Lethal host-versus-graft disease and hypereosinophilia in the absence of MHC I–T-cell interactions

Neonatal injection of semiallogeneic spleen cells in BALB/c mice induces a self-limited state of chimerism that promotes the differentiation of donor-specific CD4 T cells toward the Th2 phenotype. Here we show that injection of spleen cells from β2-microglobulin–deficient (BALB/c × C57BL/6) F1 mice into BALB/c newborns with a disrupted β2-microglobulin (β2m) gene results in a lethal lymphoproliferative disorder associated with uncontrolled Th2 response, long-term persistence of donor B cells, and sustained blood eosinophilia. Autoimmune manifestations are also enhanced and characterized by a severe autoantibody-mediated glomerulonephritis. Histological examination of the spleen shows a hyperplasia of periarteriolar lymphoid sheaths, with accumulation of eosinophils and basophils, and variable degree of fibrosis. Perivascular lymphoid infiltrates with eosinophils are also found in the lung and are correlated with disease severity. Such abnormalities are almost absent using β2m-sufficient mice. These data demonstrate that induction of lymphoid chimerism in the absence of MHC class I–T-cell interactions results in a lethal form of host-versus-graft disease that represents a unique model of Th2-dependent chronic inflammatory disease associated with an hypereosinophilic syndrome in mice.

Since the original experiments by Billingham, Brent, and Medawar (1), the model of neonatal tolerance to transplantation antigens has been extensively studied to analyze the different parameters promoting allograft acceptance. Although it has been proposed that in mice neonatally injected with semiallogeneic spleen cells, MHC class I–reactive cytotoxic T cells are tolerized by clonal deletion (2, 3), it is clear that alloantigen-specific, IL-4-secreting CD4 T cells are still present in the periphery (4–7) and are probably responsible for the upregulation of MHC class II molecules on both donor and host B cells (8). In addition, it has been shown that the differentiation of donor-specific CD4 T cells toward the Th2 phenotype is dependent on the early production of IL-4 in vivo (5, 9). Furthermore, blocking endogenous IL-4 was found to restore antidonor cytoxic T lymphocytes (CTL) activity and abrogated the lymphoid chimerism, thereby resulting in graft rejection (5, 10). Another important feature of this model is the development of autoimmune manifestations, known as host-versus-graft (HVG) disease, due to the polyclonal T–B-cell interactions that occur in this model between host Th2 and donor B cells, resulting in a dramatic increase of serum IgE level and in the production of autoantibodies, some of which are found deposited in the kidneys (11, 12). However, such serological abnormalities are spontaneously downregulated by 10 weeks of age due to the disappearance of lymphoid chimerism (11, 13).

Because IgE production and Th2 priming are critically dependent on the early production of IL-4 (14–16), we have recently analyzed the requirement for NK1.1 T cells in the generation of alloreactive Th2 responses in this model (17). For this purpose, lymphoid chimerism was induced in both normal and β2-microglobulin–deficient (β2m°) BALB/c mice. We showed that similar Th2 priming and IL-4–dependent IgE production developed in both β2m° or normal BALB/c mice injected at birth with semiallogeneic spleen cells from normal (BALB/c × C57BL/6) F1 mice (CB6F1), indicating that β2m-dependent NK1.1⁺ host T cells were not required for the generation of alloreactive Th2 responses in vivo (17). Interestingly, when β2m° CB6F1 spleen cells were injected into BALB/c β2m° mice, the IgE response was even exacerbated and associated with an increased splenomegaly (17). Thus, we hypothesized that MHC class I–T-cell interactions could play a crucial role in this model by regulating Th2 immunity and its associated immunological disorders.

To analyze the role of MHC class I–reactive host T cells interacting with donor cells, we have compared 2 combinations, in which wild-type (WT) or β2m° newborn BALB/c mice were injected either with normal or with β2m° CB6F1 splenocytes, respectively. Data presented here demonstrate that the absence of MHC class I–T-cell interactions leads to spared Th2 immunity that results in accelerated mortality and higher incidence of proteinuria due to antibody-mediated glomerular lesions. The uncontrolled Th2 response of host CD4 T cells is associated with the maintenance of B-cell lymphoid chimerism and with sustained hypereosinophilia in adult mice. Thus, our data support the hypothesis that CD8 T cells might play a more active role than previously anticipated by regulating donor B-cell persistence and/or Th2-cell development during neonatal induction of lymphoid chimerism in mice, thereby preventing chronic Th2 responses and its associated hypereosinophilic syndrome.

Effects of β2m deficiency in host versus graft disease. Injection of semiallogeneic spleen cells in newborn BALB/c mice stimulates host-derived alloreactive Th2 cells that react with MHC class II alloantigens on donor B cells. This cognate interaction induces a polyclonal B-cell activation characterized by increased levels of serum IgE and autoantibody production (11, 12). As shown in Figure 1, no difference in serum levels could be observed at 4 weeks of age between the WT (Figure 1a) and the β2m° (Figure 1b) combinations. In striking contrast, whereas serum IgE levels were spontaneously downregulated by 8 weeks of age in the CB6F1-injected WT mice (Figure 1a), induction of lymphoid chimerism in the complete absence of β2m-dependent T cells resulted in uncontrolled IgE responses that remained high until 12 weeks of age (Figure 1b). The incidence of IgE response, at 4 weeks of age, in the CB6F1→BALB/c and CB6F1β2m°→BALB/cβ2m° combinations were 65% (37/57) and 83% (44/53), respectively. Interestingly, whereas most BALB/c mice injected at birth with CB6F1 spleen cells remained healthy and survive, more than 50% of mice died by 15 weeks of age in the β2m° combination (Figure 1c). This increased mortality was associated with an higher incidence of albuminuria (Figure 1d). Urine albumin levels were above 100 μg/mL in 66% of albuminuric mice, with a maximum of 16 mg/mL in 1 mouse (data not shown). Immunofluorescence analysis on kidney cryostat sections demonstrated the presence of heavier glomerular Ig deposits in neonatally injected β2m° compared with WT mice (Figure 2, b and d), that were mainly composed of IgG1 and IgM (data not shown). Moreover, whereas in the WT combination kidney bound Ig were mostly found in the mesangium (Figure 2b), both mesangial and subendothelial granular Ig deposits were observed in 7 of the 11 neonatally injected β2m° mice examined (Figure 2d), the 4 other mice displayed only mesangial Ig deposits. Unlike BALB/c mice injected with CB6F1 splenocytes, that exhibited only mild mesangial deposits without any cell proliferation or interstitial changes (Figure 2a), glomeruli from 4 of 6 neonatally injected β2m° mice examined by light microscopy exhibited heavy mesangial and parietal, mainly subendothelial, deposits (Figure 2c). There was also a double contour appearance after silver impregnation (data not shown). In addition, perivascular infiltrates of mononuclear cells were observed in 2 of the 6 β2m° mice (data not shown). Renal function appeared normal in these mice, as urea and creatinine blood levels were unchanged (data not shown). Another striking observation of the β2m° chimeric model was that these mice developed a much larger splenomegaly compared with the WT combination, which still persisted until 10 weeks of age, with variable degrees of fibrosis (not shown). The increased spleen size in neonatally induced β2m° chimeras was associated with a significant augmentation in CD4 T and B cell numbers (data not shown).

Figure 1

Persistence of IgE production, high mortality rate, and increased incidence of albuminuria in β2m° BALB/c mice injected at birth with β2m° CB6F1 splenocytes. Normal (a) and β2m° (b) BALB/c mice were injected at birth with 50 × 10⁶ CB6F1 spleen cells from normal or β2m° donors, respectively. Mice were bled at 4–16 weeks of age and IgE were quantified in individual mice serum by ELISA. Statistically significant decrease of IgE concentration was compared between the 8-, 10-, 12-, and 16-week values and the 4-week values. Bars represent means of individual mice. (c) Mortality of neonatally injected β2m° (solid line; n = 9) and WT mice (dotted line; n = 13) was followed up until the age of 22 weeks. (d) Albumin was quantified in urine by ELISA at 10–16 weeks of age. Mice were considered albuminuric when albumin titer was greater than 15 μg/mL. Statistical analyses were performed as indicated in Methods. ^AP < 0.001; ^BP < 0.01.

Figure 2

Histological changes and immunofluorescence analysis of kidney. (a and b) Representative changes in 6 week-old BALB/c mice injected at birth with CB6F1 splenocytes. Moderate mesangial deposits are seen on H&E staining (a, ×800) and after immunostaining for mouse Ig (b, ×500). (c and d) Representative changes in 6-week-old β2m° BALB/c mice neonatally injected with β2m-deficient CB6F1 spleen cells. Strong mesangial and subendothelial immune complex deposits are seen on PAS staining (c, ×800) and after immunostaining for mouse Ig (d, ×500).

Spared Th2 immunity is associated with the maintenance of lymphoid chimerism in β2m-deficient mice. Because splenomegaly in this model is likely to be associated with uncontrolled Th2 immunity, we next compared the expression of memory/activation markers on T and B cells between these 2 combinations. Results in Figure 3 show that there was a significant increase in antigen-experienced splenic CD4 T cells in both WT and β2m° mice at 6 weeks of age after neonatal induction of lymphoid chimerism. CD44^highCD62L^low CD4 T-cell numbers were higher in CB6F1β2m°-injected BALB/cβ2m° mice compared with the WT combination. This difference was more evident at 10 weeks of age, where number of memory/effector CD4 T cells were significantly increased in neonatally injected β2m° but not WT mice. Analysis of the splenic B-cell compartment showed that the expression of I-E molecules on B cells was significantly higher in neonatally injected β2m° compared with WT mice at both 6 and 10 weeks of age (Figure 4a). Because MHC class II expression on B cells is upregulated by IL-4 (23, 24), we looked for the presence of IL-4 transcripts in splenic CD4 T cells. Data in Figure 4b demonstrate that IL-4 mRNA were readily detectable in the majority (85%) of 6-week-old neonatally induced β2m° chimeras, compared with 22% in the WT combination. IL-4 is likely to be produced by allospecific Th2 cells interacting with donor F1 B cells, resulting in donor B-cell differentiation to Ig-producing cells. Indeed, the number of plasmocytes was selectively increased in the spleen of 6-week-old neonatally injected β2m° mice (Figure 4c) and was correlated with the persistence of significantly higher levels of B-cell chimerism (Figure 4d). To analyze further the effect of the absence of MHC class I–T-cell interactions in the development of Th2 cells in this model, we measured type-2 cytokine production by splenic CD4 T cells from 6-week-old mice after stimulation with plate bound anti-CD3ε mAb. Data in Figure 5 show that there was a 4-fold upregulation of IL-4 production in F1-injected WT mice compared with uninjected control animals. In striking contrast, IL-4 secretion was dramatically increased (54-fold) in the neonatally induced β2m° chimeras. The lower IL-4 production in control β2m° compared with WT BALB/c mice may be explained by the absence of NK T cells that promptly produce this cytokine in response to TCR ligation (25). As for IL-4, IL-10 secretion upon TCR engagement was strongly increased in CD4 T cells from both combinations, but this was more evident in β2m° mice. On the other hand, the amounts of IL-5 produced by CD4 T cells was significantly upregulated only in β2m° chimeras (Figure 5). IFN-γ synthesis was slightly increased (2- to 3-fold) in neonatally injected compared with control mice both in the WT and β2m° combinations (data not shown). Thus, the induction of lymphoid chimerism in the absence of MHC class I–T-cell interactions led to the accumulation in the spleen of CD4 T cells with a memory phenotype probably as a consequence of the long-term persistence of donor B cells, resulting in chronic alloantigenic stimulation and increased type-2 cytokine production by CD4 T cells.

Figure 3

Accumulation of CD44^hi/CD62L^low T lymphocytes in neonatally injected β2m° BALB/c mice. WT and β2m° BALB/c mice were injected at birth with normal or β2m-deficient CB6F1 spleen cells as in Figure 1. Uninjected WT and β2m° BALB/c mice were used as control. Splenocytes from 6- to 10-week-old neonatally injected and control mice were analyzed by flow cytometry to quantify the CD44^hi/CD62L^low memory/effector CD4⁺ T cells. Bars represent means ± SEM of individual mice (4–12 mice per group). Numbers in brackets indicate the fold increase of neonatally induced chimera versus control mean values. ^AP < 0.05; ^BP < 0.01.

Figure 4

Upregulation of MHC class II on B cells, increased number of plasmocytes, and IL4-mRNA are associated with the maintenance of lymphoid chimerism in β2m° BALB/c mice. (a) I-E expression level on splenic B lymphocytes from 6- and 10-week-old neonatally injected and control mice was evaluated by flow cytometry. Data are expressed as fold increase versus control (mean ± SEM) of 4–12 mice per group. (b) IL4-mRNA expression in negatively enriched CD4 splenic T cells, from 6-week-old neonatally injected WT and β2m° (n = 14) mice, was quantified by semiquantitative RT-PCR. Data are expressed as the ratio of IL-4/HPRT mRNA levels from 9–14 individual mice per group. (c) Cytospins were prepared from spleen cells suspensions from 6-week-old mice; total plasmocyte numbers were calculated from 4–7 individual mice per group. (d) Percentages of I-A^b+ splenic B220⁺ cells from 6-week-old WT (n = 16) and β2m° (n = 9) mice injected at birth with semiallogeneic splenocytes were evaluated by flow cytometry analysis. Bars represent means of individual mice.

Figure 5

Th2-type cytokines production by in vitro stimulated CD4 T lymphocytes. Negatively enriched splenic CD4 T cells from 6-week-old control and neonatally injected mice were stimulated with plate-bound anti-CD3ε mAb for 48 hours. IL-4, IL-10, and IL-5 were quantified in culture supernatants by ELISA. Results are expressed as mean ± SEM cytokine concentration (ng/mL) from 4–7 mice per group. Numbers in brackets indicate the fold increase of neonatally injected versus control mean values. ^AP < 0.05; ^BP < 0.01.

Hypereosinophilic syndrome develops after induction of lymphoid chimerism in the absence of MHC class I–T-cell interactions. Because the uncontrolled Th2 response that develops in β2m° chimeric mice is associated with an increased splenomegaly with signs of fibrosis in most animals, a morphological analysis was performed. Data in Figure 6 show that although neonatally injected WT mice exhibited minimal change in spleen morphology (Figure 6a), a considerable modification of spleen architecture was observed in chimeric β2m° BALB/c mice (Figure 6b). Such changes were characterized by a strong hyperplasia of periarteriolar lymphoid sheaths (PALS) (Figure 6b) and by prominent eosinophil- (Figure 6c), basophil-, and/or mastocyte-containing (Figure 6d) infiltrates in the majority (>80%) of β2m° chimeras. These modifications were associated with variable degree of fibrosis (Figure 6b and data not shown). Massive perivascular lymphoid infiltration (Figure 6f) and tissue eosinophilia (data not shown) were also occasionally found in the lung (2 of 5 β2m° chimeric mice tested) and were correlated with disease severity (data not shown). Histological analysis of others organs such as liver, heart, and intestine did not reveal significant abnormalities (data not shown). Cytological analysis confirmed the increased eosinophilia in the spleen of neonatally induced β2m° chimeras compared with F1-injected WT mice (Figure 7a). Furthermore, a significant increase in eosinophil number was detected in the blood of 6- to 10-week-old β2m° chimeric mice (Figure 7b). Such abnormalities were barely detectable or absent in the majority of F1-injected β2m-sufficient mice. Thus, the uncontrolled type-2 immune response that develops in β2m° BALB/c mice injected at birth with semiallogeneic β2m° splenocytes leads to an hypereosinophilic syndrome.

Figure 6

Histological changes in the spleen (a–d) and lung (e and f) of 6-week-old WT (a and e) and β2m° BALB/c mice (b, c, d, and f) injected at birth with normal or β2m-deficient CB6F1 splenocytes, respectively. In comparison with the spleen of WT mice (a, ×125), a striking hyperplasia of PALS is observed on H&E staining of β2m° mice (b, ×100). On higher magnification (×800), one sees large numbers of eosinophils (arrows) (c) in a background of activated lymphoid cells, some showing immunoblastic features (arrowheads). Giemsa staining shows that cells with large basophilic granules (arrows) are also present in PALS (d). Note also the presence of scattered plasma cells (arrowheads). (e and f) H&E staining shows perivascular infiltrates (arrows) consisting of activated lymphoid cells associated with eosinophils and basophils in the lung of a β2m° mouse (f, ×100). No significant changes are seen in the lung of a WT mouse (e, ×100).

Figure 7

Increased eosinophil number in spleen and blood of β2m° BALB/c mice injected at birth with β2m-deficient CB6F1 splenocytes. Splenic (a) and blood (b) eosinophil numbers were assessed in 6- and 10-week-old control and neonatally injected WT or β2m° mice. Bars represent mean values of individual mice (3–9 mice per group). ^AP < 0.05; ^BP < 0.01.

The present report demonstrates that induction of lymphoid chimerism in BALB/c mice in the absence of MHC class I–T-cell interactions results in a lethal syndrome associated with uncontrolled Th2 immunity and sustained hypereosinophilia. In addition, these mice develop a severe form of antibody-mediated glomerulonephritis probably as a consequence of autoreactive F1 donor B-cell activation by host allohelper T cells. However, the autoimmune glomerulonephritis is not, or is not only, responsible for the mortality because biologic signs of renal failure were absent, and some mice died without proteinuria. We postulate that lung damages may contribute to the increased mortality in β2m° chimeric mice. This model markedly contrasts with the mild and self-limited form of HVG disease that occurs in WT BALB/c mice injected with CB6F1 spleen cells (11, 13) and represents the first example of a Th2-dependent lethal chronic inflammatory disease in mice associated with an hypereosinophilic syndrome.

The mechanisms by which β2m deficiency leads to an uncontrolled Th2 response are likely to be related to the lack of interactions between donor MHC class I molecules and host CD8 T cells. In previous experiments, we have shown that the injection of CB6F1 spleen cells into normal or β2m° BALB/c mice induces identical Th2-type responses (17). This could be explained by the fact that in β2m° mice there is a residual population of CD8 T cells that can readily expand after in vivo priming with cells expressing either self- or allogeneic-MHC class I molecules (26, 27). Thus, to avoid the interference of this minor population of CD8 T cells, MHC class I molecules must also be absent on donor-derived CB6F1 cells. The demonstration that MHC class I–T-cell interactions play a crucial role in the control of Th2 immunity and its associated lethal syndrome is rather unexpected. Indeed, it is widely accepted that after neonatal transplantation, tolerance alloantigen-specific CTLs are deleted in the thymus and in the periphery (2, 3). However, the failure to detect allospecific CTL activity in neonatally primed mice could also be explained by a difference in phenotype acquisition in the CD8 compartment rather than clonal deletion. Indeed, it has been shown that under the influence of IL-4, CD8 T cells can develop into noncytotoxic, type-2 cytokines producing cells (28). Alternatively, it may be possible that functional alloreactive CTL cells still persist in tolerized mice, but that their frequency is largely underestimated in limiting dilution analysis. The hypothesis that CD8 T cells might play a crucial role in this model is in apparent contradiction to previously published experiments showing that administration of anti-CD8 mAb in vivo had no effect on disease severity (29). This could easily be explained by the fact that the few CD8 T cells that might have escaped in vivo depletion with anti-CD8 mAb could still mediate efficient regulatory effector functions. Indeed, as already discussed here, the minor CD8 T-cell population that persist in β2m° mice seems to be as effective as conventional CD8 T cells in the control of Th2 responsiveness (17). Taken together, our data support the conclusion that MHC class I expression on donor spleen cells and the presence of potentially MHC class I–reactive T cells in the host are the critical factors that control HVG disease severity induced by neonatal injection of semiallogeneic spleen cells.

Several mechanisms could be proposed to explain how MHC class I–reactive T cells could regulate Th2 immunity in this lethal HVG disease. First, as these abnormalities are strongly correlated with the persistence of B-cell chimerism, it is reasonable to speculate that the main effector function of class I–reactive T cells would be to eliminate donor-derived antigen-presenting cells (APC), thereby preventing chronic allohelper T-cell activation, resulting in a mild and self-limited disease. Second, the absence of CD8 T-cell activation during the early phase of allohelper CD4 T-cell priming may result in an enhanced Th2-cell differentiation in vivo. In support of this, we have recently shown that immunization of adult parental strains with semiallogeneic APC, including dendritic cells, in the absence of MHC class I–specific T cells induces a rapid and vigorous expansion of CD4 T cells secreting type-2 cytokines and TGF-β in response to allogeneic class II MHC products (G. Foucras et al., manuscript submitted for publication). As in other experimental models (30, 31), host CD8 T cells could modulate Th2 cell priming by the secretion of IFN-γ. Indeed, it has been shown that the early administration of this cytokine inhibits the Th2-type response induced by the neonatal injection of semiallogeneic spleen cells (32). Finally, there is another subset of MHC class I–restricted T cells, expressing the NK1.1 molecule, that is absent in β2m° mice (25). NK1.1⁺ T cells have been reported to ameliorate autoimmune disease (33, 34) and acute GVH disease (35) through the production of IL-4. Whether or not this T cell population play a role in our model is not clear. Lethal HVG disease is associated with uncontrolled Th2 immunity, which would be exacerbated rather than suppressed by IL-4 production from NK T cells. However, as these T-cell subset can also produce IFN-γ, we cannot exclude that they can effectively control type-2 Th cells development in β2m-sufficient mice.

An important feature of the present study is that the injection of semiallogeneic splenocytes into newborn BALB/c mice leads, in the absence of MHC class I–T-cell interactions, to a lethal form of HVG disease. Although, the reasons why mice die during the course of the disease remain to be elucidated, it must be stressed that besides the marked glomerulonephritis, these mice also exhibit sustained blood eosinophilia and an important accumulation of eosinophils not only in their spleen but also in their lungs, making this model relevant to human hypereosinophilia (36). In addition, high numbers of basophils and/or mastocytes were found to be infiltrating the PALS. Eosinophils have been recently shown to mediate rejection of acute and chronic MHC class II–disparate skin allografts in mice (37, 38) and are probably responsible for the generation of fibrotic lesions in some chronic inflammatory diseases (39, 40). Variable degrees of fibrosis were evident in the spleen of chimeric β2m° diseased mice and were associated with eosinophil infiltrates indicating that such mechanisms could also be involved in our model. Hypereosinophilia was also associated with an increased capacity of CD4 T cells to secrete IL-5, a cytokine that plays a central role in the differentiation, proliferation, and recruitment of eosinophils (41). Whether or not eosinophils play a major effector role in the observed histological and clinical manifestations, including mortality remains to be elucidated.

In conclusion, we have demonstrated that the HVG reaction that develops in the absence of MHC class I–T-cell interactions leads to a lethal leukoproliferative disorder associated with uncontrolled Th2-type immunity and hypereosinophilic syndrome in mice. These data support the hypothesis that CD8 T cells might play a more active role than previously anticipated by regulating lymphoid chimerism and/or Th2-cell development resulting in a mild and self-limited form of HVG disease. Thus, induction of neonatal lymphoid chimerism in BALB/c mice in the absence of MHC class I–T-cell interactions represents an unique model of Th2-dependent chronic inflammatory disease that may help to understand the mechanisms involved in idiopathic hypereosinophilic syndrome in humans (36).

We thank M. Calise, S. Pilipenko, and M. Marche for their skillful technical assistance. This work was supported by Institut National de la Santé et de la Recherche Médicale (INSERM), and grants from Conseil Général de Région Midi-Pyrénées, Etablissement Français des Greffes, and Université Paul Sabatier. G. Foucras is on a leave of absence from Ecole Nationale Vétérinaire de Toulouse, France. J. Coudert is supported by the Ministère de l’Education Nationale, de la Recherche et de la Technologie. We thank D. Abramowicz for his helpful comments on the manuscript.

1. Billingham, RE, Brent, L, Medawar, PB. Actively acquired tolerance of foreign cells. Nature 1953. 172:603-606.
   View this article via: PubMed CrossRef Google Scholar
2. Feng, HM, Glasebrook, AL, Engers, HD, Louis, JA. Clonal analysis of T cell unresponsiveness to alloantigens induced by neonatal injection of F1 spleen cells into parental mice. J Immunol 1983. 131:2165-2169.
   View this article via: PubMed Google Scholar
3. Nossal, GJV, Pike, BL. Functional clonal deletion in immunological tolerance to major histocompatibility complex antigens. Proc Natl Acad Sci USA 1981. 78:3844-3847.
   View this article via: PubMed CrossRef Google Scholar
4. Powell, TJ, Streilein, JW. Neonatal tolerance induction by class II alloantigens activates IL-4–secreting, tolerogen-responsive T cells. J Immunol 1990. 144:854-859.
   View this article via: PubMed Google Scholar
5. Schurmans, S, et al. In vivo effects of anti–IL-4 monoclonal antibody on neonatal induction of tolerance and on associated autoimmune syndrome. J Immunol 1990. 145:2465-2473.
   View this article via: PubMed Google Scholar
6. Abramowicz, D, et al. Persistence of anti-donor allohelper T cells after neonatal induction of allotolerance in mice. Eur J Immunol 1990. 20:1647-1653.
   View this article via: PubMed CrossRef Google Scholar
7. Chen, N, Field, EH. Enhanced type 2 and diminished type 1 cytokines in neonatal tolerance. Transplantation 1995. 59:933-941.
   View this article via: PubMed Google Scholar
8. Abramowicz, D, et al. Increased expression of Ia antigens on B cells after neonatal induction of lymphoid chimerism in mice: role of interleukin 4. Eur J Immunol 1990. 20:469-476.
   View this article via: PubMed CrossRef Google Scholar
9. Donckier, V, et al. Critical role of interleukin 4 in the induction of neonatal transplantation tolerance. Transplantation 1995. 59:1571-1576.
   View this article via: PubMed Google Scholar
10. Gao, Q, Chen, N, Rouse, TM, Field, EH. The role of interleukin-4 in the regulation phase of allogeneic neonatal tolerance. Transplantation 1996. 62:1847-1854.
    View this article via: PubMed CrossRef Google Scholar
11. Luzuy, S, Merino, J, Engers, H, Izui, S, Lambert, PH. Autoimmunity after induction of neonatal tolerance to alloantigens: role of B cell chimerism and F1 donor B cell activation. J Immunol 1986. 136:4420-4426.
    View this article via: PubMed Google Scholar
12. Goldman, M, et al. Hyperactivity of donor B cells after neonatal induction of lymphoid chimerism in mice. Clin Exp Immunol 1988. 72:79-83.
    View this article via: PubMed Google Scholar
13. de la Hera, M, et al. Self-limited autoimmune disease related to transient donor B cell activation in mice neonatally injected with semi-allogeneic F1 cells. Int Immunol 1992. 4:67-74.
    View this article via: PubMed CrossRef Google Scholar
14. Kühn, R, Rajewsky, K, Müller, W. Generation and analysis of interleukin-4 deficient mice. Science 1991. 272:50-54.
    View this article via: PubMed CrossRef Google Scholar
15. Takeda, K, et al. Essential role of Stat6 in IL-4 signalling. Nature 1996. 380:627-630.
    View this article via: PubMed CrossRef Google Scholar
16. Shimoda, K, et al. Lack of IL-4–induced Th2 response and IgE class switching in mice with disrupted Stat6 gene. Nature 1996. 380:630-633.
    View this article via: PubMed CrossRef Google Scholar
17. Foucras, G, et al. β2-microglobulin-dependent T cells are not necessary for alloantigen-induced T helper (Th)2 responses after neonatal induction of lymphoid chimerism in mice. J Immunol 1998. 161:1751-1757.
    View this article via: PubMed Google Scholar
18. Koller, BH, Marrack, P, Kappler, JW, Smithies, O. Normal development of mice deficient in β2M, MHC class I proteins, and CD8⁺ T cells. Science 1990. 248:1227-1230.
    View this article via: PubMed CrossRef Google Scholar
19. Guéry, J-C, Galbiati, F, Smiroldo, S, Adorini, L. Selective development of T helper (Th)2 cells induced by continuous administration of low dose soluble proteins to normal and β2-microglobulin–deficient mice. J Exp Med 1996. 183:485-497.
    View this article via: PubMed CrossRef Google Scholar
20. Mason, DW, Pugh, CW, Webb, M. The rat mixed lymphocyte reaction: roles of a dendritic cell in intestinal lymph and T-cell subsets defined by monoclonal antibodies. Immunology 1981. 44:75-87.
    View this article via: PubMed Google Scholar
21. Tomonari, K, Spencer, S. Epitope specific binding of CD8 regulates activation of T cells and induction of cytotoxicity. Int Immunol 1990. 2:1189-1194.
    View this article via: PubMed CrossRef Google Scholar
22. Reiner, SL, Zheng, S, Corry, DB, Locksley, RM. Constructing polycompetitor cDNAs for quantitative PCR. J Immunol Methods 1993. 165:37-46.
    View this article via: PubMed CrossRef Google Scholar
23. Roehm, NW, et al. Interleukin-induced increase in Ia expression by normal mouse B cells. J Exp Med 1984. 160:679-694.
    View this article via: PubMed CrossRef Google Scholar
24. Noelle, R, Krammer, PH, Ohara, J, Uhr, JW, Vitetta, ES. Increased expression of Ia antigens on resting B cells: an additional role for B-cell growth factor. Proc Natl Acad Sci USA 1984. 81:6149-6153.
    View this article via: PubMed CrossRef Google Scholar
25. Bendelac, A, Rivera, MN, Park, S-H, Roark, JH. Mouse CD1-specific NK1 T cells: development, specificity, and function. Annu Rev Immunol 1997. 15:535-562.
    View this article via: PubMed CrossRef Google Scholar
26. Apasov, S, Sitkovsky, M. Highly lytic CD8+, alpha beta T-cell receptor cytotoxic T cells with major histocompatibility complex (MHC) class I antigen-directed cytotoxicity in beta 2-microglobulin, MHC class I-deficient mice. Proc Natl Acad Sci USA 1993. 90:2837-2841.
    View this article via: PubMed CrossRef Google Scholar
27. Glas, R, Öhlén, C, Höglund, P, Kärre, K. The CD8⁺ T cell repertoire in β2-microglobulin–deficient mice is biased toward reactivity against self-major histocompatibility class I. J Exp Med 1994. 179:661-672.
    View this article via: PubMed CrossRef Google Scholar
28. Le Gros, G, Erard, F. Non-cytotoxic, IL-4, IL-5, IL-10 producing CD8+ T cells: their activation and effector functions. Curr Opin Immunol 1994. 6:453-457.
    View this article via: PubMed CrossRef Google Scholar
29. Merino, J, et al. Autoimmune syndrome after induction of neonatal tolerance to alloantigens: effects of in vivo treatment with anti–T cell subset monoclonal antibodies. J Immunol 1987. 139:1426-1431.
    View this article via: PubMed Google Scholar
30. Srikiatkhachorn, A, Braciale, TJ. Virus-specific CD8⁺ T lymphocytes downregulate T helper cell type 2 cytokine secretion and pulmonary eosinophilia during experimental murine respiratory syncytial virus infection. J Exp Med 1997. 186:421-432.
    View this article via: PubMed CrossRef Google Scholar
31. Hussell, T, Baldwin, CJ, O’Garra, A, Openshaw, PJM. CD8⁺ T cells control Th2-driven pathology during pulmonary respiratory syncytial virus infection. Eur J Immunol 1997. 27:3341-3349.
    View this article via: PubMed CrossRef Google Scholar
32. Donckier, V, et al. IFN-γ prevents Th2 cell–mediated pathology after neonatal injection of semiallogenic spleen cells in mice. J Immunol 1994. 153:2361-2368.
    View this article via: PubMed Google Scholar
33. Gombert, JM, et al. Early quantitative and functional deficiency of NK1⁺-like thymocytes in the NOD mouse. Eur J Immunol 1996. 26:2989-2998.
    View this article via: PubMed CrossRef Google Scholar
34. Hammond, KJL, et al. α/β-T cell receptor (TCR)⁺CD4^–CD8^– (NKT) thymocytes prevent insulin-dependent diabetes mellitus in nonobese diabetic (NOD)/Lt mice by the influence of interleukin (IL)-4 and/or IL-10. J Exp Med 1998. 187:1047-1056.
    View this article via: PubMed CrossRef Google Scholar
35. Zeng, D, et al. Bone marrow NK1.1^– and NK1.1⁺ T cells reciprocally regulate acute graft versus host disease. J Exp Med 1999. 189:1073-1081.
    View this article via: PubMed CrossRef Google Scholar
36. Weller, PF, Bubley, GJ. The idiopathic hypereosinophilic syndrome. Blood 1994. 83:2759-2779.
    View this article via: PubMed Google Scholar
37. Le Moine, A, et al. Critical roles for IL-4, IL-5, and eosinophils in chronic skin allograft rejection. J Clin Invest 1999. 103:1659-1667.
    View this article via: JCI PubMed CrossRef Google Scholar
38. Le Moine, A, et al. IL-5 mediates eosinophilic rejection of major histocompatibility complex class II-disparate skin allografts in mice. J Immunol 1999. 163:3778-3784.
    View this article via: PubMed Google Scholar
39. Kroegel, C, Virchow, J-C, Luttmann, W, Walker, C, Warner, JA. Pulmonary immune cells in health and disease: the eosinophil leucocyte (Part I). Eur Respir J 1994. 7:519-543.
    View this article via: PubMed CrossRef Google Scholar
40. Rothenberg, ME. Eosinophilia. N Engl J Med 1998. 338:1592-1600.
    View this article via: PubMed CrossRef Google Scholar
41. Sanderson, CJ. Interleukin-5, eosinophils, and disease. Blood 1992. 79:3101-3109.
    View this article via: PubMed Google Scholar