TEXT

Top Abstract Text References

Maternal-fetal transmission of Trypanosoma cruzi, the protozoan parasite agent of Chagas' disease in Latin America, occurs in 2 to 12% of pregnancy in chronically infected mothers, inducing severe disease and significant mortality (4, 7). The factors enabling the vertical transmission to occur, as well as those allowing the vast majority of babies of infected mothers to remain uninfected, are not entirely known. Maternal-fetal transfer of antigens might influence the capacity of the progeny to respond to infection through modulating the fetal immune system (11, 12, 19, 22, 28, 39). However, there is little information on innate immunity of neonates in the case of maternal infection, whereas monocyte activation plays a central role in controlling infection (20) as well as in maintaining pregnancy (36). To better understand the maternal-fetal immunological relationship in human T. cruzi infection, we explored the capacity of uninfected neonates and their infected mothers to produce type 1 (gamma interferon [IFN-] and interleukin-2 [IL-2]) and type 2 (IL-4 and IL-5) cytokines, as well as proinflammatory (IL-1, IL-6, and tumor necrosis factor alpha [TNF-]) and anti-inflammatory (IL-10 and soluble TNF receptor [sTNFR]) factors.

Seventeen asymptomatic mothers chronically infected with T. cruzi and 58 uninfected mothers from the Maternity German Urquidi (Universidad Mayor de San Simon [UMSS]), Cochabamba, Bolivia, and their uninfected neonates were enrolled in this study. All newborns were delivered at term. Cases of known pathology of newborns or mothers during pregnancy, twins, or cesarean-born babies were excluded. Cross-reactive protein levels in maternal and cord plasma, determined by routine immunoturbidimetry, were within the normal ranges (means ± standard errors of the means [SEM]): 1.30 ± 0.17 and 1.19 ± 0.32 mg/dl for uninfected and infected mothers, respectively, and below the detection limit of 0.2 mg/dl for neonates). This study was approved by the scientific/ethic committees of UMSS and Free University of Brussels (ULB), and informed written consent of the mothers was obtained before blood collection.

Cord and maternal blood were collected just after delivery in endotoxin-free heparinized tubes (Becton Dickinson, MLS SA, Menin, Belgium) and immediately used for stimulation of whole blood cells (WBC) and study of cytokine production. Cord blood (10 ml) was also collected on EDTA disodium salt (Sigma, St. Louis, Mo.), added to 10 ml of 6 M guanidine hypochloride (Sigma), and kept at room temperature until use for T. cruzi-specific PCR.

T. cruzi trypomastigotes and amastigotes (Tehuantepec strain) were obtained from culture medium of infected 3T3 fibroblasts as previously described (18). After three washings in RPMI 1640 medium, they were submitted to eight cycles of freezing-thawing under sterile conditions. The parasite lysate was stored at 70°C until use for cell stimulation. The prepared batch of T. cruzi lysate contained no detectable endotoxin (<1 pg/10⁷ lysed parasites), as determined by the Limulus amebocyte lysate assay (BioWhittaker Europe, Verviers, Belgium). Lipopolysaccharide (LPS; from Escherichia coli O111B4) and phytohemagglutinin (PHA) were purchased from Sigma.

Maternal infection was assessed using the classical parasite-specific serological tests of enzyme-linked immunosorbent assay (ELISA), hemagglutination, and immunofluorescence (9). T. cruzi-specific immunoglobulin M (IgM) antibodies were not detected in maternal plasma by ELISA, indicating that mothers were in the chronic form of Chagas' disease. T. cruzi congenital infection in newborns was sought in cord blood by (i) direct analysis of parasites by microscopic examination of buffy coat of blood collected in four heparinized microhematocrit tubes (each of 75 µl) (23), (ii) T. cruzi-specific PCR, performed with the primers S35 (5'-AAA TAA TGT ACG GG(T/G) GAG ATG CAT GA-3') and S36 (5'-GGG TTC GAT TGG GGT TGG TGT-3') (42) as described by Centurion-Lara et al. (13), and (iii) detection of T. cruzi-specific IgM and IgA antibodies by specific ELISA. These techniques allowed us to detect one congenitally infected newborn in 18 seropositive mothers (5.6%), who was not included in this study.

The WBC stimulation test was performed as previously reported (15), with minor modifications. Tenfold-diluted blood was incubated in polypropylene tubes (Falcon) either without stimulating agents or in the presence of either LPS (10 ng/ml) plus PHA (5 µg/ml) or T. cruzi lysate (10⁶ lysed parasites/ml). After 24 or 72 h of incubation at 37°C in a 5% CO₂ atmosphere, the samples were centrifuged and the supernatants were collected and kept at 70°C. Cytokine levels were determined using the following commercially available ELISAs: for IL-2, IFN-, and TNF- (detecting free and soluble TNFR-bound TNF-), EASIA from Medgenix, BioSource Europe, Nivelles, Belgium; for IL-4 and IL-10, ultrasensitive assays from Biosource, Nivelles, Belgium; for IL-1, standard and antibody pairs from Genzyme, R&D Systems Europe, Abigdon, Oxon, United Kingdom; and for IL-5, Duoset Genzyme. Antibodies and standards kindly given by W. Buurman (Department of Surgery, University of Limburg, Maastricht, The Netherlands) were used for sTNFR type 1 (sTNFR1), sTNFR2, and IL-6 determinations, performed as previously described (16, 26, 38). The detection limits of ELISAs were as follows: for IL-1 and IL-2, 10 pg/ml; for IL-4, 0.07 pg/ml; for IL-5, 8 pg/ml; for IL-10, 0.20 pg/ml; for IFN-, 1.5 pg/ml; for IL-6, TNF-, sTNFR1, and sTNFR2, 20 pg/ml. Results after WBC stimulation are expressed as the differences between the cytokine levels of stimulated and unstimulated cell samples (supernatants of 1/10-diluted blood) at the same incubation time, without further correction by the dilution factor. Results are expressed as arithmetic means ± SEM of all individual patients tested in each group. The Mann-Whitney-Wilcoxon U test was used for comparison between groups.

(i) Cells of T. cruzi-infected mothers product IFN-, IL-2, and IL-4 in response to parasite stimulation, whereas neonate cells occasionally release IFN-. Whatever the patient group (infected or uninfected mothers and their neonates), IFN-, IL-2, IL-4, and IL-5 were not spontaneously released by blood cells or were released only at similar background levels in all patient groups. Mitogenic stimulation with LPS-PHA showed that maternal and neonatal cells were able to produce the type 1 cytokines IL-2 and IFN-, with no significant difference according to maternal infection status (Fig. 1A and C). Though neonates are generally considered poor IFN- producers (45), in this study newborn WBC were as able as WBC from the mothers to produce significant amounts of IFN- and IL-2, although with slightly different kinetics: newborns were more prone to produce IL-2 at 24 h than their mothers, whereas IFN- release was delayed. By contrast, but in agreement with previous work (46), the type 2 cytokines IL-4 and IL-5 were not produced by neonates from both groups in the presence of mitogens, whereas their mothers did produce these cytokines (Fig. 1E and G).

View larger version (13K): [in this window] [in a new window]   FIG. 1.   Production of type 1 and type 2 cytokines by WBC from T. cruzi-infected () or uninfected () mothers and their newborns. WBC were stimulated or not for 24 or 72 h with LPS (10 ng/ml) plus PHA (5 µg/ml) (A, C, E, and G) or a lysate of T. cruzi (106 lysed parasites/ml) (B, D, F, and H). Numbers of individuals in each group ranged from 26 to 43 for uninfected mothers, 11 to 16 for infected mothers, 25 to 47 for neonates from uninfected mothers, and 7 to 13 for uninfected neonates from infected mothers. Results (mean ± SEM) are expressed as the differences between levels obtained for stimulated and unstimulated cells. The Mann-Whitney-Wilcoxon U test was used for statistical comparisons between infected and uninfected mothers and between their newborns (*, P < 0.05; **, P < 0.005).

(ii) Maternal T. cruzi infection upregulates the capacity of maternal and neonate cells to produce proinflammatory cytokines. In contrast to lymphocytic cytokines, IL-1, IL-6, and TNF- were spontaneously released by blood cells from mothers as well as their neonates (data not shown), likely reflecting the monocytic activation normally associated with pregnancy and delivery (3, 37). As shown in Fig. 2A, C, and E, such spontaneous inflammatory cytokine release was enhanced in mothers and neonates upon in vitro stimulation with LPS-PHA (except for IL-6 in neonates of uninfected mothers). Moreover, IL-6, and to a lesser extent IL-1, levels were still higher in infected mothers and also, surprisingly, in their uninfected newborns. The parasite lysate strongly stimulated the production of TNF- and IL-6, and slightly that of IL-1, in both chagasic mothers and their neonates (Fig. 2B, D, and F). A detailed analysis of individual results indicated that a simultaneous higher production of these cytokines was observed in 58% of newborns. T. cruzi infection in mothers could easily explain their higher capacity to produce inflammatory cytokines. Indeed, although PHA-activated lymphocytes could be the source of IL-6, the simultaneous overproduction of three inflammatory cytokines suggests a monocytic origin. Different mechanisms could account for monocytic activation in infected mothers: (i) in vivo monocyte priming, known to occur during T. cruzi infection (8), strengthening the direct effect of parasite molecules supporting proinflammatory activities (2, 10, 21, 35) and present in the parasite lysate used to activate cells; (ii) monocytic activation due to IFN- released in vitro by specific lymphocytes after recognition of T. cruzi antigens; or (iv) proinflammatory cytokine release resulting from a cross-linking of FcR on monocytes and/or NK cells (6, 14, 27) by immune complexes formed by the T. cruzi-specific antibodies present in the blood of infected mothers and the parasite lysate added in vitro.

View larger version (17K): [in this window] [in a new window]   FIG. 2.   Production of proinflammatory cytokines by WBC from T. cruzi-infected () or uninfected () mothers and their newborns. WBC were stimulated or not for 24 or 72 h with LPS (10 ng/ml) plus PHA (5 µg/ml) (A, C, and E) or a lysate of T. cruzi (106 lysed parasites/ml) (B, D, and F). Numbers of individuals in each group ranged from 23 to 42 for uninfected mothers, 7 to 13 for infected mothers, 16 to 46 for neonates from uninfected mothers, and 8 to 15 for uninfected neonates from infected mothers. Results (mean ± SEM) are expressed as the difference between levels obtained for stimulated and unstimulated cells. The Mann-Whitney-Wilcoxon U test was used for statistical comparisons between infected and uninfected mothers and between their newborns (*, P < 0.05; **, P < 0.005).

(iii) Maternal T. cruzi infection upregulates the capacity of maternal and newborn cells to produce anti-inflammatory factors. The effect of maternal infection on proinflammatory cytokines prompted us to also investigate the production of the potent anti-inflammatory factor IL-10 (41), as well as the sTNFR1 and sTNFR2, previously shown to play an essential role in modulating TNF bioactivity (31), particularly in experimental Chagas' disease (43). Though only traces of IL-10 could be found in supernatants of unstimulated cells from newborns and mothers (data not shown), LPS-PHA stimulated IL-10 production by WBC from all patient groups (Fig. 3A), whatever the maternal infection status. By contrast, maternal infection was associated with a higher reactivity of both mother and newborn WBC to parasite lysate, which produced two- to threefold more IL-10 after 24 h of culture than the control couples (Fig. 3B). Infection status did not modify the WBC release of both sTNFRs, either spontaneously (data not shown) or after incubation with LPS-PHA (Fig. 3C and E). By contrast, T. cruzi components induced a higher release of both sTNFRs in infected mothers and their newborns (Fig. 3D and F). The overproduction of IL-10 and sTNFR by cells incubated with the parasite lysate was observed in the same newborns from infected mothers as those who responded upon proinflammatory cytokine production.

View larger version (17K): [in this window] [in a new window]   FIG. 3.   Production of anti-inflammatory cytokine and factors by WBC from T. cruzi-infected () or uninfected () mothers and their newborns. WBC were stimulated or not for 24 or 72 h with LPS (10 ng/ml) plus PHA (5 µg/ml) (A, C, and E) or a lysate of T. cruzi (106 lysed parasites/ml) (B, D, and F). Numbers of individuals in each group ranged from 22 to 41 for uninfected mothers, 9 to 15 for infected mothers, 26 to 43 for neonates from uninfected mothers, and 8 to 13 for uninfected neonates from infected mothers. Results (mean ± SEM) are expressed as the difference between levels obtained in stimulated and unstimulated cells. The Mann-Whitney-Wilcoxon U test was used for statistical comparisons between infected and uninfected mothers and between their newborns (*, P < 0.05; **, P < 0.005).