WEN LI*, DOMINIQUE BUZONI-GATEL*†, HAJER DEBBABI*, MARK S. HU*, FRANCK J. D. MENNECHET*, BRIGIT G. DURELL*, RANDOLPH J. NOELLE*, LLOYD H. KASPER*

* Departments of Medicine and Microbiology/Immunology, Dartmouth Medical School, Lebanon, N. H. 03756, USA.

† Institut Pasteur, Unite de Biologie Moleculaire du Gene, 25-28 rue du Dr Roux, 75724 Paris and Pathologie infectieuse et immunoligie, Institut National de la Recherche Agronomique, 37380 Nouzilly, France

Short title: CD40/CD154 and T gondii infection

Key words: Intestine, mucosal immunity, ileitis, CD40/CD154, T gondii

Cell to cell signaling between T lymphocytes and antigen-presenting cells (APC) strictly regulates the development of the immune response. T cell co-stimulatory molecules are critical for the induction of this process. Among these signals, CD40 and its ligand CD154 are required for T cell co-stimulation, lymphocyte activation, isotype switching, T cell maturation, activation of accessory cells to produce IL-12 and activation of the macrophages (or APC) effector function. Moreover CD40/CD154 interactions are known to be a critical element in the development of both experimental and human inflammatory bowel disease (IBD).¹

It has become increasingly apparent that cytokines play an important immunologic role in the development of human IBD. ^2-5 Among the various immune modulators, IFN-g, TNF-a, and RANTES contribute to the selective accumulation of macrophages and memory T lymphocytes inside the granulomas and inflammatory infiltrates that are characteristic of IBD. In IBD the pathogenic cytokine production appears consistent with the increased expression and upregulation of various co-stimulatory molecules, in particular CD40/CD154 interactions. For example, intestinal T cells from IBD patients show increased and prolonged expression of CD154. Increased expression of CD40 in both the B cell and macrophage populations has been described in these individuals.¹ In an experimental model of colitis, blocking the CD40-CD154 interaction can prevent the development of colitis induced by adoptive transfer of CD4 CD45RB high cells. ^{6, 7} These observations suggest that CD154 up-regulation is involved in pathogenic cytokine production in IBD.

The potential importance of CD40/CD154 interactions in both human and experimental toxoplasmosis has been reported. Toxoplasmosis in humans and other mammals is acquired by oral ingestion of either tissue cysts containing bradyzoites from infected meat products or oocyst/sporozoites from contaminated soil. Once ingested, the cyst wall is digested within the lumen of the small intestine and the parasites infect the mucosal cells of the intestine from which the parasite is disseminated to other organs throughout the host, in particular muscle and the central nervous system. Initial exposure of the host to the orally acquired parasite stimulates a strong innate immune response characterized by a marked increase in the production of IL-12 and IFN-g. Human macrophages infected with T. gondii were shown to upregulate expression of CD40. Production of IFN-g and IL-12 by human PBMC (peripheral blood mononuclear cells) exposed to the parasite is dependent upon the CD40/CD154 interaction. ⁸

In this paper, the role of CD40/154 interaction in the immune pathogenesis of this infection-driven model of acute ileitis is explored. Our findings suggest that the interaction of these co-stimulatory molecules is essential for the development of acute, lethal inflammatory ileitis.

C57BL/6 (B6) mice were obtained from Jackson Laboratories (Bar Harbor, ME). CD40^-/- (CD40 KO) and CD154^-/- (CD154 KO) mice on a B6 background were originally obtained from Jackson laboratories and were bred under pathogen free conditions in the Animal Research Facility at Dartmouth Medical School. FACS analysis of spleen cells isolated from the geneticlly altered mice demonstrate a consistent number of B cells, macrophages and dendritic cells compared with wild type B6 mice. Congenic C57BL/6 Ly5.2 (B6 Ly5.2) were obtained from the National Cancer Institute (NCI, Frederick, MD). In every experiment six mice were used in each group.

Bone marrow (BM) was isolated from donor CD40^-/- and B6 mice by flushing the femur and tibia bones. 20 • 106 BM cells were intravenously injected into 800 RADS irradiated recipient mice. B6 Ly5.2 and CD40^-/- mice were each reconstituted with either BM from CD40^-/- or B6 Ly5.1 mice (four sets).

Six weeks after reconstitution the blood from the B6 Ly5.2 chimeric mice was taken to check the quality of the reconstitution. BM donors were marked with CD45.2 (Ly5.1) whereas the recipient mice had the congenic CD45.1 (Ly5.2) marker. Blood cells were stained with both anti CD45.1 and anti CD45.2 antibodies (PharMingen, San Diego, CA) to determine chimerism. FACScan was used to find that the chimerism in the mice was ≥ 85% (Ly5.1+). There was no significant difference between the number of B cells, macrophages and dendritic cells in chimera mice than B6 mice.

B6 mice received a daily intraperitoneal injection of 150mg of anti CD154 (MR1) antibody ¹³ over a period of 7 days beginning the day before challenge. Control mice were administered hamster isotype control antibody. Mice were able to tolerate the infusion without clinically significant side effects.

The strain 76K of T. gondii was used in all studies. This Type II parasite strain has the capacity to produce tissue cysts containing bradyzoites in the brains of infected mice. For most challenge experiments, mice were infected by intragastric gavage with tissue cysts isolated from the brains of infected mice. Cysts were maintained by passage every 2 months in naïve mice. For this study, brain tissue containing cysts were suspended in saline buffer, and the mice were orally challenged with 30 cysts which is lethal to B6 mice.

At day 7 after parasite challenge, mice were sacrificed by CO₂ inhalation. The intestines were preserved in 10% buffered formalin. Fixed tissues were embedded in paraffin, serially sectioned at 5 µm and stained with hematoxylin and eosin.

Vibratome sections of ileum, Peyer’s patches and mesenteric lymph nodes were examined simultaneously by double color staining for detection of both phenotype and CD40/CD154 expression.¹⁴ At days 3, 5 and 7 post infection tissues were removed and placed immediately in sterile ice cold PBS. These tissues were trimmed and thin sections (30-70 mm) were cut using a vibratome (V1000, Energy Beam Sciences). FITC conjugated (1%) antibodies in PBS/0.1%BSA/0.1%azide (PBA) containing 5% normal mouse serum (to block nonspecific binding) were added to sections in 96-well plates and incubated overnight at 4ƒC in the dark with continuous gentle agitation. A panel of mAbs was used for direct and immunofluorescent staining: FITC conjugated antibodies to CD19 (1D3), Mac-3 (M3/84), CD11c (HL3), CD4 (L3T4 ) (GK 1.5) and CD8a (Ly-2) (53-6.7), CD40 staining was done with either a direct anti CD40-PE or indirect with unlabeled CD40 (PharMingen, San Diego, CA) and goat anti Rat-Cy3 (Jackson ImmunoResearch Laboratories, West Grove, PA) as the secondary antibody. CD154 was stained with either anti CD154-PE or anti CD154-biotin and streptavidin-FITC (PharMingen, San Diego, CA).

Either rat or hamster isotype controls were used and kept constant throughout the experiment. Unbound antibodies were removed from the sections by aspiration followed by four 20-min washes in PBA. Washed sections were then fixed over night in the same buffer containing 1% paraformaldehyde. Stained sections were wet-mounted in anti-fade (Molecular Probes, Eugene, OR), sealed with nail polish, and stored at 4ƒC in the dark before confocal examination. Confocal thresholds were set using FITC, PE or Cy3.

Intracellular staining for IL-12p40/p70 production by cells within the vibratome sections of intestine and mesenteic lymph nodes was investigated following treatment of brefeldin A. This compound will allow for the accumulation of newly synthesized proteins within the cells. The staining method is an adaptation of an indirect staining method for flow cytometry. ¹⁴ In our studies, vibratome sections of viable tissues were incubated with brefeldin A (100 mg/ml, Sigma, St Louis, MO) at 37 ƒC for 4 hours, and then for 2 hours at room temperature in 200 ml of PBA containing 0.5% saponin (Sigma, St Louis, MO) in the presence of 1% anti IL-12 p40/p70 and 5% normal mouse serum to block the nonspecific binding. Following three 20 minutes washes in PBA contaning 0.5% saponin, sections were fixed as described before.¹⁴

Immunofluorescent labeled sections were optically sectioned using a Bio-Rad MR 1024 Confocal Scanning laser Microscope System which has Zeiss Axioskop using a 40X plan neofluar 1.3NA. (Bio-Rad Laboratories, Hercules, CA) equipped with a krypton/argon laser. Laser power, PMT gain and enhancement factors were then determined for the FITC and either PE or Cy3 channels using the single fluorochrome-stained sections to ensure effective cross-channel compensation. Two-color fluorescent sections were then evaluated for the presence of the FITC labeled phenotypic and either Cy-3 or PE labeled phenotypic inflammatory cells expressing either CD40 or CD154. Double staining images are composed from two gray-scale images, for each PMT channel, with each range from 0 to 255 gray levels.

Intestinal LPMC were isolated as described. ¹⁵ Briefly following dissection and removal of Peyer’s patches, the sectioned (0.3cm) intestines were incubated in RPMI containing 25 mM EDTA, 100 UI/ml penicillin-streptomycin, 1% fungizone, 50 UI/ml gentamicin (Sigma, St Louis, MO) under a magnetic stirring (300 rpm) at room temperature (six times, 20 min). The intestine pieces were then incubated at 37°C into RPMI-10%FCS with 125 UI/ml collagenase VIII (Sigma). After 3 hours, the digested suspension containing LPMC was filtered on cell strainer and the pellet resuspended in RPMI-10%FCS. Mononuclear cells were obtained by centrifugation on a Ficoll layer (d=1.077).

All procedures were performed under sterile conditions. Peyer’s patches were carefully excised from the wall of the small intestine, and minced in culture medium. Mesenteric lymph nodes were excised carefully. The cell suspensions were passed through a cell strainer (Fisher Scitisic, Chicago, IL), then a sterilized gauze band to remove cell debris and washed twice with cold RPMI 1640 medium. Mononuclear cells were obtained by centrifugation on a Ficoll layer (d=1.077).

IEL were isolated as previously described. ¹⁶ Briefly, the small intestine was flushed with PBS, and divided longitudinally after the removal of the Peyers’s patches. The mucosa were scraped, dissociated by mechanical disruption in RPMI 1640 containing 4% fetal calf serum (FCS) and 1 mM dithioerythritol. After passage over a glass-wool column the lymphocytes were separated by Ficoll layer (d=1.077).

Phenotypic analysis

Phenotypic pattern of cells from LPMC, MLN, PP and IEL were determined by FACScan analysis using 1% FITC conjugated antibodies to CD19 (1D3), Mac-3 (M3/84), CD11c (HL3), CD4 (L3T4 ) (GK 1.5), CD8a (Ly-2) (53-6.7) (PharMingen, San Diego, CA). Dual staining was utilized to identify co-expression of either CD40 or CD154 (PE conjugated antibodies) (PharMingen, San Diego, CA).

Cell suspension containing 1•106 cells were added to 96-well plates and washed twice in PBS. Cells were incubated with 100 ml of 1% Fc-block (CD32/CD16), for 20 min at 4ƒC to prevent nonspecific Ab staining. After washing, the cells were incubated for 1 hour on ice in the presence of 1% anti CD40 or CD154 mAb (PE conjugated) together with one of FITC conjugated phenotyping antibodies (1%). After washing, the cells were fixed with 1% paraformaldehyde in PBS and analyzed by FACScan (Becton Dickinson, Mountain View, CA) the following day.

Quantification of the parasite burden was carried out by PCR. Intestine and spleen were recovered from infected mice at day 7 post infection. Genomic DNA was extracted from tissue using the Qiamp tissue kit (Qiagen, Valencia, CA), and 400 ng of each samples were analyzed. Amplification of parasite DNA was performed using primers specific for the Toxoplasma B1 gene (5' -GGAACTGCATCCGTTCATGAG-3' AND 5' -TCTTTAAAGCTTCGTGGTCT-3'), a 35-fold sequence found in all known parasite strains. ^17-18 A134 bp competitive standard containing the primer template sequences as the 194 bp B1 PCR fragment was generated and used as an internal standard for competitive PCR. Amplification of the 194 base pairs segment of the B1 gene and the 134 base pairs segment of the internal standard was performed in a 50 ml reaction mix containing 1.25 units of Amplitaq DNA polymerase, 1• buffer (Perkin Elmer), 0.2mM each of dGTP, dATP, dTTP, dCTP and 0.4mM (each) B1 primers. To quantify the number of parasites from the tissue, T gondii DNA was isolated from a known number of cell culture-derived extracellular tachyzoites. The toxoplasma B1 sequence was used as a probe to quantify tissue parasite load and run in parallel with the infected tissues, a known number of parasites were used as a standard to quantitate the number of parasites/microgram of tissue DNA. ¹⁸

The tissues from intestine were frozen in liquid nitrogen and stored at –80ƒC until RNA preparation. The samples were homogenized in 2.0 ml Trizol reagent (Life Technology, Grand Island, NY) and incubated for 5 minutes at room temperature to permit the complete dissociation of nucleoprotein complexes. After adding 0.2 ml of chloroform per 1 ml of Trizol reagent and incubating them at room temperature for 3 minutes, the samples were centrifuged at 12,000 x G for 15 minutes at 4o C. The mixture separates into a lower, red, phenol-chloroform phase, an interphase and a colorless upper aqueous phase. RNA remains exclusively in the aqueous phase. RNA precipitation was done with isopropyl alcohol, and washed with 75% ethanol. The isolated mRNA was resuspended in RNase-free water. The concentration was determined by UV spectroscope at 260 nm.

Chemokine and cytokine mRNA expression in the intestines were detected as described, ^{11, 19} 10 mg of mRNA were used and radioactively labeled for RPA analysis. Chemokine and cytokine mRNA expressions in the cells were detected with RiboQuant multi-Probe RNase Protection Assay (RPA) System Kit (PharMingen, San Diego, CA) as described by the manufacturer. Briefly, the probe either for chemokine detection: macrophage inflammatory protein-1b (MIP-1b) and MIP-2, lymphotactin, monocyte chemo-attractant protein (MCP-1/JE), MCP-3, IFN-g inducible protein (IP10/CRG-2), regulated on activation normal T cell expressed and secreted (RANTES), or for cytokine detection interleukin 12p35 (IL-12p35), IL-12p40, IL-1b, IL-18, IL-6, IFN-g were labeled with ³²P. Then, the hybridization procedure between RNA samples and the labeled probe was performed. RNA samples were digested by RNase treatment 45 min at 30°C. The RNase digests were extracted carefully. Samples were loaded in 5% acrylamide gel. For quantification, autoradiographs were scanned and band densities were be analyzed by NIH Image 1.61/ppc software in Macintosh computer. The housekeeping gene probes GAPDH, allows for normalizing sampling and technique errors to permit comparison of individual mRNA species between samples. Results were expressed as the % of the intensity of the band corresponding to the studied chemokines relative to the intensity of the housekeeping RNA.

Mice sera were collected at day 3, 5 and 7 after infection, IL-12 p70 production was evaluated by ELISA following the manufacture's instructions (Biosouce, Camarillo, CA).

Data were expressed as means and standard deviations and plotted in corresponding figures. The differences between experimental groups and comparison groups were examined with Student t-test.

At day 7 post oral infection, B6 mice exhibited acute ileitis characterized by fulminent necrosis. Cross-section of the small intestine demonstrated reduced villus length, massive hemorrhage and leukocytic infiltrates in the lamina propria and epithelium (Fig. 1B) compared to an uninfected mouse (Fig. 1A). There was no difference between infected mice treated with the hamster isotype control and the B6 untreated mice (data not shown). In contrast, the histologic analysis of CD154^-/- mice (Fig. 1C) or anti CD154 mAb treated mice (Fig. 1D) revealed nominal mononuclear infiltration in the lamina propria. These observations were consistent in two independent experiments. These data suggest that CD40/CD154 interactions are important components in the development of murine ileitis following oral infection with T. gondii.

To further explore the consequence of the CD40/CD154 interaction in the development of the inflammatory ileitis observed in B6 mice, chimeric mice including B6 Æ B6 Ly5.2,

CD40^-/- Æ B6 Ly5.2, CD40^-/- Æ CD40^-/- and B6 Æ CD40^-/- groups were generated. The chimeric mice had >85% reconstitution of the lymphoid compartment in the periphery and intestine six weeks after reconstitution when they were infected per os. At day 7 post infection, the intestines were examined for histologic evidence of inflammation. The intestines of the CD40^-/- Æ B6 Ly5.2 chimeric mice exhibited little evidence of inflammation and were without necrosis or hemorrhage (Fig. 1E). The epithelium was preserved and only mild mononuclear infiltrates were observed in the lamina propria. In contrast B6Æ B6 Ly5.2 (Fig. 1F) mice developed intestinal necrosis that was identical to that observed in the infected B6 mice. Control mice including CD40 ^-/- Æ CD40^-/- chimera exhibited no evidence of inflammation (Fig. 1G). B6 Æ CD40^-/- chimeric mice resulted in the development of intestinal inflammation (Fig. 1H) consistent with that observed in parental B6 mice. These observations were consistent within two independent experiments. Most of the mice from the group exhibiting major inflammation (B6, B6 Æ B6 Ly5.2, B6 Æ CD40^-/-) died within 10 days. Mice from the other groups (CD154^-/-, anti-CD154 mAbs treated, CD40^-/- Æ B6 Ly5.2, CD40^-/- Æ CD40^-/-) survived at least within 2 weeks after infection. These findings emphasized the critical role of the CD40/CD154 interaction in the development of hyperinflammation and more particularly expression of CD40 from hematopoietic origin. These observations illustrate that it is possible to adoptively transfer the potential to develop inflammatory ileitis in response to antigen exposure using BM derived cells that express the CD40 molecules.

##UPLOAD IMAGE: Figure 1##

Figure 1. Histological analysis. At day 7 after infection, B6 mice exhibited major inflammation (B) compared to naïve control (A). CD154^-/- mice (C) and anti CD154 treated mice (D) lacked evidence of intestinal inflammation at day 7 after oral infection. Chimeric mice were infected and sacrificed at day 7 post infection. CD40^-/- Æ B6 Ly5.2 chimeric mice (E) as well as CD40^-/- Æ CD40^-/- chimeric mice (G) exhibited only mild inflammation in comparison to B6 Æ B6 Ly5.2 chimeric mice (F) and B6 Æ CD40^-/- chimeric mice (H). Magnification is indicated by the bar: 100 m m.

A phenotypic analysis was carried out to evaluate whether the induction of ileitis correlated with the up-regulation of CD40 on various components of the GALT. A kinetic study was first performed using cells isolated from the lamina propria, Peyer’s patches and the mesenteric lymph nodes. Cells were isolated from 6 mice in each group and pooled. Increased expression of CD40 was observed in all compartments following infection with the parasite. Maximal expression of CD40 was observed at day 3 post infection among the cells from the Peyer’s patches. For cells isolated from the mesenteric lymph nodes and the lamina propria, maximal expression was observed at day 7 compared to day 3 post infection, suggesting a sequential nature to this response. In all three intestinal compartments evaluated, the increase in CD40 expression was found to be primarily within the B cell compartment. FACScan analysis indicated that a high frequency of the positive cells express both CD19 and CD40. Macrophages (Mac-3) and dendritic cells (CD11c) that expressed CD40, increased with time post infection within the lamina propria (Fig. 2).

Figure 2. Phenotypical analysis. A phenotypical analysis was performed on cells from three different intestinal compartments: lamina propria mononuclear cells (LPMC), Peyer's patches (PP) and mesenteric lymph nodes (MLN). This phenotypic analysis studied the association between different markers: CD40 and CD19 (B cells) or Mac-3 (Macrophages) or CD11c (dendritic cells) into B6 naive mice (1), B6 mice at day 3 (2) and at day 7 (3) after infection. This study examined also the expression of these markers at day 7 post infection in chimeric mice: B6 Æ B6 Ly5.2 chimeric mice (4) and CD40^-/- Æ B6 chimeric mice (5). Results are expressed as the % of CD40 positive (1 ) and among those positive for CD40 is expressed the % of double positive cells for three different phynotypic markers (› ).

To confirm the association between intestinal inflammation and expression of the CD40 molecule, similar experiments were undertaken using chimera mice at day 7 after infection. In contrast to the B6 Æ B6 Ly5.2 chimeric mice, GALT cells isolated from mice reconstituted with BM from the CD40^-/- mice displayed negligible quantity of CD40 in the lamina propria, the Peyer’s patches and the mesenteric lymph nodes. In the lamina propria, the Peyer’s patches and the mesenteric lymph nodes from B6 Æ B6 Ly5.2 chimeric mice, CD40 expression is mostly associated with the B cells (CD19). A small proportion of the CD40 expression is associated with the macrophages (Mac-3) and dendritic cells (CD11c) (Fig. 2). In CD40^-/- Æ CD40^-/- chimeras expression of CD40 remained undetectable (data not shown), and in the B6 Æ CD40^-/- chimeric mice expression of CD40 was consistent although decreased to that observed in the B6 Æ B6 Ly5.2 chimeric group (data not shown). As anticipated, phenotypic analysis of these various components of the GALT in CD154^-/- and anti CD154 treated mice failed to demonstrate any significant difference between these mice and the control B6 mice. Several unsuccessful attempts were made to evaluate the expression of CD154 by FACScan analysis in the population of either CD4+ or CD8+ cells isolated from the three different immune compartments. All studies were unremarkable except for very weak expression in the IEL CD8 population (data not shown) from the non-chimeric B6 mice at day 7 after infection.

In situ expression of the CD40 was evaluated by confocal analysis using gut derived tissue sections obtained from parasite infected mice. An increase in CD40 signal intensity was detected in the intestine, Peyer’s patches and the mesenteric lymph nodes at day 3, 5 and 7 post infection confirming the FACScan results (data not shown). Evidence of CD154 expression in associated tissue including the intestine, the Peyer's patches and the mesenteric lymph nodes was noted at day 3 and day 7 post infection (Fig. 3A, B, C). This expression was principally associated with CD4 (Fig. 3D, E) T cells and to a lesser extent with CD8a (Fig. 3F).

Figure 3. CD154 expression. At day 3 post infection, there was evidence of CD154 expression in the intestine (Green, FITC) (A), the Peyer's patches (Red, PE) (B) and the mesenteric lymph nodes (Red, PE) (C). At day 7 after infection, CD154 (Red, PE) was co-expressed within CD4+ (Green, FITC) T cells in Peyer’s patches (D) and mesenteric lymph nodes (E) or within CD8a + (Green, FITC) T cells in Peyer’s patches (F).

To evaluate whether CD40/CD154 interaction are involved in the regulation of chemokineproduction, the chemokine pattern produced within infected mice was measured at day 7 post infection (Fig. 4A). Compared to naive B6 mice, a significant upregulation of a wide variety of chemokines (P £ 0.001), including MIP-2, MCP-3, MIP-1b, MCP-1, IP-10 and RANTES were observed in B6 mice at day 7 after infection. The increase in chemokine production was abrogated significantly by treatment with either anti-CD154 mAb or by genetic deletion (CD154^-/-, P £ 0.001) (Fig. 4A), indicating that CD40/CD154 interaction is required.

These results were confirmed in the chimeric model (Fig. 4A). In B6 Æ B6 Ly5.2 chimeric mice, chemokine expression in the intestine at day 7 post infection was just below those levels observed in the control infected mice (Fig. 4A). One explanation for this is that irradiation of the recipient altered certain physiologic interactions that cannot be reconstituted with BM derived cells. This would suggest a role for non hematopoietic cells such as enterocytes bearing CD40. When B6 Ly5.2 mice were reconstituted with CD40^-/- BM, such an increase in chemokine expression was not observed (P £ 0.001) (Fig. 4A). The control CD40^-/- Æ CD40^-/- chimeric mice did not display any chemokine production after infection (data not shown). The upregulation of chemokines in B6 Æ CD40^-/- chimeric mice was significant compared to CD40^-/- Æ CD40^-/- or CD40^-/- Æ B6 Ly5.2 chimeric mice, but remained impaired compared to the B6 Æ B6 Ly5.2 chimeric mice (P £ 0.001). Cytokine analysis for IL-1b, IL-6, IL-18 and IFN-g were consistent with an important role for CD40/CD154 interaction in the regulation. (P £ 0.001) (Fig. 4B).

Figure 4. Chemokines and cytokines production. RPA analysis for chemokines (A) and cytokines (B) was performed on the whole ileum from different groups of mice at day 7 after infection.

CD40/CD154 interaction is critical for the activation of accessory cells to produce IL-12. Production of IL-12 p70 was detected in the murine sera. The B6 and B6 Æ B6 Ly5.2 chimeric mice had increased levels of IL-12 when compared to the CD40^-/- Æ B6 chimeric mice at day 3, 5 and 7 after infection (P £ 0.05). Peak production was observed at day 3 after infection (Fig. 5).

Figure 5. Systemic IL-12 production. Level of IL-12 (p70) into the sera from different groups was detected by ELISA at increasing times after infection. Basical level of IL-12 p70 in sera of naïve mice was 0.31pg/ml.

Confocal analysis was also performed to detect intracellular expression of IL-12 (p40/p70) in the intestine at different times post infection. Detection of IL-12 product was maximal at day 3 post infection in the intestine of B6 mice and decreased thereafter (day 5, 7). In the intestine, IL-12 production appeared to be predominantly localized to the lamina propria of the villi (Fig. 6A). Detection remained lower in the mesenteric lymph nodes (Fig. 6B) compared to the intestine. There appeared to be a direct relationship between the level of IL-12 detected and the degree of inflammation. In the intestine in B6 Æ B6 Ly5.2 chimeric mice (Fig. 6C), IL-12 production was high compared to CD40^-/- Æ B6 chimeric group (Fig. 6E). Whatever the chimeric group IL-12 production was low in the mesenteric lymph nodes (Fig. 6D, 6F).

Figure 6. IL-12 production at the gut level. Confocal analysis gives the evidence of the over expression of IL-12 (p40/p70) (Red, PE) into the intestine of B6 mice at day 3 after infection(A) and lower in the mesenteric lymph nodes (B). A higher expression of IL-12 was also detected in B6 Æ B6 Ly5.2 chimeric mice in the intestine (C), and remained low in the mesenteric lymph nodes (D), compared to the lower level in CD40^-/- Æ B6 Ly5.2 chimeric group in the intestine (E) and mesenteric lymph nodes (F). Magnification is indicated by the bar: 50 m m.

IFN-g is likely to be produced by CD4 cells from the lamina propria in response to antigen stimulation and to the production of IL-12, IL-18 as well as different chemokines. Mice in which the IFN-g is blocked by treatment with specific antibodies can be rescued from the inflammatory response in the intestine. Mice from both the anti CD154 treated group as well as the CD154^-/- mice produced significantly less IFN-g in the intestine compared to the mice from the control infected group (P £ 0.001) (Fig. 4B). Similarly, B6 Æ B6 Ly5.2 chimeric mice produced more IFN-g than the CD40^-/- Æ B6 chimeric mice(P £ 0.001). These results suggest that CD40/CD154 interaction is required for production of IFN-g. The CD40^-/- Æ CD40^-/- chimeric mice were unable to produce significant amount of IFN-g (data not shown). In B6 Æ CD40^-/- chimeric mice the IFN-g production is reduced compared to the B6 Æ B6 Ly5.2 chimeric mice. However the IFN-g production in this group (B6 Æ CD40^-/-) was found to be significantly higher than that observed in the CD40^-/- Æ B6 chimeric group (data not shown). The level of production of IFN-g within the intestinal mucosal is consistent with the degree of inflammation and necrosis in this tissue.

To explore whether co-stimulation can effect parasite multiplication, both CD154^-/- as well as mAbs depleted mice were infected and evaluated for parasite burden in their spleen and intestine at day 7 after infection. There was marked increase in parasite burden in the spleens of depleted mice compared to the control untreated wild type mice (P £ 0.001) (Fig. 7). Chimeric B6 Æ B6 Ly5.2 chimeric mice had significantly fewer parasites in their splenic tissue than the CD40^-/- Æ B6 chimeric mice (P £ 0.001). CD40^-/- Æ CD40^-/- chimeric mice exhibited a parasite burden in the spleen comparable to that observed with the CD154^-/- mice (data not shown). A similar analysis of parasite burden was performed using intestinal samples (data not shown). In spite of repeated sampling, there were no observed differences in the level of parasite burden among any of the tissue samples obtained from these mice which may have been due to inconsistencies of infected cells within the sample population. Long term evidence of chronic infection as measured by cyst burden in the brain could not be evaluated because B6, B6 Æ B6 Ly5.2, B6Æ CD40^-/- chimeric mice died from intestinal necrosis within 10 days post infection. These findings indicate that CD40 /CD154 interaction is an important component in the control of the parasite multiplication as impaired mice display a heavier parasite burden than the corresponding control groups.

Figure 7: Parasite Load: Parasite burden was evaluated by a PCR method in the spleen from the different experimental groups at day 7 after infection.

CD40/CD154 interactions have a central role in the induction of both humoral and cellular immunity. The interaction of these co-stimulatory molecules is critical for generation of the IL-12 and IFN-g response against a number of pathogen including T. gondii.^{8, 20} In this report we demonstrate that this interaction plays a key role in the development of acute inflammatory ileitis following oral infection with brain tissue cysts that contain bradyzoites. We observed that CD40 expression was upregulated in the intestine of B6 mice at day 7 after infection. Upregulation of CD40 was primarily associated with B cells, although expression of the CD40 molecule was also noted on both macrophages and dendritic cells as reported by others. ^20-21 Interruption of the CD40/CD154 interaction results in a reversal of the acute ileitis associated with this infection whereas reconstitution of this interaction results in development of intestinal inflammation following parasite exposure.

The induction of a Th1-type immune response is critical for resistance to many intracellular pathogens amongst which is T. gondii. Cell mediated immunity resulting in IFN-g production is central to the control of this infection during both the acute and chronic phase of the infection. Conversely, this cytokine has been incriminated in the development of the overwhelming lethal inflammatory process in the intestine of certain strains of mice. Our results indicate that IFN-g production in the intestine appears to require the expression and interaction of the co-stimulatory molecules CD40/CD154.

As in other inflammatory disease ^22-24 we demonstrate that engagement of CD40 correlates with the expression of several different chemokines and the proinflammatory cytokines IL-1b and IL-6. These chemokines and early cytokines are strong chemoattractant molecules for macrophages (MCP-1), PMN (MIP-2), dendritic cells (MIP-1b) and T lymphocytes (RANTES, IP-10) and are likely to participate in the influx of leukocytes that is the hallmark of the inflammation in the mucosal immune compartment. ²⁵ Chemokine secretion as well as chemokine receptor expression ²⁶ by enterocytes or by other cells present in the intestine may play a role in the initiation and modulation of the immune response. Both experimental and human IBD have been associated with the production of such inflammatory chemokines and cytokines. ^{1, 2}5 We have observed that upregulation of both chemokines and cytokines was associated with pro-inflammatory response following primary infection of enterocytes with T. gondii. ¹¹ Blocking engagement of CD40/CD154 results in a decrease in the production of those pro-inflammatory molecules that would in-turn lead to a decrease in migration of cells into the lamina propria.

The CD40/CD154 interaction has been shown to be important in resistance to several intracellular parasites, including Leishmania, ²⁷ Cryptosporidium parvum, ²⁸ Pneumocystis carinii. ²⁹ The protective effect was attributed primarily to its promotion of IL-12 production or alternatively to antibody production. Stimulation of macrophages and dendritic cells through CD40 leads to the production of IL-12, and consequently drives the immune response toward a Th-1 like pattern. Recently it has been demonstrated ³⁰ that B cells also can differentiate into a Th1 or Th2 pattern that results in the production of specific cytokines following oral infection with T gondii in B6 mice. These CD19+ B cells produce IL-2, IL-12, IFN-g, IL-6 and TNF-a. Extrapolating these observations to the data in this report would suggest that B cells maybe a potential source for cytokine production, in particular IL-12 and perhaps IFN-g that would lead to the necrosis following oral infection. In fact, preliminary data from our laboratory indicates that IL-12 production can be detected in a CD19+ population of cells isolated from the intestine as early as day 3 post infection. The direct or indirect role of B cells in the inflammatory process is also suggested by our preliminary data in B deficient mice (B -/-). Oral infection with T gondii of B -/- mice resulted in a statistically significant delay to death compared to wild type mice (data not shown). Histological analysis performed at different times after the infection revealed that the prolonged time to death is consistent with a parallel delay in the degree of intestinal inflammation compared with the susceptible parental C57BL/6 mice. These preliminary observations indicate that B cells might participate in the early development of the acute inflammatory response.

A number of inflammatory bowel diseases, such as Crohn's disease, ulcerative colitis, ^{1, 7} or colitis induced by enteric bacterial, ³¹ are characterized by local production of a wide array of Th-1 like cytokines that are influenced by the upregulation of CD154. In our study, we demonstrate that CD40/CD154 signaling is involved in the Th-1 pro-inflammatory response in the intestine that occurs following exposure to the parasite. Inhibition of this co-stimulatory interaction prevents the development of pathogen induced ileitis. A role for CD40/CD154 interaction in response to pathogen driven ileitis has not been previously reported. Compared to the wild type mice, CD154^-/- mice develop minimal intestinal necrosis in response to oral infection. Similarly treatment of wild type mice with anti CD154 abrogated development of the intestinal lesions. To further examine whether reconstitution of the CD40/CD154 ligation would again result in development of the inflammatory process, chimeric mice were evaluated. Mice reconstituted with bone marrow cells with the wild type phenotype developed inflammation and necrosis throughout the ileum. CD40 expression in the chimeric B6 Æ B6 Ly5.2 mice was consistent with our observation in the parental strains. In contrast, chimeric mice reconstituted with the CD40^-/- bone marrow cells were devoid of pathological inflammation that was consistent with a very low expression of the CD40 molecule at the intestinal level.

Impaired CD40/CD154 ligation was consistent with the impaired production of IL-12 and IFN-g and subsequently with an absence of inflammation. CD40 triggering of APC can elicit a spectrum of cytokines as previously reported. ⁸ Moreover, patients with HIGM (Hyper IgM) syndrome have defective IFN-g and IL-12 secretion in response to an intracellular organism, ³² that can be restored in vitro by agonistic soluble CD154 trimer. These results further suggest that CD154 modulates IFN-g production primarily through its effect on IL-12 secretion rather than through direct T cell activation. However, it appears that there are others factors that regulate IL-12 production in response to T. gondii. In an other model, Reichmann et al ⁸ provide support that splenocytes isolated from B6 mice infected 5 days previously produced high level of IFN-g and IL-12 when stimulated with T.gondii antigen in vitro, and blocking CD154 did not significantly alter the production of these cytokines. They also showed that CD154^-/- mice infected with T.gondii produced less IL-12 than wild type, but comparable levels of IFN-g. The existence of CD154 independent mechanism of IL-12 production is thus suggested. In IL-10^-/- mice, it has been shown that a concomitant blocking of the interaction between CD28-B7 together with CD40/CD154 resulted in decreased IFN-g and iNOS expression and in reduced hepatic damage. ³³

Our observations indicate that the interaction of CD40/CD154 is an important and perhaps essential process in the initiation of pathogen driven inflammatory ileitis. Blocking of this interaction by either treatment with antibody or genetic deletion results in an increase in parasite burden in the host. This would indicate that CD40/CD154 engagement can elicit an intense inflammatory response that is microbicidal ⁸ but if uncontrolled leads to a devastating form of inflammatory ileitis.