Jan 2, 1991 - Irradiated (700-radX rays) Fischer rats ..... this cell line can beadopted as an additional in vitro model .... Wachtel, and J. S. Remington. 1987.
Vol. 59, No. 9
INFECTION AND IMMUNITY, Sept. 1991, p. 3303-3308
0019-9567/91/093303-06$02.00/0
Trypanosoma cruzi but Not Trypanosoma brucei Fails To Induce a Chemiluminescent Signal in a Macrophage Hybridoma Cell Line B. VRAY,l* P. DE BAETSELIER,2 A. OUAISSI, AND Y. CARLIER4 Laboratoire de Parasitologie Experimentale, Faculte des Sciences,' and Laboratoire de Parasitologie, Faculte de Medecine,4 Universite Libre de Bruxelles, Brussels, and Unit of Cellular Immunology, Institute of Molecular Biology, Vrije Universiteit Brussel, Sint Genesius Rode,2 Belgium, and Centre d'Immunologie et de Biologie Parasitaire, Institut Pasteur, F-59000 Lille, France3 Received 2 January 1991/Accepted 27 June 1991
Macrophage-Trypanosoma cruzi interactions were studied by using a newly generated macrophage hybridoma cell line (2C11-12) that was selected for its capacity to produce high levels of reactive oxygen intermediates. This cell line was found to be a suitable host cell for T. cruzi, and intracellular parasitic development could be inhibited by activation with gamma interferon. When exposed to opsonized Trypanosoma brucei, Micrococcus lysodeikticus, or Legionella pneumophila, the activated macrophage cell line produces a high chemiluminescent signal, indicating the release of reactive oxygen intermediates. Alternatively, when opsonized T. cruzi was added to these activated macrophages, this parasite failed to stimulate a chemiluminescent response, suggesting an impairment in the triggering of the respiratory burst.
(Nunc, Roskilde, Denmark) with sterile round coverslips (Thermanox, 13-mm diameter; Miles Scientific, Naperville, Ill.). Trypomastigotes were added to the cells in various parasite-to-cell ratios. After 16 h, the cultures were washed with prewarmed medium to remove all free parasites. At different time intervals (24, 48, 72, and 96 h), the cells were fixed with methanol and stained with Giemsa stain. At least 200 cells per coverslip were microscopically counted to determine the level of infection. The parasitic index was calculated by multiplying the percentage of infected cells by the mean number of amastigotes per infected cell. The statistical significance of these results was determined by the nonparametric Mann-Whitney Wilcoxon test (39). 2C11-12 cells were activated by adding mouse recombinant IFN--y (1, 10, or 50 U/ml) to the culture medium 24 h before the cells were infected with parasites. The culture medium was removed 24 and 48 h after the initiation of the infection and replaced with fresh medium containing IFN-y. Polymyxin B sulfate (10 U/ml; Sigma) was added to inhibit the effect of lipopolysaccharide (LPS) traces which could be present in the culture medium (12, 21). For CL assays, 104 2C11-12 cells, previously activated for 48 h with mouse recombinant IFN--y (50 U/ml) or LPS B from Salmonella enteritidis (10 ,ug/ml; Difco, Detroit, Mich.), were grown in sterile polypropylene vials (Nunc). When IFN--y was used for cell activation, polymyxin B sulfate (10 U/ml) was added (see above). The opsonized or nonopsonized T. cruzi and T. brucei were added at a ratio of 20 parasites per cell unless otherwise indicated. When added to activated 2C11-12 cells, opsonized M. lysodeikticus cells induced a high CL signal and were routinely used as the positive control (we assumed that they induce 100% of the CL capacity of activated 2C11-12 cells). A ratio of 2,000 opsonized M. lysodeikticus cells per 2C11-12 cell was used (11). L. pneumophila was added at the same ratio. Each CL assay was repeated, in duplicate, at least three times. The CL signal was recorded over a 20-min period with a Biolumat (Berthold Co., Wildbab, Germany). The maximum CL emission curve (cpm x 1,000) was used to calculate the CL percentage value. An anti-T. cruzi serum (S-1) was produced by inoculating
When activated, macrophages are engaged in the respiratory burst. Following their interaction with microorganisms, activated macrophages release reactive oxygen intermediates (ROTs), which are thought to be involved in the killing of various parasites (19, 25) and yeasts (40). For example, gamma interferon (IFN--y)-activated macrophages are capable of impairing the intracellular multiplication of Trypanosoma cruzi (36, 37, 49), and hydrogen peroxide is capable of directly killing T. cruzi (15, 31, 32). However, there is a lack of direct evidence for ROI production during macrophage-T. cruzi interaction. Since the activated 2C11-12 macrophage hybridoma cell line produces high levels of chemiluminescence (CL) after treatment with phorbol myristate acetate or opsonized pathogens (9), we investigated the susceptibility of this cell line to infection by T. cruzi and its ability to produce a CL signal when interacting with this parasite. This response was compared with that generated by Trypanosoma brucei and the bacteria Micrococcus lysodeikticus and
Legionella pneumophila. T. cruzi (Y strain) was maintained in male BALB/c mice by weekly intraperitoneal inoculations. Irradiated (700-rad X rays) Fischer rats were used to produce large quantities of trypomastigote forms (17). T. cruzi epimastigotes were grown in GLSH medium at 28°C (20). T. brucei (strain ANTAT, l.l.E clone) was maintained by serial passage in mice and harvested as described previously (14). Lyophilized cells of M. lysodeikticus (ATCC 4698) were purchased from Sigma Chemical Co., St. Louis, Mo. L. pneumophila (serogroup 1, Philadelphia strain 1, ATCC 33152) was cultivated and harvested as described previously (35). The 2C11-12 cell line (10) was cultivated in RPMI 1640 medium (GIBCO, Grand Island, N.Y.) supplemented with N2-hydroxyethylpiperazine-N'-2-ethanesulfonic acid (HEPES; 25 mM), glutamine (2 mM), fetal calf serum (10%), NCTC135 medium (10%) (11), penicillin (100 IU/ml), and streptomycin (100 ,ug/ml; GIBCO) and incubated at 37°C in a 5% CO2 atmosphere. For infection experiments, 2C11-12 cells were seeded (105 cells per ml per well) in 24-well plates *
Corresponding author. 3303
3304
NOTES
INFECT. IMMUN.
BALB/c mice with irradiated trypomastigotes and then with 500 nonirradiated parasites (50). The absence of parasitemia indicated that the specific immune response was protective. Another anti-T. cruzi serum (S-2) was obtained by the intraperitoneal inoculation of BALB/c mice with 40 live trypanosomes. A very low parasitemia was observed during 25 days. The mice were then killed, and the sera were harvested, filtered, and pooled. A pool of sera (S-3) from healthy mice served as the control. The production of rabbit anti-T. cruzi epimastigote serum (S-4), anti-T. brucei serum (IRSb), anti-M. Iysodeikticus serum (IRSm), and anti-L. pneumophila serum (IRSp) has been previously described (9, 11, 18, 44). T. cruzi trypomastigotes (106 per ml) were opsonized with S-1, S-2, or S-3 (final dilutions, 1/10, 1/100, and 1/1,000, respectively) for 30 min at 37°C and then washed in RPMI medium. Fibronectin (FN) opsonization was performed by incubating T. cruzi parasites (106 per ml) with human FN at various concentrations (10, 50, 100, 200, and 400 ,ug/ml) for 45 min. Antibody opsonization was also performed with rabbit anti-FN immunoglobulin G, obtained as previously reported (33, 34), and used at a 1/40 final dilution in a second layer on previously FN-opsonized parasites. T. cruzi epimastigotes (106 per ml), T. brucei (107 per ml), M. lysodeikticus (1010 per ml), and L. pneumophila (1010 per ml) were
opsonized by incubating them with their respective rabbit specific immune sera (final dilution, 1/100) for 30 min at 37°C and then washing them in RPMI medium. In some experiments, T. cruzi trypomastigotes were opsonized by incubating them in fresh guinea pig serum (a source of complement; final dilution, 1/40) at 37°C for 30 min before being washed. Our results show that when parasites were incubated with 2C11-12 cells at different parasite-to-cell ratios (3/1, 10/1, 50/1, 100/1, and 200/1), the parasitic index increased from 243 to 2,190, in parallel with the greater number of parasites. A kinetic study of 2C11-12 cell infection by T. cruzi indicated a steady elevation of the parasitic index as a function of time after infection (data not shown). Microscopic observations of stained cells showed intracellular amastigotes progressively enlarging. After 72 h, most of them had differentiated into trypomastigotes, and numerous motile trypomastigotes in the cytoplasm, free-swimming trypomastigotes, and some amastigotes were clearly observed by phase-contrast microscopy. Additional kinetic studies with opsonized parasites were also performed (using the S-1 and S-2 anti-T. cruzi antisera; see above), and higher parasitic indexes were recorded (data not shown). There was a more statistically significant difference (P < 0.01) between nonopsonized and control serum (S-3)-treated parasites than between nonop-
o--o Si **- U S2 *.. S3
-4*
T
+
600 24 H
48 H
72 H
500-
x
400 -
z I-
(A)
300 a
0.
/ \
o--r ~ \' \
200 -
0..i*
100 -
i
.%.., %%
*_s
0U1.10 0~~~*-11
-
50
.Zl~~~~~~~~~~~~-~~
0-
II
I
0
1
10
50
0
1
10
50
0
1
10
50
I F N -F UNIT / ml FIG. 1. Parasitic index in relation to IFN--y activation. 2C11-12 cells were activated with various doses of IFN-,y and infected with parasites (parasite/cell ratio, 1:3). The parasitic indexes were recorded for nonopsonized parasites (A) or for parasites opsonized with S1 (0) or S2 (A) or treated with S3 (0).
VOL. 59,
sonized and S-1- or S-2-opsonized parasites. Parasites harvested from the infected 2C11-12 cell culture medium were inoculated in male BALB/c mice (4 x 105 parasites per mouse), and the animals developed a normal pattern of parasitemia, indicating that passage through the 2C11-12 cells did not alter the virulence of the parasites. When 2C11-12 cells were maintained in an activated state by IFN-y for three days after infection, a dose-dependent inhibitory effect on parasite multiplication was observed. As shown in Fig. 1, an extremely low parasitic index was recorded when 50 U of IFN--y per ml was used to activate host cells. When antibody-opsonized parasites were used to infect the 2C11-12 cells, IFN--y activation also induced a strong dosedependent inhibitory effect on parasite multiplication. To assess the capacity of T. cruzi to induce an oxidative burst in 2C11-12 cells, CL assays using T. cruzi and other, unrelated microorganisms were performed. No CL signal was recorded when nonactivated 2C11-12 cells were exposed to opsonized or nonopsonized T. cruzi, T. brucei, M. lysodeikticus, or L. pneumophila. When cells activated with IFN--y were exposed to nonopsonized microorganisms, a low and insignificant CL signal was recorded. Similar results were obtained with LPS (data not shown). However, a high CL response (2,320 x 103 + 1,042 x 103 cpm) was obtained when antibody-opsonized M. lysodeikticus interacted with IFN--y-activated 2C11-12 cells, confirming that these cells are capable of releasing high levels of ROIs. This bacterium-induced high CL level was obtained regularly in all experiments. Similarly, a high CL signal was also obtained with opsonized T. brucei and L. pneumophila. In contrast, T. cruzi opsonized with mouse anti-T. cruzi serum S-1 or S-2 induced a low CL emission (Table 1 and Fig. 2). FN favors the binding of T. cruzi parasites to macrophages (34, 48), including the 2C11-12 cell line (46a). Experiments using FN-opsonized parasites were performed, and they also gave weak CL signals (4.5% + 1.1%), similar to those observed with nonopsonized parasites. This indicates that antibody-independent as well as antibody-dependent receptor/ligand systems that promote the binding of T. cruzi to 2C11-12 cells were not capable of triggering a significant respiratory burst in 2C11-12 cells. Since the species specificity of the Fc receptor could play a role in the induction of antibody-dependent CL signals (T. brucei, M. lysodeikticus, and L. pneumophila were opsonized with rabbit antibodies, whereas T. cruzi was opsonized with mouse antibodies), T. cruzi was also opsonized with rabbit anti-FN immunoglobulin G after FN coating. However, the CL signal remained low (Table 1), indicating that neither the rabbit nor the mouse antibody opsonization was capable of triggering the respiratory burst. Similarly, only a very weak CL signal could be found by using complement-opsonized T. cruzi parasites. Since the high CL signal induced by antibody-opsonized bacteria was obtained with a high bacterium/macrophage ratio (2,000:1), antibody-opsonized parasite/macrophage ratios higher than the 20:1 ratio previously used were also tested. However, even at higher ratios, the CL signal remained low (Table 1). Triggering macrophages with LPS alone or with IFN--y and then with LPS has no enhancing CL effect (data not shown). Epimastigotes of T. cruzi were also unable to trigger the respiratory burst of IFN--y-activated 2C11-12 cells, except when S-4-opsonized parasites and a high parasite/cell ratio (300:1) were used; however, the CL signal was only 30% of the bacterial effect (Table 1). Our results indicate that 2C11-12 cells, similar to other
3305
NOTES
1991
TABLE 1. Percentage of CL recorded from IFN-y-activated 2C11-12 cells interacting with M. lysodeikticus, L. pneumophila, T. brucei, or T. cruzi Triggering agent and parasite/cell ratio
M. lysodeikticus 2,000:1
2,000:1 2,000: 1
Source of opsonin
% CL + SDa
IRSm Normal rabbit serum
100 3.5 0.9 3.8 ± 1.9
b
L. pneumophila 2,000:1 2,000:1 T. brucei 20:1 20:1 20:1 T. cruzi (trypomastigote) 20:1 20:1 20:1 20:1 20:1 20:1 20:1 20:1 20:1 20:1 20:1 50:1 100:1 200:1 700:1 T. cruzi (epimastigote) 10:1 100:1 300:1 10:1 100:1 300:1 10:1 100:1 300:1
IRSp
113.4 ± 10.1 1.2 1.2
IRSb Normal rabbit serum
106.8 ± 43,Oc 5.5 ± 1.6 6.7 ± 1.4
Complement S-1 S-1 S-1 S-i
4.9 5.3 9.2 9.6 8.3 8.2 9.9 5.8 4.5 5.7 2.3 5.9 11.1 13.0 10.0
S-4 S-4 S-4 Normal rabbit serum Normal rabbit serum Normal rabbit serum
0.6 0.4 4.3 1.7 3.7 30.8 1.2 0.5 6.1
S-i S-1 S-2 S-2 S-3 S-3 FN
Anti-FN IgGd
± ± ± ± ± ± ± ±
±
± ±
± ±
± ±
3.2 3.2 4.5 2.9 5.2 3.6 4.9 4.6 1.1 0.6 0.9 2.3 3.4 5.4 4.9
± 0.4 ± 0.5 ± 3.1 ± 0.8 ± 0.5 ± 8.8e ± 0.4 ± 0.3 ± 0.9
a Mean ± standard deviation of at least three experiments performed in duplicate. Unless otherwise indicated, P < 0.01 versus control as determined by the Mann-Whitney Wilcoxon test. , nonopsonized. 'P < 0.05 versus control. d IgG, immunoglobulin G. e
p
