Enhanced Interleukin-12 and CD40 Ligand Activities but Reduced Staphylococcus aureus Cowan 1-Induced Responses Suggest a Generalized and Progressively Impaired Type 1 Cytokine Pattern for Human Schistosomiasis

Patient population.Patients were examined and monitored at the outpatient clinic, Hospital das Clínicas, at the Federal University of Pernambuco, Recife, Brazil, a regional reference center for the study of tropical diseases. Patients were classified by form of schistosomiasis—intestinal, hepatointestinal, and HS—according to criteria proposed previously (13). The patients selected for the study displayed intestinal (n = 14), hepatointestinal (n = 8), and HS (n = 10) clinical forms of S. mansoni schistosomiasis (13). The subjects studied were men and women ranging from 18 to 68 years of age (average = 40.2 years [Table 1]). The patients with intestinal and hepatointestinal disease had stools positive for eggs but no splenomegaly (average, 413.9 and 27.0 eggs/g of stool, respectively [Table 1]). Patients with intestinal and hepatointestinal disease, less severe forms of chronic schistosomiasis, comprised a single group (HI) for statistical analysis because their responses were similar. The patients with HS disease (HS group) displayed typical hepatosplenomegaly with marked portal fibrosis confirmed by ultrasound examination (Table 1). They presented an average of 26.4 eggs/g of stool (Table 1).

Two of the 41 schistosomiasis patients presented concomitant intestinal parasite infections (Ascaris lumbricoides, Ancylostomus duodenalis, or Trichuris trichiura) and were treated before the study. All patients treated for schistosomiasis underwent a standard course of chemotherapy with oxamniquine at the time of study. The control group was from areas of nonendemicity in Brazil (n = 10). Healthy individuals from Brazil exhibited three negative stool examinations and had no contact with infested rivers or streams. All human studies described were reviewed by ethical and scientific review boards overseeing both the U.S. and Brazilian teams. Written informed consent was obtained from the patients before recruitment into the study. This study was approved by the Hospital das Clínicas Ethical Committee.

Antigens, antibodies, cytokines, and stimulators.Neutralizing anti-IL-10 (monoclonal antibody [MAb] JES-9D7) or rat immunoglobulin G1 (κ isotype MAb R59-40 as an isotype-matched control; Pharmingen, San Diego, Calif.) was used at a concentration of 5 μg/ml. Recombinant human IL-12 (rhIL-12) was used at a concentration of 20 ng/ml (generously provided by Genetics Institute, Cambridge, Mass.). rhIL-10 (catalog no. 19701V; Pharmingen) and human recombinant IFN-γ (rIFN-γ) (catalog no. 285-IF; R & D Systems, Minneapolis, Minn.) were used at 20 ng/ml. Lipopolysaccharides (LPS) from Escherichia coli (026:B6 L2654; Sigma Chemical Corporation, St. Louis, Mo.) were used at 1 μg/ml. S. aureus Cowan 1 strain, (SAC) (P7155; Sigma Chemical Corporation) was used at a concentration of 0.01%. Recombinant fibroblasts expressing CD40L on the surface were used to stimulate lymphocytes, whereas fibroblasts expressing surface CD32 were used as controls (both cell lines were kindly provided Brian Kelsall, National Institute of Allergy and Infectious Diseases, National Institutes of Health) (21).

Cell preparation and culture conditions.Whole blood was drawn with heparin (10 U/ml) and diluted 1:3 in RPMI plus penicillin (100 U/ml) and streptomycin (100 μg/ml). Cultures were maintained in a humidified atmosphere with 5% CO₂. Supernatants were collected at 48 h and were frozen immediately and stored at −70°C for determination of cytokine concentration.

Measurement of cytokine production by ELISA.The measurement of cytokines was performed by enzyme-linked immunosorbent assay (ELISA) using specific capture and detection MAbs following published protocols (43). Briefly, flat-bottom plates (Immulon 4; Dynatech, Chantilly, Va.) were coated with appropriate antibodies for the detection of IFN-γ (mouse anti-human #M700-A at 2 μg/ml; Endogen, Woburn, Mass.), IL-10 (anti-hu/v IL-10 9D7 at 2.5 μg/ml; Pharmingen), IL-12 (ELISA specific for p70) (anti-hu IL-12 MAb 611 at 5 μg/ml; R & D Systems), and tumor necrosis factor alpha (TNF-α) (anti-TNF-α human MAb M303-E at 2 μg/ml; Endogen). The plates were blocked with 5% evaporated milk in phosphate-buffered saline (PBS) at 37°C for 2 h and washed four times in 0.05% Tween 20 in PBS. Test samples and standards were diluted in 1% bovine serum albumin-0.05% Tween 20 in PBS (ELISA diluent), added to the plates, and incubated overnight at 4°C. Biotinylated second-step antibodies (polyclonal rabbit anti-human IFN-γ [Endogen], P-700 [1:1,000] [Endogen], anti-IL-10 JES1 12G8-biotin at 500 ng/ml [Pharmingen], anti-IL-12 p70 BAF219-biotin at 1 μg/ml [R & D], and anti-human TNF-α, M-302-B [1:1,000] [Endogen]) were diluted in ELISA diluent, added to the appropriate plates, and incubated for 2 h at 37°C. The plates were washed, and goat anti-rabbit immunoglobulin G-alkaline phosphatase (Fc-specific) 1:1,000 (Jackson ImmunoResearch, West Grove, Pa.) was added for the detection of IFN-γ, and streptavidin-horseradish peroxidase conjugate (1:750) was added for the detection of IL-10, IL-12, and TNF-α. After a final wash, the IFN-γ plates were developed with p-nitrophenyl phosphate substrate tablets (Sigma Chemical Corporation) dissolved in sodium carbonate buffer. IL-10, IL-12, and TNF-α plates were developed with the 2, 2′-azinobis(3-ethylbenzthiazolinesulfonic acid) (ABTS) one-step reagent (Kirkegaard and Perry, Gaithersburg, Md.) for approximately 20 min. The reactions were stopped with 10 μl of 0.3 N NaOH for IFN-γ plates and 10 μl of 10% sodium dodecyl sulfate for the other cytokines.

Statistical analysis.Statistical comparisons were made by Student's t test for paired samples and by analysis of variance for multigroup independent samples. A P of <0.05 was considered significant.

SAC-induced IFN-γ production is impaired, while rIL-12- and CD40L-induced responses are enhanced in schistosomiasis patients.Stimulation of WBC cultures with SAC (Fig. 1A) resulted in significantly increased production of IFN-γ in healthy controls (NC) and patients with hepatointestinal disease (HI group) in comparison with unstimulated cells (P = 0.020 and P = 0.010, respectively). NC also expressed significantly more IFN-γ than did HS patients (P = 0.019; Fig. 1A), suggesting that patients with HS schistosomiasis exhibit a defective SAC-induced cytokine response. Indeed, when SAC is used either alone or with rIL-12, anti-IL-10, or CD40L stimulation, the data clearly suggest that the IFN-γ deficiency develops progressively with the evolution of the disease (Fig. 1), because HI patients repeatedly show a response intermediate to that of NC and HS patients. Strikingly, SAC-induced responses were strongly inhibited by recombinant IL-10 (Fig. 1A), and the decrease was statistically significant for both NC and HI patients compared with SAC treatment alone (P < 0.001 and P = 0.030, respectively). Of interest, rIL-10 did not affect the already-weak response observed with HS cells. Moreover, neutralization of endogenous IL-10 failed to elevate the weak IFN-γ response in HS patients. In contrast, addition of anti-IL-10 MAb (Fig. 1A) increased IFN-γ production in SAC-stimulated cultures from NC and HI patients compared with SAC stimulation alone (P = 0.003 and P = 0.023, respectively). Similar to the results obtained following SAC treatment alone, IFN-γ levels were greater in NC than in HI patients (P = 0.011) following SAC-anti-IL-10 stimulation, with the latter group producing more IFN-γ than HS patients. The addition of rIL-12 markedly increased SAC-induced IFN-γ in all groups, although differences were significant only for NC and HI patients (for SAC versus SAC plus rIL-12, P = 0.032 and P < 0.001, respectively). Of interest, IL-12 treatment alone increased IFN-γ production in both HI and HS patients, although the differences were significant only for the HI group (for medium versus IL-12, P = 0.004). Notably, the NC consistently failed to show a significant response to exogenous rIL-12 (HI patients versus NC, P = 0.004). Surprisingly, these data suggest that schistosome-infected patients in general exhibit greater responsiveness to IL-12, despite the fact that they simultaneously manifest a significant and progressive defect in SAC- and SAC-CD40L-induced (see below) IFN-γ production. It is interesting, however, that the response of NC was dramatic when IL-12 was combined with SAC stimulation, which suggests an activation stimulus is required to promote IL-12 responsiveness in uninfected individuals. In this regard, production of IFN-γ was significantly higher in the NC group than in HS patients (P = 0.044).

• Open in new tab
• Download powerpoint

FIG. 1.

IFN-γ levels in whole-blood cultures from NC and HI and HS schistosomiasis patients. (A) WBC cultures were stimulated with SAC, rIL-10, anti-IL-10, and rIL-12. Symbols: ∗, result for the group is significantly different from that for the corresponding SAC-stimulated group; ∗∗, result is significantly different from that for medium alone. SEM, standard error of the mean. (B) WBC cultures were stimulated with SAC, CD40L-expressing fibroblasts, control CD32, or LPS. The significance level (P) for different patient groups within a specific treatment regimen is shown in the graph. ∗, result for the group is significantly different from the corresponding control for each panel (CD32, SAC-CD32, or media).

The HI and HS patients also showed a weak response to CD40L stimulation alone (Fig. 1B, left panel); this was again in contrast to NC, who displayed no response. It is important that the increase in infected individuals only reached significance with the HI cultures (P = 0.010). There was no IFN-γ detected in the control non-CD40L-expressing fibroblast cultures, which confirms the specificity of the CD40L-induced response. Thus, patient cells appear to manifest an increased responsiveness to both CD40L (Fig. 1B) and rIL-12 (Fig. 1A); however, as observed with rIL-12, a marked effect of CD40L was induced in NC when SAC was combined with CD40L. Although consistently lower than the response in NC, the response in cultures from HI patients was also elevated by this combination, while HS patients showed no additional increase over that observed with CD40L stimulation alone (SAC-CD40L versus SAC-CD32, Fig. 1B).

Finally, LPS failed to stimulate a significant IFN-γ response in any group, which was in marked contrast to SAC or CD40L. The slight LPS-induced increase in IFN-γ was significant only for NC (P = 0.007; Fig. 1B) when compared with medium alone.

Production of IL-10 is elevated in WBC cultures from HI patients following activation with multiple T-cell-dependent and T-cell-independent stimuli.Stimulation of WBC cultures with SAC (Fig. 2A) resulted in increased production of IL-10 in NC, HI, and HS patients compared with cells in medium alone (P < 0.001, P = 0.025, and P = 0.202, respectively); however, the SAC-induced response was less apparent in cultures from HS patients, which produced significantly less IL-10 than those from the HI (P = 0.018) group. More obvious, however, was the consistently more robust IL-10 response observed in SAC-stimulated cultures from HI patients. Addition of rIFN-γ failed to modify these responses, while rIL-12 slightly reduced the SAC-induced response in cultures from HI patients. It is important to note, however, that the latter changes were modest. Perhaps more intriguing, rIL-12 administration alone markedly induced IL-10 expression in both infected groups (Fig. 2A), thus mirroring what was observed for IFN-γ in these cultures (Fig. 1A) and further emphasizing the conclusion that infected individuals are more responsive to exogenous IL-12 (NC versus HS patients, P = 0.001). Here, however, in contrast to IFN-γ (Fig. 1A), the combination of SAC and rIL-12 did not have an additive effect on IL-10 expression in any patient group (Fig. 2A).

• Open in new tab
• Download powerpoint

FIG. 2.

IL-10 levels in whole-blood cultures from NC and HI and HS schistosomiasis patients. (A) WBC cultures were stimulated with SAC, rIL-12, and IFN-γ. Symbols: ∗, result for the group is significantly different from that for the corresponding SAC- or medium-stimulated control group in the same set of panels; ∗∗, result is significantly different from that for medium alone. SEM standard error of the mean. (B) WBC cultures were stimulated with SAC, CD40L-expressing fibroblasts, control CD32, LPS, or IFN-γ. The significance level (P) for different patient groups within a specific treatment regimen is shown in the graph. ∗, result for the group is significantly different from that for the corresponding control for each panel (CD32, SAC-CD32, or media).

In addition, all three groups showed a weak response to CD40L stimulation alone (Fig. 2B, far left panel), although the change was significant only for NC (CD40L versus CD32, P = 0.049). Strikingly, the effects of SAC were potentiated by CD40L (Fig. 2B). Indeed, the increase in IL-10 was significant for both HI patients (P = 0.017) and NC (P = 0.004) (SAC-CD32 versus SAC-CD40L). Moreover, the pattern in general mirrored what was observed for SAC stimulation alone (Fig. 2A), because the HI group clearly exhibited a more robust response than did HS (P = 0.008) or NC (P = 0.014) patients. Somewhat surprisingly, the addition of IFN-γ to cultures stimulated with SAC, SAC-CD40L, or LPS failed to decrease IL-10 production (Fig. 2). Finally, cultures stimulated with LPS alone or LPS/IFN-γ exhibited an IL-10-producing profile that was similar to the SAC-CD40L-induced response, thus further highlighting the conclusion that HI patients exhibit a more exuberant IL-10 response (HI versus HS, P = 0.047).

TNF-α production is similar following LPS or SAC stimulation, while CD40L-induced responses are completely absent in HS individuals.In marked contrast to results with IFN-γ and IL-10, SAC-induced TNF-α expression was significant in all patient groups (Fig. 3A), although HS patient cultures showed some diminution in their response versus HI (P = 0.006), and versus NC (P = 0.029). SAC-induced TNF-α was also strongly inhibited in all groups by the addition of rIL-10 (Fig. 3A, far left panel), although the decrease was significant only for the NC and HI groups (P < 0.001 and P < 0.001). By contrast, neutralization of endogenous IL-10 in SAC-primed cells triggered a marked increase in TNF-α production in NC (SAC versus SAC-anti-IL-10, P < 0.001), with a more variable but similar response in both HI and HS patient cultures. Indeed, no significant differences in TNF-α were observed among the three patient groups when SAC was combined with IL-10 neutralization, which suggests the partial impairment in TNF-α production in HS patients with SAC alone is primarily attributable to the downregulatory effects of IL-10. Notably, similar to addition of anti-IL-10 MAb, addition of exogenous IL-12 restored a normal TNF-α response in HS patients. In fact, although IL-12 significantly upregulated TNF-α production in both NC and HS patients, the effect in HS cultures was much more dramatic. Notably and again consistent with our results with IFN-γ (Fig. 1A) and IL-10 (Fig. 2A), administration of rIL-12 alone triggered significant cytokine expression in both HI and HS patients (medium versus IL-12, P = 0.013 and P = 0.044, respectively), and this was also significant versus NC (NC versus HI, P = 0.048; NC versus HS, P = 0.039) (Fig. 3A); however, unlike IL-12, addition of rIFN-γ to SAC-stimulated cells failed to restore normal TNF-α levels in HS patient cultures, although there was some upregulation. Thus, similar to SAC stimulation alone, the response in NC and HI patients was significantly greater than in HS patient cultures (Fig. 3A). Also unlike our rIL-12 findings, exogenous IFN-γ alone had little effect, although a modest but significant increase in TNF-α was observed in HI patients (medium versus IFN-γ, P = 0.026).

• Open in new tab
• Download powerpoint

FIG. 3.

TNF-α levels in whole-blood cultures from NC and HI and HS schistosomiasis patients. (A) WBC cultures were stimulated with SAC, IFN-γ, IL-12, IL-10, or anti-IL-10. Symbols: ∗, result for the group is significantly different from that for the corresponding SAC- or medium-stimulated control group in the same set of panels; ∗∗, result is significantly different from medium alone. SEM, standard error of the mean. (B) WBC cultures were stimulated with SAC, CD40L-expressing fibroblasts, control CD32, LPS, or IFN-γ.. The significance level (P) for different patient groups within a specific treatment regimen is shown in the graph. ∗, result for the group is significantly different from that for the corresponding control for each panel (CD32, SAC-CD32, or media)

A modest increase in TNF-α production was observed in HI patient cultures following CD40L stimulation (Fig. 3B, far left panel; compare CD32 control with CD40L). The combined stimulation with SAC-CD40L resulted in significantly increased production of TNF-α in NC (P < 0.001) and HI patients (P = 0.040) when compared with the same groups stimulated with SAC-CD32 (Fig. 3B). When these groups were compared internally, we noted that NC (P = 0.039) and HI patients (P = 0.040) produced significantly more TNF-α than HS patients. The addition of IFN-γ to these cultures did not change these responses dramatically, although the increase in the NC group was more obvious when compared with HI patients (P = 0.003) and HS patients (P = 0.006). Moreover, the modest CD40L/SAC/IFN-γ response of HI patients was now insignificant.

Finally, LPS increased TNF-α expression in NC and HI and HS patients (P = 0.002, P = 0.001, and P = 0.028, respectively) compared with unstimulated cells (Fig. 3B, far-right panels). Moreover, production of TNF-α was further enhanced in all groups with the addition of exogenous IFN-γ, especially in the NC group.

CD40L-induced IL-12 p70 expression is enhanced in infected individuals.A modest but significant increase in IL-12 p70 expression was observed in WBC cultures from NC following stimulation with CD40L (P = 0.045). Although the increase in HI and HS patients was more obvious (Fig. 4), it was significant only for the HI group (CD40L versus CD32, P = 0.043). Indeed, stimulation with CD40L produced more IL-12 in HI patients than in NC (P = 0.015). Combined stimulation with SAC and CD40L increased the response in both HI patients and HS patients more than in NC (HS versus NC, P = 0.026). Finally, IL-12 in the cultures from HI and HS patients stimulated with the combination of SAC-CD40L and IFN-γ was significantly increased over that in the control SAC-CD32 group (P = 0.009 and P = 0.027, respectively). There was also a modest but significant response in the NC group (P = 0.034), though less than in the HI (P = 0.002) and HS (P = 0.005) groups. LPS or LPS in combination with IFN-γ had little or no effect in any group (Fig. 4, far-right panel). However, HS patients were more responsive than the NC group (P = 0.025). As observed for LPS, the response in cultures stimulated with SAC alone (or SAC in combination with anti-IL-10 or IFN-γ) was extremely weak and highly variable (data not shown).

• Open in new tab
• Download powerpoint

FIG. 4.

IL-12 p70 levels in whole blood cultures from NC and HI and HS schistosomiasis patients. WBC cultures were stimulated with SAC, IFN-γ, CD40L, CD32, or LPS. ∗, result for the group is significantly different from that for the corresponding control for each panel (CD32, SAC-CD32, or media). SEM, standard error of the mean.