Immunity, Vol. 21, 43–53, July, 2004, Copyright 2004 by Cell Press
Plasmacytoid Dendritic Cells
Activate Lymphoid-Specific Genetic Programs
Irrespective of Their Cellular Origin
2002). In murine hematopoiesis, early studies showed
that CD11c?MHCII?CD11b?CD8??“myeloid” DCs can
be efficiently derived from myeloid progenitors along
with granulocytes and macrophages (Inaba et al., 1993),
whereas CD11c?MHCII?CD11b?CD8??“lymphoid” DCs
can arise from thymic T cell progenitors (Ardavin et al.,
1993). Further analyses revealed that DCs of these two
phenotypes can originate both from the earliest myeloid
progenitors (common myeloid progenitors, CMPs [Akashi
et al., 2000]) and from the earliest lymphoid progenitors
(common lymphoid progenitors, CLPs [Kondo et al.,
1997]) (Traver et al., 2000). The potential to develop
into the DC lineage is maintained in pro-T cells and
(MEPs) (Manz et al., 2001; Wu et al., 2001), although
some DC potential may be retained in more primitive B
cell precursors (Izon et al., 2001). Thus, DC develop-
monocyte and T lymphoid developmental pathways.
Plasmacytoid dendritic cells (PDCs) were initially
identified in humans (Cella et al., 1999; Grouard et al.,
These cells morphologically resemble antibody-secre-
activation, they quickly acquire mature DC morphology
and phenotype without significant cell proliferation (Liu,
2001; Shortman and Liu, 2002). PDCs secrete massive
amounts of type I interferons (IFN-? and -?) when stimu-
lated by unmethylated CpG DNA or by viruses (Cella et
al., 1999; Grouard et al., 1997; Jarrossay et al., 2001;
Kadowaki et al., 2000, 2001; Siegal et al., 1999) through
the function of Toll-like receptors such as TLR7 and
TLR9 (Kadowaki et al., 2001) and are thus believed to
also play a role in adaptive immunity by directing either
T helper cell type 1 (Th1) or Th2 development (Boonstra
et al., 2003), as well as by generating T regulatory cells
that induce T cell tolerance (Gilliet and Liu, 2002). Re-
cently, a murine counterpart of human PDCs has been
ties (Asselin-Paturel et al., 2001; Bjorck, 2001; Martin et
al., 2002; Nakano et al., 2001). Like human PDCs, mouse
PDCs express high levels of CD45RA (B220) and low
levels of MHC class II (MHCII). Mouse PDCs express
low levels of the dendritic cell marker CD11c, and the
majority of them are CD8??. In addition, mouse PDCs
are Ly-6C?, a myeloid cell marker recognized by the
Gr-1 antibody that also reacts with Ly-6G on mature
granulocytes and monocytes.
The lineage origin of PDCs is controversial. On the
one hand, murine PDCs have been reported to carry
rearranged immunoglobulin heavy chain (IgH) D-J but
not T cell receptor (TCR) D?-J? loci (Corcoran et al.,
2003). Another apparently lymphoid characteristic of
PDCs is the expression of the pre-T cell receptor ? (pT?)
in the human thymic PDCs (Bendriss-Vermare et al.,
2001; Res et al., 1999). The pT? protein pairs with the
TCR? chain to form the pre-TCR (Groettrup et al., 1993)
Hirokazu Shigematsu,1,6Boris Reizis,2,6,7
Hiromi Iwasaki,1Shin-ichi Mizuno,1Dan Hu,1
David Traver,3Philip Leder,2Nobuo Sakaguchi,4
and Koichi Akashi1,5,*
1Department of Cancer Immunology and AIDS
Dana-Farber Cancer Institute and
2Department of Genetics and
The Howard Hughes Medical Institute
Harvard Medical School
Boston, Massachusetts 02115
3Department of Hematology/Oncology
Boston, Massachusetts 02115
4Department of Immunology
Graduate School of Medical Sciences
1-1-1 Honjo, Kumamoto 860-8556
5Center for Cellular and Molecular Medicine
Kyushu University Hospital
3-1-1 Maidashi, Higashiku, Fukuoka 812-8582
The developmental origin of type I interferon (IFN)-
producing plasmacytoid dendritic cells (PDCs) is con-
globulin heavy chain (IgH) genes in murine PDCs and
the expression of pre-T cell receptor ? (pT?) gene
by human PDCs were proposed as evidence for their
“lymphoid” origin. Here we demonstrate that PDCs
capable of IFN production develop efficiently from
both myeloid- and lymphoid-committed progenitors.
found in both myeloid- and lymphoid-derived PDCs.
The human pT? transgenic reporter was activated in
both myeloid- and lymphoid-derived PDCs at a level
comparable to pre-T cells. PDCs were the only cell
population that activated murine RAG1 knockin and
human pT? transgenic reporters outside the lymphoid
lineage. These results highlight a unique develop-
mental program of PDCs that distinguishes them from
other cell types including conventional dendritic cells.
Dendritic cells (DCs) are one of the most important cell
types inthe immune systemdue to their uniqueability to
orchestrate an acquired immune response (Banchereau
and Steinman, 1998). Murine DCs have been classified
into two populations, myeloid and lymphoid, on the basis
6These authors contributed equally to this work.
New York, New York 10032.
and plays a critical role in the efficient generation of
classical lineage markers of both T and B cells, which
was proposed to reflect their lymphoid past (Corcoran
et al., 2003). On the other hand, human PDCs can be
generated at least from myeloid progenitors expressing
receptors for macrophage colony-stimulating factor
(M-CSF) (Olweus et al., 1996). Mice deficient in ICSBP,
a critical transcription factor for monocyte development
(Tamura et al., 2000), display the loss of PDCs and
CD8??DCs (Aliberti et al., 2003; Schiavoni et al., 2002),
which can be restored by enforced expression of ICSBP
(Tsujimura et al., 2003). A recent report showed that
lymphoid progenitors give rise to cells of B220?CD11clo
PDC phenotype more efficiently than myeloid progeni-
tors (D’Amico and Wu, 2003). In this study, however, the
function of the resulting cells was not evaluated. It is
thus still unclear whether the lymphoid characteristics
of PDCs reflect their lymphoid origin.
or lymphoid-restricted potentials including CMPs and
CLPs (Akashi et al., 2000; Kondo et al., 1997). In the
current study, we analyzed directly the PDC potentials
of these progenitor populations. Our data show that
lymphoid-restricted CLPs and pro-T cells as well as
myeloid-restricted CMPs and GMPs can generate
IFN-producing PDCs with IgH D-J rearrangement. Fur-
thermore, recombination activation gene-1 (RAG1) tran-
scripts were found in both myeloid and lymphoid PDCs.
Themajority ofPDCs includingCMP-derived PDCsacti-
vated RAG1 locus in mice carrying a knockin reporter
for murine RAG1. Furthermore, in transgenic mice car-
rying a human pT? reporter construct, human pT? was
or myeloid origin. Taken together, our data demonstrate
that IFN-producing PDCs can originate from both
lymphoid and myeloid pathways and that the lymphoid
characteristics of PDCs represent a unique “ectopic”
activation of the lymphoid programs. PDC-specific acti-
vation of the human pT? reporter in murine hematopoie-
are faithful counterparts of human PDC subsets.
number of PDC progeny was obtained on day 15 in both
cases. In both CMP- and CLP-injected mice, CD11c?
donor-derived progeny (CD45.2?) in the spleen con-
tained a significant fraction of B220?PDCs as well as
B220?DCs (Figure 1A). PDC and DC development was
observed also in mice injected with 1 ? 105GMPs or
1 ? 104pro-T cells, whereas 1 ? 105MEPs or pro-B
cells did not give rise to detectable numbers of PDC or
DC subsets (Figure 1A and data not shown). Thus, PDC
development appears similar to conventional DC devel-
opment (Manz et al., 2001; Traver et al., 2000; Wu et al.,
2001) in that they can be efficiently derived from both
granulocyte/monocyte and early T lymphoid pathways.
A recent study showed that a circulating population
expressing CD11c and B220 but not MHCII can give rise
to PDCs, CD8??DCs, and CD8??DCs (del Hoyo et al.,
CD11b were also found in the blood of mice injected
with CMPs or CLPs (Figure 1A, right). These CMP- or
CLP-derived CD11c?I-Ab?DC precursors gave rise to
B220?CD11c?PDCs in the culture with Flt-3L (Brawand
et al., 2002) for 2 days (data not shown), suggesting that
at least a fraction of PDCs can originate from CMPs and
CLPs via the common DC precursor stage.
Figure 1B shows phenotypic characteristics of CMP-
or CLP-derived PDCs and DCs. Within the CD11c?DC
fraction in mice reconstituted with CMPs, Gr-1 expres-
sionwas onlyfoundin B220?PDCs. CMP-derivedPDCs
expressed only negative to low levels of CD11b and DC
CD86. A majority of PDCs expressed CD4 and CD8?. In
both CMP- and CLP-derived PDCs, the expression level
of I-Abwas relatively low as compared to B220?DCs.
PDCs were identical to those of B220?CD11c?PDCs
isolated from the normal spleen (Figure 1B, bottom).
produced significant amount of IFN-? upon exposure
to herpes simplex virus (HSV) in vitro, indicating that
they are functional PDCs (Figure 2A).
On a per cell basis, transplanted CMPs and CLPs
were more efficient than GMPs and pro-T cells for both
PDC and DCproduction (Table 1). CMPsand CLPs gave
rise to 2- to 3-fold higher numbers of conventional DCs
as compared to PDCs. Total mouse bone marrow cells
contains ?10-fold more CMPs than CLPs. Accordingly,
the estimated contribution of myeloid progenitors
and ?30% in the spleen and the thymus, respectively
PDC Potentials of Murine Progenitors Correlate
with Their Conventional DC Potentials
We estimated the relative contribution of myeloid and
lymphoid pathways to PDC development by in vivo re-
constitution assays. Myeloid- and lymphoid-committed
progenitors including CMPs, GMPs, and CLPs (CD45.2)
were purified and transplanted into C57B6-CD45.1 mice
poietic stem cells (HSCs) (Spangrude et al., 1988) to
provide radioprotection. Myeloid or lymphoid differenti-
ation potentials of these lineage-restricted progenitors
have been described, previously (Akashi et al., 2000;
Kondo et al., 1997; Na Nakorn et al., 2002). We injected
2 ? 104CLPs or CMPs, and analyzed their PDC progeny
on day 10, 15, and 20 on FACS by using the phenotypic
definition of Lineage (Lin: CD19, CD3, NK1.1, TCR?, and
sIgM) -negative, B220?, and CD11c?(Asselin-Paturel et
al., 2001; Bjorck, 2001; Nakano et al., 2001). The highest
Comparison of Lineage-Specific Gene Expression
in Myeloid and Lymphoid PDCs
We next tested the myeloid or lymphoid characteristics
of CMP- and CLP-derived PDCs and DCs by evaluating
lineage-related gene expression by semiquantitative
RT-PCR (Akashi et al., 2000). As shown in Figure 2B, all
PDC and DC populations expressed Flt-3 and myeloid
cytokine receptors such as IL-3R? and GM-CSFR? at
similar levels. PDCs expressed higher levels of IL-7R?,
an essential cytokine receptor for T and B cell develop-
ment (Akashi et al., 1997; Peschon et al., 1994), regard-
less of their myeloid or lymphoid origin. Both lymphoid-
Ectopic Lymphoid Programs in Dendritic Cells
Figure 1. Development of PDCs from Myeloid and Lymphoid Progenitors In Vivo
(A) DC analyses 15 days after transplantation of each purified progenitor subset are shown. Spleen cells were enriched for positive expression
of CD11c, and blood cells were enriched for negative expression of CD3 and CD19 prior to the phenotypic analyses (see Experimental
Procedures). Lin (CD19, CD3, NK1.1, TCR?, and sIgM)?CD45.2?cells in the spleen were analyzed for B220 and CD11c. Either transplanted
CMP, GMP, CLP, or pro-T cells differentiated into both PDCs and DCs in the spleen (left panels). The CD11c?MHCII?DC precursor population
was detectable in the Lin?blood of mice reconstituted with CMPs or CLPs (right panels). CD11c?MHCII?DC precursors expressed B220
(B) DC/PDC-related antigen expression of CMP- or CLP-derived PDCs (closed histogram) and conventional DCs (open histogram) is shown.
CMP, common myeloid progenitor; GMP, granulocyte/monocyte progenitor; CLP, common lymphoid progenitor; DCp, DC precursor.
and myeloid-derived PDCs expressed higher amounts
recognizes unmethylated CpG DNA leading to IFN-?
PU.1 and RelB are essential for CD8??DC develop-
ment (Guerriero et al., 2000; Wu et al., 1998), whereas
ICSBP and Id2 are required for CD8??DC generation
tial also for PDC development (Schiavoni et al., 2002).
All of these transcription factors were expressed in both
PDCs and DCs irrespective of their origin (Figure 2B).
Pax-5 and GATA-3, the master B and T lymphoid tran-
scription factors, respectively (Nutt et al., 1997; Ting
et al., 1996), were not expressed in any DC or PDC
populations. Spi-B, a myeloid/B lymphoid transcription
factor (Ray et al., 1992) was expressed only in PDCs
as previously reported (Bendriss-Vermare et al., 2001).
Notch-1, which plays a critical role in ??T cell commit-
ment and differentiation (Deftos et al., 2000; Robey,
1999),was expressedinPDCs athigherlevels thanDCs.
Spi-B and Notch-1 were expressed in both CMP- and
CLP-derived PDCs at similar levels. Thus, the expres-
sion patterns of these dendritic cell-related genes were
similar in myeloid- and lymphoid-derived PDCs.
Myeloid- and Lymphoid-Derived PDCs
Rearrange IgH D-J Locus
We then tested the rearrangement status of IgH D-J
genes in myeloid- and lymphoid-derived PDCs. Consis-
isolated PDCs but not DCs possessed IgH D-J rear-
rangements (Figure 2C, top). Purified Gr-1?or CD4?
Lin?B220?CD11c?PDCs were also rearranged IgH
genes (Figure 2C, bottom), indicating that the IgH rear-
rangements should not be due to contamination of B
cells within the B220?CD11c?PDC gate. TCR D?-J?
gene rearrangement was not detected in PDCs or DCs
(data not shown). CMPs and CLPs did not rearrange
IgH genes (Figure 2C, top). Interestingly, a significant
fraction of both CMP-derived and CLP-derived PDCs
rearranged IgH D-J genes (Figure 2C, top), indicating
Figure 2. IFN Production and Gene Expression Profiles of PDCs and DCs of Myeloid or Lymphoid Origin
(A) IFN-? levels in the supernatant of PDCs cultured with HSV. Note that only PDCs produce significant amounts of IFN-? irrespective of their
(B) RT-PCR analyses of lineage or differentiation related genes including cytokine receptors and transcription factors. The symbols under
each lane depict the relative amounts of transcripts in each population compared to control cDNA (2 ? 105cells) by the ratio of pixel density
units of target cDNA/pixel density units of control cDNA: ?0.1 (?), 0.1–0.5 (?), 0.5–1.5 (?), ?1.5 (??).
(C) D-J rearrangement of IgH gene in PDCs. IgH D-J rearrangement was detected by genomic PCR using primer sets for DHQ52 element.
Freshly isolated PDCs but not DCs, CMPs, or CLPs rearranged IgH D-J locus. Both CMP-derived PDCs (CMP-PDC) and CLP-derived PDCs
(CLP-PDC) rearranged at least at the JH2 locus of IgH (upper panels). Samples containing 300 freshly isolated Gr-1?B220?CD11c?or
CD4?B220?CD11c?PDCs also rearranged IgH genes (bottom).
that the rearrangement of IgH genes occurs during the
transition from the CMP or CLP stages to mature PDC
stage. Thus, IgH D-J rearrangements in PDCs do not
necessarily reflect their lymphoid origin but rather arise
during late stages of PDC development.
in pre-T and pre-B cells. A fraction of NK (data not
shown), T, and B cells maintained GFP expression. We
found that the majority of spleen and thymic PDCs but
not DCs expressed a low level of GFP (Figure 3B). Inter-
estingly, in the analysis of the bone marrow and spleen,
the activation of RAG1-GFP outside the lymphoid (T, B,
and NK) system was only found in the PDC fraction.
GFP was also expressed in PDCs developed from CMP
in vivo (Figure 3C), but CD11c?MHCII?DC precursors
in the blood did not express GFP (Figure 3A). RT-PCR
not DCs, possessed RAG1 and RAG2 (data not shown)
transcripts (Figure 3D). The expression level of RAG
genes was low, since we could detect RAG1 and RAG2
transcripts in PDCs only by nested RT-PCR. Altogether,
result from the ectopic low-level expression of RAG
genes in the committed PDCs.
PDCs Activate RAG1 Transcription
Irrespective of Their Origin
Since the IgH rearrangements depend upon activation
of RAG, we hypothesized that RAG might be expressed
in PDCs. To test this model, we analyzed mice carrying
a reporter for RAG1 transcription, in which enhanced
green fluorescent protein (EGFP) is knocked into the
RAG1 locus (Kuwata et al., 1999). In heterozygous
RAG1-EGFP knockin mice, the majority of CLPs and
proT cells expressed GFP, whereas there were no de-
tectable GFP?cells in myeloid progenitors including
CMPs andGMPs (Figure3A). GFPwas highlyexpressed
Ectopic Lymphoid Programs in Dendritic Cells
PDCs Activate Human pT? Transcription
Irrespective of Their Origin
In humans, pT? is expressed in PDCs at a high level
comparable to thymocytes (Bendriss-Vermare et al.,
2001). In contrast, murine pT? expression was barely
detectable in PDCs, and its level was significantly lower
than in thymocytes (Corcoran et al., 2003). By RT-PCR,
in both CLP- and CMP-derived PDCs (Figure 4A). Thus,
it has been suggested that pT? expression may not
reflect their origin. Furthermore, this observation raised
the possibility that murine B220?CD11c?PDCs may not
the transcriptional regulation of the human and murine
pT? genes may be inherently different. To distinguish
ysis of reporter mouse strains.
Normal transcriptional regulation of human genes has
been reproduced in transgenic mice that carry large
human genomic fragments such as yeast or bacterial
artificial chromosomes (BAC) (Kaufman et al., 1999;
Okuno et al., 2002; Peterson et al., 1993). Therefore, we
analyzed BAC transgenic mice carrying either mouse or
human pT? loci in which the first exons were replaced
with an EGFP gene (Reizis and Leder, 1999, 2001). In
restricted to lymphoid cells in the thymus and was
not found at detectable levels in spleen or bone mar-
row cells including PDCs (Figure 4B). In contrast, in
human pT?-EGFP transgenic mice, GFP?cells were
found in a significant fraction of thymic cells including
B220?CD11c?PDCs and immature thymocytes includ-
ing pro-T and pre-T cells (Figure 5A). Strikingly, in the
bone marrow and the spleen, GFP?cells almost exclu-
ure 5A). These results indicate that the human pT? is
activated in PDCs during mouse hematopoiesis, possi-
bly due to some regulatory element specific to the hu-
man gene, and confirm that previously defined murine
PDCs (Asselin-Paturel et al.,2001; Bjorck, 2001; Nakano
et al., 2001) represent the human PDC counterpart.
The expression of the human pT? transgene in PDCs
was heterogeneous, with the fraction of GFP?PDCs
proportional to the transgene copy number (data not
shown). In the transgenic line analyzed, the expression
was found in ?30% of PDCs in the thymus, spleen, and
bone marrow (Figure 5B). The expression patterns of
Gr-1, CD11b, MHCII (I-Aq), CD4, and CD8? were similar
between GFP?and GFP?PDCs (Figure 5B), and both
ure 5C). Moreover, expression levels of lymphoid genes
such as IL-7R?, Spi-B, and Notch-1 did not significantly
differ between GFP?and GFP?splenic PDCs (Figure 5D).
We next tested whether the activation of human pT?
occurs specifically in lymphoid-derived PDCs. In human
pT?-EGFP transgenic mice, GFP expression was initi-
in 2% and ?60% of CLPs and pre-T cells, respectively,
whereas myeloid progenitors such as CMPs and GMPs
did not express GFP. GFP was not upregulated in circu-
lating CD11c?MHCII?DC precursors (Figure 6A). GFP?
CLPs and CMPs were purified from human pT?-EGFP
transgenic mice and were transplanted into lethally irra-
diated hosts. As shown in Figure 6A, both CMPs and
Table 1. Generation of Dendritic Cell Progeny 15 Days after Transplantation
Extrapolated Yield of PDCs
2 ? 104
0.8 ? 0.2 ? 105
1.9 ? 0.4 ? 105
0.1 ? 0.02 ? 103
2.4 ? 0.8 ? 103
20 ? 104
83.0 ? 104
0.1 ? 104
10 ? 104
0.2 ? 0.1 ? 105
0.7 ? 0.4 ? 105
40 ? 104
8.0 ? 104
2 ? 104
0.3 ? 0.1 ? 105
1.1 ? 0.2 ? 105
0.2 ? 0.06 ? 104
3.5 ? 1.3 ? 104
2 ? 104
3.1 ? 104
0.2 ? 104
1 ? 104
0.6 ? 0.5 ? 103
9.1 ? 2.3 ? 103
3 ? 104
0.19 ? 104
PDC, plasmacytoid dendritic cell; DC, conventional dendritic cell; CMP, common myeloid progenitor; GMP, granulocyte/monocyte progenitor; CLP, common lymphoid progenitor; ND, not detectable.
Results are given as means ? SD in three experiments, each consisting of three to five animals.
*Progenitor number refers to total cell numbers in the femurs and tibias of one mouse (CLP, CMP, GMP) or in the thymus (Pro-T) of 6- to 8-week-old animals.
Figure 3. Analysis of RAG1-EGFP Knockin Mice
(A) Expression of RAG1-EGFP in each purified progenitor subsets. DCp, DC precursors.
(B) Steady-state PDCs but not DCs were EGFP?.
(C) RAG1-EGFP was expressed in CMP-derived PDCs in the spleen. EGFP expression was evaluated 15 days after transplantation of purified
EGFP?CMPs into congenic hosts.
(D) The nested RT-PCR analysis of RAG1 in purified populations. Both CMP-derived PDCs (CMP-PDC) and CLP-derived PDCs (CLP-PDC)
possessed RAG1 transcripts.
CLPs gave rise to PDCs, which contained a similar pro-
pT? transcription is upregulated after cells differentiate
into PDCs irrespective of their lineal origin.
In this study, we provide a formal proof that functional
PDCs originate from both myeloid and lymphoid devel-
Figure 4. Analysis of Murine pT? Expression
(A) RT-PCR analyses of pT? in purified PDC
and DC subsets. Both myeloid and lymphoid
PDCs express murine pT? transcript albeit at
a very low level.
(B) Analysis of GFP?cells in murine (m)
pT?-EGFP transgenic mice. GFP expression
is only seen in B220?CD11c?immature thy-
and bone marrow cells express GFP at a de-
tectable level on FACS.
Ectopic Lymphoid Programs in Dendritic Cells
Figure 5. Analysis of Mice Harboring a Human pT? Transgenic Reporter
(A) Analysis of GFP?cells in human pT?-EGFP transgenic mice. In the thymus, the GFP?population consists of B220?CD11c?PDCs as well
as B220?CD11c?immature thymocytes. In contrast, GFP?cells are exclusively B220?CD11c?PDCs in the bone marrow and the spleen.
(B) GFP expression in PDCs and DCs in thymus, bone marrow, and spleen cells of human pT?-EGFP transgenic mice (left panels). The
expression of other DC-related antigens in spleen PDCs and DCs is also shown (right panels). PDCs display the similar expression pattern
of these antigens regardless of human pT?-EGFP transgene activation.
(C) IFN-? levels in the supernatant of human pT?-GFP-positive or -negative PDCs cultured with HSV.
(D) Expression profiles of lymphoid-related genes in human pT?-GFP-positive or -negative PDCs. BM, bone marrow; hu-pT?, human pT?.
opmental pathways. PDC potential is maintained along
early T lymphoid stages (i.e, CLPs and pro-T cells) and
early granulocyte/monocyte stages (i.e., CMPs and
equally functional, producing IFN-? when challenged
with a virus. Both CMPs and CLPs efficiently gave rise
to PDCs and DCs, and the frequency of PDC generation
(Table 1). Although a very small fraction (?0.1%) of
CMPs maintain B cell potential (Akashi et al., 2000),
which are completely devoid of lymphoid potential.
Pro-B cells could not differentiate into either PDCs or
DCs. Thus, PDC potential is closely associated with
conventional DC potential (Manz et al., 2001; Traver et
and CLPs could be still heterogeneous for PDC and DC
potential since Flt-3-expressing CMPs and CLPs were
Flt-3-negative ones (D’Amico and Wu, 2003; Karsunky
et al., 2003). In this study, however, we did not subfrac-
tionate each progenitor population by Flt-3 expression
because Flt-3-negative subsets still could produce sig-
nificant numbers of DCs and PDCs (D’Amico and Wu,
2003). It is possible that Flt-3 expression represents
an immediate expansion potential of each progenitor
tors that include both Flt-3-positive and -negative sub-
sets, we demonstrate here that the myeloid pathway
should produce the majority of PDCs, as in the case of
conventional DCs (Table 1) (Manz et al., 2001).
Myeloid and lymphoid cells generally differentiate
along independent pathways, where CMPs and CLPs
likely represent the earliest committed branchpoints
(Akashi etal., 2000;Kondo et al.,1997). Therefore,it was
conceivable that myeloid- or lymphoid-derived DC/PDC
subsets might be phenotypically and functionally dis-
ent phenotypic or functional differences between my-
eloid- and lymphoid-derived subsets. Interestingly,
Figure 6. Function and Origin of PDC Populations Segregated by the Activation of the Human pT? Transgene
(A) The expression of human pT?-EGFP in PDCs derived from CMPs or CLPs. In human pT?-EGFP transgenic mice, pre-T cells but not CMPs
or CLPs upregulate GFP (upper panels). GFP?CMPs and CLPs give rise to GFP?PDCs in vivo after transplantation (bottom panels).
(B) Models of PDC and DC development from myeloid and lymphoid progenitors. According to the myeloid versus lymphoid developmental
scheme (Akashi et al., 2000; Kondo et al., 1997), myeloid progenitors (CMPs and GMPs) and lymphoid progenitors (CLPs and pro-T cells) give
rise to independent DC precursors (DCp), respectively (I), where DC and PDC populations should consist of myeloid and lymphoid subtypes.
In contrast, if myeloid and lymphoid progenitors use an identical pathway for DC development via common DCp (II), their DC and PDC progeny
should also be identical regardless of their original lineages. The induction of human pT? and RAG1 expression has been proposed to occur
at late stages of PDC development from both lymphoid and myeloid progenitors in either model. HSC, hematopoietic stem cell.
myeloid-related cytokine receptors such as GM-CSFR?
and IL-3R? were expressed in CLP-derived PDCs and
DCs at levels similar to CMP-derived ones, respectively,
suggesting the reactivation of these myeloid cytokine
receptors during the lymphoid to DC/PDC differentia-
tion.Itis importanttonotethatCLPsand pro-Tcellscan
be converted into mature granulocytes and monocytes
simply byectopic expression of GM-CSFsignaling (Iwa-
saki-Arai et al., 2003; Kondo et al., 2000). In this context,
DCs and PDCs may develop from CLPs and pro-T cells
by using their residual myeloid differentiation programs
(Iwasaki-Arai et al., 2003).
ing their development from myeloid and lymphoid pro-
genitors. Indeed, B/myeloid-related Spi-B as well as T
lymphoid-related Notch-1 was expressed at higher lev-
els in CMP- and CLP-derived PDCs than conventional
DCs (Figure 2B). Moreover, IgH rearrangement was de-
tected in both myeloid- and lymphoid-derived PDCs
(Figure 2C). Importantly, PDCs were the only cell types
that expressed GFP outside the lymphoid lineage in
RAG1-GFP mice (Figure 3), and both lymphoid and my-
eloid PDCs possessed RAG1 and RAG2 transcripts, al-
progenitors. Thus, the ectopic activation of RAG genes
in both myeloid- and lymphoid-derived PDCs might re-
sult in the rearrangement of IgH genes. Another impor-
tant lymphoid marker, pT?, is also expressed in both
levels were low as compared to thymocytes (Figure 4).
It remains to be established whether the activation of
these lymphoid developmental programs is in any way
relevant for the unique functions of PDCs.
By PCR analysis, the expression of pT? in the mouse
PDCs appeared very low. Consistent with this, PDCs in
of GFP, while pre-T cells had GFP at a very high level.
In human PDCs (Bendriss-Vermare et al., 2001), pT?
levels are comparable to those observed in thymic T cell
precursors (Res et al., 1999). Indeed, the human pT?
transgene was strongly expressed in a fraction of PDCs,
as represented by the expression of GFP at levels com-
parable to thymic precursors (Figure 5). These expres-
sion patterns were observed in multiple independent
transgenic lines, including seven human pT?-EGFP and
four mouse pT?-EGFP lines (B.R. et al., unpublished
data). Therefore, the significant upregulation of pT? in
PDCs appears to result from properties intrinsic to the
human pT? gene. Our analysis showed that similar per-
centages of CMP- and CLP-derived PDCs activated the
human pT? transgene, and both human pT??and pT??
PDCs can produce equal amounts of IFN-?. Further-
more, when we crossed human pT? transgenic mice
with athymic nu/nu mice, PDCs in athymic human-
pT?-EGFP transgenic mice also became GFP?(B.R. et
tion does not require thymic T cell maturation or expo-
sure to thymic microenvironmentsand is therefore inde-
pendent of T cell development. Taken together, our
results show that the expression of the human pT? by
PDCs does not reflect their derivation from lymphoid
precursors. The fact that the activation of human-
pT?-EGFP reporter occurs exclusively in PDCs outside
the lymphoid system suggests that the phenotypically
defined murine PDCs (Asselin-Paturel et al., 2001;
Bjorck, 2001; Martin et al., 2002; Nakano et al., 2001)
Ectopic Lymphoid Programs in Dendritic Cells
fully correspond to the human PDC counterpart (Cella
et al., 1999; Grouard et al., 1997; Jarrossay et al., 2001;
Kadowaki et al., 2000).
It is important to note that the GFP protein might
and therefore, the low level of GFP expression in PDCs
could represent the RAG1 activation at a developmental
stage prior to the PDC stage. Both CMPs and CLPs
generate CD11c?MHCII?DC precursors (del Hoyo et
PDCs and conventional DCs in vitro (H.S. et al., unpub-
lished data). The RAG1 or the human pT? reporter, how-
ever, was not activated in the DC precursors, and the
level of RAG1 transcripts was very low in mature PDCs.
It is thus possible that these lymphoid genes are mainly
activated during the transition from the DC precursor to
mature PDC stages (Figure 6B). Furthermore, although
it remains unknown whether all DCs and PDCs are gen-
erated from the DC precursor population, the capability
of CMPs and CLPs to generate DC precursors suggests
the existence of the common DC/PDC pathway, where
myeloid and lymphoid DC pathways merge together at
the common DC precursor stage to give rise to mature
DCs and PDCs (Figure 6B, bottom). In this model, equiv-
alent DC subsets may thus be independently generated
by both lineal pathways, and this may explain our inabil-
ity to detect unique DC attributes from each pathway.
If early lymphocyte-specific genes such as RAG and
pT? are unlikely to play any role in PDC function, why
are they expressed in mature PDCs? Since PDCs are
part of the innate immune system, they are likely to
predate antigen receptor-bearing lymphocytes in evolu-
tion. Therefore, the genetic program of their develop-
ment would be more ancient than that of lymphocytes
and may represent a “prototype” genetic pathway of im-
mune cell development. Later in evolution, parts of such a
pathway couldbe utilizedin thedevelopment ofspecific
lineages such as T and B lymphocytes. It is tempting to
control that can activate both lymphoid- and myeloid-
specific genes during PDC development. The character-
ization of these mechanisms of gene expression in PDCs
should be a focus of future experimentation.
In conclusion, our data demonstrate that the gene
expression program of PDCs is unique and includes the
programs primarily associated with early lymphocyte
development, including RAG and pT? transcripts and
rearranged IgH genes. This lymphoid program is inde-
pendent of the cellular origin of PDCs, as it is observed
in PDCs derived from both myeloid and lymphoid pro-
genitors. Therefore, it is likely to reflect specific tran-
scriptional mechanisms operating in committed PDCs
rather than their affiliation with the lymphoid lineage.
The expression of early lymphoid genes appears as a
unique property of PDCs that clearly distinguishes them
from other mature cell types in the immune system.
Future studies should elucidate the molecular basis and
possible functional significance of lymphoid gene ex-
pression in PDCs.
C57B6 (CD45.1 or CD45.2) mice were purchased from The Jackson
Laboratory (Bar Harbor, ME). RAG1-EGFP knockin mice have been
described elsewhere (Igarashi et al., 2001; Kuwata et al., 1999).
Human or murine pT?-EGFP transgenic mice (FVB-CD45.1) were
described previously (Reizis and Leder, 1999, 2001). F1 (FVB-
of pT?-EGFP transgenic cells. To generate pT? bacterial artificial
chromosome transgenic mice, BAC clones containing either mouse
or human pT? loci (90 and 113 kb, respectively) were modified to
replace the first pT? exon with an enhanced green fluorescent pro-
was left intact. The resulting BAC clones were used to generate
multiple independent transgenic mouse lines, which manifested
copy number-dependent EGFP expression with the same pattern
for each construct (data not shown). One line derived from each
BAC with a similarly high transgene copy number (16–17 copies in
hemizygous state as determined by quantitative genomic Southern
hybridization) was chosen for detailed analysis. Mice were bred and
maintained in the Research Animal Facilities at the Dana Farber
Cancer Institute or Harvard Medical School, in accordance with
Sorting of Progenitors and Dendritic Cells
Sorting of myeloid progenitors was accomplished by staining bone
ies (A7R34) (eBioscience, San Diego, CA) and purified or PE-Cy5-
conjugated rat antibodies specific for the following lineage markers:
CD3 (CT-CD3), CD4 (RM4-5), CD8 (5H10), B220 (6B2), Gr-1 (8C5),
Ter119, and CD19 (6D5) (Caltag, Burlingame, CA). IL-7R??Lin?cells
were removed with sheep anti-rat IgG-conjugated magnetic beads
(Dynabeads M-450; Dynal A.S., Oslo, Norway), and the remaining
cells were stained with PE-Cy5-conjugated goat anti-rat IgG (Cal-
tag). Cells were then stained with PE-conjugated anti-Fc?RII/III
bodies (Pharmingen, San Diego, CA), followed by avidin-APC-Cy7
(Caltag). Myeloid progenitors were sorted as IL-7R??Lin?Sca-1?
Fc?RII/IIIhi(GMPs), and IL-7R??Lin?Sca-1?c-Kit?CD34-Fc?RII/IIIlo
were sorted as IL-7R??Lin?Sca-1hic-Kithiand IL-7R??Lin?Sca-1lo
cells were sorted as CD3?CD4?CD8?NK1.1?IL-7R??c-Kit?CD25?
and CD43?B220?IgM?NK1.1?cells, respectively.
DCs were isolated from spleen, thymus, and bone marrow as
described elsewhere (Manz et al., 2001; Wu et al., 2001). In brief,
spleens and thymi were cut into small fragments and digested with
collagenase and DNase under repeated agitation, followed by the
addition of EDTA. Enzymatic digestion was avoided when isolating
DCs frombone marrow samples.To enrich DCpopulations, CD11c?
cells were positively enriched with CD11c (N418) MicroBeads and
MACS separation columns (Miltenyi Biotec, Bergisch Gladbach,
Germany). In some experiments, DCs were enriched after immuno-
magnetic depletion of B cells and T cells with anti-CD19 (6D5), anti-
IgM (1B4B1),anti-CD3 (CT-CD3),and sheepanti-rat IgG-conjugated
magnetic beads (Dynal). Cells were stained with anti-CD11c (HL3)
and anti-B220 (RA3-6B2) as well as a lineage cocktail including anti-
CD3 (CT-CD3), anti-TCR? (H57-597), anti-CD19 (6D5) (Caltag), anti-
IgM (1B4B1) (eBioscience), and anti-NK1.1 (PK136) (Pharmingen).
PDCs and DCs were purified as Lin?B220?CD11c?cells and
Lin?B220?CD11c?cells, respectively. These DC populations were
further analyzed by anti-Gr-1 (RB6-8C5), anti-CD11b (M1/70), anti-
CD4 (GK1.5), anti-CD8? (53-6.7), anti-I-Ab(AF6-120.1), anti-CD40
(3/23), anti-CD80 (16-10A1), anti-CD86 (GL1), and anti-I-Aq(KH116).
Streptavidin-conjugated PE, Cy5-PE, and APC-Cy7 (Caltag) were
used to visualize biotinylated antibodies.
nm Enterprise II ? 647 nm Spectrum) high-speed FACS (Moflo-
software (Treestar, Inc., San Carlos, CA).
Measurement of IFN-? from PDCs
FACS-purified B220?CD11c?PDCs and B220?CD11c?DCs were
cultured with HSV (10 pfu/cell) in IMDM supplemented with 10%
natants were collected 24 hr after the initiation of cultures and were
analyzed by mouse IFN-? ELISA kit (Performance Biomedical Labo-
ratories, New Brunswick, NJ).
Ardavin, C., Wu, L., Li, C.-L., and Shortman, K. (1993). Thymic den-
dritic cells and T cells develop simultaneously in the thymus from
a common precursor population. Nature 362, 761–763.
Asselin-Paturel, C., Boonstra, A., Dalod, M., Durand, I., Yessaad,
N., Dezutter-Dambuyant, C., Vicari, A., O’Garra, A., Biron, C., Briere,
F., and Trinchieri, G. (2001). Mouse type I IFN-producing cells are
immature APCs with plasmacytoid morphology. Nat. Immunol. 2,
In Vivo Reconstitution Assays
Purified progenitors (CD45.2) from C57B6 or RAG1-GFP knockin
(C57B6) mice were intravenously transplanted into congenic mice
(CD45.1) after a lethal dose (950 rad) of irradiation, together with
300Sca-1?c-Kit?host-type (CD45.1)stemcellsto provideradiopro-
tection. Mice were sacrificed on day 15, and donor-derived progeny
were analyzed using the CD45.2 donor marker. For reconstitution
assays of human pT?-EGFP transgenic progenitors (FVB-CD45.1),
GFP?CMPs or CLPs were injected intravenously into F1 (FVB-
mice. Donor-derived cells were detected as CD45.1?CD45.2?cells.
Banchereau, J., and Steinman, R.M. (1998). Dendritic cells and the
control of immunity. Nature 392, 245–252.
Bendriss-Vermare, N.,Barthelemy, C.,Durand, I., Bruand,C., Dezut-
ing CD11c(?), myeloid CD11c(?), and mature interdigitating den-
dritic cells. J. Clin. Invest. 107, 835–844.
Bjorck, P. (2001). Isolation and characterization of plasmacytoid
dendritic cells from Flt3 ligand and granulocyte-macrophage col-
ony-stimulating factor-treated mice. Blood 98, 3520–3526.
Boonstra, A., Asselin-Paturel, C., Gilliet, M., Crain, C., Trinchieri, G.,
Liu, Y.J., and O’Garra, A. (2003). Flexibility of mouse classical and
plasmacytoid-derived dendritic cells in directing T helper type 1 and
2 cell development: dependency on antigen dose and differential
toll-like receptor ligation. J. Exp. Med. 197, 101–109.
Analysis of Gene Expression from Total RNA
of cDNA was analyzed for specific genes. PCR products were sub-
jected to electrophoresis on a 1.8% agarose gel, followed by ethid-
ium bromide staining. PCR amplification for each experiment was
repeated for two or more independently prepared cDNA samples.
Relative gene expression quantification was accomplished by com-
paring the level of any transcript in the target samples to that in
control cDNA prepared from 2 ? 105whole bone marrow cells or
thymocytes (Akashi et al., 2000). PCR cycles for each target gene
were at a point when the reaction was in the exponential phase to
obtain a linear correlation between pixel density units of the PCR
product andthe amountof controlcDNA (Akashiet al., 2000).Primer
sequences for each specific gene are shown in Supplemental Table
S1 at http://www.immunity.com/cgi/content/full/21/1/43/DC1.
Brawand, P., Fitzpatrick, D.R., Greenfield, B.W., Brasel, K., Malis-
zewski, C.R., and De Smedt, T. (2002). Murine plasmacytoid pre-
dendritic cells generated from Flt3 ligand-supplemented bone mar-
row cultures are immature APCs. J. Immunol. 169, 6711–6719.
Cella, M., Jarrossay, D., Facchetti, F., Alebardi, O., Nakajima, H.,
Lanzavecchia, A., and Colonna, M. (1999). Plasmacytoid monocytes
migrate to inflamed lymph nodes and produce large amounts of
type I interferon. Nat. Med. 5, 919–923.
Corcoran, L., Ferrero, I., Vremec, D., Lucas, K., Waithman, J.,
O’Keeffe, M., Wu, L., Wilson, A., and Shortman, K. (2003). The
lymphoid past of mouse plasmacytoid cells and thymic dendritic
cells. J. Immunol. 170, 4926–4932.
IgH Gene Rearrangement Analysis
Genomic DNAwas extractedfrom 4000double-sorted cellsby incu-
bation at 56?C for 45 min in 50 ?l of 1? PCR buffer containing 0.5%
Tween 20 and 100 ?g/ml proteinase K, followed by 10 min at 94?C
to inactivate proteinase K. PCR primers specific for DHQ52 element
were used to amplify rearranged IgH D-J as described previously
(ten Boekel et al., 1995). PCR products were separated on an aga-
rose gel, transferred onto nylon membranes, and subjected to
Southern blotting using standard procedures. To detect TCR? rear-
rangement, the extracted DNA was amplified with the primers for
D?1-J?1 and D?2-J?2 as reported (Iwasaki-Arai et al., 2003).
D’Amico, A., and Wu, L. (2003). The early progenitors of mouse
dendritic cells and plasmacytoid predendritic cells are within the
bone marrow hemopoietic precursors expressing Flt3. J. Exp. Med.
Deftos, M.L., Huang, E., Ojala, E.W., Forbush, K.A., and Bevan, M.J.
(2000). Notch1 signaling promotes the maturation of CD4 and CD8
SP thymocytes. Immunity 13, 73–84.
del Hoyo, G.M., Martin, P., Vargas, H.H., Ruiz, S., Arias, C.F., and
Ardavin, C. (2002). Characterization of a common precursor popula-
tion for dendritic cells. Nature 415, 1043–1047.
Fehling, H.J., Krotkova, A., Saint-Ruf, C., and von Boehmer, H.
(1995). Crucial role of the pre-T-cell receptor alpha gene in develop-
analysis and sorting, and M. Shinohara for the instruction for IFN-?
measurement. This work was supported in part by grants from the
NIH (DK050654, DK061320, and CA072009) and Damon-Runyon
Cancer Research to K.A., and by the Uehara Memorial Foundation
and Leukemia & Lymphoma Society to S.-i.M.
cells by CD40 ligand-activated plasmacytoid dendritic cells. J. Exp.
Med. 195, 695–704.
Groettrup, M., Ungewiss, K., Azogui, O., Palacios, R., Owen, M.J.,
Hayday, A.C., and von Boehmer, H. (1993). A novel disulfide-linked
heterodimer on pre-T cells consists of the T cell receptor ? chain
and a 33 kd glycoprotein. Cell 75, 283–294.
Received: March 6, 2004
Revised: April 27, 2004
Accepted: May 5, 2004
Published: July 20, 2004
Grouard, G., Rissoan, M.C., Filgueira, L., Durand, I., Banchereau, J.,
and Liu, Y.J. (1997). The enigmatic plasmacytoid T cells develop
into dendritic cells with interleukin (IL)-3 and CD40-ligand. J. Exp.
Med. 185, 1101–1111.
Guerriero, A., Langmuir, P.B., Spain, L.M., and Scott, E.W. (2000).
PU.1 is required for myeloid-derived but not lymphoid-derived den-
dritic cells. Blood 95, 879–885.
man, I.L. (1997). Bcl-2 rescues T lymphopoiesis in interleukin-7 re-
ceptor-deficient mice. Cell 89, 1033–1041.
Akashi, K., Traver, D., Miyamoto, T., and Weissman, I.L. (2000). A
lineages. Nature 404, 193–197.
Aliberti, J., Schulz,O., Pennington, D.J., Tsujimura, H.,Reis e Souza,
C., Ozato, K., and Sher, A. (2003). Essential role for ICSBP in the
in vivo development of murine CD8alpha ? dendritic cells. Blood
Hacker, C., Kirsch, R.D., Ju, X.S., Hieronymus, T., Gust, T.C., Kuhl,
C., Jorgas, T., Kurz, S.M., Rose-John, S., Yokota, Y., and Zenke, M.
(2003). Transcriptional profiling identifies Id2 function in dendritic
cell development. Nat. Immunol. 4, 380–386.
Igarashi, H., Kuwata, N., Kiyota, K., Sumita, K., Suda, T., Ono, S.,
Bauer,S.R.,and Sakaguchi,N.(2001).Localization ofrecombination
activating gene 1/green fluorescent protein (RAG1/GFP) expression
Ectopic Lymphoid Programs in Dendritic Cells
in secondary lymphoid organs after immunization with T-dependent
antigens in rag1/gfp knockin mice. Blood 97, 2680–2687.
Inaba, K., Inaba, M., Deguchi, M., Hagi, K., Yasumizu, R., Ikehara,
S.,Muramatsu, S.,andSteinman, R.M.(1993). Granulocytes,macro-
phages, and dendritic cells arise from a common major histocom-
patibility complex class II-negative progenitor in mouse bone mar-
row. Proc. Natl. Acad. Sci. USA 90, 3038–3042.
Iwasaki-Arai, J., Iwasaki, H., Miyamoto, T., Watanabe, S., and
Akashi, K. (2003). Enforced GM-CSF signals do not support lympho-
poiesis, but instruct lymphoid to myelomonocytic lineage conver-
sion. J. Exp. Med. 197, 1311–1322.
Izon, D., Rudd, K., DeMuth, W., Pear, W.S., Clendenin, C., Lindsley,
R.C., and Allman, D. (2001). A common pathway for dendritic cell
and early B cell development. J. Immunol. 167, 1387–1392.
Jarrossay, D., Napolitani, G., Colonna, M., Sallusto, F., and Lanza-
vecchia, A. (2001). Specialization and complementarity in microbial
molecule recognition by human myeloid and plasmacytoid dendritic
cells. Eur. J. Immunol. 31, 3388–3393.
Kadowaki, N., Antonenko, S., Lau, J.Y., and Liu, Y.J. (2000). Natural
interferon alpha/beta-producing cells link innate and adaptive im-
munity. J. Exp. Med. 192, 219–226.
Kadowaki, N., Ho, S., Antonenko, S., Malefyt, R.W., Kastelein, R.A.,
Bazan, F., and Liu, Y.J. (2001). Subsets of human dendritic cell
precursors express different toll-like receptors and respond to dif-
ferent microbial antigens. J. Exp. Med. 194, 863–869.
Karsunky, H., Merad, M., Cozzio, A., Weissman, I.L., and Manz, M.G.
(2003). Flt3 ligand regulates dendritic cell development from Flt3?
lymphoid and myeloid-committed progenitors to Flt3? dendritic
cells in vivo. J. Exp. Med. 198, 305–313.
of a 100-kb human beta-globin cluster-containing DNA fragment
propagated as a bacterial artificial chromosome. Blood 94, 3178–
Kondo, M., Weissman, I.L., and Akashi, K. (1997). Identification of
clonogenic common lymphoid progenitors in mouse bone marrow.
Cell 91, 661–672.
mura, K., and Weissman, I.L. (2000). Cell-fate conversion of
lymphoid-committed progenitors by instructive actions of cyto-
kines. Nature 407, 383–386.
Kuwata, N., Igarashi, H., Ohmura, T., Aizawa, S., and Sakaguchi, N.
(1999). Cutting edge: absence of expression of RAG1 in peritoneal
B-1 cells detected by knocking into RAG1 locus with green fluores-
cent protein gene. J. Immunol. 163, 6355–6359.
Liu, Y.J. (2001). Dendritic cell subsets and lineages, and their func-
tions in innate and adaptive immunity. Cell 106, 259–262.
Manz, M.G., Traver, D., Miyamoto, T., Weissman, I.L., and Akashi,
K. (2001). Dendritic cell potentials of early lymphoid and myeloid
progenitors. Blood 97, 3333–3341.
Martin, P., Del Hoyo, G.M., Anjuere, F., Arias, C.F., Vargas, H.H.,
Fernandez, L.A., Parrillas, V., and Ardavin, C. (2002). Characteriza-
tion of a new subpopulation of mouse CD8alpha(?) B220(?) den-
dritic cells endowed with type 1 interferon production capacity and
tolerogenic potential. Blood 100, 383–390.
Na Nakorn, T., Traver, D., Weissman, I.L., and Akashi, K. (2002).
Myeloerythroid-restricted progenitors are sufficient to confer radio-
protection and provide the majority of day 8 CFU-S. J. Clin. Invest.
Nakano, H., Yanagita, M., and Gunn, M.D. (2001). CD11c(?)
B220(?)Gr-1(?) cells in mouse lymph nodes and spleen display
characteristics of plasmacytoid dendritic cells. J. Exp. Med. 194,
Nutt, S.L., Urbanek, P., Rolink, A., and Busslinger, M. (1997). Essen-
tial functions of Pax5 (BSAP) in pro-B cell development: difference
between fetal and adult B lymphopoiesis and reduced V-to-DJ re-
combination at the IgH locus. Genes Dev 11, 476–491.
Okada, T., Lian, Z.X., Naiki, M., Ansari, A.A., Ikehara, S., and Gersh-
win, M.E. (2003). Murine thymic plasmacytoid dendritic cells. Eur.
J. Immunol. 33, 1012–1019.
Okuno, Y., Iwasaki, H., Huettner, C.S., Radomska, H.S., Gonzalez,
D.A., Tenen, D.G., and Akashi, K. (2002). Differential regulation of
the human and murine CD34 genes in hematopoietic stem cells.
Proc. Natl. Acad. Sci. USA 99, 6246–6251.
distinct changes in the expression of the PU.1-regulated molecules,
CD64 and macrophage colony-stimulating factor receptor. Blood
skovsky, E., Gliniak, B.C., Park, L.S., Ziegler, S.F., Williams, D.E.,
Ware, C.B., et al. (1994). Early lymphocyte expansion is severely
impaired in interleukin 7 receptor-deficient mice. J. Exp. Med.
Peterson, K.R., Clegg, C.H., Huxley, C., Josephson, B.M., Haugen,
H.S., Furukawa, T., and Stamatoyannopoulos, G. (1993). Transgenic
mice containing a 248-kb yeast artificial chromosome carrying the
human beta-globin locus display proper developmental control of
human globin genes. Proc. Natl. Acad. Sci. USA 90, 7593–7597.
Ray, D., Bosselut, R., Ghysdael, J., Mattei, M.G., Tavitian, A., and
Moreau-Gachelin, F. (1992). Characterization of Spi-B, a transcrip-
tion factor related to the putative oncoprotein Spi-1/PU.1. Mol. Cell
Biol. 12, 4297–4304.
Reizis, B., and Leder, P. (1999). Expression of the mouse pre-T cell
receptor alpha gene is controlled by an upstream region containing
a transcriptional enhancer. J. Exp. Med. 189, 1669–1678.
Reizis,B.,and Leder,P.(2001).Theupstream enhancerisnecessary
and sufficient for the expression of the pre-T cell receptor alpha
gene in immature T lymphocytes. J. Exp. Med. 194, 979–990.
Res, P.C., Couwenberg, F., Vyth-Dreese, F.A., and Spits, H. (1999).
in the human thymus. Blood 94, 2647–2657.
Robey, E. (1999). Regulation of T cell fate by Notch. Annu. Rev.
Immunol. 17, 283–295.
Schiavoni, G., Mattei, F., Sestili, P., Borghi, P., Venditti, M., Morse,
H.C., III, Belardelli, F., and Gabriele, L. (2002). ICSBP is essential
for the development of mouse type I interferon-producing cells and
for the generation and activation of CD8alpha(?) dendritic cells. J.
Exp. Med. 196, 1415–1425.
Shortman, K., and Liu, Y.J. (2002). Mouse and human dendritic cell
subtypes. Nat. Rev. Immunol. 2, 151–161.
Siegal, F.P., Kadowaki, N., Shodell, M., Fitzgerald-Bocarsly, P.A.,
Shah, K., Ho, S., Antonenko, S., and Liu, Y.J. (1999). The nature
of the principal type 1 interferon-producing cells in human blood.
Science 284, 1835–1837.
Spangrude, G.J., Heimfeld, S., and Weissman, I.L. (1988). Purifica-
tion and characterization of mouse hematopoietic stem cells. Sci-
ence 241, 58–62.
Tamura, T., Nagamura-Inoue, T., Shmeltzer, Z., Kuwata, T., and
Ozato, K. (2000). ICSBP directs bipotential myeloid progenitor cells
to differentiate into mature macrophages. Immunity 13, 155–165.
ten Boekel, E., Melchers, F., and Rolink, A. (1995). The status of Ig
loci rearrangements in single cells from different stages of B cell
development. Int. Immunol. 7, 1013–1019.
Ting, C.N., Olson, M.C., Barton, K.P., and Leiden, J.M. (1996). Tran-
scription factor GATA-3 is required for development of the T-cell
lineage. Nature 384, 474–478.
E.G., and Weissman, I.L. (2000). Development of CD8alpha-positive
dendritic cells from a common myeloid progenitor. Science 290,
Tsujimura, H., Tamura, T., and Ozato, K. (2003). Cutting edge: IFN
consensus sequence binding protein/IFN regulatory factor 8 drives
the development of type I IFN-producing plasmacytoid dendritic
cells. J. Immunol. 170, 1131–1135.
Wu, L., D’Amico, A., Winkel, K.D., Suter, M., Lo, D., and Shortman,
K. (1998). RelB is essential for the development of myeloid-related
CD8??dendritic cells but not of lymphoid-related CD8??dendritic
cells. Immunity 9, 839–847.
Wu, L., D’Amico, A., Hochrein, H., O’Keeffe, M., Shortman, K., and
Lucas, K. (2001). Development of thymic and splenic dendritic cell
populations from different hemopoietic precursors. Blood 98, 3376–