CXCR7 Silencing Attenuates Cell Adaptive Response to
Stromal Cell Derived Factor 1a after Hypoxia
Sufang Liu1,2, Xiaofeng Jia2,3,4, Changsheng Li4,5, Xuefei Han1, Wenhai Yan1, Ying Xing1*
1 Department of Physiology, School of Medicine, Zhengzhou University, Zhengzhou, Henan Province, China, 2 Department of Biomedical Engineering, Johns Hopkins
University School of Medicine, Baltimore, Maryland, United States of America, 3 Department of Physical Medicine and Rehabilitation, Johns Hopkins University School of
Medicine, Baltimore, Maryland, United States of America, 4 Department of Anesthesiology and Critical Care Medicine, Johns Hopkins University School of Medicine,
Baltimore, Maryland, United States of America, 5 Department of Anesthesiology, Henan Anti-cancer Hospital, Zhengzhou, Henan Province, China
Abstract
Previous studies have shown that chemotactic factor stromal-cell derived factor 1a (SDF1a) promotes cell recovery from
hypoxic injury via its main receptor C-X-C chemokine receptor type (CXCR) 4. However, the role of its new receptor CXCR7
on cell repair against hypoxia and cell response to SDF1a remains largely unknown. In this study, neurons induced from
hippocampal progenitor cells were pre-conditioned in hypoxia for 4h and subsequently monitored to investigate the
function of SDF1a on cell repair after hypoxia. Neurons were assessed for their cell morphology, actin filament
polymerization and migration capability. SDF1a protein levels increased significantly 1 h after hypoxia compared to control
(P,0.01), and it reached a peak at 24 h after hypoxia. Moreover, addition of SDF1a promoted neurite outgrowth and actin
filament polymerization both in normoxic and hypoxic cells compared to untreated cells. Cell migration showed a timedependent increase with SDF1a stimulation in both groups, and hypoxic cells illustrated a significant augment at 0.5 h, 1 h
and 12 h after SDF1a application compared to normoxic cells (P,0.01). CXCR7 expression also increased with time
dependence after hypoxia and demonstrated a two-fold upregulation compared to control at 24 h after hypoxia. With
CXCR7 silencing, axon elongation and actin filament polymerization induced by SDF1a were inhibited sharply both in
normoxic and hypoxic cells. CXCR7 silencing also leads to reduced hypoxic cell migration at 0.5 h, 1 h, 12 h, 24 h and 36 h
after SDF1a application (P,0.01), but it failed to reduce normoxic cell migration induced by SDF1a at 0.5 h, 1 h and 12 h
(P.0.05). 24 h SDF1a stimulation led to higher ERK1/2 phosphorylation compared to control, and ERK1/2 phosphorylation
increased more in hypoxic cells than that in normoxic cells. This study suggested that CXCR7 plays an important role on cell
repair processing induced by SDF1a, and CXCR7 silencing attenuates cell adaptive response to acute SDF1a stimulation
(#12 h) after hypoxia.
Citation: Liu S, Jia X, Li C, Han X, Yan W, et al. (2013) CXCR7 Silencing Attenuates Cell Adaptive Response to Stromal Cell Derived Factor 1a after Hypoxia. PLoS
ONE 8(1): e55290. doi:10.1371/journal.pone.0055290
Editor: Jo Anne Powell-Coffman, Iowa State University, United States of America
Received August 23, 2012; Accepted December 20, 2012; Published January 31, 2013
Copyright: ß 2013 Liu et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted
use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: The authors appreciate State Natural Science Funds Commission for their funding support. The funders had no role in study design, data collection and
analysis, decision to publish, or preparation of the manuscript.
Competing Interests: The authors have declared that no competing interests exist.
* E-mail: xingyingprofessor@gmail.com
CXCR7 (C-X-C chemokine receptor type 7), is also able to bind
SDF1a [10]. Many lines of evidence indicate that CXCR7 has an
important role in regulating cell signaling in culture and in vivo.
However, the function of CXCR7 in neurons toward SDF1a is
under intense debate. While some reports have suggested that
CXCR7 is necessary to regulate CXCR4 protein levels, thereby
adapting chemokine responsiveness in migrating cells [11], other
studies indicate that CXCR4 and CXCR7 have distinct roles and
signal transduction in regulating interneuron movement and
laminar positioning [12]. These results, along with similar
observations in other tissues, led to the notion that CXCR7
modulates the effects of SDF1a by regulating chemokine gradients
or counteracting with CXCR4 [13]. Nevertheless, analysis of
CXCR7 function is limited to studies that demonstrate its function
in normoxic culture conditions. The exact mechanisms involved in
cellular response to SDF1a after hypoxia still require further
investigation. Recently, CXCR7 has been shown to activate MAP
kinases in transfected cells [14]. Additionally, a study showed that
SDF1a mediates some signaling upon binding to CXCR7, leading
to an increase in phosphorylation of ERK1/2 [15]. This leads to
Introduction
Previous studies have shown that neurons are able to withstand
hypoxia by triggering endogenous survival and neuroprotective
pathways [1]. These processes are always related to the recruitment of modulators that trigger some events promoting repair
mechanisms [2,3]. Recent studies have illustrated that a chemokine
stromal cell derived factor 1a (SDF1a) is up-regulated in bone
marrow mesenchymal stromal cells and synovial fibroblast cells to
fulfill its role on cell protection against hypoxia [4,5]. It increases
neuron survival and protects them from hypoxia-induced cell
death [6,7]. Furthermore, neural chemokine SDF1a mediates
axon elongation and neurite growth [8]. SDF1a may serve as
a protective factor to promote cell repair after hypoxia, however,
the exact effect of SDF1a on cell repair after hypoxia is still
unclear.
Many studies have shown that SDF1a binds to C-X-C
chemokine receptor type 4 (CXCR4) to exert its role on neurite
outgrowth and cell migration [8,9]. This view was challenged with
the discovery that the orphan receptor RDC1, now designated as
PLOS ONE | www.plosone.org
1
January 2013 | Volume 8 | Issue 1 | e55290
Cell Adaptive Response to SDF1a
were quantitatively analyzed by microscope with Image-pro-Plus
5.0 software (Media Cybernetics, USA) [18].
a plausible hypothesis that ERK1/2 may be involved in the
cellular response to SDF1a-CXCR7 axis after hypoxic stimulation.
Here we have investigated the function of chemokine SDF1a on
hippocampal cell repair after hypoxia by determining cell
morphology, actin filament polymerization and migration capability. We found that CXCR7 silencing attenuates these processes
by decreasing cell responsiveness to acute SDF1a stimulation.
Moreover, ERK1/2 phosphorylation increased with SDF1a
stimulation after hypoxia, whereas ERK1/2 phosphorylation
decreased by pre-treatment with CXCR7 ShRNA. Thus, CXCR7
may trigger cell signaling in order to fulfill its role in modulating
cellular response to SDF1a.
Enzyme-Linked Immunosorbent Assay (ELISA)
The production of SDF1a in the supernatants of hippocampal
cells under normoxia or hypoxia was assessed by ELISA using
a commercially available ELISA kit (R&D Systems, USA)
according to the manufacturer’s recommendations. Supernatants
were obtained and measured at 0.5, 1, 12, 24, and 36 hours after
hypoxia.
Migration Assay
Millicell hanging transwell chambers (Millipore, MA) with an 8mm pore size polystyrene filter inserts were used according to the
manufacturer’s instructions as described [19,20]. Briefly, 16105
cells in serum-free DMEM/F12 medium were seeded into the
upper compartment of each chamber, and serum-free medium
with 50 ng/ml SDF1a was placed into the lower chamber. This
optimal concentration was determined by a migration assay with
normoxic cells stimulated by SDF1a at varying doses (1, 25, 50,
100 ng/ml) for 30 min (data not shown). Before migration testing,
cells were divided into four parts with different treatments followed
by SDF1a stimulation: (1) cells treated with hypoxia, (2) cells
treated with CXCR7 ShRNA, (3) cells treated with CXCR7
ShRNA followed by hypoxia, and (4) cells left untreated. Cells
seeded in the top chambers were incubated for 0.5, 1, 12, 24 and
36 hours with SDF1a stimulation, respectively. Then, nonmigrating cells were removed from the top chamber, and migrated
cells were fixed in methanol and stained with 2% toluidine. The
number of cells that had migrated to the underside of the insert
membranes was calculated by counting six random independent
fields.
Materials and Methods
Ethics Statement
Sprague Dawley (SD) rats were provided by the Experimental
Animal Center of Medical School at Zhengzhou University
(Zhengzhou, China). This study was carried out in strict
accordance with the Guidelines on the Care and Use of
Laboratory Animals issued by the Chinese Council on Animal
Research and the Guidelines of Animal Care. All procedures
involving animals were approved by the Institutional Animal Care
and Use Committees of Zhengzhou University. All efforts were
made to minimize animals’ suffering and to reduce numbers of
animals used.
Cell Harvesting and Culture
Primary hippocampal progenitor cells were obtained from
newborn SD rats as described previously [16]. Briefly, animals
were euthanized under anesthesia. Bilateral hippocampi were
carefully dissected on a cold stage and digested in 0.25% trypsin
(Sigma-Aldrich) at 37uC for 5 min. Then the cells were collected
and centrifuged at 1000 rpm for 5 min. The precipitant cells were
gathered, washed and inoculated into culture plates at a density of
16105 cells/cm2 in serum-free DMEM/F12 medium supplemented with 2% B27, 20 ng/ml bFGF, 20 ng/ml EGF, kept in an
incubator at 37uC in a 5% CO2 fully humidified atmosphere and
fed half-fresh medium every 3 days. After about 7 days, the
progenitor cells grew into oating neurospheres. These were
mechanically separated into single cells and then passaged every
7 days. On passage 3, fetal bovine serum was added to the
medium, and then the neurospheres adhered to the bottom of the
culture dish and differentiated into neurons or glia-like cells. To
further reduce growth of non-neuronal cells, cytosine arabinoside
(AraC, 1 mM, Sigma-Aldrich) was added to the cultures within
24 h of plating. The single cells were incubated for 7 days to get
the mature hippocampal neurons. On the seventh day of
culturing, the neurons underwent hypoxia with 3% O2, 5%
CO2 and 92% N2 for 4 h [17]. After the hypoxic period, the
neuronal cells were returned to the original normoxic culture
conditions at 37uC with 5% CO2 and 95% air to carry out
experiments at different culture times.
Western Blotting
Adhered hippocampal neurons were washed with ice-cold PBS
and scraped in RIPA lysis buffer including protease inhibitors.
Total cell lysates were resolved on SDS-PAGE gels and transferred
to a polyvinylidene difluoride membrane. SDF1a, CXCR7
(1:2000, Abcam), GAPDH, phospho-ERK and total-ERK antibodies (1:1000, Beyotime Institute of Biotechnology, China) were
used for the analyses. Six independent experiments were applied
for western blot. Density analysis of protein bands were calculated
by Image J 1.4 software, and each band was normalized with the
integrated density of the corresponding reference band [21].
Immunocytochemistry
Briefly, adhered hippocampal cells were fixed for 30 min with
4% paraformaldehyde. Nonspecific binding was blocked with 2%
bovine serum albumin (BSA) for 1 h, followed by incubation with
CXCR7 primary antibody overnight at 4uC (1:100, Abcam). Cells
were then incubated with fluorescently labeled rabbit secondary
antibody (Sigma, 1:200) for 1 h respectively. CXCR7 positive cells
of six fields of each independent experiment were counted
manually under a light microscope (Olympus, Japan). The number
of positive cells per field was averaged to yield a cell number with
standard deviation. Alexa Fluor 546-conjugated phalloidin was
used as a high-affinity F-actin probe (Invitrogen, USA).
Neuronal Cell Count and Neurite Outgrowth Analysis
Neuronal cells were counted in 24-well culture dishes (six
independent wells per time point) under a light microscope
(Olympus, Japan). The photographed images were analyzed in
a blinded fashion; the neuronal cells were identified by their
typical morphology and counted manually. The number of
neuronal cells per field was averaged to yield a cell number with
standard deviation. The longest axon length and soma perimeter
PLOS ONE | www.plosone.org
Lentiviral Transduction of Hippocampal Cells
Three different short hairpin RNA (ShRNA) sequences directed
against CXCR7 were purchased from Shanghai GenePharma
(China). Three ShRNA lentiviral expression vectors with green
fluorescent protein (GFP) were transfected into hippocampal cells
2
January 2013 | Volume 8 | Issue 1 | e55290
Cell Adaptive Response to SDF1a
targeting CXCR7 mRNA (GenBank NM_053352.1) sequence.
To make the constructs, a BbsI site was added to the 59 end of the
ShRNA sequence and a XohI site to the 39 end. Three ShRNA
sequences that target the rat CXCR7 sequence were designed as
follows: 59-GGAACTACTCGGACATCAACT-39 (sense) and 59AGTTGATGTCCGAGTAGTTCC-39
(antisense,
CXCR7ShRNA1);
59-GGGTGAATATCCAGGCCAAGA-39
(sense) and 59-TCTTGGCCTGGATATTCACCC-39 (antisense,
CXCR7 ShRNA 2); 59-GGTCAGTCTCGTGCAGCATAA-39
(sense) and 59-TTA TGCTGCACGAG ACTGACC-39 (antisense, CXCR7 ShRNA 3). The lentivirus-GFP that expressed
GFP only was used as a blank control. On the seventh day of
culturing, some of transfected cells were assessed via western blot
to detect expression of the CXCR7 gene, while others were
exposed to hypoxia as described above. After the hypoxic period,
the neuronal cells were returned to the original normoxic culture
to carry out the experiments for different culture periods.
Statistical Analysis
Statistical analysis was performed using a standard computerized statistical package (Statistics Program for the Social Sciences
version 16.0, Chicago IL). Parametric data are expressed as the
mean 6 standard deviation of each group. Analysis of variance
(ANOVA) was performed for parametric data with the use of least
significant difference (LSD) analysis used for multiple comparisons.
An alpha level ,0.05 was selected to consider the differences
significant.
Figure 1. Expression of SDF1a in hippocampal cells after
hypoxia. Adhered hippocampal cells were exposed to hypoxia (3%O2,
5% CO2 and 92% N2) for 4 h or left untreated on the seventh day of
culturing. Supernatants and cell lysates were obtained and measured at
0.5, 1, 12, 24, and 36 h after hypoxia for ELISA and western blot
respectively. Untreated cells were used as control. A, Concentrations of
supernatant SDF1a in the medium were determined by ELISA. *P,0.01,
vs control. B, Expression of SDF1a in the cells was measured by western
blot analysis. GAPDH was used as reference.
doi:10.1371/journal.pone.0055290.g001
Results
Expression of SDF1a in Cultured Hippocampal Cells after
Hypoxia
Expression of SDF1a in cultured hippocampal cells at 0.5, 1,
12, 24, and 36 h after hypoxia is shown in Figure 1A. Comparison
of the expression level of SDF1a at different culture stages
revealed that SDF1a secreted in the medium was increased to
618.65670.46 ng/L at 1 h after hypoxia compared to control
(513.946107.76 ng/L, P,0.01). It reached peak levels at 24 h
followed by a decrease at 36 h (P,0.01), which may be
contributed to neural cells binding and taking up secreted SDF1a
in the medium. However, analysis of protein expression in the cells
(Figure 1B), revealed an up-regulation of SDF1a at 12 h after
hypoxia, most likely due to synthesis of SDF1a in the cytoplasm.
Thus, hypoxic pre-conditioning leads to an increase of SDF1a
expression in both secreted and synthesized forms.
migratory response of both normoxic and hypoxic pre-conditioned
cells, but with differences in timing of the response. In the first
12 h of SDF1a treatment, cell migration was significantly higher
in the hypoxic pre-conditioned group (180.17612.40) versus
normoxia (155.3367.12, P,0.01). However, after 24 h exposure
of SDF1a, cell migration of normoxic cells accounted for
209.3368.55, greater than hypoxic cells (190.6765.57, P,0.01).
And 36 h SDF1a application showed a significant increase of
migrated cells in normoxic group (216.1768.98) than hypoxic
group (200.33614.3, P,0.01).Together with observations above,
these results strongly suggest that cells pre-conditioned in hypoxia
showed higher sensitivity to respond to acute SDF1a stimulation
(#12 h) compared with that of normoxic cells.
Effects of SDF1a on Cell Morphology, Actin Filament
Polymerization and Migration Capability after Hypoxia
Cells treated with hypoxia conditions displayed an overall
decrease in dendrite length and shorter branches compared with
the normoxia group (shown by arrows). However, application of
SDF1a for 24 or 36 hours almost completely repaired cell
morphologies including neurite outgrowth and neural network,
which were initially damaged in the early stages after hypoxia
(Figure 2A). In addition, 24 h SDF1a stimulation increased actin
filament polymerization in axons and dendrites both in normoxic
and hypoxic cells (Figure 2B), but not in soma (Detail data shown
in Table S1).
As shown in Figure 2C, SDF1a enhanced cell migration with
time dependence from 0.5 h to 36 h both in normoxic and
hypoxic cells. The number of migrated cells in the hypoxic group
accounted for 71.5066.60, showing a significant increase compared to normoxia group (56.566.95) with stimulation of SDF1a
for 0.5 h (P,0.01). Stimulation with SDF1a resulted in a robust
PLOS ONE | www.plosone.org
Expression of CXCR7 in Cultured Hippocampal Cells after
Hypoxia
The number of CXCR7 positive cells increased in a timedependent manner after hypoxia (Figure 3A) to 85.6764.46 at
24 h, which was two times greater than control (40.3367.76,
P,0.01). Measurement of western blot confirmed the above
mentioned results, showing a significant increase in CXCR7
expression 24 h after hypoxia, the CXCR7 expression was
doubled compared to control measurement (Figure 3B–C,
P,0.01). Conversely, a decline of CXCR7 expression 36 h after
hypoxia was observed relative to that of 24 h, which can probably
be attributed to combination with its ligand, SDF1a (Figure 3B–C,
P,0.01).
3
January 2013 | Volume 8 | Issue 1 | e55290
Cell Adaptive Response to SDF1a
Figure 2. Effects of SDF1a on cell morphology, actin filament polymerization and migration capability after hypoxia. Adhered
hippocampal cells were exposed to hypoxia or left untreated (normoxia) on the seventh day of culturing. Then SDF1a (50 ng/ml) was added to
medium for 0.5, 1, 12, 24, and 36 hours. A, Cell morphological changes were observed in normoxia and hypoxia with 24 or 36 h stimulation of SDF1a
or not. Arrow heads indicate the changes of morphology. Bar = 50 mm. B, Actin filament polymerization was measured by Alexa Fluor 546-conjugated
phalloidin under normoxia or hypoxia with SDF1a for 24 h or not. Distribution changes of actin filament polymerization are indicated by arrows.
Bar = 50 mm. C, Cell migration induced by SDF1a was determined by a transwell chamber assay. *P,0.01, vs normoxia.
doi:10.1371/journal.pone.0055290.g002
without SDF1a application. However, after 24 h SDF1a stimulation, actin polymerization was increased and redistributed in
neurites and sprouts in normoxic cells, but not in cells with
CXCR7 silencing (shown by arrows).
Unexpectedly, as shown in Figure 4F, there was not a significant
difference in the number of migrated cells between the untreated
group and CXCR7 ShRNA group with 0.5 h, 1 h, 12 h SDF1a
stimulation respectively (P.0.05). The number of migrated cells in
the untreated group was 56.5066.95 after 0.5 h application of
SDF1a. However, unlike groups with acute SDF1a exposure
(#12 h), migrated cells in CXCR7 ShRNA group (159.6767.34)
still illustrated a decline compared to untreated group
(209.3368.55) after 24 h SDF1a application.
Effects of CXCR7 ShRNA on Cell Morphology, Actin
Filament Polymerization and Migration Capability
Induced by SDF1a in Normoxic Cells
To study whether CXCR7 was involved in cell repair processes
induced by SDF1a, CXCR7 expression was firstly detected in
cultured cells treated with 24 h SDF1a application (Figure 4A),
and western blotting illustrated a strong increase of CXCR7
expression compared to untreated cells. Then three ShRNA were
used to knock down CXCR7 expression, each of which was
effective in inhibiting CXCR7 protein expression (Figure 4B).
Here, we used CXCR7 ShRNA1 as an effective inhibitor to do
following experiments. As shown in Figure 4C–D, the ratio of the
longest axon length to the soma perimeter after CXCR7 silencing
was 0.5760.02, showing almost half decline compared to control
(1.0660.02). Following 12 h and 24 h SDF1a stimulation, the
ratio in normoxic cells increased to 1.6760.03 and 1.9960.18,
showing significant increase than that of cells with CXCR7
silencing (0.5960.03 and 0.6560.09) respectively.
CXCR7 was then tested for its involvement in recombination of
cytoskeletons like actin filament polymerization. As shown in
Figure 4E, actin filament polymerization was mainly distributed in
the soma in both normoxic and CXCR7 ShRNA exposed cells
PLOS ONE | www.plosone.org
Effects CXCR7 ShRNA on Cell Morphology and Migration
Capability Induced by SDF1a in Hypoxic Cells
As shown in Figure 5A, cells with CXCR7 ShRNA exposure
followed by hypoxia showed less neurites and shorter axons
compared to those treated with hypoxia only, following 24 h
stimulation of SDF1a (shown by arrows). Then cell migration was
determined with 0.5 h, 1 h, 12 h, 24 h and 36 h SDF1a
stimulation as shown in figure 5 B. The number of migrated cells
induced by 0.5h SDF1a application was significantly greater in the
4
January 2013 | Volume 8 | Issue 1 | e55290
Cell Adaptive Response to SDF1a
group than that of CXCR7 ShRNA-hypoxia group with timedependent stimulation of SDF1a.
Expression of ERK1/2 Phosphorylation Induced by SDF1a
after Hypoxia with CXCR7 Silencing
As shown in Figure 6, ERK1/2 phosphorylation was detected
with different treatments. We observed that 24 h SDF1a
stimulation strongly caused ERK1/2 phosphorylation, which
was remarkably significant in cells pre-conditioned with hypoxia
compared to untreated group (Figure 6A–B, P,0.01). Then we
found cells transfected with CXCR7 ShRNA failed to increase
phosphorylated ERK1/2 expression with SDF1a stimulation for
24 h compared to untreated group (Figure 6C–D, P,0.01).
Though ERK1/2 phosphorylation induced by SDF1a increased
significantly in hypoxic cells, it was still inhibited following
treatment with CXCR7 ShRNA(Figure 6E–F, P,0.01). Altogether, our experiments demonstrated that ERK1/2 pathway was
activated by hypoxia and SDF1a/CXCR7 pathway.
Discussion
In this study, we have used hippocampal cells cultured in vitro
as a model system to investigate the function of atypical chemokine
receptor CXCR7 in cellular response to SDF1a after preconditioning in hypoxia. We found that CXCR7 is required in
hippocampal cells to regulate cell adaptive functions with
chemokine SDF1a stimulation, including cell morphology, actin
filament polymerization and migration capability. However, these
functions emerge as significant properties of hypoxic cells, because
the interference of CXCR7 altered cell migration ability in
hypoxic cells much more than in normoxic cells. These results
provide a clear demonstration that an atypical chemokine receptor
can modulate cell response to classical chemokine SDF1a after
hypoxia.
Our results showed that SDF1a was highly up-regulated in
cultured hippocampal cells after hypoxia, which is consistent with
previous studies findings that SDF1 is over-expressed in inflammatory processes in several brain diseases with some
neuroprotective function [4,5,20,22]. However, a study demonstrated that SDF1a expression in neural progenitor cells derived
from neonatal mouse brain showed no response to chronic
hypoxia [22,23]. Thus, SDF1a seems to be involved in cell
adaptive changes to acute hypoxia but not to chronic hypoxia
[24].
Our results have shown that 24 h application of SDF1a
repaired neurite outgrowth and neural network damaged by
hypoxia. We also found that SDF1a may change or recombine cell
cytoskeleton by increasing actin filament polymerization, thus
resulting in cell adaptive survival after hypoxia. This is consistent
with previous studies that SDF1a improves neurite outgrowth and
axon elongation in corticospinal tract axons and cultured
cerebellar granule neurons [25,26]. Drastic changes in neuronal
shape occur immediately after the exit of neuronal cells from
mitotic cycles during embryonic and postnatal development. Thus,
it is conceivable that SDF1a might be necessary for reparative
capabilities after hypoxia as a mitogen and potential target for
morphogenesis and neuroprotective strategies.
Previous studies have shown that SDF1a increases the motility
of upper rhombic lip cells and promotes the migration of granule
cell progenitors and interneurons [27,28]. However, whether
SDF1a could affect cell migration under hypoxic conditions still
needs further investigation. A recent study illustrated that SDF1a
could direct the migration of neural progenitor cells under hypoxic
circumstances [29], and guide migration after focal cerebral
Figure 3. Expression of CXCR7 in hippocampal cells after
hypoxia. Adhered hippocampal cells were exposed to hypoxia or
normoxia (control), and then CXCR7 expression was determined by
immunofluorescence and western blot at 0.5, 1, 12, 24, and 36 h after
hypoxia. Six independent experiments were applied for both cell count
and western blot. A, Hippocampal cells were digested from the culture
dish and measured for CXCR7 expression by immunofluorescence.
*P,0.01, vs control. B–C, Cell lysates were obtained to perform western
blot assay for CXCR7 expression. Data showed strong up-regulation of
CXCR7 after hypoxia. *P,0.01, vs control.
doi:10.1371/journal.pone.0055290.g003
group that was exposed to hypoxia (71.5066.60) compared to the
group in which hypoxia was preconditioned by CXCR7 ShRNA
(39.50610.21, P,0.01). The difference still was apparent in two
groups at 36 h after SDF1a application (200.33614.39 and
85.83612.01, respectively, P,0.01). In addition, number of cell
migration showed a stronger upward tendency in the hypoxic
PLOS ONE | www.plosone.org
5
January 2013 | Volume 8 | Issue 1 | e55290
Cell Adaptive Response to SDF1a
Figure 4. Effects of CXCR7 ShRNA on cell morphology, actin filament polymerization and migration capability induced by SDF1a in
normoxic cells. A, CXCR7 expression was increased with 24 h stimulation of SDF1a. Adhered hippocampal cells were cultured in normoxia for seven
days. Then cells were treated with SDF1a (50 ng/ml) for 24 h or left untreated (control). Cell lysates were obtained to detect CXCR7 expression by
western blot. B, CXCR7 protein expression was determined after CXCR7 silencing. Three ShRNA (CXCR7 ShRNA1, CXCR7 ShRNA2, CXCR7 ShRNA3)
lentiviral expression vectors with green fluorescent protein (GFP) were transfected into hippocampal cells targeting CXCR7 mRNA on the 7th day of
culture. Three days later, cell lysates of transfected and untreated cells (control) were obtained to determine CXCR7 protein expression by western
blot. The lentivirus-GFP that expressed GFP only was used as a blank control (data was not shown). Here, we used CXCR7 ShRNA1 as an effective
inhibitor to do subsequent experiments. C, D, Cell morphology was observed and the ratio of the longest axon length to the soma perimeter was
calculated by image analysis system. Cells were cultured under normoxia with or without CXCR7 ShRNA transfection followed by SDF1a (50 ng/ml)
for 24 or 36 hours, or left untreated. Neuronal cells were counted under a light microscope. Six predetermined areas in each independent well (From
a total of six wells) at each time point were selected and photographed. Then the longest axon length and the soma perimeter were counted and
calculated manually. Bar = 10 mm. *P,0.01, vs untreated. E, Actin filament polymerization was observed by staining with phalloidin and imaging with
confocal microscopy. Neural cells were transfected with CXCR7 ShRNA with following SDF1a (50 ng/ml) application for 24 h or not. Arrows indicate
the distribution changes of actin filament polymerization. Bar = 10 mm. F, Number of migrated cells was counted by transwell chamber analysis. Some
of cells were transfected with CXCR7 ShRNA, and some were left untreated. Then cells were placed in the upper chamber and SDF1a (50 ng/ml) was
added to the lower chamber for 0.5, 1, 12, 24, and 36 hours respectively for migration analysis. *P,0.01, vs untreated.
doi:10.1371/journal.pone.0055290.g004
ischemia [30]. Our observations suggested that SDF1a enhanced
cell migration especially when cells were treated with hypoxia.
Though our observations provided evidence that cells have a high
sensitivity to respond to SDF1a at early stages after hypoxia, we
PLOS ONE | www.plosone.org
still wonder if it is hypoxia or SDF1a that mainly increased cell
migration.
It has been shown that CXCR7 is expressed in most neurons of
human brain cultures [31]. In the embryonic rat cortex numerous
cells also express CXCR7, which is found to be very abundant in
6
January 2013 | Volume 8 | Issue 1 | e55290
Cell Adaptive Response to SDF1a
Figure 5. Effects of CXCR7 ShRNA on cell morphology and migration capability induced by SDF1a in hypoxic cells. A, Cell morphology
was observed after hypoxia with pre-treatment of CXCR7 ShRNA or not. Some of cells were transfected with CXCR7 ShRNA, or left untreated. Then all
of cells were treated with hypoxia followed by stimulation of SDF1a (50 ng/ml) for 24 h. Neuronal cell morphology was observed and photographed
under a fluorescence microscope. Arrows indicate the changes of dendrite length and neurite outgrowth formation. Bar = 30 mm. B, Number of
migrated cells was counted by transwell chamber analysis. Cells with or without CXCR7 silencing were treated with hypoxia, and then were placed in
the upper chamber with SDF1a (50 ng/ml) in the lower chamber for 0.5, 1, 12, 24, and 36 hours respectively. *P,0.01, vs hypoxia.
doi:10.1371/journal.pone.0055290.g005
neurons forming the cortical plate during initial stages of
corticogenesis [11,12]. In our work we found CXCR7 was
expressed in hippocampal cells and was increased significantly
with application of SDF1a for 24 hours. Indeed, CXCR7 is evenly
distributed in DRG neurons both at early and at late stages of
cultivation [8]. Moreover, previous studies have suggested that
hypoxic preconditioning increases CXCR7 expression in mesenchymal stromal cells [32], and only CXCR7 activity was
functionally linked to survival signaling in mouse neural progenitor
cell cultures during hypoxia exposure [33]. In our experiments,
both CXCR7 positive cells and CXCR7 protein expression
increased at 12 h after hypoxia. CXCR7 expression in the brain
has been shown to increase after stroke [34], and seems to exert
a prominent role in the immediate regulation of cell transduction
pathways [35]. Thus it is possible that CXCR7 might be necessary
for cell adaptive functions after hypoxia.
Our work showed no significant difference between untreated
cells and CXCR7 ShRNA transfected cells with acute stimulation
of SDF1a, but a remarkable difference was shown after 24 h
application of SDF1a. This result was partly consistent with a study
in which CXCR7 is involved in neuronal cell migration of rat
forebrain, independent of the main receptor CXCR4 [36], and
CXCR7 mutants also show a reduced rate of interneuron
PLOS ONE | www.plosone.org
migration and laminar positioning defects [12]. This suggests that
CXCR7 may be necessary, but not sufficient for cell migration, as
pre-treatment of CXCR7 ShRNA could not block the timedependent promotion of SDF1a until 24 hours later.
Indeed, CXCR7 was necessary for survival signaling in mouse
neural progenitor cells during hypoxic exposure. A recent study
suggests that down-regulation of CXCR7 results in reduced
interneuron responsiveness to SDF1a [12]. Our observations
showed that CXCR7 silencing caused neurite outgrowth defects
and migration capacity decline significantly. Taken together with
previous data, we found that hypoxia increased cell responsiveness
to acute SDF1a stimulation, but this responsiveness was abrogated
by pretreatment with CXCR7 ShRNA. These results mentioned
above strongly suggested that CXCR7 was critical and required in
cell response to available SDF1a, thereby modulating neurite
outgrowth and cell migration.
The ERK1/2 signaling pathway is critical for cell survival and
migration. Previous studies have shown that Raf/ERK upregulation enhances the migration of cultured cortical and
cerebellar granule cells [33], and that the ERK1/2 pathway is
involved in neuronal differentiation, synaptic plasticity and cell
migration [37,38]. As a mitogen and chemokine, SDF1a induces
ERK1/2 activation in activated T cells [39]. Our results,
7
January 2013 | Volume 8 | Issue 1 | e55290
Cell Adaptive Response to SDF1a
Figure 6. Expression of ERK1/2 phosphorylation induced by SDF1a after hypoxia with CXCR7 silence. Cell lysates were obtained with
different treatment to detect phosphorylated ERK1/2 (P-ERK) and total ERK1/2 (T-ERK) by western blot. Total ERK was used as a control. A–B,
Detection of ERK phosphorylation induced by hypoxia and/or SDF1a was performed. Cells were treated with hypoxia and/or SDF1a (50 ng/ml) for
24 h, or left untreated. *P,0.01 vs untreated group. C–D, Detection of ERK1/2 phosphorylation induced by SDF1a after CXCR7 silence was performed.
Cells were transfected with CXCR7 ShRNA and/or with application of SDF1a (50 ng/ml) for 24 h, or left untreated. *P,0.01 vs untreated group. E–F,
Detection of ERK1/2 phosphorylation induced by SDF1a after CXCR7 silencing and hypoxia was performed. Cells were transfected with CXCR7 ShRNA
or not, then all of cells were treated with hypoxia with presence or absence of SDF1a (50 ng/ml) for 24 h. *P,0.01 vs group with hypoxia exposure.
doi:10.1371/journal.pone.0055290.g006
consistent with these studies, also showed a significant increase of
ERK1/2 phosphorylation induced by 24 h SDF1a stimulation.
A recent study showed that hypoxia induced ERK1/2
phosphorylation [40,41], which is involved in mitochondrial
biogenesis in hippocampal cells during hypoxia exposure
[42,43]. In our work, we also found phosphorylated ERK1/2
increased significantly after hypoxia. Stimulation with SDF1a
strongly promotes the phosphorylation of ERK1/2 in migrating
interneurons, but fails to elicit phosphorylation of ERK1/2 in cells
obtained from CXCR7 mutants [11]. Thus we wondered whether
CXCR7 was involved in ERK1/2 phosphorylation activated by
SDF1a after hypoxia. Our data using cultures of hippocampal
cells supported this hypothesis. We found that ERK1/2 phosphorylation was inhibited after CXCR7 silencing, either with
treatment of hypoxia or SDF1a stimulation. Thus, it is conceivable
that SDF1a, acting through its receptor CXCR7, increased cell
adaptive function via activation of ERK1/2.
We believe that our findings may have important implications
to repair cell function after hypoxia. CXCR7 has been involved in
multiple steps induced by SDF1a stimulation, including neurite
outgrowth, actin filament polymerization and cell migration. If
one important function of CXCR7 is to alter cell response toward
SDF1a, as reported here, then modulating this receptor may
contribute to repair activities from hypoxia. While we show effects
between control cultures in approximately 20% oxygen and the
hypoxia treated cells, however, caution must be exercised in
extending the observations to in vivo responses to low oxygen
levels (e.g. 3% Oxygen), which is close to physiological levels
within tissues of an intact organism.
Supporting Information
Table S1 Fluorescence mean densities of actin filament
polymerization. OD: optical density. *P,0.05, vs normoxia;
#P,0.05, vs hypoxia.
(PDF)
Acknowledgments
The authors appreciate Daphne Hutton and Elliott Greenwald for their
editorial support.
Author Contributions
Conceived and designed the experiments: SL XJ YX. Performed the
experiments: SL CL. Analyzed the data: CL XH WY. Contributed
reagents/materials/analysis tools: YX. Wrote the paper: SL XJ.
References
5. Hitchon C, Wong K, Ma G, Reed J, Lyttle D, et al. (2002) Hypoxia-induced
production of stromal cell-derived factor 1 (CXCL12) and vascular endothelial
growth factor by synovial fibroblasts. Arthritis Rheum 46: 2587–2597.
6. Gillard SE, Lu M, Mastracci RM, Miller RJ (2002) Expression of functional
chemokine receptors by rat cerebellar neurons. J Neuroimmunol 124: 16–28.
7. Meucci O, Fatatis A, Simen AA, Bushell TJ, Gray PW, et al. (1998) Chemokines
regulate hippocampal neuronal signaling and gp120 neurotoxicity. Proc Natl
Acad Sci U S A 95: 14500–14505.
8. Opatz J, Kury P, Schiwy N, Jarve A, Estrada V, et al. (2009) SDF-1 stimulates
neurite growth on inhibitory CNS myelin. Mol Cell Neurosci 40: 293–300.
1. Feast A, Martinian L, Liu J, Catarino CB, Thom M, et al. (2012) Investigation of
hypoxia-inducible factor-1alpha in hippocampal sclerosis: A postmortem study.
Epilepsia 53: 1349–1359.
2. Wang Y, Huang J, Li Y, Yang GY (2012) Roles of chemokine CXCL12 and its
receptors in ischemic stroke. Curr Drug Targets 13: 166–172.
3. Timotijevic G, Mostarica Stojkovic M, Miljkovic D (2012) CXCL12: role in
neuroinflammation. Int J Biochem Cell Biol 44: 838–841.
4. Liu H, Liu S, Li Y, Wang X, Xue W, et al. (2012) The role of SDF-1-CXCR4/
CXCR7 axis in the therapeutic effects of hypoxia-preconditioned mesenchymal
stem cells for renal ischemia/reperfusion injury. PLoS One 7: e34608.
PLOS ONE | www.plosone.org
8
January 2013 | Volume 8 | Issue 1 | e55290
Cell Adaptive Response to SDF1a
9. Ma Q, Jones D, Borghesani PR, Segal RA, Nagasawa T, et al. (1998) Impaired
B-lymphopoiesis, myelopoiesis, and derailed cerebellar neuron migration in
CXCR4- and SDF-1-deficient mice. Proc Natl Acad Sci U S A 95: 9448–9453.
10. Balabanian K, Lagane B, Infantino S, Chow KY, Harriague J, et al. (2005) The
chemokine SDF-1/CXCL12 binds to and signals through the orphan receptor
RDC1 in T lymphocytes. J Biol Chem 280: 35760–35766.
11. Sanchez-Alcaniz JA, Haege S, Mueller W, Pla R, Mackay F, et al. (2011) Cxcr7
controls neuronal migration by regulating chemokine responsiveness. Neuron
69: 77–90.
12. Wang Y, Li G, Stanco A, Long JE, Crawford D, et al. (2011) CXCR4 and
CXCR7 have distinct functions in regulating interneuron migration. Neuron 69:
61–76.
13. Levoye A, Balabanian K, Baleux F, Bachelerie F, Lagane B (2009) CXCR7
heterodimerizes with CXCR4 and regulates CXCL12-mediated G protein
signaling. Blood 113: 6085–6093.
14. Rajagopal S, Kim J, Ahn S, Craig S, Lam CM, et al. (2010) Beta-arrestin- but
not G protein-mediated signaling by the ‘‘decoy’’ receptor CXCR7. Proc Natl
Acad Sci U S A 107: 628–632.
15. Kumar R, Tripathi V, Ahmad M, Nath N, Mir RA, et al. (2012) CXCR7
mediated Gialpha independent activation of ERK and Akt promotes cell
survival and chemotaxis in T cells. Cell Immunol 272: 230–241.
16. Brewer GJ, Torricelli JR, Evege EK, Price PJ (1993) Optimized survival of
hippocampal neurons in B27-supplemented Neurobasal, a new serum-free
medium combination. J Neurosci Res 35: 567–576.
17. Wang FZ, Ding AS, Liu ZW (1995) Effect of ginsenosides against anoxic damage
of hippocampal neurons in culture. Zhongguo Yao Li Xue Bao 16: 419–422.
18. Cui YF, Xu JC, Hargus G, Jakovcevski I, Schachner M, et al. (2011) Embryonic
stem cell-derived L1 overexpressing neural aggregates enhance recovery after
spinal cord injury in mice. PLoS One 6: e17126.
19. Zagzag D, Lukyanov Y, Lan L, Ali MA, Esencay M, et al. (2006) Hypoxiainducible factor 1 and VEGF upregulate CXCR4 in glioblastoma: implications
for angiogenesis and glioma cell invasion. Lab Invest 86: 1221–1232.
20. Rempel SA, Dudas S, Ge S, Gutierrez JA (2000) Identification and localization
of the cytokine SDF1 and its receptor, CXC chemokine receptor 4, to regions of
necrosis and angiogenesis in human glioblastoma. Clin Cancer Res 6: 102–111.
21. Carretero-Ortega J, Walsh CT, Hernandez-Garcia R, Reyes-Cruz G, Brown
JH, et al. (2010) Phosphatidylinositol 3,4,5-triphosphate-dependent Rac
exchanger 1 (P-Rex-1), a guanine nucleotide exchange factor for Rac, mediates
angiogenic responses to stromal cell-derived factor-1/chemokine stromal cell
derived factor-1 (SDF-1/CXCL-12) linked to Rac activation, endothelial cell
migration, and in vitro angiogenesis. Mol Pharmacol 77: 435–442.
22. Lazarini F, Tham TN, Casanova P, Arenzana-Seisdedos F, Dubois-Dalcq M
(2003) Role of the alpha-chemokine stromal cell-derived factor (SDF-1) in the
developing and mature central nervous system. Glia 42: 139–148.
23. Li Q, Michaud M, Stewart W, Schwartz M, Madri JA (2008) Modeling the
neurovascular niche: murine strain differences mimic the range of responses to
chronic hypoxia in the premature newborn. J Neurosci Res 86: 1227–1242.
24. Schneider C, Krischke G, Rascher W, Gassmann M, Trollmann R (2012)
Systemic hypoxia differentially affects neurogenesis during early mouse brain
maturation. Brain Dev 34: 261–273.
25. Arakawa Y, Bito H, Furuyashiki T, Tsuji T, Takemoto-Kimura S, et al. (2003)
Control of axon elongation via an SDF-1alpha/Rho/mDia pathway in cultured
cerebellar granule neurons. J Cell Biol 161: 381–391.
26. Tanaka E, Sabry J (1995) Making the connection: cytoskeletal rearrangements
during growth cone guidance. Cell 83: 171–176.
PLOS ONE | www.plosone.org
27. Yu T, Huang H, Li HF (2010) Stromal cell-derived factor-1 promotes migration
of cells from the upper rhombic lip in cerebellar development. J Neurosci Res
88: 2775–2786.
28. Lysko DE, Putt M, Golden JA (2011) SDF1 regulates leading process branching
and speed of migrating interneurons. J Neurosci 31: 1739–1745.
29. Xu Q, Wang S, Jiang X, Zhao Y, Gao M, et al. (2007) Hypoxia-induced
astrocytes promote the migration of neural progenitor cells via vascular
endothelial factor, stem cell factor, stromal-derived factor-1alpha and monocyte
chemoattractant protein-1 upregulation in vitro. Clin Exp Pharmacol Physiol
34: 624–631.
30. Robin AM, Zhang ZG, Wang L, Zhang RL, Katakowski M, et al. (2006)
Stromal cell-derived factor 1alpha mediates neural progenitor cell motility after
focal cerebral ischemia. J Cereb Blood Flow Metab 26: 125–134.
31. Shimizu S, Brown M, Sengupta R, Penfold ME, Meucci O (2011) CXCR7
protein expression in human adult brain and differentiated neurons. PLoS One
6: e20680.
32. Liu H, Xue W, Ge G, Luo X, Li Y, et al. (2010) Hypoxic preconditioning
advances CXCR4 and CXCR7 expression by activating HIF-1alpha in MSCs.
Biochem Biophys Res Commun 401: 509–515.
33. Bakondi B, Shimada IS, Peterson BM, Spees JL (2011) SDF-1alpha secreted by
human CD133-derived multipotent stromal cells promotes neural progenitor cell
survival through CXCR7. Stem Cells Dev 20: 1021–1029.
34. Schonemeier B, Kolodziej A, Schulz S, Jacobs S, Hoellt V, et al. (2008) Regional
and cellular localization of the CXCl12/SDF-1 chemokine receptor CXCR7 in
the developing and adult rat brain. J Comp Neurol 510: 207–220.
35. Sengupta R, Burbassi S, Shimizu S, Cappello S, Vallee RB, et al. (2009)
Morphine increases brain levels of ferritin heavy chain leading to inhibition of
CXCR4-mediated survival signaling in neurons. J Neurosci 29: 2534–2544.
36. Tiveron MC, Boutin C, Daou P, Moepps B, Cremer H (2010) Expression and
function of CXCR7 in the mouse forebrain. J Neuroimmunol.
37. Yang K, Cao F, Sheikh AM, Malik M, Wen G, et al. (2012) Up-regulation of
Ras/Raf/ERK1/2 signaling impairs cultured neuronal cell migration, neurogenesis, synapse formation, and dendritic spine development. Brain Struct
Funct.
38. Ortega JA, Alcantara S (2010) BDNF/MAPK/ERK-induced BMP7 expression
in the developing cerebral cortex induces premature radial glia differentiation
and impairs neuronal migration. Cereb Cortex 20: 2132–2144.
39. Reichardt LF (2006) Neurotrophin-regulated signalling pathways. Philos
Trans R Soc Lond B Biol Sci 361: 1545–1564.
40. Kremer KN, Clift IC, Miamen AG, Bamidele AO, Qian NX, et al. (2011)
Stromal cell-derived factor-1 signaling via the CXCR4-TCR heterodimer
requires phospholipase C-beta3 and phospholipase C-gamma1 for distinct
cellular responses. J Immunol 187: 1440–1447.
41. Lin SC, Wang CC, Wu MH, Yang SH, Li YH, et al. (2012) Hypoxia-Induced
MicroRNA-20a Expression Increases ERK Phosphorylation and Angiogenic
Gene Expression in Endometriotic Stromal Cells. J Clin Endocrinol Metab 97:
E1515–1523.
42. Biswas I, Garg I, Singh B, Khan GA (2012) A key role of toll-like receptor 3 in
tissue factor activation through extracellular signal regulated kinase 1/2 pathway
in a murine hypoxia model. Blood Cells Mol Dis 49: 92–101.
43. Hota KB, Hota SK, Chaurasia OP, Singh SB (2012) Acetyl-L-carnitinemediated neuroprotection during hypoxia is attributed to ERK1/2-Nrf2regulated mitochondrial biosynthesis. Hippocampus 22: 723–736.
9
January 2013 | Volume 8 | Issue 1 | e55290