Oncotarget45200
www.impactjournals.com/oncotarget
www.impactjournals.com/oncotarget/ Oncotarget, 2017, Vol. 8, (No. 28), pp: 45200-45212
Mesenchymal stem cells release exosomes that transfer miRNAs
to endothelial cells and promote angiogenesis
Min Gong
1,2
, Bin Yu
1
, Jingcai Wang
1
, Yigang Wang
1
, Min Liu
1
, Christian Paul
1
, Ronald
W. Millard
1,3
, De-Sheng Xiao
4
, Muhammad Ashraf
1,5
and Meifeng Xu
1
1
Department of Pathology and Laboratory Medicine, University of Cincinnati Medical Center, Cincinnati, Ohio, USA
2
Children’s Nutrition Research Centre, Children’s Hospital of Chongqing Medical University, Chongqing, China
3
Department of Pharmacology and Cell Biophysics, University of Cincinnati Medical Center, Cincinnati, Ohio, USA
4
Department of Preventive Medicine, School of Public Health, Guangzhou Medical University, Guangzhou, Guangdong
Province, China
5
Department of Pharmacology, University of Illinois at Chicago, Chicago, Illinois, USA
Correspondence to: Meifeng Xu, email: Meifeng.xu@uc.edu
Keywords: exosomes, miRNA transfer, mesenchymal stem cells, angiogenesis, miR-30b
Received: December 19, 2016 Accepted: March 21, 2017 Published: April 01, 2017
Copyright: Gong et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License 3.0 (CC BY
3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
ABSTRACT
Mesenchymal stem cells (MSCs) have been found to benet patients with a
variety of ischemic diseases via promoting angiogenesis. It is also well established
that exosomes secreted from MSCs deliver bioactive molecules, including microRNAs
(miRs) to recipient cells. Therefore, we hypothesized that exosomes secreted from
MSCs deliver miRs into endothelial cells and mediate angiogenesis. The pro-angiogenic
stimulatory capacity of exosomes was investigated using tube-like structure formation
and spheroid-based sprouting of human umbilical vein endothelial cells (HUVECs), and
in vivo Matrigel plug assay. The secretion of pro-angiogenic miRs (pro-angiomiRs)
from MSCs into culture medium and transfer of the miRs to HUVECs were conrmed
using real-time quantitative PCR. Supplementation of the exosome secretion blocker
GW4869 (10 μM) reduced the pro-angiomiRs in the MSC-derived conditioned medium
(CdM
MSC
). Addition of exosomes isolated from CdM
MSC
could directly 1) promote HUVEC
tube-like structure formation in vitro; 2) mobilize endothelial cells into Matrigel plug
subcutaneously transplanted into mice; and 3) increase blood ow inside Matrigel
plug. Fluorescence tracking showed that the exosomes were internalized rapidly by
HUVECs causing an upregulated expression of pro-angiomiRs in HUVECs. Loss-and-
gain function of the pro-angiomiRs (e.g., miR-30b) in MSCs signicantly altered the
pro-angiogenic properties of these MSC-derived exosomes, which could be associated
with the regulation of their targets in HUVECs. These results suggest that exosomal
transfer of pro-angiogenic miRs plays an important role in MSC mediated angiogenesis
and stem cell-to-endothelial cell communication.
INTRODUCTION
It has been estimated that more than 500 million
people will be beneted from various angiogenic therapies
in the coming decades for treating ischemic diseases
such as peripheral and coronary vascular disease [1],
cerebral infarction [2], and critical limb ischemia [3].
Angiogenesis is a complex biological process involving
interactions between vascular cells and the extracellular
environment. Cell-based pro-angiogenic therapies have
been increasingly utilized in the treatment of ischemic
diseases [4, 5]. Consequently, stem cells have been
extensively used to experimentally treat ischemic diseases
including myocardial infarction [6, 7] and stroke [8, 9].
Mesenchymal stem cells (MSCs) have been recognized as
a promising treatment option with the potential to generate
a variety of useful cell-based interventions [10, 11] and
pro-angiogenic therapies [12, 13]. Indeed, both in vitro
and in vivo models have shown that MSCs can increase
endothelial cell growth and enhance new blood vessel
formation [14], as a result of paracrine effects that are
considered as the predominant mechanism in addressing
Research Paper
Oncotarget45201
www.impactjournals.com/oncotarget
tissue damage [15]. We previously demonstrated that
conditioned medium (CdM) of MSCs promoted post-
infarction angiogenesis in ischemic myocardium and
global heart function recovery [16]. However, the exact
molecular mechanisms responsible for these benecial
paracrine effects of MSCs have not been identied.
Exosomes are cell-derived vesicles (diameter
30–100 nm) that exist in almost all biological uids
including blood, urine, saliva, cerebrospinal uid, and
cell preconditioned medium [17, 18]. They are initially
formed by fusion of a multi-vesicular body with a
plasma membrane, or released directly from the plasma
membrane [17, 19]. Exosomes shuttle mRNAs, miRs,
and other molecular constituents to achieve cell-to-cell
communication, and modulate the function of recipient
cells [20]. However, exosomes contents vary from different
cell types, pathological conditions and by preconditioning
or genetic manipulation of the parent MSCs [21, 22],
which might cause completely inversed fate of target cells.
Most recently, the existence of miRs in exosomes
has been reported [23–25], suggesting that exosomes
may serve as a vehicle for miR transfer and mediate
intercellular communication [26]. MiRs, a class of small
non-coding RNAs (containing about 18–22 nucleotides),
regulate gene expression on the posttranscriptional
level by binding to specic mRNA and inducing
their degradation and/or translational inhibition [27].
MiRs are recognized to participate in a wide range of
biological and pathological processes including the cell
cycle, hematopoiesis, neurogenesis, aging, cancer, and
cardiovascular disease [28]. Evidence has suggested that
miRs are key regulators of endothelial cell function and
are especially important modulators of angiogenesis [29].
For instance, it has been reported that miR-424 promoted
angiogenesis in vitro and in a mouse model by targeting
cullin 2 [30]. miR-30 family targeted DLL4 in endothelial
cells to promote angiogenesis [31]. The present study was
designed to investigate whether MSC-derived exosomes
shuttle various pro-angiogenic miRs and transfer
these miRs to endothelial cells resulting in promoting
angiogenesis.
RESULTS
Pro-angiogenic capacity of conditioned medium
derived from MSCs
MSCs line C3H10T1/2 cells were purchased from
ATCC (Manassas, VA, USA). MSCs adhered to the surface
of plastic culture dishes and exhibited a spindle-shaped
broblast-like morphology as shown in the Supplementary
Figure 1. The pro-angiogenic capacity of CdM obtained
from these cells (CdM
MSC
) was assessed using tube-like
structure formation, spheroid-based sprouting of HUVECs
and in vivo Matrigel plug assay. The cumulative tube
length was signicantly longer (31.80 ± 3.37 mm/eld) in
HUVECs treated with CdM
MSC
compared to those treated
with control medium (18.69 ± 2.83 mm/eld) following
culture for 16 h (Figure 1A). Sprout length per spheroid in
HUVECs treated with CdM
MSC
for 16 h was signicantly
longer (216.67 ± 36.29 μm/spheroid) than that treated
with control medium (82.66 ± 32.23 μm/spheroid)
(Figure 1B). The effect of CdM
MSC
on endothelial cell
invasion and hemoglobin concentration in Matrigel plug
was investigated following subcutaneous injection of
Matrigel into C57BL6 mice. The Matrigel plug contained
CdM
MSC
had a red gross appearance after transplanting
for 14 days (Figure 1C). The hemoglobin content (a sign
of increased new vessel formation) was signicantly
increased in the plugs containing CdM
MSC
(11.14 ±
5.01 μg/mg plug) compared to the Matrigel plugs without
CdM
MSC
(2.48 ± 1.19 μg/mg plug) (Figure 1D). The
neovasculature visualized by immunouorescence staining
of CD31 indicated that the number of CD31 positive cells
in the plugs containing CdM
MSC
was signicantly higher
than that without CdM
MSC
(Figure 1E).
miRs secreted from MSCs transfer to HUVECs
The expression of 26 most commonly
acknowledged pro-angiogenic miRs (pro-angiomiRs) in
CdM
MSC
was quantified using real-time PCR before and
after adding into HUVEC culture. The expression of
miR-424, miR-30c, miR-30b, and let-7f in conditioned
medium was significantly reduced after adding into
HUVECs culture for 48 h, indicating that extracellular
miRs, derived from MSCs, transferred into HUVECs.
Meanwhile, the expression of miR-21, miR-10a, miR-
126, miR-10b, miR-19a, miR-19b was significantly
increased after adding into HUVECs culture,
suggesting that HUVECs might release these miRs
(Table 1). The expression of other miRs was either
very low in CdM
MSC
or the change was not significant
(Supplementary Table 1).
Next, the transfer of miRs between MSCs and
HUVECs was determined using a non-contact co-culture
system. The CdM was collected from MSC-HUVEC
co-culture (CdM
MSC-HUVEC
) and its controls, MSC-MSC
co-culture (CdM
MSC-MSC
) and HUVEC-HUVEC co-
culture (CdM
HUVEC-HUVEC
), respectively. The expression
of miR-30b, 30c, 424 and let-7f in CdM
HUVEC-HUVEC
was
very low and in CdM
MSC-MSC
was very high. However,
the expression of these miRs in CdM
MSC-HUVEC
was
signicantly lower compared to those in CdM
MSC-MSC
,
indicating that these miRs transferred into HUVECs
(Figure 2A). Moreover, the expression of these miRs in
HUVECs co-cultured with MSCs was signicantly higher
than in those without co-cultured with MSCs (Figure 2B),
directly demonstrating a transfer of these miRs into
HUVECs.
Oncotarget45202
www.impactjournals.com/oncotarget
Exosomes derived from MSCs deliver pro-
angiomiRs and promote angiogenesis
To investigate whether exosomes mediated
miRs transfer, the expression of miRs in CdM was
measured after MSCs were treated with 10 μM GW4869
(an exosome release inhibitor) for 48 h [32]. As shown
in Figure 3A, the levels of miR-30b, -30c, -424, and
let-7f in the CdM collected from MSCs treated with
GW4869 (CdM
GW4869
) were signicantly decreased
compared with CdM obtained from control MSCs. In
addition, the expression of these miRs in HUVECs
treated with CdM
GW4869
for 48h was also signicantly
decreased compared to those treated with CdM
MSC
(Figure 3B), indicating that exosomes mediated miR
transportation between MSCs and HUVECs. Exosomes,
isolated from CdM
MSC
, exhibited the characteristic round
morphology with heterogeneous size under transmission
electron microscope (Figure 4A). The average size of
exosomes was 48.72 ± 2.7nm according to the results
of dynamic light scattering (Figure 4B). The expression
of CD9, CD63, and HSP70 was signicantly higher in
exosomes compared to their parent MSCs (Figure 4C).
The internalization of exosome pre-labeled with PKH26
by HUVECs was recorded using IncuCyte ZOOM Live
Content Imaging System every 2 h for 12 h. The PKH26
red uorescence intensity increased with the passage of
time and achieved its maximum after exosomes were
added into HUVECs culture for 10 h (Figure 4D). The
expression of miR-30b, 30c, 424, and let-7f in HUVECs
treated with exosomes for 24 h was signicantly increased
compared to those treated with BSA (Figure 4E).
Figure 1: CdM derived from MSCs promotes angiogenesis. (A) Representative images of capillary-like structures and quantitative
analysis of the total tube length (4× magnication microscopic elds); (B) Representative images of HUVEC spheroids sprouting and
quantitative analysis of the cumulative sprout length per spheroid; (C) Representative gross look of Matrigel plugs which were implanted
subcutaneously in mice for 14 days; (D) Hemoglobin content in the Matrigel plugs; (E) The neovasculature in Matrigel was visualized
by immunouorescence staining of CD31. Quantication of the CD31-positive cells in Matrigel plugs were determined by pixel density
(*P < 0.05 vs CON).
Oncotarget45203
www.impactjournals.com/oncotarget
The pro-angiogenic capacity of exosomes was
examined in vitro and in vivo. The cumulative tube length
was signicantly increased in HUVECs treated with
exosomes (100μg/ml) for 16 h (36.24 ± 3.65 mm/eld)
compared to that treated with BSA in the same protein
amount (15.73 ± 2.44 mm/eld) (Figure 5A). The Matrigel
plug test showed a more red appearance in those containing
exosomes (100μg) after being transplanted for 14 days
than those without exosomes (Figure 5B). The hemoglobin
concentration in the plugs containing exosomes
was signicantly higher (11.76 ± 5.61 μg/mg plug)
than those without exosomes (2.54 ± 1.45 μg/mg plug)
(Figure 5C). The immunouorescence staining showed
that the number of CD31 positive cells in the plugs
containing exosomes was also signicantly higher than
those without exosomes (Figure 5D).
To demonstrate the effect of transferred miRs on
angiogenesis, miR-30b was selected as a representative
miR. Exosomes were obtained from MSCs in which
miR-30b was overexpressed or knockdown, respectively.
MSCs were infected with lentivirus carrying the pre-miR-
30b fragment (MSC
miR-30b
). The expression of miR-30b in
MSC
miR-30b
and exosomes derived from MSC
miR-30b
(Exo
miR-
30b
) was 5.24-fold and 5.22-fold increase compared with
their counterpart MSC
scrambled
and Exo
scrambled
, respectively
(Figure 6A). The cumulative tube length was increased in
Table 1: The expression of pro-angiogenic miRNAs in CdM
MSC
after adding into HUVECs culture
for 48 hours
Downregulated Upregulated
miRNA
CdM
MSC
2
(−ΔCt)
CdM
MSC
with HUVECs 2
(−ΔCt)
miRNA
CdM
MSC
2
(−ΔCt)
CdM
MSC
with HUVECs 2
(−ΔCt)
miR-424
#
44.965 ± 5.542 10.725 ± 1.795
*
miR-21 89.021 ± 9.117 187.956 ± 27.620
*
miR-30c 6.420 ± 0.623 0.572 ± 0.140
*
miR-10a 0.435 ± 0.040 10.160 ± 0.985
*
miR-30b 5.877 ± 0.692 0.133 ± 0.012
*
miR-126 0.045 ± 0.014 6.988 ± 0.933
*
let-7f 4.592 ± 0.245 0.153 ± 0.003
*
miR-10b 0.008 ± 0.002 5.869 ± 0.442
*
miR-19a 1.623 ± 0.063 3.380 ± 0.316
*
miR-19b 1.540 ± 0.116 2.950 ± 0.225
*
(*P < 0.05 vs CdM
MSC
).
#
The mouse homologue of miR-424 sequence from human is miR-322-5p.
Figure 2: miRs secreted from MSCs transfer to HUVECs. (A) miRNA expression in the CdM of non-contact cell co-culture
system (*P < 0.05 vs CdM
HUVEC-HUVEC
;
&
P < 0.05 vs CdM
MSC-MSC
, respectively); (B) The expression of miRs in HUVECs after co-culture
with MSCs (*P < 0.05 vs HUVEC alone).
Oncotarget45204
www.impactjournals.com/oncotarget
Figure 4: Characterization of exosomes and their involvement in miRs transfer. (A) Morphology of exosomes under
transmission electron microscopy; (B) The exosome size was measured using a Zetasizer Nano instrument. (C) The expression of CD9,
CD63 and HSP70; (D) Representative images of time-lapse internalization of PKH26-labled exosomes (red) in HUVECs. (E) miR
expression in HUVECs following exosome treatment (*P < 0.05 vs HUVEC+ BSA).
Figure 3: Exosomes mediate the transfer of miRs from MSCs to HUVECs. (A) The expression of miRs in CdM
(*P < 0.05 vs
CdM
MSC
); (B) The expression of miRs in HUVECs treated with CdM (
&
P < 0.05 vs HUVEC + CdM
MSC
).