SOURCES of Listeria monocytogenes CONTAMINATION in TRADITIONAL FERMENTED sausage PROcessing plants IN ITALY

See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/233832458

Sources of Listeria monocytogenes

contamination in traditional fermented sausage

processing plants in Italy

Article in Italian Journal of Food Science · January 2012 Impact Factor: 0.29 CITATIONS

6

READS

44

5 authors, including: Domenico Meloni Università degli Studi di Sassari 63 PUBLICATIONS 127 CITATIONS SEE PROFILE Francesca Piras Università degli Studi di Sassari 46 PUBLICATIONS 120 CITATIONS SEE PROFILE Roberta Mazza Università degli Studi di Sassari 26 PUBLICATIONS 27 CITATIONS SEE PROFILE

All in-text references underlined in blue are linked to publications on ResearchGate, letting you access and read them immediately.

Available from: Domenico Meloni Retrieved on: 24 June 2016

l:.,:,.','ì...lì;,:,,:.ll.ii;:,a#W*ffiPAPER

S

OURCES

OF

TISTER

IA

MON

O

CWOGENES

CONTAMINATION

IN

TRADITIONAL

_{FERMENTED SAUSAGE}

PROCESSING

PTANTS

IN

ITALY

D. MELONI'È, F. PIRAS, A. MUREDDU,

R.MAZZA,D.

NUCERAI and R. MAZZETTE Dipartimento di Biologia Animale, Facoltà di Medicina Veterinaria, Università di Sassari,

Via Vienna 2,071.00 Sassari,

Italy

lDipartimento di Patologia Animale, Facoltà di Medicina Veterinaria, Università di Torino, Via L. Da Vinci M,1,0095 Grugliasco (TO),Italy

*Corresponding author: Tel. +39 079 229570, Fax +39 079 229458,

email: dmeloni@uniss.it

ABSTRACT

Environmental samples, raw materials and fermented sausages produced

ln

Sardinia (Italy)

were analysed

in

order to investigate the prevalence and enumeration of L. monocgtogenes. Iso-lates were identified by single PCR and characterised by multiplex PCR-based serogrouping. The contamination routes of L. monocgtogenes

in

the plants were traced using PFGE.

In

additton, a quantitative assessment of

the

in uitro biofilm formatlon was carried out. Fermented sausages seem to be regularly contaminated with L. monocgtogenes: results showed the

abil§

of the

path-ogen to overcome the hurdles of the manufacturing process and adapt to the processing plant

en-vironments, forming biofilms.

- Ke)rwords: biofilm, fermented sausages, Listeriamonocytogenes,polymerase chain reaction, pulsed-field gel electrophoresis

INTRODUCTION

Listerinmonocytogenes is a ubiquitous organ-ism, widely distributed in the environment. The

principal

reservoirs are soil, forage and water

(EFSA,

20l

f _). Other reservoirs include healthy humans

and

animals (ILSI, 2005)

or

infected

domestic and

wild

animals

(EFSA,

_2}ll).

L.

monocytogenes has been isolated throughout

the pork

processing

industry

(NESBAKKEN et

al.,

1996),

with

an increase of contamination

along the production line (CHASSEIGNAUX etal., 2OO2). One of the most recent listeriosis out-breaks (Canada 2OO8) was linked

to the

con-sumption of ready to eat (RTE) pork meat prod-ucts, causing22 deaths and 57 conflrmed cases

(PHAC, 2OO9). Fermented RTE pork meat

prod-ucts, such as dry and semi-dry sausages have

rarely been implicated

in

food poisoning.

Nev-ertheless,

in

the manufacturing of traditional

fermented products,

an

empirical application

of hurdle technologies often occurs and these

products

are

regularly

contaminated

with

L.

monocytogenes. This fact may present a maJor

public

health concern

if

the

pathogen

is

able to

multiply

and reach high levels of

contamina-tion (THEVENOT et aL.,2OOGI). Previous surveys

carried

out

on

traditional

fermented products

at

the

end of ripening showed a prevalence of I0%

in

France (THEVENOT et aL.,2OO5), LO.6o/o

in

Chile (CORDANO and ROCOURT, 2OOl) and

L5.2o/o

in Italy

(DE CESARE et al., 2OO7).

ln

a

former study,

L.

monocytogenes was found

in

4Oo/o of "Salsiccia Sarda' at the end of ripening

(MELONI et aL.,2OO9) with contamination levels

always lowerthan IOO CFU/g. "SalsicciaSarda' is the primary meat product (37o/o _of

consump-tion) of the meat supply chain

in

the Sardinia region (Italy) and is included on the national

list

of traditional food products. It is one of the typ-ical Italian semi-dry sausages (a* ranging from

O.9O to 0.95), naturally fermented (GRECO et al.,

2OO5) and marketed locally and regionally in

It-aly (COMI et qL.,2OO5). _{The main composition}

and characteristics of " _{Salsiccra Sardd' include:}

minced lean

pork

(87o/o):

_pork

_back

_fat

(8olo);

salt

(30lo);

a mixture of

sugars

and

additives

(O.8o/o): garlic (O.15%); ground pepper (O.25o/o);

and a mixture of whole pepper, nutmeg, cloves

and pimento (O.8%). The mixture, refrigerated overnight, is stuffed into natural pork casings. After they are

flrst

warmed up at 2O"-22'C for

4-6 h, the products are dried for six days

in

a fermentation chamber. On the

first

day of

dry-ing,

the

products are stored

at

2O"-22'C and

6O%o relative

humidity.

Over the next five days

of drying the temperature is gradually reduced

to

l5'C

and

the

relative

humidity is

gradual-ly

increased

to

7Oo/o. Ripening

is

then carried out

for

l5

days in store-rooms

at

15'C and

7O-75o/o relative

humidity.

The

flnal

products are

cylindrical (about 3O mm

A),

about 4O-45 cm long, shaped like horseshoes, and about half a

kg each in weisht. The normal pH of "Salsrccia

Sarda' at the end of ripening is 5.28 (GRECO et al.,2OO5). L. monocgtogenes may survive during

the processing of the fermented meat products due to

its

high tolerance to low pH conditions and high salt concentrations (FARBER _and

PE-TERKIN, 1991) and if the standard hurdle tech-nologies are ineffective. The main hurdles used

during

the processing of

dry

fermented

prod-ucts are

nitrite

and salt content, the decrease

in

redox potential and water

activity

(a.)

that

inhibit

many aerobic bacteria in the early

stag-es of production (Pseudomonas spp. and other Gram-negative bacteria), and selecting the

lac-tic

acid bacteria (l,AB)

which

cause the pH to

decrease (BARBUTI _and PAROLARI,2OO2). These

hurdles, with the coagulase negative

Staphylo-cocci development and the length of the ripen-ing period, are essentlal for the microbial

safe-ty and

stabil§

of quick-ripened fermented sau-sages,

which

are not greatly dried. Pork meat used

for the

manufacturing of sausages may be contaminated from a wide variety of sourc-es

and L,

monocgtogenes once introduced

in

the processing plants can persist over time

in

the environment, contaminating food

process-ing machines (LopEz et al.,2OO8). L.

monocg-tog ene s f orms assemblages of surface- asso

ciat-ed microbial cells that €rre enclosed in hydrated

extracellular pol5rmeric substances and grow

in

biofllms on surfaces

in

contact or not

with

the food (GANDHI and CHIKINDAS, 2OO7), such as floor drains, storage tanks, hand

trucks,

con-veyor

belts and other

food contact materials

(MAFU et aL., l99O). The presence of the

path-ogen on surfaces

in

contact and

wlthout

con-tact with food increases the food safety risk (KIM

and FRANK, 1995). Thus, L. monocgtogenes may

become an important source of secondary

con-tamination of food products, and without suit-able sanitisation procedures,

cross-contamina-tion

of the meat products may occur (SAMELIS

and METAXOPOULOS,

f999).

The

main

objec-tive of the present study was to evaluate the oc-currence of L. monocgtogenes in the fermented sausage production of the Sardinia region (It-aly). Isolates were identifled by single PCR and

further

characterised by multiplex PCR-based

serogrouping. The contamination routes of L.

monocgtogenes

in

the plants

were traced

us-ing

PFGE.

In

addition,

a

quantitative assess-ment of the tn urtro biofllm formation was

car-ried out in order to investigate the potential for

persistence.

MATERTALSAND METHODS Sampling

Atotal of l7O samples from environments, raw materials and final products from two large (A

and D > 3O0 tons of "Salsiccia Sarda' per year)

and two small (B and C from 5O

to

lOO tons of

" Salsiccra S arda' p er y ear) processing plants, lo-cated

in

different provinces of Sardinia (Italy), were sampled

in

order to investigate the preva-lence, ecologz and genetic profile of L. monocg-togenes. All of the processing plants were repre-sentative of the regional fermented sausage

pro-duction and some characteristics are

summa-rised in Table

I.

Three of the processing plants

(A, B, D) were only sampled once (Sl). In order to evaluate the role of swine carcasses as a

pos-sible source of plant contamination and tJee

per-sistence of L. moruocytogenes-adapted strains, in

plant C (randomly chosen) samplingwas

repeat-ed within three months (S2). Details on samples collected

from

each

plant

are provided

in

Ta-ble 2. A total of 132 environmental samples were

collected: 96 swabbed surfaces without contact

with

meat (SWCM) and 36 surfaces

ln

contact

with

meat (SCM). SWCM and SCM were

s€un-pled during the production stages by swabbing

with

1O cm

by

IO cm sterile ga:uze pads rehy-drated

with

10 mL of neutralising buffer (Solar-cult sampling kit, Biogenetics, Padova, Italy) and using a sterile template to delineate the swabbed area

of

lO0 cm2. Sampling locations were cho-sen

in

order

to

represent those most

likely

to present L. monocgtogenes contamination.

Sam-pling sites for SWCM included walls and floor

drains of the ground meat store-rooms, dryrng and ripening rooms, processing and packaging/ shipment rooms. Regarding SCM, the sampling

sites included

work

tables, sausage trolleys, hooks, mincing, mixing and stufflng machines. A total of 16 samples from ground meat and 1O

from fermented sausages

at the

end of ripen-ing were collected. Only during the repeat sam-pling (S2) _{in plant C12 samples from swine} car-casses were collected: 6 of raw pork meat with

and without

rind

and 6 swab samples of

verte-bral

canal between

the

first

thoracic vertebra

and the seventh lumbarvertebra, in

correspond-ence with the bone saw surface of the swine

car-casses. The carcasses were stored at +4'C and were sampled before their entry in the process-ing line. Samples were collected by swabbprocess-ing as

previously described. All of the items were

asep-tically sampled, placed in sterile bags (kept in ice boxes at +3'C) during transport and were imme-diately analysed upon arrival at the laboratory.

Detection and enumeration

of L. monocgtogenes

Detection

and enumeration

of L.

monocg-togenes was carried out using the ISO

ll290-1:1996

and

1129O-2:I998 protocols,

respec-tively. Samples of raw pork meat, ground meat, fermented sausages and swabbed samples were homogenised 1/10 witJ: Fraser broth base (Bi-olife, Milan, Italy)

in

a Stomacher Lab Blender

4OO (Seward Medical, London, UK)

for 2

min. The homogenates were incubated at 20"C

for

I

h,

in

order to resuscitate stressed microorgan-isms. For the enumeration of L. monocgtogenes,

I

ml of each inoculum was distributed over the surface of three Aloa (Biolife) 9O mm plates

us-ing a sterile spreader. The plates were

incubat-ed

at

37'C for 48

h.

For detection of

L.

mono-cgtogenes, the homogenates were

supplement-ed by Fraser half-selective supplement (Biolife)

and incubated at SO'C for 24 h. Afterwards, O.

I

mL of the primary enrichmentwas inoculated in IO mL of Fraser broth supplemented (Biolife) _by

Fraser selective supplement (Biolife) and

incu-bated at 37"C for 48 h. Cultures were streaked onto O>dord (Oxoid, Milan, Italy) and Aloa (Bi-olife) plates and incubated at 3O' and 37'C for 48

h,

respecttvely. From each plate of the

pri-mary and secondary enrichment, five colonies

presumed to be Listeria spp. were streaked on

Tsyea plates (Biolife) and incubated at 37'C for

24 h. Colonies were selected for typical

appear-ance onTsyea and submitted to Gram staining,

catalase and oxidase tests. Haemolytic activity

and CAMP tests on sheep blood agar were

per-formed

for

conflrmation

of L.

monocgtogenes.

Biochemical characterisation of all the isolates was performed using the API Listeria identifica-tion system (bioMérieux, Marcy l'Etoile, France).

Molecular identification and characterisation Svqle P C R-b as e d identific ation

The phenotypic identification

of

L.

monocg-togenes isolates was confirmed by a single

PCR-based method (sPCR) _aimed at prJA gene frag-ment detection. L. monocgtogenes isolates were

grown on BHI (Oxoid) at 37'C

for

16-18

h.T4e

cells were pelleted by centrifugation

of

I

mL at

Table I - Some characteristics of the four Sardinian processing plants.

Tons/year Characteristicsof Ripeningperiod

the ground

meat

(fermented sausages)

Fresh

15 days

Fresh

15 days

Fresh

15 days

Fresh

7-15 days

A

300-500

B

50-100

c

50-100

D

>500 Processing line complexity +++ ++ ++ ++++ Origin of the groundmeat Domeslic Domestic Domestic Domestic /European

Processing line complexity: ++ (simple equipment);+++ (more equipment, more complex); ++++ (a lot ol equipment, with a complex design).

12,OOO rpm for 5 min at 4"C and washed twice in phosphate-buffered saline (PBS). DNAwas

ex-tracted by suspending the pellet

in I

mL of PBS, which was boiled for 5

min

and centrifuged at

IO,OOO

rpm.

The supernatant was quantifled

with

a

LIV-I7OO PharmaSpec

spectrophotome-ter (Shimadzu, Kyoto, Japan) at ODruo and then stored at -20'C

until

use. Heterologous DNA of

S. .rElosns ATCC 29971was used as a negative

control and DNA of L. morwcgtogenes reference

strain

ArcC

19115 was used as a positive

con-trol. The sPCRwas carried out with

tle

primer set

Lipl

and Lip2 (DAGOSTTNO eta"L,2OO4; JOFRÉ et

aL, 2OO5) which produces, a fragment of

274be(St-MON et al., f 996). All ampliflcation reactions'ivere performed in a final volume of 5O pL containing 5 pL of DNA, 5 pL of lOX PCR buffer (Invitrogen, Carlsbad, USA), 2.5 mM of MgCl, O.3 mM each of dNTP, O.3 mM each of primer and

IU

of

Plat-inum Taq DNA polyrnerase (Invitrogen). All

am-pliflcation reactions were performed

in

a

Gene-Amp 27OO Thermal Cycler (Applied Biosystems,

Foster City, CA, USA) programmed as follows: de-naturation at 94'C for 2 min, annealing at 55"C

for 3O sec and elongation at74"C for 1 min, fol-lowed by a final extension period at 74"C for 5

min. The amplified fragments were separated by

1.50lo a§arose gel electrophoresis (Roche

diagnos-tics, Milan, Italy)

in

lX

Tris-acetate EDTA (TAE,

Invitrogen) and stained

with

ethidium bromide

(0.I mglml-) for 20 min. The gels were observed

and digitalised try the Gel-Doc LIV

trans-illumi-nator (Bio-Rad, Hercules, CA, USA).

M ultiple x P C R- b as e d s e rog r o up ing

Multiplex

PCR-based

serogrouping

was

carried

out

using the target

genes LmoO737,

\nwl118, ORF2819, ORF2L1 O and prs (DOUMITH

et aL.,2OO4). The multiplex PCR products were

resolved by electrophoresis on 1.5olo agarose gel

in

IXTAE (Invitrogen) and stained with ethidium bromide

(0.I

mglml.) for 20 min. The gel

imag-es were visualised and captured using the Gel-Doc UV trans-illuminator (Bio-Rad).

DNA macrorestriction and pulsed-field gel electrophoresis (PFGE)

Isolates were submitted to DNA

macrorestric-tion with ApaI and AscI (New England Biolabs, Beverly, MA, USA). The separation of

the

re-striction fragments was carried out by PFGE

in

a CHEF MapperXA system (Bio-Rad) using the PFGE-PulseNet protocol (GRAVES _and

SWAMI-NATFIAN, 2OOt). Gel images were visualised and captured using the Gel-Doc

UVtrans-illumina-tor (Bio-Rad). The banding patterns for each en-4/rne were assigned through visual analysis

of

tlee restriction proflles. Isolates were designated

genetically indistinguishable (same pulsotype)

when

their restriction

patterns had

the

same number of bands and the correspondingbands

were the same apparent size (TENoVER _et aL.,

1995; GRAVES etaL.,2OO5). AscI and Apalmac-rorestriction patterns were analysed using Bi-oNumerics software (Applied Maths,

Sint-Mar-tens-Platen, Belgium). The

similarity

between

restriction patterns, based on bands position,

\ ras expressed as a Dice coefflcient correlation.

The position tolerance was optimal when set at

l.O

and 2.Oo/o for AscI and ApaI, respectively.

Clustering

and construction of

dendrograms

were performed by the Unweighted Pair Group

Method

using

arithmetic

averages (UPGMA)

combining both AscI and ApaI macrorestriction

patterns into one unique PFGE profile. Quantitative assessment

of in uitro biofllm formation

Isolates were tested for their ability to attach to abiotic surfaces forming biofilm. The quanti-tative assessment of the ur uitrobiofllm formation was carried

out

on sterile 96-well polystyrene

microtiter

plates using

the

method described by STEPANOVIC et al. (2OO4), _{with some}

modifi-cations. Isolates were grown for 24 h in 2 mL of BHI broth. All the wells of a polystyrene

micro-titer plate were filled

with

230 pL of BHI broth. Afterwards, 21 wells per strain were filled with

20 1t"L of culture. Each plate included 12 wells of BHI broth without inoculum, as a negative

con-trol. Microtiter plates were incubated at 3Oo for

24h.

Atthe end of the incubation the content of the wells was removed and the microtiter plate washed three tlmes

with

3OO mL of sterile,

dis-tilled water

in

order to remove loosely attached bacteria. The remaining attached bacteria were fixed

with

25O 1tL of methanol per well, and

af-ter 15 min the wells were emptied and air-dried.

Each well was stained with 25O prl, of crystal

vio-let for 5 min. After stalning, the microtiter plates

were washed under running tap water, then air-dried and the dye bound to the adherent cells was solubilised

with

25O _1.tL of 33o/o (v/v) glacial

acetic acid per well. The microtiter plates were read spectrophotometrically (OD"ro) using a Sun-rise RC absorbance reader (Tecan, Maénnedorf, Swltzerland). The strains were divided into four

categories: no biofilm producers (NP, _OÈ<0.5),

weak producers [WP,

OÈ

>0.5<1.0), moderate producers (MP,

_OÈ

>f .O<I.5) and strong pro-ducers (SP,

OÈ

>f .5).

RESULTS

Detection and enumeration

of L. monocytogenes

The results (Table 2) showed the presence of

L. monocgtogenes

in

all

of the fermented sau-sage processing

plants

(overall prevalence in

the environments: I5%). None of the samples

from pork carcasses was positive for L.

Table 2 ' ListeriamonocAtogenes in en\^ronments, raw materials and final products of four fermented sausage processing plants. Plants A B D Grand total

swcM***' Pork carcasses Ground

meat

Fermented sausages

Positive/Total

Yo

Positive/Total

o/o

Positive/total

o/o

Positive/Total

%

Positive/Total o/o

S1 S1 s1* s2** S1

0/6

-

2t18

11.11

216 ss.33 4t18

22.22

0/10

6/18

33.33

0/8

-

a20

10

0t12

216

33

A22

I

.

4136 11.11 16i96 16.66

0t12

014

-

414

100

6/16 3750 8/10

80

2t4

50

2t2

100

214 50

0t2

214 50

U2

'r00 *S1:first

sampling; **S2: repeated sampling; ***SCM:Surfaces in contact with meat; ..-'SWCM: Surfaces without contact with meat,

cgtogenes. The occurrence was 37o/o in ground

meat and 8Oo/o in the fermented _{sausages. These}

products did show detectable levels always

be-low

lO

CFU/g, complying

with

the food safety

criteria provided for RTE foods able to support the growth of L. morrccgtogenes (EFSA, 2OO5).

According to THEVENOT et

al.

(2006r), several factors, such as the manufacturing process, the

use of spices with antioxidative or antimicrobial

properties, and the growth of the natural

com-petitive microflora should significantly reduce

the count of L. monocgtogenes in fermented sau-sages. Altogether, 170

strains

of Listeria spp.

were isolated: 39o/o were L. welshimert,23o/o L.

monocytogenes, 2lo/o L. innocua. LOo/o L. gragi,

4o/o L. iuanouìi, and 3olo L. seeligeri. A subset of 20 L, manocgtogenes isolates was selected and

primarily

identified by single PCR. From each processing plant, L. monocgtogenes strains were selected in order to include almost 5oolo of each category of positive samples. Isolates were

fur-ther characterised by multiplex PCR-based se-rogrouping, PFGE

and in

uitro

biofllm

forma-tion. A summary of the phenotypic and geno-typic characterisfies of the 20 L. monocgtogenes

isolates is reported in Table 3.

Table 3 - Distribution of 20 Li:teria manocatogenes isolates in the four processing plants, in relation to serotype, pulsotype,

PFGE profile and in uitro biofilm production.

Plant and source of contamination

A

Ground meat store room: floor drains

Ground meat Fermented Sausage

Strain

code

Serotype Pulsotypes PFGE prolile Biofilm production

Ascl

Apal A1 A2 A3 4b 4b 1l2b NP MP NP

I

11 t3

tt

ltlt

ilt

ilt

B

Ground meat store room:lloor drains

Ground meat store room:floor drains Ground meat

Drying room: floor drains

Drying room: floor drains

Ripening room: hooks Ripening room: hooks

MP NP ,,WP ,WP NP WP WP 4 4 2 2 7 1 1

1t2b

tv

tv

1l2b

lv

lv

1l2b

v

v

1l2b

v

v

1l2b

vt

vt

1l2b vil

vil

1l2b vil

vil B1 B2 83 B4 B5 B6 87

C

Drying room: floor drains

Ground meal Fermented sausage

Shipmenl room: floor drains

Ripening room: floor drains

Ripening room: floor drains

C1 c2 c3 c4 c5 CSbls

1l2b vilt

vilt

1l2b lx

tx

1l2b tx

lx

1l2b x

x

1l2b x

x

1l2b x

x 12 5 5

I

8

I

NP WP NP WP NP MP

D

Fermented

sausage

Dl

1l2a

Xl

Fermented

sausage

D2

1l2a

Xll

Processing room: work

tables

D3

1l2a

Xlll

Processing room: f loor

drains

D4

1l2a

XIV

*n.t.: _not

typeable; NP: no biofilm producers;WP: weak producers; MP: moderate producer.

WP WP WP NP 6 3 10 n.t.' xt xll xil n.t.*

Molecular identiflcation and characterisation

SirtS l-e P CR- b as e

d

ident!fi c atian

The amplification product of 274n^was found

in all the strains phenotypically

idditified

as L.

moracytogenes. The amplification products

al-ways exhibited the same electrophoretic pattern.

These results confi.rm that the prJAgene is a re-liable target gene for identiflcation of the

patho-gen (vAzouEz-Bor-AND et aL, 2oot).

M uttiplex P CR- b a s e d s e ro g r oup irtg

The prevalent serotype was

l/2b

(7Oo/o),

fol-lowed

by

L/2a

(2Oo/o) and

4b

(loolo). No

specif-ic

sero§rpe was recovered during the process-ing and ripenprocess-ing of the sausages (THEVENOT et

aL.,2OO5).

DNA macrorestriction and pulsed-field gel electrophoresis (PFGE)

A

high heterogenei§ of pulsotypes

(I4

with

AscI

and

13

with

ApaI)

occurred

within

the plants (Table 3). Restriction patterns were com-bined

in

l3

PFGE profiles appearingto be plant-speciflc. One strain from plant D was typeable only with AscI (pulsotype

)(V)

and consequent-ly was not assigned to any PFGE profile.

With-in each processing plant, particularly in the two

large ones (A _and D), a high heterogeneity distri-bution of PFGE proflles was observed. The pres-ence of PFGE proflle flve both

in

ground meat

and

in

the fermented sausages, may evidence

the

abil§

of speciflc strains of L. monocytogenes

to

survive

during

"Salsiccia Sotrda'

fermenta-tion. Furthermore, the recovery of PFGE profile eight in floor drains from different rooms in both samplings

(Si

and 52) showed the ability of the pathogen

to

adapt and persist

in

the different processing environments of plant C. The PFGE

profiles were allotted into three major PFGE

clus-ffi

§?

ffi

m

§* §t

§I

r1

m

§1

ffi

tr

§s c§ Àt

§I

*3

f;r A§

eh*6*,

C3r*erB Càm*uC

Fig. 1 - UPGMA clustering of the 19 Listeriafi@nocAtogenes

PFGE profiles.

ters (similartLy > 7Oo/o) labelled A-B-C (Table 4

and Fig.

l).

Cluster A included 9 isolates of

se-rotype |

/2b

(8) arrd |

/2a

(l) from three process-ing plants (B-C-D). Cluster B included 6 isolates belonging to different sero§pes: |

_/2a

{l),

|

_/2b

$s$

I

I

I

J

I

_i

F

I

Table 4 - Clustering of 18 ListeriamanocAtogenes strains.

Cluster Overall similarity PFGE profile N" of isolates

Serotype

Source of contamination in the plants

1

2

1l2b

Hooks in the ripening room

2

2

1l2b

Ground meat and floor drains in the drying room

A

3

1

1l2a

Fermented sausage

73.3Yo

4

2

1l2b

Floor drains in the store room

5

2

1l2b

Ground meat and Fermented sausage

77.3%

6

1

1l2a

Fermented sausage

7

1

1l2b

Floor drains in the drying room

8

3

1l2b

Floor drains in the shipping room (1) and ripening room (2)

9

1

4b

Floor drains in the store room

10

1

1l2a

Work tables in the processing room

11

1

4b

Ground meat

74.40/"

12

1

112Ò

Floor drains in the drying room

(a) _{and 4b}

(l)

from all of the processing plants (A-B-C-D). Cluster C included three isolates of

sero§4pe

l/2a,

l/2b,

and 4b from plantsA-C-D. One

strain

(A3) _of serot5rpe

l/2b

isolated from fermented sausage was an outlier, showing low

similarity (47o/o) _{wltl'r the other clusters.}

Quantitative assessment of in uitro

biofilm formation

More

than half

(6o0lo)

_{of the strains}

previ-ously identifled and characterised

by

molecu-lar methods were able to attach to abiotic

sur-faces forming biofilm. The results of this experi-ment are shown in Table 3 and Fig. 2. Serotypes

|

_/2a

and I

_/2b

showed weak OD (OD=>O.S< 1.O)

or moderate OD (OIÈ>1 .O< I .5)

abil§

in biofilm

formation. The microtiter plate assay is a useful method to assess the ability of L. manocgtogenes

strains to attach to abiotic surfaces (STEPANOVIC

et aL., 2OO4l.

DISCUSSION AND CONCLUSION

In this study, the contamination of four "sal-siccin Sardd' processing plants representative

of Sardinian fermented sausage production was

investigated

on

samples

from

surfaces (with

and

without

contact

with

meat), ground meat and fermented sausages. Our results show the overall presence

of L.

monocgtogenes

in

envi-ronmental niches, raw materials and final prod-ucts. L. manocgtogenes was not detected in the

pork carcasses prior to processing;

contamina-tion of the

final

products appears to be due to

strains already present

in

the processing envi-ronments (WPEZ et aL.,2OOS), with all of the fa-cilities serving as the source of product contam-ination (CHASSEIGNAUX et aL,2OO2). In the

Sar-dinian processing plants and their products, L. morwcgtogenes sero§pe | _/2b predominated,

fol-lowed by serotypes

l/2a

and 4b. Previous stud-ies have also reported the presence ofthese

se-ro§pes in meat processing environments

(CFIAS-SEIGNAUX et aL., 2OOL; THEVENOT et aL., 20061).

The recovery of serotypes frequently

associat-ed with epidemic or sporadic cases of listeriosis

(McI,AUCHLIN _{et al.,} 2OO4)

is

an interesting

re-sult in terms of public health protection. Previ-ous surveys have shown that lineage I stralns of sero§pe 4b belonging to a clonal group (DUP-ID

lO38) linked to several listeriosis outbreaks (DE

CESARE et al.,2OOl) were recently recovered in

the same meat products (MELONI et al.,2OO9).lt

must be emphasised that the presence of such

strains

in

the processing plants, and as a

con-sequence in fermented sausages, may represent ahazard if the pathogen is able to multiply dur-ing the ripendur-ing of the product and reach lev-els

higherthan

1,0OO CFU/g (ROSS etaL.,2OO2:

THEVENOT et aL , 20061). Thirteen different PFGE

profiles were obtained indicating a great level of

220

ltat. J. Food Sci., vol. 24 - 2012

diversity among the strains collected from the Sardinian processing plants and their products.

The high number of PFGE proflles and their het-erogeneous distribution

within

the plants is in

agreement

with

the results of previous surveys carried out in meat processing plants (THEVENOT

.."^.,"-'."--.."-1.

i

{},* *,ù 0,$ l

1.1

O§fiT§

T,4

Fig. 2 - Biofllm production of the Listena monacAtogenes

stralns.

The quantitauve assessment of the dn uitro biofflm formation was evaluated spectrophotometrically (OD62O). The strains were divided up into four categories: no biofllm producers

(NP= OD <0.5), weak producers NVP= OD >0.5<1.0), mod-erate producers (MP= OD >1,O<1.5) and strong producers

(SP= _6p >1.5). Altogether,6o0/o of the strains were able to

attach to abiotic surfaces formtng biofilm, showing weak or moderate ability in biofilm formaflon.

et aL , 20061). Such subspes appear to be unique

to each processing plant(FUGETT et al., ZOOZ).

These results may be due to the limited number

of isolates included in this study or to the great

diversity

of the L.

monocytogenes

strains

col-lected from ground meat,

but

also to the

quan-tities used by each plant. The large plants ex-hibited the highest PFGE proflle heterogeneity,

since these plants also used the greatest amount of raw meat from several different sources

(do-mesfic and European). Numerical analysis of the

PFGE profiles showed

that

the isolates includ-ed in the study could be allotted into three

ma-jor

PFGE clusters labelled A-B-C. Contrarlr to

all expectations, these three genomic divisions

were not linked with the flagellar antigen type, in

spite of what was described by several authors

(CHASSETGNAUX et aL., 2OO I ; AUTrO et aL., 2OO3t

THEVENOT et at.,2006Ì). In this study, L.

mano-cgtogenes strains were thought to be persistent when the same PFGE proflle occurred in samples

collected from the same plant equipment C after an interval of three months: such strains were found in several floor drains. The persistence of

the same PFGE proflle (eight) _{is probably due to}

two interacting factors. First of all, the nature of the strains themselves, which have a moder-ate ability to form biofilms and have adapted to environments where meat is processed.

Sever-al authors have shown

that

certain L.

manocg-togenes strains were more capable of causing

persistent contamination

of

meat processing

plants than others (AUTIO _etal.,2OO3; THEVEN-OT et aL.,20061) probably due to their ability to

form biofilms. The second factor affecting

per-sistence may be the presence of ineffective

clean-ing and disinfection measures. Treatments need

to

reach

the

contamination

site

in

sufflcient quantities and duration

in

order to be effective

(THEVENoT _{et al.,2O06'). This may not always}

occur with routine cleaning procedures because of

the

complexi§ of the processing line

struc-ture

and because environmental niches, such

as floor drains can be a site critical for control, ling contamination of the processing plant

en-vironment and food products (ToMPKIN, 2OO2).

Decontaminating floor drains is especially chal-lenging because, when entrapped

in

a biofllm,

L. manocgtogenes is afforded unusual protectlon against available disinfectants and treatments

(ZHA} et ctL, 2OO4).

L.

monocytogenes strains can become well established in the floor drains and perstst as resident microbial flora for up to several years. This study has shown the

pres-ence of adapted L. manocgtogenes strains able to

survive during sausage fermentation, overcom-ing the current hurdles of the "Salsiccia Sardd'

manufacturing process. Our results

highlight-ed

that

L. manocgtogenes

is

able to suryive in

meat processing plants by forming biofllms on abiotic surfaces. The surface used for the m

uit-ro experiment (polys§rene) approximately

mim-ics the plastic material used in the plants (e.g.,

conveyor belts). Further testing with appropriate

steel specimens of each plant is needed in order

to understand better the mechanism of biofllm formation inuiuo. To decrease the presence of L, monocgtogenes in the traditional fermented

sau-sages at the end of ripening, food business

op-erators should adhere to accurate application of hurdle technologies. Products can also become

contaminated through contact

with

work

sur-faces and equipment, even after routine

clean-ing and disinfecting operations (THEVENOT et aL, 20061). More attention should be focused on re-specting good manufacturing practices and the application of FIACCP principles.

REFERENCES

Anon5rmous 1996. Microbiology of food and animal feeding stuffs - horizontal method for the detection and enumer-ation of Listena manocgtogenes. Part l: Detection meth-od, l"t Ed. ISO method I 1290- l. Internatlonal

Organiza-Uon for Standardization, Genève, Switzerland.

Anon)'rnous 1998. Microbiology of food and animal

feed-ing stuffs - horizontal method for the detection and enu-meration of Listeria monocAtogenes. Part 2: Enumera-tion method, l$ Ed. ISO method ll29o-2. International Organizatlon for Standardization, Genève, Switzerland.

Autio T., Keto-Timonen R., Biorkroth J. and Korkeala H.

2003. Character2ation ofpersistent and sporadic Listena fiatacAtogenes strains by pulsed-fleld gel electrophore-sis (PFGE) and amplified fragment length poll..rnorphism

(AFLP). Syst. and AppI. Microbiol. 26, (4),539.

Barbuti S. and Parolari G.2OO2. Validation of manufactur-ing process to control pathogenic bacteria in typical dry fermented products. Meat. Sci. 62, (3), 323.

Chasseignaux E., Toquin M., Ragimbeau C., Salvat G.,

Co-lin P. and Ermel G. 2OO1. Molecular epidemiologr of

Lds-teria matacAtogenes lsolates collected from the enyiron-ment, raw meat and raw products in two poultry artd pork processing plants. J. ofAppl. Microbiol. 9l, (5), 888. Chasseignaux E., Gerault P., Toquin M., Salvat G., Colin P.

and Ermel G.2002. Ecologr of UsteriamonocAtogeneslrr tfre environment of raw poultry meat and raw pork meat processing plants. FEMS Microbiol. Irtt. 210, (2),271.

Comi G., Urso R., Iacumin L., Rantsiou K., Cattaneo P., Can-toni C. and Cocolin L. 2OO5. Characterisation of natural-ly fermented sausages produced in the North East of

It-aly. Meat Sci. 69, (3), 381.

Cordano A.M. and Rocourt J. 2001. Occurrence of Listerùt

manocAtogenes in food in Chile. Int. J. of Food

Micro-biol.70, (t-2),175.

DAgosUno M., Wagner M., Yazqrez-Boland J.A., Kutcha T., Karpiskova R., Hoorfar J., Novella S., Scortti A., Elli-son J., Murray A., Fernandes I., Kuhn M., Pazlarova J., Heuvelink A. and Cook N. 2004. A validated PCR-based

method to detect Listeriamorocytogenes using raw milk

as a food model-towards an international standard. J. of Food Prot. 67, (8), 1646.

De Cesare A., Bruce J. L., Dambaugh T.R., Guerzoni M.A. and

Wiedmann M. 2001. Automated Ribotyping using differ-ent en4/mes to improve discrimination of Listerta mano-cgtogenes isolates with a particular focus on serotype 4b strains. J. of Clin. Microbiol. 39, (8), 3OO2.

De Cesare A., Mioni R. and Manfreda G. 2OO7 . Prevalence of Listeriamonocytogenes in fresh and fermented sausages

and ribotyping of contaminating strains. Int. J. of Food

Microbiol. L2O, {l-2), 124.

Doumith M., Buchrieser C., Glaser P., Jacquet C. and

Mar-tin P. 2004. Differentiation of the major Listeria

mono-cAtagetEs serovars by multiplex PCR. J. of Clin. Micro-bior. 42, (8), 3819.

European Food Safety Authority, European Centre for

Disease Prevention and Control; The European Union Summary Report on Trends and Sources oT Zoonoses,

Zoonotic Agents and Food-borne Outbreaks in 200g.

2011. EFSA Journal 2011; 9 (3):2090. doi:I0.2908,tj.

efsa. 20 I l . 2090. Available online : www. efsa. europa. eu-/

efsajournal.

European Commission, Regulation (EC) No. 2O73/2OOS of

l5 November 2005 on microbiological criteria for food-stuffs. 2005. Off. J. Eur. U L 338,1.

FarberJ.M. and Peterkin P. 1991. ListeriamanocAtogenes,

a food-borne pathogen. Microbiol Mol. Biol. Rev. 55, (3),

476.

Fugett E.B., Schoonmaker-Bopp D., Dumas N.8., Corby J. and WiedmannM. 2OO7. Pulsed-Field Gel Electrophore-sis (PFGE) _{analysis o[temporally matched} Ltsteriamono-cAtogenes isolates from human clinical cases, foods,

ru-minant farms, and urban and natura-l eru,lronments

re-veals source-associated as well as widely distributed

PFGE types. J. of Clin. Microbiol. 45, (3), 865.

Gandhi M. and Chlkindas M.L. 2OO7. Listeria: a foodborne pathogen that knows how to suMve. Int. J. of Food

Micro-biol. 113, (1). 1.

Graves L.M. and Swaminathan B. 2OOl. PulseNet stand-ardized protocol for subtyping Lis brta monocgtogenes by macrorestricuon arld pulsed fleld gel electrophoresis. Int.

J. of Food Microbiol. 65, (1-2), 55.

Graves L.M., Hunter S.B., Ong A.R., Schoonmaker-Bopp D., Hise K., Kornstein L., DewittW.E., Hayes P.S., Dunne E.,

Mead P. and Swaminathan B. 20O5. Microbiological as-pects ofthe investigation that traced the 1998 outbreak of listeriosis in the United States to contaminated hot dogs and establishment of molecular subtyping-based survelllance for Listerio monocAtogenes in the PulseNet network. J. of Clin. Microbiol. 43, (5), 2350.

Greco M., Ma?zetle R., De Santis E.P.L., Corona A. and

Cosseddu A.M. 2OO5. Evolution and identiflcaflon of

lac-tic acid bacteria isolated during the ripening of Sardini-an sausages. Meat Sci. 69, (4), 733.

Ilsi Research Foundation-Risk Science Institute - Expert

Pan-el ofl Listeria monocAtogenes infoods, Achieving

continu-ous improvement in reductions in foodborne listeriosis - a

risk-based approach. 2005. J. ofFood Prot. 68, (9), 1932.

Jofré A., Martin 8., Garriga M., Hugas M., Pla M.,

Rodrigu-ez,-Làzaro D. and Al,rnerich T. 20O5. Simultaneous

detec-tion of Listerra monocatogenes and Salmonella by

mul-tiplex PCR in cooked ham. Food Microbiol. 22, (l), lO9. Kim K.Y. and FrankJ.F. 1995. Effect of nutrients on biofflm formation by Listeria monocgtogenes on stainless steel.

J. ofFood Prot. 58, (11.24.

L6puY., Villator D., Oertiz 5., l-opez P., Navas J., Dàvila

J. C. and Martinez-Suar ez J .Y. 2OO8. Molecular tracking of Listeria rnonocAtogenes in an Iberian pig abattoir and processing plant. Meat Sci. 78, (1-2), 130.

Mafu A.A., Roy D., Goulet J. and Magny P. 1990. Attach-ment of Listena fiDnocAtogenes to stainless steel, glass,

polypropylene and mbber surfaces after short contact times. J. of Food Prot. 53, (9),742.

Mc l.auchlin J., Mitchell R.T., Smerdon W.J. and Jewell K.

2OO4. ListeriamanocAtogenes and listeriosis: a review of

hazard characterization for use in microbiological risk as-sessment offoods. Int. J. ofFood Microbiol.92, (1), 15.

Meloni D., GalhtzzoP., MuredduA., Piras F., Grifffths M. and Mazzelte R. 2009. Listeria filonocAtogenes in RTE foods

marketed in Italy: prevalence and automated EcoRI

ri-botyping of the isolates. Int. J. of Food Microbiol. 129,

(2), 166.

Nesbakken T., Kapperud G. and Caugant D.A. 1996.

Path-ways of Listeria monocAtogenes contamination in the

meat processing industry. Int. J. of Food Microbiol. 31,

(l-3), 161.

Public Health Agency Canada, Lessons Learned Report:

The Canadian Food InspectionAgency's Recall Response

to the 2OO8 Listerlosis Outbreak. 20O9. Available

on-line: http: //www.phac-aspc.gc.calfs-sa/listeria /

2OO8-lessons-lecons-eng.php.

Ross R.P., Morgan S. and Hill C. 2OO2. Preservation and fermentatlon: past, present and future. Int. J. of Food

Microbiol.79, (l-2), 3.

Samelis J. and Metaxopoulos J. 1999. Incidence and

prin-cipal sources of Listena spp . and Listerta monocAtogenes

contamination ùr processed meats and a meat process-ing plant. Foed Microbtol. 16, (5), 465.

Simon M.C., Gray D.I. and Cook N. f 996. DNA extraction

and PCR methods for the detection of Listeria

rrr.noca-togenes in cold-smoked salmon. Appl. Environ. Micro-biol. (62),822.

Stepanovic S., Cirkovic I., Ranin L. and Svabic-Vlahovic S.

2004. Biofllm formation by Salmonella spp. and Listerio

fitonocAtogenes on plastic surfaces. Lett. lnAppl. Micro-biol. 38, (5't,428.

Tenover F. C., Arbeit R. D., Goering R.V., Mickelsen P.A., Mur

-ray B.E., Persing D.H. and Swaminathan B. 1995.

Inter-preting chromosomal DNA restriction patterns produced

by Pulsed-Field Gel Electrophoresis: criteria for bacterial strain typing. J. of Clin. Mierobiol. 33, (9), 2233. Thévenot D., Delignette-Muller M.L., Christieans S. and

Vernozy-Roland C. 2005. Preva-lence of

ListeriamonocA-togenes in 13 dried sausage processing plants and their

products. Int. J. ofFood Microbiol. 102, (1),85. Thévenot D., Delignette-Muller M.L., Christieans S., Leroy

S., KodJo A. and Vernozy-Rolaad C. 2006'. Serological and molecular ecolory of Lùsteriamonocgtogenes isolates

collected from l3 French pork meat salting-curing plants and thelrproducts. Int. J. of Food Microbiol. I12, (2), 153.

Thévenot D., Dernburg A., Christieans S. and

Vernozy-Ro-Iand C. 2006'. An updated review of Listeria

ffanocA-togenes ln the pork meat industry and its products. J. of Appl. Microblol. 101, 7,

Tompkln R.B. 2002. Control of Listeria monocAtogenes in the food-processing environment. J. ofFood Prot. 65, (4), 709.

Yazqwez-Boland J.A., Kuhn M., Berche P., Chakraborty T.,

Domingues-Bernal G., Goebel W., Gonzalez-Zorn 8.,

Wehlamd J. and Kreft J. 2OOl. Listeria pathogenesis and molecular virulence determinants. Clin. Microbiol. Rev. 14, (3), 584.

Zhao T., Doyle M,P. and Zhao P. 2OO4. Control of Listeria nwrocAtagenes in a biofilm by competitive-exclusion mi-croorganisms. Appl. and Environ. Microbiol, (7), 3996.

Paper received August 3,2011

222

ltal. J. Food Sci., vot.24 - 2012