original article http:/ /www.kidney-international.org
© 2012 International Society of Nephrology
see commentary on page 375
Microvesicles derived from endothelial progenitor
cells protect the kidney from ischemia–reperfusion
injury by microRNA-dependent reprogramming of
resident renal cells
Vincenzo Cantaluppi1, Stefano Gatti 2, Davide Medica 1, Federico Figliolini1, Stefania Bruno 1,
Maria C. Deregibus 1, Andrea Sordi2, Luigi Biancone 1, Ciro Tetta 3,4 and Giovanni Camussi1
1 Dialysis and Kidney Transplantation Unit, Department of Internal Medicine, Center for Experimental Medical Research (CeRMS) and
Nephrology, University of Torino, Torino, Italy; 2Center for Surgical Research, Fondazione IRCCS Cd Granda Ospedale Maggiore
Policlinico, Milano, Italy; 3Sis-Ter SpA, Palazzo Pignano (CR), Italy and 4Fresenius Medical Care, Bad Homburg, Germany
Endothelial progenitor cells are known to reverse acute
kidney injury by paracrine mechanisms. We previously found
that microvesicles released from these progenitor cells
activate an angiogenic program in endothelial cells by
horizontal mRNA transfer. Here, we tested whether these
microvesicles prevent acute kidney injury in a rat model
of ischemia-reperfusion injury. The RNA content of
microvesicles was enriched in microRNAs (miRNAs) that
modulate proliferation, angiogenesis, and apoptosis. After
intravenous injection following ischemia-reperfusion, the
microvesicles were localized within peritubular capillaries
and tubular cells. This conferred functional and morphologic
protection from acute kidney injury by enhanced tubular
cell proliferation, reduced apoptosis, and leukocyte
infiltration. Microvesicles also protected against progression
of chronic kidney damage by inhibiting capillary rarefaction,
glomerulosclerosis, and tubulointerstitial fibrosis. The
renoprotective effect of microvesicles was lost after
treatment with RNase, nonspecific miRNA depletion of
microvesicles by Dicer knock-down in the progenitor cells,
or depletion of pro-angiogenic miR-126 and miR-296
by transfection with specific miR-antagomirs. Thus,
microvesicles derived from endothelial progenitor cells
protect the kidney from ischemic acute injury by delivering
their RNA content, the miRNA cargo of which contributes to
reprogramming hypoxic resident renal cells to a regenerative
program.
Kidney International (2012) 82, 412-427; doi:10.1038/ki.2012.1 OS;
published online 11 April 2012
KEYWORDSa: cute kidney injury; exosome; ischemia-reperfusion
Correspondence: Giovanni Camussi, Cattedra di Nefrologia, Dipartimento
di Medicina lnterna, Ospedale Maggiore S. Giovanni Battista ‘Molinette’,
Corso Dog/iotti 14, 10126, Torino, Italy. E-mail: giovanni.camussi@unito.it
Received 26 April 2011; revised 19 January 2012; accepted 24 January
2012; published online 11 April 2012
412
Ischemia-reperfusion is one of the main causes of acute
kidney injury (AKI).1•2 Therapeutic strategies aimed to
inhibit ischemia-reperfusion injury (IRI) may potentially
limit AKI and the development of chronic kidney disease
(CKD).3 Several studies addressed the role of bone marrowderived
and tissue-resident stem cells in the regeneration of
ischernic kidneys.4-7 Endothelial progenitors (EPCs) are
circulating bone marrow-<lerived precursors able to localize
within sites of tissue damage inducing regeneration. 8’9 EPCs
are known to exert protective effects in experimental models
of hindlimb ischemia, myocardial infarction, and glomerular
diseases. lO-lZ Moreover, it has been recently demonstrated
that EPCs are recruited in the kidney after IRI and that they
induce tissue repair via secretion of pro-angiogenic factors_
13- 15 EPC paucity and dysfunction have been proposed as
mechanisms of accelerated vascular injury in CKD patients. 16
The regenerative effects of EPCs on ischemic tissues have
been ascribed to paracrine mechanisms including the release
of growth factors and microvesicles (MVs). 17•18 MVs are
small particles derived from the endosomal compartment
known to have an important role in cell-to-cell communication
through the transfer of proteins, bioactive lipids, and
RNA to target cells.19- 22 We recently demonstrated that MVs
released from EPCs are internalized into endothelial cells
activating an angiogenic program by horizontal transfer of
mRNAs.18
The aim of this study was to evaluate whether MV s
released from EPCs exert a protective effect in an experimental
model of acute renal IRI. Moreover, we studied
in vitro the mechanisms of MV protection from hypoxiainduced
endothelial and epithelial kidney cell injury.
RESULTS
Characterization of EPC-and fibroblast-derived MVs
Transmission electron microscopy on EPCs revealed the
shedding of MVs (Figure la and b) by a membrane-sorting
process (Figure le). Purified MVs showed a homogenous
Kidney lnternatianal (2012) 82, 412-427
V Cantaluppi et al.: EPC microvesicles and kidney ischemia–reperfusion injury O r i g i n a I a rt i C I e
c d
Shedding vesicles
;;;.’ “- .Jt ,.;rocess
Plasma ~t._J ;:- 2;–‘ A”–,rmembrane o C o 0 0 e 132 cum% con/ml E6 0% f 50 c”‘ 40 30 => 20 8 1g 50 c”‘ 40 30 :, 20 8 10 0 2 90% Results 80% ·~u; Mean: 159 nm SO: 63 nm 70% Concentration: 15.62 x 108 C 60% Q) particles/ml .£ 50% Q) > 40% ·cu 30% a, a: 20% 537 10% C) “C)Cc)f ‘C), ,,ci.C,,l c,iC,,cl ici’o C),C.,) ci’CoCl )C0:), ClCn)m
Particle size
50□ 50
.!1 40 00 40
C 30 c 30 6 20 15 20
(.) 10 O 10
0 0
100 101 102 103 104 100 10 1 102 103 104 100 10 1 102 103 104
a4 integrin ~1 integrin CD34
!!ll8LI
50
.!1 40
C 30 C 30
8:, 20 5 20 10 0 10
0 . 0
100 101 102 103 104 100 101 102 103 104 10° 101 102 103 104
CD154 L-selectin HLA class I
200 200– .!1 160 _:g1 60 “‘126000 □
C 120 § 120 C 120
5 80 o 80 6 80
(.) 40 (.) 40 (.) 40
0
~00 101 102 103 104
0
10° 101 102 103 104 100 101 102 103 104
CD42b P-selectin CD14
g
{F~iu ,W I
EPC
25 200 1000 4000 (nt)
{F~iu
:,: I I I I
EPCMV
25 200 1000 4000 (nt)
h
Figure 1 I Characterization of endothelial progenitor (EPC)-derived microvesicles (MVs). (a, b) Transmission electron microscopy
performed on cultured EPCs showing MV shedding by a membrane-sorting process. (cl Schematic representation of shedding MV formation
by budding of plasma membrane. (d) Transmission electron microscopy analysis of purified MVs showing a spheroid shape. In a, b, and d,
bars indicate 100 nm. (e) Nanosight analysis of purified MVs: curve 1 describes the relationship between particle number distribution
(left Y axis) and particle size (X axis); curve 2 describes the correlation between cumulative percentage distribution of particles (percentile
in right Y axis) and particle size (X axis). Mean size and particle concentration values were calculated by the Nanoparticle Tracking
Analysis (NTA) software that allows analysis of video images of the particle movement under Brownian motion captured by Nanosight
LMl0 and calculation of the diffusion coefficient, sphere equivalent, and hydrodynamic radius of particles by using the Strokes-Einstein
equation. (f) Fluorescence-activated cell sorting (FACS) analysis of MV protein surface expression. (g) Bioanalyzer RNA profile of EPCs and
EPC-derived MVs. (h) Analysis of microRNAs (miRNA array) present in EPCs and EPC-derived MVs (white circle: EPCs; gray circle:
EPC-derived MVs).
pattern of spheroid particles. About 90% of MVs showed a
size ranging from 60 to 160 nm (Figure ld), whereas a
minority of them were larger with a size around 1 μm. The
Kidney International (2012) 82, 412-427
purity and the size of EPC-MV preparations were confirmed
by Nanosight analysis (Figure le). By fluorescence-activated
cell sorting (FACS) analysis, £PC-derived MVs expressed cx4
413
original article V Cantaluppi et al.: EPC microvesicles and kidney ischemia;eperfusion injury
a EPC
Amplificationp lot
Wild-type “”
miR-296
miR-126
siRNA DICER
Ami R 126/296
O’l,t,.’b’b~~,0,’V’°,§>,9,,t~,§J
Cyde
3.00
2.75
2.00
2.25
2.00
1.75
&_ 1.50
1.25
1.00
0.75
Amplificationp lot
g:~ miR-296
J:~ E=======~:JmiR-126
()’I, t,. ‘o ‘b.,Q…‘l,.t-.‘b..‘b,£),0,,’/J’~’l,’b,§>’9-,!J’,§l,§>§>t,.’l- Cyc~ Amplificationp lot 3.00~——–‘—-~ 2.75 200 2.25 2.00 1.75 1.00 /i. 1.25 1.00 0.75 0.00 0.25 miR-296 J: —-–––––––––m-iR-1-2_6- -~.:–,
O’l,t-b’b~,£)H~4-n.t~4′ Cyde b “” EPCMV Amplificationp lot 13 12 11 0 10 9 8 7 6 5 4 3 2 0′ ()’l,t,.'()’b~,£),0,~,§>~4′
Cyde
Amplificationp lot
6.5 ~——–~
6.0
5.5
5.0
4.5
4.0
miR-126
miR-296
Wild-type
&. g:g RNase
2.5
2.0
1.5
6:g miR-126
o.o ‘-===============:JmiR-296
() ‘l, t,. ‘o ‘b ,t:>..‘l,.._t-..‘b,’b’l,Q,fJ,qJ-,§>4′,t,§),§J~Qi,, Cyc~ Amplificationp lot 6.5 ~——–~ 6.0 5.5 5.0 4.5 4.0 3.5 Ii_ 3.0 2.5 2.0 1.5 1.0 miR-126 Z& ..:=.=.=.=.=.=.=.=.=.=.=.=.=.=.===miR-296
o ‘l, t,. ,’b,§),fJ,rJ”~rr,’b,§>~,’l9, ,nJ-~4’
Cycle
3.00
2.75
2.00
2.25
2.00
1-75
&. ~:~
1.00
0.75
Amplificationp lot
8:~ miR-126
J:~E========:J miR-296
O’l,t-‘b’b~~r(H’,§>,§J~~
Cyc~
siRNA DICER
AmiR 126/296
Figure 21 Representative quantitative reverse transcriptase (qRT)-PCR for miR-126 and miR-296 in endothelial progenitors
(EPCs) and EPC-derived microvesicles (MVs). (a) qRT-PCR analysis of miR-126 and miR-296 content in EPCs cultured with vehicle alone
(wild-type), subjected to small interfering RNA (siRNA) for Dicer (siRNA Dicer), or transfected with anti-miR-126 and anti-miR-296 antagomiRs
(AmiR 126/296). (b) qRT-PCR analysis of miR-126 and miR-296 content in MVs derived from EPCs cultured with vehicle alone (wild-type),
treated with 1 U/ml RNase (RNase), subjected to siRNA for Dicer (siRNA Dicer), or transfected with anti-miR-126 and anti-miR-296
antagomiRs (AmiR 126/296).
and ~l integrin, CD154 (CD40-L), L-selectin and CD34 but
not human leukocyte antigen class I and class II antigens and
markers of platelets (P-selectin, CD42b) and monocytes
(CD14) (Figure lf). Bioanalyzer profile of EPC-derived MVs
showed the presence of different subsets of RNAs and in
particular enrichment for small RNAs, including microRNAs
(miRNAs) (Figure lg): miRNA array analysis showed the
presence of 131 miRNAs shared by EPCs and EPC-derived
MVs and 26 miRNAs specifically concentrated in MVs
(Figure lh, Supplementary Information Tables Sl and S2).
The presence in EPCs and EPC-derived MVs of several
pro-angiogenic and anti-apoptotic miRNAs, including
miR-126 and miR-296, was confirmed by quantitative
reverse transcriptase (qRT)-PCR with specific primer pairs
(Figure 2). The expression of miR-126 and miR-296 seen by
qRT-PCR was abrogated by RNase treatment ofMVs and was
absent in MVs derived from Dicer-silenced or antagomiRtransfected
EPCs (Figure 2). MVs derived from fibroblasts
were also characterized and used as negative experimental
414
control. Fibroblast-derived MVs were larger than those of
EPCs with a mean size of 260 nm detected by Nanosight (not
shown). By FACS analysis, fibroblast-derived MVs expressed
cx4 and Pl integrin, CD154 and L-selectin, but not CD34,
class I and class II human leukocyte antigens, and markers
of platelets (P-selectin, CD42b) and monocytes (CD14)
(Figure 3a and b). In comparison with EPC-derived MVs,
fibroblast-derived MVs expressed significantly lower levels of
L-selectin (Figure 3b). Bioanalyzer profile of fibroblastderived
MVs showed the presence of different subsets of
RNAs including miRNAs (Figure 3c). The qRT-PCR analysis
with specific primer pairs evidenced the absence of the proangiogenic
miR-126 and miR-296 within fibroblast-derived
MVs (Figure 3d).
Protective effect of EPC-derived MVs in experimental
renal IRI
We evaluated the effects of EPC-derived MVs in an experimental
model of acute renal IRI in Wistar rats (experimental
Kidney lnternatianal (2012) 82, 412-427
V Cantaluppi et al.: EPC microvesicles and kidney ischemia–reperfusion injury original article:[] § 120 0 80 O 40 0 – 100 101 102 103 104 u4 integrin 200 u, 160 § 120 8 80 40 0 100 101 102 103 104 CD154 200,–.——, w 160 ~ 1~g O 40 0 10° 101 102 103 104 CD42b C FIBRO 210600 □ E 120 . 5 80 o 40 0 100 101 102 103 104 p1 integrin 200 160 E 120 is 80 (.) 40 0 100 101 102 103 104 L-selectin 200~–~ 160 E 120 5 80 o 40 0 ._ _ _J 10° 101 102 103 104 P-selectin ,,~iu, 25 200 200 w 160 § 120 0 80 (.) 40 0 100 101 102 103 104 CD34 200 160 ‘ ~ 120 0 80 (.) 40 0 100 101 102 103 104 HLA class I 200~–~ 160 ~ 120 5 80 o 40 0 100 101 102 103 104 CD14 I~
1000 4000
b
C
0 ·u;
If)
~
a.
X “‘ c (“.) ‘
If)
~
0
:::, -=
~
C
:””;’; ‘:
(nt)
(FU)
FIBRO MV ‘! ~1::l~,~::;:,:::::::::::;;::,:::,: :·:::,;::•:::::::~:::::===
25 200 1000 4000 (nt)
60
50
40
30
20
10
0
0 EPC MV
■ FIBRO MV
a4 p1 CD34 CD154 L- HLA CD42b P- CD14
integrin integrin selectin class I selectin
d i
l8
45
40
FIBRO ii!
l& miR-296
Jc========~ miR-126
i
l8
FIBRO MV C ~
“‘ !&
Jg miR-296
8 -,c.:.:.:=====:…:..::.:.i miR-126
o ‘1-t, ‘o ~ ..0. ,4._t-,ro.~rf->#~ff
Cyde
Figure 3 I Characterization of fibroblast-derived microvesicles (MVs). (a) Fluorescence-activated cell sorting (FACS) analysis of
fibroblast-derived MV protein surface expression. (b) Comparison between FACS analysis of fibroblast- and endothelial progenitor (EPC)derived
MV protein surface expression. (c) Bioanalyzer RNA profiling of fibroblasts and fibroblast-derived MVs. (d) Representative quantitative
reverse transcriptase (qRT)-PCR analysis for miR-126 and miR-296 in fibroblasts and fibroblast-derived MVs.
plan in Figure 4). In comparison with sham-operated animals,
rats subjected to kidney IRI showed a significant rise in
serum creatinine (Figure Sa) and blood urea nitrogen (BUN)
(Figure Sb) that peaked at day 2 in association with histological
signs of tubular injury such as formation of hyaline casts,
vacuolization, widespread necrosis, and denudation of basal
membrane (Figure Sc and Table 1). When rats were treated
with £PC-derived MVs, a significant reduction of tubular
lesions in parallel with the decrease in serum creatinine and
BUN was observed at day 2 (Figure Sa-c and Table 1). The
specificity of £PC-derived MVs was indicated by the absence
of protective effect exerted by MVs derived from human
fibroblasts (Figure Sa and b and Table 1). £PC-derived MVs
enhanced the proliferation rate of tubular cells after IRI as
detected by bromo deoxy uridine (Figure 6a and c) and
proliferating cell nuclear antigen (Figure 66 and d) staining.
Moreover, as shown by TdT-mediated dUTP nick end labeling
assay (Figure 7a and b), MVs significantly reduced the number
of apoptotic tubular cells. These renoprotective effects were
significantly reduced when MVs were pre-treated with 1 U/ml
RNase (Figures Sa-c, 6a-<l, 7a-c and Table 1). When MVs
derived from Dicer knocked-down EPCs or MVs released from
EPCs transfected with anti-miR126 and anti-miR-296 antagomiRs
were used, a significant reduction of the functional and
Kidney International (2012) 82, 412-427
histological protective effects on ischemic kidneys was also
observed (Figures Sa-c, 6a-b, 7a and Table 1). Moreover, in
comparison with sham-operated animals, IRI induced a
massive infiltration of granulocytes (Figure 7c) and monocytes
(Figure 7d) within kidneys. A significant decrease in leukocyte
infiltration was observed in rats subjected to IRI and
injected with MVs but not with RNase-treated MVs (Figure 7c
and d). Similar functional and histological renoprotective
effects of MVs were observed also at day 7 after IRI (not
shown).
Six months after IRI, animals treated with MVs showed
reduced levels of serum creatinine (Figure Sa), tubulointerstitial
fibrosis, and glomerulosclerosis (Figure Sb), as
well as a preserved expression of rat endothelial cell antigen-1
antigen in the tubulo-interstitial structures (Figure Sc and d)
and within the glomeruli (Figure Se and f), suggesting an
inhibition of microvascular rarefaction and of progression
toward CKD.
In biodistribution experiments, the accumulation of
PKH26-labeled MVs was observed in the kidney 2 and 6 h
after IRI. After 2 h, MVs were detectable within the endothelial
cells of large vessels and within some peritubular
capillaries and lumen of injured tubules (Figure 9a and b).
After 6 h, the amount of tubular cells containing MV s was
415
original article V Cantaluppi et al.: EPC microvesicles and kidney ischemia;eperfusion injury
markedly enhanced (Figure 9c). When injected in shamoperated
control rats, the renal accumulation was significantly
lower than in IRI, and only a slight staining for MV s
was detected within glomeruli and tubular cells (Figure 9d).
MVs were also detected in the liver of sham-operated
controls, as well as in rats subjected to kidney IRI (Figure 9e).
In vitro effects of EPC-derived MVs on hypoxic peritubular
endothelial cells (TEnCs) and tubular epithelial cells (TEpCs)
In consideration of the in vivo localization of MVs in
peritubular capillaries and tubular cells, we evaluated the role
MaleWistar
rats
IRI
Right nephrectomy+
45 min. left pedicle
clamping
!
Dayo
l
MV injection
(30 μg) in the
tail vein immediately
after IRI
Day 2
l
Killed
n= 6 per group
(all groups)
functional and
hystological
analysis
(BrdU, PCNA,
TUNEL, leukocyte
infiltration)
of adhesion molecules in the internalization of MVs in
isolated human TEnCs and TEpCs. MVs were efficiently
internalized in TEnCs (Figure 9f) and TEpCs (Figure 9g).
Moreover, hypoxia significantly enhanced MV internalization
in both cell types (Figure 9h and i). Experiments conducted
with blocking antibodies revealed that L-selectin was the main
mediator of MV internalization in hypoxic cells (Figure 9h
and i). Internalization was not altered in RNase-treated MVs
(not shown). Control fibroblast-derived MVs showed a
reduced internalization in normoxic and hypoxic TEnCs
and TEpCs (Figure 10).
Experimental groups:
1.Normal
- Sham operated
- IRI
- IRI + MV EPC
- IRI + MV EPC RNase
- IRI + MV EPC siRNA DICER
- IRI + MV EPC siRNA control
- IRI + AmiR126/296 (antagomiRs)
- IRI + MV fibroblast
Day?
l
Killed
n=6 per group
(all groups)
functional and
hystological
analysis
(BrdU, PCNA,
TUNEL, leukocyte
infiltration)
Day 180
l
Killed
n=6 per group
(groups 1-4)
functional and
hystological
analysis
Figure 41 Representative scheme of the experimental plan of acute renal ischemia;eperfusion injury (IRI) in male Wistar rats.
Schematic representation of IRI model, experimental groups, number of animals treated, modality and dose of MV injection, timing of
killing, and functional/histological analysis performed. BrdU, bromo deoxy uridine; EPC, endothelial progenitor; MV, microvesicle; PCNA,
proliferating cell nuclear antigen.
Figure 5 I Protective effect of endothelial progenitor (EPC)-derived microvesicles (MVs) on acute kidney ischemia-f”eperfusion injury
(IRI). (a, b) Evaluation of serum creatinine (a) and blood urea nitrogen (BUN) (b) in different experimental groups. IRI induced a significant
increase in serum creatinine and BUN (*P< 0.05 IRI vs. sham or normal). EPC-derived MVs significantly decreased serum creatinine and BUN (#P<0.05 IRI + MV EPC vs. IRI). The pre-treatment of EPC MVs with 1 U/ml RNase or the use of MVs released from EPCs transfected with small interfering RNA (siRNA) Dicer or with antagomiRs-126/296 (AmiRl 26/296) did not reduce serum creatinine and BUN (tP<0.05 IRI + MV EPC RNase, IRI + MV EPC siRNA DICER or IRI + MV EPC AmiR126/296 vs. IRI + MV EPC). MVs released from EPCs transfected with an irrelevant control siRNA (siRNA control) significantly decreased serum creatinine and BUN (#P< 0.05 IRI + MV EPC siRNA control vs. IRI). The specificity of EPC-derived MVs was confirmed by the lack of renoprotective effect of MVs derived from control human fibroblasts (tP<0.05 IRI + MV fibroblasts vs. IRI + MV EPC; P> 0.05 IRI + MV fibroblasts vs. IRI). (c) Hematoxylin/eosin staining of
representative kidney sections from different experimental groups (magnification x 100).
416 Kidney lnternatianal (2012) 82, 412-427
V Cantaluppi et al.: EPC microvesicles and kidney ischemia–reperfusion injury original article
Internalization of MVs within hypoxic TEnCs was
followed by reduced apoptosis (Figure lla) and enhanced
angiogenesis on Matrigel-coated surfaces (Figure llb). The
anti-apoptotic and pro-angiogenic effects of MVs on
hypoxic TEnCs was almost completely abrogated by RNase
a 3
‘5 o, 2.5 .s
QJ C 2 ·c
~ 1.5
~ u
E
:::,
<ii
(f) 0.5
0
pre-treatment or by using MVs released by EPCs engineered
to knock-down Dicer or by EPCs transfected with the
selective anti-miR-126 and anti-miR-296 antagomiRs
(Figure lla and b). Gene array analysis revealed that MVs
restored in TEnCs the expression of pro-angiogenic and
Normal Sham IRI IRI + MV IRI + MV IRI + MV IRI + MV IRI + MV IRI + MV
b 90
80
70
‘5 60
o,
.s 50
z 40
::J
CD 30
20
10
0
Normal
C
Sham
Kidney International (2012) 82, 412-427
Sham
IRI + MV EPC
siRNA dicer
IRI
IRI
EPC EPC RNase EPC siRNA EPC siRNA EPC fibroblast
IRl+MV
EPC
dicer control AmiR126/296
IRI + MV IRI + MV IRI + MV
EPC RNase EPC siRNA EPC siRNA
dicer control
IRI +MV
EPC
AmiR126/296
IRI +MV
fibroblast
IRI + MV EPC IRI + MV EPC RNase
IRI + MV EPC
siRNA control
IRI + MV EPC
AmiR126/296
417
original article V Cantaluppi et al.: EPC microvesicles and kidney ischemia;eperfusion injury
anti-apoptotic genes that were downregulated by hypoxia
(Figure llc).
Internalization of MVs within TEpC was followed by a
significant inhibition of hypoxia-induced apoptosis as shown
by TdT-mediated dUTP nick end labeling assay (Figure 12a)
and enzyme-linked immunosorbent assay for caspase-3, -8,
and -9 activities (Figure 12b ). The anti-apoptotic effect of
MVs was inhibited by RNase pre-treatment or by using MVs
released by EPCs engineered to knock-down Dicer or by
EPCs transfected with anti-miR-126 and anti-miR-296
antagomiRs. Gene array analysis revealed that MV stimulation
of hypoxic TEpCs reduced the expression of inflammatory
and pro-apoptotic caspases (Figure 12c) and of genes
involved in both mitochondrial and death receptor pathways
of apoptosis (Figure 12d).
DISCUSSION
In this study, we demonstrated that MVs derived from EPCs
exert a protective effect on experimental acute renal IRI as
detected by the significant decrease in serum creatinine/BUN
levels and by the improvement of histological signs of
microvascular and tubular injury.
EPCs were shown to induce angiogenesis and tissue repair
in experimental models of acute glomerular and tubular
m• J• ury. 12 ‘ 13 ‘ 23-2s Th e on•g m• o f EPC s 1• s st1·1 1a matter o f d eb ate.
Some studies suggested that contamination with monocytes
and platelet-derived products of EPCs derived from circulation
may account for their pro-angiogenic potential. 26 •27 To
avoid such contamination, we purified MVs from EPCs after
3-5 passages in culture. The cells used and the derived MVs
expressed the CD34 stem cell marker and markers of
endothelium, but not of monocytes and platelets. Previous
studies suggested that EPCs do not act via a direct transdifferentiation
into mature endothelial cells, but rather by
paracrine mechanisms. 10•28 We demonstrated that MVs act
as a paracrine mediator as they may enter the target cells
through specific receptor-ligand interactions and deliver
selected patterns of mRNAs and miRNAs. 20 •29- 31 Moreover,
MV s released from mesenchymal stem cells were shown to
favor recovery from toxic and ischemic AKI.32 •33
Table 1 I Morphologic evaluation
Casts (n/HPF) Tubular necrosis (n/HPF)
Normal
Sham
IRI
IRl+MV EPC
IRl+MV EPC RNase
IRl+MV EPC siRNA Dicer
IRl+MV EPC siRNA control
IRl+MV EPC AmiR126/296
IRl+MV fibroblasts
0
0
2.6 ± 1.2
0.48 ± 0.21*
2.93 ±0.84
2.26 ± 1.28
0.38 ± 0.23*
1 .36 ± 0.56·# 2.2±0.94 0 0 2.9 ± 0.42 0.38 ± 0.16
2.82 ± 0.89
1.83 ± 1.19
0.42 ± 0.11*
1.76 ± 0.79*·#
2.3 ± 0.82
Abbreviations: EPC, endothelial progenitor; IRI, ischemia,eperfusion injury;
MV, microvesicle; n/HPF, number/high-power field; siRNA, small interfering RNA.
*P<0.05 IRl+EPC MV, IRl+EPC MV siRNA control or IRl+EPC MV AmiR126/296 vs. IRI;
‘P<0.05 IRl+EPC MV AmiR126/296 vs. IRl+EPC MV.
Renal morphology score in different experimental groups: n/HPF.
418
Herein, we demonstrated that EPC-derived MVs protected
kidney from IRI-induced functional impairment and morphologic
injury. Indeed, the administration of MVs significantly
decreased serum creatinine and BUN levels,
renal cell apoptosis, and leukocyte infiltration. Moreover,
MVs enhanced tubular cell proliferation and angiogenesis.
These renoprotective effects were specific for EPC-derived
MVs, as MVs obtained from human fibroblasts were
ineffective.
Little is known at present about the biogenesis and the
molecular composition of MVs produced by EPCs in
different physiopathological states. It is supposed that the
production of MVs is enhanced after appropriate stimulation.
In this study, we evaluated the effects of MVs
released from EPCs in basal culture conditions. However,
our preliminary results indicate that hypoxia enhances
the production from EPCs of MVs carrying miR-126 and
miR-296 (not shown).
In vivo, MVs were detected both in endothelial cells and in
tubular epithelial cells. The in vitro studies on isolated
hypoxic TEnCs and TEpCs demonstrated that L-selectin was
instrumental in MV internalization, probably through the
binding to fucosylated residues or other oligosaccharide
ligands known to be upregulated after IRI.34 •35
It is known that IRI induces both microvascular and
tubular injury and that TEnC dysfunction is associated with an
extension phase of AKI.36’37 Moreover, the rarefaction of renal
microvascular density in the presence of sustained hypoxia is
associated with an accelerated progression toward CKD.38 On
this basis, we observed the effects of EPC-derived MVs on
kidneys 6 months after IRI, suggesting that MVs significantly
reduced glomerulosclerosis, tubulo-interstitial fibrosis, and
microvascular rarefaction, thus preserving renal function.
The results of this study suggest that the protective effects
of EPC-derived MVs in experimental renal IRI seem to be
associated with the triggering of angiogenesis in TEnCs and
by the inhibition of apoptosis in TEpCs. Indeed, the
detrimental effects induced by hypoxia on TEnCs were
limited by MVs. It is interesting to note that gene array
analysis of MY-stimulated hypoxic TEnCs revealed the
upregulation of molecules involved in cell proliferation,
angiogenesis, and inhibition of apoptosis. After an ischemic
damage, TEpCs are subjected to loss of polarity with
mislocalization of proteins located at the apical or at the
basolateral membrane and finally to necrosis and/or
apoptosis. 39•40 Herein, we showed that MVs protected TEpCs
from hypoxia-induced apoptosis through the downregulation
of inflammatory and pro-apoptotic caspases and by modulation
of molecules involved in the mitochondrial and death
receptor pathways.41 ’42
We observed that RNase treatment induced the loss of
the protective effect of MVs on functional and morphological
alterations induced by IRI in vivo and on hypoxiainduced
TEnC and TEpC injury in vitro. The significant
reduction of MV biological activities after treatment with
RNase suggests a putative horizontal transfer of RNAs
Kidney International (2012) 82, 412-427
V Cantaluppi et al.: EPC microvesicles and kidney ischemia–reperfusion injury original article
a 16 I
14 I
Q)
-~ 12 I *
·u; g_:,:1, 0 I
‘ Q)
~~ 8:
e-n ai 6′ 0 “
0 4!
z
2
0
Sham IRI IRI +MV IRI +MV IRl+MV IRl+MV IRl+MV
RNase siRNAdicer siRNA control AmiR126/296
b 16 *
14
-~ 12
·u; ?-~ 10
=> lE “O (fJ 8
a-i ‘ij; 6 0 “
0 4
z
2
0
Sham IRI IRl+MV IRI +MV IRI +MV IRl+MV IRl+MV
RNase siRNAdicer siRNA control AmiR126/296
C
d
Sham IRI IRI+ MV IRI + MV RNase
Figure 6 I Enhancement of cell proliferation induced by endothelial progenitor (EPC)-derived microvesicles (MVs) in ischemic
kidneys. Count (a, b) and representative micrographs (c, d) of BrdU-positive (a, c) or PCNA-positive (b, d) cells in different experimental
conditions. MVs induced a significant increase in BrdU- and PCNA-positive cells (*P<0.05 ischemia–reperfusion injury (IRI) + MV vs. IRI). The
pre-treatment of MVs with 1 U/ml RNase or the use of MVs released from EPCs transfected with small interfering RNA (siRNA) Dicer or with
antagomiRs-126/296 (AmiRl 26/296) significantly reduced the number of proliferating cells (‘P < 0.05 IRI + MV RNase, IRI + MV siRNA DICER
or IRI + MV AmiRl 26/296 vs. IRI + MV). MVs released from EPCs transfected with an irrelevant control siRNA (siRNA control) significantly
enhanced the number of BrdU and PCNA-positive cells (‘P<0.05 IRI + MV siRNA control vs. IRI). All sections were counterstained with
hematoxylin; original magnification x 100. BrdU, bromo deoxy uridine; PCNA, proliferating cell nuclear antigen.
from MVs to injured renal cells. It is known that MVs
protect RNAs from physiological concentrations of RNase.
However, as seen in previous studies, 18•20 •32 •33 the treatment
of MVs with high concentrations of RNase inactivates the
RNAs.
We previously demonstrated that MVs released from
EPCs shuttle mRNAs involved in angiogenic pathways such
as eNOS and Akt. 18 We now identified in £PC-derived MVs
several miRNAs typical of hematopoietic stem cells and of
Kidney International (2012) 82, 412-427
the endothelium, which are associated with cell proliferation,
angiogenesis, and inhibition of apoptosis. 43 .44 In
particular, MVs carried the angiomiRs miR-126 and miR-
296.45 The role of miRNAs shuttled by MVs in renal cell
regeneration in viva and in vitro was confirmed by
experiments with MVs derived from EPCs previously
subjected to the knock-down of Dicer, the intracellular
enzyme essential for miRNA production. 46 ’47 These results
suggest that miRNAs shuttled by MVs contribute to
419
original article V Cantaluppi et al.: EPC microvesicles and kidney ischemia;eperfusion injury
a 25
“O
al *
~ 20
al
(.) .’2 15
§
Cl. 10 0
Cl. “‘ 0 5
0 z
0
Sham IRI IRI+ MV IRl+MV IRl+MV IRI+ MV IRI +MV
RNase siRNAdicer siRNA control AmiR126/296
b
Sham IRI IRI +MV IRI + MV RNase
C d
en 40
- en 50
Q) >- 35 .El 45
(.) >,
0 (.) 40
“S 30 0
C C 0 35
e> 25 E 30 Cl
Cl
Cl 20 .!: 25 C
~ 15 ~ 20
~ 15
~ 10
0 10
0 5 0 5
z0 z 0 0
*
Sham IRI IRI+ MV IRI + Sham IRI IRI + MV IRI +
MV RNase MV RNase
Figure 71 Decrease in tubular cell apoptosis and leukocyte infiltration induced by endothelial progenitor (EPC)-derived
microvesicles (MVs) in ischemic kidneys. (a, b) Count (a) and representative micrographs (b) of TdT-mediated dUTP nick end labeling
(TUNEL)-positive cells in different experimental conditions. A significant increase in TUNEL-positive cells was observed in ischemiareperfusion
injury (IRI) in comparison with sham-treated animals (P<0.05 IRI vs. sham). MVs induced a significant decrease in apoptotic cells (#P< 0.05 IRI + MV vs. IRI). The pre-treatment of MVs with 1 U/ml RNase or the use of MVs released from EPCs transfected with small interfering RNA (siRNA) Dicer or with antagomiRs-126/296 (AmiRl 26/296) significantly reduced their anti-apoptotic effect (tp < 0.05 IRI + MV RNase, IRI + MV siRNA DICER or IRI + MV AmiRl 26/296 vs. IRI + MV). MVs released from EPCs transfected with an irrelevant control siRNA (siRNA control) significantly reduced the number of TUNEL-positive cells (#P<0.05 IRI + MV siRNA control vs. IRI). All sections were counterstained with hematoxylin; original magnification: x 100. (c, d) Counts of infiltrating granulocytes (cl and monocytes (d) in different experimental conditions. IRI induced an enhancement of granulocyte and monocyte infiltration in the kidney (P < 0.05 IRI vs. sham), which was not observed in MV-treated animals (#P < 0.05 IRI + MV vs. IRI). By contrast, RNase pre-treatment inhibited
the decrease in granulocyte and monocyte infiltration induced by MVs (tP<0.05 IRI + MV RNase vs. IRI MV).
their regenerative potential. Moreover, miR-126 and
miR-296 were identified to have a key role in MY-associated
renoprotective effects, as MVs derived from EPCs
transfected with specific antagomiRs anti-miR-126 and antimiR-
296 were less effective.
In conclusion, MVs released from EPCs exert a RNAmediated
protective effect in experimental acute renal IRI
overcoming the cross-species barrier. The protective effect of
MVs released from EPCs in hypoxic tissues may find
therapeutic application in AKI, CKD, vascular diseases, and
IRI after solid organ transplantation without the potential
risks of stem cell therapy such as maldifferentiation and
tumorigenesis.
420
MATERIALS AND METHODS
Isolation and characterization of EPCs and MVs derived from
EPCs and fibroblasts
EPCs were isolated from peripheral blood mononuclear cells of
healthy donors by density centrifugation and characterized as
previously described. 9•18 EPCs from 3-5 passages were used in order
to avoid monocyte and platelet contamination. EPCs expressed the
CD34 stem cell marker and markers of endothelial cells such as
CD31, KDR, CD105, and von Willebrand factor, which was detected
by FACS and western blot analysis. Moreover, EPCs were able to
uptake acetylated low-density lipoprotein. 9 In selected experiments,
EPCs were engineered to knock-down Dicer by specific small
interfering RNA (siRNA) (Santa Cruz Biotechnology, Santa Cruz,
CA) or transfected with anti-miR-126 and miR-296 antagomiRs
Kidney lnternatianal (2012) 82, 412-427
V Cantaluppi et al.: EPC microvesicles and kidney ischemia–reperfusion injury original article
a
Q)
C
1.2
~co — 0.8
~~ 0.6
2E .SO •4
~ 0.2
0
C
<t: -~ 120
U [!! 100
w Q)
cc c 80
~ -~ 60
~ ~ 40
‘c: ~ 20
Q):::,
c.. 0 0 -=
e
<{ ~100
·;;; 90
~ C 80
cc 2 70
~ .e: 60
~ t ~g
., 0 30
E lfl 20
.2 :i 10
(!Jg 0
SHAM
SHAM IRI
SHAM IRI
b SHAM IRI IRl+MV
IRI IRI +MV
d
IRl+MV
f
IRl+MV
Figure 8 I Long-term preservation of renal function and inhibition of glomerulosclerosis, tubulo-interstitial fibrosis, and capillary
rarefaction induced by endothelial progenitor (EPC)-derived microvesicles (MVs). (a) Evaluation of serum creatinine 180 days after
ischemia–reperfusion injury (IRI) in different experimental groups. Rats treated with EPC-derived MVs showed lower serum creatinine levels
than those observed in IRI animals (P<0.05 IRI vs. sham; #P<0.05 IRI + MV vs. IRI). (b) Representative Masson’s trichrome (upper panels) and hematoxylin/eosin (lower panels) staining of kidney sections of rats killed at 180 days after IRI in different experimental groups. Original magnification: x 100 in the upper panel, x 200 in the lower panel. (c-f) Mean fluorescence intensity (c, e) and representative confocal microscopy micrographs showing the staining for rat endothelial cell antigen-1 (RECA-1) antigen (d, f) in tubulo-interstitial structures (c, d) and within the glomeruli (e, f) in kidney sections of rats killed at 180 days after IRI in different experimental groups. Original magnification: x 200 in d and f: nuclei were counterstained with 2.5 μg/ml Hoechst (P<0.05 IRI vs. sham; #P<0.05 IRI + MV vs. IRI).
(Ambion, Austin, TX). Western blot analysis for Dicer expression
was performed by using an anti-Dicer polyclonal antibody (Abeam,
Cambridge, UK). MVs from EPCs and control human fibroblasts
were obtained from supernatants by ultracentrifugation as
previously described. 18’32 MV shape and size were evaluated by
transmission electron microscopy and by Nanosight technology
(Nanosight, London, UK). Antigen expression on MVs was studied
by FACS using antibodies directed to CD14, CD34, CD42b,
L-selectin, P-selectin, CD154 (Dako, Copenhagen, Denmark), cx4
integrin (Becton Dickinson, San Jose, CA), cxv~3 integrin, cx6
integrin (BioLegend, San Diego, CA), and human leukocyte antigen
class I and II (Santa Cruz Biotechnology). RNA extraction from
MVs was performed using the mirVana isolation kit (Ambion).
RNA was analyzed using the Agilent 2100 bioanalyzer (Agilent Tech,
Santa Clara, CA). miRNA expression levels were analyzed using
the Applied Biosystems TaqMan MicroRNA Assay Human Panel
Early Access kit (Applied Biosystems, Foster City, CA) to profile 365
miRNAs by qRT-PCR (E-MEXP-2956, European Bioinformatics
Institute: www.ebi.ac.uk/arrayexpress/). All reactions were per-
Kidney International (2012) 82, 412-427
formed using an Applied Biosystems 7900HT real-time PCR
instrument equipped with a 384-well reaction plate ( detailed
protocol reported in Supplementary Information). miRNA expression
levels were analyzed by qRT-PCR in a StepOne Real Time
System (Ambion): 200 ng of RNA was reverse-transcribed and the
complementary DNA was used to detect and quantify specific
miRNAs within EPCs, fibroblasts, and MVs derived from both cell
types by qRT-PCR using the miScript SYBR Green PCR Kit (Qiagen,
Valencia, CA). In selected experiments, MVs were labeled with the
red fluorescent dye PKH26 (Sigma-Aldrich, St Louis, MO) or
treated with I U/ml RNase (Ambion) and then blocked with 10 U/ml
RNase inhibitor (Ambion). 18
TEnC and TEpC cultures
Primary TEpCs were isolated and characterized as previously
described. 48.49 Primary TEnCs were obtained by using filters with
different meshes to discard the glomeruli. Isolated cells were
cultured on gelatine-coated flasks with endothelial growth factors
(Lonza, Basel, Switzerland) 18: after three passages in culture, cells
421
original article V Cantaluppi et al.: EPC microvesicles and kidney ischemia;eperfusion injury
a MV PKH26 RECA Merge b f
C d e
h Hypoxia Hypoxia
Normoxia Hypoxia a V ~ 3 Normoxia Hypoxia a V ~ 3
100 100 100 50 50 50
80 80 80 40 40 55.26% 73.27% 65.08% 57.15% 83.75% 40
22 60 22 60 22 60 ~ 30 ~ 30 ~ 30 74.74% C C C ::, ::, ::, ::, ::, ::,
0 40. M1 0 40 M1 0 40 M1 8 20 M1 8 20 8 20 (.) (.) (.)
20 20 20 10 10 10
0 o. 0 0 0 0
10° 101 1o” 103 104 10° 101 102 103 104 10° 101 102 103 104 10° 101 1o” 103 104 10° 101 1o” 103 104 10° 101 102 .103 104
Hypoxia Hypoxia Hypoxia Hypoxia Hypoxia Hypoxia
100 ~1 100
L-selectin
100
a4 50 ~1 L-selectin
50
a4
80 80 80 40 40
22
64.23%
22
49.43%
22
68.15%
~ 30
72.26% 22.36%
~ 30
77.86%
C 60 C 60. C 60
::, ::, ::, ::, 0 40 M1 0 40 M1 0 40 M1 8 20 M1 152 0 (.) (.) (.) (.)
20 20: 20 10 10
0 0 0 0 0
10° 101 102 103 104 10° 101 102 103 104 10° 101 102 103 104 10° 101 102 103 104 10° 101 1o” 103 104 10° 101 102 103 104
TEnCs TEpCs
Figure 9 I In vivo localization and in vitro internalization of endothelial progenitor (EPC)-derived microvesicles (MVs) in
isolated human tubular endothelial cells (TEnCs) and tubular epithelial cells (TEpCs). (a, b) Confocal microscopy analysis of PKH26-
labeled MV localization in endothelial cells (green staining for rat endothelial cell antigen-1 (RECA-1); arrows) of large vessels and
peritubular capillaries 2 h after injection. (c) Confocal microscopy analysis of PKH26-labeled MV localization in tubular epithelial cells
(green staining for laminin). (d, e) Representative micrograph of PKH26-labeled MVs in kidney glomeruli (d) and liver (e) of sham-operated
animals. In merge images, nuclei were counterstained with 2.5 μg/ml Hoechst. Original magnification: x 100 in a, b, c, and d and x 200 in e.
(f, g) Confocal microscopy analysis of PKH26-labeled MVs in TEnCs (f) and TEpCs (g). Nuclei were counterstained with 2.5 μg/ml Hoechst.
Original magnification x 400. {h, i) Representative FACS analysis of PKH26-labeled MV internalization in TEnCs (h) and TEpCs (i) cultured in
normoxia or hypoxia in the presence or absence of different blocking monoclonal antibodies. Hypoxia enhanced MV internalization in
TEnCs and TEpCs (P<0.05 normoxia vs. hypoxia). Anti-L-selectin mAb significantly decreased MV internalization in both cell types (P<0.05
hypoxia+ L-selectin mAb vs. hypoxia). Three different experiments were conducted with similar results. Kolmogorov-Smirnov statistical
analysis was performed on FACS data.
were further separated by magnetic cell sorting using an anti-CD3 l
antibody coupled to magnetic beads (MACS system, Miltenyi
Biotec, Auburn, CA) and characterized for endothelial markers
(CD31, CDlOS, and vonWillebrand factor).
Cell culture in hypoxic environment
TEnCs and TEpCs were cultured for 24 h into an airtight humidified
chamber flushed with a gas mixture containing 5% COz, 94% Ni,
and 2% 0 2 at 20 atm, 37 °C.
422
Kidney IRI model
The experimental protocol is given in detail in Figure 4. Male Wistar
rats (250 g body weight) were anesthetized by using an induction
chamber with isoflurane and by intraperitoneal administration of
ketamine (100 mg/kg). A subcutaneous injection of 1-2 ml normal
saline was administered to replace fluid loss during the surgical
procedure. After midline abdominal incision, the right kidney was
removed by a sub-capsular technique. Left renal artery and vein were
then occluded by using a non-traumatic vascular clamp that was
Kidney lnternatianal (2012) 82, 412-427
V Cantaluppi et al.: EPC microvesicles and kidney ischemia–reperfusion injury original article
Normoxia Hypoxia Normoxia Hypoxia
200 200 200 200
160
48.43% I
160 71.60% 160 160
(fJ 120 J!J 120 I (fJ 120 52.45% J!J 120 78.28%
EPC MVs c C Ml c C
::, Ml ::, EPC MVs ::, ::,
0 80 0 80 0 80 Ml 0 (.) (.) (.) (.) 80 Ml
40 40 40 40
0 0 0 0
1 oo 10 1 102 103 104 1 o0 10 1 10 2 10 3 10 4 1 o0 10 1 10 2 10 3 10 4 100 10 1 102 103 104
PE PE PE PE
200 200 200 200
160 5.97% 160 160 160
J!J J!J 7.51% J!J J!J
C 120 C 120
FIBRO MVs §
120 12.75% C 120 14.30%
FIBRO MVs ::, Ml ::, Ml ::,
0 80 0 80 0 80 Ml 0 80 (.) (.) (.) (.) M1
40 40 40 40
0 0 0 0
100 10 1 102 103 104 100 10 1 102 103 104 100 10 1 102 103 104 100 10 1 102 103 104
PE PE PE PE
TEnCs TEpCs
Figure 10 I Comparison of in vitro internalization of endothelial progenitor (EPC)-and fibroblast-derived microvesicles (MVs) in
tubular endothelial cells (TEnCs) and tubular epithelial cells (TEpCs) cultured in normoxia or hypoxia. Representative fluorescenceactivated
cell sorting (FACS) analysis of PKH26-labeled MVs derived from EPCs or fibroblasts (FIBRO) internalized in TEnCs and TEpCs
cultured in normoxia or hypoxia. Three different experiments were conducted with similar results. Kolmogorov-Smirnov statistical analysis
was performed.
applied across the hilum of the kidney for 45 min. Animals were
divided in the following groups: (1) normal (untreated); (2) shamoperated
(right nephrectomy); (3) IRI (right nephrectomy + left
renal pedicle clamp); (4) IRI+EPC MVs (right nephrectomy+left
renal pedicle clamp+ intravenous (i.v.) injection of 30 μg EPC
MVs); (5) IRI + RNase EPC MVs (right nephrectomy + left renal
pedicle clamp+ i.v. injection of 30 μg EPC MVs pre-treated with
1 U/ml RNase); (6) IRI+siRNA Dicer EPC MVs (right nephrectomy+
left renal pedicle clamp+i.v. injection of 30μg MVs derived
from £PCs engineered to knock-down Dicer by siRNA); (7)
IRI + siRNA Control EPC MVs (right nephrectomy + left renal
pedicle clamp+ i.v. injection of 30 μg MVs engineered with an
irrelevant siRNA); (8) IRI + AntagomiR-126/296 EPC MVs (right
nephrectomy + left renal pedicle clamp+ i.v. injection of 30 μg MVs
derived from £PCs transfected with anti-miR-126 and anti-miR-296
antagomiRs); and (9) IRI + fibroblast MVs (right nephrectomy + left
renal pedicle clamp+ i.v. injection of 30 μg MVs derived from
cultured fibroblasts). For all groups, MVs were diluted in 0.9% saline
and injected in the tail vein immediately after IRI. Six animals from
each group were killed at day 2, day 7, and day 180 (only groups l to
4). Kidneys were removed for histology and immunohistochemistry.
For renal histology, 5-μm-thick paraffin kidney sections were
routinely stained with hematoxylin/eosin or Masson’s trichrome
(Merck, Darmstadt, Germany). Luminal hyaline casts and cell loss
(denudation of tubular basement membrane) were assessed in nonoverlapping
fields ( up to 28 for each section) using a x 40 objective
(high-power field) to evaluate the score of AKI. The number of casts
and tubular profiles showing necrosis were recorded in a single-blind
manner. 33 Proliferation was evaluated in rats injected with bromo
deoxy uridine by using anti-bromo deoxy uridine (Dako) or antiproliferating
cell nuclear antigen (Santa Cruz Biotechnology)
monoclonal antibodies. 33 TdT-mediated dUTP nick end labeling
assay (Chemicon International, Temecula, CA) for the detection of
apoptotic cells was performed according to manufacturer’s instructions.
Leukocyte infiltration was evaluated by staining with antimonocyte
(Chemicon International) or anti-granulocyte (Serotec,
Oxford, UK) antibody. Immunoperoxidase staining was performed by
using an anti-mouse HRP (Pierce, Rockford, IL). Confocal micro-
Kidney International (2012) 82, 412-427
scopy analysis was performed on frozen sections for localization of
PKH26-labeled MVs within kidneys after staining with an antilaminin
(Sigma-Aldrich) or anti-rat endothelial cell antigen-1 antibody
(Serotec).
Blood samples for measurement of serum creatinine and
BUN were collected before and 2, 7, or 180 days after IRI. Creatinine
concentrations were determined using a Beckman Creatinine
Analyzer II (Beckman Instruments, Fullerton, CA). BUN was
assessed in heparinized blood using a Beckman Synchrotron CX9
automated chemistry analyzer (Beckman Instruments).
In vitro internalization of MVs into renal cells
TEnCs and TEpCs were seeded on six-well plates in normoxic or
hypoxic culture conditions and incubated with PKH26-labeled
MVs derived from £PCs or fibroblasts. MV internalization was
evaluated by confocal microscopy (Zeiss LSM 5 PASCAL, Jena,
Germany) and FACS in the presence or absence of 1 μg/ml blocking
antibodies directed to avP3-integrin (BioLegend), a4-integrin,
a5-integrin (Chemicon International), CD29, or L-selectin (Becton
Dickinson).
In vitro assays on TEnCs and TEpCs
Angiogenesis: Formation of capillary-like structures was studied on
TEnCs (5 x 104) seeded for 6 h on Matrigel and observed under an
inverted microscope. 50- 52 Apoptosis: TEnCs or TEpCs were subjected
to TdT-mediated dUTP nick end labeling assay (Chemicon
International). Samples were analyzed under a fluorescence microscope,
and green-stained apoptotic cells were counted in 10 nonconsecutive
microscopic fields.50 The activities of caspase-3, -8, and
-9 were assessed by enzyme-linked immunosorbent assay (Chemicon
International) based on the spectrophotometric detection of the
cromophore p-nitroanilide after cleavage from the labeled substrate
Asp-Glu-Val-Asp-p-nitroanilide, which is recognized by caspases.
Cell lysates were diluted with an appropriate reaction buffer, and
Asp-Glu-Val-Asp-p-nitroanilide was added at a final concentration
of 50 mol/1. Samples were analyzed in an automatized enzymelinked
immunosorbent assay reader at a wavelength of 405 nm. Each
experiment was performed in triplicate. 49- 50
423
0 rig in a I a rt i C I e V Cantaluppi et al.: EPC microvesicles and kidney ischemia;eperfusion injury
C 4
3
()
()
r
I)
a 25
b
“O
ai 20
~
Qi
-~ 15
0
§- 10
a.
“‘ 0
0 z
5
30
., 25
~
1 (/) 20
~ ~
g–§ 15
(.) 2
o iii 10
0 z
()
X
()
[;;
5
0
;=
CD
s:
s:
-u
I)
#
z
JJ
-u
-u
m
()
}>
s:
-u
~
C
□ Hypoxia
■ H poxia + MV
(J)
-u
I
:”.::
< m
G)
JI
G”)
J”J
w
Figure 11 I Effect of endothelial progenitor (EPC)-derived microvesicles (MVs) on apoptosis, angiogenesis, and mRNA expression
profile of tubular endothelial cells (TEnCs) cultured in hypoxic conditions. (a) TdT-mediated dUTP nick end labeling (TUNEL) assay
of TEnCs cultured in different experimental conditions. With respect to normal culture (Control), hypoxia induced a significant
increase in TEnC apoptosis (P<0.05 hypoxia vs. control). MVs significantly decreased hypoxia-induced TEnC apoptosis (#P<0.05 hypoxia+ MV vs. hypoxia). By contrast, preincubation of MVs with 1 U/ml RNase or the use of MVs released from EPCs transfected with small interfering RNA (siRNA) Dicer or with antagomiRs-126/296 (AmiRl 26/296) significantly inhibited the anti-apoptotic effect of MVs (tP<0.05 hypoxia+ MV RNase, hypoxia+ MV siRNA Dicer or hypoxia+ MV AmiRl 26/296 vs. hypoxia+ MV). MVs released from EPCs transfected with an irrelevant control siRNA (siRNA control) significantly reduced the number of apoptotic cells (#P<0.05 hypoxia+ MV siRNA control vs. hypoxia). Results are given as mean± s.d. of green-stained apoptotic cells in 10 microscopic fields (magnification x 100) of five independent experiments. (b) In vitro angiogenesis assay of TEnCs cultured on Matrigel-coated plates in different experimental conditions. With respect to normal culture (control), hypoxia induced a significant decrease in TEnC angiogenesis (P<0.05 hypoxia vs.
control). MVs enhanced angiogenesis of hypoxic TEnCs (#P<0.05 hypoxia+ MV vs. hypoxia). By contrast, preincubation of MVs with
1 U/ml RNase or the use of MVs released from EPCs transfected with siRNA Dicer or with antagomiRs-126/296 (AmiR126/296) significantly
inhibited the pro-angiogenic effect of MVs (tP< 0.05 hypoxia+ MV RNase, hypoxia+ MV siRNA Dicer or hypoxia+ MV AmiRl 26/296 vs.
hypoxia+ MV). MVs released from EPCs transfected with an irrelevant control siRNA (siRNA control) significantly increased TenC
angiogenesis (#P< 0.05 hypoxia+ MV siRNA control vs. hypoxia). Results are given as mean± s.d. of 20 different microscopic fields
(magnification x 100). Three independent experiments were conducted with similar results. (c) Gene array profiling of TEnCs cultured
in different experimental conditions (angiogenesis-related genes). The graph shows the fold variation of angiogenesis-related genes
between TEnCs cultured in hypoxia in the absence (white columns) or presence (black columns) of MVs in comparison with TEnCs cultured
in normoxic conditions. Samples were normalized for the signals found in housekeeping genes (actin, GAPDH). Three independent
experiments were conducted with similar results. Gene table: CCL2, chemokine (C-C motif) ligand 2; CXCL5, C-X-C motif chemokine 5;
FGFR3, fibroblast growth factor receptor 3; IL8, interleukin-8; MMP2, matrix metalloproteinase-2; NRPl, neuropilin-1; PECAMl, platelet
endothelial cell adhesion molecule (CD31); PLAU, urokinase-type plasminogen activator; SPHKl, sphingosine kinase 1; TYMP, thymidine
phosphorylase; VEGFa, vascular endothelial growth factor A.
424 Kidney lnternatianal (2012) 82, 412-427
V Cantaluppi et al.: EPC microvesicles and kidney ischemia–reperfusion injury original article
a b 60 ■ Caspase-3
-2!-50
~ (.) 40 “‘
#
□ Caspase-8
□ Caspase-9
(J)
“O -10.00
0 u..
-15.00 0 …
-20.00
-25.00
d
15.00
10.00
C 5.00
0 -~ 0.00
·c: -5.00 “> ‘
“O -10.00 ;r -15.00 )> (JJ (JJ ~ z 0 0 0 ~ ~ ~ “U ~ –t 0 –t –t –t –t
~ )> )> ;,;; 0 0 0 :IJ -< “U z 0 z :IJ :IJ :IJ -20.00 ~ X 0 … … )> 0 (J) 0 ;,;; “TI I) “TI )> )> )>
3′ 0 0 0 r )> ” (J) 0 “TI “TI r 0 G) I) I) -25.00 G) 0 :IJ 3′ 0 “‘
0 0
-30.00
Figure 12 I Effect of endothelial progenitor (EPC)-derived microvesicles (MVs) on apoptosis and mRNA expression profile of
tubular epithelial cells (TEpCs) cultured in hypoxic conditions. (a, b) TdT-mediated dUTP nick end labeling (TUNEL) assay (a) and
enzyme-linked immunosorbent assay (ELISA) for caspase-3, -8, and -9 activities (b) of TEpCs cultured in different experimental conditions.
Hypoxia induced a significant increase in TEpC apoptosis (P<0.05 hypoxia vs. control). MVs significantly decreased hypoxia-induced TEpC apoptosis (# P < 0.05 hypoxia + MV vs. hypoxia). By contrast, preincubation of MVs with 1 U/ml RNase or the use of MVs released from EPCs transfected with small interfering RNA (siRNA) Dicer or with antagomiRs-126/296 (AmiRl 26/296) significantly inhibited the anti-apoptotic effect of MVs (tP<0.05 hypoxia+ MV RNase, hypoxia+ MV siRNA Dicer or hypoxia+ MV AmiRl 26/296 vs. hypoxia+ MV). MVs released from EPCs transfected with an irrelevant control siRNA (siRNA Control) significantly reduced the number of apoptotic cells (#P < 0.05 hypoxia+ MV siRNA control vs. hypoxia). Results are given as mean± s.d. of green-stained apoptotic cells in 10 microscopic fields (magnification x 100) of five independent experiments. Similar results were observed for caspase activities (P < 0.05 hypoxia+ MV
or hypoxia+ MV siRNA control vs. hypoxia; #P<0.05 hypoxia+ MV RNase or hypoxia+ MV siRNA Dicer vs. hypoxia+ MV). Results are
given as mean± s.d. of five independent experiments and expressed as percentage increase in caspase activity with respect to normal
culture conditions. (c, d) Gene array profiling of TEpCs cultured in different experimental conditions (apoptosis-related genes). The graph
shows the fold variation of apoptosis-related genes between TEpCs cultured in hypoxia in the absence (black columns) or presence
(white columns) of MVs in comparison with TEpCs cultured in normoxic conditions. Samples were normalized for the signals found in
housekeeping genes (actin, GAPDH). Three independent experiments were conducted with similar results. Gene table in c: CASPl, caspase-
1; CASP10, caspase-10; CASP14, caspase-14; CASP3, caspase-3; CASP4, caspase-4; CASP5, caspase-5; CASP6, caspase-6; CASP7, caspase-7;
CASP8, caspase-8; CASP9, caspase-9. Gene table in d: APAFl, apoptotic protease-activating factor 1; BAKl, BCL2-antagonist/killer 1;
BAX, BCL2-associated X protein; BIK, Bcl-2-interacting killer; NODl, nucleotide-binding oligomerization domain-containing protein 1;
CD27, CD 27; CD40, CD 40; CD40LG, CD40 ligand (CDl 54); CRADD, death domain (CARD/DD)-containing protein; FADD, Fas-associated
protein with death domain; FASLG, Fas ligand (TNF superfamily, member 6); LTA, lymphotoxin alpha; LTBR, lymphotoxin beta receptor
(TNFR superfamily, member 3); PYCARD, apoptosis-associated speck-like protein containing a CARD or ASC; RIPK2, receptor-interacting
serine/threonine-protein kinase 2; TNF, tumor necrosis factor; TNFSF10, TNF-related apoptosis-inducing ligand (TRAIL); TRADD, tumor
necrosis factor receptor type 1-associated DEATH domain protein; TRAF2, TNF receptor-associated factor 2; TRAF3, TNF receptor-associated
factor 3; TRAF4, TNF receptor-associated factor 4.
Kidney International (2012) 82, 412-427 425
original article V Cantaluppi et al.: EPC microvesicles and kidney ischemia;eperfusion injury
Gene array analysis
The human GEarray kit for the study of angiogenesis in TEnCs and
apoptosis in TEpCs (SuperArray, Bethesda, MD) was used to
characterize the gene expression profile of cells cultured in normoxia
or hypoxia in the presence or absence of MVs. Microarray data
archive: (E-MEXP-2972 for TEnC angiogenesis and E-MEXP-3086
for TEpC apoptosis, European Bioinformatics Institute: www.ebi.
ac. uk/arrayexpress/ ).
Statistical analysis
All data of different experimental procedures are expressed as
average± s.d. Statistical analysis was performed by Kruskal-Wallis
statistical test for in viva studies and by Student’s t-test or analysis of
variance with Newmann-Keuls or Dunnet’s multicomparison test
where appropriate for in vitro experiments. For FACS data, the
Kolmogorov-Smirnov nonparametric statistical test was performed.
DISCLOSURE
SG, SB, MCD, and GC received funding for research from Fresenius
Medical Care. CT (Fresenius Medical Care) is employed by a
commercial company and contributed to the study as researcher. VC,
MCD, SB, CT, and GC are named as inventors in related patents.
ACKNOWLEDGMENTS
This work was supported by Italian Government Miur PRIN project,
Regione Piemonte, Piattaforme Biotecnologiche PiSTEM project and
Converging Technologies NanolGT project, Ricerca Finalizzata and
Local University Grants (ex-60%).
SUPPLEMENTARY MATERIAL
Supplementary Information 1. Supplementary methods.
Table 51. miRNA array analysis in EPCs (A: pro-angiogenic; B: proliferative;
C: anti-apoptotic; D: stem cells).
Table 52. miRNA array analysis in EPC MVs (A: pro-angiogenic; B:
proliferative; C: anti-apoptotic; D: stem cells).
Supplementary material is linked to the online version of the paper at
http://www.nature.com/ki
REFERENCES
- Mehta RL, Pascual MT, Soroko 5 et al. Program to improve care in acute
renal disease. Spectrum of acute renal failure in the intensive care unit:
the PICARD experience. Kidney lnt 2004; 66: 1613-1621. - Hoste EA, Kellum JA, Katz NM et al. Epidemiology of acute kidney injury.
Contrib Nephral 2010; 165: 1-K - Chawla LS, Amdur RL, Amodeo 5 et al. The severity of acute kidney injury
predicts progression to chronic kidney disease. Kidney lnt 2011; 79:
1361-1369. - Benigni A, Morigi M, Remuzzi Get al. Kidney regeneration. Lancet 2010;
375: 1310-1317 Review. - Herrera MB, Bussolati B, Bruno 5 et al. Mesenchymal stem cells contribute
to the renal repair of acute tubular epithelial injury. /nt J Mo/ Med 2004;
14: 1035-1041. - Morigi M, lntrona M, lmberti Bet al. Human bone marrow mesenchymal
stem cells accelerate recovery of acute renal injury and prolong survival
in mice. Stem Cells 2008; 26: 2075-2082. - Bussolati B, Bruno 5, Grange C. Isolation of renal progenitor cells from
adult human kidney. Am J Pathol 2005; 166: 545-555. - Asahara T, Murohara T, Sullivan A et al. Isolation of putative
progenitor endothelial cells for angiogenesis. Science 1997; 275:
964-967. - Biancone L, Cantaluppi V, Duo D et al. Role of L-selectin in the vascular
homing of peripheral blood-derived endothelial progenitor cells.
J lmmunol 2004; 173: 5268-5274. - Zampetaki A, Kirton JP, Xu Q. Vascular repair by endothelial progenitor
cells. Cardiovasc Res 2008; 78: 413-421. - Charwat 5, Gyongyosi M, Lang I et al. Role of adult bone marrow stem
cells in the repair of ischemic myocardium: current state of the art. Exp
Hemato/ 2008; 36: 672-680.
426 - Uchimura H, Marumo T, Takase O et al. lntrarenal injection of bone
marrow-derived angiogenic cells reduces endothelial injury and
mesangial cell activation in experimental glomerulonephritis. J Am Sac
Nephra/ 2005; 16: 997-1004. - Kwon 0, Miller 5, Li N et al. Bone marrow-derived endothelial progenitor
cells and endothelial cells may contribute to endothelial repair in the
kidney immediately after ischemia-reperfusion. J Histochem Cytochem
201 0; 58: 687-694. - Patschan D, Plotkin M, Goligorsky MS. Therapeutic use of stem and
endothelial progenitor cells in acute renal injury: ~a ira. Curr Opin
Pharmacol 2006; 6: 176-183. - Li B, Cohen A, Hudson TE et al. Mobilized human hematopoietic stem/
progenitor cells promote kidney repair after ischemia/reperfusion injury.
Circulation 2010; 121: 2211-2220. - Goligorsky MS, Yasuda K, Ratliff B. Dysfunctional endothelial
progenitor cells in chronic kidney disease. J Am Sac Nephrol 2010; 21:
911-919. - Yang Z, von Ballmoos MW, Faessler D et al. Paracrine factors
secreted by endothelial progenitor cells prevent oxidative stressinduced
apoptosis of mature endothelial cells. Atherasc/erasis 201 O; 211:
103-109. - Deregibus MC, Cantaluppi V, Calogero R et al. Endothelial progenitor cell
derived microvesicles activate an angiogenic program in endothelial
cells by a horizontal transfer of mRNA. Blood 2007; 110: 2440-2448. - Ratajczak J, Wysoczynski M, Hayek F et al. Membrane-derived
microvesicles: important and underappreciated mediators of cell-to-cell
communication. Leukemia 2006; 20: 1487-1495. - Ratajczak J, Miekus K, Kucia M et al. Embryonic stem cell-derived
microvesicles reprogram hematopoietic progenitors: evidence for
horizontal transfer of mRNA and protein delivery. Leukemia 2006; 20:
847-856. - De Broe ME, Wieme RJ, Logghe GN et al. Spontaneous shedding of
plasma membrane fragments by human cells in viva and in vitro.
Clin Chim Acta 1977; 81: 237-245. - Valadi H, Ekstrom K, Bossios A et al. Exosome-mediated transfer of mRNAs
and microRNAs is a novel mechanism of genetic exchange between cells.
Nat Cell Biol 2007; 9: 654-659. - lkarashi K, Li B, Suwa M et al. Bone marrow cells contribute to
regeneration of damaged glomerular endothelial cells. Kidney lnt 2005;
67: 1925-1933. - Daniel C, Amann K, Hohenstein B et al. Thrombospondin 2 functions as
an endogenous regulator of angiogenesis and inflammation in
experimental glomerulonephritis in mice. J Am Sac Nephrol 2007; 18:
788-798. - Becherucci F, Mazzinghi B, Ronconi E et al. The role of endothelial
progenitor cells in acute kidney injury. Blood Purif 2009; 27: 261-270. - Aharon A, Tamari T, Brenner B. Monocyte-derived microparticles and
exosomes induce procoagulant and apoptotic effects on endothelial
cells. Thromb Haemost 2008; 100: 878-885. - Prokopi M, Pula G, Mayr U et al. Proteomic analysis reveals presence
of platelet microparticles in endothelial progenitor cell cultures. Blood
2009; 114: 723-732. - Yoder MC, Mead LE, Prater D et al. Redefining endothelial progenitor cells
via clonal analysis and hematopoietic stem/progenitor cell principals.
Blood 2007; 109: 1801-1809. - De Broe ME, Wieme RJ, Logghe GN et al. Spontaneous shedding of
plasma membrane fragments by human cells in viva and in vitro.
Clin Chim Acta 1977; 81: 237-245. - Schorey JS, Bhatnagar 5. Exosome function: from tumor immunology to
pathogen biology. Traffic 2008; 9: 871-881. - Quesenberry PJ, Dooner MS, Aliotta JM. Stem cell plasticity revisited: the
continuum marrow model and phenotypic changes mediated by
microvesicles. Exp Hematol 201 O; 38: 581-592. - Bruno 5, Grange C, Deregibus MC et al. Mesenchymal stem cell-derived
microvesicles protect against acute tubular injury. J Am Sac Nephra/ 2009;
20: 1053-1067. - Gatti 5, Bruno 5, Deregibus MC et al. Microvesicles derived from human
adult mesenchymal stem cells protect against ischaemia-reperfusioninduced
acute and chronic kidney injury. Nephro/ Dial Transplant 2011;
26: 1474-1483. - Burne MJ, Rabb H. Pathophysiological contributions of
fucosyltransferases in renal ischemia reperfusion injury. J lmmunol 2002;
169: 2648-2652. - Rabb H, O’Meara YM, Maderna P et al. Leukocytes, cell adhesion
molecules and ischemic acute renal failure. Kidney lnt 1997; 51:
1463-1468.
Kidney International (2012) 82, 412-427
V Cantaluppi et al.: EPC microvesicles and kidney ischemia–reperfusion injury
- Molitoris BA, Sutton TA. Endothelial injury and dysfunction: role in 45.
the extension phase of acute renal failure. Kidney lnt 2004; 66:
496-499. 46. - Sutton TA, Fisher CJ, Molitoris BA. Microvascular endothelial injury and
dysfunction during ischemic acute renal failure. Kidney lnt 2002; 62: 47.
1539-1549. - Legrand M, Mik EG, Johannes T et al. Renal hypoxia and dysoxia after 48.
reperfusion of the ischemic kidney. Mo/ Med 2008; 14: 502-516. - Bonventre JV, Weinberg JM. Recent advances in the pathophysiology
of ischemic acute renal failure. J Am Soc Nephrol 2003; 14: 49.
2199-2210. - Bonventre JV. Dedifferentiation and proliferation of surviving epithelial
cells in acute renal failure. J Am Soc Nephrol 2003; 14: SSS-S61. 50. - Kaushal GP, Basnakian AG, Shah SV. Apoptotic pathways in ischemic
acute renal failure. Kidney lnt 2004; 66: 500-506. - Sheridan AM, Bonventre JV. Cell biology and molecular mechanisms of
injury in ischemic acute renal failure. Curr Opin Nephro/ Hypertens 2000; 9: 51.
427-434 Review. - Nelson P, Kiriakidou M, Sharma A et al. The microRNA world: small is
mighty. Trends Biochem Sci 2003; 28: 534-540. 52. - Camussi G, Deregibus MC, Bruno S et al. Exosomes/microvesicles as a
mechanism of cell-to-cell communication. Kidney lnt 2010; 78: 838-848.
Kidney International (2012) 82, 412-427
original article
Wang S, Olson EN. AngiomiRs-key regulators of angiogenesis. Curr Opin
Genet Dev 2009; 19: 205-211.
Jaskiewicz L, Filipowicz W. Role of dicer in posttranscriptional RNA
silencing. Curr Top Microbiol lmmunol 2008; 320: 77-1:)7_
Saal S, Harvey SJ. MicroRNAs and the kidney: coming of age. Curr Opin
Nephro/ Hypertens 2009; 18: 317-323.
Conaldi PG, Biancone L, Bottelli A et al. HIV-1 kills renal tubular epithelial
cells in vitro by triggering an apoptotic pathway involving caspase
activation and Fas upregulation. J Clin Invest 1998; 102: 2041-2049.
Cantaluppi V, Biancone L, Romanazzi GM et al. Macrophage stimulating
protein may promote tubular regeneration after acute injury. J Am Soc
Nephro/ 2008; 19: 1904-1918.
Cantaluppi V, Biancone L, Romanazzi GM et al. Antiangiogenic
and immunomodulatory effects of rapamycin on islet endothelium:
relevance for islet transplantation. Am J Transplant 2006; 6:
2601-2611.
Boccellino M, Biancone L, Cantaluppi V et al. Effect of platelet-activating
factor receptor expression on CHO cell motility. J Cell Physiol 2000; 183:
254-264.
Biancone L, Cantaluppi V, Segoloni G et al. Role of platelet-activating
factor in functional alterations induced by xenoreactive antibodies in
porcine endothelial cells. Transplantation 2000; 70: 1198-1205.
427