INTRODUCTION

Top Abstract Introduction Materials & Methods Results Discussion References

Toxoplasma gondii is an obligate intracellular protozoan parasite that infects all warm-blooded animals, including humans. After the primary infection in an immunocompetent individual, the immune response of the host limits the replication of tachyzoites, resulting in the formation of the bradyzoite form, a dormant stage of the parasite. Because of this effective immune reaction against parasites, chronic infection of an immunocompetent individual with T. gondii is usually asymptomatic (3). However, in patients with AIDS as well as other immunocompromised states, reactivation of chronic toxoplasmosis results in excessive cellular destruction, often leading to severe morbidity and mortality (10).

Protective immune responses against infection with T. gondii have been studied with experimental murine systems. Vaccination with an attenuated mutant (22, 24) or irradiated tachyzoites (18, 25) induces protective responses against subsequent infection with the virulent RH strain. CD8⁺ T-cell-mediated immunity is one of the major protective mechanisms. Mice primed with T. gondii succumb to lethal infection when depleted of CD8⁺ T cells prior to challenge infection, and host resistance can be adoptively transferred to naive animals by primed CD8⁺ T cells (22, 25) or by CD8⁺ T-cell clones specific for T. gondii major surface antigen (SAG-1) (8). However, the underlying mechanisms of CD8⁺ cell-mediated protective immunity are not completely understood. The protection depends on the presence of gamma interferon, which can be secreted by CD8⁺ as well as CD4⁺ cells (8, 23). T. gondii-specific CD8⁺ T cells also lyse parasite-infected target cells in a class I major histocompatibility complex-restricted manner (7, 11, 14, 21, 26). Lysis of host cells may directly damage intracellular microbes or simply release viable parasites into the extracellular space. Apoptosis of infected host cells induced either chemically (13) or by CD8⁺ cytotoxic T lymphocytes (CTL) (20) appears to be coupled to damage to intracellular mycobacteria. Little is known, however, about the fate of T. gondii parasites in host cells lysed by CD8⁺ CTL.

The present study was therefore designed to determine whether CD8⁺ CTL-mediated lysis of host cells is associated with the killing of intracellular T. gondii tachyzoites. We developed a novel method to determine the number of live T. gondii parasites within host cells undergoing apoptosis. We used a combination of DNase treatment of the samples prior to DNA extraction, which removed DNA within dead parasites, and quantitative competitive PCR (QC-PCR) of the T. gondii SAG-1 gene (12). The results indicated that CD8⁺ CTL-mediated apoptosis of target cells does not lead to the death of intracellular T. gondii tachyzoites.

	MATERIALS AND METHODS

Top Abstract Introduction Materials & Methods Results Discussion References

Animals, parasites, and cell lines. BALB/cAnNCrj and C57BL/6NCrj mice and Lewis/Crj rats were purchased from Charles River Japan (Kanagawa, Japan). Animals were housed in the Laboratory Animal Center for Biomedical Research at the Nagasaki University School of Medicine (Nagasaki, Japan) and were used at 8 to 10 weeks of age. T. gondii RH was maintained as previously described (16, 26). M12-neo-1 cells were generated by stable transfection of M12.4.1 cells (a gift from L. Glimcher, Harvard Medical School, Boston, Mass.) (9) with linearized Rc/CMV (Invitrogen, Carlsbad, Calif.) by use of a Gene Pulser (Bio-Rad, Hercules, Calif.). Transfectants were selected in culture medium containing G418 (0.5 mg/ml) (Gibco BRL, Grand Island, N.Y.) and cloned by a limiting-dilution method.

DNase treatment of killed T. gondii and DNA preparation. Complement-mediated killing of free tachyzoites (10⁵) was performed at 37°C for 30 min with rat anti-T. gondii serum prepared from a Lewis rat after priming with X-ray-irradiated (120 Gy) RH tachyzoites and with rabbit complement. The trypan blue dye exclusion test indicated that 100% of the parasites treated with antibody plus complement were dead after the treatment (data not shown). Cells were incubated in phosphate-buffered saline (PBS) containing DNase (50 µg/ml) for 30 min at 37°C. After the cells were washed once with PBS, genomic DNA was prepared with DNAzol (Gibco BRL) in accordance with the manufacturer's instructions. Glycogen (10 µg) (Boehringer GmbH, Mannheim, Germany) was included in the DNAzol solution as a carrier.

Preparation of DNA from T. gondii-infected target cells after CD8⁺ T-cell-mediated cytolysis. M12-neo-1 cells were infected with tachyzoites at a multiplicity of infection of 3:1 for 2 h at 37°C. Free tachyzoites were removed from infected cells by two rounds of low-speed centrifugation (70 × g for 10 min) followed by use of a magnetic cell separation system (Miltenyi Biotec, Bergisch Gladbach, Germany). The infected cells were incubated with rat anti-T. gondii serum at 4°C for 10 min. After being washed with PBS, the cells were incubated with microbeads coated with mouse anti-rat immunoglobulins (20 µl of magnetic probes per 10⁶ cells) in 1 ml of PBS containing 0.5% bovine serum albumin and 0.1 mM EDTA at 8°C for 20 min. Free tachyzoites were subsequently removed in a magnet field of 0.6 T with an MS column (Miltenyi Biotec) in which the flowthrough fraction was collected. The number of contaminating free tachyzoites was less than 2% of M12-neo-1 cells, as determined by light microscopy. The proportion of the cells infected with T. gondii was 40 to 50%. After coculturing of infected M12-neo-1 cells (10⁴) with CTL (10⁵), samples were treated with DNase (50 µg/ml) for 30 min at 37°C and DNA was extracted. In some experiments (see Fig. 3), T. gondii-infected M12-neo-1 cells were passed through a 30-gauge needle several times to force the intracellular parasites out of the parasitophorous vacuoles as described previously (4). The materials were treated or not treated with anti-T. gondii serum and complement before DNase treatment.

QC-PCR of SAG-1 and Neo genes. The number of parasites was determined by QC-PCR of the SAG-1 gene as previously described (12). Briefly, genomic DNA (<1 µg) extracted from T. gondii or infected cells was coamplified with a constant amount of competitor DNA (i.e., approximately 90 copies of the truncated SAG-1 gene) by use of a set of SAG-1-specific primers for 36 cycles in a final volume of 50 µl in a TSR-300 thermal sequencer (IWAKI Glass Co. Ltd., Chiba, Japan). The amplified products were separated by agarose gel electrophoresis and stained with ethidium bromide. The ratios of the staining intensities of the amplified target and competitor sequences were determined by densitometry (IPLab Gel densitometer; Signal Analytical Corp., Vienna, Va.). By comparing the ratio thus obtained to a standard curve, the copy number of SAG-1 DNA was calculated. The detection limit was seven copies of SAG-1 DNA per sample. Because SAG-1 is a single-copy gene, the copy number of SAG-1 DNA is equal to the number of T. gondii tachyzoites. The results were expressed as the mean parasite number per template DNA, corresponding to 100 input tachyzoites or 300 infected M12-neo-1 cells ± 1 standard deviation (SD). For statistical comparisons of two groups, an unpaired two-tailed Student's t test was used. A P value of <0.05 was considered significant.

3

	RESULTS

Top Abstract Introduction Materials & Methods Results Discussion References

Estimation of the number of live T. gondii organisms by QC-PCR. The number of intracellular T. gondii organisms can be quantitated by plotting the ratios of the sequence of the QC-PCR product of T. gondii SAG-1 and its competitor sequences against a standard curve (12). Initially, we examined whether this method can selectively amplify DNA of viable parasites. DNA extracted from free viable tachyzoites and those killed by anti-T. gondii serum plus complement was used to determine the number of T. gondii organisms by QC-PCR (Fig. 1 and Table 1). The ratios of the amplified target and competitor sequences did not differ significantly (Fig. 1, lanes 1 and 3; Table 1; P = 0.207), indicating that the DNA extracted from complement-killed T. gondii can become a template for PCR, like that extracted from viable T. gondii. Therefore, we modified this method by introducing DNase treatment to exclude DNA from dead parasites and determined whether the number of live parasites can be estimated by QC-PCR. Free T. gondii was treated with DNase to remove the DNA template within killed T. gondii, assuming that DNase can gain access to the DNA within dead but not viable cells. This pretreatment reduced the amount of the PCR product from dead parasites more than 100-fold (Fig. 1, lanes 3 and 4; Table 1). This reduction was selective for dead parasites, because the same treatment did not significantly reduce the amount of the PCR product from live parasites (Fig. 1, lane 2; Table 1, P = 0.268). To determine that this method can be applied to a mixture of live and dead T. gondii, a sample containing a 1:1 mixture of live and dead parasites was treated with DNase, and the number of viable T. gondii parasites was quantitated by QC-PCR. As expected, the number of SAG-1 gene copies obtained from the DNase-treated samples (88 ± 2) was approximately half that in the untreated samples (163 ± 16) (Fig. 1, lanes 5 and 6; Table 1). In a separate experiment, we confirmed that the gene copy numbers estimated by this method had a linear relationship with the ratios of viable parasites within the mixture of dead and live T. gondii (data not shown). These results indicated that DNase treatment effectively eliminates the SAG-1 gene within dead T. gondii and that this method can be applied to the quantitation of live parasites in the presence of dead T. gondii and host tissue.

View larger version (17K): [in this window] [in a new window]   FIG. 1.   QC-PCR of DNA from cells treated with DNase. Free T. gondii organisms were untreated (lanes 1 and 2) or treated with anti-T. gondii serum and complement (lanes 3 and 4), or both untreated and treated T. gondii organisms were mixed at 1:1 (lanes 5 and 6). Cells were untreated (lanes 1, 3, and 5) or treated (lanes 2, 4, and 6) with DNase. DNA was extracted and amplified in the presence of competitor DNA with SAG-1 primers. Each lane represents QC-PCR of the DNA template corresponding to 100 tachyzoites. M, molecular size marker (X174 HaeIII digest); sizes (in bases) are shown at left. Experiments were performed in triplicate. Similar results were obtained in four independent experiments.

Viability of T. gondii in infected CTL target cells. We applied this method to determine the fate of intracellular tachyzoites after the lysis of host cells by CD8⁺ CTL. An alloreactive CTL line specific for H-2^d was generated by repeated stimulation of C57BL/6 lymphocytes with BALB/c spleen cells. The stable cell line M12-neo-1 was used as a target. The use of a neo transfectant cell line allowed quantitation of the number of viable target cells in the presence of CTL. M12-neo-1 cells were infected with T. gondii in vitro. After the removal of free T. gondii, the infected target cells were incubated with H-2^d-specific CTL for 0 to 12 h. Cells were treated with DNase, and DNA was extracted and subjected to QC-PCR to determine the number of live T. gondii parasites within the target cells (Fig. 2). During the initial 8 h of coculturing, the number of tachyzoites within the target cells did not change significantly. After 8 h, the number increased, perhaps due to the DNA synthesis of T. gondii within the infected cells. Interestingly, this increase was observed even when the target cells were lysed by CTL. We speculate that some of the parasites within the apoptotic target cells infected and multiplied within effector CTL. Indeed, light microscopic inspection of the cells after 12 h of culturing revealed that approximately 8% of the effector T cells were infected by tachyzoites (data not shown). In the same experiment, ⁵¹Cr release by the target cells reached 84% during 6 h of culturing (Fig. 2). These results indicate that CTL-mediated lysis of cells infected with tachyzoites does not lead to the death of the intracellular parasites.

View larger version (25K): [in this window] [in a new window]   FIG. 2.   Kinetics of the number of viable tachyzoites after host cells were lysed by CTL. Ten thousand M12-neo-1 cells were infected with RH tachyzoites (infection rate, 45%) and cultured in the presence (hatched bars) or absence (open bars) of CTL at an effector/target cell ratio of 10:1. The reaction was terminated at the times indicated, and the number of viable tachyzoites was determined by QC-PCR. Results are shown as the mean parasite number per 300 M12-neo-1 cells initially placed in the culture ± 1 SD (four samples/group). In the same experiment, a portion of the infected M12-neo-1 cells was 51Cr labeled, and CTL activity was assessed by the 51Cr release assay. The percentages of specific 51Cr release by the target cells during 4 and 6 h of culturing were 76 and 84%, respectively. Similar results were obtained in two independent experiments.

View larger version (28K): [in this window] [in a new window]   FIG. 3.   Resistance of the intracellular T. gondii to DNase treatment after CTL-mediated lysis of the host cells. M12-neo-1 cells (104) infected with tachyzoites for 15 h (infection rate, 40%) were cultured in the presence of an H-2d-specific CTL line at an effector/target cell ratio of 10:1 for 4 h. The samples were disrupted by repeated passage through a 30-gauge needle. One of the disrupted samples was treated with rat anti-T. gondii serum (Ab) and complement (C). Finally, the samples were treated or not treated with DNase, and DNA was extracted. QC-PCR of each DNA sample was performed, and the SAG-1 gene copy number was estimated as described in Materials and Methods. Mean data from one of two similar independent duplicate experiments are shown.

	DISCUSSION

Top Abstract Introduction Materials & Methods Results Discussion References

A novel method to selectively quantitate the number of live T. gondii parasites in a mixture of dead T. gondii and host tissue was developed. The quantitation of live T. gondii parasites was previously performed by determining their ability to grow intracellularly or lyse host cells by a ³H-uracil incorporation assay or a plaque assay, respectively (15). However, only 20 to 30% of the free tachyzoites lysed from hosts are infective for other cells, and these methods are difficult to apply in quantifying viable tachyzoites within apoptotic host cells, because individual tachyzoites cannot be segregated from the apoptotic cell, which tends to clump. Therefore, we used membrane permeability as a measure of cell viability and overcame these difficulties by a combination of DNase treatment and QC-PCR, which allowed simple and rapid quantitation of viable tachyzoites in the cell mixture. This method was applied to determine whether CD8⁺ CTL can kill intracellular tachyzoites when they lyse target cells. The results indicated that CTL specific for T. gondii are unable to kill intracellular T. gondii tachyzoites.

The induction of apoptosis in target cells by CD8⁺ CTL is mediated by perforin and granzyme B or the engagement of Fas on the target cells (7, 11). In our assay system, it is likely that the lysis of T. gondii-infected target cells was mediated by perforin and granzyme B, because the CTL used in our assay system are conventional CD8⁺ CTL and because M12-neo-1 target cells do not express Fas (reference 5 and unpublished data). Thus, we believe that the lysis of T. gondii-infected cells through the release of cytotoxic granules by CTL does not lead to the death of intracellular parasites. Indeed, perforin does not appear obligatory for protection against T. gondii infection, because T. gondii-primed perforin knockout mice retain resistance to challenge infection with the parasite (2). It is formally possible that Fas-mediated lysis of host cells has distinct effects on intracellular T. gondii. We think it is unlikely, however, because the induction of apoptosis in Fas⁺ A20 cells infected with T. gondii by anti-Fas monoclonal antibodies did not kill intracellular parasites (data not shown). Perforin- or granzyme-mediated lysis of infected macrophages by CTL has been shown to result in the death of intracellular Mycobacterium tuberculosis, whereas Fas-mediated lysis has not (20). The discrepancy between our study and theirs may be due to the differences in the pathogens used (T. gondii versus M. tuberculosis) and in the host species (mouse versus human). Alternatively, differences in host cells may explain this difference. We used a B-cell tumor which does not have the phagocytic ability of apoptotic cells, while they used macrophages as infected targets. Therefore, the possibility that the death of the intracellular bacteria was due to the phagocytosis of apoptotic macrophages by neighboring macrophages was not completely ruled out in their study. It is also possible that a subset of CD8⁺ CTL can kill intracellular T. gondii.

We previously demonstrated that protective immunity against a virulent strain of T. gondii can be transferred to naive animals by adoptive transfer of primed CD8⁺ cells (25). If CD8⁺ CTL are not themselves cytotoxic for intracellular T. gondii, how are tachyzoites cleared in vivo? The lysis of host cells by CTL may release tachyzoites from their sequestered environment into the extracellular space, where other effector molecules and cells are accessible. These include antibody, complement, NK cells, a population of CD8⁺ T cells which are directly parasitocidal (8), and macrophages (6). It is unclear which molecules and cells are the most critical for clearing tachyzoites after host cell lysis. Alternatively, tachyzoites may be cleared even prior to their release into the extracellular fluid after the induction of host cell apoptosis. Induction of apoptosis by CD8⁺ T lymphocytes may result in the phagocytosis of apoptotic cells by macrophages (17), resulting in enzymatic digestion of the parasites in phagolysosomes. Thus, one of the roles of CTL may be to prepare infected cells for engulfment by macrophages. In our CTL assay, only target cells and CTL were mixed in vitro, and intracellular T. gondii survived the CTL-mediated lysis of the host cells. We believe that it is likely that macrophages are involved in the clearance of T. gondii in vivo. Our preliminary study indeed suggests that this mechanism may be operative.