10.1161/CIRCULATIONAHA.116.023877
Distinct Regulatory Effects of Myeloid Cell and Endothelial Cell Nox2 on
Blood Pressure
Running Title: Sag et al.; Nox2 Dependent Regulation of Blood Pressure
Downloaded from http://circ.ahajournals.org/ by guest on June 16, 2017
Can Martin Sag, MD1,2*; Moritz Schnelle, MD, PhD1,3*; Juqian Zhang, MD1;
Colin E. Murdoch, PhD1; Sabine Kossmann, PhD4; Andrea Protti, PhD 1;
Celio X.C. Santos, PhD1; Greta Sawyer, PhD1; Xiaohong Zhang, MD1;
Heloise Mongue-Din, PhD1; Daniel A. Richards, PhD1; Alison C. Brewer, PhD1;
Oleksandra Prysyazhna, PhD1; Lars S. Maier, MD2; Philip Wenzel, MD4;
Philip J. Eaton, PhD1; Ajay M. Shah MD, FMedSci.1
1
King'ss College London British Heart Foundation Centre of Excellence, Cardio
King
Cardiovascular
ova
v scular
Division, London, United Kingdom; 2Klinik und Poliklinik für Innere Medizin
Med
ediz
ed
izin
iz
in II,
II,
Universitätsklinikum Regensburg, Regensburg, Germany; 3Department of Cardiology and
Pneumology, Medical Center Goettingen, Goettingen, Germany; 4Center for Cardiology and
Ce
Center
ent
nter forr Thrombosis
Thr
h om
ombosis and Hemostasis, Univer
University
rsiity Medical Cen
Center
enter Ma
M
Mainz,
inz, Mainz, Germany
*E
Equal C
onntrib
but
ution
*Equal
Contribution
Address for Correspondence:
Ajay M. Shah, MD, FMedSci
James Black Centre
King's College London BHF Centre of Excellence
125 Coldharbour Lane
London SE5 9NU, UK
Tel: 44-2078485189
Fax: 44-2078485193
Email: [email protected]
Journal Subject Terms: Animal Models of Human Disease; Endothelium/Vascular Type/Nitric
Oxide; Vascular Biology; High Blood Pressure
1
10.1161/CIRCULATIONAHA.116.023877
Abstract
Downloaded from http://circ.ahajournals.org/ by guest on June 16, 2017
Background—Hypertension due to increased renin angiotensin system (RAS) activation is
associated with elevated reactive oxygen species (ROS) production. Previous studies implicate
NADPH oxidase (Nox) proteins as important ROS sources during RAS activation, with different
Nox isoforms being potentially involved. Among these, Nox2 is expressed in multiple cell types
including endothelial cells, fibroblasts, immune cells and microglia. Blood pressure (BP) is
regulated at central nervous system, renal and vascular levels but the cell-specific role of Nox2 in
BP regulation is unknown.
Methods—We generated a novel mouse model with a Floxed Nox2 gene and used Tie2-Cre,
LysM Cre or Cdh5-CreERT2 driver lines to develop cell-specific models of Nox2 perturbation to
investigate its role in BP regulation.
Results—Unexpectedly, Nox2 deletion in myeloid but not endothelial cells resulted in a
significant reduction in basal BP. Tie2-CreNox2 knockout (KO) mice (in which Nox2 was
deficient in both endothelial cells and myeloid cells) and LysM Cre Nox2KO mice (in which
Nox2 was deficient in myeloid cells) both had significantly lower BP than littermate controls
whereas basal BP was unaltered in Cdh5-CreERT2 Nox2 KO mice (in which Nox2 is deficient
oonlyy in endothelial
e do e a cells).
ce s). Thee lower
owe BP was aattributable
bu ab e too an
a increased
c eased NO bbioavailability
oava ab y
which dynamically dilated resistance vessels in vivo under basal conditions, without
withoout change
chaang
ngee in
renal
enal function. Myeloid-specific Nox2 deletion had no effect on angiotensin II-ind
II-induced
duc
uced
ed
d
hypertension which, however, was blunted in Tie2-CreNox2KO mice along with preservation of
endothelium-dependent relaxation during angiotensin II stimulation.
Conclusions—We identify a hitherto unrecognized modulation of basal BP by myeloid cell
whereas
angiotensin
Nox22 w
h re
he
reas
as eendothelial
nddothelial cell Nox2 regulates ang
giotensin II-induced
II-induuce
c d hy
hypertension. These results
identify
Nox2
BP
den
nti
tify distinct
distincct cell-specific
ceell
l -sppeccif
ific
i roles
rol
oles
es for
fo No
N
x2 in
nB
P rregulation.
eg
gul
ulat
a io
at
on.
n
Key-Words:
NAD(P)H
oxidase;
mice
Key
y-W
-Words: NA
NAD(
D P))H oxid
idase; bloodd pressure;
id
pressuure; angiotensin
ang
nggio
iote
tennsin
te
in II; genetically
genetic
icallly
y altered
dm
icce
2
10.1161/CIRCULATIONAHA.116.023877
Clinical Perspective
What is new?
x
NADPH oxidases generate superoxide and are implicated in the pathogenesis of
hypertension.
x
The NADPH oxidase Nox2 isoform is expressed in several cell types (such as
endothelial cells and inflammatory cells) but exactly how it impacts on blood pressure
(BP) is unclear.
x
This study uses novel gene-modified mouse models to show that Nox2 in myeloid cells
modulates basal BP whereas endothelial cell Nox2 is involved in angiotensin II-
Downloaded from http://circ.ahajournals.org/ by guest on June 16, 2017
dependent hypertension.
x
This is the first demonstration that myeloid cell Nox2 regulates basal BP.
What are the clinical implications?
x
The finding that Nox2 in different cell types has distinct effects on BP suggests that
diff
di
ffer
ff
eren
er
e t di
disease conditions may alter BP through
th
o Nox2
Noox2
x in distinct cell types.
different
effectss of
x
It is conce
eiv
vab
blee tthat
hatt th
ha
thee ef
effe
fectts of mye
fe
yeloid
id cells
cel
ells
el
ls on
on basal
b saal BP
ba
B may
may
y be
be enhanced
enha
en
hanc
ha
n ed
d iin
n
conceivable
effects
myeloid
inflam
mma
m to
tory settings
settin
ngs where the
hese cel
llss are
re m
oree ac
or
acti
t vateed.
inflammatory
these
cells
more
activated.
x
On tthe
he oother
th
her
e hhand,
a d, eendothelial
an
ndot
nd
othe
ot
heli
he
lial
al ccell
elll No
el
Nox2
x aactivation
x2
c iv
ct
vat
atio
ionn ma
io
mayy be more
more
r rrelevant
elev
el
evan
ev
antt to rreninan
een
nin
ini
i system-dependent
d
d hhypertension.
i
angiotensin
x
The current results are relevant to the design of novel therapeutic approaches for
hypertension by targeting NADPH oxidases.
3
10.1161/CIRCULATIONAHA.116.023877
Introduction
The renin angiotensin system (RAS) plays a central role in blood pressure (BP) regulation and its
chronic activation contributes to hypertension. Both animal and human studies have implicated
increased reactive oxygen species (ROS) production in the pathophysiology of angiotensin II
(AngII)-dependent hypertension.1,2 ROS have complex cell-, source- and context-specific roles,
ranging from physio-pathologic redox signaling to inactivation of nitric oxide (NO) and
detrimental oxidation of cellular biomolecules.3,4 In experimental models, ROS scavengers can
attenuate the hypertensive response to AngII1 but randomized clinical trials of general
Downloaded from http://circ.ahajournals.org/ by guest on June 16, 2017
antioxidant approaches have failed to demonstrate reduction in cardiovascular morbidity and
mortality;5 such global inhibition approaches may affect beneficial redox signaling
ng aass we
well
ll aass
detrimental oxidative stress. A better understanding of the roles of ROS in BP regulation and
hypertension is therefore necessary in order to develop novel and more refined therapeutic
appr
approaches.
p oaches.
pr
oxidase
(Nox)
major
NADPH
H ox
xidaase (N
Nox)
oxx family
y pproteins
roteinss aree ma
ajo
jorr ssources
our
urces ooff ROS
O inn the cardiovascular
caard
r io
ovascular
ar
66,7
7
system
ystem and are important in redox signali
signaling.
l ng.6,
They
Th
hey contain a Nox catalytic subu
subunit
b nit th
that
hat
mediates ROS generation through electron transfer from NADPH to molecular oxygen. Five
different oxidases, each based on a distinct Nox catalytic subunit (Nox1-5), have been identified.
These have tissue-specific distribution and differing modes of activation, based on their varying
requirements for accessory subunits. Different Nox isoforms may have distinct roles even in the
same cell type, thought to be related to their coupling to different intracellular signaling
pathways and/or production of different ROS (superoxide versus hydrogen peroxide).6 This is
relevant from a therapeutic perspective because isoform-selective Nox inhibitors are currently
being developed.8 Previous work suggested an involvement of Nox1 in the genesis of AngII-
4
10.1161/CIRCULATIONAHA.116.023877
dependent hypertension. Mice globally deficient in p47phox, a subunit required for both Nox1 and
Nox2 function, display a reduced hypertensive response to AngII9 as do global Nox1 knockout
mice,10 whereas vascular smooth muscle-targeted Nox1 transgenic mice develop exaggerated
AngII-induced hypertension.11 The role of Nox2 is less clear. This isoform is expressed in
endothelial cells, fibroblasts, cardiomyocytes, inflammatory cells and microglia,6 sites that are of
interest given that BP is regulated at central nervous system (CNS), renal, vascular and cardiac
levels. Nox2 is involved in the genesis of endothelial dysfunction in diverse models.6,7
Interestingly, ROS production in the subfornical organ in the brain is implicated in the
Downloaded from http://circ.ahajournals.org/ by guest on June 16, 2017
vasopressor effects of AngII12 and the knockout of p22phox - a subunit required for Nox1, Nox2
and Nox4 function - at this site blunted AngII-induced hypertension.13 p47phox knoc
knockout
ocko
kout
ko
ut mice
mic
icee
have reduced renal afferent arteriolar constrictor responses to AngII14 and the hypertensive
effect of AngII has been shown to involve infiltration of the vasculature by p47phox-containing T
lymphocytes
ymp
mphhocytes155 an
mp
andd No
Nox2
ox2 ccompetent
om
mpe
pete
tent
te
n m
nt
monocytic
o ocytic
on
ic cel
cells.
ells.16 Wh
While
Whil
ilee th
il
thes
these
esee st
es
sstudies
udies su
ud
sugg
suggest
gges
gg
e t an
es
n
involvement
nvoolv
l ement off N
Nox
ox pproteins
rootein
ns in Ang
AngII-dependent
gII-d
-dependdent hhypertension,
yp
per
erte
teensio
ionn, the
they
hey doo not
ot directlyy es
establish
stablish th
the
he
role
ole of No
N
Nox2,
x22, in particular its potential cell-specifi
cell-specific
fic role. In this study, we have used a novel
mouse model with a Floxed Nox2 gene to study the role of endothelial and myeloid cell Nox2 in
BP regulation. Unexpectedly, we found that myeloid cell Nox2 has an essential role in the basal
regulation of BP whereas activation of endothelial Nox2 contributes to AngII-dependent
hypertension.
5
10.1161/CIRCULATIONAHA.116.023877
Methods
Generation of mice with a Floxed Nox2 allele and cell-specific knockouts
Animal studies complied with the UK Home Office Guidance on the Operation of the Animals
(Scientific Procedures) Act, 1986 and institutional guidelines. The generation of Nox2fl/fl mice
was commissioned from Genoway (France). The targeting construct was electroporated into
129sv embryonic stem cells. Recombinant clones were LGHQWL¿HGE\3&5DQG6RXWKHUQEORWting.
After successful targeting, the neomycin cassette was excised using flanking Flippase
Recognition Target sites. Clones were injected into C57BL/6 blastocysts. Heterozygous Floxed
Downloaded from http://circ.ahajournals.org/ by guest on June 16, 2017
mice obtained from germline chimeras were back-crossed >10 generations with C57BL/6 mice.
females
maale
less we
were
re
For generation of cell-specific knockout (KO) and littermate controls, Nox2fl/fl fema
crossed with male Tie2-Cre,17 LysM-Cre18 or Cdh5-CreERT219 transgenic mice. Inducible
deletion of Nox2 in the Cdh5-CreERT2 model was achieved by Tamoxifen treatment (40 mg/kg
i.p.
.p. for
for 3 conse
consecutive
ecutiive
v days).
day
ays)
s). Ad
s)
Adul
Adult
ultt male
ul
m lee mice aged
ma
age
g d 8-166 we
week
weeks
eks wi
ek
with
th
h ccell-specific
ellel
l spec
ecif
ec
ific
if
ic N
Nox2
ox22 de
dele
deletion
leti
le
tion
ti
o
were
re comparedd wi
w
with
th
hC
Cre-negative
re-neegative Floxx llittermates.
itterma
m tess.
ma
PCR
Confirmation of cell-specific Nox2 deletion by PCR was based on the amplification of a 225 bp
product that is only formed after excision of exons 1 and 2 of the cybb gene (forward:
GGAATTGAGTTGTAAGAATCAAATGAC, reverse: ATGATGTGTCCCAAATGTGC).
Primer GGGGCTGAATGTCTTCCTCT was included in the reaction to detect the 467 bp
wildtype Nox2 sequence.
Real-time RT-PCR with SYBR Green was used to quantify mRNA expression levels.
Delta delta Ct values were calculated using GAPDH as denominator. Primer sequences were
6
10.1161/CIRCULATIONAHA.116.023877
(forward, reverse): GAPDH: ATGACAACTTTGTCAAGCTCATTT,
GGTCCACCACCCTGTTGCT;
Nox2: ACTCCTTGGGTCAGCACTGG, GTTCCTGTCCAGTTGTCTTCG
p22phox: GCCCTCCACTTCCTGTT, GCAGATAGATCACACTGGCAAT;
p40phox: CTGCTTTTCTGACTACCCACAG, AAGCTGCTCAAAGTCGCTCT;
p47phox: GGACACCTTCATTCGCCATA, CTGCCACTTAACCAGGAACAT;
p67phox: TTGAACCTGTCACACAGCAAT, CCAGCACACACACAAACCTT;
superoxide dismutase-1 (SOD1): GGACCTCATTTTAATCCTCACTCTAAG,
Downloaded from http://circ.ahajournals.org/ by guest on June 16, 2017
GGTCTCCAACATGCCTCTCTTC;
SOD2: CACACATTAACGCGCAGATCA, GGTGGCGTTGAGATTGTTCA;
SOD3: ACACCTTAGTTAACCCAGAAATCTTTTC, GGGATGGATCTAGAGCATTAAGGA
Catalase: GCTGAGAAGCCTAAGAACGCAAT, CCCTTCGCAGCCATGTG.
Immunoblotting
mmu
munoblotti
mu
tiing
Snap-frozen
Snap
ap-frozen ao
ap
aortic
ort
rticc ttissue
issu
sue was
waas homog
homogenized
gen
nized for
foor imm
immunoblotting.
mmun
unob
un
oblo
ob
lott
ttin
ng. Primary
Primaary antibodies
anntibodiees were:
we No
Nox2
ox2
h
(1:1000;
1:1000;
0 BD
D Biosciences,
Biosciiences, UK), p2
p22
22phox
((1:1000;
1:100000; Santa
S nta Cruz, USA),
Sa
U A)
US
A , ȕ-actin ((1:4000;
1:40000; Si
Sigma,
UK), Nox4 (1:1000),20 eNOS (1:1000; BD Biosciences, UK). Actin (Sigma) was used as a
loading control. Protein bands were visualized using enhanced chemiluminescence and
quantified by densitometry.
Blood pressure measurement
BP telemeters (model TA11PA-C10, Data Sciences International, Netherlands) were implanted
subcutaneously under isoflurane anesthesia, with a 1-week recovery period prior to
measurements.21 Analyses were performed using Dataquest ART analysis software. AngII (1.1
mg/kg/day) was administered via subcutaneous osmotic minipumps (Model 1002, Alzet,
7
10.1161/CIRCULATIONAHA.116.023877
Cupertino, CA), implanted under 2% isoflurane.21 In some experiments, 1Ȧ-Nitro-L-arginine
methyl ester (L-NAME, Sigma-Aldrich, UK; 100 mg/kg/day) was administered in the drinking
water.
Magnetic resonance imaging (MRI)
MRI was performed in prone mice on a 7T horizontal scanner (Agilent Technologies, Varian
Inc., Palo Alto, CA) under isoflurane anesthesia.22 Body temperature was maintained at 37oC and
heart rates above 400 bpm. Temporally resolved dynamic short-axis images of the carotid
arteries were acquired with a cine-FLASH sequence using ECG and respiratory gating.
Downloaded from http://circ.ahajournals.org/ by guest on June 16, 2017
Endothelium-dependent relaxation in vivo was assessed using acetylcholine (18.8 mg/kg i.p.).22
Pixels encompassing the blood pool were clustered based on signal intensity and th
thee ve
vess
vessel
s el w
ss
wall
al
al
borders. Images were analyzed in end-systolic and end-diastolic phases using ClinicalVolumes
segmentation
egmentation software (King’s College London, UK; www.clinicalvolumes.com).
Other
Othe
her in vivo p
he
procedures
roc
oced
dur
ures
es
Echocardiography
Echo
hocardiograp
ho
aphy
ap
h w
hy
was
as perfo
performed
forrmed using
fo
ng a Vev
Vevo
vo 2100
000 System
Sys
y te
t m with
th a 40-MHz
400-MH
MHz linear
MH
a probe
ar
pro
robe
23
(Visualsonics,
Visualsonics, Canada),
) under 1.5%
% isofl
isoflurane
flurane anesthesia.23
Renall ffunction was assessed in
response to an acute saline challenge (40 ml/kg 0.9% w/vol. saline, i.p. injection). Animals were
placed in an individual metabolic chamber (Tecniplast 3600M021, UK) for 4 h, without access to
food or water, and urine was collected at hourly intervals.24 Metabolites were analyzed on an
Advia 2400 Chemistry System (Siemens AG, Germany).
Ex vivo vascular function
Isometric tension was quantified in descending thoracic aortic rings suspended in a Krebs buffer
solution containing indomethacin (3 μmol/L) at 37°C, pH 7.4.21 Endothelium-dependent
relaxation was assessed from the cumulative dose-response to acetylcholine (ACh) of rings pre-
8
10.1161/CIRCULATIONAHA.116.023877
constricted to 70% of the maximal contraction to phenylephrine. In some experiments, rings
were incubated with AngII (0.1 μmol/L, 4 h) in the presence or absence of the superoxide
scavenger MnTMPyP (10 μmol/L) prior to addition of other vasoactive agents. Basal NO
bioavailability was assessed from the response to a single dose of N-methyl-L-arginine (LNMMA, 100 μmol/L) in rings pre-constricted to 30% of the maximal phenylephrine response.
Vascular segments from mesenteric (second order) arteries were studied in a tension
myograph (Danish Myo Technology, Denmark) in Krebs buffer at 37 °C.25 Endotheliumdependent relaxation was assessed from the cumulative dose-response to ACh in rings preDownloaded from http://circ.ahajournals.org/ by guest on June 16, 2017
constricted with U46619 ȝPRO/6LJPD
ROS assays
HPLC-based detection of dihydroxyethidium (DHE) oxidation products was performed as
described pr
ppreviously.
eviously.26 Aortic segments were incubated with or without AngII (0.1 μmol/L, 3.5
h, 37
37 °C) before
re add
addition
dditio
dd
ionn of DHE
io
DHE ((100
1 0μ
10
μmol/L)
mol/L)
L) for
or 300 min
min at
a 37°C
37°
7°C
C in the
the
h dark.
dar
ark.
ar
k Tissue
k.
Tis
i suee was
was
harvested
acetonitrile,
sonicated
harvvested in ac
cet
etonnitriilee, so
oni
nicated and
d ccentrifuged.
entrifu
f ged.
fu
d Supernatants
Sup
uper
e na
er
nata
tantss were
weree ddried
riied underr vvacuum
acuuum aand
nd
pellets storedd at -80°C.
-880°
0 C.
C For analysis,
analysiis, samples were re-suspended in 120 μl PBS/DTPA
PBS
B /D
DTPA
A and
injected into the HPLC system. The superoxide specific 2-hydroxyethidium (EOH) signal was
normalized to tissue weight.
ROS production in bone marrow and blood mononuclear cells was assessed using flow
cytometry of cells loaded with DHE (10 μmol/L, 10 mins at 37 °C), after stimulation with
phorbol 12-myristate 13-acetate (PMA, 100 ng/ml) to activate the Nox2 oxidase complex.27
Bone marrow cells were harvested as described previously.28 Monocytes were isolated by Ficoll
gradient centrifugation and CD11b Microbeads (Miltenyi Biotech, Germany).
9
10.1161/CIRCULATIONAHA.116.023877
Vascular morphology
In vivo perfusion fixation with 4% PFA under pressure followed by paraffin embedding was
used.21 Vascular media thickness and intima-media area were quantified in 6 μm sections stained
with hematoxylin and eosin, using Volocity software (Volocity v5.0, Perkin Elmer, American
Fork, UT).
Coronary microvascular endothelial cells (CMEC) were isolated from mouse hearts as
described previously.29
Flow cytometry (FACS)
Downloaded from http://circ.ahajournals.org/ by guest on June 16, 2017
Quantitative analyses of leukocyte number and phenotype were performed by FACS on aortic
tissue
issue digests, using a FACS CantoII® instrument (BD Biosciences, Oxford, UK).
UK)). Briefly,
Briief
Br
efly
ly,,
ly
animals under terminal anesthesia were perfused with saline through the LV to eliminate
circulatingg blood. The blood free descending and abdominal aorta was digested in a mixture of
collagenase
coll
lag
agenase IV
IV,
V, DN
DNase
Nasse an
andd hy
hyal
hyaluronidase
alur
al
uron
ur
onid
on
das
ase at 337°C
7 C for
7°
for 30 m
min,
in, followed
in
folllow
fo
o ed
d by tr
trit
trituration
itur
it
urat
ur
atio
at
i n and
an
nd
filtration
filtra
rati
ra
t on through
throuugh a 700 μm
μm nylon
nylon mesh.. Cell
Cell suspensions
suuspen
ension
onss were
on
were
re wa
washed
ashhed an
and
nd bblocked
lockedd wi
with
h antiCD16/C
CD322 antibodies
antibo
b dies prior to staining. Monocytes (CD45+CD
CD11b
D111b+Ly6G-), macropha
macrophages
h ges
CD16/CD32
(CD45+CD11b+F4/80+), neutrophils (CD45+CD11b+Ly6G+), T-cells (CD45+7&5ȕ+) and B-cells
(CD45+CD19+) were identified. Zombie-Aqua dye (Biolegend) was used to identify dead cells
prior to fixation with 1% PFA. Fluorescence-minus-one (FMO) stained samples were used as
negative controls. Data analysis was performed with FlowJo software (Tree Star Inc., Ashland,
OR).
NO measurement by Electron Paramagnetic Resonance (EPR)
Aortic NO formation was measured as described previously30 by EPR-based spin trapping with
iron-diethyldithiocarbamate (Fe(DETC)2) colloid using a Miniscope MS400 table-top X-band
10
10.1161/CIRCULATIONAHA.116.023877
spectrometer (Magnettech, Berlin, Germany). In brief, aortae were either stimulated with
calcium ionophore for measurement of eNOS-derived NO levels or with LPS to detect iNOSderived NO formation. For LPS stimulation, freshly prepared aortae were cut into 3-mm rings
DQGLQFXEDWHGZLWKȝJPO/36LQ530,PHGLXP)&6SHQLFLOOLQ
streptomycin for 21 h at 37°C, 5% CO2. Afterwards rings were transferred in a 24-well plate
filled with 1 ml Krebs-HEPES-solution (pH 7.35, containing NaCl 99.01 mM, KCl 4.69 mM,
CaCl2 2.50 mM, MgSO4 1.20 mM, NaHCO3 25.0 mM, K2HPO4 1.03 mM, Na-HEPES 20.0 mM,
D-glucose 11.1 mM). For activation of eNOS, 10 μM calcium ionophore (A12187, Sigma) was
Downloaded from http://circ.ahajournals.org/ by guest on June 16, 2017
added to freshly prepared aortic rings in 1 ml Krebs-HEPES-solution two minutes before
Fe(DETC)2 spin trap addition. 1 ml colloid Fe(DETC)2 was then added to each we
well
lll ((0.4
0 4 mM iin
0.
n
PBS Ca2+/Mg2+) and incubated at 37°C, 10% CO2 for 1 h. After incubation aortic rings were
napfrozen in a 1 mL syringe and recordings were performed at 77°K, using a Dewar flask.
snapfrozen
Instrument
nsttruument settings
settti
ting
ngss were:
ng
weeree: B0-field=3300
B -f
B0
-fie
ield
ie
ld
d=3
=33000 G, swe
sweep=110
w ep
we
p=110
10 G
G,, sw
swee
sweep
eepp ti
ttime=30
me=3
me
= 0 s, steps
ste
teps
ps 4096,
409
096,
6, number
num
umber
passs 10, modul
modulation=7000
ulaatio
ul
on=70
7000 m
mG,
G, MW al
alten
lteen = 10 m
milliwatts,
illliwa
watt
wa
ttss, gain=9
tt
gaiin=9 E2. Levels
Leve
vels of NO
ve
N are
are
expressed as intensit
intensity
i y of signal (A
(A.U.)
A.U.) per weight of wet sampl
sample.
le.
Statistics
All data are expressed as mean±SEM. Comparisons were made by Student’s t-tests, 2-way
ANOVA or 2-way repeated measures ANOVA followed by Newman-Keuls post hoc tests as
appropriate. Concentration-response curves were fitted with a sigmoid dose response curve with
fixed Hill slope (also known as four-parameter logistic equation). Curves were compared by nonlinear regression analysis followed by the extra sum-of-squares F test. Data were analyzed on
GraphPad Prism v6 or SigmaStat v3.5. P<0.05 was considered significant.
11
10.1161/CIRCULATIONAHA.116.023877
Results
Tie2-Cre targeted deletion of Nox2 in mice in vivo reduces basal BP
We first generated a mouse model with a “Floxed” Nox2 allele on a C57Bl6 background such
that Cre-mediated recombination deletes a 3kb fragment of Nox2 including the transcription
initiation site and the first two exons (Fig. 1A,B). Homozygous Floxed mice (Nox2fl/fl) were
crossed with Tie-2 Cre mice to achieve endothelial-targeted deletion of Nox2 (Tie2-Nox2KO),
although bearing in mind that myelo-monocytic cells may also be targeted with this promoter.31
Tie2-Nox2KO mice were born at the expected Mendelian ratio and had no gross abnormalities
Downloaded from http://circ.ahajournals.org/ by guest on June 16, 2017
up to 6 months of age. Body and major organ weights were similar between Tie2-Nox2KO and
Flox littermates (e.g. body weight 26.8±0.3g vs. 26.4±0.1g at 10 weeks ageQ•JURXS7KH
ageQ•JJUR
URXS
XS
XS
7KH
KH
aorta of Tie2-Nox2KO mice showed significant reduction in Nox2 mRNA levels but no changes
inn the mRNA levels of p22phox, p47phox, p67phox, p40phox, SOD1-3 or catalase (Fig. 1C). Nox2
mRNA
mR
RNA
N levels we
w
were
r als
re
also
so su
subs
substantially
b ta
bs
tant
ntia
nt
iall
ia
llyy re
ll
reduced
edu
d ced in
n ccoronary
oronnar
aryy mi
micr
microvascular
c ovvas
a cu
c la
lar en
endothelial
ndo
doth
thel
th
elia
el
i l ce
ccells
llss
ll
(CMEC)
compared
control
(Suppl.
CME
MEC) from Tie2-Nox2KO
ME
T e2
Ti
e2-No
Nox2KO
KO mice com
mpared to co
ontro
roll (S
ro
(Sup
ppl. Fig.
Fig. 1A).
1A)
A . Nox2
Nox2 protein
pro
rottein
ro
in levelss
were significantly ddecreased
ecreased in Tie2-Nox2KO
K mouse aorta compared to control Flox mice, bu
bbutt
there were no differences in p22phox, Nox4 or endothelial NO synthase (eNOS) protein levels
(Fig. 1D). A Tie2-driven deletion of Nox2 was therefore not accompanied by significant changes
in antioxidant genes, Nox4 or eNOS.
Assessment of ambulatory BP by telemetry revealed that Tie2-Nox2KO mice had a
significantly lower basal systolic and mean BP than control Flox mice by approximately 10
mmHg (Fig. 2A,B). There were no differences between groups in heart rate, activity levels or
cardiac structure and function assessed by echocardiography (Fig. 2B and Suppl. Fig. 1B).
12
10.1161/CIRCULATIONAHA.116.023877
Lower basal BP in Tie2-Nox2 KO is not accounted for by altered renal function, vascular
remodeling or endothelium-dependent vasodilation
There were no differences between Tie2-Nox2KO and control mice in urinary volume,
electrolytes or osmolarity in response to an acute saline challenge (Fig. 3A). To look for vascular
remodeling as a basis for the lower BP, we quantified intima-media thickness and area in aortic
sections but this was also similar between groups (Fig. 3B). Endothelium-dependent relaxation
assessed from the vasodilator response to acetylcholine in isolated aortic rings was no different
between Tie2-Nox2KO and control mice (Fig. 3C). The vascular smooth muscle response to the
Downloaded from http://circ.ahajournals.org/ by guest on June 16, 2017
NO donor sodium nitroprusside and the constrictor response to phenylephrine were similar in
both groups. Acetylcholine-induced vasodilation in mesenteric arteries was also similar
sim
mil
ilar
ar between
bet
etwe
weeen
we
groups (Fig. 3D).
Vascular and BP responses of Tie2-Nox2KO mice during AngII stimulation
Afte
terr exposuree ooff ao
aaortic
rtticc ri
rrings
ngss tto
ng
o An
AngI
g I (0
((0.1
.1 μmo
m l//L, 4 h),
h), the
the phenyephrine-induced
phe
heeny
n ep
e hr
hrinee-in
indu
in
duce
du
c d
ce
After
AngII
μmol/L,
vasooconstriction
on w
as ssimilar
imilaar in both groups,
gro
roups, ind
n iccattin
nd
ng comparable
comp
co
mpaarab
blee bas
selin
nes. Acet
tylch
choline-vasoconstriction
was
indicating
baselines.
Acetylcholinenducedd vasodilati
ion was significantly greater iinn Tie2-Nox2KO
Ti
O tha
h n control (Fig. 4A,
4A
A, upper
induced
vasodilation
than
right panel). Aortic superoxide levels were similar in Tie2-Nox2KO and control aorta at baseline
but were significantly higher in the control group after AngII treatment (Fig. 4A, upper left
panel). In line with this, the superoxide scavenger MnTMPyP (10 μmol/L) improved
endothelium-dependent relaxation in the control group but had no significant effect in Tie2Nox2KO (Fig. 4A, lower panels). We also found that acute NOS inhibition with L-NMMA (100
μmol/L) induced greater vasoconstriction in AngII-treated Tie2-Nox2KO aortic rings than
AngII-treated control rings (Suppl. Fig. 2A), suggestive of a greater amount of bioactive NO in
the former setting. In vivo, the hypertensive response observed in control mice with a 2-week
13
10.1161/CIRCULATIONAHA.116.023877
infusion of AngII at 1.1 mg/kg/d was significantly blunted in Tie2-Nox2KO animals, without
difference in the heart rate response (Fig. 4B). Thus, the blunted hypertensive response to AngII
observed in Tie2-Nox2KO mice may be attributable to a lower AngII-induced increase in
endothelial superoxide, a consequent higher level of NO bioavailability and a greater extent of
endothelium-dependent vasorelaxation.
Tie2-Nox2KO mice have increased basal NO bioavailability in vivo
Although basal ex vivo endothelium-dependent vascular function was similar between Tie2Nox2KO and control, it is possible that this might be different in vivo and that an increase in
Downloaded from http://circ.ahajournals.org/ by guest on June 16, 2017
NO-bioavailability due to reduction in Nox2-derived superoxide may account for the lower basal
BP in Tie2-Nox2KO. We tested this idea by assessing the ambulatory BP response
se to
to the
th
he NOS
NOS
inhibitor,
nhibitor, L-NAME (100 mg/kg/d orally for 2 days). The hypertensive response to L-NAME was
found to be significantly greater in Tie2-Nox2KO than control mice, such that BP levels were
similar
imi
mila
lar in the two
la
two groups
group
upss after
up
afteer L
af
L-NAME
-NA
NAME
NA
ME treatme
treatment
m nt (Fig
(Fig.
ig. 5A
ig
5A).
). T
This
his
is ffinding
indi
in
ding
n ssuggests
ugge
ug
gest
ge
sts that
st
th
hatt in vvivo
ivo
iv
NO
O bbioavailability
ioavailabi
bili
bi
lity
li
y at baseline
baseliine may be higher
ba
hi
in
in Tie2-Nox2KO
Tiie2--No
Nox2
x2KO
x2
KO tha
than
han co
control
ontrool mice. To
T as
assess
ssess if LNAME-mediated
NAME-med
diated ch
changes
hanges in BP involved the
h vasculature, we usedd MRIMRI-based
I based measurement of
resistance artery calibre in vivo (by imaging the carotid artery). The carotid artery luminal area in
Tie2-Nox2KO was significantly higher than in matched control mice at baseline (0.46±0.03 mm2
vs. 0.38±0.02 mm2, n=10 each, P<0.05, Fig. 5B,D), consistent with basal vasodilatation.
Acetylcholine induced a similar extent of vasodilatation in both groups of mice (Fig. 5E),
consistent with the results in ex vivo vessels. After L-NAME treatment, however, Tie2-Nox2KO
carotid arteries constricted to a greater extent than in control mice such that luminal vessel
diameters were now similar in both groups (Fig. 5C,F). These results are consistent with the
14
10.1161/CIRCULATIONAHA.116.023877
notion that a higher in vivo NO bioavailability may increase basal resistance vessel calibre in
Tie2-Nox2KO mice.
Contrasting roles of endothelial versus myeloid immune cell Nox2 in BP regulation
The data presented so far suggest that the lower basal BP in Tie2-Nox2KO animals involves an
NO-dependent mechanism but cannot be accounted for by altered endothelial function. Since
Tie2 is expressed not only in endothelial cells but also in myelo-monocytic cells,31 we
considered the possibility that the results observed in Tie2-Nox2KO mice might be related to
Nox2 deletion in myeloid cells. Indeed, Tie2-Nox2KO mice showed clear evidence of CreDownloaded from http://circ.ahajournals.org/ by guest on June 16, 2017
mediated recombination in bone marrow-derived cells (Fig. 6A). Furthermore, Tie2-Nox2KO
Nox2bone marrow cells and circulating mononuclear cells displayed functionally deficient
deficcie
ient
nt N
ox2ox
2
2derived ROS production (Fig. 6B).
To investigate whether Nox2 in myeloid cells contributes to the effects on basal BP, we
next
xt ggenerated
enerated
dam
myeloid-specific
y lo
ye
loid
id
d-spe
speeci
cifi
f c No
fi
N
Nox2
x2 KO mo
m
model
deel using
usin
in
ng a Ly
L
LysM-Cre
sM
M-Cre
-C
Cr mouse
mouse
se lline
in
ne wh
w
which
i h ta
ic
targ
targets
rget
rg
es
myelo-monocytic
cells
myel
eloo-monocy
el
yti
tic cell
ls1177 (Suppl.
(Suuppl. Fig.
Fig
g. 2B).
2B LysM-Cre-Nox2
LyssM-Cr
C e-N
-Nox
-N
o 2K
ox
KO
Om
mice
ice sh
show
showed
wed no ob
obvi
obvious
vious gro
gross
osss
phenotype and hadd comparable amounts of myeloidd inflammatory cells in the vessel wall (Suppl
(Suppl.
Table 1). However, ambulatory BP monitoring by telemetry revealed a significant reduction in
basal systolic BP as compared to control Flox mice, similar to that observed in Tie2-Nox2KO
animals, with no alteration in heart rate (Fig. 6C). To assess if this reduction in basal BP was
related to increased in vivo NO bioavailability, LysM-Cre-Nox2 KO mice and controls were
treated with L-NAME. Indeed, we found that after L-NAME treatment, which had hypertensive
effects in both groups of mice, the systolic BP was similar in LysM-Cre-Nox2 KO and controls
(Fig. 6D). To more directly assess basal vascular NO levels, we performed EPR with NOFe(DETC)2 spin-trapping (Fig 6E). Aortic NO levels were significantly higher in LysM-Cre-
15
10.1161/CIRCULATIONAHA.116.023877
Nox2KO mice than control, consistent with the higher NO bioavailability in vivo (Fig. 6D).
Interestingly, while basal aortic NO levels were increased in LysM-Cre-Nox2KO, iNOS-derived
NO formation as assessed after lipopolysaccharide stimulation was lower than in controls (Suppl.
Fig. 3A). We also studied the response of LysM-Cre-Nox2 KO mice to AngII infusion. In
contrast to the differences in basal BP, the hypertensive response to AngII was similar in LysMCre-Nox2 KO and control mice (Suppl. Fig. 3B), indicating that myeloid cell Nox2 does not
appear to modulate AngII-dependent hypertension.
To elucidate the specific role of Nox2 in the endothelium, we then generated an inducible
Downloaded from http://circ.ahajournals.org/ by guest on June 16, 2017
endothelial Nox2KO model using a Cdh5-CreERT2 driver line.19 In this model, endothelialpecific Nox2 deletion was achieved in adult mice by tamoxifen treatment (Suppl
l. Fi
Fig.
g 22C).
C).
C)
specific
(Suppl.
Tamoxifen treatment had no effect on cardiac function (Suppl. Fig. 2D). In contrast to LysMCre-Nox2 KO mice, Cdh5-CreERT2-Nox2 KO animals showed no difference in basal BP
compared
comp
mpared to ma
mp
matche
matched
hed co
he
controls
ont
ntro
r ls aass as
asse
assessed
s sssed
e by am
aambulatory
bulato
bu
ory telemetry
tele
te
leme
le
mettry (Fig.
(Fiig.
g 7A).
7A)
A).. The
A)
The hype
hypertensive
pert
pe
rten
rt
ensi
en
sive
si
v
response
espponse to L-NA
L-NAME
N ME
NA
M w
was
as aalso
lss similar inn bothh groups
lso
grou
oupss (Fig.
(Fi
Figg. 7B),
7B), suggesting
suggeestting
su
ng a comparable
compa
paraable NO
pa
O
bioavaillabilit
i y unde
d r ba
bbasal
sal conditions in this model.
modell. The quantification
quantiification of basal aortic NO
NO levels
bioavailability
under
by EPR confirmed that these were similar between groups (Fig. 7C). AngII-induced
hypertension, however, was significantly attenuated in Cdh5-CreERT2-Nox2KO mice (Suppl.
Fig. 3C) pointing to endothelial Nox2 as a crucial player in AngII-mediated hypertension.
Discussion
This study provides significant new insights into the role of Nox2 in the regulation of BP. With
the use of several novel cell-specific Nox2 knockout models, we were able to dissect out and
distinguish between the effects of myelo-monocytic cell Nox2 on basal BP and those of
16
10.1161/CIRCULATIONAHA.116.023877
endothelial cell Nox2 on AngII-induced hypertension. We accordingly identify cell-specific and
context-specific roles for Nox2 in BP regulation which indicates that the roles of ROS are highly
complex, not only among different ROS sources but also for the same source in different cell
types.
An important and unexpected initial finding was that Tie2-Nox2KO mice, which we
generated as a model of endothelial-specific Nox2 deletion, exhibited significantly lower basal
BP compared to matched controls. While some previous studies have reported a lower BP in
global Nox2KO mice,32,33 it was not immediately obvious why the deletion of endothelial Nox2
Downloaded from http://circ.ahajournals.org/ by guest on June 16, 2017
should affect basal BP. For example, previous work from our group21 and others34 showed that
endothelial-specific overexpression of Nox2 had no effect on basal BP. Compensatory
Compenssat
ator
oryy changes
or
chan
ch
ange
an
ges
ge
in
n other enzymes such as the antioxidants SOD1-3 and catalase, eNOS or Nox4 could
conceivablyy be involved (and Nox4 was previously linked to lower BP26), but we found no
differences
diff
ferrences in tthe
he llevels
e el
ev
e s of tthese
hese
he
se eenzymes
nzym
nz
mes
e betwe
between
w en
we
n Tie2
Tie2-Nox2KO
e2
2-No
Nox2
No
x KO
O aand
n ccontrol
nd
ontrrol m
mice.
ice. T
ic
The
hee
investigation
function
nveestigation of exx vvivo
ivvo endothelium-dependent
enddothelium-dep
ott
ependent
n fun
nt
uncttio
ionn in
in Tie2-Nox2KO
Tie2--Nox22KO
O mouse
mouse aorta
aorta
ta and
mesenteric artery di
did not reveal any enhancement of relaxation, in line with the finding
finding that basal
b sa
ba
vascular O2- production was unaltered. We also found no changes in cardiac or renal function nor
any evidence of structural vascular remodeling to explain the lower BP. Nevertheless, the lower
BP in Tie2-Nox2KO mice was related to an enhanced in vivo NO bioavailability as indicated by
the normalization of the BP difference upon treatment with L-NAME. Furthermore, in vivo
assessment of resistance vessel function by MRI revealed that Tie2-Nox2KO mice had a
pronounced vasodilatation under basal conditions that could be reversed by L-NAME treatment
but showed no evidence of altered endothelium-dependent relaxation. Taken together, these
findings led us to consider the possibility that non-endothelial Nox2-containing cells may be
17
10.1161/CIRCULATIONAHA.116.023877
involved in the effect on basal BP. In this regard, it is known that Tie2 is expressed not only in
endothelial cells but also in certain hematopoietic cells, notably monocytic cells.31,35 Indeed, we
confirmed that the Tie2-Cre approach not only targeted endothelial cells but also induced
functional Nox2 deletion in myelo-monocytic cells.
The generation of novel LysM-Cre-Nox2 KO and Cdh5-CreERT2-Nox2 KO mice then
allowed us to establish that the reduction in basal BP was in fact related to deletion of Nox2 from
myelo-monocytic cells rather than endothelial cells, with an associated increase in in vivo NO
bioavailability. Using EPR with NO-Fe(DETC)2 spin-trapping, we found that the aortae of
Downloaded from http://circ.ahajournals.org/ by guest on June 16, 2017
LysM-Cre-Nox2 KO mice had increased NO levels, pointing to the vasculature as the likely site
of action of myelo-monocytic cells with disrupted Nox2. Conversely, Cdh5-CreERT2-Nox2
Cdh5-CreE
ERT
T2-No
Nox2
No
x2 KO
KO
aortae had unaltered vascular NO levels under basal conditions. It should be noted that the
magnitude of difference in NO levels observed in aorta cannot necessarily be extrapolated to the
resistance
esiisttance vascu
vasculature,
culaatu
cu
turee, wh
whic
which
icch is tthe
he mor
more
oree impo
important
ort
r an
nt vascular
vaasc
scul
u ar ssite
ul
itte for
foor BP
P reg
regulation.
gul
ulat
atio
at
ion.
io
n The
he effect
eff
ffec
ectt of
ec
myelo-monocytic
myel
eloo-monocy
el
yti
tic cells
cellls in the
he vasculature
vasculatu
ure is m
most
ost likely
lik
kel
elyy to ref
reflect
e lect th
the
he ina
inactivation
nacttiv
vation of
o endothelialend
ndotheliaalderived NO by myeloi
id Nox2-deriv
i ed superoxid
i e, before
b fore the N
be
O can aff
f ect vascular smoothh
myeloid
Nox2-derived
superoxide,
NO
affect
muscle cell relaxation. When myelo-monocytic Nox2 is disrupted, the levels of bioactive NO
would accordingly rise. Consistent with this idea, we could demonstrate the presence of myelomonocytic cells in the aortae of both LysM-Cre-Nox2 KO and control mice but with no
difference in the number of cells between groups. Myelo-monocytic cells could in principle
produce NO from inducible NOS (iNOS) but we found that iNOS-derived NO (after LPS
stimulation) was decreased in LysM-Cre-Nox2 KO mice - consistent with previous reports that
Nox-dependent signaling can increase iNOS expression36 – and making it unlikely that myeloid
cell iNOS plays a role in the observed changes in basal BP. The reason why only the deletion of
18
10.1161/CIRCULATIONAHA.116.023877
myelo-monocytic Nox2 but not the deletion of endothelial Nox2 affects basal BP is most likely
because the abundance of Nox2 is much lower in endothelial cells than myeloid cells.37 We did
note, however, that the reduction in basal BP was slightly higher in the Tie2-Nox2KO mice
(where both myeloid and endothelial cells Nox2 is deleted) than in LysM-Cre-Nox2 KO. This
could indicate that the effects of myelo-monocytic Nox2 KO may be enhanced by the
concomitant knockout of endothelial cell Nox2. The current results are to our knowledge the first
indication that myelo-monocytic cell Nox2 modulates basal BP.
AngII is a potent activator of both Nox1 and Nox2 and an impact of Nox enzymes on
Downloaded from http://circ.ahajournals.org/ by guest on June 16, 2017
AngII-induced hypertension has been documented in many previous studies, as discussed in the
Introduction.
ntroduction. However, the specific roles of different Nox isoforms and different ce
cell
ll ttypes
ypes
yp
ess hhas
as
remained
emained unclear. Quite strong evidence supports a role for vascular smooth muscle Nox1 in
AngII-induced hypertension, involving changes in vascular remodeling.10,11 The cell-specific
role
olee ooff Nox2 iiss un
unclear
uncl
c eaar - a pe
pert
pertinent
rtin
rt
inen
in
entt qu
en
question
ues
e tion because
bec
e au
ause Nox2
Nox
ox22 is
i expressed
exp
xpre
r ss
re
s ed in endothelial
endo
en
doth
do
theliaal ce
th
cell
cells,
l s,
ll
cardiomyocytes,
fibroblasts,
vascular
muscle
inflammatory/immune
card
dio
i myocytees,
es fi
fib
brooblastss, certain va
asccular ssmooth
moot
oth mu
musc
s le ccells
sc
ellss and in
nfllam
a matory
ry/imm
ry
m unee
cells.6 No
Nox2
N
x2 is involvedd in the genesis off endothelial
enddothe
h li
l al dysfunction iinn diverse models,
mode
d ls,66,7,77 and
previous studies from our lab21 and others32 found that endothelium-targeted overexpression of
Nox2 enhanced AngII-induced hypertension. However, the role of endogenous endothelial Nox2
was not established in those studies. The current results in Cdh5-CreERT-Nox2KO mice and
Tie2-Nox2KO mice clearly establish that endogenous endothelial Nox2 augments AngII-induced
hypertension, at least in the relatively short-term (2 weeks). Interestingly, AngII-induced
hypertension was comparable in LysM-Cre-Nox2 KO and respective control mice, indicating
that constitutive deficiency of Nox2 in myelomonocytic cells is apparently not important in this
setting. On the other hand, previous data suggest that T cell Nox215 as well as Nox2 in the
19
10.1161/CIRCULATIONAHA.116.023877
subfornical organ,12,13 and possibly Nox2 in renal afferent arterioles,14 also modulate AngIIinduced hypertension. Furthermore, it was also shown that mice can be protected from arterial
hypertension when LysM-positive cells are depleted prior to beginning of AngII-infusion, and
that BP can be restored by adoptive transfer of Nox2-competent monocytes into these mice.16
Taken together, these data indicate complex roles during AngII-induced hypertension for Nox2
in multiple cell types, some of which involve altered NO bioavailability and others which may
involve Nox2-dependent redox signaling events. The endothelium-dependent effects of Nox2
defined in the current study appear to involve an increased inactivation of NO by ROS, which
Downloaded from http://circ.ahajournals.org/ by guest on June 16, 2017
affects endothelium-dependent relaxation as suggested by our ex vivo studies, but could
potentially also affect other NO-dependent functions.
The major new finding of this study is the potential for myelo-monocytic cells to affect
basal BP in a reversible manner. While the absolute change in basal BP is modest, it is of a
imi
mila
lar magnitude
la
magnittud
udee to
t changes
cha
hang
n ess that
ng
tha
hatt would
w ul
wo
uldd be con
o side
on
deredd cl
clin
in
nic
ically
ly
y oorr prognostically
prog
pr
o noost
stic
ical
ic
ally
al
ly significant,
sig
gniifi
fica
cant
ca
nt,
nt
similar
considered
clinically
for ex
example in
n hhypertension.
ypperteension.
n. From a pa
athophy
ysiollogic
ical
ic
al perspective,
persp
specttivve, itt iiss in
nterestin
ng to specu
ulaate
pathophysiological
interesting
speculate
whether the effects off these cells on BP might
h be
be enhanced in di
ddisease
sease settings where
whe
h re monocytes
are activated, e.g. inflammatory conditions. In this regard, patients with chronic granulomatous
disease (CGD), who have functionally deficient Nox2 activity, might be of particular interest.
More broadly, the current results suggest the potential for different disease conditions to alter BP
through Nox2-dependent effects in distinct cell types, i.e. myeloid cells or endothelial cells. In
terms of RAAS-dependent hypertension, Nox2 inhibition would be anticipated to be beneficial
and combined Nox1/Nox2 inhibitors could be of particular value given the effects of both
isoforms to increase BP. In summary, this study identifies distinct effects of myeloid cell Nox2
20
10.1161/CIRCULATIONAHA.116.023877
and endothelial cell Nox2 on basal and AngII-dependent BP respectively, and suggests that Nox2
may be a master regulator of BP.
Sources of Funding
These studies were supported by the British Heart Foundation (CH/1999001/11735), the German
Cardiac Society, the Deutsche Forschungsgemeinschaft (SA 3282/1-1), and a Fondation Leducq
Transatlantic Network of Excellence. MS was supported by a Deutsche Forschungsgemeinschaft
Joint PhD Studentship (IRTG1816).
Downloaded from http://circ.ahajournals.org/ by guest on June 16, 2017
Disclosures
None
References
Refe
f reencess
fe
[1]
Laursen
Rajagopalan
Z,, Tarpe
Tarpey
Freeman
BA,
Harrison
Role
1] La
L
aursen JB,
B R
B,
ajag
aj
gopalan
an
n S, Galis Z
ey M, Fre
reem
re
eman
em
nB
A,, H
arrrissonn DG. Ro
ole off
superoxide
hypertension.
Circulation.
upeero
roxi
xide iin
xi
n aangiotensin
ngiioten
nsin II-induced
III-i
-ind
n ucced but
nd
ut not
nott catecholamine-induced
cattecchola
lami
la
m ne
ne-i
-ind
-i
ndducced
e hyp
yperrte
tens
n io
on. C
irc
rculatio
on.
1997;95:588-593.
1997;95:
5 5888 593.
[2] Higashi Y, Sasaki S, Nakagawa K, Matsuura H, Oshima T, Chayama K. Endothelial function
and oxidative stress in renovascular hypertension. N Engl J Med. 2002;346:1954-1962.
[3] Dröge W. Free radicals in the physiological control of cell function. Physiol Rev. 2002;82:4795.
[4] Burgoyne JR, Mongue-Din H, Eaton P, Shah AM. Redox signaling in cardiac physiology and
pathology. Circ Res. 2012;111:1091-1106.
[5] Lonn E, Bosch J, Yusuf S, Sheridan P, Pogue J, Arnold JM, Ross C, Arnold A, Sleight P,
Probstfield J, Dagenais GR; HOPE and HOPE-TOO Trial Investigators. Effects of long-term
vitamin E supplementation on cardiovascular events and cancer: a randomized controlled trial.
JAMA. 2005;293:1338-1347.
[6] Lassègue B, San Martín A, Griendling KK. Biochemistry, physiology, and pathophysiology
of NADPH oxidases in the cardiovascular system. Circ Res. 2012;110:1364-1390.
[7] Brandes RP, Weissmann N, Schröder K. Redox-mediated signal transduction by
cardiovascular Nox NADPH oxidases. J Mol Cell Cardiol. 2014;73:70-79.
[8] Cifuentes-Pagano ME, Meijles DN, Pagano PJ. Nox Inhibitors & Therapies: Rational Design
of Peptidic and Small Molecule Inhibitors. Curr Pharm Des. 2015;21:6032-6035.
21
10.1161/CIRCULATIONAHA.116.023877
Downloaded from http://circ.ahajournals.org/ by guest on June 16, 2017
[9] Landmesser U, Cai H, Dikalov S, McCann L, Hwang J, Jo H, Holland SM, Harrison DG.
Role of p47(phox) in vascular oxidative stress and hypertension caused by angiotensin II.
Hypertension. 2002;40:511-515.
[10] Matsuno K, Yamada H, Iwata K, Jin D, Katsuyama M, Matsuki M, Takai S, Yamanishi K,
Miyazaki M, Matsubara H, Yabe-Nishimura C. Nox1 is involved in angiotensin II-mediated
hypertension: a study in Nox1-deficient mice. Circulation. 2005;112:2677-2685.
[11] Dikalova A, Clempus R, Lassègue B, Cheng G, McCoy J, Dikalov S, San Martin A, Lyle A,
Weber DS, Weiss D, Taylor WR, Schmidt HH, Owens GK, Lambeth JD, Griendling KK. Nox1
overexpression potentiates angiotensin II-induced hypertension and vascular smooth muscle
hypertrophy in transgenic mice. Circulation. 2005;112:2668-2676.
[12] Peterson JR1, Burmeister MA, Tian X, Zhou Y, Guruju MR, Stupinski JA, Sharma RV,
Davisson RL. Genetic silencing of Nox2 and Nox4 reveals differential roles of these NADPH
oxidase homologues in the vasopressor and dipsogenic effects of brain angiotensin II.
Hypertension. 2009;54:1106-1114.
[13] Lob HE, Schultz D, Marvar PJ, Davisson RL, Harrison DG. Role of the NADPH oxidases
in the subfornical organ in angiotensin II-induced hypertension. Hypertension. 2013;61:382-387
[14] Lai EY, Solis G, Luo Z, Carlstrom M, Sandberg K, Holland S, Wellstein A, Welch WJ,
Wilcox
contractile
W
co CS. p47(phox)
p 7(p o ) iss required
equ ed for
o afferent
a e e arteriolar
a e o a co
ac e responses
espo ses to
o aangiotensin
g ote s II
and perfusion pressure in mice. Hypertension. 2012;59:415-420.
[15]
Goronzy
Weyand
15] Guzik
k TJ, Hoch NE, Brown KA, McCann LA, Rahman A, Dikalov S, Goronz
nzzy J, W
eyan
ey
andd
an
C, Harrison DG. Role of the T cell in the genesis of angiotensin II induced hypertension and
vascular dysfunction. J Exp Med. 2007;204:2449-2460.
[16]
16] Wenzel P, Knorr M, Kossmann S, Stratmann J, Hausding M, Schuhmacher S, Karbach SH,
Schwenk
M,, Yogev
Grabbe
Schw
wen
enkk M
Yog
o ev N, Schulz E, Oelze M, Grabb
be S, Jonuleit H,, Becker
Becke
ker C, Daiber A, Waisman
Münzel
Lysozyme
M-positive
mediate
arterial
A, M
ünzel T. Ly
Lyso
ozy
z me M
-ppos
osit
i iv
it
ivee monocytes
mono
mo
n cytees me
mediat
atee angiotensin
at
angi
an
giot
gi
oten
en
nsin
siin II-induced
III-in
i duuce
cedd ar
arte
t rial
a
al
hypertension
dysfunction.
Circulation.
2011;124:1370-1381.
hype
p rtension and vvascular
pe
ascu
cu
ula
lar dy
dys
sfuncttio
ion. Circul
latio
on. 20
011;124
24:137
24
370-13381..
[17]
Hammer
RE,
Miyazaki
Williams
Tie217]
7]] Kisanuki
Kisanuki YY,
Y , Ha
YY
Ham
mmerr R
E, Miy
yazzaki J, W
illliam
am
ms SC,
SC Richardson
Richaardsonn JA,
Ri
JA, Yanagisawa
Yanag
gis
i aw
wa M. T
ie2Cre tr
transgenic
mice:
model
Biol.
tran
ansg
an
sgen
sg
en
nic m
icce: a nnew
ew m
o ell for
od
for endothelial
end
ndothe
heliial cell-lineage
he
celll-l
-lin
i eaage aanalysis
naaly
lysi
s s in vvivo.
ivo.
iv
o Dev
Dev Bi
iol.
2001;230:230-242.
2001;230
3 :230
3 -2422.
[18] Clausen BE; Burkhardt C; Reith W; Renkawitz R; Forster I. Conditional gene targeting in
macrophages and granulocytes using LysMcre mice. Transgenic Res. 1999;8:265-277.
[19] Kusumbe AP, Ramasamy SK, Adams RH. Coupling of angiogenesis and osteogenesis by a
specific vessel subtype in bone. Nature. 2014;507:323-328.
[20] Anilkumar N, Weber R, Zhang M, Brewer A, Shah AM. Nox4 and Nox2 NADPH oxidases
mediate distinct cellular redox signaling responses to agonist stimulation. Arterioscl Throm Vasc
Biol. 2008;28:1347–1354.
[21] Murdoch CE, Alom-Ruiz SP, Wang M, Zhang M, Walker S, Yu B, Brewer A, Shah AM.
Role of endothelial Nox2 NADPH oxidase in angiotensin II-induced hypertension and
vasomotor dysfunction. Basic Res Cardiol. 2011;106:527-538.
[22] Phinikaridou A, Andia ME, Protti A., Indermuehle A, Shah A, Smith A, Warley A, Botnar
RM. Noninvasive magnetic resonance imaging evaluation of endothelial permeability in murine
atherosclerosis using an albumin-binding contrast agent. Circulation. 2012;126:707–719.
[23] Zhang M, Prosser BL, Bamboye MA, Gondim ANS, Santos CX, Martin D, Ghigo A, Perino
A, Brewer AC, Ward CW, Hirsch E, Lederer WJ, Shah AM. Increased Nox2 activity modulates
cardiac calcium handling and contractility via phospholamban phosphorylation. J Am Coll
Cardiol. 2015;66:261-272.
22
10.1161/CIRCULATIONAHA.116.023877
Downloaded from http://circ.ahajournals.org/ by guest on June 16, 2017
[24] Soodvilai S, Jia Z, Wang MH, Dong Z, Yang T. Mpges-1 deletion impairs diuretic response
to acute water loading. Am J Physiol Renal Physiol. 2009;296:F1129-1135.
[25] Charles RL, Rudyk O, Prysyazhna O, Kamynina A, Yang J, Morisseau C, Hammock BD,
Freeman BA, Eaton P. Protection from hypertension in mice by the Mediterranean diet is
mediated by nitro fatty acid inhibition of soluble epoxide hydrolase. Proc Natl Acad Sci U S A.
2014;111:8167-8172.
[26] Ray R, Murdoch CE, Wang M, Santos CX, Zhang M, Alom-Ruiz S, Anilkumar N, Ouattara
A, Cave AC, Walker SJ, Grieve DJ, Charles RL, Eaton P, Brewer AC, Shah AM. Endothelial
Nox4 NADPH oxidase enhances vasodilatation and reduces blood pressure in vivo. Arterioscler
Thromb Vasc Biol. 2011;31:1368-1376.
[27] Görlach A, Brandes RP, Nguyen K, Amidi M, Dehghani F, Busse R. A gp91phox
containing NADPH oxidase selectively expressed in endothelial cells is a major source of
oxygen radical generation in the arterial wall. Circ Res. 2000;87:26-32.
[28] Chaubey S, Jones GE, Shah AM, Cave AC, Wells CM. Nox2 is required for macrophage
chemotaxis towards CSF-1. PLoS One. 2013 ;8:e54869.
[29] Li JM, Mullen AM, Yun S, Wientjes F, Brouns GY, Thrasher AJ, Shah AM. Essential role
of the NADPH oxidase subunit p47(phox) in endothelial cell superoxide production in response
too phorbol
p o bo ester
es e aand
d tumor
u o necrosis
ec os s factor-alpha.
ac o a p a. Circ
Ci c Res.
es. 2002;90:143-150.
00 ;90: 3 50.
[30]
30] Kossmann S, Hu H, Steven S, Schönfelder T, Fraccarollo D, Mikhed Y, Brähler
Brähl
hler
er M,
M, Knorr
Knor
Kn
orrr
or
M, Brandt M, Karbach SH, Becker C, Oelze M, Bauersachs J, Widder J, Münzel T
T,, Daiber
A,,
Daib
Da
iber
ib
er A
Wenzel P. Inflammatory monocytes determine endothelial nitric-oxide synthase uncoupling and
nitro-oxidative stress inducedd by angiotensin II. J Biol Chem. 2014;289:27540-27550.
[31]
31] Tangg Y, Harrington A, Yang X, Eriessel RE, Liaw L. The contribution of the Tie2+ linage
too primitive
definitive
2010;48:563-567.
pri
riimi
miti
t ve and
ti
and def
efinitive hematopoietic cells. Genesis.
Geenesis. 2010;48:
:56
5 3-56
567.
[32]
AC,
Heymes
Gall
Shah
32]] Bendall
Bendall JK,
JK
K, Cave
Ca e A
C, H
eyme
ey
mess C,, Ga
me
G
ll N,
N, Sh
hah AM.
AM.
M Pivotal
Piv
ivottal role
rol
ole of
o a gp91(phox)gp9
p91(
1(ph
1(
phoxx)ph
containing
II-induced
hypertrophy
conntaining NADPH
H oxidase
oxi
xiidasee iinn angiotensin
angiioten
nsin II-in
induucedd ccardiac
ardiaac hype
pertro
ophyy in
i mice.
e Circulation.
e.
Circ
rculatio
ionn.
io
2002;105:293-296.
2002
02;1
02
; 05:293-2
-296
-2
9 .
96
[33]
Y,, Qu
Quinn
MT,
Cayatte
RA.
33] Wang
Wan
angg HD,
HD Xu S,, Johns
Johnss DG,
DG,, Du Y
DG
Quin
innn MT
in
T, Ca
Caya
yatt
ya
ttte AJ,
AJJ, Co
Cohenn RA
A. Role
R le of
Ro
of NADPH
NA H
oxidase in the
oxidative
h vascular hypertrophic
hypertrophhic
i and
d oxi
idative stress response to angiotensin II in mice.
Circ Res. 2001;88:947-953.
[34] Bendall JK, Rinze R, Adlam D, Tatham AL, de Bono J, Wilson N, Volpi E, Channon KM.
Endothelial Nox2 overexpression potentiates vascular oxidative stress and hemodynamic
response to angiotensin II: studies in endothelial-targeted Nox2 transgenic mice. Circ Res.
2007;100:1016-1025.
[35] Shaw JP, Basch R, Shamamian P. Hematopoietic stem cells and endothelial cell precursors
express Tie-2, CD31 and CD45. Blood Cells Mol Dis. 2004;32:168-175.
[36] Wu F, Tyml K, Wilson JX. iNOS expression requires NADPH oxidase-dependent redox
signaling in microvascular endothelial cells. J Cell Physiol. 2008;217:207-214.
[37] Li JM, Shah AM. Differential NADPH- versus NADH-dependent superoxide production by
phagocyte-type endothelial cell NADPH oxidase. Cardiovasc Res. 2001;52:477-486.
23
10.1161/CIRCULATIONAHA.116.023877
Figure Legends
Figure 1. Generation of Tie2-Nox2KO mice.
A. Targeting strategy for generation of Nox2flox mice (Flox). The endogenous Nox2 locus is
shown at the top and the targeting vector at the bottom. LoxP sites are represented by blue
triangles and FRT sites by double red triangles. After successful targeting, the Neomycin (Neo)
cassette was excised using the FRT sites. Cre-mediated recombination deletes a 3kb fragment of
Nox2 including the transcriptional initiation site and the first two exons. B. Southern blot
Downloaded from http://circ.ahajournals.org/ by guest on June 16, 2017
analysis of genomic DNA from selected ES cell clones screened for 5’-homologous (upper
7C1
11 were
weree uused
sedd fo
se
fforr
panel) and 3’-homologous recombination (lower panel). The clones 1A3 and 7C11
blastocyst injection to generate Nox2flox mice. C. mRNA levels of Nox2, accessory subunits and
antioxidant genes in Tie2-Nox2KO and Flox control aortae. D. Protein levels of Nox2, p22phox,
Nox4
x44 and eNOS
eNO
OS in
n Tie2-Nox2KO
Tie
ie2ie
2-No
2N x2
No
x2KO
KO aand
nd ccontrol
ontroll aortae.
aorrtaae. R
Representative
epre
ep
rese
re
s nt
ntaativ
i e im
immu
immunoblots
uno
nobblo
lots
t aree sshown
hown
ho
wn
above
below.
abov
ovve and mean
an ddata
atta be
below
w. #, P<0.05 vvs.
s. Flox
Flox.
x.
Figure 2. Tie2-Nox2KO mice have reduced basal blood pressure.
A. Representative telemetric BP traces in Tie2-Nox2KO mice and Flox controls. B. Mean data
for BP, heart rate and activity level. #, P<0.05 vs. Flox.
Figure 3. Renal function, vascular remodeling and in vitro vascular function in Tie2Nox2KO mice.
A. Renal function assessed by response to an acute saline challenge. Changes in urine
osmolarity, volume, and sodium and potassium concentrations are shown. RM-TW-ANOVA =
24
10.1161/CIRCULATIONAHA.116.023877
repeated measures two-way ANOVA; N.S. = not significant between groups. B. Representative
histological sections of aorta from Tie2-Nox2 KO and Flox mice (x40 magnification), and mean
intima-media area and thickness. C. Concentration-response curves for response of aortic rings
to the NO donor sodium nitroprusside (SNP), phenylephrine (PE) and acetylcholine (ACh). D.
Concentration-response curves for response of mesenteric arteries to ACh.
Figure 4. Tie2-Nox2KO mice demonstrate blunted in vitro and in vivo responses to AngII.
A. Upper left panel: Superoxide levels in Tie2-Nox2KO and Flox aorta at baseline and after
Downloaded from http://circ.ahajournals.org/ by guest on June 16, 2017
AngII stimulation, assessed by HPLC of the specific dihydroethidium oxidation product, 2hydroxyethidium (EOH). #, P<0.05 vs. Flox; *, P<0.05 vs. control conditions (TW-ANOVA).
(TW
W-AN
ANOV
OVA)
OV
A).
A)
Upper right panel: Effect of AngII pre-treatment on vasodilation of aortic rings to ACh. Lower
panels: Effect of MnTMPyP on ACh responses in AngII-treated rings from Flox mice (left) and
Tie2-Nox2KO
Tie2
2-Nox2KO
O mice
mice
ce (right).
(ri
righ
ri
ght)
gh
t). B. E
t)
Effect
ffec
ff
e t of chron
ec
chronic
nicc A
AngII
ngIII iinfusion
n us
nf
usio
i n (1.1
(1.11 m
mg/kg/day)
g kgg/d
g/
/day
ay)) on
ay
o tel
telemetric
e em
el
emet
etri
et
ricc
ri
Tie2-Nox2KO
repeated
measures
twoBP aand
nd heart rrate
ate in T
at
iee2-No
Nox2KO andd Flox mice.
No
m cee. RM-TW-ANOVA
mi
RM
M-TW
TW--ANOV
OVA = re
epeated m
easuures tw
ea
woway AN
ANOVA;
NOVA;
A # indi
indicates
d cates significant diff
difference
ference between genotypes; * indicates signif
significant
fic
i ant
interaction.
Figure 5. Tie2-Nox2KO mice have increased basal NO bioavailability in vivo.
A. BP response to L-NAME in Tie2-Nox2KO and Flox mice, assessed by telemetry. # indicates
significant difference between genotypes, * indicates significant interaction as tested by RMTW-ANOVA. B, C. Representative MRI images of the carotid artery in Tie2-Nox2KO and Flox
mice at baseline (B) and after L-NAME treatment (C). Scale bar, 1 mm. D-F. Mean data for
25
10.1161/CIRCULATIONAHA.116.023877
diastolic area of the carotid artery at baseline (D), following ACh treatment (E), and after LNAME treatment (F). #, P<0.05 vs. Flox; *, P<0.05 vs. control conditions (TW-ANOVA).
Figure 6. LysM-Cre-Nox2KO mice have reduced basal BP and increased vascular NO
bioavailability.
A. Cre-mediated recombination in bone marrow-derived cells from Tie2-Nox2KO mice. B.
Functionally deficient ROS production in phorbol ester (PMA)-stimulated bone marrow cells
and circulating mononuclear cells from Tie2-Nox2KO mice, assessed by flow cytometry in cells
Downloaded from http://circ.ahajournals.org/ by guest on June 16, 2017
loaded with dihydroethidium (DHE). C. Reduced basal BP (telemetry) in LysM-Cre-Nox2KO
mice. #, P<0.05 vs. Flox. D. In vivo response to L-NAME in LysM-Cre-Nox2KO
O mice
mic
icee and
and
controls. # indicates significance vs. Flox as tested by RM-TW-ANOVA. E. Increased aortic NO
evels in LysM-Cre-Nox2KO
Lys
y M-Cre-Nox2KO under basal conditions, as assessed by EPR. Representative EPR
levels
spectra
pecctr
tra are show
shown
own to the
ow
the right.
rig
ightt. #, P
P<0.05
< .0
<0
05 vs. Fl
Flox.
lox
o .
Figure. 7. Cd
C
Cdh5-CreERT2-Nox2KO
h5-CreE
ERT2-Nox2KO mice have unaltered ba
b
basal
sall BP.
A. Telemetric BP and heart rate at baseline. B. BP responses to L-NAME in Cdh5-CreERT2Nox2KO and Flox mice. L-NAME increased BP in both groups but there was no significant
difference between groups by RM-TW-ANOVA. C. Comparable aortic NO levels in Cdh5CreERT2-Nox2KO and controls under basal conditions, as assessed by EPR. Representative
EPR spectra are shown to the right.
26
A
B
Nox2floxed mice (Flox)
2.0
2.0
AORTA
Tie2-Nox2KO, n=5
mRNA (A.U.)
(A.U.
1.5
1.0
0.5
0.0
D
1.5
Tie2-Nox2KO, n=5
1.0
0.5
#
Nox2
AORTA
Flox, n=6
Flox, n=6
mRNA (A.U.)
Downloaded from http://circ.ahajournals.org/ by guest on June 16, 2017
C
phox
h
p22
2ph
Tie2Flox Nox2KO Flox
hox
p40
40phox
40
Tie2Nox2KO
ox
p47phox
Flox
0.0
0
p67
67phox
Tie2Tie2
Nox2KO Flox
Catalase
Ca
Cat
a ala
ase
Tie2Nox2KO Flox
Tie2Tie2
Nox2KO
Flox
S
SOD1
O
OD1
S
SO
SOD
SOD2
2
Tie2Nox2KO
Tie2Nox2KO Flox
Nox2
p22phox
Nox4
eNOS
β-actin
β-actin
β-actin
β-actin
SO
SOD3
D
Flox
Tie2Nox2KO
A
Flox
160
#
100
n=7
n=6
Flox
Tie2-Nox2KO
Diastolic BP (mmHg)
130
70
160
130
100
70
n=7
n=6
Flox
Tie2-Nox2KO
15
800
600
Activity
ctiv
i ity (A.U.)
Heart rate (min-1)
Downloaded from http://circ.ahajournals.org/ by guest on June 16, 2017
Systolic BP (mmHg)
160
Mean BP (mmHg)
B
Tie2-Nox2KO
400
10
0
5
200
0
n=7
n=6
Flox
Tie2-Nox2KO
0
n=7
n=6
Flox
Tie2-Nox2KO
130
#
100
70
n=7
n=6
Flox
Tie2-Nox2KO
800 RM-TW-ANOVA
500
N.S.
Flox, n=5
0
[Na+] (mmol/L)
1000
600
400
200
N.S.
Tie2-Nox2KO, n=4
0
1
2
3
4
Time After Injection (h)
200
N.S.
100
50
N.S.
1
2
3
4
Time After Injection (h)
1
2
3
4
Time After Injection (h)
50
37 μm
Flox
50000
25000
37 μm
Tie2-Nox2KO
Intima-media thickness (Pm)
75000
n=3
Flox
n=3
40
30
20
10
0
Tie2-Nox2KO
O
C
n=3
F
Fl
Flo
Flox
x
n=3
Tie2-Nox2KO
D
50
100
P=N.S.
-9
-8
-7
-6
SNP (log M)
-5
-4
100
00
0
80
80
60
0
Flox,
nt10
Flox, nt
Fl
t10
0
Tie2-Nox2KO,
Tie2-Nox2KO
O,
nt10
40
20
0
-10
P=N.S.
-9
-8
-7
-6
PE (log M)
-5
-4
0
Flox,
nt10
Flo
ox, nt
t10
10
% Re
Relaxation
ela
laxa
ax tion
Flox, nt10
Tie2-Nox2KO,
KO
O
nt10
%R
Re
Relaxation
ela
laxa
xati
xa
tion
ti
on
0
Contraction
% Co
C
ont
ntra
ract
ctio
ion
n
Downloaded from http://circ.ahajournals.org/ by guest on June 16, 2017
Intima-media area (μm 2)
100000
0
150
-10
100
0
0
1
2
3
4
Time After Injection (h)
B
% Relaxation
150 RM-TW-ANOVA
300 RM-TW-ANOVA
[K+] (mmol/L)
1500 RM-TW-ANOVA
Volume (PL)
Urine Osmolarity
(mOsm/Kg H2O)
A
Tie2-Nox2KO,
Tie
e2-N
Noxx2KO,
nt10
nt
t10
0
50
100
P=N.S.
-10
-9
Flox, nt5
Tie2-Nox2KO,
nt5
50
5
0
100
P=N.S.
-8
-7
-6
ACh (log M)
-5
-4
-8
-7.5 -7
-6.5 -6
ACh (log M)
-5.5
Tie2-Nox2KO
80
0
*
Flox
#
60
0
n=7
40
20
Flox, nt15
% Relaxation
100
n=8
EOH
(nmol/mg tissue)
A
Tie2-Nox2KO,
nt15
50
100
P<0.05
Basal
AngII
-10
-9
-8
-7
-6
-5
-4
0
Flox, nt7
Flox +
MnTMPyP, nt7
50
100
P<0.05
-10 -9
-8 -7 -6 -5
ACh (log M)
Relaxation (in %)
Downloaded from http://circ.ahajournals.org/ by guest on June 16, 2017
Relaxation (in %)
ACh (log M)
0
Tie2-Nox2KO +
MnTMPyP, nt5
Tie2-Nox2KO,
nt5
50
100
P=N.S.
-10 -9
-4
-8 -7 -6 -5
ACh (log M)
-4
AngII
AngII
180
160
#
140
120
Flox, n=6
T
Tie2-Nox2KO,
ie2-Nox2KO,
KO,
O,, n=5
n=
=5 RMRM
RM-TW-ANOVA
M TW--AN
NOVA
100
80
-1
1
2
5
8
11
1
D asttoli
Di
lic BP (mmHg)
li
Diastolic
Systolic BP (mmHg)
B
14
4
160
140
120
100
#
60
0
RM-TWTW
W ANO
AN VA
RM-TW-ANOVA
--1
1
2
5
160
140
#
120
*
100
RM-TW-ANOVA
2
5
days
8
11
11
14
4
AngII
11
14
Heart rate (min-1)
Mean BP (mmHg)
AngII
180
-1
8
days
days
80
*
80
750
600
N.S.
450
300
150
0
RM-TW-ANOVA
-1
2
5
days
8
11
14
A
150
B
L-NAME
0.4
0.2
0.0
n=10
n=10
Flox
Tie2-Nox2KO
Ctrl
200
*#
*
#
0.4
0.2
Flox
RM-TW-ANOVA
Ctrl
L-NAME
TTie2-Nox2KO
ie2-N
No
F
0.6
0
+ L-NAME
0.8
0.0
L-NAME
FFlox
lox
n=6, + ACh
#
70
C
n=10
Maximal diastolic area (mm2)
Ctrl
E
0.6
90
80
TTie2-Nox2KO
ie
e2-N
Nox2
2KO
0.8
*#
100
#
Baseline
Flox
D
70
L-NAME
*
Maximal diastolic area (mm2)
Downloaded from http://circ.ahajournals.org/ by guest on June 16, 2017
Ctrl
90
110
400
Tie2-Nox2KO
0.8
0.6
#
0.4
*
0.2
0.0
Flox
n=5, + L-NAME
Tie2-Nox2KO, n=7
70
100
80
Flox, n=7
80
110
120
n=10
90
120
N.S.
n=4, + L-NAME
100
130
n=10
110
130
Heart rate (min-1)
#
140
Maximal diastolic area (mm2)
120
140
n=10
130
RM-TW-ANOVA
Mean BP (mmHg)
Diastolic BP (mmHg)
Systolic BP (mmHg)
140
600
150
RM-TW-ANOVA
RM-TW-ANOVA
n=5, + ACh
150
Tie2-Nox2KO
A
B
Tie2-Nox2KO
Tie2-Nox2KO
Flox
WT
KO
Cre
-
+
-
+ -
Bone marrow cells
Flox
Tie2-Nox2KO
-PMA
600
140
rate (min-1)
Heart ra
#
120
100
80
D
n=7
Flox
F x
Flo
n=7
550
Mononuclear cells
500
450
400
LysM-Cre-Nox2KO
L M-Cre-Nox2KO
Lys
Flox
n=7
LysM-Cre-Nox2KO
Lys
sM-C
M re-Nox2KO
700
00
n=4
-1
Systolic BP (mmHg)
mmH
mH
Hg)
g
160 RM-TW-ANOVA
RMR
M TWTW
W ANO
NOVA
NO
n=7
Heart rate (min
min
n )
C
BP (mmHg)
Systolic B
50
0
150
n=3
n 3
140
n=7
130
#
n=7
120
Basal
600
Aorta
Flox
LysM-Cre-Nox2KO
n=6
n=4
n=7
N.S.
n=3
400
300
200
RM-TW-ANOVA
Basal
L-NAME
#
200
Flox
n=7
500
L-NAME
400
0
600
6
LysM-Cre-Nox2KO
EPR Signal (A.U.)
E
Aortic NO levels (A.U.)/
mg tissue weight
Downloaded from http://circ.ahajournals.org/ by guest on June 16, 2017
+PMA
Flox
n=6
LysM-Cre-Nox2KO
3250
3300
Magnetic field (G)
3350
A
n=5
n=8
120
100
120
100
600
400
200
80
Flox Cdh5-CreERT2
Nox2KO
800
Heart rate (min-1)
100
Mean BP (mmHg)
Diastolic BP (mmHg)
Systolic BP (mmHg)
120
80
140
140
140
n=5
n=8
n=5
80
Flox Cdh5CreERT2
Nox2KO
n=8
n=5
0
Flox Cdh5-CreERT2
Nox2KO
n=8
Flox Cdh5Cre-ERT2
Nox2KO
150
700
110
600
140
120
130
N.S.
120
110
Flox, n=5
5
Cdh5-CreERT2-Nox2KO,
eERT2-Nox2KO, n=8
Basal
al
100
L-NAME
C
600
0
RM-TW-ANOVA
Basal
L-NAME
N.S.
100
500
N.S.
400
0
90
80
Heart rate (min-1)
Hea
Mean BP (mmHg)
N.S.
Diastolic BP (mmHg)
Diastol
140
160
100
120
RM-TW-ANOVA
Aortic NO levels ((A.U.)/
A.U
.U.)/
.
.)
ight
mg tissue weight
Systolic BP (mmHg)
180
RM-TW-ANOVA
Basal
L-NAME
300
RM-TW-ANOVA
Aorta
Aorta
a
Cdh5-CreERT2-Nox2KO
Cdh
d 5
dh
5-C
-CreE
ERT2
2-No
N x2K
2KO
O
EPR Signal (A.U.)
U.)
Downloaded from http://circ.ahajournals.org/ by guest on June 16, 2017
B
400
200
0
n=7
n=6
Flox
Cdh5-CreERT2Nox2KO
Flox
3250
3300
Magnetic field (G)
3350
Basal
LNAME
Distinct Regulatory Effects of Myeloid Cell and Endothelial Cell Nox2 on Blood Pressure
Can Martin Sag, Moritz Schnelle, Juqian Zhang, Colin E. Murdoch, Sabine Kossmann, Andrea Protti,
Celio X. C. Santos, Greta J. Sawyer, Xiaohong Zhang, Heloise Mongue-Din, Daniel A. Richards,
Alison C. Brewer, Oleksandra Prysyazhna, Lars S. Maier, Philip Wenzel, Philip J. Eaton and Ajay M.
Shah
Downloaded from http://circ.ahajournals.org/ by guest on June 16, 2017
Circulation. published online March 15, 2017;
Circulation is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Copyright © 2017 American Heart Association, Inc. All rights reserved.
Print ISSN: 0009-7322. Online ISSN: 1524-4539
The online version of this article, along with updated information and services, is located on the
World Wide Web at:
http://circ.ahajournals.org/content/early/2017/03/15/CIRCULATIONAHA.116.023877
Free via Open Access
Data Supplement (unedited) at:
http://circ.ahajournals.org/content/suppl/2017/03/15/CIRCULATIONAHA.116.023877.DC1
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published in
Circulation can be obtained via RightsLink, a service of the Copyright Clearance Center, not the Editorial Office.
Once the online version of the published article for which permission is being requested is located, click Request
Permissions in the middle column of the Web page under Services. Further information about this process is
available in the Permissions and Rights Question and Answer document.
Reprints: Information about reprints can be found online at:
http://www.lww.com/reprints
Subscriptions: Information about subscribing to Circulation is online at:
http://circ.ahajournals.org//subscriptions/
Sag et al.
SUPPLEMENTAL MATERIAL
1
Sag et al.
Supplementary Table 1. Aortic myelo-monotic and lymphatic cells in Flox control and
LysM-Cre-Nox2KO mice.
Flox control
LysM-Cre
Nox2KO
Cell type
CD45+CD11b+ Ly6G-
Number of
n
Number of
cells / 10 mg
cells 10 / mg
aortic tissue
aortic tissue
n
P
100 ± 60
4
120 ± 20
5
N.S.
3.5 ± 2.2
4
8.1 ± 1.3
5
N.S.
540 ± 230
5
580 ± 80
5
N.S.
790 ± 340
5
810 ± 110
5
N.S.
640 ± 310
5
640 ± 90
5
N.S.
(Monocytes)
CD45+CD11b+ F4/80+
(Macrophages)
CD45+CD11b+ Ly6G+
(Neutrophils)
CD45+TCRβ+
(T-Lymphocytes)
CD45+CD19+
(B-Lymphocytes)
Data are mean±SEM.
2
0.6
0.8
CMEC
CMEC
0.6
0.4
mRNA (A.U.)
mRNA (A.U.)
A
0.2
0.2
#
n=3
0.0
0.4
#
n=3
Nox2
Flox
Tie2-Nox2KO
n=3
0.0
Flox
n=3
Nox4
Tie2-Nox2KO
B
40
20
n=4
Flox
400
200
n=5
0
Tie2-Nox2KO
Flox
n=5
0.6
0.4
0.2
n=4
Flox
n=5
Tie2-Nox2KO
Tie2-Nox2KO
n=6
n=6
Flox
Tie2-Nox2KO
n=4
n=5
5
0.8
0.6
0.4
0.2
0.0
4
0
LV enddiastolic diameter
(mm)
0.8
IVSd (mm)
n=4
1.0
LV endsystolic diameter
(mm)
1.0
0.0
LV/BW (mg/g)
Heart Rate (bpm)
Ejection fraction (%)
60
0
8
600
80
n=4
Flox
n=5
Tie2-Nox2KO
Suppl. Fig. 1
4
3
2
1
0
Flox
Tie2-Nox2KO
A
L - N M M A - in d u c e d c o n tr a c t io n ( % )
200
AO RTA
*
150
100
50
n=6
n=12
F lo x
T ie 2 -N o x 2 K O
0
B
LysM-Cre-Nox2KO
WT
KO
Cre
C
-
+
-
-
+
+
Cdh5-CreERT2-Nox2KO
WT
KO
Cre
-
D
+
-
+
-
Ejection fraction (%)
80
60
40
20
n=5
0
n=8
Flox Cdh5-CreERT2
Nox2KO
Suppl. Fig. 2
Aortic NO levels (A.U.)/
mg tissue weight
600
Aorta
400
#
200
0
Flox
EPR Signal (A.U.)
A
n=4
n=4
Flox
LysM-CreNox2KO
LysM-Cre-Nox2KO
+ LPS
3250
3300
+ LPS
B
AngII
180
RM-TW-ANOVA
N.S.
160
140
Flox, n=4
LysM-Cre-Nox2KO, n=4
120
800
3
5
7
N.S.
600
400
200
RM-TW-ANOVA
0
100
Bl
AngII
1000
Heart rate (min-1)
Systolic BP (mmHg)
200
Bl
9
3
5
AngII
RM-TW-ANOVA
160
#
140
120
Flox, n=3
Cdh5-CreERT2-Nox2KO, n=5
100
Bl
3
Heart rate (min )
180
5
9
AngII
1000
-1
Systolic BP (mmHg)
200
7
days
days
C
3350
Magnetic field (G)
7
9
800
600
N.S.
400
200
RM-TW-ANOVA
0
Bl
3
5
days
days
Suppl. Fig. 3
7
9
Sag et al.
Supplementary Figure 1. Coronary microvascular endothelial cell mRNA levels and
cardiac echocardiography.
A. mRNA levels of Nox2 and Nox4 in coronary microvascular endothelial cells (CMEC). #,
P<0.05 vs. Flox. B. Echocardiographic parameters of cardiac structure and function, and left
ventricle/body weight ratio (LV/BW). IVSD, interventricular septal diameter.
Supplementary Figure 2. Generation of LysM-Cre-Nox2KO and Cdh5-CreERT2Nox2KO mice.
A. Magnitude of L-NMMA-induced constriction in AngII-treated aortic rings from Tie2Nox2KO mice and Flox controls. *, P<0.05 vs. Flox. B. Cre-mediated recombination in Cre+
bone marrow cells from LysM-Cre-Nox2KO mice. C. Cre-mediated recombination in Cre+
lung tissue from Cdh5-CreERT2-Nox2KO mice. D. Left ventricular ejection fraction in Cdh5CreERT2-Nox2KO cf. Flox mice following Tamoxifen treatment.
Supplementary Figure 3. Hypertensive response to AngII in LysM-Cre-Nox2KO and
Cdh5-CreERT2-Nox2KO mice.
A. Reduced aortic iNOS-derived NO formation in LysM-Cre-Nox2KO after LPS stimulation. #,
P<0.05 vs. Flox. Representative EPR spectra shown to the right. B. In vivo response to AngII
infusion in LysM-Cre-Nox2KO mice and Flox control. C. In vivo response to AngII infusion in
Cdh5-CreERT2-Nox2KO mice and respective control. RM-TW-ANOVA = repeated measures
two-way ANOVA. #, P<0.05 vs. Flox control.
3