Dendritic cell function is perturbed by Yersinia enterocolitica infection in vitro

Abstract

Infection with Yersinia enterocolitica is the cause of intestinal or extraintestinal diseases. We investigated the role of dendritic cells (DC), the most potent antigen-presenting cell (APC), in the course of infection with Y. enterocolitica in vitro. For these studies, DC were isolated from human peripheral blood and infected with green fluorescent protein (GFP)-labelled Y. enterocolitica. Bacteria were found within DC by FACS analysis and viable bacteria could be cultured from lysed cells. Within 24 h after infection, DC upregulated CD83 and CD86 followed at day 3, indicating maturation of DC. In contrast, for MHC class II, a marked but transient downregulation was observed at day 3 after infection, and downregulation to a lesser extent for CD80 at day 5. To assess the immunostimulatory capacity of DC, viable infected and uninfected DC were incubated with autologous T cells in the presence of phytohemagglutinin A (PHA). T cell proliferation was significantly reduced at days 4–6 after infection but not thereafter, whereas nonpathogenic Escherichia coli was not able to mimick this suppressive effect of Y. enterocolitica. The same suppression could be observed when infected DC were used in a mixed leucocyte reaction with allogeneic T cells. Thus Y. enterocolitica is able to invade DC, does not induce necrosis or apoptosis, but affects maturation of DC. However, MHC class II-molecules are downregulated initially, which coincides with a diminished immunostimulatory capacity of DC infected with Y. enterocolitica. The diminished immunostimulatory capacity of DC following infection with Y. enterocolitica in vitro might impair or delay elimination of bacteria thereby contributing to pathogenesis of bacterial enteritis or extraintestinal manifestations such as reactive arthritis.

INTRODUCTION

Yersinia enterocolitica are worldwide occurring bacteria which account for approximately 1–3% of all cases of acute bacterial enteritis [1]. Usually, enteritis is self-limited and resolves without treatment except supportive therapy. Reactive arthritis, Reiter's syndrome, or erythema nodosum are possible secondary manifestations, but their occurrence does not require overt intestinal symptoms [2]. Pathogenicity of Y. enterocolitica is determined by plasmids, called pYV [3], which encode for outer membrane proteins, secreted proteins, regulatory proteins and a protein secretion machinery [4]. The plasmid-encoded proteins, which are secreted into the extracellular milieu, are called Yops (Yersinia outer proteins) [5]. Some pathogenic factors are believed to be chromosomally encoded [6], such as the two outer membrane proteins Inv and Ail, which are responsible for entry into mammalian cells. To date, invasiveness has been shown for a small number of cells only, e.g. CHO cells [6–8]. Finally, Yersinia produces a heat stable enterotoxin, which seems to account for diarrhea in association with Yersinia -infection [9].

After orogastric inoculation, the bacteria enter the Peyer's patches of the ileum, proliferate and disseminate in the lamina propria. The initial invasion is independent of plasmid encoded factors [10]. One of the mechanisms to alter regulation of immune and inflammatory responses, as demonstrated in murine macrophages, is to induce apoptosis by suppressing the cellular activation of NF-kB, which in turn inhibits production of TNFα [11]. For human B and T cells, adherence of Yersinia, but not invasion, was found in significant amounts as well [12].

Dendritic cells (DC) are the most potent antigen-presenting cells of the immune system. Their function is to initiate specific immune responses. After capture and processing of antigen in the periphery, DC move to the T-dependent areas of lymphoid organs to present and activate T cells [13–15]. Different DC populations have been isolated from the Peyer's patches of mouse intestine, which is one of the main sites of primary antigen contact, and from human colonic lamina propria [16–18]. These DC seem to be immature forms with weak accessory function for T cells, which develops after stimulation with GM-CSF [19]. DC are motile and they are able to cluster and activate T cells by mean of their abundance of surface markers, especially high levels of MHC class I and II products, ICAM-1, the lymphocyte costimulatory molecules CD80 and CD86, and CD83, a member of the immunoglobulin superfamily with unknown function.

Since the role of DC as APC has not been assessed following infection with Y. enterocolitica, we studied this aspect in more detail. Using DC isolated from human blood and bacteria labelled with the green fluorescent protein (GFP), we showed that Y. enterocolitica can infect DC thereby suppressing the immunostimulatory capacity of DC for T cells without induction of necrosis or apoptosis in DC. The diminished immunostimulatory capacity of DC can be correlated with downregulation of DC surface markers which are crucially involved in antigen-presentation and DC–T cell interaction.

MATERIALS AND METHODS

Isolation of dendritic cells and T cells

Peripheral blood mononuclear cells (PBMC) were obtained from buffy coats of healthy adult blood donors after informed consent from the Bavarian Red Cross (Würzburg, Germany) by density gradient centrifugation (FicoLite H; Linaris, Bettingen a.M., Germany). The interface cells were collected, washed and T cell-rosetting was performed with sheep red blood cells (BAG, Lich, Germany) pretreated with Vibrio cholerae neuraminidase (Sigma, St Louis, MO, USA) as described elsewhere [20]. Incubation for 60 min on ice was followed by another density gradient centrifugation over FicoLite. Sheep red blood cells of the pellet containing rosetting T cells were lysed with 0·8% (w/v) NH₄Cl (Sigma). The gradient interface cells were washed and incubated overnight in RPMI 1640 (Sigma) supplemented with 10% heat-inactivated foetal calf serum (FCS; Greiner, Frickenhausen, Germany) in plastic dishes. The nonadherent cells were layered over a hypertonic metrizamide (Sigma) gradient the next day, and buoyant cells were found to be DC at an approximate purity of 60–70% [20].

Bacterial strains

Y. enterocolitica O:3108P is a patient isolate harboring the virulence plasmid pYV and was obtained from J. Heesemann (Max von Pettenkoffer Institute, Munich, Germany). The nonpathogenic Escherichia coli strain used was isolated from human faeces, and was received from the culture collection of the institute for Hygiene and Microbiology of the University of Würzburg, Germany. Y. enterocolitica was grown for 24 h at 27°C [21], E. coli for 24 h at 37°C in tryptone-yeast broth. Prior to use in infection studies, bacteria were centrifuged and washed two times in PBS (phosphate buffered saline, pH 7·4). Labelling of Y. enterocolitica with the green fluorescent protein (GFP) was achieved by transformation with the shuttle plasmid pKSBCgfp-sod which carries the tetracycline resistance gene as a selection marker and the gfp mut2 gene [22] under the control of the listerial sod (superoxide dismutase) promoter. This promoter leads to constitutive expression of gfp in both Gram-negative and Gram-positive bacteria (A. Bubert, unpublished results). Transformation of yersiniae was performed using the CaCl₂ method following standard protocols [23]. The recombinant Yersinia strain was cultured in tryptone-yeast broth containing 10 µg/ml tetracycline. Infection of DC was performed at a multiplicity of infection (MOI) of 20 bacteria per cell for all bacterial preparations, if not indicated otherwise. Thirty min after infection, cells were centrifuged and resuspended in medium containing gentamicin (25 µg/ml; Sigma) and incubated for a period of 2 h in order to kill remaining extracellular bacteria. Subsequently, gentamicin concentration was adjusted to 4 µg/ml for the following assays. This gentamicin concentration prevents overgrowth of extracellular bacteria, but does not affect intracellular bacteria [7]. Uninfected controls were treated identically with the exception that the preparation added for infection was free from bacteria (mock-infection).

Assessment of cell viability and necrosis

Cell viability was determined by addition of fluorescein diacetate (FDA; Sigma: 100 ng/ml), necrosis was assessed by addition of propidium iodide (Sigma: 100 µg/ml). A volume of 300 µl cell suspension with Yersinia-infected and uninfected DC (1 × 10⁶ cells) was stained with 100 µl FDA and 100 µl propidium iodide. The proportion of viable versus necrotic cells was assessed by two-colour flow cytometry.

Analysis of apoptotic processes

1 × 10⁶ DC infected with Y. enterocolitica (MOI 20) or uninfected were fixed after 6 h of incubation with bacteriae with 4% phosphate-buffered paraformaldehyde (Sigma) and permeabilized with 0·1% Triton X-100 (Serva, Heidelberg, Germany) in 0·1% sodium citrate (Merck, Darmstadt, Germany) buffer. DNA breaks of apoptotic cells were stained with the in situ cell-death detection kit (Boehringer Mannheim, Germany) following the manufacturer's instructions. For positive control, fixed and permeabilized DC were treated with DNase I (Sigma, 500 µg/ml) for 10 min to induce DNA strand breaks. After staining, cells were analysed by flow cytometry.

Detection of viable intracellular bacteria

After the indicated time of incubation, 3 × 10³ Yersinia-infected DC (MOI 20) were lysed by addition of 0·5% (w/v) tergitol (Sigma) for 15 min at room temperature and plated on Müller-Hinton agar plates [8]. Colony-forming units were counted after an incubation period of 2 days at room temperature as described previously [7].

Flow cytometry

Uninfected and infected DC were analysed for cell surface expression of a variety of leucocyte markers by fluorescence-activated cell sorting (FACS) analysis using a FACScan flow cytometer (Becton Dickinson, Heidelberg, Germany). The antibodies were purchased from PharMingen (Hamburg, Germany) [anti-HLA ABC/FITC, anti-CD80, anti-CD86], from Dako (Hamburg, Germany) [anti-HLA DP,DQ,DR/FITC] and from Coulter-Immunotech (Krefeld, Germany) [anti-ICAM-1, anti-CD83 and the corresponding isotype controls]. 1–2 × 10⁵ cells were washed in PBS containing 0·1% BSA and incubated for 30 min on ice with the antibodies listed above at the indicated time points after isolation and culture of DC according to the conditions of the mitogen stimulation assay. Staining with CD80, CD83 and CD86 was followed by incubation with goat anti-mouse/FITC (Dianova, Hamburg, Germany). A minimum of 8000 events were analysed per sample. GFP-fluorescence of bacteria in live infected cells was measured by flow cytometry using an Epics Elite ESP cell sorter (Coulter, Krefeld, Germany). Using the 488 nm line of an argon ion laser, GFP-based fluorescence was quantified in the green light channel (525/10 nm bandpass filter) while the red light channel (630/20 nm bandpass filter) was used to exclude dead cells following staining with propidium iodide (2 µg/ml). A threshold of 0·1% was set to discriminate between GFP-positive and -negative cells using control cells infected with Y. enterocolitica without the gfp gene. All infections with the GFP-expressing Y. enterocolitica were microscopically analysed by using an epifluorescence microscope. Infections were documented using the Seescan digital imaging system (Intas, Göttingen, Germany).

Mitogen stimulation assay

Viable infected DC and uninfected, but otherwise identically treated DC (3 × 10⁴/well) were incubated with autologous T cells (3 × 10⁴/well) in the presence of phytohemagglutinin A (PHA, 1 µg/ml; Difco, Hamburg, Germany) in 96-well microtitre plates. RPMI 1640 medium was supplemented with 10% FCS, gentamicin (4 µg/ml), recombinant human IL-4 (Pharma Biotechnologie Hannover, Germany: 500 U/ml), and GM-CSF (Leucomax 400; Sandoz, Nürnberg, Germany: 800 U/ml) as described previously [24]. Every third day, fresh cytokines were added. Cytokines are necessary to keep DC alive and to induce maturation as would occur during migration from site of infection to lymph node. Proliferative responses were assessed at the different time points by the addition of 1 µCi ³H-thymidine (2 Ci/mmol; NEN Life Science Products, Bruxelles, Belgium) to each well for the last 24 h.

Mixed leucocyte reaction (MLR)

Allogenic T cells (2 × 10⁵/well) were incubated with infected or uninfected DC (1 × 10⁴/well) for 5 days. During the final 24 h, each well was pulsed with 1 µCi ³H-thymidine (2 Ci/mmol).

Statistical analysis

Triplicate values were compared using Student's t-test. P < 0·05 was considered as statistically significant in both proliferation assays.

RESULTS

Infection of DC with Y. entercolitica

For detection of DC-associated bacteria, Y. enterocolitica O:3 pYV was transformed with plasmid pKSBCgfp-sod which carries the gfp mut2 gene controlled by the yersinia-independent listerial sod promoter. The obtained strain was subsequently used in infection studies with DC isolated from human blood. As shown in Fig. 1, 2 h post infection, a clear association of fluorescent bacteria with DC could be demonstrated. To quantify these associations, FACS analysis measuring the number of DC with GFP (originated by the bacteria) were performed. At a multiplicity of infection (MOI) of 20, approximately 65% of DC were associated with bacteria, as judged by FACS analysis (Fig. 2b) and, at a MOI of 50, nearly all cells were associated with yersiniae (Fig. 2c). No difference in the percentage of associated bacteriae could be observed when DC were infected at different stages of maturation induced by the cytokines GM-CSF and IL-4 (Fig. 2d). To distinguish between cell-surface bound bacteria already killed by gentamicin and viable intracellular bacteria, DC were lysed and colony-forming units of bacteria were determined at different time points; high numbers of viable intracellular bacteria were found in the beginning of the infection with a rapid decline thereafter. At day 3 after infection, only small numbers of colony-forming units could be detected and thereafter colonies yielded no live bacteria (Table 1).

Fig. 1.

Association of green fluorescent protein (GFP)-labelled Y. enterocolitica with peripheral blood-derived DC 2 h post infection. Phase contrast pictures of uninfected DC (a) and DC infected with Y. enterocolitica (b). (c) Fluorescence image of the infected DC, showing fluorescent bacteriae only. (d) Phase contrast and fluorescence overlay image of the same infected DC as in (b).

Fig. 2.

Flow cytometry analysis of DC associated with green fluorescent protein (GFP)-labelled Y. enterocolitica. (a) Uninfected cells; (b–d) 4 h after infection. Infection was performed with different multiplicities of infection [(b), 20; (c,d) 50]. The y-axis shows intensity of GFP fluorescence; the x-axis displays the cells incorporating propidium iodide (dead cells). Maturation of DC can be induced by addition of cytokines as described in the method section. DC at different stages of maturation were infected with Y. enterocolitica to exclude a maturation-dependent mechanism of infection; (d) day 4 of maturation. Infected cells are found above the horizontal line of all four panels, dead cells are to the right of the vertical lines.

Table 1.

Number and percentage of viable yersiniae found in DC infected with Y. enterocolitica for different periods of time

Days after infection	Colony forming units	Bacteriae/100 cells
0	1·9 × 104	550
1	1·6 × 103	124
3	8·5 × 101	27

3 × 10³ DC were infected by 6 × 10⁴ yersiniae (MOI 20), treated with gentamicin to kill extracellular bacteria, and lysed at different time points after infection (4 h, 1 day, 3 days) for detection of intracellular viable bacteria. Colony-forming units were counted 2 days after plating. Mean values of one typical experiment performed in duplicate are shown, similar results were obtained in two independent experiments.

Viability of DC

To exclude adverse effects of the infection with Y. enterocolitica on DC, we investigated whether Y. enterocolitica leads to necrosis or apoptosis in infected DC. As shown in Fig. 3(a), no difference in the proportion of necrotic cells could be detected in comparison to uninfected DC and the percentage of necrotic cells in both preparations was <10% at day 1 after infection. Similar results were obtained in the beginning of the infection at 6 h and 18 h as well as 3 days, 5 days and 7 days after Yersinia-infection. Parallel staining with fluorescein diacetate confirmed the viability of infected cells. In addition, apoptosis could not be observed in infected or uninfected cells (Fig. 3b) as demonstrated by staining of DNA strand breaks.

Fig. 3.

(a) The proportion of necrotic DC was assessed by addition of propidium iodide, viability by fluoresceindiacetate at day 1 (A,B) and day 5 (C,D) for uninfected DC (A,C) or Y. enterocolitica-infected DC (B,D). Propidium iodide is represented by the y-axis, fluoresceindiacetate by the x-axis. Dead cells are found at the top left edge of each panel. Note that, at day 5, a second DC population emerges from the viable cell fraction, probably due to maturation of CD34⁺ cells following treatment with GM-CSF and IL-4. (b) Apoptosis-assay for Y. enterocolitica-infected DC to exclude the induction of programmed cell death due to infection using detection of DNA fragmentation by terminal deoxytransferase-mediated dUTP nick end labelling (TUNEL reaction). 6 h after infection, DC were fixed, permeabilized, and stained for DNA strand-breaks. (A) uninfected DC (B) Y. enterocolitica-infected DC and (C) DC treated with DNase I for control.

Changes in surface marker expression of DC in response to infection with Y. entercolitica

To investigate maturation and stimulatory capacity of DC, we assessed the surface expression of MHC class I and II, CD83 and costimulatory molecules (Fig. 4). DC undergo a maturation process in response to the supplemented cytokines, so that a marker drift can be observed even in the uninfected control. Comparison of the marker profiles of uninfected versus infected DC shows that infection had no influence on the expression of ICAM-1 (data not shown) or MHC class I-molecules over the entire time course of 5 days, MHC class II-molecule expression was downregulated compared to uninfected cells beginning 3 days after infection (Fig. 4). At day 5, the distribution of infected cells expressing class II-molecules was extremely broad, with infected DC that appear to be negative for MHC class II-molecules, whereas other cells labelled more significant for MHC class II. A weak downregulation was observed for the expression of adhesion/costimulatory molecule CD80 5 days after infection. In contrast, CD83, which is exclusively expressed on antigen-presenting-cells, was upregulated within hours after infection indicating maturation of infected DC. This upregulation of CD83 was consistent over the entire time course. Beginning at day 3 after infection, the costimulatory molecule CD86 also showed upregulation.

Fig. 4.

Characteristic pairs of FACS profiles of either uninfected (upper histograms) or Y. enterocolitica-infected DC (lower histograms). Cells were analysed for expression of HLA class I, HLA class II, CD80, CD83 and CD86. Surface markers expressed at day 1 after infection were compared with day 3 and day 5 (black histograms). Isotype-matched control antibodies represent background staining (white histograms). The y-axis of the histograms shows the relative cell number; the x-axis the log fluorescence intensity. Results are representative of three experiments performed with similar results.

Proliferative responses of T cells cultured with autologous DC

To assess the stimulatory capacity of DC after infection with Y. enterocolitica, we analysed mitogenic responses of autologous T-lymphocytes to phytohemagglutinin A. Proliferation of T cells was significantly reduced due to the infection of DC at days 4–6 of coculture. This effect was not observed with equal numbers of T cells and uninfected DC or DC infected by nonpathogenic E. coli (Fig. 5). T cell proliferation induced by the infected DC commenced to increase at day 6 after infection and was equal for uninfected versus infected DC at day 7 after infection, which is consistent with a delay of 2–3 days in initiation of T cell proliferation (Fig. 5). ³H-thymidine incorporation of T cells and standard deviations are shown in Table 2.

Fig. 5.

Proliferative response of T cells cultured with autologous uninfected DC (•), Y. enterocolitica-infected DC (▵), and E. coli-infected DC (◊) in the presence of PHA at the indicated time points. Proliferation was measured by addition of ³H-TdR for the final 24 h of each experiment. Differences between Yersinia-infected and uninfected cells were statistically significant at days 4–6 after infection. Standard deviations of E. coli-infected DC were omitted for clarity of presentation.

Table 2.

³H-thymidine incorporation (counts per minute; c.p.m.) of T cells cultured with autologous uninfected DC, Y. enterocolitica-infected DC and E. coli-infected DC in the presence of PHA over a time course of 8 days

Days after infection	Uninfected DC (c.p.m.)	Y. enterocolitica-infected (c.p.m.)	E. coli-infected (c.p.m.)
3	80376 ± 18600	57123 ± 6482	112374 ± 22319
4	103166 ± 21621	62418 ± 7615	126201 ± 22959
5	174201 ± 30103	64767 ± 6956	152778 ± 33218
6	213613 ± 49617	132290 ± 8020	250627 ± 52998
7	185757 ± 28393	156620 ± 31990	180693 ± 69049
8	171602 ± 39586	222524 ± 13691	159826 ± 48368

Mixed leucocyte reaction studies

We also looked for proliferative responses of allogeneic T cells stimulated by DC either uninfected or Yersinia-infected. Measurement of ³H-thymidine incorporation 5 days after infection showed significant suppression of stimulation by the infected DC in comparison to uninfected cells (Table 3).

Table 3.

Mixed leucocyte reaction of T cells cultured with allogenic uninfected or Y. enterocolitica-infected DC at day 5 after infection

T cells + uninfected DC	220003 ± 42941 c.p.m.
T cells + Yersinia-infected DC	90488 ± 47652 c.p.m.

Proliferative response is measured by ³H-thymidine incorporation (counts per minute; c.p.m.).

DISCUSSION

T cells are thought to be protective after Yersinia infection [25]. Therefore antigen-presenting cells (APC) including highly potent DC able to stimulate resting T cells might play a crucial role in the defence against infection with Y. enterocolitica. We aimed to elucidate the early phase of the infection when specific immune responses are initiated by DC, which are efficient stimulators of T-lymphocytes and indispensable for the primary immune response.

Recently, it was shown that DC are able to process viable Gram-negative bacteria for MHC-I and MHC-II presentation to T cells [26]. We have shown here that infection of DC with the Gram-negative human-pathogenic bacteria, Y. enterocolitica, leads to a diminished immunostimulatory capacity of DC, which are invaded by viable bacteria in the beginning of the infection. Depending on the multiplicity of infection (MOI) nearly all DC were associated with GFP-labelled bacteria, an effect which was observed for immature, blood-derived DC, as well as for cytokine-treated, mature DC. In most tissues, DC are present in immature state, with low capacity to stimulate T cells but many devices for capturing antigens. Antigens are able to induce further maturation of DC, and this differentiation is crucial for induction of immunity. Immature DC can be isolated from blood, gastrointestinal tract, heart, kidney or skin [13], particularly at the main sites of primary antigen contact. Thus, the population of immature DC is the differentiation state that enterobacteriaceae such as Y. enterocolitica meet in vivo. Upon activation by antigens, DC migrate to the lymphoid tissues such as the spleen and lymph nodes where DC complete their maturation and present antigens to T cells. However, interactions of DC with bacteria and their influence on DC numbers, localization and function are not well documented.

In contrast to macrophages, which are not invaded by Yersinia but become apoptotic by means of secreted Yersinia-derived proteins [11], lysis and subsequent plating of DC yielded viable bacteria in the beginning of the infection, underlining the intracellular localization of viable yersiniae. Listeria monocytogenes, a Gram-positive bacterium, induces apoptosis after invasion of murine DC [27], a mechanism that could be excluded for the interaction of human DC with Yersinia, as well as necrosis. Yersinia might enter DC by phagocytosis, a feature of less differentiated or immature DC [28,29] or by a phagocytosis-independent mechanism, as was described for Borrelia burgdorferi [30], and Chlamydia is killed by DC and internalized by macropinocytosis, a process accompanied by maturation of the DC [31].

We observed changes in surface marker expression after infection, i.e. MHC class II-molecules were downregulated, which was a transient effect and which correlated with the observed immunosuppressive effect at days 4–6 after infection. Day 5 showed DC that appeared to be negative for MHC class II as well as DC which seem to label more significantly for MHC class II as uninfected DC. Upregulation of the initially downregulated MHC II molecules might represent a mechanism to overcome the adverse effects of Yersinia on immunostimulation. The lymphocyte costimulatory molecules CD80, as well as CD86, are differently regulated to a lesser extent; CD80 was downregulated at day 5 after infection. Activation of human peripheral blood DC is associated with induction of the CD86 costimulatory molecule [32], and we found here that CD86 is upregulated at day 3 after Yersinia-infection. Similar findings were reported after infection of DC with measles virus, exhibiting upregulation of MHC class II, CD83 and CD86 [33]. Infection of DC with live Mycobacterium tuberculosis results in increased APC surface expression of CD54, CD40, CD80 and MHC class I molecules [34] and Bacillus Calmette-Guerin treatment of human DC leads to an enhanced stimulatory potential and upregulation of CD83 and CD86 [35]. The upregulation of CD83 immediately after infection, indicative of a maturation process, was observed in these experiments in case of Yersinia-infection as well. In addition the measles virus infection of DC coincided with immunosuppression [36]. Similar observations with reduced DC function after virus-infection were made for HIV [37] and hepatitis C-virus [38].

The immunosuppressive effect we found for Yersinia was transient in character and, at day 7 after infection, no difference in stimulation of autologous T cells by infected versus uninfected DC could be detected. Stimulation of allogenic T cells in a mixed leucocyte reaction was affected as well. Interestingly, staphylococcal enterotoxin B induces a similar early and transient state of immunodeficiency by binding to class II-molecules on APC, representing a potential mechanism for escaping host immune surveillance [39]. The suppressive effect on T cell stimulation was specific for Yersinia in that another Gram-negative bacteria (E. coli) was unable to mimick the effect under these conditions.

In conclusion, the results show a transient decrease in the immunostimulatory capacity of DC after infection with Yersinia, which can be correlated with an initially marked downregulation of MHC class II-molecules on the surface of DC. The delayed initiation of specific immune responses might facilitate the persistence of the microorganism, thereby initiating new cycles of bacterial proliferation leading to an increased bacterial burden and following increased and ongoing inflammatory process in the intestines or even in joints in predisposed individuals. On the other hand, in most patients, the self-limitation of intestinal Yersinia-infection might be explained by the transient nature of this immunosuppression.