Transplantation tolerance to a single noninherited MHC class I maternal alloantigen studied in a TCR-transgenic mouse model.
ABSTRACT The mechanisms underlying tolerance to noninherited maternal Ags (NIMA) are not fully understood. In this study, we designed a double-transgenic model in which all the offspring's CD8(+) T cells corresponded to a single clone recognizing the K(b) MHC class I protein. In contrast, the mother and the father of the offspring differed by the expression of a single Ag, K(b), that served as NIMA. We investigated the influence of NIMA exposure on the offspring thymic T cell selection during ontogeny and on its peripheral T cell response during adulthood. We observed that anti-K(b) thymocytes were exposed to NIMA and became activated during fetal life but were not deleted. Strikingly, adult mice exposed to NIMA accepted permanently K(b+) heart allografts despite the presence of normal levels of anti-K(b) TCR transgenic T cells. Transplant tolerance was associated with a lack of a proinflammatory alloreactive T cell response and an activation/expansion of T cells producing IL-4 and IL-10. In addition, we observed that tolerance to NIMA K(b) was abrogated via depletion of CD4(+) but not CD8(+) T cells and could be transferred to naive nonexposed mice via adoptive transfer of CD4(+)CD25(high) T cell expressing Foxp3 isolated from NIMA mice.
- Science 11/1945; 102(2651):400-1. · 31.03 Impact Factor
- [show abstract] [hide abstract]
ABSTRACT: Recent studies suggest that exposure of the fetus and newborn to non-inherited maternal major histocompatibility complex HLA antigens (NIMA) has a life-long effect on allograft recognition that could influence tolerance of organ grafts. NIMA also appear to influence disease susceptibility. Here, Jon van Rood and Frans Claas discuss evidence that three HLA haplotypes, those inherited from the parents plus NIMA, shape the immune response.Immunology Today 07/2000; 21(6):269-73.
- [show abstract] [hide abstract]
ABSTRACT: Patients who have received many transfusions become highly sensitized and develop antibodies against almost all HLA alloantigens, so that finding a cross-match negative kidney donor is difficult. A survey of those patients showed that 50 percent did not form antibodies against the noninherited maternal HLA antigens. Apart from the obvious clinical implications, the data indicate that a human equivalent of murine neonatal or actively acquired tolerance has now been identified.Science 10/1988; 241(4874):1815-7. · 31.03 Impact Factor
The Journal of Immunology
Transplantation Tolerance to a Single Noninherited MHC
Class I Maternal Alloantigen Studied in a TCR-Transgenic
Yoshinobu Akiyama,*,1,2Ste ´phane M. Caucheteux,†,‡,1,3Ce ´cile Vernochet,†,‡,4
Yoshiko Iwamoto,* Katsunori Tanaka,* Colette Kanellopoulos-Langevin,†,‡,5
and Gilles Benichou*,5
The mechanisms underlying tolerance to noninherited maternal Ags (NIMA) are not fully understood. In this study, we designed
a double-transgenic model in which all the offspring’s CD8+T cells corresponded to a single clone recognizing the KbMHC class I
protein. In contrast, the mother and the father of the offspring differed by the expression of a single Ag, Kb, that served as NIMA.
We investigated the influence of NIMA exposure on the offspring thymic T cell selection during ontogeny and on its peripheral
T cell response during adulthood. We observed that anti-Kbthymocytes were exposed to NIMA and became activated during fetal
life but were not deleted. Strikingly, adult mice exposed to NIMA accepted permanently Kb+heart allografts despite the presence
of normal levels of anti-KbTCR transgenic T cells. Transplant tolerance was associated with a lack of a proinflammatory
alloreactive T cell response and an activation/expansion of T cells producing IL-4 and IL-10. In addition, we observed that
tolerance to NIMA Kbwas abrogated via depletion of CD4+but not CD8+T cells and could be transferred to naive nonexposed
mice via adoptive transfer of CD4+CD25highT cell expressing Foxp3 isolated from NIMA mice.
2011, 186: 1442–1449.
ents that had been exposed to alloantigens during development. In
1945, seminal studies by Owen et al. (1) showed that fetal exposure
to alloantigens via vascular anastomoses led to indefinite survival
of allotransplants in bovine twins. A few years later, Billingham
The Journal of Immunology,
ransplantation tolerance, defined as the lack of donor-
specific inflammatory immunity associated with long-
term allograft survival, was initially described in recipi-
et al. (2) described the first experimental model of neonatal tol-
erance induction in rodents. It was reported that adult mice in-
jected with fully allogeneic splenocytes during fetal or neonatal
periods of life were rendered tolerant to skin grafts from the same
donor. These seminal studies demonstrated that exposure to Ags
during fetal/neonatal period of life impacts dramatically the future
offspring’s immune system.
Maternal cells and molecules, as well as microbes, traffic reg-
ularly from the mother to the fetus/neonate during pregnancy and
breast-feeding (3). This phenomenon has been implicated in the
offspring’s susceptibility to autoimmune diseases and infections,
as well as its ability to reject allogeneic transplants (4). The most
compelling evidence of the maternal influence on the offspring’s
immune system has been provided by studies evaluating the role
of noninherited maternal Ags (NIMA) in transplantation (4). It
is now well established that the transmission of NIMA during
fetal and neonatal periods of life has a long-term impact on the
alloimmune response and subsequent allotransplant rejection in
adult individuals. Van Rood et al. (4) provided initial evidence
showing the influence of NIMA exposure on humoral and cellular
alloimmunity in humans. It was observed that a large portion of
patients who produced anti-donor Abs after blood transfusion did
not form Abs to NIMA. In contrast, the same subjects consistently
mounted vigorous humoral responses to noninherited paternal Ags
(5). Later, Bean et al. (6) reported the absence of MLRs after
maternal transfusions but not paternal ones. Subsequent observa-
tions of both prolonged survival of kidney transplants from sibling
or cadaver donors and suppression of graft-versus-host (GVH)
reactions after bone marrow transplantation further confirmed the
tolerogenic effects of NIMA (7).
Although much evidence has been accumulated for the influence
of NIMA in transplant patients, few studies have addressed this
effect in experimental models. Iva ´nyi and De ´mant (8) showed
prolonged survival of maternal skin grafts in newborn rabbits.
*Department of Surgery, Massachusetts General Hospital/Harvard Medical School,
Boston, MA 02114;†Laboratoire d’Inflammation, Gestation et Autoimmunite ´ Institut
Jacques Monod, Centre National de la Recherche Scientifique, 75205 Paris Cedex 13,
France; and‡University Paris-Diderot, 75205 Paris Cedex 13, France
1Y.A. and S.M.C. contributed equally to this work.
2Current address: Department of Surgery, Tachikawa Hospital, Tokyo, Japan.
3Current address: Laboratory of Immunology, National Institute of Allergy and In-
fectious Diseases, National Institutes of Health, Bethesda, MD.
4Current address: Unite ´ Mixte de Recherche, Centre National de la Recherche Sci-
entifique, Institut Gustave Roussy, Villejuif, France.
5G.B. and C.K.-L. contributed equally to this work.
Received for publication September 8, 2010. Accepted for publication November 19,
This work was supported by the National Institute of Child Health and Human De-
velopment, National Institutes of Health (Grants RO1HD050484 and KO2AI53103 to
G.B.), the French Ministry of Education and Research (to S.M.C. and C.V.), the
Fondation pour la Recherche Me ´dicale and Ligue contre le Cancer (to S.M.C.),
and a North Atlantic Treaty Organization Science Program collaborative grant (to
G.B. and C.K.-L.).
Address correspondence and reprint requests to Drs. Gilles Benichou or Colette
Kanellopoulos-Langevin, Department of Surgery, Massachusetts General Hospital,
Thier 807, 55 Fruit Street, Boston, MA 02114 (G.B.) or Laboratory of Inflammation,
Gestation and Autoimmunity, Institute Jacques Monod, Centre National de la Re-
cherche Scientifique and University Paris-Diderot, 15 Rue He ´le `ne Brion, 75205 Paris
Cedex 13, France (C.K.-L.). E-mail addresses: firstname.lastname@example.org (G.B.) and
Abbreviations used in this article: GVH, graft-versus-host; IMA, inherited maternal
Ag; MST, mean survival time; NE, never exposed to Kb; NIMA, noninherited ma-
ternal Ag; pc, postcoitum; pp, postpartum; Tg, transgenic; Treg, regulatory T cell.
NIMA on semiallogeneic maternal skin transplants in mice. In this
model, both pregnancy and breast-feeding were required to achieve
long-term graft survival. In collaboration with Burlingham’s group
(10), we previously investigated the effects of NIMA on polyclonal
Tand B cell alloresponses and allotransplant rejection in mice. We
reported that the majority of H-2b/boffspring of semiallogeneic (H-
2b/d) mothers accept fully allogeneic DBA/2 heart grafts (graft
fibrosis, which are characteristic features of chronic rejection, were
breast-fed by a mother expressing NIMA (10). We also demon-
strated a specific influence of NIMA on the development of off-
spring’s B lymphocytes in a BCR transgenic (Tg) model, distinct
from the fate of self-reactive B cells in the same model (11, 12).
Collectively, these studies underscore the potent tolerogenic effects
of NIMA in allotransplantation. In contrast, Molitor-Dart et al. (13)
have recently reported that, under certain circumstances, the pre-
sentation of NIMA can result in offspring’s sensitization rather
than tolerization. However, the mechanisms by which NIMA ac-
tually drive the immune system toward transplant tolerance or re-
jection remain unclear. Elucidation of this question is likely to pave
the way for the design of novel tolerance protocols in clinical
In this study, we used a model in which a single NIMA is the
MHC class I H-2 Kbmolecule in a Kb-Tg mouse and the offspring
express an anti-KbTCR transgene on CD8+T cells. We observed
that the fetus’s anti-KbTCR Tg thymic T cells were exposed and
activated to NIMA during pregnancy and neonatal life up to 3 wk
of age, leading to the deletion of half of T cells during this period.
The adult offspring displayed long-term survival of NIMA Kb-
expressing heart allotransplants. Tolerance to NIMAwas mediated
via the suppression of the proinflammatory response by anti-Kb
CD8+T cells and the activation/expansion of CD4+CD25high
Foxp3+regulatory T cells (Tregs) recognizing the Kballoantigen.
The implications of these findings for the design of tolerance
protocols in allotransplantation are discussed.
Materials and Methods
Mice and transplantations
Mice were bred and maintained at Massachusetts General Hospital and
Institut Jacques Monod’s animal facilities under specific pathogen-free
conditions. All animal care and handling were performed according to
institutional guidelines. The day of the vaginal plug was considered as day
0.5 of gestation. CBK Tg (CBA/ca mice [H-2k] expressing a KbMHC class
I transgene) were used as donors in heart transplants (14). Offspring of
BM3.3 anti-KbTCR Tg male mice (15) and F1 (CBA/ca 3 CBK) females
were used as NIMA (offspring that do not inherit Kb) and IMA (offspring
that inherit Kbmaternal Ag) recipient mice (Fig. 1). To separate NIMA and
IMA offspring, we stained blood from orbital sinus with an anti-KbFITC
mAb. No cells from NIMA were Kb+, whereas all the cells were found to
express Kbin IMA mice. Offspring of BM3.3 anti-KbTCR Tg male and
CBK female mice were used as positive control animals for tolerance to Kb
(Kbinherited as self: IMA). Offspring of BM3.3 anti-KbTCR Tg male and
wild type CBA female mice (never exposed to Kb: NE) were used as
negative control animals (i.e., lack of tolerance to Kb). NIMA, IMA, and
NE mice were transplanted in the peritoneal cavity with a vascularized
CBK (Kb-Tg) heart using the microsurgical technique previously described
by Corry et al. (16). Graft rejection was monitored by daily palpation of
heart and confirmed by histological techniques. In some experiments,
CD4+or CD8+T cells were depleted from recipient mice with anti-CD4
(GK1.5) and anti-CD8 (53.6.72) mAbs (1 mg given i.p. at days 23 and 21
Cells were isolated from thymus and spleen of individual fetuses at
18.5 d postcoitum (pc) and neonates at 3.5 d postpartum (pp). Organs were
gently pressed through a sieve using a syringe plunger and suspended in
PBS containing 4% FCS and 0.1% sodium azide (PBS/FCS/NaN3). Viable
cells were counted by trypan blue exclusion.
Immunofluorescence staining and flow cytometric analyses
Aliquots of 5 3 105nucleated cells were incubated for 40 min at 4˚C with
an optimal amount of the following mAbs: anti-CD8 coupled to FITC,
NIMA and IMA mice, we mated Bm3.3
anti-KbTCR Tg male mice with (CBK
[CBA, KbTg] 3 CBA) F1 female mice.
The offspring, which inherited both
the anti-KbTCR transgene and the Kb
transgene, were referred to as IMA mice.
The offspring, which inherited the anti-
KbTCR transgene but not the Kbtrans-
gene, were referred to as NIMA mice.
NE control offspring were obtained by
mating BM3.3 anti-KbTCR Tg female
mice with CBA male mice (these mice
could never be exposed to Kb). B, The
expression of MHC class I glycoprotein
Kbwas assessed in positive control mice
IMA and CBK, negative control mice
NE, and experimental NIMA mice. Rep-
resentative FACS profiles obtained with
spleen cells are shown.
The model. A, To obtain
The Journal of Immunology1443
anti-CD4 coupled to PE, anti-CD25 coupled to FITC, anti-CD44 coupled
to PE, anti-TCR b-chain coupled to PE (all purchased from Pharmingen),
and anti-BM3.3 clonotype Ti98 prepared according to conventional tech-
niques and coupled to biotin. Cells were washed twice in PBS/FCS/NaN3.
Biotinylated Ab was revealed by incubating cells for 20 min at 4˚C with
streptavidin-PE or streptavidin-allophycocyanin. After washing, cells were
analyzed on a CyAn LX flow cytometer (DakoCytomation) equipped with
488- and 635-nm lasers. The cell populations analyzed were gated on the
viable lymphoid cell population on the basis of forward and side scatter
criteria. When possible, at least 104Ti98+cells were analyzed from each
T cell assays
The deletion of anti-KbTCR Tg T cells was monitored with an anti-
clonotypic mAb (Ti.98) using FACS analysis. The frequencies of type 1
and 2 cytokine-producing T cells responding to Kbvia the direct allor-
ecognition pathway were determined using an ELISPOT method as pre-
viously described (4).
Cardiac transplants were fixed in 10% buffered formalin, embedded in
paraffin, coronally sectioned, and stained with H&E for evaluation of
cellular infiltrates and myocyte damage (acute rejection) by light micros-
copy. For assessment of chronic rejection, cardiac grafts were stained with
Verhoeff’s elastin (vessel arteriosclerosis scoring) or Mason’s trichrome
(evaluation of fibrosis). Arteriosclerosis was assessed by light microscopy,
and the percentage of luminal occlusion and intimal thickening was
determined using a scoring system, as previously described (17). Only
vessels that display a clear internal elastic lamina were included in mor-
phometric analysis (five to seven vessels per section). All arteries were
scored by at least two examiners in a blinded fashion.
Statistical analyses were performed using STATView software (Abacus
Concepts, Berkeley, CA). The p values were calculated using paired t test.
A p value ,0.05 was considered statistically significant.
In this study, we used a mouse Tg model in which the NIMA is an
MHC class I transgene, Kb, and all the offspring’s T cells express
an anti-KbTCR transgene (BM3.3 Tg mice; Fig. 1). The TCR
from CD8+T cells of BM3.3 mice recognize intact KbMHC class
I molecules bound to an 8-mer peptide (INFDFNTI) called BM1
(direct allorecognition). Both CBK (KbTg) and Bm3.3 (anti-Kb
TCR Tg) mice used as parents were engineered in CBA/Ca (H-2k)
mice. To study the NIMA effect, we mated heterozygous female
mice (CBA 3 CBK) F1 with homozygous Bm3.3 TCR Tg anti-Kb
male mice. In this setting, all the offspring inherit the TCR Tg
from their father and express anti-KbTCR on their CD8+T cells.
Half the offspring are expected to inherit the Kbtransgene from
their mother and are referred to as IMA mice. The other half of the
offspring should not inherit Kbfrom their mother; these mice are
called NIMA mice. In addition, non-Tg CBA females were mated
with Bm3.3 anti-KbTCR Tg males. In the absence of a Kb
transgene in the mother, the resulting offspring do not inherit Kb
and are, therefore, never exposed to it and are referred to as NE
mice. The design of NIMA, IMA, and NE mice is depicted in Fig.
1A. The phenotype of NE, NIMA, and IMA offspring was
ascertained by staining splenocytes using anti-KbAbs. As ex-
pected, the splenocytes of CBK and IMA mice expressed MHC
class I Kbmolecules, whereas the spleen cells from NE and NIMA
mice did not display Kbon their surface (Fig. 1B).
Next, we investigated the influence of NIMA on allotransplant
rejection by the offspring. To test this, we transplanted NIMA mice
with a CBK (KbTg) allogeneic heart. Acute and chronic rejections
of the cardiac allografts were monitored by palpation and his-
tological techniques. In these experiments, NE mice, which are
never exposed to Kb, and IMA mice that inherit Kbfrom their
mothers were used as control recipients for rejection and toler-
ance, respectively. As expected, NE mice rejected CBK heart
transplants in an acute fashion (12 6 4 d; Fig. 2A), whereas IMA
mice accepted their transplants indefinitely (Fig. 2A). Twenty-two
of the 25 NIMA mice tested (.80%) accepted CBK heart trans-
plants permanently. Three mice rejected their transplants, al-
though in a markedly delayed fashion (40–60 d). As shown in Fig.
2B, massive infiltration and tissue damage were detected in the
heart transplants of control NE mice tested 12 d after grafting.
In contrast, in NIMA mice, histological examination of cardiac
transplants performed 50 d after allograft placement revealed no
inflammatory cell infiltrates and a well-preserved tissue architec-
ture (Fig. 2B). Therefore, NIMA mice are tolerant to Kb+allo-
geneic CBK heart transplants. This suggests that although NIMA
mice did not inherit KbMHC class I Ag from their mothers, they
had been exposed to this allo-MHC Ag during their development.
It was possible that tolerance to Kballografts in NIMA mice was
due to the deletion of anti-KbTCR Tg T cells during thymic se-
lection. To test this, we assessed the presence of TCR Tg T cells
in the peripheral blood of adult mice by FACS using an anti-
clonotypic mAb, Ti98. Control non-TCR Tg CBA (NE) and
CBK mice displayed a normal polyclonal population of CD8+
IMA (express Kb), and NIMA (experimental group) mice underwent het-
and histological methods. A, Percentages of graft survival at different time
points after transplantation. Graft survival was analyzed using the Kaplan–
Meiermethod, and survival curveswere compared usingthe log-rank test.B,
Histology of BALB/c heart transplants from control rejecting NE and toler-
ant NIMA mice. Microphotographs (H&E, original magnification 340) are
representative of four NE and NIMA mice tested individually.
1444TOLERANCE TO NONINHERITED MATERNAL MHC CLASS I Ags
T cells, which did not express the Ti98 clonotype (Fig. 3A),
whereas virtually all the CD8+T cells found in Bm3.3 TCR Tg
mice were Ti98+(Fig. 3B). Strikingly, no Ti98+were detected in
Bm3.3 mice, which had inherited Kbfrom their mothers (IMA)
(Fig. 3C). Therefore, NE mice did not delete their anti-KbTCRTg
T cells, whereas IMA mice in which Kbrepresents a self-antigen
deleted their anti-KbTCR Tg Ti98+T cells. Most important,
normal levels of Ti98+T cells similar to those observed in control
BM3.3 NE mice were found in NIMA mice (Fig. 3D). Therefore,
deletion of anti-KbTCR Tg CD8+T cells is not responsible for Kb-
specific tolerance in adult NIMA mice.
Next, we examined whether T cells from NIMA mice are ex-
posed to Kballoantigen during thymic development. The thymi of
NE, IMA, and NIMA mice were collected during fetal life at days
16.5 and 18.5 postcoitum (pc) (fetuses), at birth time, and 3.5 d pp
(neonates). As shown in Fig. 4A, a few T cells were detected at
day 16.5 pc in all mouse groups, but numbers were already re-
producibly decreased in NIMA fetuses. At day 18.5 pc and birth
time, control NE mice displayed high numbers of Kb-specific
thymocytes, whereas none was found in IMA mice. This is con-
sistent with a model in which BM3.3 T cells expanded in NE mice
(positive selection), whereas they were deleted in IMA mice
(negative selection). At this time point, TCR Tg T cells were
detected in NIMA mice, although at half the frequency found in
control NE mice. In turn, at day 3.5 pp, the number of anti-Kb
T cells had doubled in NIMA mice but remained significantly
lower than that observed in NE mice (Fig. 4A), and the same
observation could be made at 3 wk pp (data not shown). Virtually
no anti-KbTCRTg T cells were found at each of these time points
in IMA mice, a result consistent with their clonal deletion (Fig.
4A). Altogether, these results indicate that NIMA mice are ex-
posed to and affected by NIMA Kballo-MHC class I Ag during
fetal life. To confirm this, thymocytes from NE and NIMA mice
were permeabilized and tested at day 18.5 pc and 3.5 pp for their
proliferation rate using propidium iodide. The FACS profiles
presented in Fig. 4B show the frequencies of T cells in G0-G1
phase (left peak) and in S-M/G2phase (right peak). The results
show a marked increase in the proliferation rates in NIMA mice as
compared with NE mice in both fetuses (4 versus 10%; p , 0.05)
and neonates (18 versus 40%; p , 0.05). This observation further
supports the view that, in NIMA mice, T cells are exposed to
NIMA and activated to proliferate during fetal thymic de-
velopment. This phenomenon occurred up to weaning age at the
end of the transfer of maternal cells through suckling.
In another set of experiments, we compared anti-KbT cell-
mediated alloresponses in adult NE, IMA, and NIMA mice (Fig.
5). T cells from the spleen of naive mice and mice transplanted
with a CBK heart were isolated and placed in culture with allo-
geneic irradiated CBK stimulator cells (MLR), a test that is tra-
ditionally used to detect direct alloreactivity. The frequencies of
anti-Kballoreactive T cells producing type 1 (IL-2 and IFN-g) and
type 2 (IL-4 and IL-10) cytokines were measured using an ELI-
SPOT assay as previously described (18). In naive control NE
mice, ∼200 activated T cells per million T cells were found to
produce IL-2, IFN-g, and IL-4, but no IL-10, which is consistent
deleted from the thymus. Thymus T cells from NE or NIMA mice were
analyzed by flow cytometry, using the Ti98 mAb specific for the BM3.3 Tg
TCR. A, Compilation of cell numbers (6 SEM) of BM3.3 Tg thymic
T cells from NE control, NIMA, or IMA animals as 16.5- and 18.5-d pc
fetuses, newborns, or 3.5-d-old neonates; at least seven animals were
studied in each group in three separate experiments. *p , 0.05; **p ,
0.01. B, Flow cytometry profiles obtained from NE or NIMA BM3.3 Tg
thymic T cells from 18.5-d pc fetuses and 3.5-d-old neonates, after per-
meabilization and staining with propidium iodide (PI). Percentages of cells
in G0/G1or S/G2/M phases of the cell cycle are given in each graph. One
experiment representative of at least three for each group is shown.
Fetal and neonatal NIMA-exposed T cells are partially
in IMA and NIMA mice. The expression of the anti-KbTCR Tg protein on
the surface of peripheral blood CD3+CD8+T cells was assessed by FACS
analysis using the anti-clonotypic mAb, Ti98. Representative FACS pro-
files obtained with control non-Tg CBA mice (A), control anti-KbTCR Tg
Bm3.3 mice (B), IMA offspring that inherited both the anti-KbTCR and
the Kbprotein (C), and NIMA mice that inherited the anti-KbTCR
transgene but not the Kbtransgene (D). The results are representative of
.100 mice in each group.
Frequencies of T cells expressing the anti-KbTCR protein
The Journal of Immunology1445
with the frequencies previously reported for a primary MLR. As
expected, these frequencies were much greater (.1000 spots/
million) in NE mice that had been transplanted with a CBK
heart, whereas no T cells producing IL-10 were detected in these
mice. In contrast, no activated T cells producing IL-2, IFN-g, and
IL-4 were detected in both naive and transplanted IMA mice.
Interestingly, however, a few T cells producing IL-10 were found
in naive and transplanted IMA mice. In NIMA mice, although no
alloreactive T cells producing IL-2 and IFN-g were found, some
T cells producing IL-4 were detected. In addition, IL-10–secreting
anti-KbT cells were detected in naive NIMA mice and particularly
in NIMA mice that had received a cardiac allograft. These cyto-
kines are traditionally secreted by type 2 (TH2/CT2) cells and
Tregs. Our results also imply that, although most Ti98+TCR Tg
T cells had been deleted in developing IMA mice, some anti-Kb
T cells producing IL-10 had escaped negative selection and could
become activated in adults after exposure to Kballoantigen.
Next, we investigated the mechanisms underlying transplan-
tation tolerance in adult NIMA mice. First, NE, NIMA, and IMA
mice were treated with depleting anti-CD4 or anti-CD8 Abs
starting 3 d before transplantation with an allogeneic Kb+CBK
heart. This resulted in the near-complete depletion of CD4+and
CD8+T cell subsets for more than 2 wk after Ab administration
(data not shown). As shown in Fig. 6A, the depletion of CD8+
T cells resulted in long-term survival of Kb+heart transplants in
NE mice, whereas the anti-CD4 mAb treatment had no effect. This
is consistent with the fact that CBK allografts placed in these
BM3.3 Tg mice are rejected primarily by CD8+TCR Tg anti-Kb
T cells. Although some CD4+T cells can be found in these mice,
they apparently do not contribute to the rejection of Kb+allo-
grafts. The majority (.80%) of nontreated NIMA mice either
accepted Kb+allotransplants or exhibited marked delayed re-
jection (.60 d post-transplantation; Fig. 6B). All NIMA mice
treated with anti-CD8 Abs retained CBK cardiac transplants in-
definitely, a result that is consistent with the observations made
in NE mice. Most important, the majority of NIMA mice treated
with anti-CD4 mAbs rejected CBK hearts between 10 and 30 d
post-transplantation (Fig. 6B). Histological examination of the re-
jected transplants revealed massive inflammatory infiltrates and
tissue damage typical of acute cellular rejection (data not shown).
Therefore, depletion of CD4+T cells in NIMA mice had abolished
tolerance to Kballoantigen. Surprisingly, we observed that de-
pletion of CD4+T cells in IMA mice induced the rejection of
CBK heart transplants in .50% of the mice. Therefore, tolerance
to Kbcan be broken in IMA mice, a result suggesting that clonal
deletion of anti-KbT cells is not the sole mechanism underlying
tolerance induction and/or maintenance in these mice (data not
The results obtained in NIMA mice with anti-CD4 mAbs
prompted us to test whether these mice display CD4+Tregs re-
sponsible for inducing or maintaining tolerance to Kballoantigen.
To address this question, we isolated CD4+CD25highand CD4+
CD252T cells from the spleens of NIMA mice by FACS sorting
(.91% purity). More than 92% of CD4+CD25highT cells were
Foxp3+, whereas the CD4+CD252T cells did not express signif-
icant Foxp3 levels (data not shown). Each subpopulation was
adoptively transferred (5 3 105cells given i.v.) into CBA naive
mice 5 d before their transplantation with a CBK heart. As shown
in Fig. 7, the mice administered with CD4+CD252T cells from
NIMA mice rejected CBK cardiac allografts in an acute fashion
producing T cells in naive and transplanted
mice. The frequencies of alloreactive anti-
KbT cells secreting proinflammatory type
1 cytokines IL-2 (A), IFN-g (B), and type 2
“regulatory” cytokines IL-4 (C) and IL-10
(D) were measured by ELISPOT. Spleen
cells from nontransplanted (naive, white
bars) and mice recipient of a CBK heart
transplant (10 d post-transplant, solid bars)
were collected and stimulated in vitro with
irradiated CBK KbTg splenocytes (MLR).
The results are presented as cytokine-
producing spots per million T cells 6 SD.
The results are representative of four ex-
periments each including two to three
mice tested individually.
Frequencies of cytokine-
1446TOLERANCE TO NONINHERITED MATERNAL MHC CLASS I Ags
(mean survival time [MST]: 12 6 2 d). In contrast, adoptive
transfer of five CBA mice with CD4+CD25highT cells collected
from NIMA mice resulted in a significant increase (p , 0.02) of
allograft survival in four of five mice (MST: 89 6 18 d). Histo-
logical examination of the transplanted hearts revealed no signs of
chronic allograft vasculopathy in these mice (data not shown). It is
noteworthy that these adoptively transferred mice rejected acutely
third-party BALB/c (H-2d) cardiac allografts (data not shown). In
contrast, adoptive transfer of CD4+CD252T cells, as well as
CD4+CD25highT cells from NE mice, had no effect on graft re-
jection. Therefore, the tolerance to Kbcan be adoptively trans-
ferred to CBA NE mice using CD4+CD25highT cells derived from
the spleens of NIMA mice. In NIMA mice, anti-KbTCR Tg
T cells are not deleted and some Kb-specific CD4+Tregs may
be selected, which can ensure tolerance to Kballografts. Inter-
estingly, some modest but significant prolongation of allograft
survival was also observed on transfer of CD4+CD25highfrom
IMA mice (MST: 28 6 5 d; p = 0.04). This result further supports
the view that, in IMA mice, although CD8+anti-KbT cells are
eliminated in the developing thymus, some CD4+Tregs escape
negative selection and can confer some protection against re-
jection of Kb+allografts following adoptive transfer in naive CBA
Transplantation tolerance, defined broadly as long-term allograft
survival in the absence of immunosuppressive treatment, is regu-
larly achieved in nature during pregnancy. Mammalian pregnancy
and subsequent nursing of the newborn appears to have a profound
influence on the neonate’s developing immune system that is re-
tained in adulthood. In animal models, the passage of maternal
cells and Ags during gestation and breast-feeding is thought
to imprint long-term unresponsiveness of NIMA-specific inflam-
matory T cells in offspring. This phenomenon is clinically relevant
as exemplified by the beneficial effects of matching donors and
recipients for NIMA in human recipients of blood transfusion and
kidney allotransplants (19). In addition, there is a body of evi-
dence suggesting that the presence of NIMA also influences the
adult’s susceptibility to autoimmune disorders (4, 20–22). Alto-
gether, these observations indicate that NIMA play a critical role
in the establishment and regulation of the entire immune system.
However, the mechanisms underlying the induction of a NIMA
effect in the fetus and neonates and its maintenance in adults are
not fully understood. Gaining insights into this question will set
the path for the design of novel strategies for manipulating the
immune system in health and disease.
The elucidation of the mechanisms underlying the NIMA effect
has been difficult because of the fact that the precise nature of the
NIMA and the T cell clones recognizing these maternal Ags are
unknown. To overcome this, we designed a double-Tg model in
which the mother’s NIMA and offspring’s anti-NIMAT cells were
cardiac allotransplants. NE (A) or NIMA (B) mice were treated with de-
pleting anti-CD4 (GK1.5) or anti-CD8 (53.6.72) mAbs administered i.p.
5 d before transplantation with a heart derived from a KbTg CBK mouse.
Control mice that received no Abs (no treatment) were also transplanted.
Acute rejection of allografts was monitored for 100 d. The results are
expressed as percentages of graft survival obtained with 5–12 mice in each
group. Graft survival was analyzed using the Kaplan–Meier method, and
survival curves were compared using the log-rank test.
Effects of CD4+or CD8+T cell depletion on tolerance to Kb
tolerance to Kb+allografts. CD4+CD252and CD4+CD25+T cells were
purified from either NE (circles), IMA (squares), or NIMA (triangles)
mice. Each T cell subset (5 3 105cells) was separately injected i.v. into
naive CBA mice, which received a CBK heart transplant 3 d later. Control
mice, which received no T cells, were also tested (diamonds). The effects
of adoptive transfer of CD4+CD252(open symbols) and CD4+CD25+
(solid symbols) T cells on graft survival were monitored. Each point
corresponds to the survival data from a single mouse. Horizontal bars
represent the MST (days).
CD4+CD25+Foxp3+T cells from NIMA mice can transfer
The Journal of Immunology1447
well defined. In this model, all the offspring’s CD8+T cells cor-
responded to a single clone recognizing the KbMHC class I
protein. In contrast, the mother and the father of the offspring
differed by the expression of a single Ag, Kb, that served as NIMA.
This allowed us to study the influence of NIMA exposure on the
offspring T cell repertoire selection during ontogeny and on its
T cell response during adulthood. First, we showed that adult
NIMA mice were tolerant to Kbas they accepted Kb+heart allo-
transplants permanently. This implies that these mice have been
exposed to NIMA Kbpresumably during pregnancy or breast-
feeding, or both. It is not clear whether this results from the
transplacental passage of Kb+maternal cells or soluble Kbmole-
cules, or both. Several studies have documented the passage of
hematopoietic maternal cells from the mother to the fetus (3, 23).
Among them, T lymphocytes are regularly detected in umbilical
cord blood samples from neonates. The presence of maternal
T cellsis commonlyobservedinSCIDpatients (24–33).Astudyby
Kobayashi et al. (34) has documented that maternal CD4+T cells
are present in various tissues of a male infant with a SCID phe-
notype resulting from Artemis gene mutation. In a murine model
blastocysts to pseudopregnant female animals, Piotrowski et al.
(35) have demonstrated that in 90% of scid/scid fetuses, nucleated
maternal cells were present in at least one lymphoid organ. In an-
other study using GFP Tg female mice, Zhou et al. (36) have
reported the presence of GFP+maternal cells in fetal organs in-
cluding the thymus, spleen, and liver. In addition, a recent study by
Dutta et al. (37) demonstrates the presence of maternal hemato-
poietic microchimerism in lymphoid but also nonlymphoid organs,
with predominance in the heart.
In this study, we showed that anti-KbTCR Tg CD8+T cells pre-
sent in the fetal and neonatal thymi of NIMA offspring display an
activated phenotype. In addition, NIMA exposure is associated
with a lower frequency of Kb-specific T cells in the developing
thymus of NIMA mice compared with control CBA (NE) mice.
This suggests that from day 16.5 pc through the time of birth and
up to 3 wk pp, the presence of NIMA was associated with either
the deletion of some developing anti-KbT cells or an inefficient
positive selection of these T cells. Unexpectedly, our results also
show that thymocytes from the NIMA fetuses display a greater
rate of proliferation while they are present at a lower frequency
than their NE counterparts. The observation that TCR Tg de-
veloping thymic T cells from NIMA mice display a greater pro-
liferation rate than those of NE mice support a partial deletional
model rather than a lack of positive selection. Most important, the
presence of normal frequencies of anti-KbT cells in adult NIMA
mice demonstrates that tolerance to Kbis not ensured only via
deletion of anti-KbT cells during thymic development.
Functional analysis of anti-Kballoreactive T cells in NIMA
mice revealed the absence of proinflammatory T cells producing
IL-2 and IFN-g. In turn, although these mice were tolerant to CBK
allografts, they displayed some T cells producing IL-4 and high
numbers of IL-10–producing T cells when challenged with Kb+
allostimulators, that is, through the direct allorecognition pathway.
Therefore, exposure of fetuses or neonates, or both, to NIMA
resulted in the selection of T cells producing type 2 cytokines.
Notably, the majority of the T cells producing IL-4 and IL-10 on
stimulation with Kb+allogeneic cells displayed a CD4+pheno-
type. Indeed, the BM3.3 Tg mice used in this study were not bred
on a RAG knockout background and displayed low but significant
numbers of CD4+T cells that were not Tg T cells (Ti982). This
suggested that, in NIMA mice, anti-KbCD8+Tg T cells could not
reject CBK allografts because they were suppressed by CD4+
T cells. This was confirmed by the observation that depletion of
CD4+T cells in NIMA mice restored their ability to reject CBK
cardiac allografts. The presence of allospecific, IL-4–producing,
CD4+T cells in NIMA is consistent with the concept that neonatal
tolerance is associated with activation of Th2 cells. In support of
this, Fortshuber et al. have previously reported that neonatal tol-
erance is mediated via the positive selection of Th2 cells during
development (38). Alternatively, it is possible that Tregs ensured
tolerance to NIMA. Indeed, we showed that tolerance to Kbcould
be transferred to control NE mice by injection of CD4+CD25+
Foxp3+T cells collected from the spleens of NIMA mice. The
majority of these tolerogenic CD4+T cells secreted IL-10 and was
donor-specific in that they did not suppress the rejection of third-
party allografts (data not shown). Therefore, these Tregs are likely
to correspond to inducible regulatory Tr1 cells rather than natural
Tregs (39). This conclusion corroborates the results reported by
others showing the presence of CD4+CD25+Foxp3+Tregs secreting
IL-10 and TGF-b in the lymph nodes of mice transplanted with
a NIMA+allograft (40–42). Further supporting this view, Mold
et al. (43) have recently reported the presence of such Tregs in
human fetal lymph nodes. Our study using adoptive transfer ex-
periments demonstrates that these regulatory cells can be isolated
and mediate allotransplant tolerance in vivo. This does not, how-
ever, exclude that some other cells, including CD8+Tregs, and/or
mechanisms can contribute to NIMA tolerance in this and other
The revelation of a powerful and beneficial NIMA effect in our
Tg transplant model fully confirms and extends the original report
of Owen et al. (44) regarding a tolerogenic effect of alloantigen
pre-exposure on humoral immunity in adults. It is not clear why
maternal chimerism and subsequent T cell tolerance to NIMA has
been selected through evolution in mammals. It can be speculated
that this process prevents the fetus’s immune system from at-
tacking the mother as observed in GVH reactions. However, it
seems unlikely that a few fetal T cells that are hyporesponsive to
allostimulation could induce a life-threatening GVH-like disease
in the mother. Alternatively, the passage of maternal leukocytes
might be useful to protect the fetus against pathogens and/or
contribute to its proper immune development and maturation.
There are implications of the NIMA effect in pediatric transplan-
tation where a child receives a kidney from his or her mother. In
this setting, pretransplant maternal transfusion may reactivate and
expand NIMA-specific Tregs, thereby amplifying the NIMA tol-
erogenic effect and ensuring tolerance to the transplant. The im-
plications of the NIMA effect for a variety of other applications
are numerous and include cord blood stem cell transplantation
(45) and cadaveric organ transplantation, as well as nontransplant
fields such as autoimmunity and development of antitumor vac-
cination approaches using “self” antigenic peptides, both of which
may benefit from an understanding of the basic mechanisms of
We thank Dr A. Guimezanes (Centre d’Immunologie Marseille-Luminy,
Marseille, France) for providing CBK and BM3.3 Tg mice and Ti98 mAb.
The authors have no financial conflicts of interest.
1. Owen, R. D. 1945. Immunogenetic consequences of vascular anastomoses be-
tween bovine twins. Science 102: 400–401.
2. Billingham, R. E., L. Brent, and P. B. Medawar. 1953. Actively acquired tol-
erance of foreign cells. Nature 172: 603–606.
1448TOLERANCE TO NONINHERITED MATERNAL MHC CLASS I Ags
3. Vernochet, C., S. M. Caucheteux, and C. Kanellopoulos-Langevin. 2007. Bi-
directional cell trafficking between mother and fetus in mouse placenta. Pla-
centa 28: 639–649.
4. van Rood, J. J., and F. Claas. 2000. Both self and non-inherited maternal HLA
antigens influence the immune response. Immunol. Today 21: 269–273.
5. Claas, F. H., Y. Gijbels, J. van der Velden-de Munck, and J. J. van Rood. 1988.
Induction of B cell unresponsiveness to noninherited maternal HLA antigens
during fetal life. Science 241: 1815–1817.
6. Bean, M. A., E. Mickelson, J. Yanagida, S. Ishioka, G. E. Brannen, and
J. A. Hansen. 1990. Suppressed antidonor MLC responses in renal transplant
candidates conditioned with donor-specific transfusions that carry the recipient’s
noninherited maternal HLA haplotype. Transplantation 49: 382–386.
7. van Rood, J. J., F. R. Loberiza, Jr., M. J. Zhang, M. Oudshoorn, F. Claas,
M. S. Cairo, R. E. Champlin, R. P. Gale, O. Ringde ´n, J. M. Hows, and
M. H. Horowitz. 2002. Effect of tolerance to noninherited maternal antigens on
the occurrence of graft-versus-host disease after bone marrow transplantation
from a parent or an HLA-haploidentical sibling. Blood 99: 1572–1577.
8. Iva ´nyi, P., and P. De ´mant. 1965. Prolonged survival of maternal skin grafts in
newborn rabbits. Folia Biol. (Praha) 11: 321–323.
9. Zhang, L., and R. G. Miller. 1993. The correlation of prolonged survival of
maternal skin grafts with the presence of naturally transferred maternal T cells.
Transplantation 56: 918–921.
10. Andrassy, J., S. Kusaka, E. Jankowska-Gan, J. R. Torrealba, L. D. Haynes,
B. R. Marthaler, R. C. Tam, B. M. Illigens, N. Anosova, G. Benichou, and
W. J. Burlingham. 2003. Tolerance to noninherited maternal MHC antigens in
mice. J. Immunol. 171: 5554–5561.
11. Caucheteux, S. M., C. Vernochet, J. Wantyghem, M. C. Gendron, and
C. Kanellopoulos-Langevin. 2008. Tolerance induction to self-MHC antigens in
fetal and neonatal mouse B cells. Int. Immunol. 20: 11–20.
12. Vernochet, C., S. M. Caucheteux, M. C. Gendron, J. Wantyghem, and
C. Kanellopoulos-Langevin. 2005. Affinity-dependent alterations of mouse
B cell development by noninherited maternal antigen. Biol. Reprod. 72: 460–
13. Molitor-Dart, M. L., J. Andrassy, L. D. Haynes, and W. J. Burlingham. 2008.
Tolerance induction or sensitization in mice exposed to noninherited maternal
antigens (NIMA). Am. J. Transplant. 8: 2307–2315.
14. Tarazona, R., A. M. Sponaas, G. Mavria, M. Zhou, R. Schulz, P. Tomlinson,
J. Antoniou, and A. L. Mellor. 1996. Effects of different antigenic micro-
environments on the course of CD8+ T cell responses in vivo. Int. Immunol. 8:
15. Auphan, N., J. Curnow, A. Guimezanes, C. Langlet, B. Malissen, A. Mellor, and
A. M. Schmitt-Verhulst. 1994. The degree of CD8 dependence of cytolytic T cell
precursors is determined by the nature of the T cell receptor (TCR) and influ-
ences negative selection in TCR-transgenic mice. Eur. J. Immunol. 24: 1572–
16. Corry, R. J., H. J. Winn, and P. S. Russell. 1973. Primarily vascularized allografts
of hearts in mice. The role of H-2D, H-2K, and non-H-2 antigens in rejection.
Transplantation 16: 343–350.
17. Russell, M. E., W. W. Hancock, E. Akalin, A. F. Wallace, T. Glysing-Jensen,
T. A. Willett, and M. H. Sayegh. 1996. Chronic cardiac rejection in the LEW to
F344 rat model. Blockade of CD28-B7 costimulation by CTLA4Ig modulates
T cell and macrophage activation and attenuates arteriosclerosis. J. Clin. Invest.
18. Benichou, G., A. Valujskikh, and P. S. Heeger. 1999. Contributions of direct and
indirect T cell alloreactivity during allograft rejection in mice. J. Immunol. 162:
19. Burlingham, W. J., A. P. Grailer, D. M. Heisey, F. H. J. Claas, D. Norman,
T. Mohanakumar, D. C. Brennan, H. de Fijter, T. van Gelder, J. D. Pirsch, et al.
1998. The effect of tolerance to noninherited maternal HLA antigens on the
survival of renal transplants from sibling donors. N. Engl. J. Med. 339: 1657–
20. Bianchi, D. W. 2000. Fetal cells in the mother: from genetic diagnosis to diseases
associated with fetal cell microchimerism. Eur. J. Obstet. Gynecol. Reprod. Biol.
21. Nelson, J. L. 1998. Pregnancy immunology and autoimmune disease. J. Reprod.
Med. 43: 335–340.
22. Nelson, J. L. 2003. Microchimerism in human health and disease. Autoimmunity
23. Maurel, M. C., and C. Kanellopoulos-Langevin. 2008. Heredity—venturing
beyond genetics. Biol. Reprod. 79: 2–8.
24. O’Reilly, R. J., J. H. Patterson, F. H. Bach, M. L. Bach, R. Hong, F. Kissmeyer-
Nielsen, and A. J. Therkelsen. 1973. Chimerism detected by HL-A typing.
Transplantation 15: 505–507.
25. Pollack, M. S., N. Kapoor, M. Sorell, S. J. Kim, F. T. Christiansen, D. M. Silver,
B. Dupont, and R. J. O’Reilly. 1980. DR-positive maternal engrafted T cells in
a severe combined immunodeficiency patient without graft-versus-host disease.
Transplantation 30: 331–334.
26. Pollack, M. S., D. Kirkpatrick, N. Kapoor, B. Dupont, and R. J. O’Reilly. 1982.
Identification by HLA typing of intrauterine-derived maternal T cells in four
patients with severe combined immunodeficiency. N. Engl. J. Med. 307: 662–
27. Flomenberg, N., B. Dupont, R. J. O’Reilly, A. Hayward, and M. S. Pollack.
1983. The use of T cell culture techniques to establish the presence of an
intrauterine-derived maternal T cell graft in a patient with severe combined
immunodeficiency (SCID). Transplantation 36: 733–735.
28. Geha, R. S., and E. Reinherz. 1983. Identification of circulating maternal T and
B lymphocytes in uncomplicated severe combined immunodeficiency by HLA
typing of subpopulations of T cells separated by the fluorescence-activated cell
sorter and of Epstein Barr virus-derived B cell lines. J. Immunol. 130: 2493–
29. Thompson, L. F., R. D. O’Connor, and J. F. Bastian. 1984. Phenotype and
function of engrafted maternal T cells in patients with severe combined im-
munodeficiency. J. Immunol. 133: 2513–2517.
30. Le Deist, F., C. Raffoux, C. Griscelli, and A. Fischer. 1987. Graft vs graft re-
action resulting in the elimination of maternal cells in a SCID patient with
maternofetal GVHd after an HLA identical bone marrow transplantation. J.
Immunol. 138: 423–427.
31. Barrett, M. J., R. H. Buckley, S. E. Schiff, P. C. Kidd, and F. E. Ward. 1988.
Accelerated development of immunity following transplantation of maternal
marrow stem cells into infants with severe combined immunodeficiency and
transplacentally acquired lymphoid chimerism. Clin. Exp. Immunol. 72: 118–
32. Wahn, V., S. Yokota, K. L. Meyer, J. W. Janssen, T. E. Hansen-Hagge,
C. Knobloch, S. Koletzko, H. Stein, W. Friedrich, and C. R. Bartram. 1991.
Expansion of a maternally derived monoclonal T cell population with CD3
+/CD8+/T cell receptor-gamma/delta+ phenotype in a child with severe com-
bined immunodeficiency. J. Immunol. 147: 2934–2941.
33. Mu ¨ller, S. M., M. Ege, A. Pottharst, A. S. Schulz, K. Schwarz, and W. Friedrich.
2001. Transplacentally acquired maternal T lymphocytes in severe combined
immunodeficiency: a study of 121 patients. Blood 98: 1847–1851.
34. Kobayashi, N., K. Agematsu, H. Nagumo, K. Yasui, Y. Katsuyama,
K. Yoshizawa, M. Ota, A. Yachie, and A. Komiyama. 2003. Expansion of
clonotype-restricted HLA-identical maternal CD4+ T cells in a patient with
severe combined immunodeficiency and a homozygous mutation in the Artemis
gene. Clin. Immunol. 108: 159–166.
35. Piotrowski, P., and B. A. Croy. 1996. Maternal cells are widely distributed in
murine fetuses in utero. Biol. Reprod. 54: 1103–1110.
36. Zhou, L., Y. Yoshimura, Y. Huang, R. Suzuki, M. Yokoyama, M. Okabe, and
M. Shimamura. 2000. Two independent pathways of maternal cell transmission
to offspring: through placenta during pregnancy and by breast-feeding after
birth. Immunology 101: 570–580.
37. Dutta, P., M. Molitor-Dart, J. L. Bobadilla, D. A. Roenneburg, Z. Yan,
J. R. Torrealba, and W. J. Burlingham. 2009. Microchimerism is strongly cor-
related with tolerance to noninherited maternal antigens in mice. Blood 114:
38. Forsthuber., T., H. C. Yip, and P. V. Lehmann. 1996. Induction of TH1 and TH2
immunity in neonatal mice [see comments]. Science 271: 1728–1730.
39. Bettini, M., and D. A. Vignali. 2009. Regulatory T cells and inhibitory cytokines
in autoimmunity. Curr. Opin. Immunol. 21: 612–618.
40. Molitor-Dart, M. L., J. Andrassy, J. Kwun, H. A. Kayaoglu, D. A. Roenneburg,
L. D. Haynes, J. R. Torrealba, J. L. Bobadilla, H. W. Sollinger, S. J. Knechtle,
and W. J. Burlingham. 2007. Developmental exposure to noninherited maternal
antigens induces CD4+ T regulatory cells: relevance to mechanism of heart al-
lograft tolerance. J. Immunol. 179: 6749–6761.
41. Aoyama, K., M. Koyama, K. Matsuoka, D. Hashimoto, T. Ichinohe, M. Harada,
K. Akashi, M. Tanimoto, and T. Teshima. 2009. Improved outcome of allogeneic
bone marrow transplantation due to breastfeeding-induced tolerance to maternal
antigens. Blood 113: 1829–1833.
42. Matsuoka, K., T. Ichinohe, D. Hashimoto, S. Asakura, M. Tanimoto, and
T. Teshima. 2006. Fetal tolerance to maternal antigens improves the outcome of
allogeneic bone marrow transplantation by a CD4+ CD25+ T-cell-dependent
mechanism. Blood 107: 404–409.
43. Mold, J. E., J. Michae ¨lsson, T. D. Burt, M. O. Muench, K. P. Beckerman,
M. P. Busch, T. H. Lee, D. F. Nixon, and J. M. McCune. 2008. Maternal
alloantigens promote the development of tolerogenic fetal regulatory T cells in
utero. Science 322: 1562–1565.
44. Owen, R. D., H. R. Wood, A. G. Foord, P. Sturgeon, and L. G. Baldwin. 1954.
Evidence for actively acquired tolerance to Rh antigens. Proc. Natl. Acad. Sci.
USA 40: 420–424.
45. van Rood, J. J., C. E. Stevens, J. Smits, C. Carrier, C. Carpenter, and
A. Scaradavou. 2009. Reexposure of cord blood to noninherited maternal HLA
antigens improves transplant outcome in hematological malignancies. Proc.
Natl. Acad. Sci. USA 106: 19952–19957.
The Journal of Immunology1449