| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Laboratório de Microbiologia, CBAA/DBEB, Instituto Superior de Agronomia, Technical University of Lisbon, Tapada da Ajuda, 1349-017 Lisbon, Portugal
2 Instituto Superior de Estudos Interculturais e Transdisciplinares (ISEIT), Campus Universitário de Almada, Instituto Piaget, Quinta da Arreinela de Cima, 2800-305 Almada, Portugal
3 Institut National de la Recherche Agronomique, Infectiologie Animale et Santé Publique, UR1282, F-37380 Nouzilly, France
Correspondence
Luisa Brito
lbrito{at}isa.utl.pt
Received 2 October 2007
Accepted 11 December 2007
Abbreviations: PFA, plaque-forming assay.
 |
INTRODUCTION |
|---|
 TOP INTRODUCTION METHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
The purpose of this work was to evaluate the virulence of L. monocytogenes isolated from cheese and from its production environment (including the persistent strain) by comparing it with the virulence of isolates from different origins, including human cases of listeriosis. A previously described in vitro test of virulence, based on a plaque-forming assay (PFA), and the enumeration of viable bacteria in spleens after the subcutaneous inoculation of mice (Roche et al., 2001), were the tests used to investigate the pathogenic potential of the isolates.
|
METHODS |
|---|
|
TOPINTRODUCTIONMETHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
shows the characterization of the 51 L. monocytogenes isolates belonging to seven serovars (1/2a, 1/2b, 1/2c, 3b, 4a, 4b and 4d/4e), including strains from milk (n=6), from cheese (n=14), from cheese dairy environment (n=2), from milking environments (n=1), from delicatessen meats (n=5), from food products (n=1), from human cases of listeriosis (n=18), from animal cases (n=2) and those with uncertain origin (n=2).
|
PFGE. The 19 L. monocytogenes isolates, that were not previously pulsotyped, were characterized in this work by DNA macrorestriction analysis using PFGE in accordance with the Pulse-Net standardized protocol (Graves & Swaminathan, 2001). Bacterial cultures were embedded in chromosomal-grade agarose (SeaKem Gold agarose; Cambrex), lysed, washed and digested in situ overnight in separate reactions with 10 U AscI (New England Biolabs) and 20 U ApaI (Roche Diagnostics) at 37 and 30 °C, respectively. Resolution of the generated DNA fragments was obtained using 1 % SeaKem Gold agarose gels in 0.5xTris/borate/EDTA buffer (Invitrogen), run at 14 °C and 6 V cm–1 with time ramped from 4 to 40 s over 22 h using a CHEF DR II System (Bio-Rad). Gels were stained with ethidium bromide at a final concentration of 10 µg ml–1 (Sigma) and pattern images were acquired with a Bio-Rad Gel Doc 2000 imaging system. The TIFF images were normalized by aligning the peaks of the size-standard strains Salmonella enterica serovar Braenderup H9812 and/or L. monocytogenes CLIP 77873, which were loaded on three lanes in each gel. These strains were kindly provided by the Centre National de Référence des Listeria, Institut Pasteur, France. For each restriction enzyme, if two banding patterns differed by a single band, these patterns were considered as two different pulsotypes. The isolates were numbered after comparison with the former 25 pulsotypes described by Leite et al. (2006).
In vitro test of virulence
Cell line and culture conditions.
The human adenocarcinoma cell line HT-29 (ECACC 850611109) was used between passages 20 and 44. Cells were routinely grown in 75 cm2 flasks (Orange Scientifique) in complete medium D-MEM (Dulbecco's modified Eagle's medium: high glucose, with L-glutamine, without sodium pyruvate, sodium bicarbonate) (Gibco/Invitrogen) supplemented with 3.7 g sodium bicarbonate l–1 (Merck), 10 % (v/v) fetal calf serum (Gibco/Invitrogen) and 2 mM L-glutamine (Gibco). Penicillin (100 IU ml–1) and streptomycin (100 µg ml–1) (both from Sigma) were always added to the culture medium, except for the medium used 24 h prior to the virulence assay. Cells were kept in a humidified atmosphere using an incubator, at 37 °C under 5 % (v/v) CO2 (ShelLab).
Plaque-forming assay (PFA). PFA was performed according to Roche et al. (2001). Briefly, HT-29 cells were trypsinized and 3x104 cells were deposited per well in a 96-well tissue culture plate (Nunc). To obtain confluent monolayers, the plates were incubated for 3 days with antibiotics followed by incubation for 24 h without antibiotics.
Listeria strains grown overnight on TSA-YE were then suspended in complete medium to a concentration of 4x108 cell ml–1. HT-29 cell monolayers were infected with a dilution series of 102 to 107 L. monocytogenes per well, and incubated for 2 h at 37 °C. The number of viable bacteria was assessed by duplicate plating of the appropriate dilutions onto TSA-YE. Duplicates were incubated at 37 °C for 24 h before counting colonies.
After removing the infecting bacterial suspensions, cell monolayers were incubated for 1.5 h with complete medium containing 100 µg gentamicin ml–1 (Sigma) without penicillin/streptomycin. Each well was then covered with complete medium supplemented with 10 µg gentamicin ml–1 and containing 2.5 % (m/v) agarose (Invitrogen). Subsequently, the same liquid medium was added to the top of the agar media to prevent cell starvation. Culture plates were incubated for 24 to 48 h at 37 °C under 5 % (v/v) CO2. Enumeration of formed plaques was performed 24 h after bacteria were deposited on the HT-29 cell monolayer, using an inverted microscope (Leica). After 48 h of incubation, enumeration of the plaques was confirmed.
Mouse virulence assays. Fifteen strains belonging to serovars 1/2a, 1/2b, 4a and 4b were also tested for pathogenicity by subcutaneous injection (50 µl) into immunocompetent 7-week-old Swiss female mice (Charles River), basically according to Audurier et al. (1980). Mice were housed five per cage on sterilized wood shavings and provided with water and sterilized food ad libitum. For each isolate, groups of five mice were inoculated subcutaneously (s.c.) into the left hind footpad.
L. monocytogenes isolates were grown on BHIA slopes (Difco, Becton Dickinson) for 17 h at 37 °C. Bacteria were harvested and suspended in sterile PBS (pH 7.3). The culture was standardized turbidimetrically and diluted appropriately to obtain 4 log c.f.u. in 50 µl for s.c. injection. The number of viable L. monocytogenes cells injected was verified by plating on TSA just before administration.
Three days post-inoculation, mice were sacrificed by cervical dislocation and spleens were removed aseptically. Homogenate samples were appropriately diluted in PBS and plated onto TSA. Animal handling and care conditions were consistent for all experimental runs.
Expression of the results and data analysis. For the in vitro test, the pathogenic potential of the isolates was expressed as the mean log of the number of plaques formed (for 1x107 Listeria per well) (log PFA), in duplicate, from at least two independent trials. Strains EGDe (serovar 1/2a) and 442 (serovar 4d/4e) were used in the PFA as references for virulent and low-virulence strains, respectively. The previously reported low virulence of strain 442 (Roche et al., 2005) was the reason for its use as a reference for low virulence.
ANOVA of the log PFA values from the plaque-forming isolates was carried out using least significant differences (LSD) post hoc multiple comparison tests by running the program Statistica, version 6 (Statsoft).
For the mouse virulence assay, the virulence potential of the isolates was expressed as the mean log of the number of colony-forming units (log c.f.u.) per spleen, in addition to the ratio of infected mice to inoculated mice (I/T) in the s.c. test.
|
RESULTS AND DISCUSSION |
|---|
|
TOPINTRODUCTIONMETHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
). There were isolates with the same pulsotype but from different origins. This was the case for isolates 7–11 (serovar 1/2b) from delicatessen meats with the same pulsotype 3 as isolate 15 from cheese (Table 1). There was no evidence that these isolates from delicatessen meats were related. There was no association between the delicatessen meats and the cheese that were the source of these isolates. Furthermore, the five isolates showing pulsotype 6 (serovar 4b) were collected from three distinct cheese dairies: isolates 24 and 25 were from cheese produced by a particular cheese dairy; isolates 29 and 36 were collected in a second cheese dairy, from the dairy environment and milk, respectively; and isolate 27 was collected from the dairy environment of a third cheese dairy.
Evaluation of virulence
All the isolates showed haemolytic activity in HBA, but strain 442 produced weak haemolysis. In the PFA, the plaques were enumerated after 24 h, with confirmation after 48 h. When the plaques were large and well defined, counting after 24 h was usually enough. When the plaques were diffuse and small (for instance for EGDe strain), confirmation of the number of plaques was done after 48 h. In general, the results matched and this confirmation was done routinely for confidence. The relative size of the plaques was not recorded in this study, but small as well as large plaques were found among isolates of not significantly different (P<0.05) virulence.
The mean log PFA value determined in this study for the EGDe reference strain was 5.87±0.07 (measured from 46 independent replicate tests) (Table 1). The value reported for this strain by Roche et al. (2001) was 6.32±0.12 and Gudmundsdóttir et al. (2006) reported a mean value of log PFA of 4.88±0.38 (measured 69 times). Based on these results, these authors reported cut-off values for virulence in the PFA assay of 3.34 (Roche et al., 2001) and 1.9 (Gudmundsdóttir et al., 2006). Above the cut-off value, isolates are considered virulent and below this value isolates are considered low-virulence. The values of the mean log PFA for EGDe reported by different laboratories were the reason why no cut-off value for virulence was established in this work. Instead, ANOVA of log PFA values obtained for the plaque-forming isolates was performed in order to compare their virulence with the reference strains for virulent (EGDe) and low-virulence (442) strains used in the assays.
Strain NCTC 7973 (serovar 1/2a) formed no plaques. This strain was also incapable of spleen colonization in mice. Nevertheless, Liu et al. (2003) reported strain NCTC 7973 as one of the most virulent strains tested (LD50<108). Pine et al. (1987) reported the existence of two phenotypes for this strain, haemolytic virulent colonies and non-virulent colonies. The authors suggest that the avirulent strain is a non-haemolytic mutant of the virulent strain. However, our isolate showed haemolytic activity. This is unusual behaviour. In their review, Vázquez-Boland et al. (2001) mentioned the anomalous behaviour of NCTC 7973 as a consequence of the accumulation of several regulatory mutations. Roberts et al. (2005) proposed the classification of L. monocytogenes into four distinct virulence attenuation groups (A, B, C and D). Group D isolates were characterized by normal cytotoxicity, haemolytic activity and hlyA expression but had ‘small’ to ‘no-plaque’ phenotypes. The genetic mechanisms that cause virulence-attenuated phenotypes among group D isolates could not be determined definitively. Therefore it is possible that this isolate of NCTC 7973 can be assigned to this virulence attenuation group.
Isolates 2 and 3 (serovar 1/2a, from milk) were not significantly different (P<0.05) from the low-virulence strain (442) used as reference. The remaining isolates were significantly more virulent than strain 442. Thirty-three isolates were not significantly different (P<0.05) from the virulent strain used as reference (EGDe). Nine isolates were significantly more virulent (highly virulent) than EGDe and seven isolates were significantly less virulent than EGDe. Within these seven isolates, five were also significantly more virulent than the low-virulence 442 strain. The nine highly virulent isolates were either from cheese (isolate 35), from cheese dairy environments (isolates 27 and 29), from human cases (five) and the reference strain for serovar 1/2b (CECT 936).
Fifteen isolates selected according to different levels of pathogenicity in the PFA were also tested by s.c. inoculation in mice. After inoculation of five mice with strain ATCC 19114, it was possible to recover L. monocytogenes from the spleen of one of the animals. This strain was previously tested by intraperitoneal (i.p.) injection and the LD50 was determined (Liu et al., 2003; Cabrita et al., 2004). The high values of LD50 obtained by these authors (3 logs higher than the LD50 values for the virulent strains tested, in each study) confirm the present results from the in vivo tests. Results from Liu et al. (2003) indicate the absence in strain ATCC 19114 of eight potential virulence genes (including L. monocytogenes transcriptional regulator and internalin genes). Notwithstanding, in the PFA assay this strain, although significantly less virulent than EGDe, was also significantly more virulent (P<0.05) than the low-virulence reference strain 442. Isolates 2 and 3 (serovar 1/2a), which showed low virulence in the PFA assays, were virulent in the s.c. inoculation of mice. These two isolates were also previously characterized as virulent, following i.p. injection in mice (Cabrita et al., 2004). The agreement between a PFA with HT-29 cells and infection of immunocompetent mice was previously showed by Roche et al. (2001) and confirmed in this study for the majority of the strains tested. The discrepancy between the two models probably occurs with specific isolates. Therefore, the combination of different tests allows a better characterization of the pathogenic potential of the isolates.
In conclusion, all the isolates from food and food-related environments were potentially virulent or highly virulent. The nine highly virulent isolates (by PFA) were either from humans, from cheese or from cheese dairy environments. Of particular concern was the fact that three isolates (24, 27 and 29) of a strain from serovar 4b (pulsotype 6), found persistently in cheese and in the environments of two different cheese dairies (Leite et al., 2006), were virulent or highly virulent in the in vivo and in vitro test, respectively. These results stress the risk of listeriosis associated with the consumption of cheese contaminated with L. monocytogenes.
The implementation and maintenance of good manufacturing practices (GMPs) together with sanitation standard operating procedures (SSOPs) all along the food chain will contribute to the effective eradication of persistent strains of this pathogenic bacterium.
|
ACKNOWLEDGEMENTS |
|---|
|
REFERENCES |
|---|
|
TOPINTRODUCTIONMETHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| INT J SYST EVOL MICROBIOL | J MED MICROBIOL | MICROBIOLOGY | J GEN VIROL | ALL SGM JOURNALS |
