International Journal of Food Microbiology 301 (2019) 51–60

Contents lists available at ScienceDirect

International Journal of Food Microbiology

journal homepage: www.elsevier.com/locate/ijfoodmicro

Assessment of microbial communities on freshly killed wild boar meat by T

MALDI-TOF MS and 16S rRNA amplicon sequencing
⁎
M.F. Peruzya,b,c, N. Murrub, , Z. Yua,c, P.-J. Kerkhofa, B. Neolad, M. Joossensc, Y.T.R. Prorogae,
K. Houfa
a
Department of Veterinary Public Health and Food Safety, Faculty of Veterinary Medicine, Ghent University, Salisburylaan 133, 9820 Merelbeke, Belgium
b
Department of Veterinary Medicine and Animal Production, University of Naples “Federico II”, Via Delpino 1, 80137 Napoli, Italy
c
Laboratory of Microbiology, Department of Biochemistry and Microbiology, Ghent University, Karel Lodewijk Ledeganckstraat 35, Ghent, Belgium
d
Istituto Zooprofilattico Sperimentale del Mezzogiorno, Via Salute, 2, Portici, NA, Italy
e
Department of Food Microbiology, Istituto Zooprofilattico Sperimentale del Mezzogiorno, Via Salute, 2, Portici, NA, Italy

A R T I C LE I N FO A B S T R A C T

Keywords: Wild boars (Sus scrofa) are the most widely distributed large mammals and recent increase in consumption of
Wild boar meat wild boar meat urges the need of microbiological quality criteria. The aim of the study was to characterize the
MALDI-TOF MS initial bacterial contamination on freshly-killed wild boar meat using a culture-dependent approach with ISO-
16S rRNA amplicon sequencing methods combined with matrix-assisted laser desorption/ionization time-of-flight mass spectrometry identifi-
Salmonella spp.
cation and 16S rRNA amplicon sequencing. Moreover, the presence of foodborne pathogens was examined using
Real-Time-PCR and confirmed by classical isolation. Analysing 22 unrelated wild boar meat samples showed a
higher bacterial contamination level compared to pork, with Salmonella present in almost one third of the
samples. A great variability of the microbial contamination between the samples was recorded, as well as
complementary results between culturing and 16S rRNA amplicon sequencing as frequently isolated genera were
not always detected, and vice versa. Furthermore, the foodborne pathogen Salmonella was never detected with
16S rRNA amplicon sequencing, demonstrating the necessity for a cautious approach in the implementation of
new analysis techniques in food safety. The present work determines that attention should be paid to the trade of
non-inspected meat directly to retail or consumers.

1. Introduction diseases and predation by wolves, but nowadays, hunting has the
highest impact on the wild population density (Massei et al., 2015;
Wild boar (Sus scrofa) is currently considered one of the most de- Nores et al., 2008; Okarma et al., 1995; Toïgo et al., 2008).
structive and invasive mammal species in the world (Lowe et al., 2000). The meat of wild hogs eaten only by hunters and their families does
Wild boar are present on all continents, except for Antarctica (Sales not require any inspection whatsoever. According to Regulation (EC)
et al., 2017), and thrive in almost any condition, climate or ecosystem, No. 853/2004, large wild game hunted with the intention to sell on the
including urban areas (Keuling et al., 2018). In 2012, based on the market must be eviscerated (stomach and intestines) as soon as possible
hunting bag study, a total of 2.2 million wild boars were hunted in 18 after killing and, if necessary, be bled. The temperature of the carcasses
European countries (Massei et al., 2015). However, the number of free- must be brought down to a maximum of 7 °C and within a reasonable
range wild boars is increasing and an accurate estimation of the po- time frame. Moreover, the carcass and viscera must be examined on the
pulation density seems to be a difficult task (Keuling et al., 2018). Wild spot by a “trained person” in order to identify potential health risks.
hogs can cause significant economic losses as they may cause damage to This trained person must have sufficient knowledge of the anatomy,
crops and natural vegetation (Massei et al., 2015). Moreover, they can physiology, behaviour, pathology and further processing of wild game.
spread diseases to both livestock and humans, including African Swine Furthermore, no meat can be sold without passing veterinary inspection
fever, Salmonella spp., Shiga toxin-producing Escherichia (E.) coli and by a competent authority in a game-handling establishment. However,
Campylobacter (Keuling et al., 2018; Ruiz-Fons, 2017). The main causes depending on national legislation, the supply of a small number of wild
of natural mortality are starvation due to extreme weather conditions, game or small quantities of meat directly to the final consumer or to

⁎
Corresponding author.
E-mail address: [email protected] (N. Murru).

https://doi.org/10.1016/j.ijfoodmicro.2019.05.005
Received 11 October 2018; Received in revised form 6 May 2019; Accepted 6 May 2019
Available online 09 May 2019
0168-1605/ © 2019 Elsevier B.V. All rights reserved.
M.F. Peruzy, et al. International Journal of Food Microbiology 301 (2019) 51–60

local retail establishments does not require official post-mortem in- (Baer et al., 2013). Samples were transported to the lab at +2 °C, and
spection. examined within a maximum of 24 h after collection.
Nowadays, wild boar meat is considered as healthy and delicious
food for its high nutritional value and special sensory proprieties and is 2.2. Culture-dependent bacteriological examination
available throughout the year (Atanassova et al., 2008; Naya et al.,
2003; Strazdina et al., 2014). In Italy, consumption of wild boar in- Ten grams of meat and 90 ml (1:10 (W/W)) of sterilized Peptone
creased in recent years to approximately 0.2 kg per capita/year Water (PW, CM0009, OXOID, Basingstoke, UK) were placed in a sterile
(Pedrazzoli et al., 2017). Data on the consumption of game meat at a stomacher bag and homogenized for 3 min at 230 rpm using a peri-
European level are still limited. Furthermore, within the European staltic homogenizer (BagMixer®400 P, Interscience, Saint Nom, France).
Union, there are no specific microbiological criteria for wild boar meat Subsequently, ten-fold serial dilutions of each homogenate were pre-
products. Therefore, microbiological criteria for pork, included in pared in PW, followed by quantitative bacterial isolation in duplicate
Regulation (EC) No. 1441/2007 of 5 December 2007 amending Reg- for: a) total aerobic bacterial (TAB) counts performed according to ISO
ulation (EC) No. 2073/2005 on microbiological criteria for foodstuffs, 4833-2:2013 on Plate Count Agar (PCA; CM0325, Oxoid), incubated at
are commonly applied to assess microbiological quality (Atanassova 30 °C for 48 to72-h; b) psychotropic aerobic bacterial counts on PCA
et al., 2008). Microbial contamination of pork has already been studied incubated at 7 °C for 10 days (Ercolini et al., 2009); c) total anaerobic
extensively (Koo et al., 2016; Mann et al., 2016; Tian et al., 2017). bacterial counts (TANAB) on PCA by pour plating and incubation in an
Pseudomonas spp., Enterobacteriaceae, Brochothrix thermosphacta and anaerobic atmosphere (anaerobic GasPak jar system, Oxoid) at 30 °C for
lactic acid bacteria have been identified as the dominant bacterial taxa 48/72-h; d) lactic acid bacteria (LAB) according to ISO 15214:1998 on
present, contributing to meat flavouring but spoilage as well. In con- De man, Rogosa and Sharpe agar (MRS, CM0361, Oxoid) incubated
trast to domesticated pigs, wild boars roam free and their diet is un- aerobically at 30 °C for 72-h; e) presumptive Pseudomonas spp. ac-
controlled. Furthermore, when animals are shot, bled and eviscerated in cording to ISO 13720:2010 on Cephalothin-Sodium Fusidate-Cetrimide
the environment, there is no access to hygienic conditions like those Agar with Modified CFC Selective Supplement (CFC, CM0559B with
present in modern slaughterhouse facilities. The microbiological con- SR0103E, Oxoid) incubated aerobically at 25 °C for 48-h; f) E. coli ac-
tamination therefore depends on the circumstances in which the ani- cording to ISO 16649-2:2001 selectively isolated on Tryptone Bile X-
mals are killed (e.g. different hunting methods), dressed and further Glucuronide (TBX, CM0945, Oxoid) incubated at 44 °C for 24/48-h, and
handled from the collection to the chilling point. Various studies have g) total Enterobacteriaceae (EB) according to ISO 21528-2:2017 selec-
explored the microbiological quality of game meat, (Atanassova et al., tively isolated on Violet Red Bile Glucose Agar (VRBG, CM1082, Oxoid)
2008; Avagnina et al., 2012; Gill, 2007; Mirceta et al., 2015) but still incubated at 37 °C for 24-h.
little is known regarding the microbial communities present in wild After incubation, all colonies were counted and subsequently all
boar meat. picked from the agar plates with bacterial growth between 30 and
Conventional bacteriological isolation methods and identification 300 CFU/plate on PCA, and between 15 and 150 CFU/plate from all
using biochemical tests are laborious and biased by specific culture and others (Bernadin et al., 2018; Sutton et al., 2011). All harvested co-
laboratory conditions. Detection and identification of microorganisms lonies were subsequently subcultured on Tryptic Soy Agar (TSA,
using culture-independent approaches, such as 16S rRNA amplicon CM0131, Oxoid) or MRS and incubated in the appropriate conditions as
sequencing, are now commonly used to assess the microbial ecology of described above.
different foods like cheeses, seafood, meat and meat products (Gori For the detection of relevant foodborne pathogens, 25 grams por-
et al., 2013; Roh et al., 2009; Tian et al., 2017). The use of matrix- tions of each sample were homogenized once in 225 ml (1:10 (W/W))
assisted laser desorption/ionization time-of-flight mass spectrometry buffer peptone water (BPW, CM0509, Oxoid) and incubated at 37 °C for
(MALDI-TOF MS) in microbiology has revolutionized routine identifi- 24-h for the detection of Salmonella, in 225 ml Half Fraser broth (HF,
cation of a huge amount of isolates, allowing the exploration of abun- CM1053, Oxoid) and incubated at 30 °C for 24-h for the detection of
dances and relations in microbiota studies (Höll et al., 2016; Hilgarth Listeria (L.) monocytogenes, and once in 225 ml Peptone Sorbitol Bile
et al., 2018; Lagier et al., 2016). The application of MALDI-TOF MS in Broth (PSB, 17192, Sigma-Aldrich) for the detection of Yersinia (Y.)
food microbiology is rather preliminary due to the still clinical-oriented enterocolitica. The “iQ-Check Real-Time PCR Kits” were applied for the
database. Nevertheless the combination of MALDI–TOF MS and 16S detection of Salmonella (S.) (BR3578123, Bio-Rad, Hercules, CA, USA)
RNA sequencing has proven a promising approach in studying micro- and L. monocytogenes (BR3578124, Bio-Rad, Hercules, CA, USA), fol-
bial populations in food (Lagier et al., 2016; Peruzy et al., 2019). lowing manufacturer's recommendations. For the detection of Y. en-
Considering the increasing consumption of wild boar meat and the terocolitica, DNA was extracted using the Chelex-100-resin method
related public health risk for zoonotic pathogens, the aim of the present (1422822, Bio-Rad, Hercules, CA, USA) whereby two ml of each in-
study was to assess the microbial communities on hunted wild boar cubated homogenate was transferred into a two ml centrifuge tube and
meat using both a culture dependent and independent approach re- centrifuged for 10 min at 10,000×g at 4°C. The supernatant was dis-
sulting in more complete picture of initial microbial contamination. carded, the pellet re-suspended in 300 μl of 6% Chelex 100 by vor-
texing, and incubated for 20 min at 56°C and again for 8 min at 100°C.
2. Materials and methods The suspension was immediately chilled on ice for 1 min and cen-
trifuged for 5 min at 10,000 ×g at 4°C. In order to evaluate the pre-
2.1. Sampling sence of Y. enterocolitica 4/O:3 and biotype 1A, a SYBR green PCR-assay
was conducted, with the gene ystA as target for the pathogenic biotype
From October to December 2017, meat samples of 22 wild boars (Peruzy et al., 2017). Therefore, 3 μl of DNA extract was added to 22 μl
(W1 to W22), 10 males and 12 females, aged between 1 and 6 years, of PCR mix. The mastermix contained 12.5 μl of Qiagen QuantiTect
were collected in the province Campania, southern Italy (Table 1). On SYBR Green PCR Kit (1x), 100 nM of both primers ystA-F (5′-ATCGA
different occasions, the animals were shot by official hunters and im- CACCAATAACCGCTGAG-3′) and ystA-R (5′-CCAATCACTACTGACTTC
mediately bled in the field. They were brought to collection places GGCT-3′). To evaluate the presence of the biotype 1A, the presence of
where the evisceration and skinning were performed. Subsequently, the target gene ystB gene was examined (Peruzy et al., 2017). Three μl
approx. 100 cm2 meat was aseptically cut out the shoulder area (at least of DNA extract was added to 22 μl of PCR mastermix containing 12.5 μl
100 g each) and individually placed in sterile stomacher blender bags. of Qiagen QuantiTect SYBR Green PCR Kit (1x), 150 nM of both primers
This area was selected as it has been reported as more frequently ystB-F (5′-GTACATTAGGCCAAGAGACG-3′) and ystB-R (5′-GCAACAT
contaminated due to the inverted position of the pig carcass at slaughter ACCTCACAACACC-3′). The fluorescence of SYBR Green and the melting

52
M.F. Peruzy, et al. International Journal of Food Microbiology 301 (2019) 51–60

Table 1
Gender, weight (kg) and age (years) of the shooted animals, total bacterial counts (log CFU/g) in twenty two wild boar meat samples (W1–W22) on different media:
mesophilic bacteria on PCA (TAB 30°C), psychotropic bacteria on PCA (TAB 7 °C), anaerobic bacteria (TANAB 30 °C), E.coli on TBX, Enterobacterales on VRBG,
presumptive Pseudomonas spp. on CFC and Lactic Acid Bacteria (LAB) on MRS and number of isolates per sample picked from the different agar plates and
subsequently analysed by MALDI-TOF.
Gender Weight (kg) Age (years) TAB 30 °C TAB 7 °C TANAB 30 °C E. coli Enterobacterales spp. Pseudomonas spp. LAB Isolates
(log CFU/g) (log CFU/g) (log CFU/g) (log CFU/g) (log CFU/g) (log CFU/g) (log CFU/g) (n.)

W1 M 40 1 4.69 4.69 5.10 2.46 3.49 4.66 3.93 231

W2 M 25 1 6.05 3.76 4.24 1.96 2.57 3.56 0.00 163
W3 M 70 3 6.01 5.12 5.75 4.56 5.34 3.82 4.64 363
W4 F 60 2 3.61 2.56 4.46 3.67 2.86 0.00 2.66 74
W5 M 60 2 3.26 1.96 3.63 2.13 2.36 2.56 3.62 162
W6 F 30 1 5.55 3.12 5.36 4.26 5.44 2.80 3.41 176
W7 F 108 5 6.72 6.69 5.98 2.10 5.67 6.62 5.28 209
W8 M 125 5 5.44 4.75 4.37 2.11 3.92 2.59 2.66 182
W9 M 70 3 5.59 3.88 4.76 4.36 4.61 4.37 3.55 166
W10 F 60 2 6.03 5.54 4.67 2.41 4.69 5.16 3.07 164
W11 F 60 3 4.32 4.80 5.28 2.96 4.45 3.89 4.37 184
W12 F 40 1 6.51 5.34 5.72 3.49 5.57 4.88 5.61 156
W13 M 60 3 5.50 4.56 5.46 4.44 4.86 3.66 4.98 169
W14 F 65 4 4.08 3.06 3.03 2.96 3.41 3.04 3.36 130
W15 M 125 6 5.51 5.36 4.87 2.30 4.62 4.39 4.76 164
W16 F 30 1 5.74 5.05 5.34 2.21 4.93 3.75 5.28 211
W17 F 30 1 6.95 4.99 3.44 2.26 4.48 4.69 3.62 248
W18 F 53 2 5.73 3.37 5.10 1.66 0.00 2.28 0.00 74
W19 M 70 3 6.65 3.74 5.71 2.96 4.98 4.24 4.00 106
W20 F 40 1 5.85 4.04 3.88 1.26 3.64 3.26 3.04 166
W21 F 40 1 4.66 4.55 3.44 1.91 3.82 4.13 4.26 162
W22 M 40 1 6.67 6.58 5.21 3.47 5.05 5.74 4.14 129

curve were generated using the CFX96 system (Bio-Rad). A specific procedure from Bruker Daltonics (03.04.2006). Therefore, colonies
melting temperature (Tm) of 78.5 ± 1 °C indicated a positive result. were suspended in 800 μl of TSB or MRS broth and incubated at 28 °C
While awaiting the RT-PCR results, the enrichment broths were stored for 24 h. Subsequently, samples were centrifuged (1533g at 4 °C) for
at 4 °C. RT-PCR positive results for Salmonella spp., L. monocytogenes or 10 min, the supernatants discarded and the pellets washed twice in
Y. enterocolitica were confirmed using the corresponding normalized 500 μl of Milli-Q water and centrifuged (1533g at 4 °C) for 10 min. After
microbiological isolation methods ISO 6579-1:2017, 11290-1:2017, the second centrifugation, the supernatants were removed and the
and 10273:2017. pellets suspended in 100 μl of Milli-Q water. Next, 50 μl of formic acid
Salmonella isolates were sent to the Salmonella Typing Centre of the and 50 μl of acetonitrile were added and mixed thoroughly by pipetting
Campania Region (Department of Food Microbiology, Istituto followed by centrifugation (1533g at 4 °C) for 10 min. One microliter of
Zooprofilattico Sperimentale del Mezzogiorno, Portici, NA, Italy) for each sample was spotted onto a 96-spot plate and allowed to dry at
serotyping following the Kaufmann-White scheme (Le Minor and room temperature. Afterwards, 1 μl of matrix solution was added. The
Popoff, 1987). analysis was repeated when the spots resulted in ‘no peaks found’.
Isolates for which a MALDI-TOF MS score below 1.7 was obtained, were
imported into the BioNumerics 7.2.6 software (Applied Maths, Sint-
2.3. Isolate identification strategy
Martens-Latem, Belgium) to perform a dereplication in order to select
representatives for further analysis. For this, the Pearson correlation
All isolates, except those on MRS plates, were analysed by MALDI-
coefficient was applied and curve-based analysis was performed using
TOF MS first using the “direct colony identification method” (Alatoom
Unweighted Pair Group Method with Arithmetic Mean (UPGMA) clus-
et al., 2011). In brief, bacterial growth was smeared in duplicate onto a
tering algorithm. Based on dendrogram distance level settings and best
96-spotsteel plate (Bruker Daltonics, Bremen, Germany) and allowed to
matches ranking, representative isolates were selected for subsequent
dry at room temperature. Subsequently, the sample was covered with
16S rRNA gene sequencing. DNA was extracted using alkaline lyses
1 μl matrix solution containing 10 mg/ml α-cyano-4-hydroxycinnamic
where one colony was suspended into 20 μl of lysis buffer (2.5 ml 10%
acid in acetonitrile, deionized water, and trifluoracetic acid
SDS, 5 ml 1 N NaOH and 92.5 ml Milli-Q water) and heated for 15 min
(50:47.5:2.5, [vol/vol/vol]). Bruker's Bacterial Test Standard (BTS
at 95 °C. After a short spin, 180 μl of Milli-Q water was added. Subse-
Bruker Daltonics) was used as mass calibration and reference standard
quently, the suspension was centrifuged for 5 min at 10,000×g at 4 °C.
in each series of MALDI measurements. Mass spectra were generated
To amplify part of the 16S rRNA gene, the oligonucleotide primers pA
with the Microflex™ LT MALDI-TOF mass spectrometer, equipped with
(5′-AGA GTT TGA TCC TGG CTC AG-3′) and pH (5′-AAG GAG GTG ATC
a nitrogen laser (l1/4337 nm) and Flex Control 3.4 software using re-
CAG CCG CA-3′) were used. The PCR mixture (final volume 25 μl)
commended settings in a linear positive ion detection mode (Bruker
contained 2.5 μl template DNA, 0.25 μl of each primer at a concentra-
Daltonics). Identifications were obtained by comparing the mass
tion of 10 μM, 2.5 μl of each deoxynucleoside triphosphate at a con-
spectra to the Bruker MSP database (MBT Compass Library, 5989 en-
centration of 2 μM each, 0.5 μl AmpliTaq DNA polymerase (1 U/μl) and
tries) using the Bruker Compass software at default settings. Identifi-
16.5 μl of Milli Q water. PCR conditions consisted of 30 cycles. Ampli-
cation score criteria were classified according to Jeong et al. (2016): a
cons were collected and submitted to Eurofins for Sanger sequencing.
score of ≥2.3 indicated highly probable species identification, between
Taxonomic identity was assessed using the nucleotide BLAST algorithm
2.0 and 2.3 secure genus and probable species identification, a score
as implemented within the NCBI web service (https://www.ncbi.nlm.
between 1.7 and 1.99 probable genus and < 1.7 non-reliable identifi-
nih.gov).
cation. Isolates for which a score of < 2.0 was obtained with the “direct
colony method” and all isolates from the MRS agar plates (Alatoom
et al., 2011) were analysed using the ethanol/formic acid extraction

53
M.F. Peruzy, et al. International Journal of Food Microbiology 301 (2019) 51–60

Table 2 2.5. Statistical analysis

Presence of foodborne bacteria in wild boar meats detected using RT-PCR and
Salmonella serotypes identified by using the Kaufmann-White scheme. To compare the bacterial counts, a one-way analysis of variance
Samples Salmonella spp. Salmonella L. monocytogenes Y. enterocolitica (ANOVA) was calculated using PAST software package (https://folk.
(S.) serotypes uio.no/ohammer/past/). PAST software was also used to calculate the
ystA ystB richness expressed by the Chao1 index and diversity indices of com-
munity information obtained from 16S rRNA amplicon sequencing. A
W1 N.D. N.D. N.D. 78.5b
W2 N.D. N.D. N.D. N.D. probability value of < 0.05 (p < 0.05) was defined as statistically
W3 N.D. N.D. N.D. N.D. significant.
W4 N.D. N.D. N.D. N.D.
W5 N.D. N.D. N.D. N.D.
3. Results
W6 N.D. N.D. N.D. N.D.
W7 N.D. N.D. N.D. N.D.
W8 26.25a Monophasic S. N.D. N.D. N.D. 3.1. General bacterial indicators
Typhimurium
W9 N.D. N.D. N.D. 78.5b Counts of the different bacterial parameters are shown in Table 1.
W10 N.D. N.D. N.D. 78.5b
Total aerobic bacteria determined at 30 °C ranged from 3.26 to 6.95 log
W11 N.D. N.D. N.D. N.D.
W12 29.25a S. kasenyi N.D. N.D. N.D. CFU/g (mean ± SE = 5.5 ± 0.22 log CFU/g) and from 1.96 to 6.69
W13 N.D. N.D. N.D. N.D. log CFU/g (mean ± SE = 4.43 ± 0.25 log CFU/g) for the determi-
W14 33.78a S. stanleyville N.D. N.D. N.D. nation at 7 °C. Within each sample, the viable aerobic count determined
W15 31.59a S. stanleyville N.D. N.D. N.D.
at 30 °C and 7 °C showed significant difference (p < 0.5). Total anae-
W16 N.D. N.D. N.D. N.D.
W17 26.11a S. kasenyi N.D. N.D. N.D.
robic bacteria ranged from 3.03 to 5.98 log CFU/g (mean ±
W18 N.D. N.D. N.D. N.D. SE = 4.76 ± 0.18 log CFU/g), and the mean number of presumptive
W19 37.86a NI N.D. N.D. N.D. Pseudomonas spp. and lactic acid bacteria were 3.82 ± 0.29 and
W20 N.D. N.D. N.D. N.D. 3.65 ± 0.31 log CFU/g respectively. The count of typical blue E. coli
W21 25.78a S. kasenyi N.D. N.D. N.D.
colonies on TBX and the purple/pink Enterobacterales colonies on VRBG
W22 N.D. N.D. N.D. N.D.
showed a mean of 2.81 ± 0.21 and 4.12 ± 0.28 log CFU/g respec-
N.D. Not detected. tively. No significant correlation was observed between the bacterial
N.I. Not isolated. count and the gender, age and weight of the animals.
a
Ct: threshold cycle.
b
Tm: melting temperature. 3.2. Presence of foodborne bacteria

2.4. Culture independent community profiling by 16S rRNA amplicon Salmonella was detected in 7 samples (31.82%) and all, except
sequencing sample W19, were confirmed using the reference isolation method
(Table 2). Three serovars were identified (monophasic S. Typhimurium,
To independently culture and identify microorganisms present in S. Stanleyville and S. Kasenyi). Regarding Y. enterocolitica, the gene ystA
wild boar meat, DNA was extracted from 1.8 ml of each initial sto- was not detected in the samples, though ystB was present in W1, W9
machered sample using the PowerFood Microbial DNA Isolation kit and W10. Also L. monocytogenes was never detected (Table 2).
(Qiagen, Germany) following manufacturer's recommendations. The
DNA quantity was measured using the dual-channel Quantus™ 3.3. Community profiling using isolation and MALDI-TOF MS identification
Fluorometer (Promega USA) and for purity, the ratio of absorbance at
260 nm and 280 nm was evaluated with a NanoDrop™ 2000/2000c A total of 3789 bacterial isolates were analysed by MALDI-TOF MS
Spectrophotometer (Thermo Fisher Scientific Inc., Waltham, MA, USA). (Table 1), of which 956 were picked from the PCA plates incubated at
DNA extracts of ≥100 ng, concentration ≥ 5 ng/μl and volume ≥ 20 μl, 30 °C, 526 from PCA at 7 °C, 240 from PCA incubated anaerobically,
(OD260/280 = 1.8–2.0) were submitted to Novogene (HK) Company 701 from VRBG plates, 582 from CFC plates, 452 from TBX and 332
Limited for 16S rRNA amplicon sequencing (https://en.novogene.com). from the MRS agar plates. With the “direct colony identification
DNA was used to construct the library in which V3-V4 amplicons were method” applied on 3457 colonies (=all except isolates on MRS),
amplified with primers 341F (5′-CCTAYGGGRBGCASCAG-3′) and 806R 19.79% (n = 684) were identified at a highly probable species level,
(5′-GGACTACNNGGGTATCTAAT-3′). Truseq-DNA-PCR-free-library- 36.71% (n = 1269) at secure genus and probable species level, 18.48%
prep kit was used to construct the DNA libraries of paired ends with (n = 639) at genus level and 25.02% (n = 865) did not yield any
single index. Amplicons were mixed in equimolar amounts and se- identification. Colonies with a score value < 2 (n = 1504) were re-ex-
quenced on the Illumina2500 platform with Sequencing strategy amined using “the extraction method” of which 7.65% (n = 115) were
PE250. additionally identified at species level, 32.85% (n = 494) at genus level
Qiime2 (version 2018.6) software pipeline (https://qiime2.org) was and 36.37% (n = 547) at probable genus level but 23.14% (n = 348)
used for data analysis. Reads were demultiplexed with q2-demux remained unidentified. With this second extraction analysis method,
(https://github.com/qiime2/q2-demux). Then q2-dada2 plugin was about 67.91% of the isolates (n = 1022) obtained a higher MALDI-TOF
implemented for the quality control process and all phiX reads and MS score value then by the direct colony identification method but
chimeric sequences were filtered. Based on demux summary, sequences 11.23% (n = 169) scored values lower than with the previous direct
of 244 bases of both forward and reverse reads were truncated. After colony identification method. The same score value was obtained with
denoising the data using the dada2 denoise-paired method, re- both methods in 20.80% (n = 313) of the isolates. Interestingly, the
presentative sequences of each sample were retained. Before assigned identification outcome at species level of 182 isolates (179 Pseudo-
to taxa, bacterial representative sequences were sifted out by against monas, two Raoultella, one Staphylococcus) differed between the two
with Greengene 99% based on 97% identity using method vsearch. analysis methods.
Then, both original representative sequences and filtered sequencing Analysis of the isolates on MRS agar plates (n = 332) was per-
were assigned to taxa using Naive Bayes classifiers pre-trained on formed only with the “extraction method” and resulted in the identi-
Greengenes 13_8 99% OTUs full-length sequences. (https://docs. fication at species level, secure genus and probable genus level in
qiime2.org/2017.12/data-resources/). 27.11% (n = 90), 28.01% (n = 93), and 28.31% (n = 94) respectively.

54
M.F. Peruzy, et al. International Journal of Food Microbiology 301 (2019) 51–60

100%

90%

80%
Percentage of genera identified

70%

60%

50%

40%

30%

20%

10%

0%
TAB 30°C TAB 7°C TANAB CFC TBX VRBG MRS

Fig. 1. Most abundant genera identified by MALDI-TOF (≥5%) and 16S gene sequencing on different agars: mesophilic bacteria on PCA (TAB 30 °C), psychotropic
bacteria on PCA (TAB 7 °C), anaerobic bacteria (TANAB), E.coli on TBX, Enterobacterales on VRBG, presumptive Pseudomonas spp. on CFC and Lactic Acid Bacteria
(LAB) on MRS.

No identification was obtained for 55 isolates (16.57%). were however also isolated and displayed the same phenotypic features.
Isolates without a reliable MALDI-TOF MS identification were On the MRS plates, Leuconostoc and Candida were the genera most
analysed using 16S rRNA gene sequencing (n = 118 isolates) and two frequently isolated (59.09%) followed by the Gram positive Lactococcus
families (Comamonadaceae and Deinococcaceae) and nine genera (36.36%).
(Acidovorax, Deinococcus, Exiguobacterium, Frigoribacterium, Gibbsiella,
Micrococcus, Okibacterium, Psychrobacter and Rothia) were additionally
identified. 3.4. Culture-independent community analysis
The dominant genera (with a percentage ≥ 5%) isolated on the
different media is shown in Fig. 1. In general, bacteria identified by Using 16S rRNA amplicon sequencing, a total of 1,209,441 reads
MALDI-TOF MS could be assigned to 28 families and 44 genera. were obtained. After filtering out all non-bacterial and non-archaea
Moreover, the yeasts Candida spp., Cryptococcus spp. Rhodotorula spp. sequences, using the quality-control plugin in QIIME2, 535,513 reads
and Yarrowia were also isolated. Members of genera Acinetobacter, En- were taken into account as bacterial reads for further analysis. Only
terobacter, Enterococcus, Escherichia, Lactococcus, Lelliottia, Leuconostoc, exact amplicon sequence variants (OTUs) accounting for > 0.5% of the
Macrococcus, Pantoea, Pseudomonas and Rahnella were isolated in at total reads of each sample were further retained. These sequences were
least 50% of the samples (Supplementary material 1). In particular, clustered into 112 genera encompassing 64 families (Supplementary
Pseudomonas was present in 17 samples (77.27%) among the mesophilic material 3). However, groups of bacteria were identified only at family
and psychrotrophic population on PCA plates. At 30 °C, Macrococcus or even at a higher taxonomic level and 3.5% of the sequences were
and Acinetobacter were also frequently isolated and, in particular, they identified only at kingdom level. The relative abundance of each OTU
were present in 54.55% and 45.45% of the samples respectively. When varied among the 22 samples (Fig. 2). Richness expressed by the Chao1
incubation at 7 °C was applied, Pantoea was frequently isolated (72.73% index showed sample W10 as that with the lowest richness and sample
of the samples). Except for one Propionibacterium isolate in sample W10, W5 as the richest sample. The equitability (evenness) index showed that
anaerobic incubation showed no additional contribution as all bacteria no genera clearly dominate (Table 3) the bacterial communities on wild
present anaerobically turned out to be facultative anaerobic and were boar meat, though the genera Acinetobacter and Pseudomonas were de-
also isolated in aerobic condition. In particular, on PCA plates in- tected in all samples. Macrococcus and Propionibacterium were present in
cubated anaerobically, the genus Escherichia was isolated in 59.09% of 90.91%, Pediococcus and Enterobacter in 86.36%, Escherichia in 77.28%,
the samples (13 samples). Faecalibacterium and Oscillospira both at 72.73%, Bacteroides, Prevotella
Concerning the specificity of selective isolation media on CFC and Roseburia in 68.18%, Janthinobacterium and Rahnella in 63.64%,
plates, besides the dominant presence of Pseudomonas spp., other Brochothrix and Streptococcus in 59.09%, Pseudobutyrivibrio and Ci-
genera were isolated or detected by 16S rRNA gene sequencing trobacter in 54.55% and Fusobacterium and an uncharacterized member
(Supplementary material 2) but except for the genus Shewanella, all of of the Dethiosulfovibrionaceae family in 50% of the samples.
them were oxidase negative. On the TBX plates all typical blue colonies Comparing both analytical approaches, among the 22 samples
were confirmed as E. coli. On the VRBG plates, 98.72% of the purple/ Pseudomonas was always detected and isolated. However, it resulted as
pink colonies were confirmed as members of the order Enterobacterales, a dominant genus with the culture-dependent approach but not with
with Escherichia present in 72.73% and Patoea in 63.64% of the sam- culture-independent one. Acinetobacter was detected and isolated in
ples. Members of the genera Pseudomonas and the yeast Rhodotorula 81.82% of the samples and only detected in 5 (W9, W12, W16, W18 and
W21). 86 genera were only detected using 16S amplicon sequencing

55
M.F. Peruzy, et al. International Journal of Food Microbiology 301 (2019) 51–60

100%

90%

80%

70%
Relative abundance (%)

60%

50%

40%

30%

20%

10%

0%
W1 W2 W3 W4 W5 W6 W7 W8 W9 W10 W11 W12 W13 W14 W15 W16 W17 W18 W19 W20 W21 W22

Fig. 2. Top 10 abundant microbial communities identified in wild boar meat samples by 16S rRNA amplicon sequencing.
The relative proportion of the most abundant microbial communities identified upon the highest taxonomic level (o: order; f: family; g: genus.)

(Table 4). In contrast, 28 genera were only isolated and not detected totally surprising as the animals were not slaughtered and further
(Table 4). Even when OTUs with < 0.5% of total reads were taken into processed in modern slaughter facilities. Total aerobic bacteria levels
account, only 6 genera (Arthrobacter, Erwinia, Exiguobacterium, Leuco- were in line with those previously reported for wild boar meat (Mirceta
nostoc, Serratia and Weissella) could be additionally obtained. et al., 2015), but levels of Enterobacterales, and E. coli in particular, were
higher than in other studies (Membré et al., 2011; Mirceta et al., 2015).
This identifies a faecal-meat or faecal-skin-meat transmission during
4. Discussion processing and an intrinsic risk for public health. This is also under-
pinned by the high presence (31.82%) of Salmonella in the samples. The
In the present study, though not completely comparable due to the correlation of lower levels of Enterobacteriaceae resulting in less Sal-
larger region that has to be sampled on pork carcasses, levels of the monella positive wild boar meat samples was demonstrated in the stu-
total aerobic bacteria and Enterobacteriaceae, both indicators commonly dies of Mirceta et al. (2017) where Salmonella was detected in only
applied in hygiene evaluation, were respectively higher in 16 and 18 1.9% of the samples, in 2.15% positive Japanese wild boar meat sam-
wild boar meat samples than the EU microbiological criteria for pork ples at retail (Kanai et al., 1997) and in the studies of Atanassova et al.
(Aerobic colony count: m < 4.0 log CFU/cm2 and M < 5.0 log CFU/ (2008) and Paulsen and Winkelmayer (2004) where Salmonella was not
cm2; Enterobacteriaceae m < 2.0 log CFU/cm2 and M < 3.0 log CFU/ detected in any of the 127 and 50 meat samples respectively. However,
cm2) (Regulation 1441/07). These higher contamination levels are not

56
M.F. Peruzy, et al. International Journal of Food Microbiology 301 (2019) 51–60

Table 3 Typhimurium, S. Stanleyville and S. Kasenyi. Monophasic S. Typhi-

Richness (expressed by Chao1 index) and diversity (expressed by Shannon and murium represents one of the five most reported serovars in human
Evenness indexes) of the bacterial communities identified by 16 S amplicon salmonellosis (EFSA, 2017). Serovar Salmonella Kasenyi has not been
sequencing. reported from wild boar meat yet.
Chao-1 Shannon Evenness Pigs are considered a major reservoir for pathogenic Y. en-
terocolitica, but this foodborne pathogen was not isolated in the present
W1 239 7.42478 0.939742
study. Only the gene ystB was detected in 13.63% of the samples, in-
W2 247 7.570253 0.952429
W3 199 7.111292 0.931209
dicating the presence of non-pathogenic strains (biotype 1A) (Peruzy
W4 342 7.271982 0.863876 et al., 2017). This finding is consistent with previous wild boar meat
W5 348 8.027176 0.950756 studies in which only non-pathogenic Y. enterocolitica strains were re-
W6 212 6.710971 0.868406 covered (Gill (2007), Avagnina et al. (2012)). Also the absence of L.
W7 285 7.738421 0.948938
monocytogenes is in accordance with other studies (Membré et al., 2011;
W8 228 5.740676 0.732894
W9 146 6.573831 0.914324 Paulsen and Winkelmayer, 2004).
W10 115 6.257811 0.914151 For the identification of the different bacterial colonies with MALDI-
W11 286 7.534417 0.92335 TOF MS, the “direct identification colony method” and the “extraction
W12 302 7.691718 0.933642
method” were used. With the latter method, the individual score value
W13 231 6.892995 0.877893
W14 282 6.768272 0.831529
increased for almost 40% of the isolates. However, for almost 90% of
W15 360 8.087605 0.952396 the isolates, the identification obtained with the extraction method did
W16 270 6.826319 0.845175 not differ, not even at species level.
W17 299 6.672652 0.811363 Combining the two methods, 89.93% of the isolates were identified
W18 254 6.223689 0.779063
at genus level of which 23.11% were identified at species level. This
W19 189 6.945976 0.918507
W20 134 6.693117 0.947217 effect has also been reported by Alatoom et al. (2011) showing that
W21 250 7.449607 0.935201 “extraction” resulted in higher level identifications than with the “di-
W22 186 4.862044 0.644905 rect colony method”. However, identification of some microorganisms,
in particular Pseudomonas species, remains unreliable since different
species identification was obtained after using the direct identification
wild hogs acting as a carrier of Salmonella have been demonstrated by colony method and extraction method in 182 isolates. Rapid and reli-
its isolation from 24.8% of wild boar intestinal samples, 10.8% of wild able identification of Pseudomonas isolates remains however a challenge
boar faeces and from the surfaces and the deep tissues of 4.2% samples in many studies due the high number of species and variable taxonomy
in three previous Italian studies (Chiari et al., 2013; Zottola et al., 2013; (Mulet et al., 2012). Analysis of the isolates present on MRS was per-
Decastelli et al., 1995). Besides slaughter procedure, these differences formed only with the extraction method as it has been reported that this
may be related to the fact that Salmonella prevalence can vary between method is more efficient for this group of bacteria (Alatoom et al.,
regions and the time of the year (Gill, 2007). Furthermore, Chiari et al. 2011).
(2013) showed that wild boars can host of a variety of Salmonella ser- Among the isolates without a reliable MALDI-TOF MS identification
ovars. In the present work, three serovars were isolated: monophasic S.

Table 4
Genera only detected using 16S amplicon sequencing (Culture independent method) and genera only and identified with MALDI TOF-MS and 16S gene sequencing
(Culture dependent methods).
Culture independent Culture dependent

Actinobacillus Geobacillus Rhodoplanes Aerococcus

Agrobacterium Gluconacetobacter Roseburia Aeromonas
Akkermansia Granulicatella Rubellimicrobium Agrococcus
Anaerolinea HA73 Rubrobacter Arthrobacter
Anaerovorax Haemophilus Ruegeria Buttiauxella
Asticcacaulis Inquilinus Ruminococcus (Lachnospiraceae) Cronobacter
Bacteroides Kocuria Ruminococcus (Ruminococcaceae) Curtobacterium
Balneimonas Lachnospira Rummeliibacillus Erwinia
Bifidobacterium Luteimonas Shuttleworthia Exiguobacterium
Blautia Massilia SMB53 Filobasidium
Bradyrhizobium Megamonas Sphingobium Frigoribacterium
Burkholderia Megasphaera Sphingomonas Gibbsiella
Caldicellulosiruptor Methylobacterium Succinivibrio Hafnia
Campylobacter Methyloversatilis Sutterella Kluyvera
Candidatus Solibacter Morganella Syntrophomonas Kurthia
Catenibacterium Nocardioides Syntrophus Leclercia
CF231 Oscillospira T78 Lelliottia
Clostridium (Clostridiaceae) Paenibacillus Tepidimonas Leucobacter
Clostridium (Lachnospiraceae) Parabacteroides Treponema Leuconostoc
Clostridium (Ruminococcaceae) Paracoccus Turicibacter Micrococcus
Comamonas Pediococcus vadinHB04 Okibacterium
Coprococcus Pedobacter Vagococcus Pantoea
Corynebacterium Phascolarctobacterium Veillonella Pseudoclavibacter
Dermabacter Phenylobacterium Vibrio Raoultella
Devosia Photobacterium Wohlfahrtiimonas Rhodotorula
Dietzia Prevotella YRC22 Rothia
Dorea Propionicimonas Serratia
Enhydrobacter Providencia Weissella
Faecalibacterium Pseudobutyrivibrio
Fusobacterium Pseudonocardia

57
M.F. Peruzy, et al. International Journal of Food Microbiology 301 (2019) 51–60

and further analysed with 16S gene sequencing, three genera were still Bacteroides, Faecalibacterium, Fusobacterium, Oscillospira, Roseburia,
not present in the Bruker database (Frigoribacterium, Gibbsiella and Prevotella and Pseudobutyrivibrio, bacteria frequently detected with 16S
Okibacterium). Six others were already included but were not identified. amplicon sequencing but not isolated, were all obligate anaerobic, in-
An explanation for the latter can be that spectra generated for certain dicating that the current commonly applied anaerobic culture method
species can differ between isolates due to media and growth conditions using Gaspak did not support bacterial recovery.
(Williams et al., 2003). Other discording result between culturomics and 16S amplicon se-
Applying culture independent 16S amplicon sequencing, the re- quencing occurred for the genus Pantoea belonging to the Erwiniaceae
lative abundance of each OTU varied among the samples and there family that was frequently isolated but never detected. Moreover, also
were no same genera or even same families dominating the different the percentage of isolation and detection of the genus Pseudomonas also
samples. differs between the two approaches. It could be explained by the fact
Pseudomonas spp. and Acinetobacter spp. were detected in all sam- that organisms with high growth rates, even if they are present with a
ples and are indicative of a soil or water contamination (Chaillou et al., smaller percentage in an ecosystem may overlay on organisms present
2015). Members of the genera occur commonly on fresh meat and were with a higher number but with low growth rates (Großkopf et al.,
also abundant in the Japanese study of Asakura et al. (2017). Pseudo- 1998). It is certain that culture-dependent methods are biased due to
monas present on meat stored under refrigerated conditions often be- the media and the condition used but the direct 16S rRNA analysis also
comes the dominant bacterial flora and causes spoilage (Doulgeraki seems biased according to the primer used, variable region selection,
et al., 2012), while Acinetobacter contributes more to the breakdown of amplicon size and the number of PCR cycles (Fuks et al., 2018; Knight
food components and may produce off-odours (Rawat, 2015). et al., 2018). Also Salmonella, detected by RT- PCR and confirmed by
Macrococcus and Propionibacterium were also frequently detected isolation, was not detected using 16S amplicon sequencing applied on
with 16S amplicon sequencing (90.9%) but while Macrococcus was the first meat homogenate. It is certain that the portions analysed with
isolated from 14 samples of which 12 samples on plates incubated at RT-PCR have not been the same portions as for the other analysis and
30 °C and in other two samples on the plates incubated anaerobically, moreover, an enrichment step was used for the pathogen detection.
Propionibacterium was only isolated from one sample. The latter is more However it has been previously reported that these bacteria usually are
associated with dairy products, silages and vegetables, ruminal content present at a very low concentration and only deep Next Generation
and faeces, human stools and faeces from poultry and pigs (Freitas Sequencing seems to overcome this problem (Rouger et al., 2017). Al-
et al., 2015). Instead, macrococci are typically isolated from animal though the aim of this study was to characterize for the first time the
skin and food of animal origin (Baba et al., 2009). Pediococcus, a lactic microbial diversity and richness in fresh wild boar meat samples,
acid bacteria commonly present in the mammalian gut (Araya et al., taxonomic resolution could be improved in follow up studies by using,
2002), was detected in 19 samples but never isolated, probably because for example, full length 16S amplicon sequencing (Martínez-porchas
the MRS plates were incubated aerobically and without the addition of et al., 2016). Nevertheless, even when applying improved methods for
inhibiting antibiotics (Simpson et al., 2006). taxonomic resolution, it is important to keep the impact of the DNA
The high rate of detection of members of the Enterobacterales order extraction protocol on the results in mind (Yu et al., 2019). Moreover,
(Enterobacter, Escherichia and to a slightly lesser extent, Rahnella and in the present research, it is confirmed again that culture-dependent
Citrobacter) provide again, along with the high contamination levels and independent approaches are complementary in the study of the
described above, evidence of a faecal contamination of the samples. complex relationship between microorganisms in a food as already
Moreover they are bacterial groups that contain human pathogens reported by Lagier et al. (2018) and Peruzy et al. (2019).
(Leight et al., 2018). However, even if in the present study a selective In conclusion, in the present study, the initial bacterial con-
media for the enumeration of Enterobacterales was used, the isolation tamination on wild boar meat was studied, and showed to be a greater
rate of Enterobacter Citrobacter and Rahnella was low. public health risk compared to pork. The need for better slaughter
In almost 70% of the wild boar meat samples two genera belonging hygiene was demonstrated and the non-inspected distribution of meat
to the family of Ruminococcaceae (Faecalibacterium and Ocillospira) were directly to consumers is not recommended. Although often claimed as
present. Particularly Ruminococcaceae is one of the most abundant fa- such, culture-independent analysis proved not to be a reliable alter-
milies from the order Clostridiales present in the mammalian gut en- native as it fails to detect the foodborne pathogen Salmonella. Only in
vironment and has been associated with the maintenance of gut health the study of bacterial communities, a combination of both approaches
(Biddle et al., 2013). might be relevant.
Two genera belonging to the order of Bacteroidales (Bacteroides Supplementary data to this article can be found online at https://
(Bacteroidaceae family) and Prevotella (Prevotellaceae family)) and one doi.org/10.1016/j.ijfoodmicro.2019.05.005.
genus belonging to the order of Clostridiales (Roseburia (Lachnospiraceae
family)) were detected in 15 samples applying 16S amplicon sequen- Acknowledgements
cing. None of the three were retrieved by cultivation. Bacteroides and
Roseburia have been detected in the faeces of a variety food animals This research did not receive any specific grant from funding
(Park et al., 2014; Young Ko et al., 2018) and Prevotella has been found agencies in the public, commercial, or not-for-profit sectors.
in the gut microbiota of ruminants and chimpanzees (Holman et al.,
2017). Janthinobacterium, Brochothrix and Streptococcus were detected References
in around 60% of the samples. Janthinobacterium was once identified in
water, soil and spoiled raw milk (Chaillou et al., 2015; Liu et al., 2018; Alatoom, A.A., Cunningham, S.A., Ihde, S.M., Mandrekar, J., Patel, R., 2011. Comparison
Molin, 2000) but, the role of this genus in the spoilage process is still of direct colony method versus extraction method for identification of gram-positive
cocci by use of Bruker biotyper matrix-assisted laser desorption ionization-time of
unknown (Liu et al., 2018). Instead, the species Brochothrix thermo- flight mass spectrometry. J. Clin. Microbiol. 49, 2868–2873. https://doi.org/10.
sphacta, known as common meat contaminant, is considered one of the 1128/JCM.00506-11.
major cause of spoilage (Rouger et al., 2017). However, Janthino- Araya, M., Morelli, L., Reid, G., Sanders, M.E., Stanton, C., Pineiro, M., Ben Embarek, P.,
2002. Guidelines for the evaluation of probiotics in food. Jt. FAO/WHO Work. Gr.
bacterium and Brochothrix were isolated on PCA plates incubated at 7 °C Rep. Draft. Guidel. Eval. Probiotics Food 1–11. https://doi.org/10.1111/j.1469-
only in one and three samples respectively. Streptococcus was isolated 0691.2012.03873.
from four samples on PCA incubated at 7 °C and also in anaerobic Asakura, H., Kawase, J., Ikeda, T., Honda, M., Sasaki, Y., Uema, M., Kabeya, H.,
Sugiyama, H., Igimi, S., Takai, S., 2017. Microbiological quality assessment of game
conditions and MRS incubated aerobically. This genus contains several meats at retail in Japan. J. Food Prot. 80, 2119–2126. https://doi.org/10.4315/0362-
species that are part of the normal flora of human and animals but some 028X.JFP-17-137.
are also pathogenic (Lu et al., 2014; Lun et al., 2007). Atanassova, V., Apelt, J., Reich, F., Klein, G., 2008. Microbiological quality of freshly shot

58
M.F. Peruzy, et al. International Journal of Food Microbiology 301 (2019) 51–60

game in Germany. Meat Sci. 78, 414–419. https://doi.org/10.1016/j.meatsci.2007. Fournier, P.E., Levasseur, A., Raoult, D., 2016. Culture of previously uncultured
07.004. members of the human gut microbiota by culturomics. Nat. Microbiol. 1. https://doi.
Avagnina, A., Nucera, D., Grassi, M.A., Ferroglio, E., Dalmasso, A., Civera, T., 2012. The org/10.1038/nmicrobiol.2016.203.
microbiological conditions of carcasses from large game animals in Italy. Meat Sci. Lagier, J.-C., Dubourg, G., Million, M., Cadoret, F., Fournier, P., 2018. Culturing the
91, 266–271. https://doi.org/10.1016/j.meatsci.2012.01.025. human microbiota and culturomics. Nat. Rev. Microbiol. 16, 540–550. https://doi.
Baba, T., Kuwahara-Arai, K., Uchiyama, I., Takeuchi, F., Ito, T., Hiramatsu, K., 2009. org/10.1038/s41579-018-0041-0.
Complete genome sequence of Macrococcus caseolyticus strain JSCS5402, reflecting Le Minor, L., Popoff, M.Y., 1987. Designation of Salmonella enterica. Int. J. Syst.
the ancestral genome of the human-pathogenic staphylococci. J. Bacteriol. 191, Bacteriol. 37, 465–468.
1180–1190. https://doi.org/10.1128/JB.01058-08. Leight, A.K., Crump, B.C., Hood, R.R., 2018. Assessment of fecal indicator bacteria and
Baer, A.A., Miller, M.J., Dilger, A.C., 2013. Pathogens of interest to the pork industry: a potential pathogen co-occurrence at a shellfish growing area. 9, 1–13. https://doi.
review of research on interventions to assure food safety. Compr. Rev. Food Sci. Food org/10.3389/fmicb.2018.00384.
Saf. 12, 183–217. https://doi.org/10.1111/1541-4337.12001. Liu, X., Li, D., Li, K., Luo, Y., 2018. Monitoring bacterial communities in ε-polylysine-
Bernadin, K.K., Witabouna, K.M., Julien, C.K., 2018. Microbial Contamination of the Stem treated bighead carp (Aristichthys nobilis) fillets using culture-dependent and cul-
Bark of Mitragyna Ciliata, A Commercially Available Medicinal Plant in the District of ture-independent techniques. J. Food Microbiol. 76, 257–266. https://doi.org/10.
Abidjan (Cote d' Ivoire). 5. pp. 404–415. 1016/j.fm.2018.06.001.
Biddle, A., Stewart, L., Blanchard, J., Leschine, S., 2013. Untangling the genetic basis of Lowe, S., Browne, M., Boudjelas, S., De Poorter, M., 2000. 100 of the world's worst in-
fibrolytic specialization by lachnospiraceae and ruminococcaceae in diverse gut vasive alien species. A selection from the Global Invasive Species Database. In:
communities. Diversity 5, 627–640. https://doi.org/10.3390/d5030627. Invasive Species Spec. Gr. a Spec. Gr. Species Surviv. Comm. World Conserv. Union
Chaillou, S., Chaulot-Talmon, A., Caekebeke, H., Cardinal, M., Christieans, S., Denis, C., 12, https://doi.org/10.1614/WT-04-126.1.
Hélène Desmonts, M., Dousset, X., Feurer, C., Hamon, E., Joffraud, J.-J., La Carbona, Lu, X.M., Lu, P.Z., Zhang, H., 2014. Bacterial communities in manures of piglets and adult
S., Leroi, F., Leroy, S., Lorre, S., Macé, S., Pilet, M.-F., Prévost, H., Rivollier, M., Roux, pigs bred with different feeds revealed by 16S rDNA 454 pyrosequencing. Appl.
D., Talon, R., Zagorec, M., Champomier-Vergès, M.-C., 2015. Origin and ecological Microbiol. Biotechnol. 98, 2657–2665. https://doi.org/10.1007/s00253-013-5211-4.
selection of core and food-specific bacterial communities associated with meat and Lun, Z.R., Wang, Q.P., Chen, X.G., Li, A.X., Zhu, X.Q., 2007. Streptococcus suis: an
seafood spoilage. ISME J 9, 1105–1118. https://doi.org/10.1038/ismej.2014.202. emerging zoonotic pathogen. Lancet Infect. Dis. 7, 201–209. https://doi.org/10.
Chiari, M., Zanoni, M., Tagliabue, S., Lavazza, A., Alborali, L.G., 2013. Salmonella ser- 1016/S1473-3099(07)70001-4.
otypes in wild boars (Sus scrofa) hunted in northern Italy. Acta Vet. Scand. 55 (1). Mann, E., Wetzels, S.U., Pinior, B., Metzler-Zebeli, B.U., Wagner, M., Schmitz-Esser, S.,
https://doi.org/10.1186/1751-0147-55-42. 2016. Psychrophile spoilers dominate the bacterial microbiome in musculature
Decastelli, L., Giaccone, V., Mignone, W., 1995. Bacteriological examination of meat of samples of slaughter pigs. Meat Sci. 117, 36–40. https://doi.org/10.1016/j.meatsci.
Wild boars shot down in Piedmont and Liguria, Italy. IBEX J. Mt. Ecol. 3, 88–89. 2016.02.034.
Doulgeraki, A.I., Ercolini, D., Villani, F., Nychas, G.J.E., 2012. Spoilage microbiota as- Martínez-porchas, M., Villalpando-canchola, E., Vargas-albores, F., 2016. Significant Loss
sociated to the storage of raw meat in different conditions. Int. J. Food Microbiol. of Sensitivity and Specificity in the Taxonomic Classification Occurs When Short 16S
157, 130–141. https://doi.org/10.1016/j.ijfoodmicro.2012.05.020. rRNA Gene Sequences Are Used. HLYhttps://doi.org/10.1016/j.heliyon.2016.
EFSA, 2017. The European Union summary report on trends and sources of zoonoses, e00170.
zoonotic agents and food-borne outbreaks in 2016. EFSA J. 15 (12), 5077. Massei, G., Kindberg, J., Licoppe, A., Gačić, D., Šprem, N., Kamler, J., Baubet, E.,
Ercolini, D., Russo, F., Nasi, A., Ferranti, P., Villani, F., 2009. Mesophilic and psychro- Hohmann, U., Monaco, A., Ozoliņš, J., Cellina, S., Podgórski, T., Fonseca, C., Markov,
trophic bacteria from meat and their spoilage potential in vitro and in beef. Appl. N., Pokorny, B., Rosell, C., Náhlik, A., 2015. Wild boar populations up, numbers of
Environ. Microbiol. 75, 1990–2001. hunters down? A review of trends and implications for Europe. Pest Manag. Sci. 71,
Freitas, R. de, Chuat, V., Madec, M.N., Nero, L.A., Thierry, A., Valence, F., de Carvalho, 492–500. https://doi.org/10.1002/ps.3965.
A.F., 2015. Biodiversity of dairy Propionibacterium isolated from dairy farms in Membré, J.M., Laroche, M., Magras, C., 2011. Assessment of levels of bacterial con-
Minas Gerais, Brazil. Int. J. Food Microbiol. 203, 70–77. https://doi.org/10.1016/j. tamination of large wild game meat in Europe. Food Microbiol. 28, 1072–1079.
ijfoodmicro.2015.03.006. https://doi.org/10.1016/j.fm.2011.02.015.
Fuks, G., Elgart, M., Amir, A., Zeisel, A., Turnbaugh, P.J., Soen, Y., Shental, N., 2018. Mirceta, J., Petrovic, J., Blagojevic, B., Malesevic, M., Antic, D., 2015. The micro-
Combining 16S rRNA Gene Variable Regions Enables High-resolution Microbial biological status of carcasses from wild boar in Serbia. Procedia Food Sci 5, 199–202.
Community Profiling. pp. 1–13. https://doi.org/10.1186/s40168-017-0396-x. https://doi.org/10.1016/j.profoo.2015.09.056.
Gill, C.O., 2007. Microbiological conditions of meats from large game animals and birds. Mirceta, J., Petrovic, J., Malesevic, M., Blagojevic, B., Antic, D., 2017. Assessment of
Meat Sci. 77, 149–160. https://doi.org/10.1016/j.meatsci.2007.03.007. microbial carcass contamination of hunted wild boars. Eur. J. Wildl. Res. 63. https://
Gori, K., Ryssel, M., Arneborg, N., Jespersen, L., 2013. Isolation and Identification of the doi.org/10.1007/s10344-017-1096-3.
Microbiota of Danish Farmhouse and Industrially Produced Surface-ripened Cheeses. Molin, K., 2000. Contamination Routes of Gram-negative Spoilage Bacteria in the
pp. 602–615. https://doi.org/10.1007/s00248-012-0138-3. Production of Pasteurised Milk, Evaluated by Randomly Amplified Polymorphic DNA
Großkopf, R., Janssen, P.H., Liesack, W., 1998. Diversity and structure of the methano- (RAPD). 10. pp. 325–331.
genic community in anoxic rice paddy soil microcosms as examined by cultivation Mulet, M., Gomila, M., Scotta, C., Sánchez, D., Lalucat, J., García-Valdés, E., 2012.
and direct 16S rRNA. Gene Sequence Retrieval 64, 960–969. Concordance between whole-cell matrix-assisted laser-desorption/ionization time-of-
Hilgarth, M., Behr, J., Vogel, R.F., 2018. Monitoring of Spoilage-associated Microbiota on flight mass spectrometry and multilocus sequence analysis approaches in species
Modified Atmosphere Packaged Beef and Differentiation of Psychrophilic and discrimination within the genus Pseudomonas. Syst. Appl. Microbiol. 35, 455–464.
Psychrotrophic Strains. https://doi.org/10.1111/jam.13669. https://doi.org/10.1016/j.syapm.2012.08.007.
Höll, L., Behr, J., Vogel, R.F., 2016. Identification and growth dynamics of meat spoilage Naya, Y., Horiuchi, M., Ishiguro, N., Shinagawa, M., 2003. Bacteriological and genetic
microorganisms in modified atmosphere packaged poultry meat by MALDI-TOF MS. assessment of game meat from Japanese wild boars. J. Agric. Food Chem. 51,
Food Microbiol. 60, 84–91. https://doi.org/10.1016/j.fm.2016.07.003. 345–349. https://doi.org/10.1021/jf020579f.
Holman, D.B., Brunelle, B.W., Trachsel, J., Allen, H.K., 2017. Meta-analysis to define a Nores, C., Llaneza, L., Álvarez, Á., Nores, C., Llaneza, L., Álvarez, M.Á., Wild, M.Á., 2008.
core microbiota in the swine gut. mSystems 2, e00004–e00017. https://doi.org/10. Wild boar Sus scrofa mortality by hunting and wolf Canis lupus predation: an ex-
1128/mSystems.00004-17. ample in northern Spain. Wild boar Sus scrofa mortality by hunting and wolf Canis
Jeong, S., Hong, J.S., Kim, J.O., Kim, K.H., Lee, W., Bae, I.K., Lee, K., Jeong, S.H., 2016. lupus predation: an example in northern Spain 14, 44–51.
Identification of acinetobacter species using matrix-assisted laser desorption ioniza- Okarma, H., Jędrzejewska, B., Jędrzejewski, W., Krasiński, Z.A., Miłkowski, L., 1995. The
tion-time of flight mass spectrometry. Ann. Lab. Med. 36, 325–334. https://doi.org/ Roles of Predation, Snow Cover, Acorn Crop, and Man-related Factors on Ungulate
10.3343/alm.2016.36.4.325. Mortality in Białowieża Primeval Forest, Poland Introduction Ungulates in Central
Kanai, Y., Hayashidani, H., Kaneko, K., Ogawa, M., Takahashi, T., Nakamura, M., 1997. Europe Have Long Experienced Man-caused Mortality in Addition. 40. pp. 197–217.
Occurrence of zoonotic bacteria in retail game meat in Japan with special reference Park, S., Kim, J., Lee, J., Rhee, S., Kim, H., 2014. Anaerobe characterization of the fecal
to Erysipelothrix. J. Food Prot. 60, 328–331. https://doi.org/10.4315/0362-028X- microbiome in different swine groups by high-throughput sequencing. Anaerobe 28,
60.3.328. 157–162. https://doi.org/10.1016/j.anaerobe.2014.06.002.
Keuling, O., Sange, M., Acevedo, P., Podgorski, T., Smith, G., Scandura, M., Apollonio, M., Paulsen, P., Winkelmayer, R., 2004. Seasonal variation in the microbial contamination of
Ferroglio, E., Vicente, J., 2018. Guidance on estimation of wild boar population game carcasses in an Austrian hunting area. Eur. J. Wildl. Res. 50, 157–159. https://
abundance and density: methods, challenges, possibilities. EFSA Support. Publ. 15. doi.org/10.1007/s10344-004-0054-z.
https://doi.org/10.2903/sp.efsa.2018.EN-1449. Pedrazzoli, M., Bosco, A.D., Castellini, C., Ranucci, D., Mattioli, S., Pauselli, M., Roscini,
Knight, R., Navas, J., Quinn, R.A., Sanders, J.G., Zhu, Q., 2018. Best practices for ana- V., Pedrazzoli, M., Bosco, A.D., Castellini, C., 2017. Effect of Age and Feeding Area on
lysing microbiomes. Nat. Rev. Microbiol. https://doi.org/10.1038/s41579-018- Meat Quality of Wild Boars. https://doi.org/10.1080/1828051X.2017.1292114.
0029-9. Peruzy, M.F., Murru, N., Perugini, A.G., Capuano, F., Delibato, E., Mercogliano, R.,
Koo, O.-K., Baker, C.A., Kim, H.-J., Park, S.H., Ricke, S.C., 2016. Metagenomic assessment Korkeala, H., Proroga, Y.T.R., 2017. Evaluation of virulence genes in Yersinia en-
of the microbial diversity in ground pork products from markets in the North Central terocolitica strains using SYBR Green real-time PCR. Food Microbiol. 65, 231–235.
Region of South Korea. J. Environ. Sci. Heal. Part B 51, 622–627. https://doi.org/10. https://doi.org/10.1016/j.fm.2017.03.004.
1080/03601234.2016.1181910. Peruzy, M.F., Murru, N., Yu, Z., Cnckaert, M., Joossens, M., Houf, K., 2019. Determination
Lagier, J.C., Khelaifia, S., Alou, M.T., Ndongo, S., Dione, N., Hugon, P., Caputo, A., of the microbiological contamination in minced pork by culture dependent and 16S
Cadoret, F., Traore, S.I., Seck, E.H., Dubourg, G., Durand, G., Mourembou, G., amplicon sequencing analysis. Food Microbiol. 290, 27–35. https://doi.org/10.1016/
Guilhot, E., Togo, A., Bellali, S., Bachar, D., Cassir, N., Bittar, F., Delerce, J., Mailhe, j.ijfoodmicro.2018.09.025.
M., Ricaboni, D., Bilen, M., Dangui Nieko, N.P.M., Dia Badiane, N.M., Valles, C., Rawat, S., 2015. Food spoilage: microorganisms and their prevention. Asian J. plant
Mouelhi, D., Diop, K., Million, M., Musso, D., Abrahão, J., Azhar, E.I., Bibi, F., Yasir, ascience Res. 5, 47–56.
M., Diallo, A., Sokhna, C., Djossou, F., Vitton, V., Robert, C., Rolain, J.M., La Scola, B., Roh, S.W., Kim, K., Nam, Y., Chang, H., Park, E., 2009. Investigation of archaeal and

59
M.F. Peruzy, et al. International Journal of Food Microbiology 301 (2019) 51–60

bacterial diversity in fermented seafood using barcoded pyrosequencing. ISME J 4, doi.org/10.1016/j.meatsci.2016.09.007.

1–16. https://doi.org/10.1038/ismej.2009.83. Toïgo, C., Servanty, S., Gaillard, J.M., Brandt, S., Baubet, E., 2008. Disentangling natural
Rouger, A., Moriceau, N., Prévost, H., Remenant, B., Zagorec, M., 2017. SC. Food from hunting mortality in an intensively hunted wild boar population. J. Wildl.
Microbiol. https://doi.org/10.1016/j.fm.2017.08.013. Manag. 72, 1532–1539. https://doi.org/10.2193/2007-378.
Ruiz-Fons, F., 2017. A review of the current status of relevant zoonotic pathogens in wild Williams, T.L., Andrzejewski, D., Lay, J.O., Musser, S.M., 2003. Experimental factors
swine (Sus scrofa) populations: changes modulating the risk of transmission to hu- affecting the quality and reproducibility of MALDI TOF mass spectra obtained from
mans. Transbound. Emerg. Dis. 64, 68–88. https://doi.org/10.1111/tbed.12369. whole bacteria cells. J. Am. Soc. Mass Spectrom. 14, 342–351. https://doi.org/10.
Sales, L.P., Ribeiro, B.R., Hayward, M.W., Paglia, A., Passamani, M., Loyola, R., 2017. 1016/S1044-0305(03)00065-5.
Niche conservatism and the invasive potential of the wild boar. J. Anim. Ecol. 86, Young Ko, H., Cho, K., Park, S., Kim, J.H., Kang, J.-H., Jeong, S., Choi, J.D., Sin, Y., Lee,
1214–1223. https://doi.org/10.1111/1365-2656.12721. C., Ko, G., 2018. Host-specific bacteroides markers-based microbial source tracking in
Simpson, P.J., Fitzgerald, G.F., Stanton, C., Ross, R.P., 2006. Enumeration and identifi- aquaculture areas. Microbes Env. 00, 0. https://doi.org/10.1264/jsme2.ME17166.
cation of pediococci in powder-based products using selective media and rapid PFGE. Yu, Z., Peruzy, M.F., Dumolin, C., Joossens, M., Houf, K., 2019. Assessment of food mi-
J. Microbiol. Methods 64, 120–125. https://doi.org/10.1016/j.mimet.2005.04.019. crobiological indicators applied on poultry carcasses by culture combined MALDI-
Strazdina, V., Safety, F., Health, A., Bior, E., Technology, F., Jemeljanovs, A., Sterna, V., TOF MS identification and 16S rRNA amplicon sequencing. Food Microbiol. https://
Stende, S., Breeding, C., Safety, F., Health, A., Bior, E., 2014. Nutritional doi.org/10.1016/j.fm.2019.01.018.
Characteristics of 32–36. Zottola, T., Montagnaro, S., Magnapera, C., Sasso, S., De Martino, L., Bragagnolo, A.,
Sutton, S., Sutton, S., Sutton, S., 2011. Accuracy of Plate Counts. pp. 42–46. D’amici, L., Condoloeo, R., Pisanelli, G., Iovane, G., Pagnini, U., 2013. Prevalence and
Tian, X., Wu, W., Yu, Q., Hou, M., Gao, F., Li, X., Dai, R., 2017. Bacterial diversity analysis antimicrobial susceptibility of Salmonella in European wild boar (Sus scrofa); Latium
of pork longissimus lumborum following long term ohmic cooking and water bath Region – Italy. Comp. Immunol. Microbiol. Infect. Dis. 36, 161–168. https://doi.org/
cooking by amplicon sequencing of 16S rRNA gene. Meat Sci. 123, 97–104. https:// 10.1016/j.cimid.2012.11.004.

60