The FASEB Journal Research Communication
CXCL12/CXCR4 signaling in the osteoblast regulates
the mesenchymal stem cell and osteoclast
lineage populations
Mohammad Shahnazari,*, Vivian Chu,* Thomas J. Wronski, Robert A. Nissenson,*,
and Bernard P. Halloran*,,1
*Division of Endocrinology, Veterans Affairs Medical Center, and Department of Medicine,
University of California, San Francisco, California, USA; and Department of Physiological Sciences,
University of Florida, Gainesville, Florida, USA
ABSTRACT The chemokine CXCL12 and its recep-
tor CXCR4 play a key role in regulation of hematopoi-
etic stem cells and cell migratory function during
morphogenesis. Osteoblasts express both the ligand
and the receptor, but little is known about the role of
CXCL12-CXCR4 signaling in maintaining skeletal ho-
meostasis. Using Cre-Lox technology to delete CXCR4
in mature osteoblasts in mice, we show here a signifi-
cant decrease in bone mass and alterations in cancel-
lous bone structure. CXCR4 gene ablation increased
the number of colony-forming units (CFU), CFU-posi-
tive for alkaline phosphatase (CFU-AP), and mineral-
izing nodules in bone marrow stromal cell (BMSC)
cultures. The adipocyte precursor population de-
creased in BMSCs harvested from the KO animals. The
nonadherent population of BMSCs harvested from the
long bone diaphysis of KO animals formed more
osteoclasts, a finding that was associated with increased
circulatory levels of pyridinoline, a marker of bone
resorption. Our data show that osteoblast-specific
CXCR4 deletion has profound effects on the mesen-
chymal stem cell pool and allocation to the osteoblastic
and adipocytic cell lineages. They also show that
CXCL12/CXCR4 signaling in the mature osteoblast can
feedback to regulate the osteoclast precursor pool size
and play a multifunctional role in regulating bone
formation and resorption.Shahnazari, M., Chu, V.,
Abbreviations: ALP, alkaline phosphatase; BFR/BS, bone
formation rate/bone surface; BMSC, bone marrow stromal
cell; BV/TV, bone volume/total volume; CFU, colony-form-
ing unit; CFU-AP, colony-forming unit positive for alkaline
phosphatase; Col2.3, collagen 2.3; CXCR4, C-X-C chemokine
receptor type 4; CXCL12, C-X-C chemokine ligand 12; GFP,
green fluorescent protein; HSC, hematopoietic stem cell; KO,
knockout; M-CSF, macrophage colony-stimulating factor;
MSC, mesenchymal stem cell; Ob.S, osteoblast surface; Ob.S/
BS, osteoblast surface/bone surface; Oc.S, osteoclast surface;
Oc.S/BS, osteoclast surface/bone surface; OPG, osteoprote-
gerin; P1NP, N-terminal propeptide of type 1 procollagen;
RANKL, receptor activator of nuclear factor-B ligand; ROI,
region of interest; ; Runx2, runt-related transcription factor 2;
SDF-1, stromal derived factor 1; SMI, structure model index;
TFJ, tibial-fibular junction; TRAP, tartrate-resistant acid phos-
phatase; WT, wild type
0892-6638/13/0027-3505 FASEB
Downloaded from www.fasebj.org to IP 137.132.250.12. The FASEB Journal Vol.27, No.9 , pp:3505-3513, August, 2017
Wronski, T. J., Nissenson, R. A., Halloran, B. P. CXCL12/
CXCR4 signaling in the osteoblast regulates the mesenchy-
mal stem cell and osteoclast lineage populations. FASEB J.
27, 35053513 (2013). www.fasebj.org
Key Words: bone chemokine skeletal homeostasis
The mesenchymal stem cells (MSCs) that give rise to
osteoblasts and the hematopoietic stem cells (HSCs)
that give rise to osteoclasts are associated with one
another in microenvironmental structures or niches (1,
2). The niche is composed not only of the stem cells but
also numerous other cell types within the bone marrow
stromal cell (BMSC) population, as well as endothelial
cells and specialized extracellular matrices that sustain
and regulate their activity.
C-X-C chemokine ligand 12 [CXCL12; also known as
stromal derived factor 1 (SDF-1)], is a member of a
large family of structurally related chemoattractive cy-
tokines that binds to its receptor C-X-C chemokine
receptor type 4 (CXCR4) and plays a key role in
maintenance of the bone marrow niche (3, 4). CXCL12
is expressed by BMSCs, cells of the osteoblast lineage,
and perivascular reticular and endothelial cells, but not
hematopoietic cells (5). Expression of CXCL12 in the
osteoblast lineage has been reported to occur in both
immature and mature cells (3). The ligand has broad
effects on cell proliferation and differentiation and acts
as a chemotactic agent to direct cell migration (6 8).
Niche formation and activity appear to be guided, in
part, by microenvironmental gradients of CXCL12.
Hematopoietic stem cell activity, including osteoclast
recruitment, chemotaxis, fusion, cell survival, tartrate-
resistant acid phosphatase (TRAP) activity, and bone
resorption are highly regulated by CXCL12/CXCR4
signaling (9 11). Global ablation of CXCL12 or
1
Correspondence: Division of Endocrinology, Veterans Af-
fairs Medical Center, 4150 Clement St., San Francisco, CA
94123, USA. E-mail: bernard.halloran@ucsf.edu
doi: 10.1096/fj.12-225763
3505
CXCR4 is lethal in utero with abnormalities in bone and
marrow formation, the immune system, the heart,
cerebral development, and formation of the gastroin-
testinal tract (8, 1214).
Despite the prominent role that CXCL12 plays in
formation and maintenance of the bone marrow niche
and its importance in regulating HSC activity and
osteoclastogenesis, much less is known of its role in
osteoblastogenesis. Osteoblasts express CXCR4, and a
recent study by Zhu et al. (15) in mice reported a
significant impairment in osteoblast development and
bone formation when CXCR4 was conditionally inacti-
vated in osteoprecursor cells, concluding that CXCR4
signaling is necessary for normal bone formation. In
addition, the osteogenic differentiation of mesenchy-
mal and marrow stromal cells in response to bone
morphogenetic protein-2 was suppressed when CXCL12/
CXCR4 signaling was inhibited (9).
In the present study, we used Cre-Lox technology to
delete CXCR4 in mature osteoblasts. Our findings
suggest that CXCL12/CXCR4 signaling in the osteo-
blast regulates the mesenchymal stem cell osteopro-
genitor and osteoclast precursor pool sizes, and osteo-
blast/osteoclast recruitment and translocation to the
bone surface. Our data reveal novel signaling para-
digms and regulatory networks that expand our under-
standing of the complex mechanisms that are involved
in maintaining skeletal homeostasis.
Figure 1. Deletion of CXCR4 under control of the col 11
2.3-kb promoter fragment. A) Representative genotyping
MATERIALS AND METHODS PCR for Cre CXCR4fl/fl and the Cre transgene (Cre) and
excision product (Exc) relative to WT CXCR4 mice (WT).
Successful CXCR4 ablation was judged by the presence of the
Animals KO-only allele, the Cre recombinase gene, and excision
product. B) Representative images for immunoreactivity of
Heterozygotic mice (C57BL/6) carrying 2 loxP sites flanking osteoblasts to CXCR4 monoclonal antibody in WT and KO
exon 2 of the Cxcr4 gene (kindly provided by Dr. Yong-Rui mice. Arrows indicate osteoblasts on endosteal (panel 1) and
Zou, Columbia University, New York, NY, USA) were bred trabecular (panel 2) bone surfaces examined in 4-m-thick
with transgenic mice carrying the Cre-recombinase under the sections at 63 view.
control of a collagen 1a1 2.3-kb promoter fragment [collagen
2.3 (Col2.3)-Cre; kindly provided by Dr. Barbara Kream,
University of Connecticut]. Heterozygotic Cre mice (CXCR4fl/wt) second set of CXCR4-specific primers (forward, 5=-CACTACGCAT-
were bred with heterozygotic Cre mice (CXCR4fl/wt-Cre) GACTCGAAATG-3=, and reverse, 5=-CCTCGGAATGAAGAGAT-
to generate conditional-knockout (KO) mice (Col2.3-Cre- TATGC-3=) was used to detect deletion of the floxed allele (190-bp
CXCR4fl/fl) and littermate controls (CXCR4wt/wt-Cre, product).
CXCR4fl/fl-Cre, CXCR4fl/wt-Cre). Standard mouse chow
and water were provided ad libitum, and all experiments Immunohistochemical analysis
were carried out using 6-wk-old mice. Studies were con-
ducted in accordance with the U.S. National Institutes of
Health Guide for the Care and Use of Laboratory Animals To determine the presence of CXCR4 in osteoblasts in bone,
and were approved by the Animal Care and Use Committee immunohistochemistry was performed on 4-m paraffin-
at the Veterans Affairs Medical Center, San Francisco. embedded bone sections from cortical midshaft and the distal
femoral metaphysis. Paraffin sections were analyzed using a
monoclonal rat anti-mouse CXCR4 antibody (R&D Systems,
Genotyping Minneapolis, MN; dilution 1:50) and detected using R&D
Systems cell and tissue staining kit based on formation of
At 3 wk of age, DNA was extracted from tail tissue and avidin-biotin complex with the primary antibody. Visualiza-
amplified by PCR using a Sigma-Aldrich REDExtract-N-Amp tion was based on enzymatic conversion of a chromogenic
tissue PCR kit (Sigma-Aldrich, St. Louis, MO, USA). Primers substrate 3,3-diaminobenzidine into a brown precipitate by
(forward, 5=-GCAAAACAGGCTCTAGCGTTCG-3=, and re- horseradish peroxidase at the sites of antigen localization.
verse, 5=-CTGTTTCACTATCCZGGTTACGG-3=) were used to Sections were counterstained with Mayers hematoxylin (Sig-
detect the Cre recombinase gene (560-bp product; Fig. 1A), ma-Aldrich, St. Louis, MO, USA), mounted in Permount
CXCR4 gene [forward, 5=-CACTACGCATGACTCGAAATG-3=, medium (Fisher Scientific, Pittsburgh, PA, USA), and in-
and reverse, 5=GTGTGCGGTGGTATCCAGC-3=; wild-type (WT) spected on a Zeiss AX10 microscope (Carl Zeiss, Oberkochen,
317-bp product] or the loxP-inserted allele (437-bp product). A Germany) at 63 view.

3506 Vol. 27 September 2013 The FASEB Journal www.fasebj.org SHAHNAZARI ET AL.
Downloaded from www.fasebj.org to IP 137.132.250.12. The FASEB Journal Vol.27, No.9 , pp:3505-3513, August, 2017
Osteoblast culture
Tibiae and femurs were cleaned of adherent tissue, and
diaphyseal BMSCs were harvested and plated at 1.3 105
nucleated cells/cm2 in 100-mm dishes, as described previ-
ously (16, 17). On d 2, nonadherent cells were removed and
were cultured under osteoclast induction conditions, as de-
scribed below. The culture medium in the adherent cells was
changed to secondary medium (MEM supplemented with
10% FBS, 1% P/S antibiotics, 0.1% fungizone, 50 g/100 ml
l-ascorbic acid, and 3 mM -glycerophosphate) to induce
osteoblastogenesis. Subsequent medium changes were per-
formed every 2 d for up to 28 d. The number of colony-
forming units (CFUs), alkaline phosphatase (ALP)-positive
CFUs (CFU-AP) with diameter 3 mm, and mineralizing
nodules were quantified as described previously (18 20).
Osteoclast culture
The nonadherent cell fraction from the BMSC cultures was
removed on d 2 of culture and washed with PBS. The cells
were suspended in PBS, counted using a hemocytometer, and
seeded into 24-well tissue culture plates at 1 106 cells/well.
The cultures were carried for 6 d in the stromal cell culture
medium supplemented with RANK-L (40 ng/ml) and M-CSF
(10 ng/ml). Medium was changed every 2 d, and on d 6, cells
were washed twice with PBS and stained for TRAP using a
commercial kit from Sigma-Aldrich. Dark, reddish-purple
multinucleated cells (3 nuclei) were counted as TRAP
osteoclasts.
Adipocyte culture
BMSCs harvested from tibiae and femurs were plated at 20
106 nucleated cell/well in a 6-well plate in a primary medium
(MEM supplemented with 10% FBS, 1% P/S antibiotics,
0.1% fungizone). On d 5 of the culture, the nonadherent
cells were removed, and the medium was replaced and
changed every 2 d. On d 10, the cells were treated with 1 M
rosiglitazone, 1 M dexamethasone, 5 g/ml insulin, and 500
M 3-isobutyl-1 methylxanthine; and on d 12 with 5 g/ml
insulin; and on d 14 to 19 with primary medium. The cultures
were then washed with PBS, fixed in 10% neutral buffered
formalin, and stained with oil red O (Sigma-Aldrich).
MicroCT analysis
The left distal femur and the tibial-fibular junction (TFJ) were
scanned ex vivo with a Scanco VivaCT 40 (Scanco Medical,
Basserdorf, Switzerland) microCT (70 kVp, 85 A, isotropic
resolution of 10.5 m in all 3 spatial dimensions). The region
of interest (ROI) for the distal femur included the cancellous
bone compartment beginning 0.5 mm proximal to the
growth plate and extending proximally 1.5 mm. This ensures
exclusion of the primary spongiosa from the analyses. A
threshold was determined as 22% of the maximal gray scale to
distinguish mineralized from soft tissue. Trabecular bone
volume expressed as a percentage of total volume (BV/TV),
trabecular number (Tb.N; 1/mm), trabecular thickness
(Tb.Th; m), and trabecular pacing (Tb.Sp; m); connectiv-
ity density (Conn-dens; 1/mm3), structure model index (SMI;
ranges from 0 to 3 with 0plate-like and 3rod-like), degree
of anisotropy (DA) and mineral density (mg hydroxyapatite/
cm3) were evaluated using software provided by Scanco. The
ROI for the TFJ was defined as the region beginning 1.0 mm
proximal to the bifurcation of the TFJ and extending proxi-
mally 0.2 mm.
CXCL12/CXCR4 SIGNALING IN THE OSTEOBLAST
Downloaded from www.fasebj.org to IP 137.132.250.12. The FASEB Journal Vol.27, No.9 , pp:3505-3513, August, 2017
Bone histomorphometry
Mice were injected subcutaneously with calcein (10 mg/kg)
and demeclocycline (10 mg/kg) on d 7 and 2, respectively,
before euthanasia. The distal end of the right femur was
processed undecalcified for quantitative bone histomor-
phometry. Bones were fixed in 10% phosphate buffered
formalin for 24 h, dehydrated in increasing concentrations of
ethanol, and embedded undecalcified in modified methyl
methacrylate. Osteoblast surface (Ob.S) and osteoclast sur-
face (Oc.S) were measured in 4-m-thick sections stained
according to the Von Kossa technique with a tetrachrome
counterstain. Fluorochrome-based indices of bone formation,
including mineralizing surface/bone surface (MS/BS), min-
eral apposition rate (MAR), and surfaced-based bone forma-
tion rate/bone surface (BFR/BS) were measured in 8-m-
thick unstained sections using an image analysis system
(Osteometrics, Decatur, GA, USA). Bone histomorphometric
terminology is in accordance with recommendations by the
American Society for Bone and Mineral Research Histomor-
phometry Nomenclature Committee (21). The ROI was the
same as that used for bone microCT analysis.
Measurement of bone metabolic markers in serum
Blood was collected from the heart at the time of euthanasia,
and serum was separated and stored at 80C until pro-
cessed. Enzyme immunoassays were performed for measure-
ment of N-terminal propeptide of type 1 collagen (P1NP;
rat/mouse P1NP EIA, LOD7 ng/ml; Immunodiagnostic
Systems, London, UK) and pyridinoline (PYD; MicroVue
Serum PYD EIA kit, LOD4 nM; Quidel Corp., San Diego,
CA, USA), according to the manufacturers protocols. All
assays were performed in duplicate. Serum levels of total
calcium, phosphorus, magnesium, albumin, alkaline phos-
phatase, creatinine, and BUN were measured using an auto-
analyzer.
Chemotaxis assay/CXCL12-induced migration of
osteoblasts
To examine CXCL12-CXCR4 signaling and the chemotactic
behavior of osteoblasts in response to CXCL12, calvarial cells
were prepared from 1-wk-old mice expressing green fluores-
cent protein (GFP) under the regulatory control of the 2.3-kb
promoter fragment of type 1 collagen, a marker of osteo-
blast maturation. To generate these mice, we mated mice
expressing the tetracycline transactivator under control of the
2.3-kb promoter fragment of type 1 collagen (22), with mice
expressing histone-GFP under control of a tetracycline-re-
sponsive promoter element (tetO; Jackson Laboratory, Sacra-
mento, CA, USA). Calvarial bone samples were washed 3
times in a sterile PBS solution and examined for GFP cells
using a fluorescent microscope at 10 view. The calvariae
were then subjected to 4 sequential 15-min digestions at 37C
in an enzyme mixture containing 0.05% trypsin, 0.05 mM
EDTA, and 1.5 U/ml collagenase P in a PBS solution at 1
ml/calvaria. The cell fraction from the first digestion was
discarded, and the calvaria were washed with PBS prior to
beginning the second digestion. Cell fractions following
subsequent digestions were collected into a 45-ml tube and
neutralized with an equal volume of cold PBS medium. Cells
were centrifuged at 300 g for 10 min at 4C, resuspended in
3 ml DMEM and 10% FCS, filtered through a 70-m Falcon
cell strainer (BD Biosciences, Bedford, MA, USA) to generate
a single-cell suspension, and counted for viability.
Migration in vitro was assessed using a transwell assay.
Calvarial cells prepared as described above were loaded
3507
into the upper chamber (2.5105 cells in 0.1 ml of RESULTS
medium) or into both the upper and lower chambers (to
test for nonspecific effects of cytokinesis) of a transwell Genotyping of the Col2.3 Cre-specific CXCR4fl/fl, and
culture plate (polyester membrane thickness of 10 m and littermate control mice is shown in Fig. 1A. Successful
a pore size of 8 m; Transwell Permeable Supports; CXCR4 ablation was judged by the presence of the
Corning Inc., Corning, NY, USA). The cells were incubated
KO-only allele, the Cre recombinase gene, and excision
at 37 for 4 h in 0.6 ml of medium containing MEM and
0.1% BSA. After 4 h, CXCL12 (R&D Systems) at a concen- product. We validated the Col2.3-controlled CXCR4
tration of 100 ng/ml was added to the bottom chamber of deletion in osteoblasts by immunohistochemistry using
the wells. The incubation was continued for another 4 h, a monoclonal antibody to CXCR4 in cortical and
and the GFP-labeled cells on the underneath side of the trabecular bone. CXCR4 expression was low or unde-
transmembrane were quantified in accordance to the man- tectable in osteoblasts of KO vs. littermate control mice
ufacturers protocols. (Fig. 1B). This was confirmed using quantitative RT-
PCR for CXCR4 in cortical bone extracts (data not
Measurement of mRNA expression in bone and in ex vivo shown). Interestingly, some but not all osteocytes con-
cell cultures tained CXCR4. This was true for both WT and KO
mice. The percentages of osteocytes in WT and KO
RNA was extracted from cell cultures on d 14 (n8/group) mice expressing CXCR4 were 52 3 and 13 3%,
using the RNeasy Mini Kit (Qiagen, Valencia, CA USA), respectively. Body weight was not affected by CXCR4
according to the manufacturers recommendations, and ablation. At 6 wk of age, KO male and female mice
reverse transcribed with reverse transcription reagents weighed 23.1 1 and 19.4 0.2 g, and WT male and
(Applied Biosystems, Foster City, CA, USA). Bones, without female were 24.1 0.7 and 19.7 0.4 g, respectively.
marrow, were frozen in liquid nitrogen and crushed, and To examine the chemotactic behavior of osteoblasts
RNA was extracted with the same reagent. Total RNA (2 in response to CXCL12, we used mice expressing GFP
g) was reverse transcribed in 100 l of reaction mixture
under the control of the Col2.3 promoter. Calvarial
containing 1 PCR buffer, 1 mM deoxynucleoside triphos-
phate (DNTP) mix, 5 M random primers, 7.5 mM MgCl2,
cells in transwell culture plates were treated with
0.4 U/l RNase inhibitor, and 2.5 U/l MultiScribe re- CXCL12 for 4 h, and GFP-labeled cells on the under-
verse transcriptase. The following reverse transcription neath surface of the membrane were quantified (Fig. 2).
protocol was used: 25C for 10 min, 48C for 40 min, 95C The presence of CXCL12 stimulated migration of os-
for 5 min. Inventoried TaqMan assays (Assay IDs in paren- teoblasts by 2-fold (migrated cells45.63.5 and
theses) for murine L19 (Mm02601633_g1), CXCL12
(Mm00445553_m1), CXCR4 (Mm01292123_m1), CXCR7
(Mm00432610_m1), runt-related transcription factor 2
(Runx2; Mm00501584_m1), OSX (Mm00504574_m1),
ALP (Mm00475834_m1), Col2.3 (Mm00801666_g1), OCN
(Mm03413826_mH), receptor activator of nuclear fac-
tor-B ligand (RANKL; Mm00441906_m1), osteoprote-
gerin (OPG; Mm01205928_m1), and macrophage colony-
stimulating factor (M-CSF; Mm00432686_m1) were purchased
from Applied Biosystems. Expression of these genes was
measured using real-time PCR in triplicate 20-l reaction
volumes containing 10 l TaqMan Universal PCR master mix
(Applied Biosystems), 8 l of water, 1 l of inventoried
primer/probe gene assay, as detailed above, and 1 l of cDNA
template. The reactions were performed in a 7300 RT-PCR
System (Applied Biosystems) under the following conditions:
95C for 10 min, 40 cycles of 95 for 15 s, and 60C for 1 min.
Analysis was carried out using the software supplied with the
system and the number of threshold cycles (Ct) required for
the FAM fluorophore intensities to exceed a threshold just
above background were calculated. Ct values for triplicate
reactions were averaged, and Ct was calculated by subtract-
ing the Ct value of L19 from the Ct value of the target gene.
Expression of the target genes were quantified by calculating
2Ct relative to L19 as the endogenous control.

Figure 2. CXCL12 controls osteoblast migration. Osteoblasts

Data analysis
All data are reported as means se and were analyzed using
Students t test or 2-way ANOVA. Significant intergroup
differences for serum biomarkers, cell culture variables, and
histomorphometric and architectural parameters were deter-
mined using the Tukey range test, and values of P 0.05 were
considered to be significantly different.
expressing GFP under control of 2.3- type 1 collagen pro-
moter were harvested from calvarial bones from 1-wk-old
mice and incubated with or without CXCL12. Chemotaxis was
evaluated by the number of cells that migrated across the
permeable membrane. Data are presented as mean se of 3
wells Two separate experiments using independent cell
sources were performed; fluorescent microscopy images are
representative; 63 view. **P 0.005.

3508 Vol. 27 September 2013 The FASEB Journal www.fasebj.org SHAHNAZARI ET AL.
Downloaded from www.fasebj.org to IP 137.132.250.12. The FASEB Journal Vol.27, No.9 , pp:3505-3513, August, 2017
21.62.7103/cm2, with and without CXCL12, respec-
tively).
CXCR4 gene ablation increased the number of CFU,
CFU-AP, and mineralizing nodules in BMSC cultures
(Fig. 3). No difference in the size of the colonies was
observed. Gene ablation decreased the number and
size of adipocyte colonies (Fig. 4). The nonadherent
population of BMSCs harvested from the long bone
diaphysis of KO mice formed more osteoclasts in cul-
ture (Fig. 5).
We used microCT analysis to examine the effects of
CXCR4 gene ablation on bone structure. Gender-
specific differences in bone volume and structure were
observed, necessitating independent analyses for each
sex (Fig. 6). Loss of CXCR4 signaling in the mature
osteoblasts was accompanied by reduced BV/TV in
both male (25%) and female (30%) mice (Fig. 6). This
was associated with decreased connectivity density, tra-
becular number and thickness, and increased trabecu-
lar spacing in both sexes (Fig. 6). Gene ablation also
increased the SMI and degree of anisotropy, but only in
female mice. Ob.S/bone surface (Ob.S/BS) was not
affected by gene deletion, while Oc.S/BS was increased
by 70% but only in male mice (Fig. 6). BFR/BS and
mineral apposition rate (data not shown) did not vary
significantly (Fig. 6). Despite the absence of any differ-
ences in Ob.S/BS or BFR/BS between control and KO
mice, serum P1NP, a global marker of bone-forming
activity, was lower (30%) in gene-ablated mice (Fig. 7).
Serum pyridinoline was elevated (20%) in KO mice
CXCL12/CXCR4 SIGNALING IN THE OSTEOBLAST
Downloaded from www.fasebj.org to IP 137.132.250.12. The FASEB Journal Vol.27, No.9 , pp:3505-3513, August, 2017
(Fig. 7). Serum levels of total calcium, phosphorus,
magnesium, albumin, alkaline phosphatase, creatinine,
and BUN were unaffected by CXCR4 gene deletion.
To assess the effects of CXCR4 gene deletion on
metabolic activity in BMSC cultures, we measured ex-
pression of key genes associated with bone formation
and resorption. CXCR4 gene expression was lower
(50%) and CXCL12 gene expression was higher (50%)
in cells from KO mice (Fig. 8). CXCR7 was unaffected.
Rankl and m-csf expression were unaffected by gene
deletion, but OPG expression was higher (Fig. 8).
Markers of osteoblast formation (Runx2, osterix) and
activity (Alp, col2.3, and osteocalcin) were all elevated
in cell culture.
DISCUSSION
Defining the regulatory mechanisms controlling the
osteoprogenitor and osteoclast precursor populations
and their maturation and translocation to functional
sites is essential for understanding the pathogenesis of
skeletal diseases. CXCL12/CXCR4 signaling has been
shown to play important roles in bone development
and hematopoiesis, and previous studies have shown
that early deletion of the CXCR4 gene in osteoblasto-
genesis disrupts osteoblast formation and bone devel-
opment. In this study, we set out to examine the effects
of CXCR4 gene deletion in the mature osteoblast.
Using immunohistochemistry, we first demonstrated
Figure 3. CFU (top panels), CFU-AP (middle
panels), and mineralized nodules (bottom pan-
els) from bone marrow stromal cell cultures in
6-wk-old WT and Col2.3-Cre-CXCR4fl/fl mice
(meansse; n20). **P 0.001 vs. WT.
3509

Figure 4. Adipocyte colonies from BMSC cultures in 6-wk-
old WT and Col2.3-Cre-CXCR4fl/fl mice (meansse; n8).
Images are representative of colonies in 6-well plates and
microscopic images at 20 view from d 19 of culture. *P
0.05 vs. WT.
successful deletion of CXCR4 in osteoblasts (Fig. 1).
CXCR4 mRNA expression is still detected in whole
bone and in BMSC cultures because we have selectively
deleted CXCR4 only in a subset of cells (osteoblasts) in
the cell populations. That osteocytes from both WT and
KO mice expressed CXCR4 is puzzling. Because we
have permanently deleted CXCR4 from osteoblasts and
osteocytes arise from further differentiation of osteo-
blasts, we did not expect to observe CXCR4 in osteo-
cytes. The reason for this is not evident but it suggests
that some degree of CXCR4 signaling may persist in
osteocytes of CXCR4 KO mice. The data are also
consistent with the notion that not all osteocytes go
through a stage where collagen is expressed.
Functional responsiveness of mature osteoblasts to
CXCL12 was demonstrated using a chemotaxis assay.
The data clearly show a chemoattractant effect of
CXCL12 on osteoblast migration. This suggests that
local gradients of CXCL12 may function to direct
osteoblast migration to sites of needed bone formation.
CXCL12/CXCR4 signaling may also function to induce
homing of circulating osteogenic cells to sites of frac-
ture healing (23).
Osteoblast-specific CXCR4 deletion had profound
3510 Vol. 27 September 2013 The FASEB Journal www.fasebj.org
Downloaded from www.fasebj.org to IP 137.132.250.12. The FASEB Journal Vol.27, No.9 , pp:3505-3513, August, 2017
effects on the MSC pool and allocation to the osteo-
blastic and adipocytic cell lineages. It appears that
CXCL12/CXCR4 signaling in the mature osteoblast
can feedback to regulate the MSC pool size and alloca-
tion to either the osteoblast or adipocyte lineages. This
provides a unique signaling pathway, whereby mature
functioning osteoblasts can regulate the availability of
precursors to meet demands for bone formation. Prox-
imity of the niche to bone forming sites may facilitate
this signaling. The molecular signaling pathways that
feedback to regulate the progenitor pool size are not
evident.
CXCR4 is a Gi-protein-coupled receptor, and the
balance between Gs and Gi signaling in osteoblast
lineage cells expressing Col2.3 has been shown to have
profound effects on bone (22, 24, 25). Col2.3 directed
constitutive activation of the Gs-coupled PTH/PTH-
related protein receptor (increase in Gs:Gi) has been
shown to increase the alkaline phosphatase-positive
area in cultures of bone marrow cells (25). Our data are
consistent with these findings. Reduction of Gi signal-
ing through ablation of CXCR4 (increase in Gs:Gi) also
increases the alkaline phosphatase-positive area in cul-
tures of bone marrow cells (Fig. 3).
Although loss of CXCR4 signaling in the osteoblast
increases the osteoprogenitor number, it does not
affect Ob.S/BS. The Ob.S/BS is clearly not dictated by
the size of the progenitor pool. Other regulatory mech-
anisms must come into play to regulate recruitment
from the pool, osteoblast maturation, and migration to
the bone surface. The migratory responsiveness of the
osteoblasts to CXCL12 (Fig. 2) suggests that CXCR4
Figure 5. Osteoclasts formed in culture of nonadherent
marrow stromal cells from 6-wk-old WT and Col2.3-Cre-
CXCR4fl/fl mice (osteoclast size: small 310, medium 10
30, large 30 TRAP multinucleated cells). Representative
cell culture images for WT and KO animals (n8). *P
0.05 vs. WT.
SHAHNAZARI ET AL.
 feedback to regulate the osteoclast precursor pool size.
That the osteoclast size distribution (number of nuclei)
is unaffected suggests that osteoblast-specific CXCR4
signaling does not affect cell fusion and formation of
mature osteoclasts. These previously unrecognized sig-
naling pathways suggest that mature functioning osteo-
blasts may play an important role in maintaining ade-
quate precursor pools to meet the needs for changes in
both bone formation and resorption.
BV/TV was reduced in both male and female KO
mice. The decrease in the serum concentration of
P1NP and increase in the serum concentration of
pyridinoline suggests that this deficit reflects an imbal-
ance between formation and resorption (Fig. 7). De-
spite these relationships, BFR did not differ between
control and KO mice. This anomaly may reflect site-
related differences in bone-forming activity. Our mea-
surements of BFR/BS were taken in the cancellous
bone of the distal femur, whereas the serum concen-
tration of P1NP reflects global forming activity of the
entire skeleton. That bone formation in the distal
femoral metaphysis was not affected by CXCR4 ablation
suggests that the deficit in BV/TV does not reflect a
deficiency in bone formation.
The increase in serum pyridinoline was associated
with an increase in Oc.S/BS but only in the male. These

Figure 6. Histomorphometric and architectural variables of

distal femoral metaphyseal trabecular bone in 6-wk-old WT
(CXCR4wt/wt, CXCR4fl/fl; n6), CreCXCR4wt/wt (n6),
and Col2.3-Cre-CXCR4fl/fl (KO; n10) mice. BV/TV and
structural variables are reported from microCT analyses.

P0.05, P 0.01 for male vs. female within same genotype.

*P 0.05, **P 0.01 for KO vs. WT or CreCXCR4wt/wt
within the same gender; 1-way ANOVA.

signaling may play a role in regulating migration to and

maintenance of the osteoblast population on the bone
surface. Expression data from osteoblasts in culture
(Fig. 8) suggest that CXCR4 signaling may also play a
role in osteoblast activity. Collectively, the data suggest
that CXCR4 signaling in the mature osteoblasts plays a Figure 7. Bone formation and resorption markers in serum in
multifunctional role in regulating bone formation and 6-wk-old WT and Col2.3-Cre-CXCR4fl/fl mice (meansse;
may contribute to maintaining niche function. n10). *P 0.05 vs. WT. P 0.01 for male vs. female within
Osteoblast-specific CXCR4 signaling appears to also same genotype.

CXCL12/CXCR4 SIGNALING IN THE OSTEOBLAST 3511

Downloaded from www.fasebj.org to IP 137.132.250.12. The FASEB Journal Vol.27, No.9 , pp:3505-3513, August, 2017

Figure 8. mRNA expression levels of cxcr4, cxcl12, cxcr7, rankl,
opg, mcsf, runx2, osterix, alp, col 2.3, and osteocalcin from d 14 of
osteoblast cell culture in 6-wk-old WT and Col2.3-Cre-
CXCR4fl/fl mice relative to expression level of L19 and
normalized to WT expression level (meansse; n12 for KO,
n14 for WT). *P 0.05.
data suggest that CXCR4 signaling in the mature osteo-
blast can regulate the osteoclast precursor pool size,
Oc.S/BS, and resorption. That Oc.S/BS in the male is
elevated suggests that the deficit in BV/TV in the male
may reflect an inappropriately high resorption rate. In
the female, however, CXCR4 gene deletion had no
effect on Oc.S/BS despite an increase in serum pyr-
idinoline. This discrepancy may again reflect site-re-
lated differences in bone resorption or be related to
differences in individual cellular activity. Previous stud-
ies have shown that resorption pit depth (single-cell
resorbing activity) in in vitro resorption pit assays can be
regulated by estrogen (26). These observations suggest
that, like the male, the deficit in BV/TV in the female
may reflect an inappropriately high resorption rate.
3512 Vol. 27 September 2013 The FASEB Journal www.fasebj.org
Downloaded from www.fasebj.org to IP 137.132.250.12. The FASEB Journal Vol.27, No.9 , pp:3505-3513, August, 2017
The values of the structural variables are consistent
with the differences in bone volume. The SMI is
increased, and the degree of anisotropy is decreased in
KO female mice, suggesting that trabecular structure
and bone distribution may be influenced by CXCR4
signaling, but the effects are small (10%). We exam-
ined the effects of gender on the response of CXCR4
ablation for all variables. Gender effects were noted for
bone volume, Ob.S/BS, and Oc.S/BS and the bone
structural variables but not for any of the other vari-
ables.
The mRNA expression data from bone and BMSC
cell cultures (Fig. 8) are consistent with successful
deletion of the CXCR4 gene. CXCL12 expression was
higher in KO mice suggesting that there may be
regulatory mechanisms, whereby insufficient basal
CXCR4 signaling drives increased CXCL12 expression.
This does not appear to involve CXCR7. That OPG
expression was increased in CXCR4 KO mice suggests
that CXCR4 signaling in the osteoblasts may not only
regulate the osteoclast precursor pool size but may also
function to regulate osteoclast recruitment (10, 27).
Osteoblast recruitment and activity also appear to be
regulated by CXCR4 (Fig. 8).
Collectively, our data suggest that CXCR4 signaling
in the mature osteoblast can regulate the osteoprogeni-
tor and osteoclast precursor populations, recruitment
to the osteoblast, and osteoclast lineages, as well as their
maturation and translocation to functional sites on the
bone surface. The proposed proximity of the MSC/
HSC niche to sites of osteoblast activity and bone
formation suggests that CXCR4 signaling in osteoblasts
may regulate niche structure and function.
The CXCR4wt/f mice were kindly provided by Dr. Yong-Rui
Zou (Department of Microbiology, Columbia University Col-
lege of Physicians and Surgeons, New York, NY, USA). The
authors are grateful to Alicia Leeper and Alyssa Williams
(University of Florida, Gainesville, FL, USA) for technical
assistance with bone histomorphometry. Funds for these
studies were provided by the U.S. Department of Veterans
Affairs Merit Review program and the Northern California
Institute for Research and Education.
REFERENCES
1. Nagasawa, T., Omatsu, Y., and Sugiyama, T. Control of hema-
topoietic stem cells by the bone marrow stromal niche: the role
of reticular cells. Trends Immunol. 32, 315320
2. Hoggatt, J., and Pelus, L. M. Mobilization of hematopoietic stem
cells from the bone marrow niche to the blood compartment.
Stem Cell Res. Ther. 2, 13
3. Jung, Y., Wang, J., Schneider, A., Sun, Y. X., Koh-Paige, A. J.,
Osman, N. I., McCauley, L. K., and Taichman, R. S. (2006)
Regulation of SDF-1 (CXCL12) production by osteoblasts; a
possible mechanism for stem cell homing. Bone 38, 497508
4. Sugiyama, T., Kohara, H., Noda, M., and Nagasawa, T. (2006)
Maintenance of the hematopoietic stem cell pool by CXCL12-
CXCR4 chemokine signaling in bone marrow stromal cell
niches. Immunity 25, 977988
5. Christopher, M. J., Liu, F., Hilton, M. J., Long, F., and Link,
D. C. (2009) Suppression of CXCL12 production by bone
marrow osteoblasts is a common and critical pathway for cyto-
kine-induced mobilization. Blood 114, 13311339
SHAHNAZARI ET AL.
 6. Kortesidis, A., Zannettino, A., Isenmann, S., Shi, S., Lapidot, T.,
and Gronthos, S. (2005) Stromal-derived factor-1 promotes the
growth, survival, and development of human bone marrow
stromal stem cells. Blood 105, 37933801
7. Dar, A., Goichberg, P., Shinder, V., Kalinkovich, A., Kollet, O.,
Netzer, N., Margalit, R., Zsak, M., Nagler, A., Hardan, I.,
Resnick, I., Rot, A., and Lapidot, T. (2005) Chemokine receptor
CXCR4-dependent internalization and resecretion of functional
chemokine SDF-1 by bone marrow endothelial and stromal
cells. Nat. Immunol. 6, 1038 1046
8. Nagasawa, T., Hirota, S., Tachibana, K., Takakura, N., Ni-
shikawa, S., Kitamura, Y., Yoshida, N., Kikutani, H., and
Kishimoto, T. (1996) Defects of B-cell lymphopoiesis and bone-
marrow myelopoiesis in mice lacking the CXC chemokine
PBSF/SDF-1. Nature 382, 635638
9. Zhu, W., Boachie-Adjei, O., Rawlins, B. A., Frenkel, B., Boskey,
A. L., Ivashkiv, L. B., and Blobel, C. P. (2007) A novel regulatory
role for stromal-derived factor-1 signaling in bone morphogenic
protein-2 osteogenic differentiation of mesenchymal C2C12
cells. J. Biol. Chem. 282, 18676 18685
10. Wright, L. M., Maloney, W., Yu, X., Kindle, L., Collin-Osdoby, P.,
and Osdoby, P. (2005) Stromal cell-derived factor-1 binding to
its chemokine receptor CXCR4 on precursor cells promotes the
chemotactic recruitment, development and survival of human
osteoclasts. Bone 36, 840 853
11. De Klerck, B., Geboes, L., Hatse, S., Kelchtermans, H., Meyvis,
Y., Vermeire, K., Bridger, G., Billiau, A., Schols, D., and Matthys,
P. (2005) Pro-inflammatory properties of stromal cell-derived
factor-1 (CXCL12) in collagen-induced arthritis. Arthritis Res.
Ther. 7, R1208 R1220
12. Nie, Y., Waite, J., Brewer, F., Sunshine, M. J., Littman, D. R., and
Zou, Y. R. (2004) The role of CXCR4 in maintaining peripheral
B cell compartments and humoral immunity. J. Exp. Med. 200,
11451156
13. Lu, M., Grove, E. A., and Miller, R. J. (2002) Abnormal
development of the hippocampal dentate gyrus in mice lacking
the CXCR4 chemokine receptor. Proc. Natl. Acad. Sci. U. S. A. 99,
7090 7095
14. Tachibana, K., Hirota, S., Iizasa, H., Yoshida, H., Kawabata, K.,
Kataoka, Y., Kitamura, Y., Matsushima, K., Yoshida, N., Ni-
shikawa, S., Kishimoto, T., and Nagasawa, T. (1998) The chemo-
kine receptor CXCR4 is essential for vascularization of the
gastrointestinal tract. Nature 393, 591594
15. Zhu, W., Liang, G., Huang, Z., Doty, S. B., and Boskey, A. L.
(2011) Conditional inactivation of the CXCR4 receptor in
osteoprecursors reduces postnatal bone formation due to im-
paired osteoblast development. J. Biol. Chem. 286, 26794 26805
16. Rodda, S. J., and McMahon, A. P. (2006) Distinct roles for
Hedgehog and canonical Wnt signaling in specification, differ-
entiation and maintenance of osteoblast progenitors. Develop-
ment 133, 32313244
CXCL12/CXCR4 SIGNALING IN THE OSTEOBLAST
Downloaded from www.fasebj.org to IP 137.132.250.12. The FASEB Journal Vol.27, No.9 , pp:3505-3513, August, 2017
17. Cao, J. J., Wronski, T. J., Iwaniec, U., Phleger, L., Kurimoto, P.,
Boudignon, B., and Halloran, B. P. (2005) Aging increases
stromal/osteoblastic cell-induced osteoclastogenesis and alters
the osteoclast precursor pool in the mouse. J. Bone Miner. Res.
20, 1659 1668
18. Huang, J. C., Sakata, T., Pfleger, L. L., Bencsik, M., Halloran,
B. P., Bikle, D. D., and Nissenson, R. A. (2004) PTH differen-
tially regulates expression of RANKL and OPG. J. Bone Miner.
Res. 19, 235244
19. Kostenuik, P. J., Halloran, B. P., Morey-Holton, E. R., and Bikle,
D. D. (1997) Skeletal unloading inhibits the in vitro prolifera-
tion and differentiation of rat osteoprogenitor cells. Am. J.
Physiol. Endocrinol. Metab. 273, E1133E1139
20. Cao, J., Venton, L., Sakata, T., and Halloran, B. P. (2003)
Expression of RANKL and OPG correlates with age-related bone
loss in male C57BL/6 mice. J. Bone Miner. Res. 18, 270 277
21. Parfitt, A. M. (1988) Bone histomorphometry: proposed system
for standardization of nomenclature, symbols, and units. Calcif.
Tissue Int. 42, 284 286
22. Peng, J., Bencsik, M., Louie, A., Lu, W., Millard, S., Nguyen, P.,
Burghardt, A., Majumdar, S., Wronski, T. J., Halloran, B.,
Conklin, B. R., and Nissenson, R. A. (2008) Conditional expres-
sion of a Gi-coupled receptor in osteoblasts results in trabecular
osteopenia. Endocrinology 149, 1329 1337
23. Shinohara, K., Greenfield, S., Pan, H., Vasanji, A., Kumagai, K.,
Midura, R. J., Kiedrowski, M., Penn, M. S., and Muschler, G. F.
(2011) Stromal cell-derived factor-1 and monocyte chemotactic
protein-3 improve recruitment of osteogenic cells into sites of
musculoskeletal repair. J. Orthop. Res. 29, 1064 1069
24. Hsiao, E. C., Boudignon, B. M., Chang, W. C., Bencsik, M.,
Peng, J., Nguyen, T. D., Manalac, C., Halloran, B. P., Conklin,
B. R., and Nissenson, R. A. (2008) Osteoblast expression of an
engineered Gs-coupled receptor dramatically increases bone
mass. Proc. Natl. Acad. Sci. U. S. A. 105, 1209 1214
25. Ono, N., Nakashima, K., Schipani, E., Hayata, T., Ezura, Y.,
Soma, K., Kronenberg, H. M., and Noda, M. (2012) Constitu-
tively active PTH/PTHrP receptor specifically expressed in
osteoblasts enhances bone formation induced by bone marrow
ablation. J. Cell. Physiol. 227, 408 415
26. Parikka, V., Lehenkari, P., Sassi, M. L., Halleen, J., Risteli, J.,
Harkonen, P., and Vaananen, H. K. (2001) Estrogen reduces
the depth of resorption pits by disturbing the organic bone
matrix degradation activity of mature osteoclasts. Endocrinology
142, 53715378
27. Diamond, P., Labrinidis, A., Martin, S. K., Farrugia, A. N.,
Gronthos, S., To, L. B., Fujii, N., OLoughlin, P. D., Evdokiou,
A., and Zannettino, A. C. (2009) Targeted disruption of the
CXCL12/CXCR4 axis inhibits osteolysis in a murine model of
myeloma-associated bone loss. J. Bone Miner. Res. 24, 1150 1161
Received for publication January 8, 2013.
Accepted for publication May 14, 2013.
3513
CXCL12/CXCR4 signaling in the osteoblast regulates the
mesenchymal stem cell and osteoclast lineage populations
Mohammad Shahnazari, Vivian Chu, Thomas J. Wronski, et al.

FASEB J 2013 27: 3505-3513 originally published online May 23, 2013
Access the most recent version at doi:10.1096/fj.12-225763

References This article cites 25 articles, 9 of which can be accessed free at:
http://www.fasebj.org/content/27/9/3505.full.html#ref-list-1

Subscriptions Information about subscribing to The FASEB Journal is online at

http://www.faseb.org/The-FASEB-Journal/Librarian-s-Resources.aspx

Permissions Submit copyright permission requests at:

http://www.fasebj.org/site/misc/copyright.xhtml

Email Alerts Receive free email alerts when new an article cites this article - sign up at
http://www.fasebj.org/cgi/alerts

FASEB

Downloaded from www.fasebj.org to IP 137.132.250.12. The FASEB Journal Vol.27, No.9 , pp:3505-3513, August, 2017