NOD2/CARD15 on bone marrow CD34? hematopoietic cells
mediates induction of cytokines and cell differentiation
Mouldy Sioud1and Yngvar Fløisand
Department of Immunology, Institute for Cancer Research, The Norwegian Radium Hospital, Montebello,
etic cells were found recently to express functional
TLRs and TLR signaling-induced cytokine produc-
tion and cell differentiation. Here, we have asked
whether signals other than those from TLRs could
instruct BM CD34? cells to produce cytokines and
differentiate by uncovering the role of nucleotide
oligomerization domain (Nod)-like receptor (NLR)
family members, NOD1 and NOD2. We show that
NOD2 is expressed by freshly isolated human BM
CD34? cells, whereas the expression of its close
homologue NOD1 is very weak, if it has any ex-
pression at all. Stimulation of the cells by the mu-
ramyl dipeptide (MDP), but not its inactive D–D
enantiomer, is sufficient to trigger the expression
of TNF-?, GM-CSF, CD11c, CD14, CD206, and
the transcription factor PU.1, which is indispens-
able for cell differentiation toward the myeloid
lineage. MDP differentiated CD11c? cell subset-
activated T cells in MLR. Furthermore, NOD2
stimulation enhanced the CD34? response to TLR
ligands (e.g., LPS, palmitoyl-3-cysteine-serine-ly-
sine-4) and increased intracellular ?-defensin pro-
tein levels. Although the best-known function of
NLRs involves mature cells, our data highlight for
the first time the functionality of these receptors in
human BM CD34? hematopoietic cells. J. Leukoc.
Biol. 85: 000–000; 2009.
Human bone marrow (BM) hematopoi-
Key Words: hematopoiesis ? myeloid cells ? Toll-like receptors
Hematopoiesis is regulated by a complex network of cytokines
and extracellular matrix provided by the bone marrow (BM).
The best-characterized biological function of cytokines is their
ability to regulate the survival, proliferation, and functional
activation of hematopoietic progenitors and mature cells .
Because of their limited lifespan, mature blood cells are con-
stantly generated from a pool of self-renewing progenitor cells
known as hematopoietic stem cells (HSC) through step-wise
developmental stages and commitment to myeloid or lymphoid
lineages [2, 3]. In addition to lineage-specific hematopoietic
cytokines, which instruct an uncommitted cell to differentiate
into a given cell type, hematopoiesis is regulated by intrinsic
genetic programs and interactions of stem/progenitor cells with
the BM microenvironment [2, 4]. Among the nonlymphohema-
topoietic cells, mesenchymal stem cells (MSC) have the ability
to support and maintain hematopoiesis in vitro . These cells
can also give rise to a variety of cell types including myocytes
(muscle), adipocytes (fat), fibroblasts (connective), endothelial,
and osteoblasts (bone) [6, 7]. Although MSC reside predomi-
nantly in the BM, they are also distributed through many other
tissues, where they are considered to serve as local sources of
dormant stem cells .
Previous studies have shown that the commitment of HSC/
progenitor cells to different lineages and progression within the
lineages require well-defined growth factors . Indeed, the
remarkable effects of certain cytokines such as erythropoietin,
G-CSF, stem cell factor, thrombopoietin, and others on a
variety of hematopoietic progenitor cells have highlighted the
significant and critical role that cytokines play in the prolifer-
ation and differentiation of hematopoietic CD34? cells [1, 9,
10]. However, recent studies have implicated TLR signaling in
these processes and have further emphasized the linkage be-
tween innate immunity and hematopoiesis [11–13]. In this
respect, we have shown that human BM CD34? cells ex-
pressed functional TLRs and associated signaling adaptor mol-
ecules. These included TLR9, TLR7, TLR8, TLR4, TLR2, as
well as MyD88 and Toll/IL-1R domain-containing adaptor-
inducing IFN-? adaptor molecules (ref.  and M. Sioud,
unpublished data). Human BM CD34? cells can differentiate
into myeloid cells just by stimulation with TLR7/8 or TLR2
synthetic agonists. More importantly, TLR signaling induced
the generation of dendritic cells (DC) capable of activating T
cells . Consistent with our findings in humans, mouse
hematopoietic progenitor cells expressed functional TLRs (e.g.,
TLR2, TLR4), and their activation resulted in the production of
monocytes and DC . Also, recent studies demonstrated the
expression of TLRs by MSC (refs. [14, 15] and M. Sioud,
unpublished data). Treatment of human MSC to TLR ligands
induced cytokine/chemokine gene expression and modulated
MSC function, again underlying a novel role of TLRs [14, 15].
1Correspondence: Department of Immunology, Institute for Cancer Re-
search, The Norwegian Radium Hospital, Montebello, N0310, Oslo, Norway.
Received October 21, 2008; revised January 16, 2009; accepted February 1,
0741-5400/09/0085-0001 © Society for Leukocyte Biology
Journal of Leukocyte Biology
Volume 85, June 2009
Uncorrected Version. Published on February 24, 2009 as DOI:10.1189/jlb.1008650
Copyright 2009 by The Society for Leukocyte Biology.
In addition to TLRs, vertebrates have evolved strategies to
sense pathogens such as bacteria in the cytosol [16–18].
Nucleotide oligomerization domain 1 (NOD1) and NOD2 are
members of the NOD-leucine-rich repeat (LRR) protein family,
whose members share a tripartite domain structure consisting
of a C-terminal peptide recognition domain, a central NOD
domain, and an N-terminus made up of protein–protein inter-
action domains, such as caspase recruitment domains or pyrin
domains [17, 18]. These proteins are also structurally related to
the nucleotide-binding site–LRR class of plant R gene prod-
ucts found in plants . Although their C-terminal LRR motif,
NOD1 and NOD2 recognize active entities of bacterial pepti-
doglycans, resulting in the activation of the transcription factor
NF-?B . Given the significance of NOD receptors in innate
immunity and cell signaling, in the present study, we investi-
gated their functionality in BM CD34? cells. We show that the
NOD2 signaling pathway can instruct BM hematopoietic
CD34? cells to express cytokines and growth factors and
differentiate into CD11c? cells capable of activating T cells.
Moreover, NOD2 and TLR2 activation increased intracellular
?-defensin levels, a potent antimicrobial peptide.
MATERIALS AND METHODS
Muramyl dipeptide (MDP; a NOD2 ligand) and control peptide were purchased
from InvivoGen (San Diego, CA, USA). Palmitoyl-3-cysteine-serine-lysine-4
(Pam3CSK4) was purchased from Calbiochem (San Diego, CA, USA). Anti-
CD11c-, anti-CD14-, anti-CD38-, anti-CD80-, anti-CD86-, and anti-CD206-
conjugated antibodies were purchased from DakoCytomation (Denmark). Con-
jugated anti-CD34, anti-CD105, and anti-NOD2 mAb (2D9) were purchased
from BD Biosciences (San Jose, CA, USA). Anti-NOD1 polyclonal antibodies
were purchased from Cell Signaling (Beverly, MA, USA). The CD34? selec-
tion kit was purchased from Miltenyi Biotec (Auburn, CA, USA). CD4 immu-
nomagenetic beads and related reagents were purchased from Dynal Invitrogen
(Norway). Total RNA isolation kit Trizol and cDNA synthesis kit were pur-
chased from Invitrogen (Carlsbad, CA, USA) and GE Healthcare (Waukesha,
WI, USA), respectively.
Isolation of BM CD34? and MSC cells
BM CD34? cells were obtained by iliac crest aspiration from normal adult
volunteers with informed consent of all donors. Local ethical committee
approval was received for the study with BM cells. BM mononuclear cells were
prepared by gradient centrifugation (Lymphoprep). Subsequently, CD34?
cells were positively selected using the CD34? kit according to the manufac-
turer’s instructions (Miltenyi Biotec). The cells were passed at least twice
through an LS column to obtain a pure CD34? cell population. The purity of
the CD34? cells isolated by this method was reproducibly ?97%, as deter-
mined by immunostaining and flow cytometric analyses.
Phenotypic analysis of NOD2-stimulated
BM CD34? cells were cultured in RPM medium supplemented with 5% FCS
in the absence or presence of MDP in a 5% CO2-humidified atmosphere for 6
days. Subsequently, they were double-stained with anti-CD11c in combination
with anti-CD14, anti-CD86, and anti-206, as described previously .
Briefly, aliquots of cultured cells 50 ? 105were stained with respective
fluorochrome-conjugated antibodies in PBS containing 1% FCS and 0.1%
sodium azide (FACS buffer) for 1 h at 4°C. Subsequently, the cells were
washed three times and resuspended in 300 ?l FACS buffer. Flow cytometry
was done on a FACSCalibur (Becton Dickinson, San Diego, CA, USA) with
CellQuest data acquisition and analysis software.
Cell treatments, cytokines, and ?-defensin
Freshly isolated BM CD34? cells were prepared and cultured in RPMI
medium in the absence or presence of various concentrations of MDP. After
overnight incubation, culture supernatants were collected, and cytokine levels
were measured by ELISA (R&D Systems, Minneapolis, MN, USA, and BD
Biosciences), according to the manufacturer’s instructions. To investigate the
cooperation between NOD2 and TLR, CD34? progenitor cells were stimulated
for 18 h with LPS (100 ng/ml) or Pam3CSK4(5 ?g/ml) in the presence of
various concentrations of MDP (0, 1, 5, or 10 ?g/ml). Subsequently, secreted
TNF-?, GM-CSF, and IL-1? were assayed by ELISA. Human ?-defensins
[human neutrophil peptides 1–3 (HNP1–3)] were measured by the ELISA test
kit (Hycult Biotechnology, The Netherlands).
Allogeneic MLR and cell morphology analysis
Allogeneic CD4? T cells were isolated from PBMC with CD4 Dynabeads? in
combination with Detachabead? reagent to ?98% purity, as determined by
flow cytometry. T cells (105cells/well) were cocultured in 96-well U-bottom
microtiter plates with various numbers of sorted and irradiated NOD2-derived
CD11c? cells. Proliferation of T cells was assessed by3H-thymidine incor-
poration, as described previously . To prepare the CD11c? cell subset,
BM CD34? cells were stimulated with MDP peptide for 6–7 days. Subse-
quently, cells were stained with anti-CD11c-FITC antibody for 60 min at 4°C,
and then positive cells were sorted using FACSDiva. In some experiments,
sorted CD11c? cells were spun down on microscopic slides by cytocentrifu-
gation (2?104 cells/0.1 ml), air-dried, and stained with Giemsa, and then the
slides were analyzed by light microscope.
Total RNA was isolated from cells using TRIzol (Invitrogen) and treated with
DNase I (Promega, Madison, WI, USA) to remove contaminating DNA. cDNA was
synthesized from 1 ?g total RNA using the first-strand cDNA synthesis kit and
oligo-dT primer in 15 ?l vol according to the manufacturer (GE Healthcare). PCR
was conducted in 50 ?l on 1/15 on the cDNA using 2 units of Tap polymerase
using the following human-specific primers: NOD1 forward, 5?-GTACGTCAC-
CAAAATCCTGGA-3?, and reverse, 5?-CAGTCCCCTTAGCTGTGATC-3?; NOD2
forward, 5?-CTGGCAAAGAACGTCATGCTA-3?, and reverse, 5?-CCTGGGATT-
GAATCTTGGGAA-3?; HNP1–3 forward*, 5?-CTGAGCCACTCCAGGCAAGA-3?,
and reverse*, 5?-GCTCAGCAGCAGAATGCCCA-3?; HNP1–3 forward, 5?-CTTGC-
human defensin 5 (HD5) forward, 5?-TGAGGCTACAACCCAGAAGC-3?, and re-
GAGAACCCTCACC-3?, and reverse, 5?-TGGCAATGTATGGGACACAC-3?; human
?-defensin 1 (HBD1) forward, 5?-TGAGTGTTGCCTGCCAGT-3?, and reverse, 5?-
TCTTCTGGTCACTCCCAG-3?; HBD2 forward, 5?-CATCAGCCATGAGGGTCT-3?,
and reverse, 5?-AGGCAGGTAACAGGATCG-3?; CD56 forward, 5?-CACTCAGGT-
GAAAATCCATC-3?, and reverse, 5?-CATCCCCCTTAAACTTTCTG-3?; CD177
forward, 5?-CACCAGATTCTTTCCCATTCTG-3?, and reverse, 5?-TCTAACGA-
CACGTCCCTGTC-3?; PU.1 forward*, 5?-CCGCCTGTACCAGTTCCTGTTG-3?, and
reverse*, 5?-CTTCTTGACCTCGCCCGTCTTG-3?; ?-actin forward, 5?-ATCTG-
GCACCACACCTTCTAC-3?, and reverse, 5?-CGTCATACTCCTGCTTGCTGATC-3?;
GAGGAGGGGAGATTCAG-3?. After 35 cycles, RT-PCR products were separated
on 1.5% agarose gels, visualized by staining with ethidium bromide, and photo-
graphed using the Molecular Imager ChemiDoc XRS system (Bio-Rad, Hercules,
CA, USA). The expression of ?-defensins 1–3 and PU.1 transcription factor was
analyzed by real-time PCR using SYBR Green PCR Master Mix (Invitrogen) and
ABI PRISM? 7700 equipment. GAPDH was used as an internal normalizer gene.
For each sample, comparative threshold (Ct) differences between control and
treated cells were calculated. The fold difference for each gene was calculated
using the ??Ctmethod; *, primers that were used in real-time PCR.
Western blot analysis
Cytoplasmic protein extracts were prepared from BM CD34? cells according
to standard protocols. Protein concentrations were measured by the Bio-Rad
protein assay kit. Protein lysates (10–50 ?g) were resolved on 10% SDS-
polyacrylamide gels and transferred subsequently to nitrocellulose by electro-
2Journal of Leukocyte Biology
Volume 85, June 2009
phoresis. Membranes were blocked with 5% nonfat dried milk in PBS con-
taining 0.1% Tween overnight. Subsequent to washing, membranes were
immunoprobed with mAb against the investigated proteins, followed by HRP-
conjugated rabbit or mouse secondary antibodies. Antibody–protein complexes
were visualized after exposure to X-ray film by ECL reagent.
Statistical significance was determined by a two-tailed unpaired Student t-test.
P values of ?0.05 were considered to indicate significance.
Expression of NOD1 and NOD2 by freshly
isolated BM CD34? cells
Although a number of studies have reported about the expres-
sion and functionality of NOD receptors in various human cell
types, to our knowledge, no information is available about their
function and expression in freshly isolated human BM CD34?
cells. Therefore, we have investigated the expression of NOD1
and NOD2 in CD34? cells by RT-PCR and Western blots.
Figure 1A shows that approximately 97% of the purified
CD34? cells expressed the CD34? marker. RT-PCR analysis
revealed a high level of NOD2 (Fig. 1B, as a representative
example). In general, NOD1 was undetectable; however, in
some experiments, a weak band was detected. To support the
RT-PCR data further, cytoplasmic protein extracts from
CD34? cells were analyzed by Western blots. Consistent with
the RT-PCR data, the CD34? cell subset expressed NOD2
protein (Fig. 1C).
Analysis of the functionality of NOD2 in CD34?
Signaling though NOD receptors induces the activation of the
receptor-interacting serine-threonine kinase, leading to NF-??
activation and cytokine production . To evaluate the func-
tionality of NOD2 in BM CD34? cells, the cells were stimu-
lated with different MDP concentrations, and cytokine contents
were measured in culture supernatants (Fig. 2). Stimulation
with MDP induced the secretion of TNF-? and GM-CSF as
assayed by ELISA. Subsequently, we investigated the outcome
of NOD2 activation pathway on cell differentiation by analyz-
ing the expression of CD11c, CD14, CD86, and CD206 mark-
ers. CD14 is expressed on monocytes and most macrophages,
but not on myeloid (m)DC or plasmacytoid (p)DC. CD11c is
expressed on monocytes and mDC, but not on pDC . When
compared with cells cultured in the presence of control peptide
(D–D isomer-inactive), MDP induced CD34? cell differentia-
tion along the myeloid lineage (Fig. 3A, as a representative
example). On Day 6, a significant fraction (30?8%) of MDP-
treated cells expressed the CD11c marker, whereas the level of
expression in control-peptide-treated cells was only 4 ? 2%
(P?0.01). Furthermore, a high proportion of the CD11c-posi-
tive cells also expressed the CD86 and CD206 markers (Fig.
3A and Table 1).
To evaluate the morphological characteristics of MDP-stim-
ulated cells, we treated CD34? cells for 6–7 days with MDP
peptide, and the CD11c? cell subset was FACS-sorted,
stained with Giemsa, and then analyzed by light microscope
(Fig. 3B as a representative example). The cells showed mono-
cytoid/macrophage characteristics with basophilic cytoplasm
and widespread vacuolization. A significant fraction of the cells
also showed dendritic morphology with multiple fine dendrites
as indicated by the arrows.
Consistent with the phenotypic and morphologic changes
shown above, MDP-treated cells up-regulated the expression of
the transcription factor PU.1 at mRNA and protein levels when
compared with cells treated with the control peptide (Fig. 3, C
and D). Previous studies have reported about the up-regulation
of PU.1 gene expression during myeloid differentiation .
Collectively, these data indicate that the NOD signaling path-
way is functional in BM CD34? cells, and its activation is
sufficient to induce cell differentiation.
Fig. 1. Expression of NOD1 and NOD2 in human BM CD34? progenitors
and MSC. (A) Purity of CD34? progenitor cells. Human BM immunomagnetic-
selected CD34? progenitor cells were stained with anti-human CD34? anti-
body and then analyzed by flow cytometry. Nonspecific staining was controlled
by isotype-matched, PE-conjugated antibody. SSC, Side-scatter. (B) Expres-
sion of NOD1 and NOD2 was determined by RT-PCR using cDNA from BM
CD34? cells. After 35 cycles, RT-PCR products were separated on 1.5%
agarose gels, visualized by staining with ethidium bromide. Comparable
amounts of cDNA were verified by amplification of ?-actin. The arrow indi-
cates the expected electrophoretic mobility of NOD1 and NOD2 PCR bands.
M, . (C) Protein levels were investigated by Western blots (50 ?g/lane) using
NOD1- and NOD2-specific antibodies. The arrows indicate the expected
electrophoretic mobility of NOD1 and NOD2.
Sioud and Fløisand
NOD2 and CD34? hematopoietic cells3
Having demonstrated that BM CD34? cells can respond to
the NOD2 agonist, next, we tested the ability of the NOD2-
derived CD11c? cell subset to activate T cells in a MLR
assay. In these experiments, CD34? cells were stimulated with
MDP for 6–7 days, and CD11c? and CD11c– cell subsets
were sorted, irradiated, and cocultured with allogeneic CD4?
T cells (Fig. 4). The CD11c? subset stimulated T cells to
proliferate as opposed to the CD11c– subset, indicating that
the NOD2 signaling pathway can license BM hematopoietic
CD34? cells to differentiate into functional innate immune
Cross-talk between TLR and NOD agonists in
In previous studies, we have shown that BM CD34? cells
express functional TLR2, TLR4, TLR7, and TLR8 [11, 13].
The finding of NOD2 expression prompted us to investigate
Fig. 2. Effects of NOD2 signaling on cytokine expression. Human BM
CD34? cells (105/well/200 ?l) were stimulated for 18 h with various concen-
trations of MDP. Subsequently, culture supernatants were collected, and
TNF-? and GM-CSF levels were measured by ELISA. Results are presented as
means ? SD of four independent experiments.
Fig. 3. Phenotypic and morphological anal-
ysis of MDP-stimulated CD 34? cells. (A)
Analysis of CD11c, CD14, CD86, and CD206
markers in control peptide- and MDP-stimu-
lated cells. Marker expression was determined
on Days 6–7 following stimulation with MDP
(5 ?g/ml) or control peptide (5 ?g/ml). Non-
specific staining was controlled by PE and
FITC isotype-matched antibodies. Percent-
ages of positive cells are indicated in each
quadrant. Results are representative of one of
three donors. The percentages of cell subsets
induced by MDP are summarized in Table 1.
(B) Morphologic examination of MDP-induced
CD11c? cells. Subsequent to 6–7 days in
culture, the CD11c? cell subset was sorted
and stained as described in Materials and
Methods. (C) NOD-stimulated cells up-regu-
lated the expression of the PU.1 transcription
factor. Cells were stimulated for 3 h with con-
trol peptide or MDP before total RNA prepa-
ration and PCR amplification for 35 cycles.
(D) CD34? cells were cultured in the pres-
ence of control peptide (Cont P) or MDP for
18 h, and then PU.1 expression was analyzed
by Western blots.
TABLE 1. Percentage of Cell Subsets Induced by NOD2
Signaling in BM CD34? Cells
2.8% ? 1%
1.5% ? 1%
1.5% ? 1%
1.5% ? 1%
14% ? 6%a
16% ? 5%a
18% ? 6%a
30% ? 8%a
BM CD34? cells were cultured in the presence of either control peptide
MDP, and then the expression of the indicated markers was analyzed on Days
6–7 by flow cytometry. Results are expressed as means ? SD.aSignificance as
compared with controls (P?0.05).
4 Journal of Leukocyte Biology
Volume 85, June 2009
whether it synergizes with TLR signaling at the level of cyto-
kines, and cytokine responses to MDP and LPS were signifi-
cantly higher than those obtained with MDP or LPS alone (Fig.
5, A and B). For example, stimulation of CD34? cells with
LPS (a TLR4 ligand) and MDP (5 ?g/ml) resulted in a fourfold
increase in IL1-? levels compared with cells treated with only
LPS or MDP (P?0.01). Similarly, TNF-?, IL-1?, and GM-CSF
protein levels were increased significantly when MDP was
combined with Pam3CSK4, a TLR2 ligand (P?0.02; Fig. 5C),
indicating that there is enhancement of TLR signaling by
MDP/NOD2 activation in freshly isolated human BM CD34?
cells. To rule out that the response to MDP peptide was not a
result of the presence of endothelial cells, we have analyzed
the expression of CD105 (endoglin), a type I transmembrane
protein that is highly expressed on human vascular endothelial
cells . No significant expression was detected by flow
cytometry in the analyzed samples (Fig. 5D).
BM CD34? cells up-regulate intracellular ?-
defensins 1–3 in response to NOD2 agonists
Unlike adaptive immunity, innate immunity confers broad pro-
tection against pathogens without previous exposure, and most
multicellular organisms depend on it to fight microbial infec-
tions. We hypothesized that BM CD34? or MSC may express
and secrete defensins in response to bacterial infections, given
the imperative need to neutralize microbes rapidly. First, we
analyzed the expression of ?- and ?-defensins by RT-PCR.
CD34? cells expressed ?-defensins 1–3 (HNP; Fig. 6A, Lane
1). However, they did not express ?-defensin HD5 and ?-de-
fensins HBD1 and HBD2 (Fig. 6A, Lanes 2–4, respectively).
As shown in Figure 6A, MDP stimulation did not increase the
?-defensin 1–3 mRNA levels (Lanes 1 and 6) or induce HD5,
HBD1, and HBD2 gene expression significantly (Lanes 7–9,
respectively). Real-time PCR quantification showed a moder-
ate increase in ?-defensin 1–3 mRNA levels in response to
MDP, although this increase was found not to be significant
To rule out that ?-defensin 1–3 expression is not originating
from contaminating cells such as neutrophils and NK cells,
CD34? cells were purified further using FACS (?99.8 pure).
Comparable with the immnunomagnetic-selected cells, sorted
CD34? cells expressed ?-defensins 1–3 (Fig. 6C, Lanes 1 and
2). They did not express CD56 (Lane 3) or CD177 (Lane 4),
markers for NK cells and neutrophils, respectively. The
Fig. 4. Analysis of the allogeneic stimulatory capacity of the NOD2-induced
CD11c? cell subset. BM CD34? cells were stimulated for 7 days with MDP
(5 ?g/ml). Subsequently, the cells were stained with anti-CD11c mAb, and the
CD11c? and CD11c– cell subsets were sorted, irradiated, and then cocultured
for 6 days with purified, allogeneic CD4 T cells (105) in 96-well U-bottom
plates. The numbers of sorted and irradiated cells were 3 ? 103, 1 ? 104, and
2 ? 104in Lanes 1–3, respectively. Cell proliferation was measured by
[3H]-thymidine incorporation, and data are expressed as the mean ? SD of
three independent experiments.
Fig. 5. NOD2 cooperates with TLR signaling. Human
BM CD34? (5?104) cells were stimulated with various
MDP concentrations without (A) or with a combination
with LPS (100 ng/ml; B). After 18 h incubation time at
37°C, culture supernatants were collected, and cytokine
contents were determined by ELISA. (C) The cells were
also stimulated with Pam3CSK4(5 ?g/ml) in combination
with various MDP concentrations. Results are presented
as means ? SD of four independent experiments. (D) Flow cytometry analysis of endoglin (CD105) expression in the CD34? purified cells.
Sioud and Fløisand
NOD2 and CD34? hematopoietic cells5
CD34? CD38– cells, which are enriched for most immature
hematopoietic progenitors, expressed ?-defensins 1–3 as de-
tected by PCR (Fig. 6D, Lanes 1 and 2).
As defensins have been described as secreted peptides
, next, we analyzed ?-defensin 1–3 levels in culture
supernatants using ELISA. MDP stimulation did not induce
a significant amount of ?-defensin 1–3 secretion (Fig. 6E).
One of the reasons for the failure of secretion could be
related to the poor ability of NOD2 ligands to promote
?-defensin release in BM CD34? cells. Therefore, an ad-
ditional experiment was performed to validate the intracel-
lular expression of ?-defensins1–3 in CD34? cells. Anal-
ysis of protein extracts by ELISA confirmed the expression
of ?-defensins 1–3 (Fig. 6F). NOD2 stimulation alone re-
sulted in intracellular up-regulation of ?-defensin 1–3 pro-
tein levels relative to those obtained in unstimulated cells
(P?0.05). The levels of ?-actin in both samples were com-
parable. Furthermore, no reactivity was detected when cy-
toplasmic proteins from MSC were used, confirming the
specificity of the immunoreaction (data not shown).
The innate immune system contains several classes of pattern
recognition receptors (PRRs), including TLRs, Nod-like recep-
tors (NLRs), and retinoic acid-inducible gene protein-1 recep-
tors (RLRs) [17, 18, 25, 26]. The detection of pathogen-
associated molecular patterns by TLRs, NLRs, and RLRs
activates multiple proinflammatory signaling pathways to
mount antiviral/bacterial responses to invading pathogens. Re-
cent studies demonstrated that TLRs not only form the link
between microbial recognition by innate immunity and activa-
tion of adaptive immunity but also contribute in production of
myeloid cells from BM CD34? cells in the absence of external
growth and differentiation factors [11–13]. In the present study,
we show that BM hematopoietic CD34? cells express a func-
tional NOD2 receptor. MDP-stimulated cells produced cyto-
kines and gave rise to innate immune cells such as monocytes/
macrophages (CD11c?/CD14?) and DC (CD11c?/CD14–).
BM CD34? cells not only produce cytokines, growth factors,
and innate immune cells but also up-regulate intracellular
Fig. 6. Expression of defensins in BM CD34? cells. (A) Immunomicrobead-purified CD34? cells (?97% pure) were stimulated with MDP peptide (10 ?g/ml)
for 10 h. Subsequently, total RNA was purified, and RT-PCR was performed for 35 cycles to analyze the expression of ?-defensins 1–3 (Lanes 1 and 6), HD5 (Lanes
2 and 7), HBD1 (Lanes 3 and 8), HBD2 (Lanes 4 and 9), and ?-actin (Lanes 5 and 10). Results are representative of one of four donors. (B) Real-time PCR
quantification of ?-defensins in control peptide- and MDP-stimulated CD34? cells. (C) FACS-sorted CD34? cells expressed ?-defensins 1–3. In these
experiments, immunomicrobead-purified CD34? progenitor cells were stained with PE-conjugated anti-human CD34? mAb and purified further using cell sorting.
Subsequently, total RNA was purified, and the expression of ?-defensins 1–3 (Lanes 1 and 2), CD56 (Lane 3), CD177 (Lane 4), and ?-actin (Lane 5) was analyzed
by RT-PCR. Two sets of PCR primers were used to amplify ?-defensins 1–3. (D) FACS-sorted CD34? CD38– cells expressed ?-defensins 1–3, as detected by
RT-PCR (Lanes 1 and 2). Lane 3 ? ?-actin. In these experiments, 40 cycles were performed. (E) Secretion of ?-defensins 1–3. CD34? cells were stimulated for
18 h with MDP or control peptide, and then ?-defensin 1–3 levels in culture supernatants were assayed by ELISA (Hycult Biotechnology). (F) Intracellular
?-defensin 1–3 levels in control- and MDP-stimulated CD34? cells. Cell pellets derived from the experiments shown in E were lysed in water and sonicated, and
then total protein contents were determined using the Bio-Rad protein assay kit. Approximately 10 ?g protein extracts from each sample were added to 200 ?l
PBS buffer, and then ?-defensin 1–3 protein levels were quantified by ELISA as in E (100 ?l/well). Results are presented as the mean ? SD of three independent
experiments. To control for equal protein content, 10 ?g protein lysates from each sample were separated on SDS-PAGE, and the expression of ?-actin was analyzed
6 Journal of Leukocyte Biology
Volume 85, June 2009
?-defensin 1–3 levels, suggesting that they might kill patho-
genic bacteria directly. In contrast to NOD2, the expression of
NOD1 was weak. It would be interesting to investigate whether
NOD1 is expressed specifically in distinct hematopoietic
The provision of blood cells and maintenance of hematopoi-
esis require cellular proliferation with expansion of stem/pro-
genitor cells but suppression of proliferation and quiescence of
these cells when blood cells are not needed. Among the stim-
ulatory cytokines induced by MDP in BM CD34? cells are
GM-CSF and TNF-?, which have been identified as key factors
for the induction of myeloid precursors . Previously, we
have found that neutralization of TNF-? activity inhibited the
induction of the lineage-negative DC (CD11c?/CD14–) but
not the monocyte-enriched subset (CD11c?/CD14?) induced
by TLR7/8 agonists. In contrast, the neutralization of GM-CSF
activity showed no significant effects on cell differentiation,
suggesting that intrinsic signals are involved in directing the
fate of hematopoietic CD34? cells. The involvement of intrin-
sic pathways is highlighted by the observation that MDP-
stimulated CD34? cells up-regulated the expression of the
PU.1 transcription factor quickly. More importantly, the MDP-
induced CD11c? cells are capable of activating T cells. Re-
cently, Larcher and colleagues  found that mDC derived
from human Herpes virus 8-infected umbilical cord blood
CD34? cells induced higher proliferation of allogeneic T cells
than those obtained from noninfected cells, thus suggesting
that infection not only modifies hematopoietic cell fates but
also the quality of the generated innate immune cells.
Host-pathogen interactions presumably involve several
PRRs rather then one single receptor as a result of the numer-
ous TLR and NOD ligand motifs present on a given pathogen.
Our data indicate that NOD2 is essential in the detection of
bacterial MDP and is capable of activating cytokine expression
in hematopoietic CD34? cells directly or in collaboration with
TLRs. The latter finding is consistent with previous reports
demonstrating synergistic responses in blood immune cells
. Dual signaling via TLR and NOD pathways may be
necessary for efficient innate immune responses. In some re-
ports, however, NOD2 was found to function as a negative
regulator of TLR2 signaling . In testing the co-response of
NOD2 and TLR2 signaling in BM CD34? cells for the pro-
duction of inflammatory cytokines, including TNF-? and IL-
1?, we failed to detect any significant attenuation of the
response to the TLR2 ligand. In contrast, we observed a
synergistic effect in BM CD34? cells. Therefore, any potential
negative role of NOD2 in the TLR2 response is not a common
function. In contrast to CD34? cells, MSC from the same
donors expressed the NOD1 but not NOD2 receptor (data not
shown). Also, significant changes in gene expression in
CD34? cells and MSC in response to NOD and TLR ligands
were confirmed by microarray analysis (M. Sioud et al., in
preparation). Collectively, the expression data indicated that
TLRs and NOD receptors have an important biological role in
In addition to the demonstration that BM CD34? cells
express a functional NOD2 signaling pathway, a significant
observation in this study is the finding that BM CD34? cells
can express ?-defensins 1–3 at mRNA and protein levels,
further underlying their role in innate immunity. Defensins are
small (3–6 kDa), cationic, and cysteine-rich peptides with a
broad antimicrobial spectrum . Human ?-defensins 1–3
and 4 are expressed by neutrophils, whereas ?-defensins 5 and
6 are expressed by small intestine Pameth cells [32–34].
Beside their activity as effector molecules of the innate immu-
nity, defensins link innate and adaptive immunity by acting as
chemotactic factors for T cells, monocytes, and immature DC
. In NOD2-activated CD34? cells, ?-defensins 1–3 may
be more relevant in the cytoplasm to interfere with intracellular
bacteria. In line with this observation, a study showed that
infection of CD34? cells with Listeria monocytogenes or Yes-
ninia enterocolitica did not kill the cells, but instead, it in-
creased cell number . These pathogenic bacteria are known
to kill host cells upon encounter. In agreement with this
finding, MDP treatment also increases cell number.
In conclusion, the NOD2 signaling pathway alone was found
to be sufficient to drive BM CD34? cells to secrete cytokines
and differentiate into innate effector cells. In addition to TLRs,
NOD2 may contribute to BM CD34? innate immunity by
mediating the expression of antimicrobial peptides. Further
studies should help in understanding how BM CD34? cells
defend themselves against bacteria and whether they can man-
age this task without having the adaptive immune system.
This work is supported in part by a grant from the Norwegian
Cancer Society to M. S. The authors thank Dr. Anne Dybwad
for editing the manuscript, Lise Forfang for technical assis-
tance with flow cytometry, and Tommy Karlsen for providing
MSC. The authors declare no conflict of interest.
1. Zhu, J., Emerson, S. G. (2002) Hematopoietic cytokines, transcription
factors and lineage commitment. Oncogene 21, 3295–3313.
2. Kaushansky, K. (2006) Lineage-specific hematopoietic growth factors.
N. Engl. J. Med. 354, 2034–2045.
3. Loose, M., Swiers, G., Patient, R. (2007) Transcriptional networks regu-
lating hematopoietic cell fate decisions. Curr. Opin. Hematol. 14, 307–
4. Li, Z., Li, L. (2006) Understanding hematopoietic stem-cell microenviron-
ments. Trends Biochem. Sci. 31, 589–595.
5. Li, Y. P., Paczesny, S., Lauret, E., Poirault, S., Bordigoni, P., Mekhloufi,
F., Hequet, O., Bertrand, Y., Ou-Yang, J. P., Stoltz, J. F., Miossec, P.,
Eljaafari, A. (2008) Human mesenchymal stem cells license adult CD34?
hemopoietic progenitor cells to differentiate into regulatory dendritic cells
through activation of the Notch pathway. J. Immunol. 180, 1598–1608.
6. Le Blanc, K., Ringden, O. (2006) Mesenchymal stem cells: properties and
role in clinical bone marrow transplantation. Curr. Opin. Immunol. 18,
7. Kawada, H., Fujita, J., Kinjo, K., Matsuzaki, Y., Tsuma, M., Miyatake, H.,
Muguruma, Y., Tsuboi, K., Itabashi, Y., Ikeda, Y., Ogawa, S., Hotta, T.,
Ando, K., Fukuda, K. (2004) Nonhematopoietic mesenchymal stem cells
can be mobilized and differentiate into cardiomyocytes after myocardial
infarction. Blood 104, 3581–3587.
8. Stagg, J. (2008) Mesenchymal stem cells in cancer. Stem Cell Rev. 4,
9. Ulich, T. R., del Castillo, J., McNiece, I. K., Yi, E. S., Alzona, C. P., Yin
S. M., Zebo, K. M. (1991) Stem cell factor in combination with granulocyte
colony-stimulating factor (CSF) or granulocyte-macrophage CSF synergis-
tically increases granulopoiesis in vivo. Blood 78, 1954–1962.
Sioud and Fløisand
NOD2 and CD34? hematopoietic cells7
10. Encabo, A., Solves, P., Mateu, E., Sepu ´lveda, P., Carbonell-Uberos, F.,
Minana, M. D. (2004) Selective generation of different dendritic cell
precursors from CD34? cells by interleukin-6 and interleukin-3. Stem
Cells 22, 725–740.
11. Sioud, M., Fløisand, Y., Forfang, L., Lund-Johansen, F. (2006) Signaling
through Toll-like receptor 7/8 induces the differentiation of human bone
marrow CD34? progenitor cells along the myeloid lineage. J. Mol. Biol.
12. Nagai, Y., Garrett, K. P., Ohta, S., Bahrun, U., Kouro, T., Akira, S.,
Takatsu, K., Kincade, P. W. (2006) Toll-like receptors on hematopoietic
progenitor cells stimulate innate immune system replenishment. Immunity
13. Sioud, M., Fløisand, Y. (2007) TLR agonists induce the differentiation of
human bone marrow CD34? progenitors into CD11c? CD80/86? DC
capable of inducing a Th1-type response. Eur. J. Immunol. 37, 2834–
14. Tomchuck, S. L., Zwezdaryk, K. J., Coffelt, S. B., Waterman, R. S., Danka,
E. S., Scandurro, A. B. (2008) Toll-like receptors on human mesenchymal
stem cells drive their migration and immunomodulating responses. Stem
Cells 26, 99–107.
15. Liotta, F., Angeli, R., Cosmi, L., Filı `, L., Manuelli, C., Frosali, F.,
Mazzinghi, B., Maggi, L., Pasini, A., Lisi, V., Santarlasci, V., Consoloni,
L., Angelotti, M. L., Romagnani, P., Parronchi, P., Krampera, M., Maggi,
E., Romagnani, S., Annunziato, F. (2008) Toll-like receptors 3 and 4 are
expressed by human bone marrow-derived mesenchymal stem cells and
can inhibit their T-cell modulatory activity by impairing Notch signaling.
Stem Cells 26, 279–289.
16. Janeway Jr., C., Medzhitov, R. (2002) Innate immune recognition. Annu.
Rev. Immunol. 20, 197–216.
17. Inohara, Chamaillard, McDonald, C., Nun ˜ez, G. (2005) NOD-LRR pro-
teins: role in host-microbial interactions and inflammatory disease. Annu.
Rev. Biochem. 74, 355–383.
18. Kanneganti, T. D., Lamkanfi, M., Nu ´nez, G. (2007) Intracellular NOD-like
receptors in host defense and disease. Immunity 27, 549–559.
19. Baker, B. J., Friedman, A. R. (2007) The evolution of resistance genes in
multi-protein plant resistance systems. Curr. Opin. Genet. Dev. 17, 493–
20. Steiner, H. (2004) Peptidoglycan recognition proteins: on and off switches
for innate immunity. Immunol. Rev. 198, 83–96.
21. Wu, L., Liu, Y. J. (2007) Development of dendritic-cell lineages. Immu-
nity 26, 741–750.
22. Miranda, M. B., Johnson, D. E. (2007) Signal transduction pathways that
contribute to myeloid differentiation. Leukemia 21, 1363–1377.
23. Middleton, J., Americh, L., Gayon, R., Julien, D., Mansat, M., Mansat, P.,
Anract, P., Cantagrel, A., Cattan, P., Reimund, J.M., Aguilar, L., Amalric,
F., Girard, J.P. (2005) A comparative study of endothelial cell markers
expressed in chronically inflamed human tissues: MECA-79, Duffy anti-
gen receptor for chemokines, von Willebrand factor, CD31, CD34, CD105
and CD146. J. Pathol. 206, 260–268.
24. Ganz, T. (2003) Defensins: antimicrobial peptides of innate immunity.
Nat. Rev. Immunol. 3, 710–720.
25. Sioud, M. (2006) Innate sensing of self and non-self RNAs by Toll-like
receptors. Trends Mol. Med. 12, 167–176.
26. Yoneyama, M., Kikuchi, M., Matsumoto, K., Imaizumi, T., Miyagishi, M.,
Taira, K., Foy, E., Loo, Y-M., Gale Jr., M., Akira, S., Yonehara, S., Kato,
A., Fujita, T. (2005) Shared and unique functions of the DExD/H-box
helicases RIG-I, MDA5, and LGP2 in antiviral innate immunity. J.
Immunol. 175, 2851–2858.
27. Lardon, F., Snoeck, H-W., Berneman, Z. N., van Tendeloo, V. F. I., Nijs,
G., Lenjou, M., Henckaerts, E., Boeckstaens, C. J., Vandenabeele, P.,
Kestens, L. L., Van Bockstaele, K. R., Vanham, G. L. (1997) Generation
of dendritic cells from bone marrow progenitors using GM-CSF, TNF-?,
and additional cytokines: antagonistic effects of IL-4 and IFN-? and
selective involvement of TNF-? receptor-1. Immunology 91, 553–559.
28. Larcher, C., Nguyen, V. A., Fu ¨rhapter, C., Ebner, S., So ¨lder, E., Sto ¨ssel,
H., Romai, N., Sepp, N. (2005) Human herpesvirus-8 infection of umbil-
ical cord-blood-dervied CD34? stem cells enhances the immunostimula-
tory function of their dendritic cell progeny. Exp. Dermatol. 14, 41–49.
29. van Heel, D. A., Ghosh, S., Butler, M., Hunt, K., Foxwell, B. M., Mengin-
Lecreulx, D., Playford, R. J. (2005) Synergistic enhancement of Toll-like
receptor responses by NOD1 activation. Eur. J. Immunol. 35, 2471–2476.
30. Watanabe, T., Kitani, A., Murray, P. J., Strober, W. (2004) NOD2 is a
negative regulator of Toll-like receptor 2-mediated T helper type 1 re-
sponses. Nat. Immunol. 5, 800–808.
31. Menendez, A., Finlay, B. B. (2007) Defensins in the immunology of
bacterial infections. Curr. Opin. Immunol. 19, 385–391.
32. Ouellette, A. J., Selsted, M. E. (1996) Paneth cell defensins: endogenous
peptide components of intestinal host defense. FASEB J. 10, 1280–1289.
33. Ayabe, T., Satchell, D. P., Wilson, C. L., Parks, W. C., Selsted, M. E.,
Ouellette, A. (2000) Secretion of microbicidal ? defensins by interstinal
Paneth cells in response to bacteria. Nat. Immunol. 1, 113–118.
34. Ganz, T., Selsted, M. E., Szklarek, D., Harwig, S. S., Daher, K., Bainton,
D. F., Lehrer, R. I. (1985) Defensins: natural peptide antibiotics of human
neutrophils. J. Clin. Invest. 76, 1427–1435.
35. Yang, D., Chen, Q., Chertov, O., Oppenheim, J. J. (2000) Human neutro-
phil defensins selectively chemoattract naive T cells and immature den-
dritic cells. J. Leukoc. Biol. 68, 9–13.
36. Kolb-Ma ¨urer, A., Weissinger, F., Kurzai, O., Ma ¨urer, M., Wilhelm, M.,
Goebel, W. (2004) Bacterial infection of human hematopoietic stem cells
induces monocytic differentiation. FEMS Immunol. Med. Microbiol. 40,
8 Journal of Leukocyte Biology
Volume 85, June 2009