics of the main yeast spe- which were considered to be lactic acid bacteria (LAB) and

Published by NRC Research Press

 Larpin-Laborde et al. 653

hence are not considered further, most isolates were also (PCR–RFLP) analysis. Isolates giving ambiguous results
characterized using Biotype 100 strips (Biomérieux, Marcy were subjected to DNA sequencing. For these purposes,
l’Étoile, France) (Grimont et al. 1996), which were incubated DNA was prepared as described by Vasdinyei and Deak
at 30 °C for up to 6 days. Recognizer® software (Taxolab, (2003), except that the nucleic acid pellet was dissolved in
Pasteur Institute, Paris, France) and reference databases for 100 µL of water containing 2 µL of RNase (20 µg/µL). For
Gram-negative and coryneform bacteria were used to identify RAM-PCR, the oligonucleotide primer (CAC)5 was used to
the isolates. In addition, 16S rDNA sequencing of 26 isolates amplify the intersingle sequence repeats according to Gente
from the middle stage of ripening in dairy C was done to et al. (2002), and for PCR–RFLP, ITS1 and ITS4 primers
confirm their identity. were used to amplify the internal transcribed spacer region
(Guillamón et al. 1998). For sequencing purposes, the D1–D2
Can. J. Microbiol. Downloaded from www.nrcresearchpress.com by YALE MEDICAL LIBRARY on 12/26/14

Identification of dereplicated isolates domain of the LSU rDNA was amplified with primers NL1
The dereplicated isolates were grown on tryptic soy broth and NL4m (O’Donnell 1993), according to Abliz et al. (2004).
(TSB; BBL, Sparks, Maryland, USA) containing agar (15 g/L; The amplified LSU rDNA products were sent to Genome Ex-
Oxoid, Basingstoke, England). Representative strains were press (Meylan, France) for purification and direct sequencing.
maintained at –80 °C and deposited in the public BCCM/ The sequence data were blasted against the NCBI database.
LMG Bacteria Collection (http://bccm.belspo.be/about/lmg. For PCR–RFLP, CfoI, HaeIII, and HinfI restriction en-
php). All of the strains were grown for 1–2 days at 30 °C zymes were used independently to digest the amplification
on TSB (Difco, Detroit, Michigan, USA) containing agar at products of the internal transcribed spacer PCR. PCR prod-
15 g/L (Oxoid). DNA was extracted from rod-shaped iso- ucts (25 µL) were digested in a 40 µL final volume consist-
lates according to Gevers et al. (2001) and from coccus- ing of 0.5 µL of restriction enzyme (Invitrogen), 3.6 µL of
shaped isolates using the same protocol but with lysostaphin the buffer specific for that enzyme, and 0.4 µL of enzyme of
(5000 U/mL, Sigma, St. Louis, Missouri, USA) replacing bovine serum albumin (Invitrogen), and the mixture was then
mutanolysin. A set of reference strains, known to be present incubated at 37 °C for 2 h. The restriction fragments were
on red smear cheeses, was included for identification by separated by electrophoresis on 3% (m/v) agarose (Invitrogen)
rep-PCR. PCR was performed according to Versalovic et gels in 1× TAE (Tris – acetate – EDTA) buffer.
al. (1994) using the primers BOXA1R for rods and (GTG)5
For personal use only.

for cocci. If a strain gave an unsatisfactory band pattern Results

with one primer set, the other primer set was tested. Band
patterns were analysed using the Pearson coefficient and Microbial counts
UPGMA cluster analysis with BioNumerics software (Ap- Surface bacterial counts prior to salting ranged from
plied Maths). Dereplicated isolates were subjected to 16S 106 cfu/cm2 cheese in dairy C to >108 cfu/cm2 cheese in
rRNA gene sequence analysis for identification, using pri- dairy A (Fig. 1). At the middle stage of ripening, bacterial
mers described by Coenye et al. (1999). Representative levels up to 109–1010 cfu/cm2 cheese were obtained. They
strains were subjected to SDS–PAGE analysis as described were then either quite stable or increased (dairy C) until the
by Pot et al. (1994), and numerical analysis was performed end of ripening. The yeast population reached 107 cfu/cm2
using the Gel Compare version 4.2 (Applied Maths, Kor- cheese in the early stage of ripening, except for dairy C
trijk, Belgium). Isolates giving an ambiguous Gram stain where the level was only 105 cfu/cm2 cheese. Similar levels
were the subject of fatty acid analysis as described by (about 5 × 107 cfu/cm2) were reached for the three batches
Vancanneyt et al. (1996). at the middle stage of ripening, thereafter, the yeast popula-
tion decreased very slowly.
Identification of yeasts
The yeasts were dereplicated and identified by Fourier Biodiversity of surface bacteria
transform infrared (FTIR) spectroscopy. Sample preparation Of the 450 bacteria isolated, 292 (65%) were Gram-
and spectral evaluation were performed according to positive, 145 (32%) were Gram-negative, and 13 (3%) were
Kümmerle et al. (1998) using an IFS-28B FTIR spectrometer not determined.
and opus 3.17 software from Windows (Bruker, Karlsruhe, Three hundred and eighty-six of the 450 isolates were suc-
Germany). Identification was done as described by Goerges cessfully screened by rep-PCR and assigned to 17 clusters
et al. (2008). The threshold value for cluster delineation was (data not shown). Depending on the size of the cluster, one
set at a spectral distance of 0.5–1.5, depending on the spe- to 20 bacteria were chosen per cluster, so that 156 (40%) iso-
cies. One to six isolates were randomly selected as represen- lates were finally selected. Molecular analysis, using rep-PCR
tatives of each cluster, depending on cluster size. The 46 (BOX or (GTG)5 primers) and (or) 16S rRNA gene sequence
resulting dereplicated yeast isolates were characterized using analysis and (or) SDS–PAGE profiling, led to the assignment
physiological tests according to Yarrow (1998), including fer- of 10 out of the 156 isolates to putatively novel taxa. The
mentation of D-glucose, D-galactose, and lactose; assimilation Gram-positive bacteria were identified mainly as Arthro-
of 23 carbon sources and nitrite, ethylamine, lysine, cadaver- bacter arilaitensis, Brevibacterium aurantiacum, Corynebac-
ine, and glucosamine; tolerance to 1% (v/v) acetic acid; and terium casei (Table 1) and as members of the Staphylococcus
protease and lipase production. The resulting phenotypic pro- saprophyticus group. Among the non-Enterobacteriaceae
files were compared with those described by Barnett et al. Gram-negative bacteria, the genera Pseudomonas, Alcali-
(2000). Identity of the 46 dereplicated yeast isolates was con- genes, Acinetobacter, and Psychrobacter were identified.
firmed using Random Amplified Microsatellites-PCR (RAM- Among the 450 isolates, 411 were subjected to phenotypic
PCR) and PCR – restriction fragment length polymorphism analysis. Seventy-five Gram-positive and catalase-negative

Published by NRC Research Press

 654 Can. J. Microbiol. Vol. 57, 2011

Fig. 1. Changes in bacterial (solid lines) and yeast (broken lines) populations during ripening of Livarot cheese manufactured in three dairies:
•, dairy A; ▪, dairy B; ▴, dairy C.
11

10

9
Log of cells per cm2 cheese

8
Can. J. Microbiol. Downloaded from www.nrcresearchpress.com by YALE MEDICAL LIBRARY on 12/26/14

4
0 10 20 30 40 50 60

Days of ripening

Table 1. Identity of dereplicated bacterial strains, using molecu- contained strains identified as Candida catenulata, Debaryo-
lar analysis. myces hansenii, Geotrichum candidum, Kluyveromyces lactis
or Kluyveromyces marxianus, Yarrowia lipolytica, and Can-
Identity LMG No.
dida glabrata or Pichia jadinii. The five remaining isolates
For personal use only.

Agrococcus casei 22447 formed single-membered clusters, one of which was identi-
Arthrobacter arilaitensis 22417 fied as Cryptococcus humicolus and another as Torulaspora
Brachybacterium spp. 22434 delbrueckii. Using physiological tests and molecular tools,
Brevibacterium aurantiacum 22422 the identity of 46 dereplicated isolates was confirmed in
Corynebacterium casei 22430 most cases. One out of the eight isolates identified as Y. lip-
Lactic acid bacteria 22794 olytica using the phenotypic data was finally identified as
Macrococcus sp. 22445 Candida deformans, based on the percentage identity of its
Microbacterium gubbeenense 22428 LSU rDNA sequence. It was not possible to differentiate be-
New taxon close to the genus Leucobacter 22448, 22412 tween the type strains K. lactis and K. marxianus using phe-
Staphylococcus saprophyticus group 22436 notypic data. However, these strains were distinguished using
Acinetobacter spp. — RAM-PCR, and all of the Livarot isolates belonged to the
Alcaligenes spp. — species K. lactis (Fig. 2). Among four isolates identified by
Pseudomonas spp. — FTIR spectroscopy as Candida catenulata, one was finally
Psychrobacter spp. — identified as a strain of Candida natalensis, as it showed a
Enterobacteriaceae 22792 LSU rDNA sequence that was 99% similar to Candida nata-
Other Gram-negative bacteria 22790 lensis. One isolate, which was identified as Crytococcus hu-
Note: Numbers for each representative bacterium are from the Labor- micolus by both FTIR spectroscopy and by the physiological
atorium voor Microbiologie (LMG), Universiteit Gent. data, was found to show 99% identity with Cryptococcus
musci using the LSU rDNA data.
isolates were considered to be LAB and were not studied fur-
ther. The 336 remaining isolates were examined using the Population dynamics
Biotype 100 strips (Table 2), and 247 of them were success- By merging data obtained using (i) FTIR spectroscopy and
fully identified. Forty isolates were shown to be coagulase- molecular analysis for yeasts, (ii) Biotype 100 analysis of 247
negative staphylococci, which were mainly identified as bacterial isolates, (iii) additional 16S rDNA sequencing of 26
Staphylococcus xylosus or Staphylococcus spp. The 97 iso- of the bacterial isolates, and (iv) molecular analysis of 386
lates found to be coryneform bacteria were notably assigned bacterial isolates, including the 156 dereplicated ones, an
to two taxa, namely Arthrobacter arilaitensis and Brevibacte- overview of surface microbial diversity of Livarot cheese
rium spp. Gram-negative isolates were assigned to 11 species was obtained (Table 4).
or genera. The predominant taxa were identified as Alcali- For bacteria, members of 20 different taxa were identified
genes faecalis, Alcaligenes piechaudii, Proteus vulgaris, and at all stages of ripening for dairy A. The number of bacterial
Hafnia alvei. groups remained constant during ripening but dominant
groups changed. At the early stage of ripening the S. sapro-
Biodiversity of yeasts phyticus group dominated, P. vulgaris dominated at the mid-
Four hundred and fifty-two yeast isolates were assigned to dle stage, and S. xylosus and members of the new taxon close
28 clusters by FTIR spectroscopy (Table 3). The clusters to Leucobacter sp. were the most numerous groups at the late

Published by NRC Research Press

 Larpin-Laborde et al. 655

Table 2. Identification of bacterial strains isolated from Livarot Table 3. Yeast diversity assessed by Fourier transform infrared
cheese using Biotype 100 strips. spectroscopy.

No. of No. of No. of

Identification isolates Identified species clusters isolates (%)
Staphylococci Candida catenulata 4 23 (5)
Staphylococcus equorum 2 Debaryomyces hansenii 2 56 (12)
Staphylococcus spp. 17 Geotrichum candidum 10 278 (62)
Staphylococcus xylosus 21 Kluyveromyces lactis or K. marxianus 3 38 (8)
Total 40 Pichia jadinii or Candida glabrata 1 2 (<1)
Can. J. Microbiol. Downloaded from www.nrcresearchpress.com by YALE MEDICAL LIBRARY on 12/26/14

Yarrowia lipolytica 8 50 (11)

Coryneform bacteria Cryptococcus humicolus 0 1 (<1)
Arthrobacter arilaitensis 32 Geotrichum candidum 0 3 (<1)
Brachybacterium alimentarium 10 Torulaspora delbrueckii 0 1 (<1)
Brachybacterium faecium 2
Brevibacterium spp./Arthrobacter spp. 1
Brevibacterium spp. /Corynebacterium spp. 1 Fig. 2. Random amplified microsatellites PCR patterns of isolates
Brevibacterium spp. 34 assigned to the genus Kluyveromyces, using phenotypic data, and of
Corynebacterium casei 15 the type strains Kluyveromyces lactis and Kluyveromyces marxianus,
Microbacterium gubbeenense 2 based on (CAC)5 primer. Lanes: 1, K. lactis CBS 683T; 2–5, yeasts
Total 97 strains isolated from Livarot cheese; and 6, K. marxianus CBS 712T.
1
Gram-negative bacteria
Alcaligenes eutrophus 1 2
Alcaligenes faecalis 30 3
Alcaligenes piechaudii 21
Alcaligenes spp. 2 4
For personal use only.

Citrobacter freundii sensu stricto 1 5

Escherichia coli 4
Hafnia alvei 16 6
Proteus vulgaris 21
Providencia heimbacheae 1 cheese, the diversity level was quite constant. Candida cate-
Pseudomonas putida 9 nulata and G. candidum were found throughout the ripening
Serratia liquefaciens 4 period. In contrast, K. lactis and P. jadinii were isolated only
Total 110 at the beginning of ripening and Y. lipolytica at the middle
stage. Members of six different species were identified in
Not determined 19 dairy C cheese. As for dairy A cheese, both D. hansenii and
No growth in biotype 100 medium 70 G. candidum were present at all stages of ripening; K. lactis
Total isolates 336 was detected only at the first stage of ripening. The highest
diversity was recorded before packaging. Yarrowia lipolytica
stage of ripening. In cheese from dairy B, 10 Gram-positive was first detected at the middle stage of ripening and showed
and six Gram-negative groups were delineated from all stages a higher incidence at the end of ripening. Members of nine
of ripening. The bacterial diversity increased at the middle taxa were common to the three cheeses: Arthrobacter arilai-
stage of ripening mainly with respect to the Gram-positive tensis, Brachybacterium alimentarium, Brevibacterium aur-
flora. Hafnia alvei strains dominated the early stage of ripen- antiacum, Brevibacterium sp., the taxon closely related to
ing, Corynebacterium casei the middle stage of ripening, and the genus Leucobacter, and Pseudomonas sp. for the bacte-
Alcaligenes faecalis the late stage. In cheese from dairy C, ria; and G. candidum, K. lactis, and Y. lipolytica for the
Gram-negative taxa were the most numerous. The greatest di- yeasts.
versity was found at the middle stage of ripening. Arthro-
bacter arilaitensis and Pseudomonas putida isolates were Discussion
dominant before packaging. At the end of ripening, the bac- Maximum viable bacterial counts obtained during the rip-
terial flora was dominated by Arthrobacter arilaitensis, Bre- ening of Livarot cheese were 1–2 logs higher than those pre-
vibacterium aurantiacum, and Psychrobacter sp. viously reported for Gubbeen cheese (108 cfu/cm2) (Brennan
The diversity of the yeast flora was lower than that of the et al. 2002) or one Livarot cheese from the retail market
bacteria (Table 4). Members of five species were found at the (Mounier et al. 2009) but similar to those of Tilsit, which
early stage of ripening in dairy A cheese, with only two and had 109 cfu/cm2 (Eliskases-Lechner and Ginzinger 1995a).
three species detected at the middle and late stages, respec- A variety of morphologically different bacterial colonies
tively. Debaryomyces hansenii and G. candidum were present grew on PCA+. The Gram-positive bacteria isolated from
at all stages of ripening. Cryptococcus musci and K. lactis the surface of Livarot cheese belonged either to the LAB or
were present only at the beginning of ripening, and Y. lipoly- to commonly described genera, such as Arthrobacter, Brevi-
tica was detected at the late ripening stage. In dairy B bacterium, Corynebacterium, Microbacterium, and Staphylo-

Published by NRC Research Press

 656 Can. J. Microbiol. Vol. 57, 2011

Table 4. Identification of isolates at different stages of ripening based on samples taken from the three dairies.

Dairy A Dairy B Dairy C

Early Middle Late Early Middle Late Early Middle Late Total
Bacteria
Gram-positive bacteria
Agrococcus casei 1 1
Arthrobacter arilaitensis 3 6 1 5 9 10 34
Arthrobacter spp. 1 1
Brachybacterium alimentarium 1 1 1 5 1 1 10
Can. J. Microbiol. Downloaded from www.nrcresearchpress.com by YALE MEDICAL LIBRARY on 12/26/14

Brachybacterium spp. 2 2
Brevibacterium aurantiacum 5 1 2 5 3 1 8 25
Brevibacterium spp. 4 3 3 1 2 1 3 17
Corynebacterium casei 3 1 16 20
Macrococcus sp. 1 1
Microbacterium gubbeenense 1 3 4
New taxon close to Leucobacter spp. 6 11 1 2 1 21
Staphylococcus epidermidis group 1 1
Staphylococcus equorum 1 1
Staphylococcus saprophyticus group 14 3 1 1 19
Staphylococcus pasteuri 2 2
Staphylococcus xylosus 2 2 17 21
Staphylococcus spp. 1 1 2
Unidentified Gram-positive bacteria 1 1 12 5 2 9 5 35
Gram-negative bacteria
Acinetobacter johnsonii 1 1
For personal use only.

Acinetobacter spp. 1 1
Alcaligenes faecalis 2 26 28
Alcaligenes piechaudii 1 1
Alcaligenes spp. 1 1 3 5
Enterobacteriaceae 1 1 2 1 5
Escherichia coli 2 1 3
Ewingella americana 2 2
Hafnia alvei 14 2 1 17
Marinomonas spp. 3 3
Proteus vulgaris 14 2 1 17
Proteus spp. 2 2
Pseudomonas putida 8 8
Pseudomonas stutzeri 1 1
Pseudomonas spp. 2 1 1 1 1 6
Psychrobacter celer 2 2
Psychrobacter spp. 1 1 15 17
Raoultella planticola 1 1
Serratia grimesii 1 1
Serratia liquefaciens 2 1 3
Unidentified Gram-negative bacteria 7 2 2 1 1 4 2 2 21
Other unidentified bacteria 3 4 4 2 13

Yeasts
Candida catenulata 11 7 3 21
Candida natalensis 2 2
Cryptococcus musci 1 1
Debaryomyces hansenii 12 14 3 21 3 3 56
Geotrichum candidum 23 36 40 30 42 34 6 42 28 281
Kluyveromyces lactis 10 5 23 38
Pichia jadinii 2 2
Torulaspora delbrueckii 1 1
Yarrowia lipolytica 4 7 1 18 1 19 50

Total microbial isolates 92 94 98 71 99 100 78 99 96 827

Published by NRC Research Press

 Larpin-Laborde et al. 657

coccus (Beresford et al. 2001). Less common bacteria were where their presence has no relationship to food safety (Jay
also found on Livarot cheese, e.g., Brachybacterium alimen- et al. 2005).
tarium, as described for hard cheeses (Schubert et al. 1996), Pseudomonas and Psychrobacter strains were found at the
but also bacteria belonging to the genera Agrococcus and surface of Livarot cheese. Members of these genera have
Macrococcus. The latter have been encountered in cow’s and been described as spoilage bacteria responsible for the ap-
goat’s milks and were detected in the non-smear-ripened Si- pearance of sticky or fatty rind on blue-veined cheeses and
cilian cheese Ragusano (Randazzo et al. 2002). Agrococcus of a slimy film on cottage-cheese (Bergère and Lenoir 1997)
casei sp. nov. has been identified not only from Livarot or for the development of a fluorescent stain on the cheese
cheese but also from two other smear-ripened cheeses, surface. No defects were recorded on Livarot cheeses used in
namely, Gubbeen and Tilsit (Bora et al. 2007; Rea et al. this study. Some of these Gram-negative bacteria are well
Can. J. Microbiol. Downloaded from www.nrcresearchpress.com by YALE MEDICAL LIBRARY on 12/26/14

2007). The technological impact of these bacteria remains to known for exhibiting strong proteolytic and lipolytic activ-
be understood. ities. They also produce in food (e.g., Pseudomonas putida)
Unlike what has been done in most previous studies focus- (Morales et al. 2004; Jay et al. 2005), including cheese (Dee-
ing on surface bacteria from cheese (Eliskases-Lechner and tae et al. 2009a), volatile sulphur compounds such as di-
Ginzinger 1995a; Valdés-Stauber et al. 1997; Rademaker et methyl disulfide, which significantly contribute to the aroma
al. 2005), the bacterial isolates in this investigation were of cheese (Kagkli et al. 2006). Such activities may have a
chosen to represent the macroscopic diversity irrespective of positive effect during cheese ripening. Given such results,
whether strains were Gram-positive or Gram-negative. Little we think that it would be interesting to explore the contribu-
is known about the occurrence of Gram-negative bacteria, tion of Gram-negative bacteria to cheese ripening, provided it
which are mainly regarded as indicators of either poor hy- is balanced by safety assessments.
giene (e.g., coliform bacteria) (Jay et al. 2005) or as spoil- Maximum viable yeast counts reached levels (∼107 cfu/cm2)
age-contaminating bacteria (e.g., Pseudomonas fluorescens) comparable to those reported for other red smear cheeses,
(Neugebauer and Gilliland 2005). Milk used for making the such as Limburger and Münster (Wyder and Puhan 1999).
Livarot cheeses was either pasteurized — and can be consid- The yeast flora consisted mainly of D. hansenii, G. candi-
ered as free from or as containing very low levels of non- dum, K. lactis, and Y. lipolytica. These species have been
thermoresistant bacteria such as coliform bacteria or pseudo- described to as agents of lipolytic, proteolytic, or esterase
For personal use only.

monads — or unpasteurised. Raw milk used in France nowa- activities during ripening of smear cheeses (Choisy et al.
1997). Geotrichum candidum, D. hansenii, and K. lactis
days commonly has total bacterial counts of less than
can be used as commercial starter cultures in the manufac-
5000 cfu/mL (Beuvier and Buchin 2004).
turing of Livarot cheese but not, to our knowledge, Y. lip-
Hygiene practices in the three dairy plants were very satis-
olytica. Nevertheless this organism was present in greater
factory, all having efficient cleaning procedures, good manu-
numbers than were K. lactis strains. The predominant yeasts
facturing practices, and personnel hygiene policy. The
isolated from Livarot cheese have been found in this and
sanitary and hygienic qualities of Livarot cheese can be con-
other smear cheeses, such as brick cheeses (Valdés-Stauber
sidered satisfactory, as potential pathogenic bacteria, such as et al. 1997), Limburger and Münster cheeses (Wyder and
Listeria monocytogenes, Staphylococcus aureus or other co- Puhan 1999), Tilsit cheese (Eliskases-Lechner and
agulase positive staphylococci, and Salmonella sp., were not Ginzinger 1995b), and Münster and Maroilles cheeses
identified amongst the isolates. Some Escherichia coli strains (Mounier et al. 2010). Nevertheless, if D. hansenii is con-
were isolated from the early stage but not from the later stage sidered to be by far the dominant yeast occurring in virtu-
of ripening in dairy A and C cheeses. It is interesting, there- ally all cheeses (Beresford et al. 2001), it was not as
fore, that a high population of Gram-negative bacteria be- dominant as G. candidum at the end of ripening of Livarot
longing to 20 identified taxa was detected on the surface of cheese. It would seem that Livarot cheese is a biotope to
all Livarot cheeses. Enterobacteriaceae have been isolated which G. candidum is well adapted.
from various cheeses (Tornadijo et al. 1993; Duthoit et al. High diversity, with as much as 46 different microbial
2005), while H. alvei has been shown to be the major Gram- groups, has been demonstrated at the surface of Livarot
negative species during ripening on the surface of Camem- cheese using culture-dependent polyphasic analysis. While it
bert, a soft cow’s milk cheese (Richard and Zadi 1983). A can be inferred from various studies done on German cheeses
similar result was reported for a soft goat’s milk cheese (Maoz et al. 2003) and Camembert cheese (Henri-Dubernet
(Sablé et al. 1997). et al. 2008) that uncultivable microorganisms are not so com-
Some strains of H. alvei have been implicated as agents of mon in cheese and milk, an approach coupling both culture-
gastroenteritis (Rodríguez et al. 1999), but others are sold as dependent and culture-independent rRNA analysis would
ripening agents for cheesemaking (Skovmose 2003). In Li- have probably yielded more diversity.
varot, H. alvei was only found on dairy B cheese, represent- Several bacterial and yeast commercial cultures are used in
ing 33% of its Gram-negative flora. It has been suggested Livarot cheese making. A strain of Staphylococcus sp., which
that H. alvei is involved in the synthesis of volatile aromatic was added as a commercial culture in dairy C, was not recov-
compounds in Camembert (Richard and Gratadoux 1984) ered from the surface microbial flora at any point during rip-
and shown that Enterobacteriaceae strains of dairy origin ening of Livarot cheese. A similar phenomenon has been
are able to produce volatile compounds in experimental His- reported for Gubbeen cheese (Brennan et al. 2002) for a
pánico (Morales et al. 2004) and red smear (Deetae et al. commercial strain of Brevibacterium linens and for Limbur-
2009b) cheeses. Several studies have demonstrated that coli- ger cheese (Goerges et al. 2008). In the same way, Feurer et
forms and fecal coliforms exist in many raw food materials al. (2004) concluded that the bacterial composition of the

Published by NRC Research Press

 658 Can. J. Microbiol. Vol. 57, 2011

surface flora of a French red smear cheese at the end of rip- useful bacteria in some cheese technologies. The assessment
ening did not reflect the composition of the smearing inocu- of their positive versus negative role should be objectively
lum. In contrast, isolates belonging to the same taxon as done.
another commercial Staphylococcus sp. strain were recovered
at two sampling points from cheese manufactured in dairy A, Acknowledgements
using this commercial strain. While commercial strains of This study was supported in part by the European Union
K. lactis were added to the three cheeses, this yeast was only (project QLK1-CT-2001-02228: Biodiversity and anti-listerial
detected in the early stage of ripening and thus seemed un- activity of surface microbial consortia from Limburger, Re-
able to establish in Livarot cheese, as confirmed using real- blochon, Livarot, Tilsit and Gubbeen cheese). Part of this
time PCR (Larpin et al. 2006). Thus it seems clear that work was made possible by a Ph.D. scholarship from Higher
Can. J. Microbiol. Downloaded from www.nrcresearchpress.com by YALE MEDICAL LIBRARY on 12/26/14

microorganisms, deliberately added as smearing agents, do Education Commission (HEC) of Pakistan and received a fi-
not necessarily become dominant, depending on the strain nancial support from the Conseil Régional de Basse-Normandie.
and maybe on the species (K. lactis). This last point needs to
be further investigated by cheese makers to select, if neces- References
sary, appropriate commercial strains.
Abliz, P., Fukushima, K., Takizawa, K., and Nishimura, K. 2004.
Whatever the milk used (pasteurized or unpasteurized),
Identification of pathogenic dematiaceous fungi and related taxa
similar levels of biodiversity were observed in the three dair- based on large subunit ribosomal DNA D1/D2 domain sequence
ies, suggesting a nonnegligible role of the resident microflora analysis. FEMS Immunol. Med. Microbiol. 40(1): 41–49. doi:10.
of the cheese dairies. Members of some species were found 1016/S0928-8244(03)00275-X. PMID:14734185.
in the cheeses from all three dairies, as exemplified by Ar- Barnett, J.A., Payne, R.W., and Yarrow, D. 2000. Yeasts: character-
throbacter arilaitensis, Brevibacterium aurantiacum, the pre- isation and identification. Cambridge University Press, Cam-
sumptive novel taxon closely related to the genus bridge, UK.
Leucobacter, G. candidum, K. lactis, and Y. lipolytica. In Beresford, T.P., Fitzsimons, N.A., Brennan, N.L., and Cogan, T.M.
contrast, members of other species were specific to a particu- 2001. Recent advances in cheese microbiology. Int. Dairy J. 11(4–
lar dairy. Each cheese had its own microbial flora at species 7): 259–274. doi:10.1016/S0958-6946(01)00056-5.
level, a phenomenon that has been reported for LAB and cor- Bergère, J.-L., and Lenoir, J. 1997. Les accidents de fromagerie et les
For personal use only.

yneform bacteria in Salers cheese (Duthoit et al. 2003) and, défauts des fromages. In Le fromage. Edited by A. Eck and J.-C.
at the strain level, for lactobacilli in Comté cheese (Depouilly Gillis. Lavoisier Tec and Doc, Paris, France. pp. 509–543.
et al. 2004) and Camembert cheese (Henri-Dubernet et al. Beuvier, E., and Buchin, S. 2004. Raw milk cheeses. In Cheese —
2008). The microbiological differences between the three chemistry, physics and microbiology. Vol 1. 3rd ed. Edited by P.F.
cheeses may be influenced by milk composition, practices or Fox, P.L.H. McSweeney, T.M. Cogan, and T.P. Guinee. Elsevier
technology, and interactions between starters and nonstarter Academic Press, London, UK. pp. 319–345.
Bora, N., Vancanneyt, M., Gelsomino, R., Swings, J., Brennan, N.,
microorganisms (Buchin et al. 1998; Tessier et al. 2003). For
Cogan, T.M., et al. 2007. Agrococcus casei sp. nov., isolated from
all three cheeses (although the effect is not as strong in dairy
the surfaces of smear-ripened cheeses. Int. J. Syst. Evol. Microbiol.
A cheese), the middle sample was characterized by a high 57(1): 92–97. doi:10.1099/ijs.0.64270-0. PMID:17220448.
bacterial diversity. This can be explained by the fact that the Bora, N., Vancanneyt, M., Gelsomino, R., Snauwaert, C., Swings, J.,
smear was not developed at an early stage of ripening, so Jones, A., et al. 2008. Mycetocola reblochoni sp. nov., isolated
many bacterial isolates particularly in dairies B and C were from the surface microbial flora of Reblochon cheese. Int. J. Syst.
LAB. Few species (probably commercial strains) dominated Evol. Microbiol. 58(12): 2687–2693. doi:10.1099/ijs.0.64201-0.
at the early stage (mainly in dairies A and B) when the pH PMID:19060041.
was low. In the middle stage, when the pH was more favor- Brennan, N.M., Ward, A.C., Beresford, T.P., Fox, P.F., Goodfellow,
able, it can be assumed that the cheese surface was colonized M., and Cogan, T.M. 2002. Biodiversity of the bacterial flora on the
by microorganisms originating from milk and from the resi- surface of a smear cheese. Appl. Environ. Microbiol. 68(2): 820–
dent dairy microbiota. Selective forces (competition for nu- 830. doi:10.1128/AEM.68.2.820-830.2002. PMID:11823224.
trient, lowering of temperature) probably occur at the end of Buchin, S., Delague, V., Duboz, G., Berdagué, J.-L., Beuvier, E.,
ripening. Finally, regarding dairy C, as 26 additional isolates Pochet, S., and Grappin, R. 1998. Influence of pasteurization and
were subjected to rRNA phylogenetic study, a bias was prob- fat composition of milk on the volatile compounds and flavor
ably introduced in the level of diversity compared with other characteristics of a semi-hard cheese. J. Dairy Sci. 81(12): 3097–
sampling points and other dairies. 3108. doi:10.3168/jds.S0022-0302(98)75874-6.
Compared with other smear cheeses, Livarot cheese seems Choisy, C., Desmazeaud, M., Gripon, J.C., Lamberet, G., and Lenoir,
J. 1997. La biochimie de l’affinage. In Le fromage. Edited by A.
to be characterized by the presence of G. candidum as the
Eck and J.-C. Gillis. Lavoisier Tec and Doc, Paris, France. pp. 86–
dominant yeast, by members of a new taxon that is most
105.
closely related to the genus Leucobacter, and by high num- Coenye, T., Falsen, E., Vancanneyt, M., Hoste, B., Govan, J.R.W.,
bers of Gram-negative bacteria. Describing microbial diver- Kersters, K., and Vandamme, P. 1999. Classification of Alcali-
sity and dynamics can lead to improvements in cheese genes faecalis-like isolates from the environment and human
technology. Indeed, members of all these taxa may contribute clinical samples as Ralstonia gilardii sp. nov. Int. J. Syst.
to the aromatic properties of the cheese, the hurdle effect re- Bacteriol. 49(2): 405–413. doi:10.1099/00207713-49-2-405.
garding pathogenic bacteria, and cheese typicality. Knowing PMID:10319461.
these data, technological parameters can be used to influence Deetae, P., Spinnler, H.E., Bonnarme, P., and Helinck, S. 2009a.
the balance of microbial communities. It appears that some Growth and aroma contribution of Microbacterium foliorum,
Gram-negative bacteria should now be regarded as potentially Proteus vulgaris and Psychrobacter sp. during ripening in a cheese

Published by NRC Research Press

 Larpin-Laborde et al. 659

model medium. Appl. Microbiol. Biotechnol. 82(1): 169–177. of RDO Camembert cheese. Can. J. Microbiol. 54(3): 218–228.
doi:10.1007/s00253-008-1805-7. PMID:19083231. doi:10.1139/W07-137. PMID:18388993.
Deetae, P., Mounier, J., Bonnarme, P., Spinnler, H.E., Irlinger, F., and Jay, J.M., Loessner, M.J., and Golden, D.A. 2005. Modern food
Helinck, S. 2009b. Effects of Proteus vulgaris growth on the microbiology. 7th ed. Springer Science, New York, USA.
establishment of a cheese microbial community and on the Kagkli, D.M., Bonnarme, P., Neuveglise, C., Cogan, T.M., and
production of volatile aroma compounds in a model cheese. J. Casaregola, S. 2006. L-Methionine degradation pathway in
Appl. Microbiol. 107(4): 1404–1413. doi:10.1111/j.1365-2672. Kluyveromyces lactis: identification and functional analysis of
2009.04315.x. PMID:19426267. the genes encoding L-methionine aminotransferase. Appl. Environ.
Depouilly, A., Dufrene, É., Beuvier, É., and Berthier, F. 2004. Microbiol. 72(5): 3330–3335. doi:10.1128/AEM.72.5.3330-3335.
Genotypic characterisation of the dynamics of the lactic acid 2006. PMID:16672474.
Can. J. Microbiol. Downloaded from www.nrcresearchpress.com by YALE MEDICAL LIBRARY on 12/26/14

bacterial population of Comté cheese. Lait, 84(1–2): 155–167. Kümmerle, M., Scherer, S., and Seiler, H. 1998. Rapid and reliable
doi:10.1051/lait:2003036. identification of food-borne yeasts by Fourier-transform infrared
Duthoit, F., Godon, J.J., and Montel, M.C. 2003. Bacterial spectroscopy. Appl. Environ. Microbiol. 64(6): 2207–2214.
community dynamics during production of registered designation PMID:9603836.
of origin Salers cheese as evaluated by 16S rRNA gene single- Larpin, S., Mondoloni, C., Goerges, S., Vernoux, J.-P., Guéguen, M.,
strand conformation polymorphism analysis. Appl. Environ. and Desmasures, N. 2006. Geotrichum candidum dominates in
Microbiol. 69(7): 3840–3848. doi:10.1128/AEM.69.7.3840-3848. yeast population dynamics in Livarot, a French red-smear cheese.
2003. PMID:12839752. FEMS Yeast Res. 6(8): 1243–1253. doi:10.1111/j.1567-1364.
Duthoit, F., Tessier, L., and Montel, M.C. 2005. Diversity, dynamics 2006.00127.x. PMID:17156021.
and activity of bacterial populations in ‘Registered Designation of Maoz, A., Mayr, R., and Scherer, S. 2003. Temporal stability and
Origin’ Salers cheese by single-strand conformation polymorph- biodiversity of two complex antilisterial cheese-ripening microbial
ism analysis of 16S rRNA genes. J. Appl. Microbiol. 98(5): 1198– consortia. Appl. Environ. Microbiol. 69(7): 4012–4018. doi:10.
1208. doi:10.1111/j.1365-2672.2005.02575.x. PMID:15836490. 1128/AEM.69.7.4012-4018.2003. PMID:12839776.
Eliskases-Lechner, F., and Ginzinger, W. 1995a. The bacterial flora Morales, P., Feliu, I., Fernández-García, E., and Nuñez, M. 2004.
of surface ripened cheese with special regard to coryneforms. Lait, Volatile compounds produced in cheese by Enterobacteriaceae
75(6): 571–584. doi:10.1051/lait:1995644. strains of dairy origin. J. Food Prot. 67(3): 567–573. PMID:
Eliskases-Lechner, F., and Ginzinger, W. 1995b. The yeast flora of 15035375.
For personal use only.

surface ripened cheese. Milchwissenschaft, 50: 458–462. Mounier, J., Monnet, C., Jacques, N., Antoinette, A., and Irlinger, F.
Feurer, C., Vallaeys, T., Corrieu, G., and Irlinger, F. 2004. Does
2009. Assessment of the microbial diversity at the surface of
smearing inoculum reflect the bacterial composition of the smear
Livarot cheese using culture-dependent and independent ap-
at the end of the ripening of a French soft, red-smear cheese? J.
proaches. Int. J. Food Microbiol. 133(1-2): 31–37. doi:10.1016/j.
Dairy Sci. 87(10): 3189–3197. doi:10.3168/jds.S0022-0302(04)
ijfoodmicro.2009.04.020. PMID:19481828.
73454-2. PMID:15377597.
Mounier, J., Le Blay, G., Vasseur, V., Le Floch, G., Jany, J., and
Fontana, C., Cappa, F., Rebecchi, A., and Cocconcelli, P.S. 2010.
Barbier, G. 2010. Application of denaturing high-performance
Surface microbiota analysis of Taleggio, Gorgonzola, Casera,
liquid chromatography (DHPLC) for yeasts identification in red
Scimudin and Formaggio di Fossa Italian cheeses. Int. J. Food
smear cheese surfaces. Lett. Appl. Microbiol. 51(1): 18–23.
Microbiol. 138(3): 205–211. doi:10.1016/j.ijfoodmicro.2010.01.
017. PMID:20167385. PMID:20477955.
Gente, S., Desmasures, N., Panoff, J.M., and Guéguen, M. 2002. Neugebauer, K.A., and Gilliland, S.E. 2005. Antagonistic action of
Genetic diversity among Geotrichum candidum strains from Lactobacillus delbrueckii ssp. lactis RM2-5 toward spoilage
various substrates studied using RAM and RAPD-PCR. J. Appl. organisms in Cottage cheese. J. Dairy Sci. 88(4): 1335–1341.
Microbiol. 92(3): 491–501. doi:10.1046/j.1365-2672.2002.01553.x. doi:10.3168/jds.S0022-0302(05)72799-5. PMID:15778300.
PMID:11872125. O’Donnell, K. 1993. Fusarium and its near relatives. In The fungal
Gevers, D., Huys, G., and Swings, J. 2001. Applicability of rep-PCR holomorph: mitotic, meiotic and pleomorphic speciation in fungal
fingerprinting for identification of Lactobacillus species. FEMS systematics. Edited by D.R. Reynolds and J.W. Taylor. CAB
Microbiol. Lett. 205(1): 31–36. doi:10.1111/j.1574-6968.2001. International, Wallingford, UK. pp. 225–233.
tb10921.x. PMID:11728712. Pot, B., Vandamme, P., and Kersters, K. 1994. Analysis of
Goerges, S., Mounier, J., Rea, M.C., Gelsomino, R., Heise, V., electrophoretic whole-organism protein fingerprints. In Chemical
Beduhn, R., et al. 2008. Commercial ripening starter microorgan- methods in procaryotic systematics. Edited by M. Goodfellow and
isms inoculated into cheese milk do not successfully establish A.G. O’Donnell. John Wiley and Sons, Ltd., Chichester, UK. pp.
themselves in the resident microbial ripening consortia of a South 493–521.
German red smear cheese. Appl. Environ. Microbiol. 74(7): 2210– Rademaker, J.L.W., Peinhopf, M., Rijnen, L., Bockelmann, W., and
2217. doi:10.1128/AEM.01663-07. PMID:18281427. Noordman, W.H. 2005. The surface microflora dynamics of
Grimont, P.A.D., Vancanneyt, M., Lefevre, M., Vandemeulebroecke, bacterial smear-ripened Tilsit cheese determined by T-RFLP DNA
K., Vauterin, L., Brosch, R., et al. 1996. Ability of Biolog and population fingerprint analysis. Int. Dairy J. 15(6–9): 785–794.
Biotype-100 systems to reveal the taxonomic diversity of the doi:10.1016/j.idairyj.2004.08.027.
pseudomonads. Syst. Appl. Microbiol. 19: 510–527. Randazzo, C.L., Torriani, S., Akkermans, A.D.L., de Vos, W.M., and
Guillamón, J.M., Sabaté, J., Barrio, E., Cano, J., and Querol, A. 1998. Vaughan, E.E. 2002. Diversity, dynamics, and activity of bacterial
Rapid identification of wine yeast species based on RFLP analysis communities during production of an artisanal Sicilian cheese as
of the ribosomal internal transcribed spacer (ITS) region. Arch. evaluated by 16S rRNA analysis. Appl. Environ. Microbiol. 68(4):
Microbiol. 169(5): 387–392. doi:10.1007/s002030050587. PMID: 1882–1892. doi:10.1128/AEM.68.4.1882-1892.2002. PMID:
9560418. 11916708.
Henri-Dubernet, S., Desmasures, N., and Guéguen, M. 2008. Rea, M.C., Görges, S., Gelsomino, R., Brennan, N.M., Mounier, J.,
Diversity and dynamics of lactobacilli populations during ripening Vancanneyt, M., et al. 2007. Stability of the biodiversity of the

Published by NRC Research Press

 660 Can. J. Microbiol. Vol. 57, 2011

surface consortia of Gubbeen, a red-smear cheese. J. Dairy Sci. 90(5): Skovmose, E. 2003. Cheese flavour options. Available from http://
2200–2210. doi:10.3168/jds.2006-377. PMID:17430918. www.danlac.com/news/cheese-flavour-options [accessed 20 July
Richard, J., and Gratadoux, J.J. 1984. Evolution de la flore 2010].
microbienne à la surface des Camemberts fabriqués avec du lait Tessier, L., Bugaud, C., and Buchin, S. 2003. Tentative hierarchization
cru. Lait, 64(645–646): 496–520. doi:10.1051/lait:1984645-64638. of the influence of milk properties and technological practices on
Richard, J., and Zadi, H. 1983. Inventaire de la flore bactérienne rheological properties of Abondance cheese. J. Dairy Res. 70(4):
dominante des Camemberts fabriqués avec du lait cru. Lait, 63(623– 403–411. doi:10.1017/S0022029903006228. PMID:14649411.
624): 25–42. doi:10.1051/lait:1983623-6243. Tornadijo, E., Fresno, J.M., Carballo, J., and Martin-Sarmiento, R.
Rodríguez, L.A., Vivas, J., Gallardo, C.S., Acosta, F., Barbeyto, L., 1993. Study of Enterobacteriaceae throughout the manufacturing
and Real, F. 1999. Identification of Hafnia alvei with the and ripening of hard goat’s cheese. J. Appl. Bacteriol. 75(3): 240–
Can. J. Microbiol. Downloaded from www.nrcresearchpress.com by YALE MEDICAL LIBRARY on 12/26/14

Microscan WalkAway System. J. Clin. Microbiol. 37(12): 4186– 246. PMID:8244902.

4188. PMID:10565961. Valdés-Stauber, N., Scherer, S., and Seiler, H. 1997. Identification of
Roth, E., Miescher Schwenninger, S., Hasler, M., Eugster-Meier, E., yeasts and coryneform bacteria from the surface microflora of
and Lacroix, C. 2010. Population dynamics of two antilisterial brick cheeses. Int. J. Food Microbiol. 34(2): 115–129. doi:10.
cheese surface consortia revealed by temporal temperature 1016/S0168-1605(96)01171-3. PMID:9039559.
gradient gel electrophoresis. BMC Microbiol. 10(1): 74. doi:10. Vancanneyt, M., Witt, S., Abraham, W.R., Kersters, K., and
1186/1471-2180-10-74. PMID:20222967. Fredrickson, H.L. 1996. Fatty acid content in whole-cell hydro-
Sablé, S., Portrait, V., Gautier, V., Letellier, F., and Cottenceau, G. lysates and phospholipid fractions of Pseudomonas: a taxonomic
1997. Microbiological changes in a soft raw goat’s milk cheese evaluation. Syst. Appl. Microbiol. 19: 528–540.
during ripening. Enzyme Microb. Technol. 21(3): 212–220. Vasdinyei, R., and Deak, T. 2003. Characterization of yeast isolates
doi:10.1016/S0141-0229(97)00271-8. originating from Hungarian dairy products using traditional and
Schmidt, V.S.J., Mayr, R., Wenning, M., Glöckner, J., Busse, H.J., molecular identification techniques. Int. J. Food Microbiol. 86(1–2):
and Scherer, S. 2009. Bavariicoccus seileri gen. nov., sp. nov., 123–130. doi:10.1016/S0168-1605(03)00251-4. PMID:12892927.
isolated from the surface and smear water of German red smear Versalovic, J., Schneider, M., de Bruijn, F.J., and Lupski, J.R. 1994.
soft cheese. Int. J. Syst. Evol. Microbiol. 59(10): 2437–2443. Genomic fingerprinting of bacteria using repetitive sequence-
doi:10.1099/ijs.0.006601-0. PMID:19622668. based polymerase chain reaction. Methods Cell Biol. 5: 25–40.
Schubert, K., Ludwig, W., Springer, N., Kroppenstedt, R.M., Wyder, M.T., and Puhan, Z. 1999. Investigation in the yeast flora in
For personal use only.

Accolas, J.P., and Fiedler, F. 1996. Two coryneform bacteria smear ripened cheeses. Milchwissenschaft, 54: 330–333.
isolated from the surface of French Gruyère and Beaufort cheese Yarrow, D. 1998. Methods for the isolation, maintenance and
are new species of the genus Brachybacterium: Brachybacterium identification of yeasts. In The yeasts: a taxonomic study. 4th
alimentarium sp. nov., and Brachybacterium tyrofermentans sp. ed. Edited by C.P. Kurtzman and J.W. Fell. Elsevier Science B.V.,
nov. Int. J. Syst. Bacteriol. 46(1): 81–87. doi:10.1099/00207713- Amsterdam, the Netherlands. pp. 77–100.
46-1-81. PMID:8573524.

Published by NRC Research Press