In situ immunoassay for the assessment of Trypanosoma cruzi interiorization and growth in cultured cells

Acta Tropica, 57(1994)301 306 © 1994 Elsevier Science B.V. All rights reserved 0001-706X/94/$07.00

301

ACTROP 00404

In situ immunoassay for the assessment of Trypanosoma cruzi interiorization and growth in cultured cells Roxane Maria F. Piazza a'*, Heitor Franco de Andrade Jr. b, Eufrosina S. U m e z a w a b, A l e j a n d r o M i g u e l K a t z i n °, A n n a M a r i a S. S t o l f c "Laboratrrio de Parasitologia, Instituto Butantan, S~o Paulo, bInstituto de Medicina Tropical de S~o Paulo, CDepartamento de Parasitologia, Instituto de Ci~ncias BiomOdicas, Universidade de Sao Paulo, Sao Paulo, Brazil Received 27 January; revision received 26 April 1994; accepted 26 April 1994

Interiorization and multiplication of Trypanosorna cruzi within its host cells are usually assessed by counting parasites in fixed and stained cover slip preparations, a subjective and time-consuming method. Here we describe an immunoenzymatic assay (ELISA) for assessing the number of internalized parasites in infected LLC-MK2 seed on chamber slides (NUNC). ELISA was performed employing a rabbit polyclonal serum against trypomastigote components (MOP) and anti-rabbit IgG conjugated to peroxidase. The bottom of the chamber slide was then detached and processed for quantification of internalized parasites by the conventional method. Data analysis showed a linear relationship between optical densities and number of internalized parasites (r 2 = 93.99, p < 0.001 ). The assay was also efficient to assess inhibition of parasite interiorization induced by the monosaccharide NAc-t)-glucosamine. Key words: Trypanosorna cruzi; In situ ELISA; In vitro infection

Introduction American trypanosomiasis or Chagas' disease, endemic in South America, is caused by the protozoan Trypanosoma cruzi, which can develop within many cell types (Brener, 1973; de Souza, 1984; Trager, 1986). Host cell interaction with T. cruzi is studied by incubating infective forms of the flagellate with in vitro cultured cells. The numbers of internalized parasites, the percentage of infected cells and other infectivity indexes are evaluated by microscopical analysis, after fixing and staining infected cells seed on cover slips (Andrews and Colli, 1982; Piras et al., 1983). Besides the subjectiveness in counting, this method is tedious and time consuming and frequently affected by fixation and staining problems. An enzyme immunoassay for evaluation of Toxoplasma gondii growth in tissue culture has been described, where the authors showed a correlation between ELISA *Corresponding author. Correspondence address: R.M.F. Piazza, Laborat6rio de Parasitologia, Instituto Butantan, Av. Vital Brazil, 1500, 05503-900, S~o Paulo, SP, Brazil TeL: (55) (11) 8137222 ext. 2242/2080; Fax: (55) (11) 8151505. SSDI 0 0 0 1 - 7 0 6 X ( 9 4 ) 0 0 0 3 9 - 4

302 data and either the number of trophozoites in culture supernatant or the percentage of rosettes inside cells (Merli et al., 1985). A similar method was also used for the evaluation of chemotherapeutic drugs against T. cruzi, showing a linear correlation between number of internalized parasites and optical densities of infected cells (De Titto and Arafijo, 1988; Luz et al., 1993). In these communications, however, cells were cultivated and infected in two different systems, one for each analysis: cover slips for microscopical counting and ELISA in microplates for immunoenzymatic assay, employing human chagasic sera. Here we describe an in situ immunoassay, in which the efficiency in quantifying internalized parasites was highly improved by employing chamber slides that allow the assessment by both immunoenzymatic method and microscopical counting in the same system. The assay was also efficient when applied to test inhibition of T. cruzi interiorization into LLC-MK2 cells by increased concentrations of NAc-oglucosamine in the culture medium (Andrews and Colli, 1981; Crane and Dvorak, 1982; Piras et al., 1983, Colli et al., 1984).

Materials and methods Parasites and cells

LLC-MK2 cells were seeded at several cell densities (2.5, 5, 10 x 1 0 4 cells/cm2), in a 16-well chamber slide (NUNC) and infected at different parasite/cell ratios (1:1, 5:1, 10:1), with tissue culture trypomastigote (tct), Y strain (Pereira da Silva and Nussenzweig, 1953). These forms were obtained from infected LLC-MKz cells as described previously (Andrews and Colli, 1982). After 3 h of parasite-cell interaction, non-internalized trypomastigotes were removed along with culture medium. Subsequently, infected cells were re-incubated for 3, 6, 24 and 48 h with fresh medium and then the wells were carefully washed for 5 min with: (a) PBSA (25 mM Na2HPO4, 25 mM KHzPO4 pH 7.2, 150 mM NaCI); (b) PBS (PBSA plus 1 mM Ca 2+ and 0.4 mM Mg2+); or (c) 199 serum-free medium. Infected cells were then fixed by incubating either with 2% paraformaldehyde (freshly prepared in 0.01 M lysine, 0.05 M phosphate buffer pH 7.2) for 20 min, or 0.05% glutaraldehyde in PBSA for 5 min. In situ E L I S A and parasite counting

Before enzymatic reaction, cells were made permeable by incubation with 0.01% saponin in PBSA containing 1% bovine serum albumin for 30 min (Tardieux et al., 1992). After blocking the wells for 30 min with 5% defatted milk (Nestlb) in PBSA, the cells were incubated for 1 h at 37°C with serum from a rabbit immunized with 8-methoxypsoralen (8-MOP)-treated tct, collected on the 360th day after infection (Andrews et al., 1985; Umezawa et al., 1993), The serum employed was diluted at 1:500 and pre-absorbed with 10% fetal calf serum (FCS) for 30 rain at 22°C. After washing the wells three times with 0.05% Tween 20 in PBSA, peroxidase-conjugated goat anti-rabbit IgG (Sigma Chemical Co.), diluted 1:3000 in 5% defatted milk, was added over l h at 37°C. After washing again as described above, o-phenylenediamine (0.4mg/ml) and 0.05% H20/ was added. The solution was then transferred to ELISA microplates, the reaction stopped by addition of 1 M HC1 and read at

303 492 nm in a Titertek Multiskan ELISA reader. The upper chamber wells were then detached and the bottom slide was stained with Giemsa, allowing microscopical counting of both ELISA pre-titered parasites and the number of infected cells. Adequate controls were pertbrmed in three independent reactions, with eightfold replicates in each one. In order to verify applicability of the in situ ELISA, NAcD-glucosamine (Sigma Chemical Co.) was employed as an inhibitor of trypomastigote interiorization: LLC-MKz cells were seeded at the density of 10 × 104 cells/cm2 and incubated with tct (Y strain) at a parasite/cell ratio of 10:1 in culture medium containing increasing sugar concentrations (10, 50, 100 and 200 raM). After 3 h incubation, the medium was removed and both ELISA and parasite counting performed as described above. Data analysis

ELISA optical densities and infectivity index data were analyzed by analysis of mean and ANOVA with Bonferroni tests for individual groups, using Statgraphics ® 5.1 software. Correlation among data was performed by using the Pearson linear test. Differences were considered significant when the probability of equality was less than 0.05 (p<0.05).

Results and discussion

Several conditions were analyzed to optimize the in situ ELISA assay. A cell density of 10 × 10 4 cell/cm2 was shown to be more convenient since it lead to a higher average of internalized parasites, although at lower cell density parasite load per cell was higher (not shown). At the above cell density, parasite/cell ratios of 1:1, 5:1 and 10:1 were assayed, 10:1 being the most efficient (Fig. 1). PBS was also the most suitable washing medium, since microscopical analysis showed that it removed adherent parasites and supported cells in monolayers apparently at the same density as before infection. On the other hand, washes with PBSA and 199 medium without serum promoted cell detachment and preserved some adherent parasites. The best fixative was freshly prepared buffered 2% paraformaldehyde as compared to 0.05% glutaraldehyde, which limited antibody penetration, probably by creating cross-links within cells (Sesso, 1989). We also tested other rabbit anti-T, cruzi hyperimmune sera, which resulted in higher cross-reactivity when assayed with non-infected cells (not shown). The employed MOP serum reacts mainly with osidic and non-osidic tct surface components (Umezawa et al., 1993) and was previously absorbed with 10% FCS. This was done in order to decrease background to significantly lower levels, as described by Luz et al., (1993), who employed normal mice serum for this purpose. The average number of internalized parasites, parasites per infected cell and optical densities in ELISA after 3, 6, 24 and 48 h incubation of T. cruzi with LLCMK2 cells can be seen in Fig. 2 (A, B, C, respectively). Plotted data of these infectivity indexes indicative of parasite interiorization and growth depicted similar shapes, with ascending curves within the time period analyzed. Data of average number of internalized parasites and optical densities (Fig. 2A and C) were nearly overlapped and presented thereby a closer linear correlation (r2= 93.99, p < 0.001), since both represent actually the total amount of parasites available in the system.

304 Optical densibj (492 nm) 0.8 Parasite/cell ratio

FJCelle only F11:1 ~J5:1 ~'-~I0:1 0.6

0,4

0,2

L

3

6

24

Hours Fig. 1. LLC-MK 2 cells (10 x 104/cm2) incubated with tct at several parasite/cell ratios assayed by in situ ELISA (see Material and methods).

B

A

400

C

6r-" 0.9

350 300

~5[

0.8

~4

0.7

~ 3~-

O.6

w

o0 250 0

p. 8o0 N

150 I00

50

L

O.5 0I .

I

I0

..

I

I

20

30

/

40

I

50

Time (hours) Fig. 2. Data of microscopical counting and in situ ELISA readings (O.D.) of T. cruzi internalized in LLC-MK2 cells (10 x 104/cm2) at 3, 6, 24 and 48 h after infection; parasite/cell ratio 10:1. Bars represent 95% confidence intervals. (A) Total number of parasites microscopically counted inside 400 randomly chosen cells (infected plus non-infected) seeded on chamber slide ([]). (B) Parasites per infected cell counted microscopically as in A (0). (C) Optical densities obtained by in situ ELISA performed in the same wells analyzed in A (U]).

305 30

0.6

25

0.5

~20

0.4

¢ 0,3 d~lO

0

i 0

50

100

150

200

0.2

O

0.1

0

N-acetyI-D-glu cosamine (raM)

Fig. 3. (A) Percentageof infectedcells. (B) Total number of parasites inside 100 cells. (C) In situ ELISA reading (O.D.) obtained when the assay was performedas in Fig. 2, except that differentconcentrations of NAc-oglucosaminewere added in culture medium,in a 3-h incubationschedule. Data of parasites per infected cell were less correlated with other indexes and presented some irregular higher counts at early time periods after infection (Fig. 2B). When these parameters were employed to analyze the effect of 10, 50, 100 and 200 mM of N-acetyl-Dglucosamine in T. cruzi interiorization, in situ ELISA data showed corresponding decreasing values, as shown in Fig. 3. Again, the ELISA data and the average number of internalized parasites presented the highest correlation (rZ=88.26, p<0.0001), confirming the efficiency of this ELISA in assessment of interiorized parasites and corroborating the previously described inhibitory effect of the monosaccharide (Andrews and Colli, 1981; Crane and Dvorak, 1982; Piras et al., 1983; Colli et al., 1984). These results show the feasibility and reliability of this in situ ELISA in reflecting intracellular parasite load and its applicability to study inhibition of T. cruzi interiorization, which has been assessed by conventional microscopical technique (Andrews and Colli, 1982; Crane and Dvorak, 1982; Piras et al., 1983; Colli et al., 1984). Although careful standardization of additional reactive antibody probes and conjugates must be performed before employment, the in situ ELISA avoids the subjectiveness of microscopical examination and its restraining effects on experimental design. Besides, it discriminates between parasite loads inside cells after different incubation times, and is shown to be at least as efficient as microscopical analysis for studying T. cruzi internalization and multiplication inside cells. We are currently employing the method to compare infectivity and development of different T. cruzi strains.

Acknowledgments

We are grateful to Dr. Judith K. Kloetzel and Dr. Luciana C.C. Leite for reading the manuscript critically. We thank Carlos Nascimento for expert technical help and

306 A l m i r R . F e r r e i r a f o r scientific d o c u m e n t a t i o n . T h i s i n v e s t i g a t i o n was s u p p o r t e d by Fundag~o Butantan, FAPESP, CNPq and LIMHCFMUSP-49.

References Andrews, N.W. and Colli, W. (1981) lnteriorization of Trypanosoma cruzi in cultured mammalian cells. Inhibition by N-acetyl-Dglucosamine. In: VIII Reuni~o Anual Pesquisa Basica em Doenqa de Chagas, Caxambu, Brasil, p. 81. Andrews, N.W. and Colli, W. (1982) Adhesion and interiorization of Trypanosoma cruzi in mammalian cells. J. Protozool. 29, 264-269. Andrews, N.W., Alves, M.J.M., Schumacher, R.I. and Colli, W. (1985) Trypanosoma cruzi protection in mice immunized with 8-methoxypsoralen-activated trypomastigotes. Exp. Parasitol. 60, 255 262. Brener, Z. (1973) Biology of Trypanosoma cruzi. Ann. Rev. Microbiol. 27, 347 383. Colli, W., Katzin, A.M., Kuwajima, V.Y., Gonqalves, M.F., Abuin, G., Alves, M.J.M. and Zingales, B. (1984) Perspectives of vaccination against Chagas' disease through biotechnology. 1. Definition of relevant antigens possibly involved in parasite-host cell interaction. In: R. Bell and G. Torrigiani (Eds.), New Approaches to Vaccine Development, Schwabe, Basel, pp. 418 441. Crane, M.S.J. and Dvorak, J.A. (1982) Influence of monosaccharides on the infection of vertebrate cells by Trypanosoma cruzi and Toxoplasma gondii. Mol. Biochem. Parasitol. 5, 333 341. De Titto, E.H. and Aratijo, F.G. (1988) Enzyme immunoassay in cell monolayers for evaluation of in vitro activity of chemotherapeutic agents against Trypanosoma cruzi. Am. J. Trop. Med. Hyg. 39, 274 281. De Souza, W. (1984) Cell biology of Trypanosoma cruzi. Int. Rev. Cytol. 86, 197 283. Luz, M.R.M.P., Soeiro, M.N.C. and Arafijo-Jorge, T.C. (1993) Improved enzyme-linked immunosorbent assay (ELISA) for the study of Trypanosoma cruzi-host cell interaction in vitro. Mem. Inst. Oswaldo Cruz 88, 235-241. Merli, A., Canessa, A. and Melioli, G. (1985) Enzyme assay for evaluation of Toxoplasma gondii growth in tissue culture. J. Clin. Microbiol. 21, 88-91. Pereira da Silva, L.H. and Nussenzweig, V. (1953) Sobre uma cepa de Trypanosoma cruzi altamente virulenta para o camundongo branco. Folia Clin. Biol. 20, 191-207. Piras, M.M., Piras, R. and Henriquez, D. (1983) Trypanosoma cruzi fibroblastic cell interactions necessary for cellular invasion. In: Cytopathology of Parasitic Disease (Ciba Foundation Symposium), Pitman Books, London, pp. 31 51. Sesso, A. (1989) Fixa~o de sistemas bioldgicos. In: W. De Souza (Ed.), Manual sobre T6cnicas Bfisicas em Microscopia Eletronica 1, pp. 1-36. Tardieux, I., Webster, P., Ravesloot, J., Boron, W., Lunn, J.A., Heuser, J.E. and Andrews, N.W. (1992) Lysosome recruitment and fusion are early events required for trypanosome invasion of mammalian cells. Cell 71, 1117-1130. Trager, W. (1986) Living together: The biology of animal parasitism. Plenum Press, New York, pp. 1-467. Umezawa, E.S., Stolf, A.M.S. and Zingales, B. (1993) Trypanosoma cruzi: Different surface antigens of trypomastigotes are targets of lytic antibodies. Acta Trop. 54, 41-53.