Human V62' -y6 T-cell tolerance to foreign antigens of Toxoplasma gondii

Little is known about the mechanisms involved in human y6 T-cell tolerance to self or to foreign antigens. Patients with congenital toxoplasmosis offer a unique opportunity to examine V62+ yS T-cell tolerance. Analysis of y6 T cells in patients with congenital toxoplasmosis revealed evidence for anergy of these cells with or without clonal V62+ y5 T-cell expansion in the acute phase of the Toxoplasma infection. T cells in general were unresponsive and did not proliferate upon exposure to mitogens or to Toxoplasma lysate antigens or in response to live Toxoplasma-infected cells when the congenitally infected infants were 1 month of age, and they exhibited selective anergy to Toxoplasma lysate antigens and live Toxoplasma-infected cells when the infants were aged 5 months. During the chronic phase of congenital toxoplasmosis in the patients who were more than 1 year of age, the repertoires of the yS T-cell receptors were found to be within normal ranges. In addition, in Cambridge, MA. FITC-conjugated Ti-yA (anti-Vy9), phycoerythrin (PE)-conjugated anti-HLA-DR (I2), and PE-conjugated anti-CD56 (NKH1) mAbs were from Coulter. FITC- and PE-conjugated anti-CD3 (NU-T3), FITC-conjugated anti-HLA-DR (NU-Ia), and FITC-conjugated anti-CD45RO (UCHL1) were from Nichirei Company, Tokyo. FITC-conjugated anti-aj3 TCR (WT-31), PE-conjugated anti-CD4 (Leu3), anti-CD8 (Leu2), and anti-CD45RO (Leu45RO) mAbs were from Becton Dickinson.

the Toxoplasma infection. T cells in general were unresponsive and did not proliferate upon exposure to mitogens or to Toxoplasma lysate antigens or in response to live Toxoplasmainfected cells when the congenitally infected infants were 1 month of age, and they exhibited selective anergy to Toxoplasma lysate antigens and live Toxoplasma-infected cells when the infants were aged 5 months. During the chronic phase of congenital toxoplasmosis in the patients who were more than 1 year of age, the repertoires of the yS T-cell receptors were found to be within normal ranges. In addition, in the chronic phase, the yS T cells proliferated and secreted -y-interferon in response to exposure to live Toxoplasma-infected cells. By contrast, af3 T cells remained anergic. V62+ yS T cells have been considered to undergo extrathymic maturation and thus to be subject to development of peripheral tolerance. Our findings indicate that V62+ yS T-cell tolerance was lost in these infected infants earlier than 43] T-cell tolerance. These findings suggest that 'yS T cells play a role in protection against Toxoplasma gondii in the chronic phase when congenitally infected children are more than 1 year of age, especially in those in whom c38 T cells continue to exhibit deficits in specific immune responses to Toxoplasma antigens.
af3 T-cell tolerance to self and foreign antigens has been extensively studied in mice (1). Clonal deletion or anergy of reactive murine af3 T cells has been reported when self or foreign antigens are introduced into the thymus at an early stage of T-cell development (2). Only a few reports have appeared concerning yS T-cell tolerance toward self antigens in mice. Elimination and immunologic inactivation have recently been reported for murine self-reactive transgenic -ye T cells during intrathymic maturation (3,4). Functional anergy after activation has been demonstrated in murine self-reactive transgenic yS T cells during extrathymic maturation (5,6).
In humans, few data are available on ye T-cell tolerance to self or foreign antigens. Congenital infections in humans often result in long-term T-cell unresponsiveness to the pathogens (7). However, the precise mechanisms of both af3 and -yS T-cell tolerance in the course of congenital infections remain to be clarified. Among the pathogens capable of causing congenital infections in humans, Toxoplasma gondii is of special interest. In patients with acquired T. gondii infection, expansion of cells bearing a particular V region, V62, of 'y8 T-cell receptor (TCR) has been observed (8,9). Therefore, the repertoire and function of y6 T cells were analyzed, and the mechanism of V82+ yeS T-cell tolerance to foreign antigens of T. gondii was examined in patients with congenital toxoplasmosis.

MATERIAL AND METHODS
Patients. The diagnosis of congenital toxoplasmosis was made in infants who expressed characteristic clinical features including chorioretinitis and central nervous system abnormalities (cerebral calcification and/or hydrocephalus) as well as characteristic serological evidence of congenital Toxoplasma infection and who did not have any evidence of cytomegalovirus infection. By serological studies, patients 1-6 had Toxoplasma-specific IgG and IgM antibodies, even in early infancy. Patient 7 had persistent Toxoplasma-specific IgG antibody and exhibited serologic evidence of intrauterine Toxoplasma infection (at 12 weeks of gestation with a titer of <1/128 and at 38 weeks a titer of 1/2048). In patients 1 and 2, T. gondii DNA was found in the cerebrospinal fluid by polymerase chain reaction (A. Yano, Nagasaki University, Nagasaki, Japan; personal communication). Patients 1-3 with acute phase of congenital toxoplasmosis were studied at the ages of 2 weeks, 1 month, and 3 weeks, respectively. Patients with chronic phase of congenital toxoplasmosis were examined at the age of 1 year (patients 4 and 5) and 6 years (patients 6 and 7). Ten seronegative, age-matched children aged 1 month to 7 years, six seronegative adults, two seropositive children with no evidence of congenital infection, and four seropositive mothers of the patients served as controls. Blood was obtained from the patients and controls after obtaining informed consent for our study.
Flow Cytometric Analysis. Cells were stained with an appropriate mAb at 4°C for 30 min. After washing, single-color or two-color flow cytometric analysis was done using a FAC- Scan (Becton Dickinson Immunocytometry Systems) as described (12). Preparation of Live T. gondii and Toxoplasma Lysate Antigens (TLAs). The RH strain of T. gondii was passaged in 8to 10-week-old female ddY mice (Shimizu Laboratory Supplies, Kyoto, Japan). Free tachyzoites were recovered from the peritoneal cavity after 5 ml of PBS was instilled. Isolated tachyzoites were passed through a 27-gauge needle and then a 3-,um filter (Costar) to remove murine macrophages, and they were then centrifuged at 200 x g for 10 min according to the method described (13). The pellet containing free tachyzoites was washed and resuspended in RPMI 1640 medium with 10% FBS and was used as a live T. gondii preparation. TLA was prepared as described (7). Its protein concentration was determined by the Bradford method (Bio-Rad).
In Vitro Infection with T. gondii. Phytohemagglutinin (PHA)-activated blasts (PHA blasts) were obtained by incubating PBMC and 5 ,ug of PHA-P (Difco) per ml of RPMI 1640 medium with 10% FBS for 4 days. Autologous PBMC or PHA blasts were infected with RH strain tachyzoites at a multiplicity of 2:1 to 5:1 for 2 h in RPMI 1640 medium with 10% FBS. Extracellular tachyzoites were removed by Ficoll/Hypaque density gradient centrifugation in experiments with PBMC. Extracellular tachyzoites and a few contaminating murine macrophages were removed by centrifugation with a continuous gradient of Percoll (Pharmacia) in experiments with PHA blasts (14). After irradiation with 12,000 rad (1 rad = 0.01 Gy; ref. 15), T. gondii-infected PBMC or PHA blasts were used as stimulators.
In Vitro Stimulation of T Cells. PBMC or separated T cells were cultured at 2 x 105 cells per ml in RPMI 1640 medium with 10% FBS/1 mM glutamine in the presence or absence of 1 ,ug of mitogenic CD3 mAb per ml that was prepared in our laboratory (10), 5 gg of TLA per ml, and 4 x 104 cells per ml of irradiated Toxoplasma-infected PBMC or PHA blasts for the indicated days. In some experiments, recombinant human interleukin-2 from Shionogi Pharmaceutical Company, Osaka, Japan, was added at a final concentration of 100 units/ml. For proliferation assays, the cells were pulsed with [3H]thymidine during the last 12 h and harvested onto glass fiber filters. Incorporated radioactivity was determined by liquid scintillation counting as described (16). y-Interferon Assay. Supernatants of PBMC or separated T cells cultured in the presence or absence of a stimulator were obtained 2 days following initiation of culture. -y-Interferon .:A
Functional studies in vitro showed that the patients' T cells were not responsive to stimulation with a T-cell mitogen, CD3 mAb, TLA, or irradiated, live Toxoplasma-infected cells when obtained from the infected infants at 1 month of age, and they became weakly reactive to CD3 mAb (patients versus controls; P < 0.1) and remained nonreactive to irradiated, live Toxoplasma-infected cells or TLAs at 5 months of age (Fig. 4).
Functional Recovery of y6 T cells in Chronic Phase of Congenital Toxoplasmosis. In four patients (patients 4-7) with congenital toxoplasmosis studied at ages 1-6 years, no increases in HLA-DR+CD3+ cells or CD45RO+CD3+ cells were observed compared with those of age-matched controls (mean ± 2 SD; percent HLA-DR+CD3+ cells, 2.1% ± 1.7%; , and four patients (P) were cultured and assayed in the same way as described in Fig. 4. percent CD45RO+CD3+ cells, 14.8% + 7.8%; n = 10). y8 T cells showed neither deletion nor expansion, but they showed age-dependent increases as compared to those of age-matched controls (4.3% 4.1%) as shown in Table 1.
Functional studies in vitro showed that the four patients with congenital toxoplasmosis exhibited normal T-cell responses to CD3 mAb (patients versus seropositives or seronegatives, P > 0.1), a slightly weakened response to irradiated, live Toxoplasma-infected cells (patients versus seronegatives, P > 0.1; patients versus seropositives P < 0.01), and no response to TLA (Fig. 5). Seronegative controls showed proliferative responses to irradiated, live Toxoplasma-infected cells but not to TLAs, whereas seropositive controls had responses to TLAs as well as to irradiated, live Toxoplasma-infected cells. To analyze which cells responded to respective antigens, flow cytometric analysis was performed (Fig. 6). In patients with congenital toxoplasmosis and seronegative controls, y8 T cells were selectively activated and expanded in response to live Toxoplasma-infected PHA blasts, and no significant responses of a/3 T cells to TLAs or live Toxoplasma-infected cells were observed. In seropositive controls, a43 T cells were responsive to TLAs and live Toxoplasma-infected cells, and y6 T cells were responsive to live Toxoplasma-infected cells.
To assess protective T-cell functions against T. gondii, the supernatants stimulated by TLAs or live Toxoplasma-infected , and four patients with congenital toxoplasmosis (P) were cultured in the absence (NIL) or presence of TLA or irradiated, live Toxoplasma-infected PHA blasts (LT) for 8 days. Two-color flow cytometric analysis of the cultured PBMC was performed with FITC-conjugated anti-pan a43 or y8 TCR mAb and PE-conjugated HLA-DR mAb. Solid areas show percentages of HLA-DR+ a43 or -y5 T cells. For experiment 2. After 6-day culture in the presence or absence of a stimulator, PBMC were treated with control or -y8 TCR mAb for y6 T-cell depletion. Then, control mAb-depleted or y6 TCR mAbdepleted PBMC were cultured for an additional day in the presence of recombinant human interleukin-2. Supernatants were assayed for y-interferon. y-Interferon data is expressed as mean ± SD. cells were assayed for y-interferon production. As shown in Table 2, the patients' T cells produced significant amounts of y-interferon in the presence of irradiated, live Toxoplasmainfected cells (patients versus seronegatives, P > 0.1; patients versus seropositives, P < 0.01) but not in the presence of TLAs. The fact that depletion of y6 T cells markedly decreased -y-interferon production indicated that y6 T cells were responsive for most -y-interferon production in the presence of live Toxoplasma-infected cells.

DISCUSSION
Human y8 T cells comprise two major subsets, V61 and V62. The V81 subset predominates in the thymus, whereas the V82 subset becomes predominant in the periphery and increases with age (17). The V82 subset appears to undergo extrathymic maturation and to be implicated in defenses against various microorganisms (17,18). Patients with congenital Toxoplasma infection offer the unique possibility to examine V82' y6 T-cell tolerance because T. gondii induces expansion of a particular V region, V82, of y6 TCR in acquired infection (8,9). To study the mechanisms of V62' -y6 T-cell tolerance in congenital Toxoplasma infection, an in vitro system to evaluate specific y6 T-cell function was established. In response to irradiated (12,000 rad), live Toxoplasma-infected cells, but not in response to TLA, V62+ -y6 T cells predominantly proliferated and secreted ,y-interferon. Irradiation with 3000 rad was not sufficient to kill nonattenuated T. gondii. Subauste et al. (19) (20), it was suggested that human y8 T cells might play a role in protection against T. gondii. In the mouse, y6 T cells have been reported to play an important role in protection against Toxoplasma infection (21). In vivo and in vitro studies of y6 T cells in patients with congenital toxoplasmosis revealed by the analyses reported herein have shown that V62+ y8 T cells became anergic with or without clonal expansion during the acute phase of congenital infection. V62 clonal expansion was not associated with -yS TCR down-regulation (Fig. 3), as was observed in an animal study (22). No deletion of V82+ y6 T cells was observed. Patients 2 and 3 showed total T-cell anergy at 1 month of age and Toxoplasma-specific anergy at 5 months of age. T-cell unresponsiveness to Toxoplasma antigens was observed during the acute phase of acquired Toxoplasma infection as well (23,24). It is likely that V82 clonal expansion followed by unresponsiveness might be a more frequent phenomenon in acute phase than no clonal expansion. This concept is attractive because all four patients with congenital toxoplasmosis in a recent report showed increased CD45RO+ (memory) T cells in the cord blood (25), and most of the CD45RO+ T cells were y6 T cells in our series of patients. The present study also revealed that, in addition to the findings by Michie and Harvey (25), CD45RO+ T cells were not always elevated in congenital toxoplasmosis, especially when the disease was very severe as in our patient 1. Clonal expansion of reactive T cells preceding anergy has been documented only in extrathymic peripheral tolerance in mice (22,26,27). Indeed, the V82 subset of yS T cells is considered to undergo extrathymic maturation and to be subject to development of peripheral tolerance rather than central tolerance in the thymus (17).
Long-term (5-19 years) reduction of T-cell response in vitro has been demonstrated in congenital infections as well as in prenatal exposure to foreign antigens (7,28). McLeod et al. (7) reported that infants with severe congenital toxoplasmosis tended to show low T-cell responsiveness to TLAs. Because it is difficult to investigate differences between a/3 T-cell and -y8 T-cell tolerance in mild patients with early recovery of T-cell responsiveness to Toxoplasma antigens, we elected to investigate only severe patients with persistent T-cell unresponsiveness to TLA. Despite persistent afp T-cell unresponsiveness, y6 T cells became reactive to live Toxoplasma-infected cells and produced y-interferon after the infants with congenital toxoplasmosis reached 1 year of age. In severe patients, aG T cells may be subject to development of a central tolerant state, whereas V82+ y6 T cells are subject to development of peripheral tolerance. Peripheral tolerance can be reversed when antigen is absent or upon exposure to certain other infectious agents (27,29,30). After 1 year, it is likely that Toxoplasma antigens have been removed from direct contact with y8 T cells. Further, infectious agents that induce V62 expansion may contribute to loss of V82 tolerance in congenital toxoplasmosis. Because the treatment of congenital toxoplasmosis is usually discontinued after the children reach 1 year of age without evidence of relapse, the reversal of peripheral tolerance of y6T cells may contribute to protection against spread of Toxoplasma after 1 year of age in severe patients in which a3 T-cell unresponsiveness to TLA persists.
To our knowledge, the present study has demonstrated for the first time that human V82+ y6T cells may become anergic with or without clonal expansion during the acute phase of congenital toxoplasmosis. In the chronic phase in infants more than 1 year old, V62+ y8 T-cell tolerance that occurs during extrathymic maturation is lost earlier than ca3 T-cell tolerance in severe patients. Additional analyses of the precise mechanism or mechanisms involved in T-cell tolerance in congenital infections should contribute to better understanding of the pathophysiology of persistent infections as well as better understanding of late complications that may be mediated by immune mechanisms.
We are grateful to Drs. Rumi Yamakawa, Kazue linuma, and Munehiro Yamasaki for providing us with samples from patients. We thank Ms. Tazim Verjee for manuscript preparation. This work was supported in part by grants from the Ministry of Health and Welfare of Japan.