| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
INRA, Infectiologie Animale et Santé Publique, 37380 Nouzilly, France
Correspondence
Etienne Zundel
zundel{at}tours.inra.fr
Received 5 May 2006
Accepted 12 August 2006
Abbreviations: GIT, gastrointestinal tract; MPN, most probable number; PI, post-inoculation; ROI, region of interest.
Scintigraphic imaging of the spread of L. monocytogenes radiolabelled with 111In oxine in sheep nos 5 and 6 is shown in Supplementary Fig. S1 with the online version of this paper.
 |
INTRODUCTION |
|---|
 TOP INTRODUCTION METHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
The present study was designed to localize the sites of infection in the digestive tract concomitant with Listeria faecal excretion in a sheep model. Radiolabelled or non-radiolabelled L. monocytogenes was orally administered to animals, and dissemination was assessed by in vivo imaging, bacteriological detection and gamma radioactivity measurements at slaughter. Our results indicate that brief and low-level faecal excretion of L. monocytogenes is concomitant with a transitory asymptomatic infection in sheep, after translocation from every segment of the gastrointestinal tract (GIT), with the rumen digesta serving as a reservoir.
|
METHODS |
|---|
|
TOPINTRODUCTIONMETHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
L. monocytogenes inoculum and labelling with 111indium oxine. The LCCN 95-962 L. monocytogenes strain serotype 1/2a had previously been isolated in our laboratory from a ewe with clinical listeriosis. In vivo and in vitro virulence tests (Roche et al., 2001) have confirmed the virulence observed in the field. We labelled bacteria with 111In oxine (half-life 67 h), because radioactivity is incorporated very rapidly (90 % within 10 min) with little damage to bacteria, and 111In released from the cells would not be reutilized (Ardehali & Mohammad, 1993). In brief, a 24 h culture on trypticase soy agar (TSA; bioMérieux) was suspended, washed and spectrophotometrically titrated to 4x1010 c.f.u. ml–1 in 4.5 ml PBS. The bacterial suspension was centrifuged (4000 g, 4 °C for 30 min). The supernatant was discarded and the pellet was mixed with 0.2 ml 74 mBq 111In oxine (Mallinckrodt). The solution was corrected to pH 6 with 0.2 M Tris. After 15 min at room temperature, the reaction was stopped with 20 ml PBS with 10 % fetal calf serum (Invitrogen). The labelled bacteria were washed twice in PBS by centrifugation. We achieved a labelling efficiency of 64 % (percentage of the initially added radioactivity), which corresponds to an excellent incorporation of radiolabel in bacteria (Perin et al., 1997). Radiolabelled bacteria were resuspended in 0.110 ml PBS, and 10 µl was used for enumeration of viable organisms. The viability of the radioactive inoculum was not different from that of mock-labelled bacteria that had been processed under the same conditions, except for the absence of the radioactive oxine.
Cold and radiolabelled inocula were adjusted to 1011 c.f.u. ml–1. In order to mimic natural oral contamination, cotton swabs (bioMérieux) were impregnated with 100 µl inoculum (1011 c.f.u. ml–1) and fed to the two animals selected for in vivo imaging on D0, 2 h after normal feeding.
Scintigraphic imaging. We chose scintigraphy to monitor the spread of bacteria because it is a non-invasive technique usable in vivo. The sheep were anaesthetized with a 5–2 % halothane in air influx at 2 l min–1. On D0 [30 min and 4 h post-inoculation (PI)], D1, D2, D3 and D6, ventral and lateral (right decubitus) 2–10 min static images were acquired using a gamma camera (Opti-CGR) equipped with a high resolution, low-energy, parallel collimator. Data were recorded using a 15 % window centred on the 170 keV photopeak of 111In into a 256x256 pixel matrix on a computer system specialized in digital display and analysis (Mirage, Segami). Distribution of radioactive bacteria was assessed from static images using regions of interest (ROI) for different body areas (head, thorax, abdomen and pelvis). After correction for radioactive decrease, the count rate recorded within these regions was used to calculate the percentage of radioactivity injected on D0 for each ROI, and to edit images (available as Supplementary Fig. S1 in JMM Online).
Sample radioactivity at slaughter. Sheep nos 5 and 6 were slaughtered on D6, and samples were collected as described below. Immediately before culturing for Listeria, 1 ml of each sample suspension in half-Fraser broth was counted (Cobra gamma counter, Packard). The radioactivity was expressed as disintegrations per minute per gram of sample (d.p.m. g–1). The residual radioactivity in organs at slaughter indicated the presence of live or formerly live bacteria. By counting the radionuclide and bacterial counts in serial dilutions of the labelled suspension, it was determined that a count of 3 d.p.m. corresponded to approximately one viable bacterium. The lower limit of detection of the gamma-counting system was 50–100 bacteria, depending on the background radioactivity level. Survival of bacteria was defined as the ratio of (numbers of Listeria counted on plates)/(numbers of Listeria calculated from d.p.m. counts) (Eaves-Pyles & Alexander, 2001). A ratio below one suggested that the overall number of bacteria decreased after translocation, and above one that it increased.
Sample collection.
Sheep were subsequently monitored for symptoms, and their faeces, feed and drinking water were checked for Listeria once a month for 6 months before inoculation. Faeces and saliva were regularly taken in the morning before feeding, from live animals after inoculation (Table 1
). Faeces were collected directly from the rectum, and placed in individual screw-capped specimen bottles, with gloves changed before each individual collection. Saliva samples (125 µl maximum volume) were taken with a plastic-rod rayon swab (Coplan). Samples were processed in our laboratory within 30 min of collection. Two sheep per day were euthanized with sodium thiopental on D1, D2, D6, D10 and D14 PI. Samples (up to 25 g when possible) were aseptically collected from 25 organs and digesta from all ten sheep. GIT segment samples included the oesophagus, rumen, duodenum, jejunum, ileum, caecum and colon (distal loop). A digesta sample was collected from each GIT segment, except the oesophagus. GIT tissue from a 20 cm (ileum) to 60 cm (jejunum) sample was rinsed with saline to remove visible digesta. Palatine tonsils, retropharyngeal and mesenteric lymph nodes as local lymphoid organs, and liver, spleen and blood [about 20 ml in evacuated blood-collecting tubes with lithium heparin (Venoject; Terumo)] as systemic organs were harvested.
|
Listeria from rumen digesta samples were counted using the most probable number (MPN) method (Blodgett, 2001) in Fraser broth tubes. Listeria from saliva were counted with a direct plating method on TSA, and those from other samples were serially diluted and plated on Oxford plates (AFNOR, 1997). Quantitative results were expressed in c.f.u. g–1. When necessary, Listeria-like colonies were streaked on TSA plates and confirmed by examining the colonies under obliquely reflected light (McClain & Lee, 1988), and by standard methods: catalase reaction (positive), haemolysis of horse blood agar (bioMérieux), Christie–Atkins–Munch–Petersen (CAMP) test, and acid production from D-xylose (negative), L-rhamnose (positive) and mannitol (negative). The results were confirmed with API Listeria (bioMérieux) when necessary.
Because of dilution factors, the lowest detectable counts were 1 c.f.u. per 25 g (MPN), 50 c.f.u. per g or ml (direct plating) and 200 c.f.u. per ml (direct plating with saliva). The limit of detection of the enrichment technique was 1 c.f.u. per 25 g. Therefore samples positive after enrichment contained between 0.04 and 50 or 200 c.f.u. g–1. To facilitate reading of graphical representations, we arbitrarily assumed enrichment results to correspond to 10 c.f.u. g–1 (Fig. 1).
|
|
RESULTS AND DISCUSSION |
|---|
|
TOPINTRODUCTIONMETHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
Listeria excretion
Listeria were not recovered from the faeces, feed or drinking water prior to inoculation. The experimental inoculation resulted in a low-level excretion in saliva for 13 days, and in faeces for a shorter time: up to 10 days (Table 1
), in line with previous results (Lhopital et al., 1993). In another study, a 104 or 106 c.f.u. dose in sheep appeared to have no effect on faecal excretion (Zundel et al., 2006). All ten animals remained well, without symptoms, throughout the study, even when they shed Listeria, and thus were considered as asymptomatic carriers.
Spread of Listeria
After inoculation, scintigraphic images taken successively from the ROI were grouped in one figure (Supplementary Fig. S1). Body outline was added to the images (white line). The head and neck showed very little labelling 30 min or 4 h post-inoculation in sheep no. 5. The white spot in its rumen (arrows, Supplementary Fig. S1) corresponds to the swab head containing the inoculum, swallowed without chewing, as is usual for ruminants. It was seen again on D1 in the right subcostal area, and on D2 in the lower left side of the belly. The mouth of sheep no. 6 was labelled immediately after inoculation. Its inoculum swab having been chewed, Listeria spread more rapidly and more efficiently in its forestomachs during the first 4 h. In comparison, Escherichia coli O157 : H7 was isolated from the duodenum digesta 1 h after 2x1011 c.f.u. E. coli was directly inoculated into the rumen of steers with cannulae inserted into both the rumen and duodenum (Grauke et al., 2002). But precise tomographic localization could only have been obtained by single photon emission computer tomography (SPECT), and was not possible using planar scintigraphic acquisition, although the camera used in this study facilitated quantitative analysis. The spread of L. monocytogenes in the lumen of the GIT, as seen with scintigraphy, is concomitant with bacterial translocation throughout the different segments of the GIT, as observed by culturing samples collected within 48 h of inoculation (Fig. 1). It is worthy of note that detectable Listeria dissemination was maximal and the rectum was labelled on D1 for both sheep (dashed lines, Supplementary Fig. S1).
Listeria translocation to deep organs
Bacteriological analyses at slaughter confirmed the presence of Listeria in all samples of two sheep within 24 h of inoculation, with the exception of blood and bile. The mean numbers of Listeria in samples are shown in Fig. 1 for digesta, GIT walls and lymphoid organs. The greatest number of Listeria in the GIT was approximately 105 c.f.u. per gram of digesta on D1 and of walls on D2. Very few Listeria were isolated from digesta and walls in the distal part of the GIT. Listeria were recovered from mesenteric lymph nodes from D1 onwards, with a maximum number on D2 (Fig. 1). Numbers in the tonsils and the retropharyngeal lymph nodes were higher than those of mesenteric lymph nodes, probably because the doses were given on a swab and not with a worming gun. A few bacteria reached the liver and spleen (<50 c.f.u. g–1), as systemic lymphoid organs. Thus our study revealed the ability of Listeria to translocate from every segment of the GIT.
Samples collected at slaughter (D6) from sheep nos 5 and 6 were measured for gamma radioactivity and bacterial enumeration (Fig. 2). Samples from sheep no. 6 were more often contaminated or infected with Listeria than those of sheep no. 5, although numbers were close to the lower detection limit. There was no real difference between the Listeria counts in the lymphoid organs of the head, although sheep no. 5 did not chew the inoculum swab. Twelve samples (saliva, blood, coeliac, jejunal, ileo-caecal and colic lymph nodes, oesophagus, rumen, duodenum, jejunum, ileum and colon tissues) had very low radioactivity, close to the background level, in these two sheep. Therefore, these samples were considered ‘double-negative’. The scatter diagram (not shown) of the radioactive samples from faeces, GIT digesta, caecum wall, palatine tonsils, retropharyngeal lymph nodes, spleen, liver and bile showed that both sheep dealt with L. monocytogenes in the same way (R2=0.932).
|
). Radiolabelled bacteria differentiated the role of the wall permeability to bacteria of the GIT lumen from that of the immune clearance of bacteria (Gianotti et al., 1995). The low to moderate radioactivity on D6 in the liver and spleen confirmed that Listeria cultured on D1 and D2 (Fig. 1) were eliminated. Therefore, our results show that the efficiency of Listeria translocation was low to moderate in sheep. The efficiency of listerial penetration is lower in mice (Marco et al., 1992) and in a rat ligated-ileal-loop model (Pron et al., 1998). The main point is that GIT lymph nodes were initially sites of temporary multiplication and then became sites of lysis and elimination of Listeria, in the same way as the liver, spleen and intestinal lumen.
Listeria systemic infection
The presence of L. monocytogenes in lymphoid organs, including spleen and liver, confirmed the systemic though subclinical course of infection after oral inoculation. An important finding was that no Listeria were isolated from bile, although sheep bile experimentally inoculated (2x102 c.f.u. ml–1) and incubated (24 h, 37 °C) allowed L. monocytogenes LCCN 95-962 to grow up to 2x108 c.f.u. ml–1. The same growth was observed in C57/Bl6 mice bile. Therefore it does not seem that excretion with bile would cause faecal excretion of Listeria in sheep, as shown in mice with a sublethal dose (Briones et al., 1992; Marco et al., 1997), or an enterohepatic circulation of L. monocytogenes (Hardy et al., 2004). Radioactivity in tonsils and retropharyngeal lymph nodes was moderate, but Listeria counts remained high for 2 weeks, thus suggesting that bacteria increased in number in these lymphoid organs, which can therefore be considered as sites of multiplication (Fig. 2). This result is strengthened by the particularly high survival ratio of L. monocytogenes (over 400) in tonsils. Tonsils provide good evidence of food-borne contamination, as shown in pigs, in which 22 % of tonsils have been demonstrated to be Listeria positive in different slaughterhouses (Autio et al., 2004). Tonsils can also be used by bacteria such as Salmonella Typhimurium as a site of chronic infection up to 140 days after inoculation, resulting in asymptomatic carrier animals (Horter et al., 2003). However, under our conditions, the sheep tonsils were not able to contribute significantly to Listeria faecal excretion, since they were situated upstream of the GIT, and they remained infected at up to a total count of 5x104 c.f.u. after the faecal excretion of Listeria had stopped.
Persistence of Listeria infection
Listeria numbers in tonsils and retropharyngeal lymph nodes were stable until D14, probably due to local growth of bacteria and rumination. Ruminants regurgitate foodstuff from the rumen, and chew it again for 8 h per day. In fact, Listeria continued to be isolated from rumen digesta and the duodenum wall on D14. There were more than 1.1x104 Listeria g–1 in rumen digesta (volume 
10 l) on D1, decreasing to 15 c.f.u. g–1 on D14. Radioactivity in rumen digesta was moderate, with 103 c.f.u. ml–1 at D6, i.e. at least 107 c.f.u. Listeria in the total content, with a survival ratio of 9. The renewal of the rumen content is continuous, and takes on average about 20 h for dry matter (Cannas et al., 2003). Therefore, Listeria probably multiplied to make up in part for the permanent through-flow, the negative effects of interaction with rumen flora (around 106 ciliate protozoa, 108–1011 bacteria and 108 bacteriophages per millilitre), and even anti-Listeria bacteriocins (Laukova & Czikkova, 1998). Conversely, E. coli O157 : H7 does not seem to persist in the sheep rumen (Grauke et al., 2002). Therefore, a remarkable finding in our study was that the rumen content, like tonsils and retropharyngeal lymph nodes, was a long-lasting site of Listeria multiplication. The rumen can thus be considered as a reservoir for L. monocytogenes, discharging Listeria downstream into the GIT for at least 2 weeks after a single per os inoculation. However, faecal excretion of Listeria stopped before the bacterium disappeared from the head lymphoid organs, the rumen content and the duodenum wall (the enrichment limit of detection was 1 c.f.u. per 25 grams). These data support the hypothesis that a threshold number of Listeria is needed at the proximal part of the bowel to obtain Listeria in faeces. On the farm, the rumen can thus buffer heavy food-borne contamination, as in heterogeneous silage (Fenlon, 1999), and contribute to a low-level excretion of L. monocytogenes in faeces.
In conclusion, our results strongly suggest that the sporadic, intermittent and short-term faecal excretion of L. monocytogenes in ruminants given contaminated food on the farm is not a simple passage of the bacteria, but is associated with a transitory multiplication in the rumen content and a transitory infection, i.e. an asymptomatic carrier state.
|
ACKNOWLEDGEMENTS |
|---|
|
REFERENCES |
|---|
|
TOPINTRODUCTIONMETHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
ebam/bam-a2.html (accessed 3 November 2003).
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| INT J SYST EVOL MICROBIOL | J MED MICROBIOL | MICROBIOLOGY | J GEN VIROL | ALL SGM JOURNALS |
