Journal of Applied Microbiology ISSN 1364-5072

ORIGINAL ARTICLE

The long-term survival of Propionibacterium freudenreichii

in a context of nutrient shortage
F.F. Aburjaile1,2,3, M.-N. Madec2,3, S. Parayre2,3, A. Miyoshi1, V. Azevedo1, Y. Le Loir2,3 and
H. Falentin2,3
1 Department of General Biology, Federal University of Minas Gerais, Belo Horizonte, Minas Gerais, Brazil
2 INRA, UMR 1253, Science et Technologie du Lait et de l’Oeuf, Rennes, France
3 AGROCAMPUS OUEST, UMR1253, UMR Science et Technologie du Lait et de l’Oeuf, Rennes, France

Keywords Abstract
adaptation, bacteria, culturability, growth
phases, long-term survival, Propionibacterium Aims: Propionibacterium freudenreichii is an actinobacterium widely used in
freudenreichii, viable but nonculturable. dairy industry during the ripening process of Swiss-type cheeses and which
presents probiotic properties. P. freudenreichii is reportedly a hardy bacterium,
Correspondence able to survive during the cheese-making process and when subjected to
H
el
ene Falentin, INRA, UMR 1253, Science et
digestive stresses. During this study the long-term survival (LTS) of
Technologie du Lait et de l’OEuf, Rennes,
P. freudenreichii was investigated for 11 days by means of phenotypic
France.
E-mail: helene.falentin@rennes.inra.fr characterization in a culture medium without the addition of any nutrients.
Methods and Results: For 11 days, in a non-nutrient supplemented culture
YLL and HF share credit as senior co-authors medium, eight strains were monitored by measuring their optical density,
counting colony-forming units (CFU) and using LIVE/DEAD staining and
2015/1229: received 22 June 2015, revised microscopy observation. Under these conditions, all strains displayed high
15 October 2015 and accepted 3 November
survival rates in the culture medium, their culturability reaching more than
2015
9 log10 CFU ml 1 after 2 days. After 11 days, this value ranged from 7
8 to
doi:10.1111/jam.13000
8
2 log10 CFU ml 1 depending on the strain, and at least 50% of the
P. freudenreichii population displayed an intact envelope. As lysis of part of a
bacterial population may be a microbial strategy to recover nutrients, in
CIRM-BIA 138 (the strain with the highest population at day 11), cell lysis was
assessed by quantifying intact bacterial cells using qPCR targeting the
housekeeping gene tuf. No lysis was observed.
Conclusion: Taken together, our results suggest that P. freudenreichii strains
use a viable but nonculturable state to adapt to the LTS phase.
Significance and Impact of the Study: Assessing the viability of
P. freudenreichii and understanding their mechanisms for survival should be of
great interest regarding their potential probiotic applications.

various carbohydrates, the main product being propionic

Introduction
acid (Houwen et al. 1991; Loux et al. 2015). PF is report-
Propionibacterium freudenreichii is an Actinobacterium edly a hardy bacterial species, which can stay alive and
that is responsible for the opening and aroma of Swiss- metabolically active for long periods under stressful condi-
type cheeses. PF is also used as a source of probiotics tions (i.e. cave-aged Swiss-type cheeses at cool tempera-
because PF has bifidogenic properties (Isawa et al. 2002), tures) (Dalmasso et al. 2012). The genome sequence of
PF is resistant to digestive stress (Jan et al. 2000), and the type-strain CIRM-BIA1 revealed the genetic basis for
some strains are endowed with anti-inflammatory capabil- its hardiness (Falentin et al. 2010a). In the laboratory, PF
ities that could be used to prevent inflammatory bowel can be stored in a spent medium at room temperature for
diseases (Okada et al. 2013; Le Mar
echal et al. 2015). Its more than 6 months and remains culturable if inoculated
metabolic activity is characterized by its ability to ferment in a fresh medium. Actinobacteria are known to be hardy

432 Journal of Applied Microbiology 120, 432--440 © 2015 The Society for Applied Microbiology
F.F. Aburjaile et al. P. freudenreichii survival

and extremely resistant to nutrient shortage (Greenblatt

Conditions for bacterial growth
et al. 2004). However, except for the Mycobacterium genus
which has been extensively explored (Ghodbane et al. All the strains were routinely cultured in YEL medium
2014), little is known about the mechanisms underlying (pH = 7
0) (Malik et al. 1968) with no agitation and at
the long-term survival (LTS) of this class. 30°C. Each strain was grown in a different test tube for
During their growth and survival, Gram-positive and each manipulation, thus preventing the interference of
Gram-negative bacteria pass through different physiologi- oxygen in the environment and maintaining the
cal phases (Kolter et al. 1993; Hicks 2005; Finkel 2006; P. freudenreichii strains under microaerophilic conditions.
Bruno and Freitag 2011; Wen et al. 2013). In some bacte- At T0, an inoculum of 7 log10 CFU ml 1 was added to
rial species, conventional exponential and stationary all the tubes.
phases are followed successively by the death phase and a Growth kinetics were followed by Optical Density mea-
LTS phase (Finkel 2006; Helmus et al. 2012). LTS can be surements using a Model DU 640 spectrophotometer
defined as a phase during which the bacteria develop (Beckman Coulter, Fullerton, CA) at 650 nm (OD650)
mechanisms to adapt and remain viable, even after a long and by counting the number of CFU (CFU ml 1) using
period of nutritional shortage (Finkel 2006; Meena and the micromethod as described previously (Baron et al.
Rajni 2010; Bharati et al. 2012; Justice et al. 2014): (i) 2006). For enumerations, all the P. freudenreichii strains
sporulation (Fimlaid and Shen 2015), (ii) quiescence/dor- were plated in YEL agar culture medium at 30°C, under
mancy or (iii) programmed cell death (Finkel 2006; Ritter- microaerophilic conditions, until the visualization of
shaus et al. 2013). During sporulation, a vegetative colonies (6 days). In all cases, two technical replicates
bacterial cell differentiates into a metabolically dormant and two biological replicates were performed.
cell-type known as a spore, which can reawaken when
favourable conditions return (Fimlaid and Shen 2015).
Epifluorescence microscopy
Quiescence is characterized by decelerated or arrested
growth in a viable nonreplicating state (Rittershaus et al. The eight strains were monitored for 11 days using the
2013), which can evolve to a viable but nonculturable LIVE/DEADâ BacLightTM Bacterial Viability Kit – Molec-
(VBNC) state (Ramamurthy et al. 2014). In programmed ular Probesâ (Molecular Probes, Eugene, OR), as recom-
cell death (bacterial apoptosis), the death and lysis of part mended by the supplier. Microscopic observations were
of a bacterial population may be profitable to surviving performed at each of the time points used for the OD650
cells, which can catabolize the detritus. Cell death and lysis and CFU ml 1 measurements. Briefly, SYTO9 (which
can be observed by monitoring the fall in the viable count stains all cells green, regardless of their membrane integ-
using standard plating assays associated with a leakage of rity) and propidium iodide (which stains permeabilized
intracellular content that can be assessed biochemically cells red) were mixed at a ratio of 1: 1 (5 ll SYTO 9 and
(lactate dehydrogenase quantification) (Wilkinson et al. 5 ll propidium iodide). An 0
5 ll aliquot of the mix was
1994) or using qPCR methods (Falentin et al. 2010b). added to 100 ll of the bacterial culture, vortexed and
PF is a nonsporulating species, but PF is as yet
unknown whether it survives by adopting a strategy of Table 1 Propionibacterium freudenreichii strains used during this
quiescence/dormancy or lysis of part of the bacterial pop- work
ulation. During this study, we investigated the LTS (OD,
Strains of P. freudenreichii* Source
culturability, membrane permeability and lysis) of eight
P. freudenreichii strains throughout an 11-day period of CIRM-BIA 1 Reference strain
culture without the addition of any nutrients, in order to CIRM-BIA 9 Emmental cheese 1989 Van Niel
investigate inter-strain variability in LTS capacities and to CIRM-BIA 119 Gruyere cheese 1973
CIRM-BIA 121 Swiss cheese 1937
better understand the strategies used to ensure survival.
CIRM-BIA 127 Emmental cheese 1992
CIRM -BIA 129 alias ITG-P20† Emmental cheese 1992
Materials and methods CIRM-BIA 138 alias ITG-P9† Emmental cheese 1992
CIRM-BIA 139 alias ITG-P23† Emmental cheese 1992

Bacterial strains *Bacterial strains obtained from the Centre International de Res-
sources Microbiennes (CIRM) dedicated to Bacteria of Food Interest
Eight P. freudenreichii strains (Table 1) were used during
(Bact
eries d’Int

er^
et Alimentaire, BIA) - (CIRM-BIA, UMR1253, INRA
this work. The strains were supplied by the International Rennes, France).
Centre for Microbial Resources - Bacteria of Food †Strain CIRM-BIA (alias ITG-P20), strain CIRM-BIA 138 (alias ITG-P9)
Interest – (Centre International de Ressources Microbien- and strain CIRM-BIA 139 (alias ITG-P23) were kindly supplied by Acta-
nes - Bact
eries d’Int
er^et Alimentaire; INRA Rennes). lia (Rennes, France).

Journal of Applied Microbiology 120, 432--440 © 2015 The Society for Applied Microbiology 433
P. freudenreichii survival F.F. Aburjaile et al.

incubated in the dark for 15 min. The bacteria were then

Results
analysed using epifluorescence microscopy with a phase
contrast Nikon G11 microscope (Nikon D300) and CAM-
Growth and survival rate of Propionibacterium
ERA CONTROL PRO 2 software (Nikon Corporation, Tokyo,
freudenreichii under laboratory conditions
Japan). For each sample, six photos (three photos with a
940 objective and three with a 9100 objective) were The growth of the eight P. freudenreichii strains was
analysed so as to visually estimate the percentage of per- monitored spectrophotometrically by measuring the opti-
meabilized cells (red/red + green cells) at each time cal density at 650 nm (OD650), and by plate counting
point. (CFU ml 1) throughout an 11 day period. Strains were
inoculated at 7 log10 CFU ml 1 in fresh YEL medium.
Both of the monitoring methods showed that P. freuden-
DNA extraction of nonlysed cells
reichii growth curves had roughly similar profiles despite
The DNA extraction method only enables the recovery of variations in terms of their final culturability at 11 days.
DNA from nonlysed cells. Samples were collected at three The curves established over a prolonged incubation
time points (79, 144 and 168 h postinoculation) during revealed several phases, as shown for CIRM-BIA129 and
the growth of P. freudenreichii CIRM-BIA 138, as described CIRM-BIA138 (Fig. 1. See all growth curves in the sup-
previously (Parayre et al. 2007). Briefly, cells were har- plemental data, Fig. S1). Under the conditions applied,
vested from 1 ml of the culture sample by centrifugation the cultures entered the exponential growth phase to
(5000 g, 10 min.). The cell pellets were re-suspended in reach a maximum OD650 of 3–4 (for CIRM-BIA 138)
lysis buffer (Qiagen, Courtaboeuf, France) supplemented and a population of 9–9
6 log10 CFU ml 1 between 50
with lysozyme (Promega, France; 8 mg sample 1) and and 72 h. The OD650 curves then decreased steadily for
mutanolysin (Sigma, France; 20 U sample 1). These sam- about 100 h, which corresponded to entry into the sta-
ples were incubated for 30 min. at 37°C for 1 h, before tionary phase, and reached a plateau of culturability at
being treated with proteinase K (0
5 mg sample 1) prior around 168 h, for a population of 8–8
7 log10 CFU ml 1,
to purification using the DNeasy Tissue Kit, as recom- depending on the strain (Fig. 1) but always with a lower
mended by the supplier (Qiagen, Courtaboeuf, France). OD650. The latter phase was designated the LTS phase.
Three biological replicates and two technical replicates The other six P. freudenreichii growth curves were similar
were used for each time point. (Fig. S1). In terms of culturability, all the strains survived
well with a population of around 8 log10 CFU ml 1 after
11 days in the spent medium, i.e. without the addition of
Quantification of bacterial cells using quantitative PCR
any nutrients (Fig. 2, Table S1). Interestingly, two strains,
To determine any cell lysis of CIRM-BIA 138 after entry CIRM-BIA121 and CIRM-BIA138, survived better than
into stationary phase and just before entering the LTS the six others (Fig. 2) and maintained an average popula-
phase, quantitative PCR (qPCR) was used to quantify the tion of 8
2 log10 CFU ml 1 at day 11, suggesting a
nonlysed bacterial cells throughout growth, according to strain-dependency in ability of P. freudenreichii for long-
the method described by Falentin et al. (2010b). Briefly, term survival. On Petri plates, colonies of P. freudenre-
the qPCR targeted tuf, a gene present in a single copy in ichii maintained the same size and aspect, whereas pho-
the P. freudenreichii chromosome, using primers designed ton microscopy images revealed bacteria of the same size
on the tuf sequence (F: CGAACGAGTTCCACTGCGGGT; (not shown) but single cells in the exponential phase and
R: GCAACATCGGCACCATCGGAC). Amplifications groups of two to five cells in the stationary phase.
were performed exactly as described previously. Cycle
thresholds (Ct) were determined after manual adjustment
Evaluation of cell envelope permeability
of the threshold within the linear part of qPCR logarith-
mic amplification curves. CIRM-BIA 138 samples were LIVE/DEAD staining assays showed that the numbers of
collected at 79, 144, and 168 h for qPCR determination. intact cells (stained green by SYTO9) remained stable
Each experiment was carried out in triplicate (indepen- (100% index) for each of the eight strains throughout the
dent biological replicates). We checked by a spiking exponential and early stationary phases (Table S2). The
experiment that lysed cells could not contaminate qPCR numbers of intact cells then started decreasing at 50 or
results. A sample of entire cells (8 log10 CFU) was spiked 150 h postinoculation and dropped to 80 and 58
6% for
with (9 log10 CFU) lysed cells (the lysis was performed as CIRM-BIA 129 and CIRM-BIA 138, respectively, 11 days
described below). Extraction and qPCR on entire cells postinoculation. The numbers of cells with a nonperme-
and spiked sample gave the same Ct. The qPCR results abilized membrane (stained in green) decreased from 50–
will give a quantification of entire cells only. 150 h onwards, depending on the strains, to reach a min-

434 Journal of Applied Microbiology 120, 432--440 © 2015 The Society for Applied Microbiology
F.F. Aburjaile et al. P. freudenreichii survival

CIRM-BIA 129 CIRM-BIA 138

10 10

1 1
DO650

DO650
0·1 0·1

0·01 0·01
0 50 100 150 200 250 300 0 50 100 150 200 250 300
Time (h) Time (h)

CIRM-BIA 129 CIRM-BIA 138

1·E+10 1·E+10
1·E+08 1·E+08
CFU ml–1

CFU ml–1
1·E+06 1·E+06
1·E+04 1·E+04
1·E+02 1·E+02
1·E+00 1·E+00
0 50 100 150 200 250 300 0 50 100 150 200 250 300
Time (h) Time (h)

Figure 1 Growth curves of Propionibacterium freudenreichii strains CIRM-BIA129 and CIRM-BIA138 determined at OD650 (upper panels), and
plate counting (CFU ml 1, lower panels).

1·E+10 was used to assess bacterial lysis. CIRM-BIA 138 (the

strain with the highest population at day 11) cells were
pelleted at three different time points before and after the
1·E+09 observed loss of culturability. Quantitative PCR targeting
the tuf gene was carried out on the DNA extracted from
CFU ml–1

the pelleted cells. Under these conditions, lysed cells can-

1·E+08
not normally be recovered (Falentin et al. 2012). The
copy number of the tuf gene did not vary significantly in
1·E+07 the CIRM-BIA138 culture, demonstrating that the bacte-
rial cells did not undergo lysis when entering the LTS
phase (Fig. 4).
1·E+06
Cirm Cirm Cirm 9 Cirm Cirm Cirm Cirm 1 Cirm
121 138 119 127 139 129
Discussion
Strains

Figure 2 Viability of the eight Propionibacterium freudenreichii Propionibacterium freudenreichii has a high survival rate
strains (in CFU ml 1) after 11 days in the spent culture medium (YEL) at day 11
based on a mean of two technical replicates and two biological repli-
cates. Upper bars are standard deviations. Propionibacterium freudenreichii belongs to the Actinobac-
teria, a group of Gram-positive bacteria with high GC
content genomes. This group is known for its consider-
imum proportion of 50% at day 11 (Fig. 3), suggesting able versatility (Gao and Gupta 2012), and includes
that the number of dying or dead (but nonlysed) cells members that can survive under extreme conditions
increased during the stationary and LTS phases. (Greenblatt et al. 2004). Propionibacterium freudenreichii
is reported to be a hardy species widely used in the
cheese-making industry and is also well-documented for
Assessment of lysis by quantifying bacterial cells using
its probiotic potential (Jan et al. 2000). Genome sequence
qPCR
analysis has revealed the genetic basis for its hardiness:
Because both plate counting assays and epifluorescence (i) its ability to store polyphosphates as an energy reserve
microscopy showed a reduction in culturability and an and to utilize them instead of ATP, (ii) its ability to store
increase in membrane permeabilization, quantitative PCR glycogen as a carbon source, and (iii) several copies of

Journal of Applied Microbiology 120, 432--440 © 2015 The Society for Applied Microbiology 435
P. freudenreichii survival F.F. Aburjaile et al.

(a) (b)
CIRM-BIA 129 CIRM-BIA 138
Percent of green cells

Percent of green cells

120% 120%
100% 100%
80% 80%
60% 60%
40% 40%
20% 20%
0% 0%
0 50 100 150 200 250 300 0 50 100 150 200 250 300
Time (h) Time (h)

(c) (d)

Figure 3 (a) Proportion of nonpermeabilized cells during the growth and survival of Propionibacterium freudenreichii strains CIRM-BIA129 (a)
and 138 (b) measured by LIVE/DEAD assay followed by epifluroscence microscopy (mean values of three epifluorescence microscopy fields at 940
magnification, and three fields at 9100 magnification). Representative epifluorescence microscopy views of P. freudenreichii (CIRM-BIA 138) cul-
tures after LIVE/DEAD staining during exponential growth (c) and after 11 days of growth (d).

1·00E+10

1·00E+09
Cells ml–1

1·00E+08 Figure 4 Quantification of bacterial cells

during the growth of Propionibacterium
freudenreichii CIRM-BIA138. Each point gives
the copy number of tuf gene determined by
1·00E+07 qPCR on DNA extracted from
0 25 50 75 100 125 150 175 200 P. freudenreichii CIRM-BIA138 (three
Time (h) biological replicates).

chaperones to cope with different stresses (Falentin et al. mittent cryptic regrowth might be responsible for this
2010a). The eight P. freudenreichii strains studied here variability. These cryptic growths may result from some
survived well during a period of 11 days under nutri- specific metabolisms induced to adapt to starvation or
tional shortage, according to the optical density and plate the growth of a subpopulation of mutants derived from
counting assays performed (around 8 log10 CFU ml 1). the inoculated strain and with a growth advantage that
In YEL medium, lactate - the preferential carbon source was explained in the review by Finkel (2006).
for PF- was exhausted 2 days after inoculation (Dalmasso
et al. 2012). LIVE/DEAD staining followed by epifluores-
The best surviving strain did not use partial lysis to
cence microscopy ascertained that more than 50% of the
sustain viability
P. freudenreichii populations had an intact (nonpermeabi-
lized) membrane at this date. The growth curves for Previous studies had shown that some lactic acid bacteria
CIRM-BIA 129 and 138 (Fig. 1) displayed some fluctua- present in dairy products can lyse under difficult condi-
tions during long-term incubation. We suggest that inter- tions and that the cell debris released into the culture

436 Journal of Applied Microbiology 120, 432--440 © 2015 The Society for Applied Microbiology
F.F. Aburjaile et al. P. freudenreichii survival

medium may serve as a nutrient source to ensure survival fluorescent microscopy, and especially use of the Bac-
of the nonlysed subpopulation (Lortal et al. 1997; Light LIVE/DEAD kit; and molecular techniques (DNA
Deutsch et al. 2002; Lortal and Chapot-Chartier 2005). A or RNA) (Oliver 2009; Fakruddin et al. 2013). All of
study of the autolysis of Lactobacillus and dairy propionic the eight P. freudenreichii strains studied here displayed
bacteria showed that this autolysis is associated with sev- a loss of culturability of approx. 1 log10 CFU ml 1
eral factors, including among others the type of strain, (Table S1 and Fig. 5), when this parameter was com-
autolysins, pH, temperature and ionic strength (Lortal pared between 50 and 72 h and 11 days postinocula-
et al. 1997). For CIRM-BIA 138, the quantification of tion. LIVE/DEAD staining showed that about half of
intact cells using quantitative PCR showed that under the the bacterial cells had a permeabilized membrane. Using
study conditions applied here, lysis could be ruled out. quantitative PCR, an absence of lysis was shown on
Whether this is a general feature of P. freudenreichii still strain CIRM-BIA 138. These findings suggest that at
needs to be determined. The absence of lysis (i) could be least a part of the P. freudenreichii population entered a
confirmed by quantitative PCR on cells exposed to pro- VBNC state and that the entry into LTS is concomitant
pidium monoazide, so as to prevent any contamination with entry into a VBNC state, at least for 1 log10 (i.e.
by DNA from dead cells, (ii) needs to be confirmed in 90%) CFU ml 1. As the total bacterial population was
other P. freudenreichii strains and under different condi- stained either green (50%) or red (50%), we suggest
tions before it is generally accepted. that VBNC cells (90%) can take up both colours and
that LIVE/DEAD staining cannot be used to quantify
VBNC cells.
Propionibacterium freudenreichii probably enters a viable
The VBNC state has been described in various Gram-
but nonculturable state
positive bacteria, such as Enterococcus faecalis, Enterococ-
The experimental data reported here offer a range of cus hirae, Listeria monocytogenes and Mycobacterium
presumptions suggesting that Propionibacterium freuden- tuberculosis (Oliver 2009), and in Gram-negative bacteria
reichii enters into a VBNC state. VBNC is a survival such as Vibrio cholerae (Gonzalez-Escalona et al. 2006),
strategy adopted by numerous bacteria in response to Escherichia coli (Darcan et al. 2009) and Helicobacter
adverse conditions such as starvation, extreme tempera- pylori (Casasola-Rodr
ıguez et al. 2013). Many of the stud-
ture, pH, oxygen stress, antibiotic pressure, etc. Once in ies on the VBNC state have focused on pathogenic bacte-
a VBNC state, the bacteria no longer grow in routine ria (Ramamurthy et al. 2014). This state has also recently
laboratory medium and conditions, but they neverthe- been reported for bacteria of food interest and particu-
less remain viable and can be resuscitated when more larly in ripened cheese, suggesting a role for these bacteria
favourable conditions are provided (Ramamurthy et al. in the organoleptic qualities of cheeses (Martin-Platero
2014). The methods generally used to detect VBNC bac- et al. 2008; Desfoss
es-Foucault et al. 2012; Ruggirello
teria include bright field microscopy with nalidixic acid; et al. 2014).

Inoculation + Inoculation +
2 days 11 days
(a)

Culturability
(CFU ml–1)
9 log10 8 log10

(b)

LIVE/DEAD

50 % permeabilized
Figure 5 Fate of CIRM-BIA138 bacterial cells cells
between 50–72 h postinoculation and the
long-term survival phase 11 days (c)
postinoculation assessed by (a) culturability on No lysis
qPCR*
Petri plates, (b) LIVE/DEAD staining followed
(Cells ml–1)
by epifluorescence and (c) quantification of 9 log10 9 log10
intact cells by quantitative PCR.

Journal of Applied Microbiology 120, 432--440 © 2015 The Society for Applied Microbiology 437
P. freudenreichii survival F.F. Aburjaile et al.

In conclusion, our findings suggest that one strategy Bruno, J.C. and Freitag, N.E. (2011) Listeria monocytogenes
for bacterial survival is dormancy, rather than partial bac- adapts to long-term stationary phase survival without
terial lysis, in order to sustain the remaining cells. During compromising bacterial virulence. FEMS Microbiol Lett
the past decade, many studies have investigated the bacte- 323, 171–179.
rial mechanisms underlying LTS in various fields: envi- Casasola-Rodr
ıguez, B., Orta de Vel
asquez, M.T., Luque~ no-
ronmental, industrial, medical and biotechnological Mart
ınez, V.G. and Monje-Ram
ırez, I. (2013)
microbiology. This study represents the first work involv- Quantification of Helicobacter pylori in the viable but
ing the phenotypic characterization of P. freudenreichii nonculturable state by quantitative PCR in water
disinfected with ozone. Water Sci Technol 68, 2468–2472.
over an 11-day period without the addition of nutrients,
Dalmasso, M., Aubert, J., Briard-Bion, V., Chuat, V., Deutsch,
and revealed different phases of growth, membrane per-
S.-M., Even, S., Falentin, H., Jan, G. et al. (2012) A
meabilization and entry into dormancy and a VBNC state
temporal -omic study of Propionibacterium freudenreichii
to ensure LTS. A VBNC state may, however, have an
CIRM-BIA1T adaptation strategies in conditions
important impact on the technological uses of these bac-
mimicking cheese ripening in the cold. PLoS ONE 7,
teria, as either starters in cheese-making or probiotics, as e29083.
viability and metabolic activity are important traits in Darcan, C., Ozkanca, R., Idil, O. and Flint, K.P. (2009) Viable
their wide range of potential uses. This therefore opens but non-culturable state (VBNC) of Escherichia coli related
the way for further investigations on a molecular basis of to EnvZ under the effect of pH, starvation and osmotic
the LTS of P. freudenreichii. The availability of genome stress in sea water. Pol J Microbiol 58, 307–317.
sequences (Loux et al. 2015) will greatly assist such inves-
Dussault-Lepage, V., Le Boucher, C.,
Desfoss
es-Foucault, E.,
tigations using comparative genomics and transcriptomics Savard, P., LaPointe, G. and Roy, D. (2012) Assessment of
approaches in order to understand the organization, probiotic viability during Cheddar cheese manufacture
genes and metabolic pathways required for LTS. Assessing and ripening using propidium monoazide-PCR
viability and understanding the mechanisms underlying quantification. Front Microbiol 3, 1–11.
the survival of P. freudenreichii should be of considerable Deutsch, S.-M., Ferain, T., Delcour, J. and Lortal, S. (2002)
interest regarding the development of effective bacterial Lysis of lysogenic strains of Lactobacillus helveticus in
probiotics. Swiss cheeses and first evidence of concomitant
Streptococcus thermophilus lysis. Int Dairy J 12, 591–600.
Fakruddin, M., Mannan, K.S.B. and Andrews, S. (2013) Viable
Acknowledgements but non culturable bacteria: food safety and public health
perspective. ISRN Microbiol 2013, 1–6.
This work was supported by the CAPES-COFECUB
Falentin, H., Deutsch, S.-M., Jan, G., Loux, V., Thierry, A.,
French-Brazilian Cooperation Programme. We also thank
Parayre, S., Maillard, M.-B., Dherb
ecourt, J. et al.
the Institut National de la Recherche Agronomique
(2010a) The complete genome of Propionibacterium
(INRA, UMR 1253, STLO, Rennes, France) for its finan- freudenreichii CIRM-BIA1T, a hardy Actinobacterium
cial and practical support. Our thanks also go to Profes- with food and probiotic applications. PLoS ONE 5,
sor Mary Bret and Vicky Hawken for English revision of e11748.
the paper. Falentin, H., Postollec, F., Parayre, S., Henaff, N., Le Bivic, P.,
Richoux, R., Thierry, A. and Sohier, D. (2010b) Specific
Conflict of Interest metabolic activity of ripening bacteria quantified by
real-time reverse transcription PCR throughout
The authors have no conflict of interests to declare. Emmental cheese manufacture. Int J Food Microbiol 144,
10–19.
Falentin, H., Henaff, N., Le Bivic, P., Deutsch, S.-M., Parayre,
References
S., Richoux, R., Sohier, D., Thierry, A. et al. (2012)
Baron, F., Cochet, M.-F., Ablain, W., Grosset, N., Madec, M.- Reverse transcription quantitative PCR revealed
N., Gonnet, F., Jan, S. and Gautier, M. (2006) Rapid and persistency of thermophilic lactic acid bacteria metabolic
cost-effective method for micro-organism enumeration activity until the end of the ripening of Emmental cheese.
based on miniaturization of the conventional plate- Food Microbiol 29, 132–140.
counting technique. Le Lait 86, 251–257. Fimlaid, K.A. and Shen, A. (2015) Diverse mechanisms
Bharati, B.K., Sharma, I.M., Kasetty, S., Kumar, M., regulate sporulation sigma factor activity in the
Mukherjee, R. and Chatterji, D. (2012) A full-length Firmicutes. Curr Opin Microbiol 24, 88–95.
bifunctional protein involved in c-di-GMP turnover is Finkel, S.E. (2006) Long-term survival during stationary phase:
required for long-term survival under nutrient starvation evolution and the GASP phenotype. Nat Rev Microbiol 4,
in Mycobacterium smegmatis. Microbiology 158, 1415–1427. 113–120.

438 Journal of Applied Microbiology 120, 432--440 © 2015 The Society for Applied Microbiology
F.F. Aburjaile et al. P. freudenreichii survival

Gao, B. and Gupta, R.S. (2012) Phylogenetic framework and strains of Propionibacterium freudenreichii. BMC Genom
molecular signatures for the main clades of the phylum 16, 296.
Actinobacteria. Microbiol Mol Biol Rev 76, 66–112. Malik, A.C., Reinbold, G.W. and Vedamuthu, E.R. (1968) An
Ghodbane, R., Medie, F.M., Lepidi, H., Nappez, C. and evaluation of the taxonomy of Propionibacterium. Can J
Drancourt, M. (2014) Long-term survival of tuberculosis Microbiol 14, 1185–1191.
complex mycobacteria in soil. Microbiology 160, 496–501. Martin-Platero, A.M., Valdivia, E., Maqueda, M., Martin-
Gonzalez-Escalona, N., Fey, A., Hofle, M.G., Espejo, R.T. and Sanchez, I. and Martinez-Bueno, M. (2008) Polyphasic
Guzman, C.A. (2006) Quantitative reverse transcription approach to bacterial dynamics during the ripening of
polymerase chain reaction analysis of Vibrio cholerae cells Spanish farmhouse cheese, using culture-dependent and -
entering the viable but non-culturable state and starvation independent methods. Appl Environ Microbiol 74, 5662–
in response to cold shock. Environ Microbiol 8, 658–666. 5673.
Greenblatt, C.L., Baum, J., Klein, B.Y., Nachshon, S., Meena, L.S. and Rajni. (2010) Survival mechanisms of
Koltunov, V. and Cano, R.J. (2004) Micrococcus luteus - pathogenic Mycobacterium tuberculosis H37Rv. FEBS J
survival in Amber. Microb Ecol 48, 120–127. 277, 2416–2427.
Helmus, R.A., Liermann, L.J., Brantley, S.L. and Tien, M. Okada, Y., Tsuzuki, Y., Narimatsu, K., Sato, H., Ueda, T.,
(2012) Growth advantage in stationary-phase (GASP) Hozumi, H., Sato, S., Hokari, R. et al. (2013) 1,4-Dihydroxy-
phenotype in long-term survival strains of Geobacter 2-naphthoic acid from Propionibacterium freudenreichii
sulfurreducens. FEMS Microbiol Ecol 79, 218–228. reduces inflammation in interleukin-10-deficient mice with
Hicks, W.M. (2005) Recovery from long-term stationary phase colitis by suppressing macrophage-derived proinflammatory
and stress survival in Escherichia coli require the L- cytokines. J Leukoc Biol 94, 473–480.
isoaspartyl protein carboxyl methyltransferase at alkaline Oliver, J.D. (2009) Recent findings on the viable but
pH. Microbiology 151, 2151–2158. nonculturable state in pathogenic bacteria. FEMS
Houwen, F.P., Dijkema, C., Stams, A.J.M. and Zehnder, A.J.B. Microbiol Rev 34, 415–425.
(1991) Propionate metabolism in anaerobic bacteria; Parayre, S., Falentin, H., Madec, M.-N., Sivieri, K., Le Dizes,
determination of carboxylation reactions with 13C-NMR A.-S., Sohier, D. and Lortal, S. (2007) Easy DNA
spectroscopy. Biochim Biophys Acta 1056, 126–132. extraction method and optimisation of PCR-temporal
Isawa, K., Hojo, K., Yoda, N., Kamiyama, T., Makino, S., temperature gel electrophoresis to identify the
Saito, M., Sugano, H., Mizoguchi, C. et al. (2002) predominant high and low GC-content bacteria
Isolation and identification of a new bifidogenic growth from dairy products. J Microbiol Methods 69, 431–441.
stimulator produced by Propionibacterium freudenreichii Ramamurthy, T., Ghosh, A., Pazhani, G.P. and Shinoda, S.
ET-3. Biosci Biotechnol Biochem 66, 679–681. (2014) Current perspectives on viable but non-culturable
Jan, G., Rouault, A. and Maubois, J.-L. (2000) Acid stress (VBNC) pathogenic bacteria. Front Public Health 2, 103.
susceptibility and acid adaptation of Propionibacterium Rittershaus, E.S.C., Baek, S.-H. and Sassetti, C.M. (2013) The
freudenreichii subsp. shermanii. Le Lait 80, 325–336. normalcy of dormancy: common themes in microbial
Justice, S.S., Harrison, A., Becknell, B. and Mason, K.M. quiescence. Cell Host Microbe 13, 643–651.
(2014) Bacterial differentiation, development, and disease: Ruggirello, M., Dolci, P. and Cocolin, L. (2014) Detection and
mechanisms for survival. FEMS Microbiol Lett 360, 1–8. viability of Lactococcus lactis throughout cheese ripening.
Kolter, R., Siegele, D.A. and Tormo, A. (1993) The stationary PLoS ONE 9, e114280.
phase of the bacterial life cycle. Annu Rev Microbiol 47, Wen, J., Karthikeyan, S., Hawkins, J., Anantheswaran, R.C. and
855–874. Knabel, S.J. (2013) Listeria monocytogenes responds to cell
Le Mar
echal, C., Peton, V., Pl
e, C., Vroland, C., Jardin, J., density as it transitions to the long-term-survival phase. Int
Briard-Bion, V., Durant, G., Chuat, V. et al. (2015) J Food Microbiol 165, 326–331.
Surface proteins of Propionibacterium freudenreichii are Wilkinson, M.G., Guinee, T.P. and Fox, P.F. (1994) Factors
involved in its anti-inflammatory properties. J Proteomics which may influence the determination of autolysis of
113, 447–461. starter bacteria during cheddar cheese ripening. Int Dairy J
Lortal, S. and Chapot-Chartier, M.-P. (2005) Role, 4, 141–160.
mechanisms and control of lactic acid bacteria lysis in
cheese. Int Dairy J 15, 857–871.
Supporting Information
Lortal, S., Lem
ee, R. and Valence, F. (1997) Autolysis of Additional Supporting Information may be found in the
thermophilic lactobacilli and dairy propionibacteria: a online version of this article:
review. Le Lait 77, 133–150.
Loux, L., Mariadassou, M., Almeida, S., Chiapello, H., Figure S1 Kinetics of growth and survival measured by
Hammani, A., Buratti, J.J., Gendrault, A., Barbe, V. optical density in eight strains of Propionibacterium
et al. (2015) Mutations and genomic islands explain the freudenreichii
strain dependancy of sugar utilization among 21

Journal of Applied Microbiology 120, 432--440 © 2015 The Society for Applied Microbiology 439
P. freudenreichii survival F.F. Aburjaile et al.

Table S1 Kinetics of the growth and survival of eight freudenreichii strains measured by LIVE/DEAD staining
Propionibacterium freudenreichii strains measured by cul- followed by epifluroscence microscopy.
turability in Petri dishes.

Table S2 Proportion of nonpermeabilized cells during

the growth and survival of eight Propionibacterium

440 Journal of Applied Microbiology 120, 432--440 © 2015 The Society for Applied Microbiology