Published online before print May 31, 2007
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Institutos de
* Microbiologia Prof. Paulo de Góes,
Bioquímica Médica, and
Biofísica Carlos Chagas Filho, Universidade Federal do Rio de Janeiro (UFRJ) CCS, Cidade Universitária, Rio de Janeiro, RJ, Brazil; and
Department of Microbiology and Immunology, Vanderbilt University School of Medicine, Nashville, Tennessee, USA
1 Correspondence: Depto. de Imunologia, Instituto de Microbiologia Prof. Paulo de Góes (IMPPG), Universidade Federal do Rio de Janeiro (UFRJ), CCS Bloco I, 2° andar Sala: I2-051, Avenida Carlos Chagas Filho, 373, Cidade Universitária, Ilha do Fundão, CEP: 21941-902, Rio de Janeiro, RJ, Brazil. E-mail: belliom{at}acd.ufrj.br
ABSTRACT
We have demonstrated recently that the glycoinositolphospholipid (GIPL) molecule from the protozoan Trypanosoma cruzi is a TLR4 agonist with proinflammatory effects. Here, we show that GIPL-induced neutrophil recruitment into the peritoneal cavity is mediated by at least two pathways: one, where IL-1ß acts downstream of TNF-
, and a second, which is IL-1ß- and TNFRI-independent. Moreover, NKT cells participate in this proinflammatory cascade, as in GIPL-treated CD1d–/– mice, TNF- and MIP-2 levels are reduced significantly. As a consequence of this inflammatory response, spleen and lymph nodes of GIPL-treated mice have an increase in the percentage of T and B cells expressing the CD69 activation marker. Cell-transfer experiments demonstrate that T and B cell activation by GIPL is an indirect effect, which relies on the expression of TLR4 by other cell types. Moreover, although signaling through TNFRI contributes to the activation of B and 
+ T cells, it is not required for increasing CD69 expression on ß+ T lymphocytes. It is interesting that T cells are also functionally affected by GIPL treatment, as spleen cells from GIPL-injected mice show enhanced production of IL-4 following in vitro stimulation by anti-CD3. Together, these results contribute to the understanding of the inflammatory properties of the GIPL molecule, pointing to its potential role as a parasite-derived modulator of the immune response during T. cruzi infection.
Key Words: neutrophils • TNF- • IL-1ß • T cells • CD69
INTRODUCTION
Free glycoinositolphospholipid (GIPL) is found in large amounts on the surface of trypanosomatid parasites but is not present in higher eukaryotic cells, a characteristic that makes it a potential pathogen-associated molecular pattern (PAMP), capable of being recognized by the pattern-recognition receptors (PRR) of the innate immune response. The GIPL used in the present study was purified from Trypanosoma cruzi, the etiologic agent of Chagas’ disease, which is endemic in Latin America and affects 16–18 million individuals. Its oligosaccharide sequence contains a unit of galactofuranose linked to a mannotetraose main chain, which is attached glycosidically to an inositol-phosphoryldihydroceramide, predominantly an N-lignoceroylsphinganine [1 , 2 ]. Evidence suggests the occurrence of free GIPL within the infective trypomastigote and intracellular amastigote forms of T. cruzi, although in smaller quantities than found in the epimastigotes [3 ]. It is important to note that most of the GPI anchors of mucin-like glycoproteins from the infective metacyclic trypomastigote have the same composition of the GIPL obtained from the epimastigote form used in the present study [4 ].
Thirteen members of the TLR family, which are conserved PRR have been identified in mammals [5 ]. Distinct TLRs recognize different PAMPs, having overlapping but also diverse functions in the innate immune response. In particular, TLR4 responds to LPS of Gram-negative bacteria. In addition to LPS, the best-studied TLR4 agonist, several other PAMPs have been described to signal through TLR4 [6 ]. TLR-induced activation culminates with the production of proinflammatory cytokines and the up-regulation of costimulatory molecules, linking innate to adaptive immune responses. We have shown that T. cruzi GIPL is an agonist of TLR4 and that a mutant mouse strain with a nonfunctional TLR4 (C3H/HeJ) is more susceptible to acute T. cruzi infection [7 ]. These results suggest that GIPL-triggered TLR4 signaling may be responsible for conferring the increased resistance to T. cruzi infection observed in wild-type (wt) mice, although testing this hypothesis directly will need to await the generation of a T. cruzi strain, or clone, lacking GIPL expression. How GIPL recognition by TLR4 would lead to higher resistance to infection remains poorly understood. The aim in this work was to further characterize the in vivo proinflammatory properties of the GIPL molecule and to investigate the consequences of this response on the activation of T and B lymphocytes.
MATERIALS AND METHODS
Isolation and purification of T. cruzi GIPL
Isolation and purification of GIPL have been described previously in detail [1
]. Briefly, T. cruzi epimastigotes (G strain) were grown in brain heart infusion-hemin medium with 5% FCS. After PBS washing, cells were thawed and extracted three times with cold water and centrifuged (7000 g, 10 min), and the pellet was extracted with 45% aqueous phenol at 75°C x 15 min. The aqueous layer from the phenol extraction was dialyzed, freeze-dried, redissolved in water, and applied to a column of Bio-Gel P-100. Excluded material was lyophilized, and GIPL was extracted by chloroform/methanol/water (10:10:3). Purified GIPL was -irradiated for sterilization and resuspended in endotoxin-free water (Gibco-BRL, Grand Island, NY, USA). GIPL appeared on SDS-PAGE as a fast-moving, single molecular species. Virtual absence of contaminating material was confirmed by absence of peptide-derived signals in nuclear magnetic resonance spectroscopy and mass spectrometry analyses. It tested negative for LPS using a Limulus amebocyte lysate test, with a limit of detection of 0.125 EU/ml (E-Toxate, Sigma Chemical Co., St. Louis, MO, USA).
Mice
C57BL/6 and C57BL/10ScCr strains were from the Universidade Federal Fluminense (Brazil). C3H/HeJ and C3H/HePas were from Instituto de Ciências Biomédicas, Universidade de São Paulo (Brazil). CD1d–/– [8
] and TNFR1–/– [9
] mice were described previously and were from Vanderbilt University School of Medicine (Nashville, TN, USA) and from the Universidade Federal de Minas Gerais (Brazil; gift from Dr. Leda Quercia), respectively. Experiments were performed with sex- and aged-matched mice, in accordance with current guidelines for the care of laboratory animals and ethical guidelines for experiments in conscious animals.
Reagents and treatments
Mice were injected with the indicated doses of Salmonella typhimurium LPS (Difco, Detroit, MI, USA), GIPL, or vehicle (PBS) in the posterior footpads or i.p. Cells from peritoneal lavage were obtained by injecting 3 ml cold PBS and centrifugation at 200 g for 10 min. Peritoneal fluid was then stored at –70°C.
In vivo IL-1ß neutralization
Mice were injected i.p. with 20.0 µg/mouse purified polyclonal goat anti-mouse (m)IL-1ß antibody (R&D Systems, Minneapolis, MN, USA), and control mice received 20.0 µg/mouse purified goat IgG (R&D Systems). Immediately afterwards, three mice of each group were treated with 50.0 µg/mouse GIPL or 50.0 µl PBS i.p. as control. Peritoneal fluid and cells were obtained 3 h later, as described above. IL-1ß in the peritoneal fluid was measured by ELISA, and flow cytometry was performed as follows.
Flow cytometry
Flow cytometry was conducted with FACSCalibur (PharMingen-BD Biosciences, San Jose, CA, USA). Cells (106) were resuspended in FACS buffer (PBS, 1% BSA, 0.01% sodium azide). FcRs were blocked with anti-CD16/32 (Clone 2.4G2) and 5% normal mouse serum. Anti-Gr-1-PE (Ly-6G), anti-membrane-activated complex 1 (Mac-1)-FITC (CD11b), anti-IgM-FITC, anti-B220-FITC, anti-CD3-FITC, or anti- TCR-FITC and anti-CD69-PE were from PharMingen-BD Biosciences, and staining was performed at 4°C for 30 min.
Cell-transfer experiments
After loading total spleen cells from C57BL/10ScCr mice with 1 µM CFSE (Molecular Probes, Junction City, OR, USA) in PBS for 15 min at 37°C, 107 CFSE+ cells/mouse were injected i.v. into C57BL/10 recipients. Four hours after cell transfer, recipients mice were injected with GIPL, LPS, or PBS in the posterior footpads. Sixteen to 20 h later, cells of the draining lymph nodes (LN) were collected and stained for cytometry analysis, using anti-CD69-PE and anti-B220-APC or anti-CD8-APC (PharMingen-BD Biosciences), as described above.
In vitro T cell activation
Anti-CD3 mAb (0.5 µg/ml) was immobilized in flat-bottomed 96-well microplates (Costar, Cambridge, MA, USA) by overnight incubation at 4°C. After washing, 1.0 x 106 spleen cells obtained from control or GIPL-injected mice were added in RPMI 1640 (Sigma Chemical Co.), supplemented with 10% FCS (Gibco) and antibiotics (Sigma Chemical Co.). Cells were cultured in triplicates for 24 h at 37°C, 5% CO2. Supernatants were collected and frozen at –70°C.
Cytokine and chemokine assays
MIP-2, TNF-, IL-1ß, and IL-4 were analyzed with specific ELISA, as specified by the manufacturer (R&D Systems). All assays were performed in duplicate. Sensitivity was 15 pg/ml for IL-1ß, 15 pg/ml for TNF-, 31.25 pg/ml for MIP-2, and 12.5 pg/ml for IL-4.
Statistical analysis
Statistical analysis was performed with StatView software (Abacus Concepts, Berkeley, CA, USA). Data were compared using a two-tailed Student’s t-test, expressed as mean ± SEM, and considered statistically significant if P values were <0.05.
RESULTS
TNF- and MIP-2 induced by GIPL are reduced in CD1–/– mice
We have shown previously that highly purified and LPS-free GIPL from T. cruzi has TLR4-dependent, proinflammatory properties: At early time-points after i.p. injection, GIPL induces TNF- at comparable levels with the response observed during LPS challenge, and neutrophils are recruited into the peritoneal cavity [7
]. Attempting to characterize this response further, we investigated GIPL-induced neutrophil recruitment and TNF-, as well as MIP-2 production in CD1d–/– mice, which lack NKT cells. In fact, it has been demonstrated that GIPL from T. cruzi binds to recombinant mCD1d [10
], and CD1d-restricted NKT cells secrete cytokines that stimulate innate and adaptive responses, including neutrophil recruitment [11
, 12
]. Cells obtained from the peritoneal lavage were stained and analyzed by flow cytometry to identify neutrophils (Gr-1hi/Mac-1int cells), as shown in Figure 1A
and described previously [13
]. The increase in total cell numbers in the peritoneal cavity induced by GIPL in CD1d–/– mice (from 3.81x106±0.32 to 4.25x106±0.37 in a representative experiment) was in accordance with the increase in the percentage of the neutrophil population, which was approximately 13% in the same experiment. Also, no difference in the increase of total cell numbers induced by GIPL administration was found when comparing CD1d–/– and wt mice (data not shown). Moreover, as shown in Figure 1B
, the increase in the percentage of recruited neutrophils was statistically equal in CD1d–/– and wt mice. It is interesting, however, that 1.5 h following GIPL injection, TNF- and MIP-2 levels are reduced significantly in CD1d –/– mice compared with B6 wt controls. These results indicate that NKT cells play a role in the amplification of these cytokine and chemokine responses elicited by GIPL and raise the question about the role of TNF- and MIP-2 in GIPL-induced neutrophil recruitment.
 View larger version (47K): [in a new window] |
Figure 1. Neutrophil recruitment, TNF-, and MIP-2 induced by GIPL in CD1d–/– mice. (A) Cytometry dot-plot analysis of cells from the peritoneal lavage recovered from C57BL/6 wt, TNFRI (p55)–/–, or CD1d–/– mice, 3 h following injection of 100.0 µg GIPL or PBS i.p. Cells were stained with anti-Gr-1/PE and anti-Mac-1/FITC mAb. Neutrophils are Gr-1hi/Mac-1int cells. KO, Knockout. (B–D) Time course of neutrophil recruitment (B), TNF- (C), and MIP-2 (D) induced by GIPL in CD1d–/– mice. Data are expressed as the mean ± SEM of the percentages of Gr-1hi/Mac-1int cells in three mice analyzed individually at each point. MIP-2 and TNF- levels in the peritoneal fluid were measured by ELISA. These results are representative of two independent experiments. Asterisks indicate that the values are statistically different between C57BL/6 and CD1d–/– mice (*, P<0.05).
|
signaling in the inflammatory response elicited by GIPL, we investigated the kinetics of neutrophil recruitment in TNFRI–/– mice. In fact, the majority of biologic responses classically attributed to TNF- is mediated by TNFRI (p55), including neutrophil recruitment [9
, 14
15
16
]. The cytometry profile of the Gr-1hi/Mac-1int population obtained in GIPL-injected TNFRI–/– mice is shown in Figure 1A
. As depicted in Figure 2A
, in the absence of TNFRI expression, a significant decrease in the mean neutrophil accumulation, which corresponds to approximately 55% of the control mice response, was observable 4 h following treatment with GIPL. However, no difference in neutrophil recruitment between TNFRI–/– and wt mice was observable at earlier or later time-points of GIPL treatment. This indicates that TNF- signaling through the p55 receptor contributes to GIPL-induced neutrophil recruitment to the peritoneal cavity, although a substantial part of the response is maintained even in its absence. Conversely, neutrophil recruitment in TNFRI–/– mice treated with the prototype TLR4 activator LPS is diminished significantly (Fig. 2B)
.
 View larger version (20K): [in a new window] |
Figure 2. Neutrophil recruitment, TNF-, and MIP-2 induced by GIPL in TNFRI–/– mice. (A and B) Time course of neutrophil recruitment induced by 100.0 µg GIPL (A) or 10.0 µg LPS (B) in C57BL/6 and TNFRI (p55)–/– mice. Data are expressed as the mean ± SEM of the percentages of Gr-1hi/Mac-1int cells in three individually analyzed mice at each point. (C and D) MIP-2 and TNF- levels in the peritoneal fluid induced by 100.0 µg GIPL (C) or 10.0 µg LPS (D) in C57BL/6 and TNFRI–/–, measured by ELISA. These results are representative of two independent experiments. Asterisks indicate that the values are statistically different between C57BL/6 and TNFRI–/– mice (*, P<0.05).
|
IL-1ß production induced by GIPL is amplified by TNFRI signaling
IL-1ß and TNF- are cytokines with pleiotropic effects and some redundant functions. As in several circumstances, IL-1ß is a downstream effector of TNF-, we investigated whether the same holds true in the present system. For this, we compared the induction of IL-1ß following GIPL or LPS administration in B6 wt and TNFRI–/– mice. The kinetics of IL-1ß induction (Fig. 3B
) shows that IL-1ß levels were maximal at 3–4 h after treatment in LPS-injected wt mice, whereas in response to GIPL, the higher cytokine levels were detected at a 1-h time-point. Moreover, although substantial IL-1ß is detectable at the first hour following GIPL or LPS treatment, its levels are decreased in TNFRI–/– mice in response to GIPL and to a larger extent, in LPS-treated mice (Fig. 3B)
. Thus, an early IL-1ß production in response to GIPL could be detected in TNFRI–/– mice, although as observed for the response to LPS, the production of IL-1ß is prolonged in the presence of TNFRI signaling. Moreover, as demonstrated previously for neutrophil recruitment [7
], we found that GIPL-induced IL-1ß is totally dependent on the expression of functional TLR4 (data not shown).
 View larger version (15K): [in a new window] |
Figure 3. Kinetics of IL-1ß secretion induced by GIPL or LPS treatment. (A and B) Groups of three C57BL/6 and three TNFRI–/– mice were injected i.p. with PBS (circles), 100.0 µg GIPL (A) or 50.0 µg LPS (B), as indicated. Peritoneal lavage fluid was collected at the indicated time-points, and IL-1ß levels were measured by ELISA. Results obtained from individually analyzed mice are expressed as the mean ± SEM at each data point. (C) Neutrophil recruitment following IL-1ß neutralization. Groups of three C57BL/6 and three TNFRI–/– mice were injected i.p. with 20.0 µg/mouse specific goat anti-mouse IL-1ß or 20.0 µg/mouse purified goat IgG as control (ctr Ab), followed by 50.0 µg/mouse GIPL. Three hours later, peritoneal lavage fluid was collected, and the percentages of Gr-1hi/Mac-1int cells were evaluated by cytometry. The percentages of neutrophils in PBS-injected mice treated with anti-IL-1ß or control IgG were below 0.6%. Data are expressed as the mean ± SEM from three individually analyzed mice for each experimental condition.
|
.
In vivo activation of B and T lymphocytes following GIPL administration
As we have shown that TNF- and IL-1ß are induced by GIPL treatment, and it is known that both cytokines can enhance T cell activation in vivo, we raise the question of whether GIPL administration, by itself or as a consequence of the induced proinflammatory response, would have effects on lymphocyte activation. Although a direct action of GIPL on B and T cell activation in vitro has been reported previously [17
, 18
], little is known about its potential role in lymphocyte activation in vivo [19
]. To investigate this point, mice were injected with 50 µg GIPL in each footpad, and cells of the draining LN were collected 16–20 h afterwards for cytometry analysis. The percentage of the IgM+CD69+ cells in LN was 45.73% 16 h after GIPL treatment, as compared with 28.35% found in PBS-injected controls (50% increase; Fig. 4E
). Similarly, the increase in the percentage of CD3+CD69+ cells was close to 50%. The absolute numbers of cells in LN were not affected by this treatment (data not shown). It is interesting, however, that the increase in CD69+ T cells was much more pronounced (over 200%). Gating for lymphocytes in the forward-scatter x side-scatter dot plot showed that approximately 2% of cells were T lymphocytes in LN of control mice (Fig. 4A)
, and of these, 
9% were T cells naturally expressing the CD69 activation marker (Fig. 4B)
. It is remarkable that 20 h after GIPL injection, 21.15% of T cells expressed CD69 (Fig. 4D)
, but no increase in the percentage of total T cells was observed (Fig. 4C)
.
 View larger version (36K): [in a new window] |
Figure 4. Increase in the percentage of CD69+ lymphocyte T cells in LN after GIPL treatment. Four B6 mice in each group were injected with 50.0 µg GIPL or with PBS in each posterior footpad. Popliteal LN cells were collected 20 h later and stained with anti-IgM/FITC, anti-CD3/FITC, or anti- TCR/FITC and anti-CD69/PE mAb and analyzed by cytometry, as described in Materials and Methods. Dot plots with the corresponding subpopulations of +CD69+ and +CD69– cells from PBS-injected mice (A and B) and from GIPL-injected mice (C and D) are shown. Numbers indicate percentages of cells within the gated areas. (B) PBS and (D) GIPL show the percentages of CD69+ and CD69– cells of gated T cells. (E) The percentages of CD69+ cells among the IgM+, CD3+, and + populations are shown. Data are expressed as the mean ± SEM. Asterisks indicate that the values are statistically different between PBS- and GIPL-treated mice (*, P<0.05). These results are representative of five independent experiments.
|
ß and T cells is TLR4-dependentß T cells subsets was also examined. The ratio of CD69+ cells was increased in CD8+ and CD4+ populations (Fig. 5B
and 5C
), but CD8+ cells were more sensitive to GIPL treatment. The total percentage of CD8+ and CD4+ T cells in LN, however, was not augmented in GIPL-treated mice (not shown). We then investigated whether the GIPL-induced CD69 expression on T cells was also TLR4-dependent. As observed for the induction of IL-1ß, the activation of ß and T cells by GIPL was completely dependent on TLR4 expression, as no effect of GIPL injection was observed in the TLR4-deficient strain C57BL/10ScCr [20
] (Fig. 5A
5B
5C)
. Similar results were obtained with C3H/HeJ mice (not shown).
 View larger version (21K): [in a new window] |
Figure 5. GIPL-triggered CD69 expression is not observed in TLR4-deficient mice. Four C57BL/10 (B10; wt) or C57BL/10ScCr (ScCr; TLR4-deficient) mice in each group were injected with 50.0 µg GIPL or with PBS control, in each posterior footpad. Popliteal LN cells were collected 20 h later and stained with anti- TCR/FITC, anti-CD8/FITC, or anti-CD4/FITC and anti-CD69/PE mAb and analyzed by cytometry, as described in Materials and Methods. Data are expressed as the mean ± SEM. Asterisks indicate when the values are statistically different between PBS- and GIPL-treated mice (*, P<0.05). These results are representative of three independent experiments.
|
 View larger version (18K): [in a new window] |
Figure 6. TLR4 expression on T and B lymphocytes is not required for CD69 expression in response to GIPL. C57BL/10ScCr (107; TLR4-deficient) spleen cells were labeled with 1.0 µM CFSE and transfered i.v. to C57BL/10 (wt) mice. Four hours later, groups of four mice were subsequently injected with 50.0 µg GIPL or 20.0 µg LPS or PBS as control in each posterior footpad. Popliteal LN cells were collected 20 h later and stained with anti-CD8/APC or anti-B220/APC and anti-CD69/PE mAb and analyzed by cytometry. (A) Cytometry dot-plot analysis of CFSE+ and CFSE– cells following different treatments. (B) Mean percentages ± SEM of B220+/CD69+/CFSE+, B220+/CD69+/CFSE–, CD8+/CD69+/CFSE+, and CD8+/CD69+/CFSE– cells after PBS or GIPL treatment.
|
-dependent [21
], we investigated whether the GIPL-mediated activation of T and B lymphocytes would also be dependent on the TNF- pathway. With that scope, we compared the percentage of CD69+ lymphocytes in GIPL-treated TNFRI–/– and wt mice. As early as 5 h after treatment, we observed an increase in CD69+ T cells in B6 mice but a significantly reduced effect in the TNFRI–/– animals. Twenty hours after GIPL injection, this difference was even more evident (Fig. 7A
). Similar results were obtained for B cells (Fig. 7B)
. It is interesting, however, that signaling through TNFRI was not involved in ß T lymphocyte activation, as shown in Figure 7A
.
 View larger version (22K): [in a new window] |
Figure 7. TNFRI requirement for CD69-induced expression mediated by GIPL treatment. Groups of four C57BL/6 or TNFRI (p55)–/– mice were injected with 50.0 µg GIPL or with PBS in each posterior footpad. Popliteal LN cells were collected 5 h or 20 h later and stained with anti- TCR/FITC or anti-ß TCR/FITC (A) or anti-B220/FITC (B) and anti-CD69/PE mAb and analyzed by cytometry, as described in Materials and Methods. Percentages of +CD69+, ß+CD69+, and B220+CD69+ cells in PBS-injected controls were 8.4%, 7.5%, and 6.2% in B6 mice and 11.3%, 10.9%, and 6.8% in (p55)–/– mice, respectively. Data are expressed as the mean ± SEM. Asterisks indicate when the values are statistically different between C57BL/6 and TNFRI (p55)–/– mice (*, P<0.05). These results are representative of two independent experiments.
|
ß T cells increased 20 h following GIPL injection. To test the possibility that GIPL treatment also affects the capacity of splenic T cells to respond to subsequent TCR-dependent stimulation, we injected B6 mice with 100 µg GIPL i.p., and 20 h after treatment, spleen cells were stimulated with plate-bound anti-CD3 mAb. Supernatants were collected 24 h later, and cytokines were analyzed by ELISA. It is notable that IL-4 levels were highly augmented in cultures from GIPL-treated mice (Fig. 8B)
, suggesting an immunomodulatory action of GIPL on T lymphocytes.
 View larger version (21K): [in a new window] |
Figure 8. CD25 expression and IL-4 production by spleen cells of GIPL-treated mice. Three C57BL/6 mice were injected i.p. with 100.0 µg GIPL or with PBS, and spleen cells were isolated 20 h later. (A) Cells were stained with anti-ß TCR/FITC and anti-CD25/PE or anti-CD69/PE mAb and analyzed by cytometry, as described in Materials and Methods. Each data point is expressed as the mean ± SEM and represents the result obtained from mice individually analyzed (*, P<0.05). (B) Spleen cells were stimulated with immobilized anti-CD3 mAb for 24 h in triplicates, and the IL-4 present in the supernatants was quantified by ELISA in duplicates. Each data point is expressed as the mean ± SEM and represents the result obtained from mice individually analyzed. The asterisk indicates when the values are statistically different between PBS- and GIPL-treated mice (*, P<0.05). These results are representative of two independent experiments.
|
The magnitude and quality of the innate immune response exert a profound impact on the ensuing adaptive immune response. Neutrophils, macrophages, and dendritic cells secrete a series of mediators at the early stages of an immune response, which initiate a cascade of events that culminates in the generation of T and B cell responses and long-term immunity. Activation of cells through TLRs elicits the synthesis and release of a selected subset of inflammatory cytokines and chemokines, depending on the cell type and specific TLR being stimulated [22
]. We have described recently that T. cruzi GIPL elicits a TLR4-dependent, proinflammatory response, in which TNF- and MIP-2 are induced, followed by neutrophil recruitment [7
]. These results, together with the finding that TLR4-deficient mice are more susceptible to T. cruzi infection [7
], suggest that GIPL may act as a PAMP and play a critical role in determining the quality of the adaptive immune response to the parasite. A better characterization of the effects of GIPL on cell types involved in innate and adaptive immunity is therefore of great interest.
Evidence suggests that neutrophils make important contributions to adaptive immune responses, modulating cellular and humoral immunity. Experimental Chagas’ disease is one example of an infectious disease where neutrophils play a critical role in the generation of adaptive immunity [23
]. Particularly, CD1d-restricted NKT cells can help control infection by T. cruzi [24
] and promote the trafficking of neutrophils to the site of bacterial infection [11
]. Moreover, binding of T. cruzi GIPL to the CD1d molecule has been described [10
]. Here, we have shown that although no significant difference in the percentages of recruited neutrophils was observed between CD1–/– and B6 wt following GIPL treatment, TNF- and MIP-2 levels were reduced drastically in NKT-deficient mice. These results raise important questions about the proinflammatory response elicited by GIPL: The first concerns the mode of GIPL recognition by NKT cells and how this cell population influences cytokine and chemokine levels in response to GIPL; the second points to the role of TNF- and MIP-2 in the cascade, which leads to neutrophil recruitment in this system, as the diminished levels of these factors in GIPL-treated CD1–/– were not followed by a decrease in neutrophil attraction to the peritoneal cavity.
Although the activation of NKT cells as a result of the recognition of GIPL in the context of CD1d cannot be dismissed completely, this is certainly not the mandatory pathway of activation, as we have shown previously that GIPL-induced neutrophil recruitment and chemokine production are abrogated completely in TLR4-deficient mice, which have normal levels of CD1d expression and NKT cells [7 ]. Hence, NKT cells might be activated directly by GIPL recognition via the TLR4/MD-2 complex, which was shown to be expressed and functional in these cells [25 ]. Alternatively, NKT cells would be activated by the combined action of TCR signaling, given by CD1d-restricted GIPL recognition and cytokines released by other TLR4+ GIPL-responsive cells. Although the exact mechanisms remain elusive, our results clearly implicate CD1d-restricted NKT cells in an early amplification step of cytokine and chemokine production during the innate response elicited by T. cruzi GIPL, as demonstrated recently for the LPS-induced response [26 ].
The role of TNF- in GIPL-induced neutrophil recruitment was addressed here by analyzing GIPL- and LPS-induced responses in TNFRI (p55)-deficient mice. As reported previously [9
, 14
15
16
], we found that LPS-elicited production of MIP-2 and neutrophil recruitment are reduced significantly in TNFRI–/– mice; interestingly, however, neutrophil recruitment in response to GIPL was affected significantly only at intermediary time-points. Moreover, although diminished, a substantial neutrophil recruitment was still observable at 4 h following GIPL treatment in TNFRI–/– mice. Accordingly, MIP-2 levels, which are only detectable around the first hour of GIPL treatment in wt mice, are not affected in TNFRI–/– mice. These results demonstrate that the absence of TNFRI signaling may affect but is not essential for GIPL-induced neutrophil recruitment. Although the possibility remains that neutrophil recruitment in TNFRI-deficient mice is dependent on TNFRII signaling, this is a less probable hypothesis, taking into account the literature in the field [9
, 14
15
16
]. Therefore, other inflammatory cytokines may be involved in the process. In fact, distinct chemokines, cytokines, and other mediators can trigger neutrophil migration.
IL-1ß is a natural candidate to mediate neutrophil recruitment in GIPL-treated, TNFRI-deficient mice. The typical LPS-induced 3-h peak of IL-1ß production was not observable in TNFRI–/– mice, which have IL-1ß reduced markedly, in accordance with IL-1ß being induced by TNF- [27
]. Conversely, although IL-1ß levels in response to GIPL are also decreased in TNFRI–/– mice, the highest IL-1ß production in wt C57BL/6 mice is recorded at the first hour after GIPL injection. This indicates that at least part of the early IL-1ß production is independent of TNFRI signaling, and this response seems to be more pronounced in response to GIPL. The difference between GIPL- and LPS-induced responses regarding the kinetics of IL-1ß production may be explained by our previous results, showing that much lower TNF- is found in circulation after GIPL treatment, even if comparable production of TNF- is observed at a 1-h time-point in the peritoneal cavity of LPS- and GIPL-injected mice [7
]. It is possible that these two TLR4 agonists, LPS and GIPL, attain different circulation levels and clearance ratios after i.p. injection, and this could explain the observed difference. Conversely, it is also possible that although acting through the same receptor (TLR4), GIPL and LPS induce distinct signals, an issue that merits further investigation. Nevertheless, IL-1ß neutralization in wt and TNFRI–/– mice allowed us to clearly characterize the existence of two components in the proinflammatory response elicited by GIPL, which leads to neutrophil recruitment: one of which is IL-1ß- and TNFRI-independent, and the other is conditional on the TNF-/IL-1ß axis.
We have shown here that GIPL treatment induces the expression of the CD69 activation antigen in B and T cells: CD4+, CD8+, and particularly T cells in a TLR4-dependent manner. Although it was not formally excluded that the increase in the percentage of activated B and T cells may be a result of GIPL-induced, selective interference with cell migration, we find this possibility unlikely, as no significant changes in absolute cell numbers or in the percentage of distinct subpopulations in the LN were observed. Cell transfer experiments between TLR4-deficient and wt mice were done to establish whether CD69 expression on B and T cells was induced by direct signaling through TLR4 expressed on lymphocytes or alternatively, whether B and T cell activation in response to GIPL was secondary to cytokine secretion induced by GIPL in other cell types such as TNF- and IL-1ß production by macrophages. Our results demonstrated that the second hypothesis was correct. Also, we have shown here that the response of B and T cells in TNFRI-deficient mice is impaired severely, indicating that the GIPL-induced CD69 expression in these cells is, in large part, mediated by TNF-. Curiously, however, the picture is not the same for ß T cells, which attained similar levels of CD69 expression following GIPL treatment in TNFRI-deficient and wt mice. Hence, it is possible that for ß T cells, the GIPL-induced CD69 expression is a result of exclusive signaling of TNF- through TNFRII or to cytokines other than TNF-. Our results concerning GIPL-induced CD69 expression on T cells exhibit parallels to prior studies which investigated the in vivo effects of LPS on T lymphocytes [21
, 28
29
30
]. In these reports, it was demonstrated that LPS-induced CD69 expression on T cells relies on TNF- [21
] or type I IFN [29
], derived from monocytes/macrophages. It is interesting that one of these reports found that CD8+ T cells, when compared with CD4+ T cells, were more sensitive to the in vivo activation induced by LPS [29
], and another study found that T cells were preferentially stimulated by LPS and TNF- [21
]. Similar to TNF-, the GIPL-induced IL-1ß may have an important role in costimulating T cells, particularly T cells [31
]. The data presented here support the notion that GIPL-induced cytokine activation of lymphocytes may contribute to the massive polyclonal activation of B and T cells observed during T. cruzi infection [32
]. It is interesting that a large body of evidence points to a critical role of CD8+ T cells in the control of T. cruzi infection [33
], and it is possible that the preferential activation of CD8+ T lymphocytes by GIPL might contribute to the increased resistance of TLR4-expressing mice to T. cruzi infection [7
].
We also demonstrated that spleen cells of GIPL-treated mice secrete higher levels of IL-4 when activated by anti-CD3 mAb in vitro. This finding is distinct from studies in endotoxin-injected mice, where a general diminished capacity to produce cytokines, including IL-4, was described [30 ]. The current view is that TLR4 activation by LPS induces APC to produce cytokines that favor Th1-type immune responses. However, there is also evidence that some TLR-induced activation pathways may be important for Th2-type responses [34 ]. Our results raise the question of whether GIPL administration can modulate immune responses during antigenic challenge, and recent findings obtained by our group are in accordance with this hypothesis (Ramon Lemos, unpublished). This would be particularly important, as there is increasing interest in developing pharmacological tools as TLR antagonists and agonists, with the purpose of modulating the immune response and its application as adjuvants in vaccines.
ACKNOWLEDGEMENTS
This work was supported by Fundação Carlos Chagas Filho de Amparo à Pesquisa do Estado do Rio de Janeiro, Fundação Universitária José Bonifácio, Conselho Nacional de Pesquisas (CNPq), Programa Núcleo de Excelência, Mizutani Foundation for Glycoscience, and The Millennium Institute for Vaccine Development and Technology (CNPq, 420067/2005-1). M. M. M. and J. R. P. were M. Sc. students, and A-C. O. is a Ph.D. student of the Instituto de Microbiologia Prof. Paulo de Góes, Universidade Federal do Rio de Janeiro (UFRJ), and were supported by CNPq or Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) fellowships. L. C-R. is a M. Sc. student of the Instituto de Bioquímica, UFRJ, and is supported by a CAPES fellowship.
Received July 27, 2006; revised March 27, 2007; accepted April 25, 2007.
REFERENCES
14+ natural killer T cells in the innate phase of host protection against Streptococcus pneumoniae infection Eur. J. Immunol. 33,3322-3330[CrossRef][Medline] T cells by TNF- J. Immunol. 160,5221-523014+J18+ NKT cells to initiate contact sensitivity in vivo J. Immunol. 175,6390-6401/ß and / T cells from Listeria-infected mice: synergistic effects of interleukin 1 and 7 on / T cells J. Exp. Med. 178,985-996This article has been cited by other articles:
 |
|
 |
|
  F. F. Tuon, V. S. Amato, H. A. Bacha, T. AlMusawi, M. I. Duarte, and V. Amato Neto Toll-Like Receptors and Leishmaniasis Infect. Immun., March 1, 2008; 76(3): 866 - 872. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
