Th17 Alloimmunity Prevents Neonatal Establishment of Lymphoid Chimerism in IL-4-Deprived Mice


Véronique Flamand,


Immune responses in newborn mice are known to be biased toward the helper type 2 phenotype. This may account for their propensity to develop tolerance. Herein, we evaluated the effects of IL-4 deprivation on CD4+ T-cell activities elicited by neonatal exposure to allogeneic spleen cells. We showed that chimerism, Th2-type polarization and pathology, as well as skin allograft acceptance were inhibited in BALB/c mice immunized at birth with (A/J x BALB/c) F1 spleen cells upon in vivo IL-4 neutralization. While IL-4 neutralization inhibited the development of Th2 cells in this model, it led to the accumulation of IL-17A, IL-17F, IL-22, IL-6 and RORγt mRNA in the spleen or graft tissues. Moreover, IL-4 deprivation led to the differentiation of donor-specific Th17 cells with a concomitant Th1 response characterized by IFN-γ production. The Th17-type response emerging in IL-4-deprived mice was found to mediate both intragraft neutrophil infiltration and the abrogation of B-cell chimerism. Neutralization of this Th17 response failed however to restore functional skin graft acceptance. Collectively, our observations indicate that the neonatal Th2 response opposes the development of Th17 cells, and that Th17 cells are responsible for controlling lymphoid chimerism in mice neonatally injected with semiallogeneic cells.


allophycocyanin; FITC, fluorescein isothiocyanate; IFN-γ, interferon-γ; IL-13R, IL-13 receptor; NS, not significant; PE, phycoerythrin; RORγt, retinoic orphan receptor γt; TGF-β, transforming growth factor β; Treg, regulatory T cell


More than 50 years ago, Medawar and coworkers discovered that transplantation tolerance across complete MHC disparities can be induced by neonatal injection of semiallogeneic spleen cells (1). This status of neonatal tolerance was later associated with a differentiation of alloantigen-specific Th2-type cells (2–5). Unique properties of neonatal CD4+ T cells are likely to contribute to this Th2-type polarization, as shown by hypomethylation of a regulatory region in the Th2 genomic locus, allowing a rapid and high production of the Th2 cytokine IL-4 (6). Moreover, apoptosis of primary Th1 cells induced by IL-4 (7) and a delayed maturation of IL-12-producing dendritic cells in early life (8) were further involved in the neonatal Th2 bias. We previously demonstrated that the neonatal Th2-type immune deviation and associated transplantation tolerance can be prevented by perinatal neutralization of IL-4 (9).

Th17 cells produce IL-17A, IL-17F and IL-22 (10,11). Their differentiation from naive CD4+ T cells requires two cytokines, TGF-β and IL-6 (12–14), and the expression of the retinoic orphan receptor RORγt, which is now defined as the master regulator of the Th17 pathway (15). The lineage-defining cytokine IL-17 is characterized by a high inflammatory potential, as it can mobilize, recruit and activate neutrophils (16). Neutrophils have been involved in allograft rejection. Indeed, neutrophil infiltrates were observed in rejected MHC class II disparate skin graft of IL-4-deficient mice (17), compatible with a Th17-dependent pathway of allogeneic graft rejection.

Based on the findings that blocking of IL-4 breaks neonatal transplantation tolerance (9), we hypothesized that early neutralization of IL-4 could allow the differentiation of alloreactive Th17 cells in newborn mice and be involved in rejection of donor-type cells or tissues.

Materials and Methods


BALB/c (H-2d) and C57BL/6 (H-2b) were purchased from Harlan (The Nederlands), A/J (H-2k) and IL-4−/− (H-2d) from The Jackson Laboratory (Bar Harbor, ME, USA). BALB/c, IL-4−/− and (A/J x BALB/c) F1 hybrids were bred and housed in our specific pathogen-free facility. All animal studies were approved by the institutional animal care and local use committee.

In vivo treatments

Neonatal tolerance was induced in BALB/c mice by injection of 2 × 107 (A/J x BALB/c) F1 spleen cells intravenously (i.v.) or intraperitoneally (i.p.) within the first 24 hours (h) of life (day 0). Newborns were also injected i.p. on days 0, 2, 7, 14 with 30 to 60 μg of purified rat antimouse IL-4 mAb (clone 11B11, BD Biosciences, Franklin Lakes, NJ, USA) or control purified rat IgG (Sigma-Aldrich, St. Louis, MO, USA or Jackson ImmunoResearch Laboratories, West Grove, PA, USA). For graft survival experiments, mice were injected i.v. or i.p. with 200 μg antimouse IL-17A mAb (clone 50104, R&D Systems, Minneapolis, MN, USA) or control purified IgG on days 1, 4, 7, 10, 13, 16 and 19 following transplantation. For graft neutrophilia experiments, mice were injected i.v. or i.p. with 100 μg of rat antimouse IL-17A mAb or control purified rat IgG the day before transplantation and i.p. with 50 μg on days 2 and 7 following transplantation. For chimerism assessment, neonates were injected i.p. with 50 μg anti-IL-6 (clone MP5–20F3, BioXCell), anti-IL-17A or anti-Gr-1 mAb (clone RB6-8C5, BD Biosciences), or purified rat IgG on day 1 and with 25 μg on days 3, 8, 11 based on supplier recommendations and our own neutralizing tests.

Skin grafting

Skin grafting was performed as previously described (18).

Mixed lymphocyte culture

MLC were prepared in RPMI 1640 medium (Lonza Research Products, Switzerland) containing 20 mM HEPES, 2 mM L-glutamine, 1 mM nonessential amino acids (Lonza Research Products), 5% heat-inactivated fetal calf serum, 100 mM sodium pyruvate (Lonza Research Products), penicillin (10 U/mL)-streptomycin (10 μg/mL) and 10−5M 2-ME. Cells (2.5 × 106/well) from pooled axillary, inguinal and mesenteric lymph nodes or from spleen of BALB/c mice were stimulated with 1.5 to 2.5 × 106 irradiated (2000 rad) syngeneic BALB/c, donor-type (A/J x BALB/c) F1 or third-party (C57BL/6) spleen cells in 1 mL of culture medium. Cultures were kept at 37°C in a 5% CO2 atmosphere. Supernatants were harvested after 72 h for cytokine detection or cells were collected after 5 days for intracellular cytokine assay.

Cytokine production assays and IgE levels

Cytokine concentrations in supernatants were assessed by enzyme-linked immunosorbent assay using commercially available kits (Duoset, R&D Systems, for IL-4, IL-13, IFN-γ and IL-17A and Opt EIA set, BD Biosciences, for IL-5). IgE serum levels were determined as previously described (19).

Antibodies and flow cytometry

Microchimerism was assessed by flow cytometry using anti-H-2Kk phycoerythrin (PE; clone 36–7-5), anti-I-Ak fluorescein isothiocyanate (FITC; clone 11–5.2), anti-CD45R/B220 Pacific Blue (clone RA3-6B2), anti-CD19 allophycocyanin (APC; clone 1D3), anti-TCRβ FITC (clone H57–597) and anti-CD3 Pacific Blue (clone 500A2). For intracellular staining, cells were stimulated with 5 ng/mL phorbol myristate acetate, 500 ng/mL ionomycin and 1 μl/mL Golgi Plug (BD Biosciences) for 4 h at 37°C. Cells were stained with FITC-conjugated anti-CD4 (clone GK1.5) and fixed with Cytofix/Cytoperm (BD Biosciences). After washing with PermWash buffer (BD Biosciences), staining with anti-IFN-γ APC (clone XMG1.2) and anti-IL-17A PE (clone TC11-18H10.1) or isotype controls was performed, before analysis on a Cyan ADP flow cytometer (DakoCytomation, Glostrup, Denmark). All antibodies were purchased from BD Biosciences.

Quantification of cytokine transcripts

Total RNA was extracted using the MagNA Pure LC RNA Isolation Kit III (Tissue) (Roche Diagnostics). Reverse transcription and quantitative real-time PCR were performed using the Lightcycler RNA Master Hybridization Probe Kit (Roche Diagnostics) on a Lightcycler apparatus (Roche Diagnostics). For individual samples, mRNA levels were normalized to those of β-actin used as reference. The sequences of primers and probes are available on request (; group name: FLAMAND).


Skin graft sections were stained with hematoxylin and eosin, after fixation in 10% formalin neutral solution and paraffin embedding. Tissue neutrophilia and IL-17A-producing cells were determined by immunostaining. Sections were incubated with purified rat antimouse Ly-6G mAb (clone 1A8, BD Biosciences) or rat antimouse IL-17A mAb (clone 50104, R&D Systems) and with biotin-SP-conjugated AffiniPure goat antirat IgG (H+L) (Jackson Immunoresearch Laboratories). Streptavidin-HRP (LSAB2 System-HRP kit, DakoCytomation) was added, before staining with DAB substrate (Liquid DAB+Substrate–Chromogen System, DakoCytomation) and hematoxylin (Hematoxylin Code S2020, Dako REAL). When indicated, background staining corresponds to labeling with biotin-conjugated secondary Ab and Streptavidin-HRP alone.


Data were expressed as mean ± standard error of the mean. Statistical analysis was performed using the two-tailed nonparametric Mann–Whitney test or Student t-test when indicated. Graft survival curves were compared using the Kaplan–Meier/log-rank test (p < 0.05 considered as significant).


Perinatal neutralization of IL-4 prevents neonatal transplantation tolerance, inhibits Th2-type polarized response and allows the emergence of antidonor Th17 cells

The injection of (A/J x BALB/c) F1 spleen cells into BALB/c neonates induces lymphoid chimerism, immune polarization and transplantation tolerance of donor-type skin allografts (9,18–20). We first evaluated the effects of perinatal administration of anti-IL-4 mAb on neonatal transplantation acceptance and Th2-type responses. We grafted mice at 6 weeks of age and monitored graft survival during 6 weeks. As expected, all untreated mice rejected their skin grafts within 18 days (Figure 1A). In contrast, 58% of mice neonatally injected with F1 cells and control IgG retained their graft for more than 40 days. As previously reported (9), early neutralization of IL-4 resulted in acute rejection of skin allografts, as 88% of mice treated at birth with anti-IL-4 mAb and F1 cells rejected their graft with kinetics similar to that of untreated mice (median survival time: 16, >40 and 16 days in untreated, F1-treated, and F1- and anti-IL-4-treated mice respectively, p < 0.001 between F1-treated mice and untreated or F1- and anti-IL-4-treated mice). Allogeneic skin grafts from F1-treated mice displayed intact structure with no detectable mononuclear infiltrates while grafts from untreated mice or mice having received anti-IL-4 mAb in addition to F1 cells at birth displayed altered structure and a massive infiltrate in derma and epiderma (Figure 1B). We assessed the status of the chimerism by monitoring the presence of donor-type lymphoid cells in 2-week-old BALB/c recipient mice. As shown in Figure 1C, CD45R-B220+ cells expressing donor I-Ak and H-2k molecules were detected in the spleen of F1-treated mice. B220/I-Ak+ and B220/H-2k+ cells were also present. Neutralization of IL-4 significantly decreased the percentages of all these donor cell types without completely eliminating them. We next investigated the effects of early neutralization of IL-4 on the Th2-type pathology induced by neonatal alloimmunization (19). As expected, F1-treated mice developed splenomegaly and produced high levels of serum IgE when compared to untreated mice (Figure 1D). Both immune alterations were prevented by perinatal administration of anti-IL-4 mAb, confirming the major role of IL-4 in the development of this allo-induced, Th2-type pathology. Accordingly, lymph node cells collected from F1-treated mice produced high amounts of IL-4, IL-5 and IL-13 in response to donor-type alloantigens, a Th2-response that was significantly decreased upon early neutralization of IL-4 (Table 1).

Figure 1.

Neutralization of IL-4 inhibits transplantation acceptance, microchimerism and Th2-type pathology. (A) Survival of A/J x BALB/c skin grafts. Each group contained 9–13 mice. ***p < 0.001 compared with uninjected mice or mice injected with F1 cells and anti-IL-4 mAb. (B) Hematoxylin/eosin staining of allografts (original magnification, upper panels: ×40, bottom panels: ×200). Representative sections from untreated mice (n = 22), mice injected with F1 cells and control IgG (n = 5) or anti-IL-4 mAb (n = 15), were selected on days 8, 42 and 8 posttransplantation respectively. Data were collected from at least three individual experiments. (C) Assessment of microchimerism in spleen cells from 2-week-old mice. Groups contained 11–13 mice. (D) Measure of spleen weight and IgE serum levels of 6-week-old mice. Groups contained 18–24 mice. Data were collected from at least four individual experiments. ***p < 0.001, **p < 0.005, *p < 0.05. NS, not significant.

Table 1.  Effect of early neutralization of IL-4 on cytokine production in MLC
Neonatal treatmentAnti-BALB/c1Anti-A/J1Anti-C57BL/61
F1 SCmAbIL-42 (pg/mL)IL-52 (pg/mL)IL-132 (pg/mL)IL-42 (pg/mL)IL-52 (pg/mL)IL-132 (pg/mL)IL-42 (pg/mL)IL-52 (pg/mL)IL-132 (pg/mL)
  1. 1Lymph nodes cells from 5-week-old uninjected mice (n = 16), mice injected with F1 cells and control IgG (n = 14) or anti-IL-4 mAb (n = 10) were cocultured with either syngeneic BALB/c, donor-type (A/J x BALB/c) F1 or third-party C57BL/6 irradiated spleen cells as stimulators.

  2. 2Cytokine levels in supernatants were measured by ELISA after 72h of culture. Data were collected from at least two independent experiments.

  3. 3p < 0.001 vs. uninjected mice.

  4. 4p < 0.001 vs. mice injected with F1 cells and control IgG.

  5. 5p < 0.05 vs. mice injected with F1 cells and control IgG.

<3042 ± 7<40<3071 ± 24 176 ± 61 40 ± 7183 ± 79394 ± 136
+Control IgG<3036 ± 5<40 794 ± 1683733 ± 12033582 ± 5873 74 ± 15181 ± 64707 ± 232
+Anti-IL-4<30196 ± 98161 ± 55139 ± 424270 ± 1205 582 ± 244443 ± 5 281 ± 131258 ± 92 

In order to study the impact of IL-4 neutralization on the in vivo differentiation of Th17 cells, we first screened for the RNA expression of Th17-type master regulator and cytokines. At 2 weeks of age, expression of RORγt, IL-17A, IL-17F and IL-22 mRNA was significantly increased in the spleen of mice treated with alloantigens and anti-IL-4 mAb when compared to F1-treated mice (Figure 2A). As expected (21), IL-4 mRNA was downregulated in anti-IL-4 treated mice, with no significant alteration in IFN-γ mRNA levels. Likewise, we detected a significantly higher proportion of splenic IL-17A-producing CD4+ T cells in F1-treated mice neutralized for IL-4 compared with F1-treated mice and untreated mice (Figure 2B) with minimal effects on the proportion of IFN-γ-producing cells.

Figure 2.

Neutralization of IL-4 allows Th17 differentiation. (A) RORγt and cytokines gene expression in the spleen of 2-week-old mice. Data are expressed as relative expression against untreated mice values. Each group contained 5–10 mice. (B) Intracellular staining of splenic T cells from 2-week-old mice. Results are expressed as percentages of cytokine-producing cells among the CD4+ T cells. Each group contained at least 6 mice. ***p < 0.001, **p < 0.005, *p < 0.05.

To determine the specificity of the allorecognition of neutralized IL-4-induced Th17 cells, we performed MLC before intracellular staining of IL-17A and IFN-γ. As shown in Figure 3A, mice treated at birth with anti-IL-4 mAb and F1 cells presented a significantly higher percentage of donor-specific IL-17A-producing CD4+ T cells as compared with F1-treated mice. The proportion of IFN-γ-producing CD4+ T cells in response to donor-type or third-party stimulators was not significantly altered in both F1-treated mice compared to untreated mice. To extend these results, we injected IL-4−/− neonates with F1 cells. Lymph node cells from WT F1-treated mice displayed a donor-specific Th2-type response as assessed by high levels of IL-4 and low levels of IL-17A and IFN-γ, compared to WT untreated mice (Figure 3B). In contrast, IL-4−/− F1-treated mice presented elevated production of IL-17A and IFN-γ in response to alloantigens. Altogether, these data suggest that deprivation of IL-4 allows the development of an alloreactive Th17-type response in neonates immunized with allogeneic cells.

Figure 3.

Development of an antidonor Th17-type response in the absence of IL-4. (A) Intracellular staining of T cells cocultured with A/J x BALB/c or C57BL/6 stimulators. Results are expressed as percent values of cytokine-producing cells among CD4+ T cells from spleen of 4-week-old mice. Each group contained at least 6 mice. (B) Cytokines levels in MLC. Lymph node cells from 5-week-old WT or IL-4−/− mice were stimulated with BALB/c, A/J x BALB/c or C57BL/6 stimulators. Cytokines levels in supernatants were measured by ELISA after 72 h of culture. Each group contained at least 4 mice. **p < 0.005, *p < 0.05. NS, not significant.

Acute rejection of skin allograft in F1-treated mice neutralized for IL-4 is associated with neutrophil infiltration and local accumulation of IL-6 and IL-17A

In order to correlate the Th17 alloimmune response developing in F1-treated IL-4-deprived mice with the allograft rejection, we first characterized the allograft infiltrates and cytokines. Compared to the rejected skin grafts from untreated mice, allografts from F1-treated mice displayed very low number of neutrophils (Figure 4A). In striking contrast, skin allografts from F1-treated mice neutralized for IL-4 were massively infiltrated with neutrophils and displayed a high number of IL-17A-producing cells (Figure 4B). Interestingly, the presence of neutrophils within the allograft from control recipients was not associated with a massive presence of IL-17A-producing cells. We next performed RT-PCR assay on graft mRNA. As shown in Figure 4C, a high expression of IL-4 mRNA was predominantly observed in skin allografts of F1-treated mice. In vivo neutralization of IL-4 counteracted IL-4 mRNA expression in F1-treated mice, while leading to elevated levels of IL-17A and IL-6 mRNA. Neutrophil recruitment into the acutely rejected graft of F1-treated mice neutralized for IL-4 was therefore associated with an increased expression of Th17-type cytokines.

Figure 4.

Cellular infiltrates and cytokines mRNA in skin allografts. (A) Anti-Ly-6G immunostaining of A/J x BALB/c skin grafts (original magnification, upper panels: ×40, bottom panels: ×200). Representative sections from untreated mice (n = 18), mice injected with F1 cells and control IgG (n = 5) or anti-IL-4 mAb (n = 15), were selected on days 8, 42 and 8 posttransplantation respectively. Inlaid sections show background staining. Data are representative of at least three individual experiments. (B) Anti-IL-17A immunostaining of allografts (original magnification, upper panels: ×40, bottom panels: ×200). Representative sections from untreated mice (n = 7), mice injected with F1 cells and control IgG (n = 3) or anti-IL-4 mAb (n = 4), were selected on days 10, 42 and 10 posttransplantation respectively. (C) Cytokine mRNA expression in allografts. Data are expressed as relative expression against values obtained with syngeneic skin graft collected from uninjected BALB/c mice. Groups contained 10–19 mice. Data were collected from four independent experiments. *p < 0.05.

Allograft rejection by F1-treated mice is associated with a decreased Th2-type response and an increased Th17-type response to donor-type alloantigens

Since 42% of neonatally F1-treated mice rejected their skin graft (Figure 1A), we sought to determine if this rejection could be due to an increased induction of Th17-type response. Compared to ungrafted controls, lymph node cells from transplanted neonatally untreated mice produced similar levels of IL-17 in response to donor-type alloantigens, while they were highly primed toward a Th1-type response as assessed by IFN-γ production (Table 2). F1-treated mice accepting their graft displayed a clear Th2-type response to donor-type alloantigens, with a deficiency in Th17-type response and a lower priming of Th1-type response compared to grafted untreated mice. Conversely, we observed limited production of IL-4, elevated production of IL-17 and no significant alteration of IFN-γ production in F1-treated mice rejecting their graft compared to F1-treated mice accepting their graft. We may therefore conclude that neonatally conditioned mice that escape Th2-type tolerance increased their Th17-type response and not their Th1-type response to donor-type alloantigens.

Table 2.  Skin graft rejection by neonatally treated mice is associated with Th17-type response
Neonatal treatmentMiceGraft statusCytokine production1
IL-42 (pg/mL)IL-17A2 (pg/mL)IFN-γ2 (pg/mL)
  1. 1Graft-draining or left-side lymph node cells from BALB/c ungrafted untreated mice (n = 3), grafted neonatally untreated mice (n = 7) or grafted mice neonatally injected with (A/J x BALB/c) F1 spleen cells (n = 7) were cocultured with donor-type (A/J x BALB/c) F1 irradiated spleen cells as stimulators.

  2. 2Cytokines levels in MLC supernatants were measured after 72 h of culture.

  3. 3ND, not done.

  4. 4p < 0.05 between groups of F1-treated mice accepting their graft, and ungrafted untreated or grafted untrated mice (Unpaired t-test).

  5. 5p < 0.05 between groups of F1-treated mice rejecting their graft, and F1-treated mice accepting their graft.

  6. 6Non significant between groups of grafted untreated and ungrafted untreated mice.

  7. 7p < 0.001 between groups of grafted untreated and ungrafted untreated mice.

  8. 8p < 0.001 between groups of F1-treated mice accepting their graft, and grafted untreated mice.

  9. 9Non significant between groups of F1-treated mice rejecting their graft, and F1-treated mice accepting their graft.

F1 SC1Acceptance2077445482418

Inhibiting the Th17-type response reduces intragraft neutrophil infiltration and restores chimerism in F1-treated IL-4-deprived mice

To evaluate the functional role of the antidonor Th17-type response induced upon perinatal IL-4 neutralization, neonatally F1-treated mice neutralized for IL-4 were injected with anti-IL-17A mAb and grafted with F1 skin tissues at 6 weeks of age. As shown in Figure 5A, neutralization of IL-17 during transplantation did not delay allograft rejection in IL-4 neutralized F1-treated mice (median survival time: 18 and 22 days in F1-treated mice neutralized for IL-4 or for IL-4 and IL-17, respectively, NS). Similar results were obtained when we neutralized both IL-17A and IL-6 (data not shown). However, anti-IL-17A mAb-treated mice displayed a strong decrease in intragraft neutrophil infiltration that persisted at days 10 and 12 posttransplantation, suggesting that neutrophils were recruited in the allografts through an IL-17-dependent pathway (Figure 5B). We then evaluated the effects of neutralizing the Th17 response on the status of chimerism by injecting anti-IL-6, anti-IL-17A or anti-Gr-1 mAb at birth. As shown in Figure 6, chimerism in the spleen is mostly composed of donor B and T cells. Compared with WT F1-treated mice, WT mice having received F1 cells and anti-IL-6 mAb presented a reduced proportion of donor-type cells, donor B cells being more affected, while neutralization of IL-17 had no effect. We confirmed the decreased chimerism after IL-4 neutralization and in IL-4−/− mice. IL-6 neutralization in IL-4-deprived mice restored the level of donor-type F1 cells present in the spleen, with its major effect on B cells. Anti-IL-17A treatment also restored lymphoid chimerism. Both cytokine neutralizing treatments were significantly affecting the splenic IL-17A mRNA level of IL-4-deprived F1-cell injected mice (data not shown). Partial depletion of neutrophils with anti-Gr-1 mAb (0.36 ± 0.03% Gr-1+ cells in IL-4−/− F1-treated mice depleted for neutrophils compared to 1.07 ± 0.09% Gr-1+ cells in IL-4−/− F1-treated mice, p < 0.005) led to restored proportion of donor B cells, while it decreased the percentage of donor T cells. These data demonstrate that the Th17-type response induced in IL-4-deprived neonates inoculated with F1 spleen cells controls the B-cell chimerism and the graft neutrophil infiltration.

Figure 5.

IL-17A neutralization reduces intragraft neutrophilia. (A) Survival of A/J x BALB/c skin grafts in untreated mice (n = 29), in mice injected with anti-IL-17A mAb (n = 10), with F1 cells and control IgG (n = 23), anti-IL-4 mAb (n = 22) or anti-IL-4 and anti-IL-17A mAb (n = 10). **p < 0.005 compared with F1- and anti-IL-4-treated mice. (B) Anti-Ly-6G immunostaining of representative allografts from mice injected with F1 cells and anti-IL-4 mAb and treated with control IgG (n = 8) or anti-IL-17A mAb (n = 10), on day 8 or 10 and 12 posttransplantation. Original magnification, upper panels: ×40, bottom panels: ×200. Inlaid sections show background staining. Data are representative of two independent experiments.

Figure 6.

Inhibition of the Th17-type response restores chimerism in F1-treated IL-4−/− mice. Assessment of microchimerism in spleen of 2-week-old mice. Groups contained 3–12 mice. ***p < 0.001, **p < 0.005, *p < 0.05. ND, not done. Stars ahead of columns represent statistics between neutralized mice groups and F1-treated mice from the same strain.


Development of neonatal tolerance to donor-type cells and tissues is known to be critically dependent on IL-4 activities (4,9). The present study shows that, concomitantly with the development of a Th1-type response, deprivation of IL-4 upon F1 cells treatment at birth allows the emergence of an antidonor Th17-type response. We indeed detected higher proportions of IL-17A-producing cells in lymphoid organs and rejected grafts of F1-treated IL-4-deprived mice. This Th17 response could be detected as soon as two weeks of age, suggesting that inhibition of the neonatal Th2 response promotes Th17 differentiation.

Other pieces of evidence highlighted this cross-regulation between the Th2 and Th17 pathways. It was shown in vitro that IL-23-induced naive precursor cells differentiate into Th17 cells upon IFN-γ or IL-4 inhibition (22,23). More recently, in a model of allergic asthma, mice lacking IL-4 signaling presented an increased IL-17 production after antigen challenge, associated with a reduction of lung inflammation and Th2-type response (24). This was suggested to be due to direct RORγt downregulation by GATA-3 (25) or through Gfi-1, an IL-4/STAT-6 signaling induced transcription repressor that may suppress Th17 differentiation (26). Furthermore, in vitro polarized Th17 cells express a functional IL-13 receptor (27), as do neonatal primary Th1 cells (7). Whether IL-4 and IL-13 could negatively regulate the Th17 pathway through their binding to IL-13R on neonatal Th17 cells remains to be elucidated.

Myelin antigens exposure in early life was demonstrated to prime myelin-reactive IL-17-producing cells that induce experimental autoimmune encephalomyelitis upon immunization in adulthood (28). In agreement with our data, neonates thus seem capable of developing effector Th17 immune responses. Optimal Th17 differentiation in adults was reported to need IL-6 and regulatory T (Treg) cells derived TGF-β that inhibits the Th1 and Th2 pathways (29). In newborn mice, Treg cells controlling CD8 T cell anergy can be induced upon alloimmunization (30), suggesting the presence of a cellular source of TGF-β in our model. IL-6 could be produced by the adult immunocompetent semiallogeneic spleen cells inoculum, as well as by neonatal antigen-presenting cells or Th17 cells themselves. Accordingly, we detected an induction of IL-6 mRNA expression in rejected grafts of neonatally immunized IL-4-deprived mice that were infiltrated by IL-17-producing cells.

Multiple pathways, including allospecific Th17 responses, can lead to allograft failure and rejection. Indeed, Th17 cells mediated allograft rejection in T-bet-deficient mice (31) and in minor antigen disparities (32). In the current study, inhibition of the neonatal Th2-type polarized response induced by major MHC-mismatched allogeneic cells, elicited antidonor Th17 activities involved in the rejection of F1 cells and in the intragraft neutrophil recruitment, without significantly prolonging allograft survival. Other reports also described a partial effect of IL-17 in several transplantation settings. In graft-versus-host disease models, neutralization of IL-17 (33) or transfer of IL-17-deficient T cells (34) did not prevent mortality. In cardiac transplantation, blocking IL-17 did not prevent allograft rejection triggered by a TLR9 signaling inflammatory context (35). This suggests that other effector cells or cytokines can act with Th17 cells to induce graft or F1 cell rejection. It was previously demonstrated that neonatally primed antidonor Th1 cells can effectively mediate skin allograft rejection. Detection of antidonor IFN-γ-producing T cells upon IL-4 deprivation (Figure 3) suggests a role for the Th1 pathway as rejection mechanism. This can occur without any abrogation or decrease of chimerism (18,36) confirming that in our model, chimerism is not sufficient to promote transplantation tolerance (37). In agreement with the known positive effects of IL-6 on survival, proliferation and maturation of B cells (38,39), IL-6 neutralization at birth decreased B-cell chimerism in WT mice. On the contrary, it prevented neonatal Th17 differentiation and restored F1 cells engraftment in IL-4-deprived mice. This suggests a role of Th17 cells as effector cells controlling the chimerism. We further demonstrated a direct role for IL-17A and neutrophils in that mechanism (Figure 6). Activation of allospecific CD8+ T cells by neonatal Th17 cells could play a role here. Indeed, it was shown in tumor immunity that cytotoxic CD8+ T cells can be efficiently induced by Th17 cells (40). We may therefore consider that in our setting, Th1 cells would be sufficient to mediate skin graft rejection either by favoring cytotoxicity or by enhancing macrophage activity and that Th17 cells and cytotoxic T lymphocytes would be the major effectors promoting the deletion of donor lymphocytes.

In conclusion, we showed that the induction of effector Th17 cells in neonates is controlled by the Th2 pathway and that the allospecific Th17 cells are necessary for controlling intragraft neutrophil recruitment and allogeneic cell rejection. Since this model of neonatal tolerance associated with a donor-specific Th2-type response was confirmed in other strains of mice (41,42) and since this donor-specific Th2 bias characterizes the neonatal immunity, we therefore propose to further explore IL-4 as a major regulator of Th17 differentiation in early life.


Financial support: The Institute for Medical Immunology is sponsored by the government of the Walloon Region and GlaxoSmithKline Biologicals. This study was also supported by the Fonds National de la Recherche Scientifique (FNRS, Belgium) and an Interuniversity Attraction Pole of the Belgian Federal Science Policy. I.D. is a research fellow of the FNRS.

We thank Alain Le Moine for helpful discussion, Frédéric Paulart, Frédéric Lhommé, Nicolas Passon and Marie-Claude Lalmand for technical expertise and Philippe Horlait, Laurent Depret, Christophe Notte, Grégory Waterlot and Samuel Vander Bist for animal care.


The authors of this manuscript have no conflicts of interest to disclose as described by the American Journal of Transplantation.