Hypoxia Triggers Subcellular Compartmental Redox
Signaling in Vascular Smooth Muscle Cells
Gregory B. Waypa, Jeremy D. Marks, Robert Guzy, Paul T. Mungai, Jacqueline Schriewer,
Danijela Dokic, Paul T. Schumacker
Downloaded from http://circres.ahajournals.org/ by guest on June 18, 2017
Rationale: Recent studies have implicated mitochondrial reactive oxygen species (ROS) in regulating hypoxic
pulmonary vasoconstriction (HPV), but controversy exists regarding whether hypoxia increases or decreases
ROS generation.
Objective: This study tested the hypothesis that hypoxia induces redox changes that differ among subcellular
compartments in pulmonary (PASMCs) and systemic (SASMCs) smooth muscle cells.
Methods and Results: We used a novel, redox-sensitive, ratiometric fluorescent protein sensor (RoGFP) to assess the
effects of hypoxia on redox signaling in cultured PASMCs and SASMCs. Using genetic targeting sequences,
RoGFP was expressed in the cytosol (Cyto-RoGFP), the mitochondrial matrix (Mito-RoGFP), or the mitochondrial intermembrane space (IMS-RoGFP), allowing assessment of oxidant signaling in distinct intracellular
compartments. Superfusion of PASMCs or SASMCs with hypoxic media increased oxidation of both CytoRoGFP and IMS-RoGFP. However, hypoxia decreased oxidation of Mito-RoGFP in both cell types. The
hypoxia-induced oxidation of Cyto-RoGFP was attenuated through the overexpression of cytosolic catalase in
PASMCs.
Conclusions: These results indicate that hypoxia causes a decrease in nonspecific ROS generation in the matrix
compartment, whereas it increases regulated ROS production in the IMS, which diffuses to the cytosol of both
PASMCs and SASMCs. (Circ Res. 2010;106:526-535.)
Key Words: mitochondria 䡲 reactive oxygen species 䡲 ratiometric fluorescent protein sensor
䡲 hypoxic pulmonary vasoconstriction
S
mall pulmonary arteries constrict in response to a
decrease in alveolar oxygen tension through a physiological response termed hypoxic pulmonary vasoconstriction (HPV). Despite intensive study since the first description of HPV in 1946, the mechanism of O2 sensing remains
unclear. Mitochondria have been considered a putative site
of cellular oxygen sensing because they consume O2 and
therefore represent the intracellular site with the lowest
oxygen tension. Mitochondria generate superoxide and
hydrogen peroxide during cellular respiration, so changes
in the production of reactive oxygen species (ROS) during
hypoxia could potentially signal changes in O2 tension and
trigger cellular responses. However, whether hypoxia increases or decreases ROS generation during HPV remains
unclear.
Using chemiluminescence in whole perfused lungs and
endothelium-denuded rings of distal pulmonary arteries,
Michelakis, Archer, and Weir and colleagues detected
decreases in ROS generation during hypoxia.1– 4 Similarly,
in cultured pulmonary arterial pulmonary artery smooth
muscle cells (PASMCs), Mehta et al detected a decrease in
ROS generation as determined by dihydrodichlorofluorescein diacetate (DCF), dihydroethidium, and Amplex red.5
By contrast, using cultured PASMCs, we found that
hypoxia triggers an increase in ROS generation.6 – 8 Our
studies used the ROS-sensitive indicator dichlorofluorescein (DCF), mitochondrial inhibitors, enzymatic and
chemical antioxidants and mitochondria-deficient (␳0)
cells. Other investigators have detected similar increases in
ROS generation during hypoxia by using lucigenin-derived
chemiluminescence and electron paramagnetic resonance
spectrometry in small pulmonary arteries, as well as DCF
in PASMCs.8,9
Attempts to resolve the question of whether ROS levels
increase or decrease during hypoxia have been hindered by
the technical limitations of the tools used to assess oxidant
stress. Although lucigenin can detect ROS, lucigenin itself is
capable of producing superoxide in cells.9 Furthermore,
Original received July 30, 2009; revision received December 8, 2009; accepted December 8, 2009.
From the Department of Pediatrics (G.B.W., R.G., P.T.M., J.S., D.D., P.T.S.), Division of Neonatology, Northwestern University Feinberg School of
Medicine; and Department of Pediatrics (J.D.M.), Section of Neonatology, University of Chicago, Ill.
Correspondence to Paul T. Schumacker, PhD, Department of Pediatrics, Northwestern University, Morton Building 4-685, 303 E Chicago Ave,
Chicago, IL 60611. E-mail [email protected]
© 2010 American Heart Association, Inc.
Circulation Research is available at http://circres.ahajournals.org
DOI: 10.1161/CIRCRESAHA.109.206334
526
Waypa et al
Downloaded from http://circres.ahajournals.org/ by guest on June 18, 2017
questions regarding how these probes distribute between
intracellular and extracellular compartments5 have led to
concerns regarding the interpretation of data arising from
their use. Other probes such as DCF and dihydroethidium
lack specificity and can potentially accumulate within organelles.10 In addition, none of these probes exhibits ratiometric behavior, so changes in fluorescence arising from
changes in intracellular dye concentration, excitation intensity, or fluorescence path length caused by changes in cell
volume cannot be distinguished from changes in redox state.
Finally, obtaining a global measure of tissue ROS production
during hypoxia may mask important differences in ROS
production occurring among cellular subcompartments. The
goal of the present study was to test the hypothesis that
hypoxia differentially alters ROS generation, depending on
the subcellular compartment. To address this, we used a novel
redox-sensitive protein sensor, which we separately targeted
to the mitochondrial matrix, the mitochondrial intermembrane space, and the cytosol. This sensor, RoGFP, contains
engineered cysteine residues, enabling dithiol formation in
response to oxidant stress.11–15 Oxidation resulting in dithiol
formation causes reciprocal changes in emission intensity
when excited at two different wavelengths. Unlike DCF or
lucigenin, the fluorescence excitation spectrum of RoGFP is
strongly and reversibly dependent on the redox state of the
introduced cysteines.12 Moreover, by exposing cells to strong
reducing and oxidizing agents at the conclusion of an experiment, RoGFP oxidation state can be calibrated on a cell-bycell basis in situ. We used these targeted sensors to assess
redox changes in cellular subcompartments induced by hypoxia. We also compared the hypoxia-induced redox responses
in PASMCs with those of systemic arterial smooth muscle
cells (SASMCs).
Methods
An expanded Methods section is available in the Online Data
Supplement at http://circres.ahajournals.org.
Pulmonary Microvessel Myocyte Isolation
The Northwestern University Institutional Animal Care and Use
Committee approved all animal studies. PASMCs were isolated from
rat lungs as described previously6 using a modification of the method
of Marshall et al.16 Cells isolated by this method were confirmed to
be PASMCs as previously described.6 Using a similar method,
SASMCs were isolated from rat renal arteries and confirmed to be
SASMCs as previously described.6 The isolated cells were cultured
for 2 weeks (4 passages) and then used for the next 4 weeks (4 to 12
passages).
RoGFP Sensors
The RoGFP protein contains two engineered cysteine thiols, as
first described by Remington et al (RoGFP2).11 The cDNA
encoding the protein was created by introducing four mutations in
the mammalian GFP expression vector (pEGFP-N1) (C48S,
Q80R, S147C, and Q204C) using a QuikChange Multi Sitedirected mutagenesis kit (Strategene). The RoGFP construct was
ligated into the VQ Ad5CMV K-NpA adenoviral shuttle vector
between the KpnI and NotI sites; after sequencing and amplification, this plasmid was used to generate a recombinant adenovirus
to permit widespread expression in our cells (ViraQuest Inc,
North Liberty, Iowa). The resulting redox-sensitive protein has
excitation maxima at 400 and 484 nm, with emission at 525 nm.
In response to changes in redox conditions, RoGFP exhibits
Compartmental Redox Responses to Hypoxia
527
Non-standard Abbreviations and Acronyms
Cyto-RoGFP
DCF
DTT
FRET
GPD
HPV
IMS
IMS-RoGFP
Mito-RoGFP
PASMC
RoGFP
ROS
SASMC
tBH
␳0
cytosolic RoGFP
dihydrodichlorofluorescein diacetate
dithiothreitol
fluorescence resonance energy transfer
glycerol phosphate dehydrogenase
hypoxic pulmonary vasoconstriction
mitochondrial intermembrane space
mitochondrial intermembrane space RoGFP
mitochondrial matrix RoGFP
pulmonary arterial smooth muscle cell
redox-sensitive, ratiometric fluorescent protein sensor
reactive oxygen species
systemic arterial smooth muscle cell
tert-butyl hydroperoxide
mitochondria-deficient cells
reciprocal changes in intensity at the two excitation maxima,12 and its
ratiometric characteristics render it insensitive to expression levels.13–15
Although the fluorescence behavior of RoGFP is relatively independent
of pH and it does not respond to authentic NO, reduced NADH, or the
antioxidant N-acetyl-L-cysteine, its spectrum is slightly affected by
reduced glutathione possibly attributable to thiol-disulfide exchange
(Online Figures I and II).
RoGFP was expressed in the mitochondrial matrix (Mito-RoGFP)
by appending a 48-bp region encoding the mitochondrial targeting
sequence from cytochrome oxidase subunit IV, at the 5⬘ end of the
coding sequence. This construct was then ligated into the VQ
Ad5CMV K-NpA plasmid between the KpnI and NotI sites, and used
to generate an adenoviral vector. RoGFP was targeted to the
mitochondrial intermembrane space (IMS-RoGFP) by appending it
to glycerol phosphate dehydrogenase (GPD). A cDNA construct
encoding GPD, an integral protein of the inner mitochondrial
membrane whose C terminus protrudes into the intermembrane
space,17 was ligated in-frame with cDNA encoding RoGFP.17 The
corresponding polypeptide includes amino acids 1 to 626 of GPD,
with RoGFP at the carboxy terminus. This method has been used
previously to express YFP in the intermembrane space.18 (See the
Online Data Supplement for characterization of the RoGFP sensors
and experimental protocols.)
Statistics
Changes in RoGFP oxidation and YC2.3 fluorescence resonance energy
transfer (FRET) ratio were analyzed by using a 2-way ANOVA with
repeated measures. A Newman–Keuls multiple-range test was used to
evaluate significant differences between groups and times. To control
for differences in the hypoxic responses of cultured myocytes, experimental studies and control experiments were always carried out on the
same day. Statistical significance was set at P⬍0.05.19
Results
Hypoxia Induces Progressive Cytosolic Oxidation
in PASMCs
To determine the effect of hypoxia on ROS production in
the cytosol, isolated rat PASMCs expressing cytosolic
RoGFP (Cyto-RoGFP) were subjected to hypoxia. Under
baseline normoxic conditions, the Cyto-RoGFP sensor was
18.6⫾2.0% oxidized (Figure 1A). During hypoxia, CytoRoGFP demonstrated significantly progressive oxidation
528
Circulation Research
February 19, 2010
Downloaded from http://circres.ahajournals.org/ by guest on June 18, 2017
Figure 1. Hypoxia shifts the cytosol to a more oxidized state in PASMCs. A, Combined results from multiple experiments in PASMCs
expressing Cyto-RoGFP and superfused with either normoxic (21% O2) or hypoxic (1.5% O2) media. B through E, False-color ratiometric images of PASMCs demonstrate the dynamic redox range of the Cyto-RoGFP sensor. PASMCs expressing Cyto-RoGFP were
superfused with normoxic (21% O2) media (B), then with hypoxic (1.5% O2) media for 30 minutes (C), followed by media containing
DTT to fully reduce the sensor (D) and then with media containing tBH to fully oxidize the sensor (E). F, Representative time course,
quantitative analysis of the fluorescence intensity ratio from representative individual PASMCs imaged in B through E. Img Fig indicates
the time at which images B through E were captured during the experiment. G, Representative time course, quantitative analysis of the
reciprocal changes in Cyto-RoGFP fluorescence intensity at the two excitation maxima (484 and 400 nm) for a single PASMC. Values
are means⫾SE (n⫽6 cover slips, 4 to 10 cells/cover slip). *P⬍0.05 compared to normoxia.
over a period of 30 minutes compared to normoxic
PASMCs and reached 34.5⫾3.2% oxidized by the end of
the experiment (Figure 1A). False-color ratiometric representative images (Figure 1B through 1E) and quantitative
changes of the fluorescence ratio from individual PASMCs
(Figure 1F) demonstrate the dynamic responsiveness of the
Cyto-RoGFP sensor to hypoxia, dithiothreitol (DTT)
(1 mmol/L), and tert-butyl hydroperoxide (tBH) (1 mmol/
L). In addition, Figure 1G demonstrates the reciprocal
changes in fluorescence intensity for the two excitation
maxima (484 and 400 nm) of Cyto-RoGFP in response to
hypoxia, DTT, and tBH in a single PASMCs. Taken
together, these results indicate that hypoxia shifts the
cytosol to a more oxidized state.
The Mitochondrial IMS Becomes More Oxidized
During Hypoxia in PASMCs
Having confirmed that hypoxia oxidizes thiol targets in the
cytosol, we asked whether hypoxia induces ROS production in the IMS of the mitochondria. RoGFP expression
was therefore targeted to the intermembrane space by
appending it to GPD, an enzyme localized to the IMS.18
Confocal imaging demonstrated that IMS-RoGFP expression colocalized with mitochondrial cytochrome c, which
is known to be localized to the intermembrane space
(Figure 2A through 2C). To further verify correct targeting
of IMS-RoGFP to the intermembrane space, we performed
immuno-electron microscopy of cells using an antibody
against GFP. At the dilutions used, no background labeling
Waypa et al
Compartmental Redox Responses to Hypoxia
529
Downloaded from http://circres.ahajournals.org/ by guest on June 18, 2017
Figure 2. Hypoxia shifts the mitochondrial IMS to a more oxidized state in PASMCs. A through C, Confocal images of
PASMCs expressing IMS-RoGFP (A), immunostained for cytochrome c (B), and demonstrating colocalization in mitochondria
(C). D, IMS-RoGFP targeted to the mitochondrial intermembrane space of rat PASMCs as determined by electron microscopy. Primary antibody: mouse monoclonal anti-GFP; secondary antibody: goat anti-mouse IgG, conjugated to 10-nm gold
particles (arrows). Note the restriction of labeling to the mitochondrial membrane on the mitochondrial periphery. E, Combined results from multiple experiments in PASMCs expressing
IMS-RoGFP and superfused with normoxic (21% O2) or hypoxic
(1.5% O2) media. F through I, False-color ratiometric images of
PASMCs demonstrate the dynamic redox range of the IMSRoGFP sensor. PASMCs expressing IMS-RoGFP were superfused with normoxic (21% O2) media (F) and then with hypoxic
(1.5% O2) media for 30 minutes (G), followed by media containing DTT to fully reduce the sensor (H) and then with media containing tBH to fully oxidize the sensor (I). Values are means⫾SE
(n⫽6 cover slips, 4 to 10 cells/cover slip). *P⬍0.05 compared
to normoxia.
outside mitochondria was observed in multiple images of
separate mitochondria. Rather, the Immunogold labeling
was restricted to the mitochondrial membrane at the
mitochondrial periphery and cristae, demonstrating expression restricted to the mitochondrial intermembrane space
(Figure 2D).
During normoxia, the RoGFP in the IMS was more
oxidized (47.7⫾4.5%; Figure 2E) compared with the oxidation state demonstrate with Cyto-RoGFP (18.6⫾2.0%; Figure
1A). In response to hypoxia, IMS-RoGFP also significantly
shifted to a more oxidized state compared to normoxic
PASMCs and reached a level of 62.1⫾3.4% oxidized at the
end of the experiment (Figure 2E). These results indicate that
hypoxia shifts the mitochondrial IMS to a more oxidized
state.
The Mitochondrial Matrix Becomes More Reduced
During Hypoxia in PASMCs
Having found that hypoxia induces RoGFP oxidation in
the intermembrane space, we next assessed the effect of
hypoxia on redox status in the mitochondrial matrix.
RoGFP was targeted to the mitochondrial matrix using the
mitochondrial targeting sequence from cytochrome oxidase subunit IV. Confocal microscopy studies of expression of Mito-RoGFP and endogenous manganese superoxide dismutase, a protein expressed exclusively in the
mitochondrial matrix, demonstrated colocalization of these
two proteins at the resolution afforded by light microscopy
(Figure 3A through 3C). Further assessment of its targeting
was examined in Immunogold-labeled cryosections examined under electron microscopy (Figure 3D). Unlike the
530
Circulation Research
February 19, 2010
Downloaded from http://circres.ahajournals.org/ by guest on June 18, 2017
Figure 3. Hypoxia shifts the mitochondrial matrix to a more
reduced state in PASMCs. Confocal images of PASMCs
expressing IMS-RoGFP (A), immunostained for manganese
superoxide dismutase (MnSOD) (B), and demonstrating colocalization in mitochondria (C). D, Mito-RoGFP targeted to the
mitochondrial matrix of rat PASMCs as determined by electron
microscopy. Primary antibody: mouse monoclonal anti-GFP;
secondary antibody: goat anti-mouse IgG, conjugated to 10-nm
gold particles (arrows). E, Combined results from multiple
experiments in PASMCs expressing Mito-RoGFP and superfused with normoxic (21% O2) or hypoxic (1.5% O2) media. F
through I, False-color ratiometric images of PASMCs demonstrate the dynamic redox range of the Mito-RoGFP sensor.
PASMCs expressing Mito-RoGFP were superfused with normoxic (21% O2) media (F) and then with hypoxic (1.5% O2)
media for 30 minutes (G), followed by media containing DTT to
fully reduce the sensor (H) and then with media containing tBH
to fully oxidize the sensor (I). Values are means⫾SE (n⫽6 cover
slips, 4 to 10 cells/cover slip). *P⬍0.05 compared to normoxia.
pattern of Immunogold labeling seen for IMS-RoGFP
(Figure 2D), the Immunogold labeling in Mito-RoGFPexpressing cells was restricted to the matrix and was not
associated with either cristae or peripheral mitochondrial
membranes.
Under baseline normoxic conditions, the Mito-RoGFP
sensor was 62.6⫾1.5% oxidized (Figure 3E), significantly
more oxidized than RoGFP localized to the mitochondrial
intermembrane space (Figure 2E). However, unlike the
Cyto- and IMS-RoGFP, which became more oxidized,
during hypoxia Mito-RoGFP became progressively more
reduced compared to normoxic PASMCs decreasing to
44.4⫾1.9% oxidized by the end of the experiment (Figure
3E). These results reveal that the mitochondrial matrix
undergoes a reductive stress during hypoxia.
Expression of Antioxidant Proteins Attenuates
Hypoxia-Induced Oxidation of RoGFP
Our observation of hypoxia-induced oxidation of RoGFP in the
cytosol and mitochondrial IMS, but not the matrix, suggested
that hypoxia-induced ROS generation is localized to the cytosol
and and/or intermembrane space. To determine whether H2O2
signaling contributes to the oxidation response during hypoxia,
catalase was first overexpressed in the cytosolic compartment.7
Cytosolic catalase overexpression completely blocked the hypoxia-induced increase in cytosolic oxidation (Figure 4A), indicating that hypoxia-induced oxidative stress in the cytosolic
compartment is ultimately H2O2. Overexpression of catalase in
the mitochondrial matrix, as previously achieved,7 had no effect
on the hypoxia-induced decrease in oxidation of Mito-RoGFP
during hypoxia (Figure 4B). These results indicate that the
Waypa et al
Compartmental Redox Responses to Hypoxia
531
during hypoxia, PASMCs and SASMCs expressing YC2.3 were
compared. Hypoxia significantly increased the YC2.3 FRET
ratio in PASMCs by 15 minutes (2.31⫾0.03 to 2.62⫾0.05;
Figure 5D), whereas it significantly decreased the ratio in
SASMCs within 30 minutes (2.29⫾0.04 to 2.05⫾0.08; Figure
5D). Taken together, these results indicate that hypoxia induces
a similar pattern of redox modulation in SASMCs and PASMCs,
and that there are cell type–specific downstream effectors that
respond to this pattern of redox modulation.
Discussion
Downloaded from http://circres.ahajournals.org/ by guest on June 18, 2017
Figure 4. Catalase attenuates hypoxia-induced oxidation of
Cyto-RoGFP in PASMCs. Cytosolic catalase was overexpressed
in cells expressing Cyto-RoGFP (A), whereas mitochondrial
catalase was overexpressed in cells expressing Mito-RoGFP (B).
Values are means⫾SE (n⫽6 cover slips, 4 to 10 cells/cover slip).
*P⬍0.05 compared to hypoxia.
progressive reduction in the mitochondrial matrix during hypoxia is not mediated by H2O2 signaling.
Hypoxia Shifts the Redox Status of Subcellular
Compartments in SASMCs
Unlike pulmonary arteries, systemic arteries relax in response
to hypoxia.4 We assessed the redox response to hypoxia in
smooth muscle cells isolated from renal arteries to determine
whether they respond differently from PASMCs during identical hypoxic stimulation. Just as in the PASMCs, hypoxia
induced a significant oxidation of Cyto-RoGFP from normoxic baseline (21.6⫾1.6% to 46.7⫾5.9% oxidized; Figure
5A). Similarly, hypoxia also induced oxidation of RoGFP
localized to the mitochondrial IMS from normoxic baseline
(27.5⫾3.5% to 50.4⫾4.6% oxidized; Figure 5B). Finally,
hypoxia significantly reduced RoGFP targeted to the mitochondrial matrix from normoxic baseline (48.7⫾3.3% to
21.7⫾2.3% oxidized; Figure 5C), as we observed in the
PASMCs. To examine the effect of the hypoxia-induced increase in ROS signaling on cytosolic calcium ([Ca2⫹]i) signaling
In this study, we genetically targeted a reporter of redox
state to subcellular compartments. The expression of this
sensor in vascular smooth muscle cells reveals important
differences in protein thiol redox state between the cytosol,
IMS and the mitochondrial matrix. Under normoxic conditions we found that the cytosol is the most reduced and
the mitochondrial matrix is the most highly oxidized. The
intermembrane space demonstrates an intermediate level
of protein thiol oxidation.
We previously used a FRET redox sensor (HSP-FRET) to
detect increases in oxidant stress in the cytosol of cultured
PASMCs during hypoxia.7 Those measurements provided
evidence of an increase in ROS signaling during hypoxia, but
slow refolding of the HSP-FRET protein after removal of the
oxidant stress hindered our ability to calibrate its redox state
during the experiment. Moreover, attempts to target its
expression to the mitochondria were unsuccessful, impeding
its usefulness for assessing subcellular redox signaling. By
contrast, the ratiometric nature of RoGFP, and its successful
targeting to distinct subcellular compartments, make it a
useful probe for the molecular dissection of redox changes in
response to hypoxia. The high level of basal oxidation
detected by RoGFP in the matrix is consistent with a large
literature identifying that compartment as the primary site of
ROS production in the cell.20 During hypoxia, the matrix
undergoes a reductive shift. The decrease in oxidation during
hypoxia suggests that ROS generation in the matrix arises
from the nonspecific, [O2]-dependent generation of superoxide at a number of sites, such that a decrease in O2 availability
decreases ROS production in a nonspecific manner. In
contrast to the mitochondrial matrix, the cytosol and the IMS
demonstrate an oxidative shift during hypoxia. The increase
in oxidant stress in the IMS and the cytosol occurs despite the
decrease in [O2], and therefore appears to reflect a regulated
increase in the generation of ROS in response to physiological hypoxia. Nevertheless, the oxidant signal we detect is
subtle and it produces a relatively small change in protein
thiol oxidation state.
To our knowledge, this is the first report comparing redox
conditions in the cytosol, IMS and matrix compartments, and
their responses to hypoxia. If mitochondria are the source of the
oxidant signaling in the IMS and cytosol during hypoxia, the
most likely site(s) of ROS generation could be within the IMS,
at the outer membrane, or at the outer surface of the inner
mitochondrial membrane. Importantly, the site of ROS generation cannot be in the matrix because hypoxia decreased
oxidant stress in that compartment. We cannot exclude the
possibility that ROS generation originates in the cytosol
532
Circulation Research
February 19, 2010
Downloaded from http://circres.ahajournals.org/ by guest on June 18, 2017
Figure 5. Hypoxia shifts the redox status of subcellular compartments in SASMCs. SASMCs expressing Cyto-RoGFP (A), IMS-RoGFP
(B), or Mito-RoGFP (C) were superfused under controlled O2 conditions. Effects of hypoxia on [Ca2⫹]i in PAMSC and SASMCs
assessed by YC2.3 (D). Values are means⫾SE (n⫽6 cover slips, 4 to 10 cells/cover slip). *P⬍0.05 compared to hypoxia (A through C)
and *P⬍0.05 compared to normoxic baseline (D).
during hypoxia and diffuses into the IMS, because the
oxidation response in these 2 compartments developed simultaneously. Weissmann et al recently demonstrated a role for
nonphagocytic NADPH oxidase function during the acute
phase of HPV, and suggested that NADPH oxidase may
contribute to the generation of ROS during hypoxia.21 It is
therefore possible that NADPH oxidase, or an alternative
system, contributes to the hypoxia-induced ROS generation
that affects both the cytosol and IMS. We favor the conclusion that the mitochondrial electron transport chain is the
principal source of hypoxia-induced ROS signals, based on
previous studies using mitochondrial inhibitors,
mitochondria-deficient ␳0 cells, and RNAi suppression of the
electron transport chain.6 – 8,22–28 Interestingly, Rathore et al
recently reported that hypoxia increases mitochondriaderived ROS in PASMCs, which leads to the activation of
NAD(P)H oxidase through PKC␧ activation, suggesting that
both mitochondria and cytosolic oxidant systems may contribute to the overall response.24
Our findings conflict directly with the model of HPV by
Michelakis, Archer, and Weir and colleagues, who find that
hypoxia-induced decreases in mitochondrial ROS generation
lead to redox-dependent alterations in Kv channel activation
in the plasma membrane, thereby triggering depolarization
and subsequent contraction.2,4,29 –32 One possible explanation
for this conflict relates to their use of lucigenin chemiluminescence to assess oxidant stress. Our data reveal that redox
responses can differ markedly among subcellular compartments. The unknown distribution of lucigenin between the
intracellular and extracellular spaces, and its possible accumulation in mitochondria, cytosol, sarcoplasmic reticulum or
other intracellular compartments, makes it difficult to interpret data obtained when the luminescence responses from all
these compartments are recorded as a single signal. Our data
demonstrate that, during hypoxia, oxidation increases slightly
in some compartments, whereas it decreases significantly in
others. Therefore, it is likely that the lucigenin signal reports
a decrease during hypoxia because the collective oxidant
stress across all of these compartments undergoes a net
decrease. However, that combined signal would not detect
important increases in oxidant stress in specific compartments where oxidant signaling acts as a critical mediator of
cellular responses.
Waypa et al
Figure 6. Model depicting spatial differences in hypoxia-induced
ROS signaling.
Compartmental Redox Responses to Hypoxia
533
ated the hypoxia-induced increase in cytosolic oxidant signaling underscores the importance of H2O2 signaling in that
response.
In summary, these results are consistent with a model for
HPV in which hypoxia augments oxidant signaling by
H2O2 in the cytosol and IMS, whereas it decreases oxidant
stress in the mitochondrial matrix (Figure 6). The cytosolic
ROS signal triggers an increase in cytosolic calcium,
which mediates PASMC contraction. Smooth muscle cells
from systemic arteries demonstrate a similar pattern of
oxidant signaling, indicating that the differences between
PASMCs and SASMCs in terms of their response to
hypoxia cannot be attributed to differences in mitochondrial function.
Acknowledgments
Downloaded from http://circres.ahajournals.org/ by guest on June 18, 2017
For PASMCs, hypoxia causes an increase in [Ca2⫹]i and
vasoconstriction,33– 44 whereas for SASMCs, the result is a
decrease in [Ca2⫹]i and vasodilation.4,45,46 Michelakis et al
reported that ROS generation increases during hypoxia in
systemic arterial tissue, but decreases in pulmonary arteries,
leading them to conclude that tissue-specific differences in
mitochondrial function explain their opposite responses.4 Those
findings were contradicted by Mehta et al, who detected a
decrease in hypoxia-induced ROS generation in both PASMCs
and coronary artery smooth muscle cells.5 Our comparison of
PASMCs and SASMCs revealed that both cell types exhibit
increases in cytosolic ROS signaling during hypoxia, yet the
functional responses were dissimilar, with PASMCs exhibiting
an increase in [Ca2⫹]i, whereas SASMCs showed a decrease
(Figure 5). Our results suggest that the differing physiological
responses to hypoxia in the two types of vascular smooth muscle
must therefore be mediated by cell-specific expression of downstream signaling pathways, allowing the same upstream oxidant
signal to activate different tissue-specific responses to a hypoxic
stimulus. This conclusion is supported by a growing number of
studies that implicate hypoxia-induced increases in ROS signaling in mediating diverse tissue-specific responses across a wide
range of cell types.6 – 8,27,47–54
The ability of cytosolic catalase to attenuate the hypoxiainduced oxidation of the Cyto-RoGFP sensors is consistent
with previous studies.7,8 However, the overexpression of
mitochondrial catalase had no effect on the hypoxia-induced
decrease in oxidation of Mito-RoGFP (Figure 4B). One
possibility is that Mito-RoGFP is oxidized primarily by
superoxide in the matrix. If so, then overexpression of
mitochondrial catalase might not affect Mito-RoGFP oxidation because catalase degrades H2O2 but not superoxide.
Alternatively, ROS production in the matrix may be largely
absent during hypoxia. If so, then the hypoxia-induced
decrease in Mito-RoGFP oxidation would reflect the rate at
which the protein dithiols are rereduced by matrix reductases.
If matrix H2O2 formation is minimal during hypoxia, then the
rate of Mito-RoGFP oxidation would be minimal and mitochondrial catalase overexpression would not affect its redox
state. The fact that cytosolic catalase overexpression attenu-
We thank Dr Christopher Rhodes (The University of Chicago,
Chicago, Ill) for the adenovirus expressing the YC2.3 FRET sensor.
We also thank Dr Jotham R. Austin II and Janice Wang (The
University of Chicago, Chicago, Ill) for their assistance with the
electron microscopic studies.
Sources of Funding
This work was supported by the NIH (grants HL66315, HL35440,
and HL079650 to P.T.S. and grant NS056313 to J.D.M.) and
American Heart Association Grant 0235457Z (to G.B.W.).
Disclosures
None.
References
1. Archer SL, Nelson DP, Weir EK. Simultaneous measurement of O2
radicals and pulmonary vascular reactivity in rat lung. J Appl Physiol.
1989;67:1903–1911.
2. Archer SL, Huang J, Henry T, Peterson D, Weir EK. A redox-based O2
sensor in rat pulmonary vasculature. Circ Res. 1993;73:1100 –1112.
3. Mohazzab KM, Wolin MS. Properties of a superoxide anion-generating
microsomal NADH oxidoreductase, a potential pulmonary artery PO2
sensor. Am J Physiol. 1994;267:L823–L831.
4. Michelakis ED, Hampl V, Nsair A, Wu X, Harry G, Haromy A, Gurtu R,
Archer SL. Diversity in mitochondrial function explains differences in
vascular oxygen sensing. Circ Res. 2002;90:1307–1315.
5. Mehta JP, Campian JL, Guardiola J, Cabrera JA, Weir EK, Eaton JW.
Generation of oxidants by hypoxic human pulmonary and coronary
smooth-muscle cells. Chest. 2008;133:1410 –1414.
6. Waypa GB, Chandel NS, Schumacker PT. Model for hypoxic pulmonary
vasoconstriction involving mitochondrial oxygen sensing. Circ Res.
2001;88:1259 –1266.
7. Waypa GB, Guzy R, Mungai PT, Mack MM, Marks JD, Roe MW,
Schumacker PT. Increases in mitochondrial reactive oxygen species
trigger hypoxia-induced calcium responses in pulmonary artery smooth
muscle cells. Circ Res. 2006;99:970 –978.
8. Waypa GB, Marks JD, Mack MM, Boriboun C, Mungai PT, Schumacker
PT. Mitochondrial reactive oxygen species trigger calcium increases
during hypoxia in pulmonary arterial myocytes. Circ Res. 2002;91:
719 –726.
9. Spasojevic I, Liochev SI, Fridovich I. Lucigenin: redox potential in
aqueous media and redox cycling with O-2 production. Arch Biochem
Biophys. 2000;373:447– 450.
10. Thannickal VJ, Fanburg BL. Reactive oxygen species in cell signaling.
Am J Physiol Lung Cell Mol Physiol. 2000;279:L1005–L1028.
11. Dooley CT, Dore TM, Hanson GT, Jackson WC, Remington SJ, Tsien
RY. Imaging dynamic redox changes in mammalian cells with green
fluorescent protein indicators. J Biol Chem. 2004;279:22284 –22293.
534
Circulation Research
February 19, 2010
Downloaded from http://circres.ahajournals.org/ by guest on June 18, 2017
12. Hanson GT, Aggeler R, Oglesbee D, Cannon M, Capaldi RA, Tsien RY,
Remington SJ. Investigating mitochondrial redox potential with redoxsensitive green fluorescent protein indicators. J Biol Chem. 2004;279:
13044 –13053.
13. Cannon MB, James Remington S. Redox-sensitive green fluorescent
protein: probes for dynamic intracellular redox responses. A review.
Methods Mol Biol. 2009;476:50 – 64.
14. Lohman JR, Remington SJ. Development of a family of redox-sensitive
green fluorescent protein indicators for use in relatively oxidizing subcellular environments. Biochemistry. 2008;47:8678 – 8688.
15. Jiang K, Schwarzer C, Lally E, Zhang S, Ruzin S, Machen T, Remington
SJ, Feldman L. Expression and characterization of a redox-sensing green
fluorescent protein (reduction-oxidation-sensitive green fluorescent
protein) in Arabidopsis. Plant Physiol. 2006;141:397– 403.
16. Marshall C, Mamary AJ, Verhoeven AJ, Marshall BE. Pulmonary artery
NADPH-oxidase is activated in hypoxic pulmonary vasoconstriction.
Am J Respir Cell Mol Biol. 1996;15:633– 644.
17. Brown LJ, MacDonald MJ, Lehn DA, Moran SM. Sequence of rat
mitochondrial glycerol-3-phosphate dehydrogenase cDNA. Evidence
for EF-hand calcium-binding domains. J Biol Chem. 1994;269:
14363–14366.
18. Porcelli AM, Ghelli A, Zanna C, Pinton P, Rizzuto R, Rugolo M. pH
difference across the outer mitochondrial membrane measured with a
green fluorescent protein mutant. Biochem Biophys Res Commun. 2005;
326:799 – 804.
19. Wallenstein S, Zucker CL, Fleiss JL. Some statistical methods useful in
circulation research. Circ Res. 1980;47:1–9.
20. Lambert AJ, Brand MD. Reactive oxygen species production by mitochondria. Methods Mol Biol. 2009;554:165–181.
21. Weissmann N, Zeller S, Schafer RU, Turowski C, Ay M, Quanz K,
Ghofrani HA, Schermuly RT, Fink L, Seeger W, Grimminger F. Impact
of mitochondria and NADPH oxidases on acute and sustained hypoxic
pulmonary vasoconstriction. Am J Respir Cell Mol Biol. 2006;34:
505–513.
22. Weissmann N, Ebert N, Ahrens M, Ghofrani HA, Schermuly RT, Hanze
J, Fink L, Rose F, Conzen J, Seeger W, Grimminger F. Effects of
mitochondrial inhibitors and uncouplers on hypoxic vasoconstriction in
rabbit lungs. Am J Respir Cell Mol Biol. 2003;29:721–732.
23. Wang QS, Zheng YM, Dong L, Ho YS, Guo Z, Wang YX. Role of
mitochondrial reactive oxygen species in hypoxia-dependent increase in
intracellular calcium in pulmonary artery myocytes. Free Radic Biol Med.
2007;42:642– 653.
24. Rathore R, Zheng YM, Niu CF, Liu QH, Korde A, Ho YS, Wang YX.
Hypoxia activates NADPH oxidase to increase [ROS]i and [Ca2⫹]i
through the mitochondrial ROS-PKCepsilon signaling axis in pulmonary artery smooth muscle cells. Free Radic Biol Med. 2008;45:
1223–1231.
25. Guzy RD, Hoyos B, Robin E, Chen H, Liu L, Mansfield KD, Simon MC,
Hammerling U, Schumacker PT. Mitochondrial complex III is required
for hypoxia-induced ROS production and cellular oxygen sensing. Cell
Metab. 2005;1:401– 408.
26. Guzy RD, Mack MM, Schumacker PT. Mitochondrial complex III is
required for hypoxia-induced ROS production and gene transcription in
yeast. Antioxid Redox Signal. 2007;9:1317–1328.
27. Chandel NS, Maltepe E, Goldwasser E, Mathieu CE, Simon MC, Schumacker PT. Mitochondrial reactive oxygen species trigger hypoxiainduced transcription. Proc Natl Acad Sci U S A. 1998;95:11715–11720.
28. Leach RM, Hill HM, Snetkov VA, Robertson TP, Ward JP. Divergent
roles of glycolysis and the mitochondrial electron transport chain in
hypoxic pulmonary vasoconstriction of the rat: identity of the hypoxic
sensor. J Physiol. 2001;536:211–224.
29. Weir EK, Archer SL. The mechanism of acute hypoxic pulmonary vasoconstriction: the tale of two channels. FASEB J. 1995;9:183–189.
30. Archer SL, Souil E, Dinh-Xuan AT, Schremmer B, Mercier JC, El
Yaagoubi A, Nguyen-Huu L, Reeve HL, Hampl V. Molecular identification of the role of voltage-gated K⫹ channels, Kv1.5 and Kv2.1, in
hypoxic pulmonary vasoconstriction and control of resting membrane
potential in rat pulmonary artery myocytes. J Clin Invest. 1998;101:
2319 –2330.
31. Archer SL, Weir EK, Reeve HL, Michelakis E. Molecular identification
of O2 sensors and O2-sensitive potassium channels in the pulmonary
circulation. Adv Exp Med Biol. 2000;475:219 –240.
32. Moudgil R, Michelakis ED, Archer SL. Hypoxic pulmonary vasoconstriction. J Appl Physiol. 2005;98:390 – 403.
33. Morio Y, McMurtry IF. Ca2⫹ release from ryanodine-sensitive store
contributes to mechanism of hypoxic vasoconstriction in rat lungs. J Appl
Physiol. 2002;92:527–534.
34. Dipp M, Evans AM. Cyclic ADP-ribose is the primary trigger for hypoxic
pulmonary vasoconstriction in the rat lung in situ. Circ Res. 2001;89:
77– 83.
35. Dipp M, Nye PC, Evans AM. Hypoxic release of calcium from the
sarcoplasmic reticulum of pulmonary artery smooth muscle. Am J Physiol
Lung Cell Mol Physiol. 2001;281:L318 –L325.
36. Eu JP, Xu L, Stamler JS, Meissner G. Regulation of ryanodine
receptors by reactive nitrogen species. Biochem Pharmacol. 1999;57:
1079 –1084.
37. Evans AM, Dipp M. Hypoxic pulmonary vasoconstriction: cyclic adenosine diphosphate-ribose, smooth muscle Ca2⫹ stores and the endothelium. Respir Physiolo Neurobiol. 2002;132:3–15.
38. Kumasaka S, Shoji H, Okabe E. Novel mechanisms involved in
superoxide anion radical-triggered Ca2⫹ release from cardiac sarcoplasmic reticulum linked to cyclic ADP-ribose stimulation. Antioxid
Redox Signal. 1999;1:55– 69.
39. Snetkov VA, Aaronson PI, Ward JP, Knock GA, Robertson TP.
Capacitative calcium entry as a pulmonary specific vasoconstrictor
mechanism in small muscular arteries of the rat. Br J Pharmacol.
2003;140:97–106.
40. Wang J, Shimoda LA, Sylvester JT. Capacitative calcium entry and
TRPC channel proteins are expressed in rat distal pulmonary arterial
smooth muscle. Am J Physiol Lung Cell Mol Physiol. 2004;286:
L848 –L858.
41. Sweeney M, Yuan JX. Hypoxic pulmonary vasoconstriction: role of
voltage-gated potassium channels. Respir Res. 2000;1:40 – 48.
42. Robertson TP, Hague D, Aaronson PI, Ward JP. Voltage-independent
calcium entry in hypoxic pulmonary vasoconstriction of intrapulmonary
arteries of the rat. J Physiol. 2000;525:669 – 680.
43. Gonzalez-Pacheco FR, Caramelo C, Castilla MA, Deudero JJ, Arias J,
Yague S, Jimenez S, Bragado R, Alvarez-Arroyo MV. Mechanism of
vascular smooth muscle cells activation by hydrogen peroxide: role of
phospholipase C gamma. Nephrol Dial Transplant. 2002;17:392–398.
44. Weissmann N, Dietrich A, Fuchs B, Kalwa H, Ay M, Dumitrascu R,
Olschewski A, Storch U, Mederos y Schnitzler M, Ghofrani HA,
Schermuly RT, Pinkenburg O, Seeger W, Grimminger F, Gudermann
T. Classical transient receptor potential channel 6 (TRPC6) is essential
for hypoxic pulmonary vasoconstriction and alveolar gas exchange.
Proc Natl Acad Sci U S A. 2006;103:19093–19098.
45. Saitoh S, Zhang C, Tune JD, Potter B, Kiyooka T, Rogers PA, Knudson
JD, Dick GM, Swafford A, Chilian WM. Hydrogen peroxide: a feedforward dilator that couples myocardial metabolism to coronary blood
flow. Arterioscler Thromb Vasc Biol. 2006;26:2614 –2621.
46. Lopez-Barneo J, Pardal R, Montoro RJ, Smani T, Garcia-Hirschfeld J,
Urena J. K⫹ and Ca2⫹ channel activity and cytosolic [Ca2⫹] in oxygensensing tissues. Respir Physiol. 1999;115:215–227.
47. Vanden Hoek TL, Becker LB, Shao Z, Li C, Schumacker PT. Reactive
oxygen species released from mitochondria during brief hypoxia
induce preconditioning in cardiomyocytes. J Biol Chem. 1998;273:
18092–18098.
48. Guzy RD, Sharma B, Bell E, Chandel NS, Schumacker PT. Loss of the
SdhB, but not the SdhA, subunit of complex II triggers reactive oxygen
species-dependent hypoxia-inducible factor activation and tumorigenesis.
Mol Cell Biol. 2008;28:718 –731.
49. Mansfield KD, Guzy RD, Pan Y, Young RM, Cash TP, Schumacker PT,
Simon MC. Mitochondrial dysfunction resulting from loss of cytochrome
c impairs cellular oxygen sensing and hypoxic HIF-alpha activation. Cell
Metab. 2005;1:393–399.
50. Liu JQ, Sham JS, Shimoda LA, Kuppusamy P, Sylvester JT. Hypoxic constriction and reactive oxygen species in porcine distal pulmonary arteries. Am J Physiol Lung Cell Mol Physiol. 2003;285:
L322–L333.
51. Pastukh V, Ruchko M, Gorodnya O, Wilson GL, Gillespie MN.
Sequence-specific oxidative base modifications in hypoxia-inducible
genes. Free Radic Biol Med. 2007;43:1616 –1626.
52. Lavani R, Chang WT, Anderson T, Shao ZH, Wojcik KR, Li CQ,
Pietrowski R, Beiser DG, Idris AH, Hamann KJ, Becker LB, Vanden
Waypa et al
Hoek TL. Altering CO2 during reperfusion of ischemic cardiomyocytes
modifies mitochondrial oxidant injury. Crit Care Med. 2007;35:
1709 –1716.
53. Gusarova GA, Dada LA, Kelly AM, Brodie C, Witters LA, Chandel NS,
Sznajder JI. Alpha1-AMP-activated protein kinase regulates hypoxia-
Compartmental Redox Responses to Hypoxia
535
induced Na, K-ATPase endocytosis via direct phosphorylation of protein
kinase C zeta. Mol Cell Biol. 2009;29:3455–3464.
54. Duranteau J, Chandel NS, Kulisz A, Shao Z, Schumacker PT. Intracellular signaling by reactive oxygen species during hypoxia in cardiomyocytes. J Biol Chem. 1998;273:11619 –11624.
Novelty and Significance
What Is Known?
●
●
●
Downloaded from http://circres.ahajournals.org/ by guest on June 18, 2017
The mitochondria of pulmonary arterial smooth muscle cells function
as the O2 sensor underlying HPV by triggering a change in ROS
signaling.
Conflicting reports from various groups have shown that mitochondrial ROS signaling either increases or decreases in response to
hypoxia.
Attempts to resolve the question of whether ROS levels increase or
decrease during hypoxia have been hindered by the technical
limitations of the tools used to assess changes in oxidant stress.
What New Information Does This Article Contribute?
●
●
●
A novel, redox-sensitive, ratiometric fluorescent protein sensor,
RoGFP, was used to assess the effects of hypoxia on redox
signaling in cultured pulmonary artery smooth muscle cells.
For the first time, genetic targeting sequences were used to express
RoGFP in the cytosol, the mitochondrial matrix, and the mitochondrial intermembrane space, allowing assessment of oxidant
signaling in distinct subcellular compartments.
The results are consistent with a model in which hypoxia induces the
release of ROS from the outer surface of the inner mitochondrial
membrane, thereby eliciting an increase in oxidant signaling in the
intermembrane space. This ROS signal subsequently diffuses into the
cytosol, where it elicits an increase in Ca2⫹, thereby activating HPV.
In the lung, alveolar hypoxia triggers constriction of small
pulmonary arteries in a response known as hypoxic pulmonary
vasoconstriction (HPV). Although this response has been studied
extensively, the underlying mechanism of cellular oxygen sensing is not established and the signaling pathways that link the O2
sensor to the contractile response have been a focus of debate.
Our model proposes that hypoxia triggers the release of reactive
oxygen species (ROS) signals from the outer surface of the inner
mitochondrial membrane; these oxidants initially affect the
intermembrane space and subsequently enter the cytosol where
they trigger increases in cytosolic calcium. To test, this we
measured redox status in the mitochondrial matrix, the intermembrane space, and the cytosol of pulmonary artery myocytes
during hypoxia, using targeted expression of a thiol redoxsensitive fluorescent protein sensor, RoGFP. Consistent with
the model, we detected increases ROS signaling in the
intermembrane space and the cytosol during hypoxia,
whereas oxidant stress in the mitochondrial matrix decreased. The increase in cytosolic ROS signaling is then
required for the subsequent increase in ionized calcium.
These findings reveal that hypoxia activates compartmentalized subcellular redox signaling, and that increases in
cytosolic ROS are required for activating HPV.
Downloaded from http://circres.ahajournals.org/ by guest on June 18, 2017
Hypoxia Triggers Subcellular Compartmental Redox Signaling in Vascular Smooth
Muscle Cells
Gregory B. Waypa, Jeremy D. Marks, Robert Guzy, Paul T. Mungai, Jacqueline Schriewer,
Danijela Dokic and Paul T. Schumacker
Circ Res. 2010;106:526-535; originally published online December 17, 2009;
doi: 10.1161/CIRCRESAHA.109.206334
Circulation Research is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Copyright © 2009 American Heart Association, Inc. All rights reserved.
Print ISSN: 0009-7330. Online ISSN: 1524-4571
The online version of this article, along with updated information and services, is located on the
World Wide Web at:
http://circres.ahajournals.org/content/106/3/526
Data Supplement (unedited) at:
http://circres.ahajournals.org/content/suppl/2009/12/17/CIRCRESAHA.109.206334.DC1
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published
in Circulation Research 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 Research is online at:
http://circres.ahajournals.org//subscriptions/