Molecular Cardiology

Hypoxia Induces Myocardial Regeneration in Zebrafish

Chris Jopling, PhD; Guillermo Suñé, PhD; Adèle Faucherre, PhD;
Carme Fabregat; Juan Carlos Izpisua Belmonte, PhD

Background—Hypoxia plays an important role in many biological/pathological processes. In particular, hypoxia is associated
with cardiac ischemia. which, although initially inducing a protective response, will ultimately lead to the death of
cardiomyocytes and loss of tissue, severely affecting cardiac functionality. Although myocardial damage/loss remains an
insurmountable problem for adult mammals, the same is not true for adult zebrafish, which are able to completely regenerate
their heart after extensive injury. Myocardial regeneration in zebrafish involves the dedifferentiation and proliferation of
cardiomyocytes to replace the damaged/missing tissue; at present, however, little is known about what factors regulate
this process.
Methods and Results—We surmised that ventricular amputation would lead to hypoxia induction in the myocardium of
zebrafish and that this may play a role in regulating the regeneration of the missing cardiac tissue. Using a combination
Downloaded from http://circ.ahajournals.org/ by guest on February 19, 2018

of O2 perturbation, conditional transgenics, in vitro cell culture, and microarray analysis, we found that hypoxia induces
cardiomyocytes to dedifferentiate and proliferate during heart regeneration in zebrafish and have identified a number of
genes that could play a role in this process.
Conclusion—These results indicate that hypoxia plays a positive role during heart regeneration, which should be taken into
account in future strategies aimed at inducing heart regeneration in humans. (Circulation. 2012;126:3017-3027.)
Key Words: cardiac myocyte dedifferentiation 䡲 cardiac myocyte hypoxia 䡲 cardiac myocyte proliferation
䡲 regeneration zebrafish

H ypoxia has been found to regulate a number of cellular

processes that are also associated with successful heart
regeneration in zebrafish such as cellular dedifferentiation
and proliferation.1–5 The family of transcription factors,
hypoxia-inducible factors (HIFs), are the direct effectors of
the hypoxic response. HIF-induced transcription involves the
formation of a dimer between either HIF1␣ or HIF2␣ and
HIF1␤ to subsequently target a plethora of downstream
genes.6 HIF1/2␣ transcriptional activity is regulated primarily
by their level of expression, which, under normoxic condi-
tions, remains relatively low but will be rapidly elevated
when hypoxia ensues. The level of HIF1/2␣ expression is
maintained by the prolyl hydroxylase dioxygenase family of
proteins. Under normoxic conditions, prolyl hydroxylase
dioxygenase enzymes hydroxylate HIF1/2␣, which subse-
quently binds to the von Hippel Lindau E3 ligase and is
targeted for proteasomal degradation. During periods of
hypoxia, the lack of oxygen inhibits prolyl hydroxylase
dioxygenase activity; subsequently, the expression levels of
HIF1/2␣ become elevated.6
Clinical Perspective on p 3027
Cardiac ischemia is typically caused by an occlusion of the
coronary vasculature. The resulting reduction in the blood
supply to the tissue leads to a drastic decrease in oxygen
(hypoxia).5 Consequently, HIF1␣ is no longer targeted for
degradation, and its expression levels rise sharply, allowing it
to dimerize with the HIF1␤ and activate an array of down-
stream targets. Initially, this response serves to preserve the
cardiomyocytes by inducing a range of cardioprotective
genes that allow the cells to survive in the hypoxic environ-
ment that ensues after an ischemic episode.7 Ultimately,
however, prolonged exposure to hypoxic conditions will
result in apoptosis and necrosis and the subsequent loss of
cardiac tissue. Several studies have examined the conse-
quences of perturbing HIF1␣ expression levels in cardiomyo-
cytes. For example, transgenic mice overexpressing HIF1␣ in
the myocardium show a reduction in the amount of damage
caused by cardiac ischemia and show a marked improvement
in cardiac function.8 Similarly, reducing prolyl hydroxylase

Received March 27, 2012; accepted November 1, 2012.

From the Center of Regenerative Medicine in Barcelona, Barcelona, Spain (C.J., G.S., C.F., J.C.I.B.); Salk Institute for Biological Studies, La Jolla,
CA (J.C.I.B.); and CNRS, UMR-5203, Institut de Génomique Fonctionnelle, Molecular Mechanisms of Regeneration, LabEx ICST, Département de
Physiologie, INSERM, U661, Universités de Montpellier 1 and 2, Montpellier, France (C.J., A.F.).
The online-only Data Supplement is available with this article at http://circ.ahajournals.org/lookup/suppl/doi:10.1161/CIRCULATIONAHA.
112.107888/-/DC1.
Correspondence to Juan Carlos Izpisua Belmonte, PhD, The Salk Institute for Biological Studies, 10010 N Torrey Pines Rd, La Jolla, CA 92037. E-mail
belmonte@salk.edu/izpisua@cmrb.edu
© 2012 American Heart Association, Inc.
Circulation is available at http://circ.ahajournals.org DOI: 10.1161/CIRCULATIONAHA.112.107888

3017
 3018 Circulation December 18/25, 2012
dioxygenase expression levels by siRNA-mediated knock-
down results in elevated HIF1␣ expression and consequently
a similar beneficial response to that seen in HIF1␣-
overexpressing mice.9
Hypoxia has also been found to play a role in promoting the
proliferation of a variety of mammalian cell types such as
embryonic stem cells,2 smooth muscle cells,10 endothelial
cells,11 and fibroblasts.12 Furthermore, in adult zebrafish, expo-
sure to chronic hypoxia leads to an increased number of
cardiomyocytes in the heart.13 Because ventricular amputation
damages the myocardium and impairs heart function, we hy-
pothesized that it will lead to hypoxia induction and that this may
play a role in regulating heart regeneration in adult zebrafish.
Methods
Constructs and Zebrafish Lines
All constructs and transgenic lines were generated with the Tol2 Kit
as described.14,15 Dominant-negative (dn) zf HIF1␣b was generated
Downloaded from http://circ.ahajournals.org/ by guest on February 19, 2018
as described.16 For the Tg(cmlc2a:LrL:dnHIF1␣) construct, the 5⬘
entry clone cmlc2a contained a 1-kb polymerase chain reaction
fragment of the cmcl2a promoter,17 the middle entry clone contained
a floxed red fluorescent protein stop cassette amplified from pBOB-
LRL-CBReGFPpA (a kind gift from Geoff Whal), and the 3⬘ entry
clone contained zebrafish dnHIF1␣b.
Zebrafish Oxygen Conditions
For maintaining adult zebrafish in hyperoxic conditions, 3 fish were
placed in a transport bag (SERA) with 500 mL aquaria water; 100%
O2 was used to fill the bag before it was tightly sealed with a rubber
band. Sealed bags were then placed in an incubator and maintained
at 28°C. Each subsequent day for the duration of the experiment, the
water was changed; during this time, the fish were also fed (a
procedure lasting ⬇1 hour) before the bag was refilled with 100% O2
and sealed. The average O2 concentration in the water was 17.21
mg/L (213% saturation; average calculated from 5 consecutive
days).
For maintaining adult zebrafish in normoxic conditions (and to
ensure that any differences observed between normoxia and hyper-
oxia were not due to the way the zebrafish were maintained), we
followed the exact same procedure as for the hyperoxic conditions,
but instead of adding 100% O2, we left the bag open and not sealed.
The average O2 concentration in the water was 6.22 mg/L (77%
saturation; average calculated from 5 consecutive days).
Phenylhydrazine Treatment
For induction of anemic hypoxia, 5 adult fish were placed in a
crossing box containing 1 mg phenylhydrazine (Sigma) per liter of
aquarium water for 1 hour and then returned to the aquarium to
recover for 3 days. This procedure was then repeated before
amputation.
Hypoxyprobe Treatment
Hypoxyprobe was dissolved in PBS at a concentration of 5 mg/mL.
For in vitro cell culture, 50 ␮L was added to 3 mL culture medium.
For in vivo analysis, 20 ␮L was injected intraperitoneally each
day for the duration of the experiment.
Cre/Tamoxifen Induction and Heart Amputation
Transgenes were expressed by inducing Cre-mediated recombination
with tamoxifen as described.15 All amputations were performed as
described.15
BrdU Labeling
BrdU treatment was performed essentially as described.18 Meta-
morph software (Molecular Devices) was used to count BrdU-
labeled cardiomyocytes.
Cardiomyocyte Isolation
Fish were euthanized in tricaine. Hearts were collected and put in
PBS with penicillin, streptomycin, and 10 U/mL heparin. The
outflow tracts were then removed and ventricles and atriums were
opened to get rid of the blood. They were then washed 3 times in
perfusion buffer (PBS; 10 mmol/L HEPES, 30 mmol/L taurine,
5.5 mmol/L glucose, and 10 mmol/L 2,3-butanedione monoxime
[BDM]) and placed in digestion buffer (perfusion buffer plus
12.5 ␮mol/L calcium chloride and collagenase II and IV, 5 mg each
[Gibco]) to digest for 2 hours at 32°C in a thermomixer at 800 rpm.
Next, an equal volume of stop buffer 1 (perfusion buffer plus
12.5 ␮mol/L calcium chloride and 10% FBS) was added, and cells
were mechanically separated. Undigested material was left to sedi-
ment, and cells suspended in the supernatant were pelleted by
centrifugation at 250g for 5 minutes. The cells were then resus-
pended in stop buffer 2 (perfusion buffer plus 12.5 ␮mol/L and 5%
FBS), and calcium reintroduction was performed by gradually
raising the concentration to 62, 112, 212, 500, and 1000 ␮mol/L.
Cells were then pelleted again and resuspended in plating medium
(minimum essential medium; 5% FBS, 10 mmol/L BDM, 2 mmol/L
Glutamax, 100 U/mL penicillin, and 100 ␮g/mL streptomycin). Cell
preparations were plated onto collagen type I (BD Biosciences;
reference No. 354236) –treated chamber slides (Nunc; reference No.
177445) and allowed to attach overnight.
Cardiomyocyte Culture Oxygen Conditions
For maintaining adult zebrafish cardiomyocytes in hyperoxic condi-
tions, cultures were placed in a hypoxia chamber (Stem Cell
Technologies) filled with 100% O2. Each subsequent day, the
chamber was flushed and filled with 100% O2.
For maintaining adult zebrafish cardiomyocytes in hypoxic con-
ditions, cultures were placed in a hypoxia chamber (Stem Cell
Technologies) filled with 3% O2. Each subsequent day, the chamber
was flushed and filled with 3% O2.
For maintaining adult zebrafish cardiomyocytes in normoxic
conditions, cultures were placed in a standard cell culture incubator
at 21% O2 and 5% CO2.
Immunohistochemistry
Immunohistochemistry was performed on 10-␮m cryosections as
previously described.19 The antibodies used are anti– green fluores-
cent protein (GFP; GFP-1020, AVES), anti-BrdU (sc-70441, Santa
Cruz), anti–phospho-histone H3 (PHH3; 06-570, Upstate), anti–␣-
sarcomeric actin (ASA; A2172, Sigma), anti-tropomyosin (TPM;
T2780, Sigma), anti-MEF2c (ab79436, Abcam), and Hypoxyprobe
Mab1 (Hypoxyprobe-1 kit).
In Situ Hybridization
In situ hybridization was performed as described previously.18 The
dnHIF1␣ probe was generated by subcloning dnHIF1␣ into the Pst1
and Xba1 sites of pBSK-.
Confocal Microscopy
Confocal microscopy was performed with a Leica SP5. For colocal-
ization analysis, the following formula was used to calculate the
axial resolution:
Dz⫽公[(l⫻n/NA2)⫹( AU⫻n⫻公2⫻1.22⫻l/NA2)2],
where emission (l)⫽500 nm, refractive index⫽1.518, NA⫽1.4, and
AU (airy units)⫽1. This results in a section thickness of
z⫽0.773 ␮m.
Statistical Analysis
BrdU-Positive Cardiomyocyte Assay
For each condition, 5 animals were analyzed. For each animal,
BrdU-positive cardiomyocytes were counted on 3 separate sections,
and the average was then calculated. Multiple-sample statistical
analysis was performed by 1-way ANOVA with the Holm-Sidak
 Jopling et al Hypoxia Induces Myocardial Regeneration in Zebrafish
multiple-comparisons test using the average number of BrdU-
positive cardiomyocytes per animal.
PHH3 Cardiomyocyte Assay
For each condition, 3 experiments were performed. For each exper-
iment, 105 cardiomyocytes were analyzed. The number of PHH3-
positive cardiomyocytes was subsequently calculated for each ex-
periment. Multiple-sample statistical analysis was performed by
1-way ANOVA with the Holm-Sidak multiple-comparisons test
using the number of PHH3-positive cardiomyocytes per experiment.
Dedifferentiated Cardiomyocyte Assay
For each condition, 6 experiments were performed. For each exper-
iment, 5 random nonoverlapping images were analyzed. The average
percentage of dedifferentiated cardiomyocytes was subsequently
calculated for each experiment. Multiple-sample statistical analysis
was performed by 1-way ANOVA with the Holm-Sidak multiple-
comparisons test using the average number of dedifferentiated
cardiomyocytes per experiment.
Microarray and Quantitative Polymerase Chain
Reaction Analysis
Downloaded from http://circ.ahajournals.org/ by guest on February 19, 2018
A detailed description of the microarray and quantitative poly-
merase chain reaction analysis is provided in the online-only Data
Supplement.
Results
Ventricular Amputation Induces Hypoxia in
Cardiac Tissue
To determine whether ventricular amputation induces hypox-
ia in adult zebrafish, we first established a protocol for
detecting hypoxia using the commercially available chemical
Hypoxyprobe as an indirect measure of tissue oxygen content
(Pimonidazole binds to thiol groups at low oxygen tension,
which can subsequently be detected with immunohistochem-
istry).20 To establish that this system was functioning in
zebrafish, we cultured adult zebrafish cardiomyocytes in
either normoxic or hypoxic conditions and treated the culture
with Hypoxyprobe for 24 hours. We were unable to detect
any Hypoxyprobe signal in cardiomyocytes cultured under
normoxic culture (Figure 1A–1C). However, we were able to
detect a robust Hypoxyprobe signal in cardiomyocytes cul-
tured under hypoxic conditions (Figure 1D and 1E). This
indicates that the Hypoxyprobe system functions reliably in
zebrafish cells.
To determine whether hypoxia is a feature of heart regen-
eration, we treated unamputated adult zebrafish with Hypoxy-
probe for a period of 7 days, during which time they were
maintained under normoxic conditions. Subsequently, we
were unable to detect any Hypoxyprobe signal in the hearts of
these animals (Figure 1G–1I). For hypoxia, we tested a
number of different techniques (data not shown) and found
that the most reliable and reproducible results were obtained
by inducing anemia in adult zebrafish with the chemical
phenylhydrazine21 (It is well established that anemia, a lack
of red blood cells, leads to hypoxia22). Consequently, after
anemia induction, adult zebrafish were treated with Hypoxy-
probe for a period of 7 days. Immunohistochemical analysis
revealed a robust Hypoxyprobe signal in the hearts of all the
anemic zebrafish we analyzed (n⫽5), indicating that
phenylhydrazine-induced anemia leads to hypoxia induction
in the heart (Figure 1J–1L). We next sought to determine
whether ventricular amputation also leads to hypoxia induc-
3019
tion. After ventricular amputation, adult zebrafish were
treated with Hypoxyprobe throughout the 7-day postamputa-
tion period. We were able to detect a strong Hypoxyprobe
signal throughout the heart, which was at its most intense in
both the fibrin clot, similar to a previous report,23 and the
cardiomyocytes/cells close to the site of amputation (n⫽5;
Figure 1M–1O). To ensure the validity of using Hypoxy-
probe, we also amputated adult zebrafish and maintained
them in hyperoxic conditions throughout the 7-day recovery
period. Consequently, we were unable to detect any Hypoxy-
probe signal in the hearts of these animals (Figure IA–IC in
the online-only Data Supplement). These results indicate that
ventricular amputation of adult zebrafish leads to hypoxia
induction in cardiac tissue.
Next, we analyzed what effect perturbation of normoxic
conditions has on heart regeneration. To achieve this, we
adopted several strategies that involved either modulating the
O2 concentration or using transgenic zebrafish. Because
HIF1␣ is the direct effector of hypoxia signaling, we gener-
ated transgenic lines in which we could specifically and
conditionally express dnHIF1␣16,24,25 in cardiomyocytes us-
ing the Cre/tamoxifen system we have described previously.15
The transgenic zebrafish were generated by crossing the
previously described Tg(cmlc2a:Ert2-Cre-Ert2/cmlc2a:LnL:
GFP) line15 with a Tg(cmlc2a:LrL:dnHIF1␣) line that contains a
floxed red fluorescent protein stop cassette (LrL) between the
cardiomyocyte-specific promoter (cmlc2a) and dnHIF1␣. Treat-
ment of embryos or adult zebrafish with 4-hydroxytamoxifen
results in efficient recombination of the floxed stop cassette
and subsequently a rapid, cardiomyocyte-specific induction
of dnHIF1␣. To establish that our system was functional, we
induced dnHIF1␣ expression in embryos by administering
4-hydroxytamoxifen and monitored them throughout early
development (up to 5 days after fertilization). In situ hybrid-
ization using a dnHIF1␣ antisense probe revealed a strong
expression of dnHIF1␣ in the hearts of 3dpf embryos treated
with 4-hydroxytamoxifen (Figure IID and IIE in the online-
only Data Supplement). Although untreated embryos naturally
express HIF1␣ in the brain, we could not detect any endogenous
signal in the heart (Figure IIA and IIB in the online-only Data
Supplement). Furthermore, we were unable to detect any ob-
servable defects in cardiogenesis between wild-type embryos
and embryos expressing dnHIF1␣ in their cardiomyocytes (Fig-
ure IIC and IIF in the online-only Data Supplement).
Hypoxia Positively Regulates Heart Regeneration
in Adult Zebrafish
To establish what effect perturbation of the normoxic
conditions has on regeneration, we amputated adult ze-
brafish and maintained them in normoxic conditions, hypoxic
conditions (phenylhydrazine-induced anemia), or hyperoxic
conditions. In parallel, we also induced cardiomyocyte-
specific expression of dnHIF1␣ in adult Tg(cmlc2a:Ert2-Cre-
Ert2/cmlc2a:LnL:GFP)(cmlc2a:LrL:dnHIF1␣) transgenic ze-
brafish before ventricular amputation. To establish a baseline,
we first amputated wild-type zebrafish and maintained them
under normoxic conditions throughout the postamputation
period. Subsequently, we were able to detect a robust regen-
erative response after 14 days (Figure 2A and 2B). We
 3020 Circulation December 18/25, 2012

A B C

Figure 1. Hypoxyprobe analysis of adult

Normoxia
D E
Downloaded from http://circ.ahajournals.org/ by guest on February 19, 2018
Normoxia Normoxia zebrafish cardiomyocytes in vitro and in
vivo. Adult zebrafish cardiomyocytes cul-
tured under normoxic conditions were
F immunolabeled for tropomyosin (green;
B) and Hypoxyprobe (red; C). A, A
merged image with tropomyosin (green)
and Hypoxyprobe (red). Adult zebrafish
cardiomyocytes cultured under hypoxic
conditions were immunolabeled for tro-
pomyosin (green; E) and Hypoxyprobe
(red; F). D, A merged image with tropo-
myosin (green) and Hypoxyprobe (red). A

Hypoxia Hypoxia Hypoxia section from a wild-type adult zebrafish

heart maintained under normoxic condi-
G H I tions immunolabeled for ␣-sarcomeric
actin (green), DAPI (blue), and Hypoxy-
probe (red; G). A higher-magnification
image of the same heart with
␣-sarcomeric actin (green), DAPI (blue),
and Hypoxyprobe (red; H). I, The same
image in H with only Hypoxyprobe (red).
A section from a wild-type adult
zebrafish heart maintained under hypoxi-
c(anemia) conditions immunolabeled for
Normo ia
Normoxia Normo ia
Normoxia Normo ia
Normoxia ␣-sarcomeric actin (green), DAPI (blue),
and Hypoxyprobe (red; J). A higher-
J K L magnification image of the same heart
with ␣-sarcomeric actin (green), DAPI
(blue), and Hypoxyprobe (red; K). L, The
same image in K with only Hypoxyprobe
(red). A section from a wild-type adult
zebrafish heart 7 days after amputation
immunolabeled for ␣-sarcomeric actin
(green), DAPI (blue), and Hypoxyprobe
(red; M). A higher-magnification image of
the same heart with ␣-sarcomeric actin
Hypoxia Hypoxia Hypoxia (green), DAPI (blue), and Hypoxyprobe
(red; N). O, The same image in N with
M N O only Hypoxyprobe (red). DAPI indicates
4=-6=-diamidino-2-phénylindole.

7dpa 7dpa 7dpa

confirmed this by BrdU labeling the regenerating animals and
analyzing the number of BrdU-positive cardiomyocytes (Fig-
ure 2I and Figure IIIA and IIIB in the online-only Data
Supplement). In this manner, we were able to detect numer-
ous BrdU-positive cardiomyocytes in the hearts of regener-
ating zebrafish. Next, we preconditioned adult zebrafish by
treating them with phenylhydrazine to induce anemia/hypox-
ia 7 days before amputation. These hypoxic zebrafish showed
an even more profound regenerative response (Figure 2C and
2D) that was associated with a significantly increased number
of BrdU-positive cardiomyocytes (Figure 2I). Conversely,
adult zebrafish maintained under hyperoxic conditions failed
to regenerate their hearts substantially after 14 days (Figure
2E and 2F) and had significantly fewer BrdU-positive car-
diomyocytes than the normoxic controls (Figure 2I). Simi-
larly, adult Tg(cmlc2a:Ert2-Cre-Ert2/cmlc2a:LnL:GFP)
(cmlc2a:LrL:dnHIF1␣) transgenic zebrafish, in which cardio-
myocyte-specific expression of dnHIF1␣ had been induced
before amputation, also failed to regenerate their hearts
(Figure 2G and 2H), resulting in significantly fewer BrdU-
positive cardiomyocytes (Figure 2I). As an additional control,
we repeated these experiments using an anti-MEF2c antibody
 Jopling et al Hypoxia Induces Myocardial Regeneration in Zebrafish 3021

A B C D

14dpa Normoxia Normoxia 14dpa Hypoxia Hypoxia

E F I
*

14dpa Hyperoxia Hyperoxia

G H
Downloaded from http://circ.ahajournals.org/ by guest on February 19, 2018

14dpa dnHif1α dnHif1α

Figure 2. Perturbing normoxia effects heart regeneration in adult zebrafish. A section from a 14-day postamputation (dpa) regenerating
wild-type adult zebrafish heart maintained under normoxic conditions immunolabeled for ␣-sarcomeric actin (green), BrdU (red), and
DAPI (blue). The white dashed line indicates the plane of amputation (A). The same heart at higher magnification (B). A section from a
14-day postamputation regenerating wild-type adult zebrafish heart maintained under hyperoxic conditions immunolabeled for
␣-sarcomeric actin (green), BrdU (red), and DAPI (blue). The white dashed line indicates the plane of amputation (C). The same heart at
higher magnification (D). A section from a 14-day postamputation regenerating wild-type adult zebrafish heart maintained under
hypoxic conditions immunolabeled for ␣-sarcomeric actin (green), BrdU (red), and DAPI (blue). The white dashed line indicates the
plane of amputation (E). The same heart at higher magnification (F). A section from a 14-day postamputation regenerating
Tg(cmlc2a:Ert2-Cre-Ert2/cmlc2a:LnL:GFP)(cmlc2a:LrL:dnHIF1␣) adult zebrafish heart immunolabeled for ␣-sarcomeric actin (green),
BrdU (red), and DAPI (blue). The white dashed line indicates the plane of amputation (G). The same heart at higher magnification (H). I,
A graph representing the average number of BrdU-positive cardiomyocytes per section in each of the 4 conditions⫾SEM. dnHIF1␣
indicates dominant-negative hypoxia-inducible factor-1␣. P⬍0.0001, ANOVA. *P⬍0.05, Holm-Sidak test. DAPI indicates 4=,6=-diamidino-2-
phénylindole; and BrdU, 5-bromo-2=-deoxyuridine.

(which labels cardiomyocyte nuclei) and obtained similar results
(Figure IVA–IVG in the online-only Data Supplement). These
findings were further confirmed by repeating the experiments
and allowing a 30-day recovery period. At this stage, regenera-
tion is virtually complete; thus, amputated hearts taken from fish
maintained under normoxic conditions appeared virtually indis-
tinguishable from unamputated controls (Figure 3A and 3B).
However, adult zebrafish, both those maintained in hyperoxic
conditions and Tg(cmlc2a:Ert2-Cre-Ert2/cmlc2a:LnL:GFP)
(cmlc2a:LrL:dnHIF1␣) transgenic zebrafish, failed to regenerate
normally with obvious areas of missing myocardium (Figure
3C–3F). These results indicate that hypoxia positively regulates
heart regeneration in adult zebrafish and furthermore that this is
achieved via HIF1␣.
Hypoxia Positively Regulates
Cardiomyocyte Proliferation
To elucidate the effect that hypoxia has on cardiomyocytes, we
examined these cells in vitro to determine how they responded to
changes in environmental oxygen. Under normoxic conditions,
we established that adult zebrafish cardiomyocytes, unlike adult
mammalian cardiomyocytes, proliferate as evidenced by numer-
ous PHH3-positive cardiomyocytes (46.6 PHH3-positive cardio-
myocytes per 105 cells), which is substantially more than
reported for adult mammalian cardiomyocytes26 (Figure 4A–
4C). Having established this basal proliferation rate, we cultured
adult zebrafish cardiomyocytes in either hyperoxic or hypoxic
conditions. Under hyperoxic conditions, we found that the
number of PHH3-positive cardiomyocytes was significantly
reduced compared with normoxia (4.1 PHH3-positive cardio-
myocytes per 105 cells; Figure 4A– 4C). However, under hy-
poxic conditions, we observed a significant increase in the
number of PHH3-positive cardiomyocytes compared with the
control normoxic conditions (59.3 PHH3-positive cardiomyo-
cytes per 105 cells; Figure 4A-4C). This indicates that hypoxia
positively regulates the proliferation of adult zebrafish
cardiomyocytes.
Hypoxia Induces Cardiomyocytes
to Dedifferentiate
We have previously reported that zebrafish cardiomyocytes
dedifferentiate during heart regeneration, a process involving
 3022 Circulation December 18/25, 2012
A B
30dpa Normoxia 30dpa Normoxia
C D
Downloaded from http://circ.ahajournals.org/ by guest on February 19, 2018
30dpa Hyperoxia Hyperoxia
E F
30dpa dnHif1 dnHif1
Figure 3. Perturbing normoxia affects heart regeneration in adult
zebrafish. A section from a 30-day postamputation (dpa) regen-
erating wild-type adult zebrafish heart maintained under nor-
moxic conditions immunolabeled for ␣-sarcomeric actin (green),
BrdU (red), and DAPI (blue). The white dashed line indicates the
plane of amputation (A). The same heart at higher magnification
(B). A section from a 14-day postamputation regenerating wild-
type adult zebrafish heart maintained under hyperoxic condi-
tions immunolabeled for ␣-sarcomeric actin (green), BrdU (red),
and DAPI (blue). The white dashed line indicates the plane of
amputation (C). The same heart at higher magnification (D). A
section from a 14-day postamputation regenerating
Tg(cmlc2a:Ert2-Cre-Ert2/cmlc2a:LnL:GFP)(cmlc2a:LrL:dnHIF1␣)
adult zebrafish heart immunolabeled for ␣-sarcomeric actin
(green), BrdU (red), and DAPI (blue). The white dashed line indi-
cates the plane of amputation (E). The same heart at higher
magnification (F). dnHIF1␣ indicates dominant-negative hypoxia-
inducible factor-1␣; DAPI, 4=,6=-diamidino-2-phénylindole; and
BrdU, 5-bromo-2=-deoxyuridine.
the disassembly of the sarcomere.15 This phenomenon has
also been reported in mammalian cardiomyocytes both in
vitro and in vivo; however, there is little or no proliferation/
regeneration associated with this process in mammals.26 –29
Although mammalian cardiomyocytes do not proliferate nat-
urally, they can be stimulated to divide in vitro.30 To
proliferate, the adult cells first need to dedifferentiate, which
involves disassembly of the sarcomere and re-expression of
fetal genes.26 In vitro and under normoxic conditions, we also
readily observed dedifferentiation of zebrafish cardiomyo-
cytes (similar to a previously published report31). Immuno-
histochemical analysis of differentiated zebrafish cardiomyo-
cytes using anti-ASA and anti-TPM antibodies reveals a
typical organized/striated sarcomere (Figure 5A–5C). How-
ever, dedifferentiated cardiomyocytes, in which the sarco-
mere has been disassembled, display reduced/disorganized
ASA expression and loss of z-disk striations26 yet still retain
a clear TPM signal (Figure 5D–5F). This difference between
ASA and TPM allowed us to quantify the number of
dedifferentiated cardiomyocytes (differentiated⫽ASA⫹/
TPM⫹; dedifferentiated⫽ASA⫺/TPM⫹; Figure 5G). In this
manner, we were able to determine that 46% of zebrafish
cardiomyocytes were dedifferentiated after 7 days of culture
under normoxic conditions (Figure 5H). Surprisingly, when
zebrafish cardiomyocytes were cultured under hypoxic con-
ditions, we observed a significant increase in the number of
dedifferentiated cardiomyocytes (75%; Figure 5H). Further-
more, if zebrafish cardiomyocytes were cultured under hy-
peroxic conditions, then the number of dedifferentiated cells
dropped significantly (28%; Figure 5H). These results indi-
cate that the environmental oxygen concentration plays a
significant role in regulating cardiomyocyte dedifferentiation.
This also explains the differences in the amount of prolifer-
ating zebrafish cardiomyocytes that we observed under the
different oxygen conditions in vitro. Because the cardiomyo-
cytes need to dedifferentiate in order to proliferate, enhancing
the amount of dedifferentiation will undoubtedly increase the
number of proliferating cells.
Mechanisms of Hypoxia During
Heart Regeneration
To understand more about how hypoxia regulates zebrafish
heart regeneration, we performed microarray analysis to
identify genes that could play a role in this process. We
decided to examine only the differences in gene expression
between wild-type and Tg(cmlc2a:Ert2-Cre-Ert2/cmlc2a:LnL:
GFP)(cmlc2a:LrL:dnHIF1␣) transgenic zebrafish because
this would provide a cardiomyocyte-enriched list of genes
as a result of the restricted expression of dnHIF1␣. To
further refine the identification of genes involved in
zebrafish heart regeneration, we decided to focus only on
7-day postamputation zebrafish to avoid the initial wound/
inflammatory response.31 To this end, we amputated both
wild-type and Tg(cmlc2a:Ert2-Cre-Ert2/cmlc2a:LnL:GFP)
(cmlc2a:LrL:dnHIF1␣) zebrafish, allowed them to regenerate
for 7 days, and then compared their transcriptomic profile
with those of unamputated wild-type and Tg(cmlc2a:Ert2-
Cre-Ert2/cmlc2a:LnL:GFP)(cmlc2a:LrL:dnHIF1␣) zebrafish,
respectively. Our analyses focused on genes that were not
significantly different between the wild-type and dnHIF1␣
in the unamputated samples (P⬍⫽0.05 and logFCⱕ1 or
logFCⱖ⫺1) but were subsequently found to be either
significantly differentially expressed between unampu-
tated and 7-day postamputation wild-type zebrafish
(P⬍⫽0.05 and logFCⱖ1 or logFCⱕ⫺1) but showed no
significant change in expression between unamputated and
7-day postamputation Tg(cmlc2a:Ert2-Cre-Ert2/cmlc2a:LnL:
GFP)(cmlc2a:LrL:dnHIF1 ␣ ) zebrafish (logFCⱕ0.5 or
logFCⱖ⫺0.5) or vice versa because this would identify
putative HIF1␣-regulated genes. In this manner, we were able
to identify 114 differentially expressed genes in the wild-type
versus dnHIF1␣ samples (Tables I and II in the online-only
 Jopling et al Hypoxia Induces Myocardial Regeneration in Zebrafish
A Prophase
TRPM/PH3
B Anaphase C histone H3 (PHH3; red), and DAPI (blue). C, A
* phospho-histone H3–positive cardiomyocytes per
PHH3
Downloaded from http://circ.ahajournals.org/ by guest on February 19, 2018
TRPM/PH3
Merge
Data Supplement). Furthermore, we were also able to detect
another 20 genes that were differentially expressed between
the dnHIF1␣ and wild-type samples (Tables III and IV in the
online-only Data Supplement). To confirm the validity of our
microarray data, we independently performed real-time quan-
titative polymerase chain reaction analysis on a subset of
differentially expressed genes and confirmed that the changes
in gene expression were the same (Figure V in the online-
only Data Supplement). Of the 114 genes found to be
differentially expressed in the wild-type versus dnHIF1␣
sample, 102 of these (55 annotated and 47 unannotated) are
upregulated and most likely represent genes that are either
directly or indirectly positively regulated by HIF1␣ (Table I
in the online-only Data Supplement). Within this group, we
were able to find genes that are known to be induced by
hypoxia (the Table), including several that are directly
regulated by HIF1, namely jak2a, mcl1a, cdkn1b, cxcr4b, and
gata1a.6,32– 40 Furthermore, we were able to identify genes
within the upregulated group that have a heart-associated
function (the Table) such as thbs4b, jak2a, txnipa, pim1,
slc4a1a, and cxcr4b.39,41– 45 A number of proproliferative
genes were also shown to be upregulated in the wild-type
sample compared with the dnHIF1␣ sample (the Table).46 –51
Although the positive aspects of dedifferentiation have not
been intensely investigated, we were able to identify a couple
of genes that have been linked to this phenomenon, namely
jak2a and cxcr4b, albeit these are most commonly associated
with cancer cell dedifferentiation.52,53 Additionally, we able
to identify numerous components of the Jak-STAT signaling
pathway (the Table). Jak2a expression is significantly in-
creased in the wild-type sample but shows no significant
change in the dnHIF1␣ sample; stat2 expression also was
increased in the wild-type sample. Although this was just
below our stringency (logFC⫽0.99), we were able to confirm
the change in expression via quantitative polymerase chain
reaction (Figure V in the online-only Data Supplement).
3023
Figure 4. Hypoxia increases the number of prolif-
erating adult zebrafish cardiomyocytes in vitro. A,
A prophase adult zebrafish cardiomyocyte immu-
nolabeled for tropomyosin (TRPM; green),
phospho-histone H3 (PH3; red), and DAPI (blue).
B, An anaphase adult zebrafish cardiomyocyte
immunolabeled for tropomyosin (green), phospho-
graph representing the average number of
105 cells⫾ SEM. P⬍0.0001, ANOVA. *P⬍0.05,
Holm-Sidak test. DAPI indicates 4=,6=-diamidino-2-
phénylindole.
Conversely, we also detected downregulation of a negative
regulator of Jak-STAT signaling, cish (or socs),54 which,
although just below our level of stringency in the dnHIF1␣
sample (logFC⫽⫺0.51), was dramatically reduced in the
wild-type sample (logFC⫽⫺2.18). Upstream of the Jak-
STAT pathway, cxcr4b55 was found to be upregulated in the
wild-type sample but showed no change in the dnHIF1␣
sample (Table I in the online-only Data Supplement), al-
though a putative downstream target of the Jak-STAT path-
way, pim1,56 is upregulated in the wild-type sample
(logFC⫽1.44) but does not change in the dnHIF1␣ sample
(logFC⫽0.32; Table I in the online-only Data Supplement).
Taken together, these results indicate that numerous genes are
regulated by hypoxia to positively influence the regenerative
process. Furthermore, our analyses have uncovered various
components of the Jak-STAT pathway.
Discussion
Here, we have shown that ventricular amputation in zebrafish
leads to rapid and robust induction of hypoxia in the heart and
that, by inhibiting this process either through hyperoxia or by
expressing dnHIF1␣, we can effectively block regeneration.
Conversely, preconditioning adult zebrafish with hypoxia
(via anemia) produces a more profound regenerative re-
sponse. In mammals, cardiac ischemia also results in hypoxia
induction and elevated levels of HIF1␣.57 However, mam-
mals fail to mount any significant kind of regenerative
response, which suggests that whatever factors are missing or
are inhibiting this process are downstream of the hypoxic
response. Furthermore, our data indicate that hyperoxic con-
ditions inhibit the regenerative process in zebrafish. In
humans, current therapeutic strategies aim to reduce the
hypoxia associated with cardiac ischemia.58 Although this
undoubtedly has a beneficial effect in the immediate after-
math of a cardiac ischemia/infarction, prolonged treatment
may obscure any favorable response associated with hypoxia.
 3024 Circulation December 18/25, 2012
A D
ASA ASA
B E
Downloaded from http://circ.ahajournals.org/ by guest on February 19, 2018
TPM TPM
C F
Merge Merge
G H
* *
* *
*
*
Merge
We have also shown here that hypoxia increases the amount
of dedifferentiated zebrafish cardiomyocytes in vitro, a pro-
cess involving disassembly of the sarcomere, which would
otherwise physically impede cell division. It is not unusual to
find hypoxia playing a role in regulating cellular dedifferen-
tiation. Indeed, hypoxia is well known to be able to induce
dedifferentiation in a wide variety of cell types such as
adipocytes,59 renal cells,1 and astrocytes.60 Hypoxia is also
notorious for inducing tumors to dedifferentiate and subse-
Figure 5. Hypoxia increases the number of dedif-
ferentiated adult zebrafish cardiomyocytes in vitro.
Differentiated adult zebrafish cardiomyocytes
immunolabeled for ␣-sarcomeric actin (ASA; green;
A) and tropomyosin (TPM; red; B). C, A merged
image with ␣-sarcomeric actin (green), tropomyo-
sin (red), and DAPI (blue). Dedifferentiated adult
zebrafish cardiomyocytes immunolabeled for
␣-sarcomeric actin (green; D) and tropomyosin
(red; E). F, A merged image with ␣-sarcomeric
actin (green), tropomyosin (red), and DAPI (blue).
G, A lower-magnification merged image of adult
zebrafish cardiomyocytes in vitro with
␣-sarcomeric actin (green), tropomyosin (red), and
DAPI (blue). H, A graph representing the average
percent of dedifferentiated cardiomyocytes in 3
conditions⫾SEM. P⬍0.0001, ANOVA. *P⬍0.05,
Holm-Sidak test. DAPI indicates 4=,6=-diamidino-2-
phénylindole.
quently to become more aggressive61 (Hypoxyprobe is most
commonly used in tumor diagnostics). In mammals, 1 feature
often associated with cardiac ischemia (or indeed hypoxia) is
cardiomyocyte dedifferentiation (also called hibernating
myocardium), which involves the induction of fetal gene
expression accompanied by a disassembly of the sarcomeric
structure.29,62 Although this is widely regarded as being part
of the cardioprotective program, other reports in which adult
mammalian cardiomyocytes have been induced to proliferate
 Jopling et al Hypoxia Induces Myocardial Regeneration in Zebrafish 3025

Table. Functional Groups of Differentially Expressed Genes is blocking heart regeneration from occurring in mammals
Gene Identification Gene LogFC Reference
(whether it be a lack of specific factors or an active inhibition
of this process) is downstream of the hypoxic response and of
Hypoxia induced
the dedifferentiation process. Future strategies aimed at trying
30307 jak2a 1.44 32, 33 to induce myocardial regeneration in mammals should take
497283 hif1al2 2.06 6 into account that hypoxia signaling appears to be a key
58122 mcl1a 1.01 34 positive component of this process.
368359 txnipa 1.15 35
58054 pim1 1.44 36 Sources of Funding
Dr Faucherre is supported by the Fondation pour la Recherche Médicale
368329 cdkn1b 1.06 37, 38
postdoctoral fellowship. This work was supported by INSERM ATIP-
114447 cxcr4b 1.41 39 AVENIR, MINECO (PLE2009-0147), Subprograma Juan de la Cierva
30481 gata1a 1.08 40 (JCI-2008-2808), Fondo Social Europeo, ISCIII, Tercel (RD06/0010/
0016), FEDER, GENCAT, Fundacion Cellex, The IPSEN Foundation,
Heart associated The Leona M. and Harry B. Helmsley Charitable Trust, and The G.
252850 thbs4b 1.94 41 Harold and Leila Y. Mathers Charitable Foundation.
30307 jak2a 1.44 54
368359 txnipa 1.15 43 Disclosures
None.
58054 pim1 1.44 42
Downloaded from http://circ.ahajournals.org/ by guest on February 19, 2018

84703 slc4a1a 2.33 44 References

114447 cxcr4b 1.41 45 1. Sahai A, Mei C, Schrier RW, Tannen RL. Mechanisms of chronic hyp-
oxia-induced renal cell growth. Kidney Int. 1999;56:1277–1281.
Proliferation
2. Forristal CE, Wright KL, Hanley NA, Oreffo RO, Houghton FD. Hypoxia
30307 jak2a 1.44 46 inducible factors regulate pluripotency and proliferation in human
58054 pim1 1.44 47 embryonic stem cells cultured at reduced oxygen tensions. Reproduction.
2010;139:85–97.
114447 cxcr4b 1.41 48 3. Voss MJ, Niggemann B, Zanker KS, Entschladen F. Tumour reactions to
30766 tal1 1.01 49 hypoxia. Curr Mol Med. 2010;10:381–386.
4. Jogi A, Ora I, Nilsson H, Poellinger L, Axelson H, Pahlman S. Hypox-
321283 tacc3 1 50 ia-induced dedifferentiation in neuroblastoma cells. Cancer Lett. 2003;
30481 gata1a 1.08 51 197:145–150.
5. Cassavaugh J, Lounsbury KM. Hypoxia-mediated biological control.
Dedifferentiation
J Cell Biochem. 2011;112:735–744.
30307 jak2a 1.44 52 6. Semenza GL. Hypoxia-inducible factors in physiology and medicine.
114447 cxcr4b 1.41 53 Cell. 2012;148:399 – 408.
7. Santos CX, Anilkumar N, Zhang M, Brewer AC, Shah AM. Redox
signaling in cardiac myocytes. Free Radic Biol Med. 2011;50:777–793.
8. Kido M, Du L, Sullivan CC, Li X, Deutsch R, Jamieson SW, This-
indicate that dedifferentiation is a necessary prelude to tlethwaite PA. Hypoxia-inducible factor 1-alpha reduces infarction and
proliferation.26 Mechanistically, our microarray analyses has attenuates progression of cardiac dysfunction after myocardial infarction
in the mouse. J Am Coll Cardiol. 2005;46:2116 –2124.
uncovered numerous genes that are potentially involved in
9. Natarajan R, Salloum FN, Fisher BJ, Ownby ED, Kukreja RC, Fowler
hypoxia-mediated heart regeneration. We found that pim1 AA 3rd. Activation of hypoxia-inducible factor-1 via prolyl-4
expression was increased during heart regeneration in wild- hydoxylase-2 gene silencing attenuates acute inflammatory responses in
type zebrafish; however, this effect was subdued in the postischemic myocardium. Am J Physiol Heart Circ Physiol. 2007;293:
H1571–H1580.
presence of dnHIF1␣. Pim1 is a serine/threonine kinase; its 10. Cogo A, Napolitano G, Michoud MC, Barbon DR, Ward M, Martin JG.
classification as an oncogene is indicative of it proprolifera- Effects of hypoxia on rat airway smooth muscle cell proliferation. J Appl
tive capacity.56 Interestingly, pim1 has previously been Physiol. 2003;94:1403–1409.
11. Li W, Petrimpol M, Molle KD, Hall MN, Battegay EJ, Humar R. Hyp-
shown to be capable of inducing adult mammalian cardio- oxia-induced endothelial proliferation requires both mTORC1 and
myocytes to proliferate.42,63 Furthermore, we have uncovered mTORC2. Cir Res. 2007;100:79 – 87.
what appears to be an entire signaling pathway. Numerous 12. Mizuno S, Bogaard HJ, Voelkel NF, Umeda Y, Kadowaki M, Ameshima
components of the Jak-STAT pathway are shown to be S, Miyamori I, Ishizaki T. Hypoxia regulates human lung fibroblast
proliferation via p53-dependent and -independent pathways. Respir Res.
differentially expressed in our microarray analyses. More- 2009;10:17.
over, a recent report has indicated that HIF1␣-induced acti- 13. Marques IJ, Leito JT, Spaink HP, Testerink J, Jaspers RT, Witte F, van
vation of the Jak-STAT pathway induces pulmonary arterial den Berg S, Bagowski CP. Transcriptome analysis of the response to
chronic constant hypoxia in zebrafish hearts. J Comp Physiol B. 2008;
smooth muscle cells to dedifferentiate and proliferate.32 178:77–92.
14. Kwan KM, Fujimoto E, Grabher C, Mangum BD, Hardy ME, Campbell
Conclusions DS, Parant JM, Yost HJ, Kanki JP, Chien CB. The tol2kit: a multisite
gateway-based construction kit for tol2 transposon transgenesis con-
It appears then that the hypoxia associated with both ischemia structs. Dev Dyn. 2007;236:3088 –3099.
in mammals and, as we have shown here, ventricular ampu- 15. Jopling C, Sleep E, Raya M, Marti M, Raya A, Belmonte JC. Zebrafish
tation in zebrafish induces cardiomyocytes to dedifferentiate, heart regeneration occurs by cardiomyocyte dedifferentiation and prolif-
the obvious difference being that adult zebrafish cardiomyo- eration. Nature. 2010;464:606 – 609.
16. Elks PM, van Eeden FJ, Dixon G, Wang X, Reyes-Aldasoro CC, Ingham
cytes are able to subsequently proliferate whereas adult PW, Whyte MK, Walmsley SR, Renshaw SA. Activation of hypoxia-
mammalian cardiomyocytes cannot. It is likely that whatever inducible factor-1alpha (HIF-1alpha) delays inflammation resolution by
 3026 Circulation December 18/25, 2012
reducing neutrophil apoptosis and reverse migration in a zebrafish inflam-
mation model. Blood. 2011;118:712–722.
17. Huang CJ, Tu CT, Hsiao CD, Hsieh FJ, Tsai HJ. Germ-line transmission
of a myocardium-specific GFP transgene reveals critical regulatory
elements in the cardiac myosin light chain 2 promoter of zebrafish. Dev
Dyn. 2003;228:30 – 40.
18. Raya A, Koth CM, Buscher D, Kawakami Y, Itoh T, Raya RM, Sternik
G, Tsai HJ, Rodriguez-Esteban C, Izpisua-Belmonte JC. Activation of
notch signaling pathway precedes heart regeneration in zebrafish. Proc
Natl Acad Sci U S A. 2003;100(suppl 1):11889 –11895.
19. Poss KD, Wilson LG, Keating MT. Heart regeneration in zebrafish.
Science. 2002;298:2188 –2190.
20. Rosenberger C, Rosen S, Paliege A, Heyman SN. Pimonidazole adduct
immunohistochemistry in the rat kidney: detection of tissue hypoxia.
Methods Mol Biol. 2009;466:161–174.
21. Sun X, Hoage T, Bai P, Ding Y, Chen Z, Zhang R, Huang W, Jahangir
A, Paw B, Li YG, Xu X. Cardiac hypertrophy involves both myocyte
hypertrophy and hyperplasia in anemic zebrafish. PloS One. 2009;4:
e6596.
22. Cafaro RP. Hypoxia: Its causes and symptoms. J Am Dent Soc Anesthesiol.
1960;7:4–8.
23. Lokmic Z, Darby IA, Thompson EW, Mitchell GM. Time course analysis
of hypoxia, granulation tissue and blood vessel growth, and remodeling in
Downloaded from http://circ.ahajournals.org/ by guest on February 19, 2018
healing rat cutaneous incisional primary intention wounds. Wound Repair
Regen. 2006;14:277–288.
24. Jiang BH, Rue E, Wang GL, Roe R, Semenza GL. Dimerization, DNA
binding, and transactivation properties of hypoxia-inducible factor 1.
J Biol Chem. 1996;271:17771–17778.
25. Woodley DT, Fan J, Cheng CF, Li Y, Chen M, Bu G, Li W. Participation
of the lipoprotein receptor LRP1 in hypoxia-HSP90alpha autocrine sig-
naling to promote keratinocyte migration. J Cell Sci. 2009;122:
1495–1498.
26. Engel FB, Schebesta M, Duong MT, Lu G, Ren S, Madwed JB, Jiang H,
Wang Y, Keating MT. P38 MAP kinase inhibition enables proliferation
of adult mammalian cardiomyocytes. Genes Dev. 2005;19:1175–1187.
27. Stephen MJ, Poindexter BJ, Moolman JA, Sheikh-Hamad D, Bick RJ. Do
binucleate cardiomyocytes have a role in myocardial repair? Insights
using isolated rodent myocytes and cell culture. Open Cardiovasc Med J
2009;3:1–7.
28. Wijns W, Vatner SF, Camici PG. Hibernating myocardium. N Engl
J Med. 1998;339:173–181.
29. Ausma J, Thone F, Dispersyn GD, Flameng W, Vanoverschelde JL,
Ramaekers FC, Borgers M. Dedifferentiated cardiomyocytes from
chronic hibernating myocardium are ischemia-tolerant. Mol Cell
Biochem. 1998;186:159 –168.
30. Bicknell KA, Coxon CH, Brooks G. Can the cardiomyocyte cell cycle be
reprogrammed? J Cell Mol Med. 2007;42:706 –721.
31. Lien CL, Schebesta M, Makino S, Weber GJ, Keating MT. Gene
expression analysis of zebrafish heart regeneration. PLoS Biol. 2006;
4:e260.
32. Liu T, Li Y, Lin K, Yin H, He B, Zheng M, Wang G. Regulation of
s100a4 expression via the JAK2-STAT3 pathway in rhomboid-phenotype
pulmonary arterial smooth muscle cells exposure to hypoxia. Int
J Biochem Cell Biol. 2012;44:1337–1345.
33. Wang GS, Qian GS, Zhou DS, Zhao JQ. JAK-STAT signaling pathway
in pulmonary arterial smooth muscle cells is activated by hypoxia. Cell
Biol Int. 2005;29:598 – 603.
34. Iqbal S, Zhang S, Driss A, Liu ZR, Kim HR, Wang Y, Ritenour C, Zhau
HE, Kucuk O, Chung LW, Wu D. PDGF upregulates Mcl-1 through
activation of beta-catenin and HIF-1alpha-dependent signaling in human
prostate cancer cells. PloS one. 2012;7:e30764.
35. Le Jan S, Le Meur N, Cazes A, Philippe J, Le Cunff M, Leger J, Corvol
P, Germain S. Characterization of the expression of the hypoxia-induced
genes neuritin, TXNIP and IGFBP3 in cancer. FEBS Lett. 2006;580:
3395–3400.
36. Chen J, Kobayashi M, Darmanin S, Qiao Y, Gully C, Zhao R, Kondo S,
Wang H, Yeung SC, Lee MH. Hypoxia-mediated up-regulation of PIM-1
contributes to solid tumor formation. Am J Pathol. 2009;175:400 – 411.
37. Pillai MS, Sapna S, Shivakumar K. P38 MAPK regulates G1-S transition
in hypoxic cardiac fibroblasts. Int J Biochem Cell Biol. 2011;43:919 –927.
38. Hackenbeck T, Knaup KX, Schietke R, Schodel J, Willam C, Wu X,
Warnecke C, Eckardt KU, Wiesener MS. HIF-1 or HIF-2 induction is
sufficient to achieve cell cycle arrest in nih3t3 mouse fibroblasts inde-
pendent from hypoxia. Cell Cycle. 2009;8:1386 –1395.
39. Schioppa T, Uranchimeg B, Saccani A, Biswas SK, Doni A, Rapisarda A,
Bernasconi S, Saccani S, Nebuloni M, Vago L, Mantovani A, Melillo G,
Sica A. Regulation of the chemokine receptor cxcr4 by hypoxia. J Exp
Med. 2003;198:1391–1402.
40. Zhang FL, Shen GM, Liu XL, Wang F, Zhao YZ, Zhang JW. Hypoxia-
inducible factor 1-mediated human GATA1 induction promotes erythroid
differentiation under hypoxic conditions. J Cell Mol Med. 2012;16:
1889 –1899.
41. Frolova EG, Sopko N, Blech L, Popovic ZB, Li J, Vasanji A, Drumm C,
Krukovets I, Jain MK, Penn MS, Plow EF, Stenina OI.
Thrombospondin-4 regulates fibrosis and remodeling of the myocardium
in response to pressure overload. FASEB J. 2012;26:2363–2373.
42. Cottage CT, Bailey B, Fischer KM, Avitable D, Collins B, Tuck S,
Quijada P, Gude N, Alvarez R, Muraski J, Sussman MA. Cardiac pro-
genitor cell cycling stimulated by pim-1 kinase. Circulation research.
2010;106:891–901.
43. Yoshioka J, Schulze PC, Cupesi M, Sylvan JD, MacGillivray C, Gannon
J, Huang H, Lee RT. Thioredoxin-interacting protein controls cardiac
hypertrophy through regulation of thioredoxin activity. Circulation. 2004;
109:2581–2586.
44. Alvarez BV, Kieller DM, Quon AL, Robertson M, Casey JR. Cardiac
hypertrophy in anion exchanger 1-null mutant mice with severe hemolytic
anemia. Am J Physiol Heart Circ Physiol. 2007;292:H1301–H1312.
45. Penn MS, Pastore J, Miller T, Aras R. SDF-1 in myocardial repair. Gene
Ther. 2012;19:583–587.
46. Jakel H, Weinl C, Hengst L. Phosphorylation of p27Kip1 by JAK2
directly links cytokine receptor signaling to cell cycle control. Oncogene.
2011;30:3502–3512.
47. Merkel AL, Meggers E, Ocker M. PIM1 kinase as a target for cancer
therapy. Exp Opin Investig Drugs. 2012;21:425– 436.
48. Teicher BA, Fricker SP. CXCL12 (SDF-1)/CXCR4 pathway in cancer.
Clin Cancer Res. 2010;16:2927–2931.
49. Dey S, Curtis DJ, Jane SM, Brandt SJ. The Tal1l transcription factor
regulates cell cycle progression and proliferation in differentiating murine
bone marrow monocyte precursors. Mol Cell Biol. 2010;30:2181–2192.
50. Schneider L, Essmann F, Kletke A, Rio P, Hanenberg H, Wetzel W,
Schulze-Osthoff K, Nurnberg B, Piekorz RP. The transforming acidic
coiled coil 3 protein is essential for spindle-dependent chromosome
alignment and mitotic survival. J Biol Chem. 2007;282:29273–29283.
51. Zheng J, Kitajima K, Sakai E, Kimura T, Minegishi N, Yamamoto M,
Nakano T. Differential effects of GATA-1 on proliferation and differen-
tiation of erythroid lineage cells. Blood. 2006;107:520 –527.
52. Chang KT, Tsai CM, Chiou YC, Chiu CH, Jeng KS, Huang CY. IL-6
induces neuroendocrine dedifferentiation and cell proliferation in
non-small cell lung cancer cells. Am J Physiol Lung Cell Mol Physiol.
2005;289:L446 –L453.
53. Wehler TC, Graf C, Biesterfeld S, Brenner W, Schadt J, Gockel I, Berger
MR, Thuroff JW, Galle PR, Moehler M, Schimanski CC. Strong
expression of chemokine receptor CXCR4 by renal cell carcinoma cor-
relates with advanced disease. J Oncol. 2008;2008:626340.
54. Kurdi M, Booz GW. JAK redux: a second look at the regulation and role
of JAKs in the heart. Am J Physiol Heart Circ Physiol. 2009;297:
H1545–H1556.
55. Vila-Coro AJ, Rodriguez-Frade JM, Martin De Ana A, Moreno-Ortiz
MC, Martinez AC, Mellado M. The chemokine SDF-1alpha triggers
CXCR4 receptor dimerization and activates the JAK/STAT pathway.
FASEB J. 1999;13:1699 –1710.
56. Wang Z, Bhattacharya N, Weaver M, Petersen K, Meyer M, Gapter L,
Magnuson NS. PIM-1: a serine/threonine kinase with a role in cell
survival, proliferation, differentiation and tumorigenesis. J Vet Sci. 2001;
2:167–179.
57. Jurgensen JS, Rosenberger C, Wiesener MS, Warnecke C, Horstrup JH,
Grafe M, Philipp S, Griethe W, Maxwell PH, Frei U, Bachmann S,
Willenbrock R, Eckardt KU. Persistent induction of HIF-1alpha and
-2alpha in cardiomyocytes and stromal cells of ischemic myocardium.
FASEB J. 2004;18:1415–1417.
58. Wijesinghe M, Perrin K, Ranchord A, Simmonds M, Weatherall M,
Beasley R. Routine use of oxygen in the treatment of myocardial infarc-
tion: systematic review. Heart. 2009;95:198 –202.
59. Liao YJ, Gao JH, Jiang P, Lu F. Effect of hypoxia on dedifferentiation of
mature adipocytes: an experimental study [in Chinese]. Nan Fang Yi Ke
Da Xue Xue Bao. 2008;28:339 –342.
 Jopling et al Hypoxia Induces Myocardial Regeneration in Zebrafish 3027

60. Schmidt-Ott KM, Xu AD, Tuschick S, Liefeldt L, Kresse W, Verkhratsky
A, Kettenmann H, Paul M. Hypoxia reverses dibutyryl-camp-induced
stellation of cultured astrocytes via activation of the endothelin system.
FASEB J. 2001;15:1227–1229.
61. Axelson H, Fredlund E, Ovenberger M, Landberg G, Pahlman S. Hyp-
oxia-induced dedifferentiation of tumor cells: a mechanism behind het-
erogeneity and aggressiveness of solid tumors. Semin Cell Dev Biol.
2005;16:554 –563.
62. Razeghi P, Essop MF, Huss JM, Abbasi S, Manga N, Taegtmeyer H.
Hypoxia-induced switches of myosin heavy chain iso-gene expression in
rat heart. Biochem Biophys Res Commun. 2003;303:1024 –1027.
63. Katare R, Caporali A, Zentilin L, Avolio E, Sala-Newby G, Oikawa A,
Cesselli D, Beltrami AP, Giacca M, Emanueli C, Madeddu P. Intravenous
gene therapy with PIM-1 via a cardiotropic viral vector halts the pro-
gression of diabetic cardiomyopathy through promotion of prosurvival
signaling. Circ Res. 2011;108:1238 –1251.

CLINICAL PERSPECTIVE
Cardiac ischemia leads to a reduction in oxygen supply and subsequent induction of hypoxia. At present, the consensus is
that the hypoxia associated with cardiac ischemia initially induces a cardioprotective response; however, prolonged
exposure ultimately leads to necrosis and loss of myocardial tissue. Treatment of patients who have suffered an ischemic
episode aims to reduce hypoxia. Our data now indicate that hypoxia could also play a beneficial role. We have found that
hypoxia positively regulates myocardial regeneration in zebrafish. We have also found that some of the processes that are
regulated by hypoxia during zebrafish heart regeneration have also been reported in mammals/humans after cardiac
ischemia. In particular, we have observed that hypoxia induces zebrafish cardiomyocytes to dedifferentiate, a process that
is also associated with cardiomyocytes in ischemic patients. Unfortunately, although they are capable of this initial
dedifferentiation, human cardiomyocytes fail to proliferate and subsequently regenerate the heart. Future strategies
Downloaded from http://circ.ahajournals.org/ by guest on February 19, 2018

attempting to regenerate a damaged human heart should take into account that hypoxia can play a beneficial role in
inducing a regenerative response.
 Hypoxia Induces Myocardial Regeneration in Zebrafish
Chris Jopling, Guillermo Suñé, Adèle Faucherre, Carme Fabregat and Juan Carlos Izpisua
Belmonte

Circulation. 2012;126:3017-3027; originally published online November 14, 2012;

Downloaded from http://circ.ahajournals.org/ by guest on February 19, 2018

doi: 10.1161/CIRCULATIONAHA.112.107888
Circulation is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Copyright © 2012 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/126/25/3017

Data Supplement (unedited) at:

http://circ.ahajournals.org/content/suppl/2012/11/14/CIRCULATIONAHA.112.107888.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/
SUPPLEMENTAL MATERIAL

Microarray analysis

RNA samples were obtained from 10 wildtype and 10 Tg(cmlc2a:Ert2-Cre-

Ert2/cmlc2a:LnL:GFP)(cmlc2a:LrL:dnHIF1α) unamputated zebrafish and 10 wildtype and 10

Tg(cmlc2a:Ert2-Cre-Ert2/cmlc2a:LnL:GFP)(cmlc2a:LrL:dnHIF1α) zebrafish at 7dpa. Total

heart RNA was extracted with Trizol (Invitrogen), purity and integrity of the RNA were

assessed by spectrophotometry and nanoelectrophoresis using the NanoDrop ND-1000

spectrophotometer (NanoDrop Technologies, Wilmington, DE) and the Nano lab-on-a-chip

assay for total eukaryotic RNA using Bioanalyzer 2100 (Agilent Technologies, Palo Alto,

CA), respectively. Only samples with high purity and high integrity (RNA integrity number

(RIN) >8.7) were subsequently used in microarray experiments. Microarray expression

profiles from these samples were obtained using the Affymetrix GeneChip® Zebrafish

Genome array (Affymetrix, Santa Clara, CA). Briefly, 200ng of total RNA from each sample

was processed, labelled and hybridized to GeneChip® Zebrafish Genome arrays according to

the Affymetrix manuals GeneChip® 3’ IVT Express Kit User Manual (P/N 702646 Rev.8)

and Expression Wash, Stain and Scan User Manual (P/N 702731 Rev. 3). Firstly, 200ng of

RNA were reverse-transcribed with T7 oligo(dT) primer to synthesize first-strand cDNA

containing a T7 promoter sequence. This cDNA was then converted into a double-stranded

DNA and subsequently used as a template to produce many copies of biotin-labeled cRNA

though an in-vitro transcription reaction. The cRNA was then purified, to remove

unincorporated NTPs, salts, enzymes and inorganic phosphate, and later fragmented. Finally,

12.5 µg of the fragmented and biotinylated cRNA was added to a hybridization cocktail,

loaded on a GeneChip® Zebrafish Genome array and hybridized for 16 hours at 45ºC and 60

rpm in an Affymetrix GeneChip® Oven 645. Following hybridization, the array was washed

and stained in the Affymetrix GeneChip® Fluidics.Station 450. The stained array was
scanned using an Affymetrix GeneChip® Scanner 3000 7G, generating CEL files for each

array. Microarray data analysis was performed as follows: After quality control of raw data, it

was background corrected, quantile-normalized and summarized to a probe set using the

robust multi-chip average (RMA) obtaining a total of 15617 probe sets. Linear Models for

Microarray (LIMMA), a moderated t-statistics model, was used for detecting differentially

expressed genes between the four conditions in study. Correction for multiple comparisons

was performed using false discovery rate. Genes with an adjusted p-value less than 0.05 and

with an absolute logFC>1 were selected as differentially expressed genes. All data analysis

was performed in R (version 2.11.1) with packages Biobase, Affy, limma and annotation

packages. All expression data was submitted to the NCBI Gene Expression Omnibus

(http://www.ncbi.nlm.nih.gov/geo). The series accession number is GSE39842.

Real-time quantitative RT-PCR

RNA samples were obtained from 5 wildtype and 5 Tg(cmlc2a:Ert2-Cre-

Ert2/cmlc2a:LnL:GFP)(cmlc2a:LrL:dnHIF1α) unamputated zebrafish and 5 wildtype and 5

Tg(cmlc2a:Ert2-Cre-Ert2/cmlc2a:LnL:GFP)(cmlc2a:LrL:dnHIF1α) zebrafish at 7dpa. Total

heart RNA was extracted with Trizol (Invitrogen) and RT-PCR was performed using the

Cloned AMV First-Strand cDNA Synthesis Kit (Invitrogen). For real-time quantitative PCR,

we used the SYBR Green I Master (Roche). Cycling parameters were as follows : 95ºC x 10

min, 45 cycles of 95ºC x 15 sec and 60ºC x 30 sec followed by 95ºCx 10 min. A measure of

2 ng of total RNA was used per reaction. A standard curve for β-actin was used for

normalization. Quantitative PCR was performed on a Roche LightCycler 480. Samples were

run in triplicate. The primer3 software (http://www.bioinformatics.nl/cgi-

bin/primer3plus/primer3plus.cgi) was used to design primers for short amplicons between 100
 and 200 bases, except for β-actin20. Every primers pair has been verified by Absolute

quantification/Fit points and by Tm calling using the LightCycler 480 software. The primers

used in this analysis are shown in table.2.

Table.2.

Gene Primer Sequence

Gene name Accession number
symbol (5’-3’)

β -actin beta-actin For CGAGCAGGAGATGGGAAC AF057040

Rev CAACGGTTTCGCTCATTGC
stat2 signal transducer and activator of For AGGGCTTCAGAAAATGCTCA XM_688485
transcription 2 Rev CTCCGCTTCCAGTCTACCAG
cxcr4b chemokine (C-X-C motif), receptor 4b For GCTGCCTGGTTTGATCATTT NM_131834
Rev GGATGACGGTGGTCTTCAGT
cish cytokine inducible SH2-containing For ATGCCTCAGGTTGGTACTGG NM_001076617
protein Rev ATGTAGTCAGGGTGGCTGCT
fstl1b follistatin-like 1b For GTTGTTCCTGATGGCTGGTT NM_001039621
Rev GTGCTGGATGAACTTGAGCA
mibp muscle-specific beta 1 integrin binding For AAAACCACCCTGACGAACAG NM_131693
protein Rev TGTTCACCATTGCCTCCATA
cebpb CCAAT/enhancer binding protein For GTATGCAAGCAGCCAGTCAA NM_131884
(C/EBP), beta Rev GTACTGGGGCAAAGAGTCCA
Supplemental  table  1

Locus  ID Gene  Symbol Affy  ID logFC t-­‐stat p-­‐val Adj,  p-­‐val
252850 thbs4b Dr.4572.1.S1_at 1,94 12,68 0 0
553330 npc1 Dr.4869.1.A1_at 1,18 11,82 0 0
571355 ampd2 Dr.3596.1.A1_at 1,44 10,91 0 0
352931 sepp1a Dr.1330.1.S1_at 1,22 10,72 0 0
561082 ptrfb Dr.2168.1.A1_at 1,28 10,59 0 0
327462 hsd3b7 Dr.10542.1.S1_at 1,14 10,43 0 0
563972 gstt1a Dr.14058.1.A1_at 1,54 10,15 0 0
493607 net1 Dr.23465.1.S1_at 1,38 9,7 0 0
768177 deptor Dr.6362.1.A1_at 1,47 9,58 0 0
327288 cndp2 Dr.6571.1.S1_at 1,26 9,55 0 0
393130 reep3 Dr.17514.1.S1_at 1,2 9,18 0 0
558924 abca1a Dr.7365.1.A1_at 1,01 8,92 0 0
393146 slmapb Dr.9918.1.S1_at 1,34 8,53 0 0
394051 sh2d3cb Dr.17453.1.A1_at 1,15 8,51 0 0
322846 slc26a5 Dr.4948.1.A1_at 1,08 8,38 0 0
114466 tnfsf10l Dr.10736.1.S1_at 1,05 8,25 0 0
394146 dusp4 Dr.9304.1.S1_at 1,01 8,23 0 0
100151215 pde7a Dr.8750.1.A1_at 1,05 8,16 0 0
796731 slc25a43 Dr.1805.1.A1_at 1,63 8,1 0 0
327200 asb8 Dr.26472.1.S1_at 1,08 8,08 0 0
550475 ctsk Dr.4048.1.S1_at 1,03 8,03 0 0
64281 psmb9b Dr.8213.1.S1_at 2,57 7,57 0 0
560026 dsg2 Dr.24943.1.S1_at 1,03 7,36 0 0
337132 anxa5b Dr.20555.1.S2_s_at 1,63 7,08 0 0
58122 mcl1a Dr.4957.1.S1_at 1,01 7,05 0 0
323696 f11r Dr.4412.17.A1_at 1,08 6,95 0 0
566443 fxyd6l Dr.1942.1.A1_at 1,11 6,95 0 0
550571 fibinb Dr.159.1.A1_at 1,08 6,61 0 0
565402 col11a1a Dr.3536.1.A1_at 1,03 6,54 0 0
30286 ndrg3a Dr.20517.1.S1_at 1,18 6,3 0 0
30307 jak2a Dr.4151.1.S1_at 1,44 6,25 0 0
447810 nt5c2l1 Dr.3959.1.A1_at 2,34 6,01 0 0
431765 nfil3-­‐6 Dr.9849.1.A1_at 1,61 5,9 0 0
497283 hif1al2 Dr.25766.1.A1_at 2,06 5,75 0 0
373102 mcl1b Dr.25133.1.S1_at 1,23 5,67 0 0
406458 slc20a1a Dr.23911.1.A1_at 2,06 5,62 0 0
326287 epb41 Dr.3027.1.S1_at 1,99 5,41 0 0
566325 thbs2b Dr.25095.1.A1_at 1,11 5,36 0 0
335722 dfna5 Dr.21797.1.S1_at 1,49 5,23 0 0
368359 txnipa Dr.18396.1.S1_at 1,15 5,18 0 0
58054 pim1 Dr.6191.1.S1_at 1,44 5,14 0 0
436959 blvrb Dr.5399.1.S1_at 1,8 4,82 0 0
415146 hprt1l Dr.15574.1.A1_at 1,38 4,76 0 0
30331 cahz Dr.450.1.S1_at 1,87 4,59 0 0,01
84703 slc4a1a Dr.20971.1.S2_at 2,33 4,55 0 0,01
368329 cdkn1b Dr.9024.1.A1_at 1,06 4,49 0 0,01
 114447 cxcr4b Dr.6798.1.S1_at 1,41 4,25 0 0,01
30104 klfd Dr.12569.1.S1_at 1,65 4,17 0 0,01
30766 tal1 Dr.1458.1.S1_at 1,01 4,13 0 0,01
373093 hdr DrAffx.1.39.S1_at 1,03 4,03 0 0,01
321283 tacc3 Dr.3179.1.A1_at 1 4,01 0 0,01
100144628 shmt2 Dr.24150.1.A1_at 1,21 3,79 0 0,02
30481 gata1a Dr.355.1.S1_at 1,08 3,52 0 0,02
387290 smox Dr.11707.2.A1_at 1,02 6,9 0 0
393221 slc43a2a Dr.13039.1.S1_at 1,21 5,02 0 0
492647 si:rp71-­‐39b20.7
Dr.12817.1.A1_at 1,26 3,33 0,01 0,03
Dr.25422.1.S1_s_at 1,24 3,32 0,01 0,03
Dr.11214.1.A1_at 1,01 3,73 0 0,02
445317 zgc:101000 Dr.20334.1.S1_at 1,81 3,68 0 0,02
378866 zgc:64114 Dr.13972.1.S1_at 1,54 3,63 0 0,02
406286 zgc:56095 Dr.1398.1.S1_at 1,15 3,62 0 0,02
572469 zgc:153997 Dr.25850.2.A1_at 1,06 4,12 0 0,01
558746 wu:fj58g06 Dr.23743.1.S1_at 1,45 4,1 0 0,01
Dr.2623.1.A1_at 1,21 4,3 0 0,01
Dr.11205.1.A1_at 1,05 4,58 0 0,01
445037 zgc:92880 Dr.6751.1.S1_at 2,44 4,68 0 0,01
553723 zgc:112384 Dr.13067.1.S1_at 1,53 4,68 0 0,01
799483 LOC799483 Dr.24179.1.A1_at 1,07 4,77 0 0
323636 wu:fc05g01 Dr.5111.1.A1_at 1,27 4,84 0 0
Dr.25874.1.A1_at 2,09 5,37 0 0
Dr.21275.1.A1_at 1,45 5,59 0 0
Dr.17111.1.A1_at 1,21 5,58 0 0
Dr.25957.1.A1_at 1,06 5,47 0 0
386792 zgc:66317 Dr.8670.1.A1_at 1,12 5,43 0 0
555357 si:ch211-­‐197g15.10
Dr.16166.1.A1_at 1,93 5,77 0 0
327304 wu:fd20h09 Dr.15343.1.A1_at 1,1 6,15 0 0
335667 wu:fk54d01 Dr.10592.1.A1_at 1,69 6,11 0 0
Dr.16496.1.A1_at 2,41 6,25 0 0
336380 si:dkey-­‐242g16.2
Dr.1911.1.A1_at 1,74 6,43 0 0
566329 si:ch211-­‐11c15.3
Dr.12559.1.S1_at 1,49 6,4 0 0
100148818 LOC100148818Dr.20839.1.A1_x_at 2,68 6,7 0 0
619255 zgc:114181 Dr.5167.1.A1_at 1,02 6,78 0 0
Dr.21541.1.A1_at 2,26 7,03 0 0
Dr.22905.1.A1_at 1,07 7,3 0 0
557472 wu:fj55b04 Dr.18171.1.S1_at 1,37 7,53 0 0
Dr.7870.1.A1_at 1,36 7,94 0 0
557472 wu:fj55b04 Dr.13462.1.A1_at 1,16 7,87 0 0
100307097 si:ch73-­‐110p20.1
Dr.1380.1.A1_at 1,06 7,89 0 0
368719 si:busm1-­‐116l04.2
Dr.12951.1.A1_x_at 1,46 7,81 0 0
100148818 LOC100148818Dr.20839.1.A1_at 2,77 7,72 0 0
100037361 zgc:158463 Dr.24285.1.A1_at 1,44 7,65 0 0
Dr.5234.1.A1_at 1,37 8,1 0 0
559078 si:dkey-­‐238c7.16
Dr.2595.1.A1_at 1,21 8,28 0 0
Dr.3742.1.A1_at 1,11 8,6 0 0
406828 zgc:55420 Dr.8390.1.S1_at 1,49 8,59 0 0
323342 wu:fb95f11 Dr.26323.1.A1_at 1,82 9,03 0 0
569958 LOC569958 Dr.18186.1.S1_at 1,98 8,97 0 0
Dr.22713.1.A1_at 1,11 11,72 0 0
407625 si:dkey-­‐7c22.2Dr.4316.1.S1_at 1,59 11,84 0 0
Dr.7559.1.S1_at 1,83 12,8 0 0
324589 wu:fc33f12 Dr.3224.1.S1_at 1,73 19,76 0 0
768289 zgc:153629 Dr.1361.1.S1_at 1,12 13,28 0 0
Supplemental  table  2

Locus  ID Gene  Symbol Affy  ID logFC t-­‐stat p-­‐val Adj,  p-­‐val
436751 sf3b5 Dr.15894.1.S1_at -­‐1,27 -­‐10,13 0 0
553417 ppm1e Dr.12899.1.A1_at -­‐1,7 -­‐8,56 0 0
373113 bcl2l10 Dr.15057.1.S2_at -­‐1,01 -­‐7,02 0 0
192331 snrpd1 Dr.10241.1.S1_at -­‐1,09 -­‐6,47 0 0
550549 eif4e1c Dr.7525.1.A1_at -­‐1,02 -­‐5,87 0 0
321364 id2b Dr.12836.1.S1_at -­‐1,01 -­‐5,23 0 0
114426 odc1 Dr.5578.2.S1_a_at -­‐1,2 -­‐4,43 0 0,01
394188 qars Dr.20108.1.S1_a_at -­‐1,24 -­‐5,13 0 0
387591 ivns1abpa Dr.9512.1.A1_at -­‐1,06 -­‐4,43 0 0,01
Dr.631.1.S1_at -­‐1,08 -­‐5,19 0 0
Dr.1689.2.A1_a_at -­‐1,05 -­‐4,58 0 0,01
393492 zgc:66337 Dr.24337.1.A1_at -­‐1,24 -­‐10,63 0 0
Supplemental  table  3

Locus  ID Gene  Symbol Affy  ID logFC t-­‐stat p-­‐val Adj,  p-­‐val
386787 itga5 Dr.22498.1.A1_at 1,27 10,55 0 0
352944 cxcl12a Dr.822.1.S2_at 1,01 8,97 0 0
322614 arg2 Dr.2022.1.A1_at 1,04 8,95 0 0
554826 crip2 Dr.5660.1.S1_at 1,04 7,91 0 0
326935 ier5l Dr.25657.1.A1_at 1,44 7,64 0 0
333938 myh9a Dr.9531.1.A1_at 1,12 6,75 0 0
550445 acta1a Dr.4837.1.A1_at 1,43 6,68 0 0
560560 ywhag2 Dr.2009.1.A1_at 1,15 6,35 0 0
791647 pnp5a Dr.3216.1.A1_at 1,49 5,43 0 0
393910 sulf2 Dr.12717.1.S1_at 1,09 4,63 0 0,01
373081 mvp Dr.24562.1.S1_a_at 1,23 3,68 0 0,04
406541 slc25a25a Dr.2863.1.A1_at 1,37 4,66 0 0,01
558499 chka Dr.13284.1.A1_at 1,3 6,15 0 0
Dr.17653.2.S1_a_at 1,07 10,62 0 0
100009644 zgc:158345 Dr.23447.1.A1_at 1,08 10,15 0 0
Dr.17653.1.S1_at 1,05 8,3 0 0
386881 wu:fc35d06 Dr.21676.1.A1_at 1 7,9 0 0
566752 LOC566752 Dr.15575.1.A1_at 1,58 4,39 0 0,02
Supplemental  table  4

Locus  ID Gene  Symbol Affy  ID logFC t-­‐stat p-­‐val Adj,  p-­‐val
550134 ubb Dr.25166.1.S1_at -­‐1,29 -­‐12,41 0 0
544665 stc2 Dr.23461.1.A1_at -­‐1,48 -­‐10,58 0 0
 Jopling et al suppl.fig.1.

A B C

Hyperoxia Hyperoxia Hyperoxia

 Jopling et al suppl.fig.2.

A B C

WT WT WT
D E F

dnHIF1D dnHIF1D dnHIF1D

 Jopling et al suppl.fig.3.

ASA/BRDU
B

ASA/BRDU
 Jopling et al suppl.fig.4.

A MEF/BRDU D BRDU

B MEF/BRDU E MEF

C MEF/BRDU F MEF/BRDU

G
Hypoxia
80  
Normoxia
Hyperoxia
%  BrdU  Cardiomyocytes  

60   * dnHIF

40  

20  
* *
 Jopling et al suppl.fig.5.

16,00

14,00

12,00

10,00

8,00

6,00 *  
4,00
*  
*   *   *  
2,00 **  
0,00
stat2 mibp cxcr4b fst1b cish cebpb
wt  7  /wt  0 12,96 10,75 7,51 11,20 0,44 0,97
HIF  7  /HIF  0 2,29 2,43 3,76 2,10 1,46 5,62

*  :  p<0.02  

**  :  p<0.05  
Supplemental F ig.1. +\SR[\SUREHΠDQDO\VLV RI DGXOW ]HEUDILVK FDUGLRP\RF\WHV in vivo

under hyperoxic conditions . A section from an wildtype adult zebrafish heart 7 days post

amputation, maintained under hyperoxic conditions, immunolabelled for D sarcomeric actin

(green), DAPI(blue) and +\SR[\SUREHŒUHGA). A higher magnification image of the same

heart with D sarcomeric actin (green) and DAPI(blue) and +\SR[\SUREHŒUHG (B). The same

image seen in (B) with only +\SR[\SUREHŒUHGC).

Supplemental F ig.2. A nalysis of Tg(cmlc2a: E rt2-C re-E rt2/cmlc2a: L n L : G F P)

(cmlc2a: L r L :dn H I F1D) embryos. In situ hybridisation for dnHIF1D in, wild type 3dpf

embryos (A) the black circle indicates approximately where the heart is. The same heart at higher

magnification (B). A Tg(cmlc2a:GFP) embryo at 3dpf, GFP(green)(C). In situ hybridisation for

dnHIF1D in, Tg(cmlc2a:Ert2-Cre-Ert2/cmlc2a:LnL:GFP) (cmlc2a:LrL:dnHIF1D) 3dpf embryos

treated with 4OHT (D) the black circle indicates approximately where the heart is. The same

heart at higher magnification (E). Tg(cmlc2a:Ert2-Cre-Ert2/cmlc2a:LnL:GFP)

(cmlc2a:LrL:dnHIF1D) 3dpf embryo treated with 4 OHT, GFP(green)(F).

Supplemental F ig.3. B rdU analysis of heart regeneration.

(A)A 14 dpa regenerating wildtype adult zebrafish heart maintained under normoxic conditions

immunolabelled for D sarcomeric actin (green), BrdU (red) and DAPI (blue). Brdu positive

cardiomyocytes are identified by yellow nuclei (D sarcomeric actin (green) and BrdU (red)). The

white circle denotes the area seen at higher magnification in (B). (B) Higher magnification image

showing numerous BrdU positive cardiomyocytes (circled yellow nuclei).

Supplemental F ig.4. B rdU analysis of heart regeneration.

(A)A 14 dpa regenerating wildtype adult zebrafish heart maintained under normoxic conditions

immunolabelled for mef2c (green), BrdU (red) and DAPI (blue). BrdU positive cardiomyocytes

are identified by yellow nuclei (mef2c (green) and BrdU (red)). (B,C) Higher magnification

images of the heart depicted in (A), the region of interest is denoted with a square. The white

circle indicates an individual BrdU/mef2c positive cardiomyocyte(C). (D,E) Higher

magnification images of the individual BrdU/mef2c positive cardiomyocyte labelled with

BrdU(red)(D) and mef2c (green)(E). (F) Composite Z-stack image of the individual BrdU/mef2c

positive cardiomyocyte. (G)A graph representing the average number of BrdU positive

cardiomyocytes/section in each of the four conditions (n=average of 1 section from 4 animals),

hypoxia (blue bar), normoxia (white bar), hyperoxia (black bar) and dnHIF1D (grey bar)   +/-

SEM, t-test * p<0.01.

Supplemental F ig.5. V erification of gene expression changes by qPC R.

Genes found to be differentially expressed on the microarray were subsequently analysed via

qPCR. Normalized relative quantification was performed with the ''Ct method. The ratio of

gene expression of each target gene between 7 days after amputation (Sample S) and before

amputation (Calibrator C) was normalized to E-actin (Reference) (R=ETarget 'Cp (Target)/ R=Eref 'Cp
(ref)
, with 'Cp = Cp(C)-Cp(S)). Each value corresponds to the mean of the values obtained by

triplicate measurements. The asterisks(*) indicate p-YDOXHV GHWHUPLQHG E\ VWXGHQW¶V W-test and

show that the gene expression changes observed in the dnHIF1Dsamples (black bars) are
significantly different from the gene expression changes observed in the wildtype samples (white

bars), error bars indicate SEM.

Supplemental table.1.

List of genes found to be significantly upregulated between the unamputated wildtype and 7dpa

wildtype samples which showed no significant change in expression between the unamputated

dnHIF1Dand 7dpa dnHIF1Dsamples.

Supplemental table.2.

List of genes found to be significantly downregulated between the unamputated wildtype and

7dpa wildtype samples which showed no significant change in expression between the

unamputated dnHIF1Dand 7dpa dnHIF1Dsamples.

Supplemental table.3.

List of genes found to be significantly upregulated between the unamputated dnHIF1D and 7dpa

dnHIF1D samples which showed no significant change in expression between the unamputated

wildtype and 7dpa wildtype samples.

Supplemental table.4.

List of genes found to be significantly downregulated between the unamputated dnHIF1D and

7dpa dnHIF1D samples which showed no significant change in expression between the

unamputated wildtype and 7dpa wildtype samples.