Myocardial ß2-adrenoceptor gene delivery promotes coordinated cardiac adaptive remodelling and angiogenesis in heart failure
summarize and critiquing the following scientific paper, in terms of it both its merits and flaws. Marking Scheme: Well organised and well expressed answer which shows very good understanding, evidence of independent study and critical evaluation. Covers all aspects of the subject that could reasonably be expected, accurately and in sufficient detail.

Myocardial ß2-adrenoceptor gene delivery promotes coordinated cardiac adaptive remodelling and angiogenesis in heart failure
We investigated whether b2-adrenoceptor overexpression could promote angiogenesis and improve blood perfusion and left
ventricular (LV) remodeling of the failing heart.
We explored the angiogenic effects of b2-adrenoceptor overexpression in a rat model of post-myocardial infarction (MI) heart
failure (HF). Cardiac adenoviral-mediated b2-adrenoceptor overexpression was obtained via direct intramyocardial injection
4-weeks post-MI. Adenovirus(Ad)-GFP and saline injected rats served as controls. Furthermore, we extended our observation
to b2-adrenoceptor -/- mice undergoing MI.
Transgenes were robustly expressed in the LV at 2 weeks post-gene therapy, whereas their expression was minimal at 4-weeks
post-gene delivery. In HF rats, cardiac b2-adrenoceptor overexpression resulted in enhanced basal and isoprenaline-stimulated
cardiac contractility at 2-weeks post-gene delivery. At 4 weeks post-gene transfer, Ad-b2-adrenoceptor HF rats showed
improved LV remodeling and cardiac function. Importantly, b2-adrenoceptor overexpression was associated with a markedly
increased capillary and arteriolar length density and enhanced in vivo myocardial blood flow and coronary reserve. At the
molecular level, cardiac b2-adrenoceptor gene transfer induced the activation of the VEGF/PKB/eNOS pro-angiogenic pathway.
In b2-adrenoceptor-/- mice, we found a ~25% reduction in cardiac capillary density compared with b2-adrenoceptor+/+ mice.
The lack of b2-adrenoceptors was associated with a higher mortality rate at 30 days and LV dilatation, and a worse global
cardiac contractility compared with controls.
b2-Adrenoceptors play an important role in the regulation of the angiogenic response in HF. The activation of VEGF/PKB/eNOS
pathway seems to be strongly involved in this mechanism.
Ad, adenovirus; ECs, endothelial cells; eNOS, endothelial NOS; FS, LV fractional shortening; GFP, green fluorescent
protein; GRK2, G-protein coupled receptor kinase-2; HF, heart failure; ICI118551, erythro-dl-1-(7-methylindan-4-yloxy)-
3-isopropylaminobutan-2-ol; ISO, isoprenaline; KO, knockout; LAD, left anterior descending coronary artery; LV, left
ventricular; MI, myocardial infarction; SNS, sympathetic nervous system
BJP British Journal of
2348 British Journal of Pharmacology (2012) 166 2348.2361 c 2012 The Authors
British Journal of Pharmacology c 2012 The British Pharmacological Society
Heart failure (HF) is a complex clinical syndrome characterized
by left ventricular (LV) dysfunction accompanied by
generalized hyperactivation of neurohormonal status (Rengo
et al., 2004; Mann and Bristow, 2005; Lymperopoulos et al.,
2007a). Although the sympathetic nervous system (SNS)
overdrive is beneficial immediately after cardiac injury,
aiming to preserve cardiac output, over time, it becomes
detrimental, facilitating the worsening of cardiac contractility.
The prolonged and sustained HF-related SNS activation
induces several noxious effects on myocardial and vascular
function, and is responsible for the dysregulation of cardiac
b-adrenoceptors, including b1-adrenoceptor down-regulation
and b1- and b2-adrenoceptor uncoupling from signaltransducing
G-proteins (Rockman et al., 2002; Lohse et al.,
2003; Feldman et al., 2005; Leosco et al., 2007; Triposkiadis
et al., 2009), via increased cardiac G-protein coupled receptor
kinase-2 (GRK2) expression/activity (Lefkowitz, 1993;
Ungerer et al., 1993; 1994; Rengo et al., 2011). The molecular
dysfunction of b-adrenoceptor signalling accounts for the
reduced inotropic response to adrenergic stimulation in the
failing myocardium.
Over the past two decades, several data have been
produced proving different effects between b1- and
b2-adrenoceptor signalling on cardiac physiology and pathophysiology
(Lands et al., 1967; Communal et al., 1999; Zaugg
et al., 2000; Xiao et al., 2004). However, the role of
b2-adrenoceptors in HF has not been completely elucidated
(Zhu et al., 2005). Previous evidence indicated that cardiac
the overexpression of b2-adrenoceptors improves contractility
in the healthy heart (Milano et al., 1994; Dorn et al., 1999;
Maurice et al., 1999; Liggett et al., 2000), and potentiates the
functional recovery of unloaded rabbit failing myocardium
(Tevaearai et al., 2002). Furthermore, other evidence has been
obtained indicating that pharmacological stimulation of
b2-adrenoceptors in the post-ischaemic failing heart induces
cardioprotection preventing maladaptive remodelling, and
curbing HF progression (Ahmet et al., 2004; Xydas et al.,
Recently, b2-adrenoceptor signalling has been demonstrated
to be involved in angiogenesis regulation. In an experimental
model of hindlimb ischaemia, it has been
demonstrated that b2-adrenoceptors play a pivotal role in the
control of endothelial cells (ECs) function. Adenoviralmediated
b2-adrenoceptor overexpression in the ischaemic
hindlimb leads to enhanced EC proliferation and migration
with the final effect of improved ischaemia-induced angiogenesis
(Iaccarino et al., 2005). This pro-angiogenic effect
seems to be ascribed to increased b2-adrenoceptor-dependent
vascular endothelial growth factor (VEGF) production and
release (Iaccarino et al., 2005). Moreover, further findings supporting
the important role of b2-adrenoceptors in angiogenesis
have been obtained in b2-adrenoceptor knockout mice
subjected to chronic hindlimb ischaemia (Ciccarelli et al.,
2011). Importantly, the impaired angiogenic response to
ischaemia, observed in these mice, is restored by intravascular
b2-adrenoceptor gene transfer (Ciccarelli et al., 2011).
However, whether b2-adrenoceptors regulate angiogenesis in
the heart has never been previously tested either in physiological
or in pathological conditions. With respect to this,
previous findings from our group (Leosco et al., 2008) and
others (Karam et al., 1990) indicate that angiogenic responses
in the failing heart are inadequate and that impairments in
the cardiac capillary and arteriolar network during HF plays a
key role in post-ischaemic heart dysfunction. At the molecular
level, this is associated with a blunted activation of the VEGF/
PKB (Akt)/endothelial-NOS (eNOS) pathway. Nevertheless,
the angiogenesis impairment can be reverted through interventions,
such as exercise training, leading to an improvement
in cardiac b-adrenoceptor signalling and reactivation of
the PKB pro-angiogenic pathway (Leosco et al., 2008).
We aimed to evaluate for the first time whether the beneficial
effects of b2-adrenoceptor overexpression in the
failing myocardium may be correlated to the restoration of
HF-related impairment of cardiac angiogenesis. We utilized
an adenoviral-mediated intramyocardial gene delivery technique,
which allows the robust expression of the transgene in
the LV (Rengo et al., 2009; Zincarelli et al., 2010). To further
support the notion that the b2-adrenoceptor plays a crucial
role in ischaemia-induced angiogenesis, we also studied
the phenotype of post-myocardial infarction (MI) HF
b2-adrenoceptor knockout (KO) mice.
Experimental procedures
The study protocol was designed in accordance with The
Guide for Care and Use of Laboratory Animals of the National
Institutes of Health (NIH Publication No. 85.23, Revised
1996), and was approved by the Ethics Committee for the Use
of Animals in Research of our Institution. An expanded
Methods section appears in the online-only Data Supplement.
The results of all studies involving animals are reported
in accordance with the ARRIVE guidelines (Kilkenny et al.,
2010; McGrath et al., 2010).
For experimental MI induction, rats were anaesthetized
with 4% isoflurane. Deep anaesthesia was confirmed by the
lack of a response to noxious stimuli. Next, rats were intubated
and ventilated with a mixture of O2 and 2% isoflurane
with a pressure-controlled ventilator. After thoracotomy, MI
was induced by permanent ligation of the left anterior
descending coronary artery (LAD). Subsequently, chest and
skin were closed in layers. Rats were observed until they
awakened, and then returned to the Animal Care Unit. Standard
postoperative care, including analgesic (buprenorphine
400 IUEkg-1 i.p. 20 min before surgery, ketoprofen 10 mgEkg-1
s.c. injection for 2 days after surgery), was provided. MI in
mice was induced under general anaesthesia with isoflurane(
2%), the heart was exposed and temporarily dislocated
via a small left thoracotomy to place a suture ligation of the
LAD (Lymperopoulos et al., 2009; 2010). Myocardial gene
transfer in rats was achieved by direct intramyocardial injection
4 weeks post-MI (Rengo et al., 2009). Echocardiographic
evaluations were performed at 4 weeks post-MI and at the end
of the study (Leosco et al., 2008). Basal and isoprenalinestimulated
LV contractility was evaluated by invasive haemodynamic
studies at 2 and 4 weeks after gene delivery. At the
end of the study period, animals were killed by cervical dislocation
under deep anaesthesia.
b2-adrenoceptor regulates angiogenesis in heart failure BJP
British Journal of Pharmacology (2012) 166 2348.2361 2349
Myocytes contractility
Myocytes were isolated from the non-infarcted zone of the LV
by a standard enzymatic digestion (Leosco et al., 2008; Rengo
et al., 2009). Contractility was evaluated under baseline and
after agonist-mediated b1-adrenoceptor stimulation by NA at
10-7 M or higher concentrations in the presence of the
a1-adrenoceptor antagonist, prazosin (10-6 M).
Myocardial perfusion studies
Myocardial perfusion was determined using 15 mm fluorescent
microspheres (Triton Technology, Inc., San Diego, CA,
USA). Cardiac and blood samples were processed for microsphere
determination. Myocardial blood flow was measured
at basal and after maximal vasodilatation.
Capillary and arteriolar length densities were evaluated in
border and remote zones of the infarcted area. Capillaries
were detected by Lectin Bandeiraea simplicifolia-I staining.
Arterioles were identified by immunofluorescence using
anti-SM a-actin antibody (Leosco et al., 2008).
b-Adrenoceptor signalling
Receptor binding, adenylyl cyclase activity and GRK2 protein
assays were obtained as previously described (Rengo et al.,
VEGF/PKB/eNOS measurement
VEGF, PKB, Ser473-phospho-PKB, eNOS and Ser1177-phosphoeNOS
protein levels were performed by Western blot analysis
(Lymperopoulos et al., 2007b; 2008).
Statistical analysis
Data are summarized as mean  SEM. Comparisons were
made with the use of Studentfs t-tests or ANOVA as appropriate.
A Bonferroni correction was applied to the probability
values whenever multiple comparisons arose. Values of P <
0.05 were considered significant.
Study design and transgenes expression
At 1 month after MI, a time point with established HF (Leosco
et al., 2008), rats were randomly allocated to three different
groups receiving cardiac gene transfer of: adenovirus (Ad).b2-
adrenoceptor, fluorescent protein (GFP), or vehicle
(saline). One day before gene delivery, all groups were analysed
by echocardiography to confirm the presence of similar levels
of LV dysfunction and HF. All groups were then studied over
the course of 4 more weeks (Figure 1A), and all assays in the 3
HF groups were compared with a control sham-operated group
that received neither MI nor gene transfer (4 experimental
Haemodynamic Echo
4 weeks
2 weeks after
gene delivery
4 weeks after
gene delivery
450 *
Total ƒÀAR density
pmol mg.1 of membrane protein
% Green cardiomyocyte
1 day 2 weeks 2 weeks
A MI Gene delivery
Figure 1
(A)Overall design of the 8-week study. (B) Representative GFP fluorescence microscopy (left), light microscopy (middle) and overlay of both (right)
of LV myocardium 2 weeks after intramyocardial Ad-GFP delivery. (C) b2-Adrenoceptor (AR) immunohistochemistry in LV tissue from Ad-GFP- (left,
control) and Ad-b2-adrenoceptor-treated (right) rats (2 weeks post-gene delivery). Magnification \40. (D) Total b-adrenoceptor density (B) in
cardiac homogenates purified from hearts of HF Ad-b2-adrenoceptor and HF control (HF-saline and HF Ad-GFP) groups at 2 weeks after gene
therapy (n = 6 and 8 for each group); *P < 0.001 vs. HF control. (E) Right panel; percentage (%) of GFP-stained isolated myocytes assessed 2 and
4 weeks following Ad-GFP in vivo gene delivery to HF rats by direct intra-myocardial injection (n = 5 for each time point). Green myocytes from
each rat heart were counted in five randomly selected fields and expressed as percentage of the total number of myocytes per field. *P < 0.001
vs. 4 weeks after gene delivery. Data are presented as means  SEM. Left panel: representative GFP fluorescence microscopy (upper), light
microscopy (middle) and overlay of both (lower) of myocytes 2 weeks after Ad-GFP delivery.
BJP G Rengo et al.
2350 British Journal of Pharmacology (2012) 166 2348.2361
groups in total). At 2 weeks post-gene delivery, both transgenes
(b2-adrenoceptor and GFP) were robustly expressed in the LVof
the respective groups. GFP fluorescence in cardiac sections
from Ad-GFP-treated rats confirmed that a large area of the LV
free wall was transduced, although the expression was not
homogeneous (Figure 1B). b2-Adrenoceptor immunohistochemistry
of cardiac sections from Ad-b2-adrenoceptor-treated
HF rats showed comparable areas of the LV transduced
(Figure 1C), and the b-adrenoceptor binding experiment
showing a 15-fold increase in membrane receptors compared
with HF-saline and HF-GFP hearts (Figure 1D). Moreover, to
confirm that our gene delivery technique supported efficient
cardiac expression in vivo, 2 and 4 weeks after gene delivery of
Ad-GFP, GFP expression was evaluated by detection of green
fluorescence in cardiomyocytes isolated from the LV as previously
described (Rengo et al., 2009). Interestingly, at 2 weeks
after Ad-GFP gene delivery we found a transduction efficiency
to the LV that was >35% of total isolated LV myocytes,
whereas, at 4 weeks post-gene delivery, GFP expression was
minimal (less than 5%) (Figure 1E), which is consistent with
the duration of in vivo expression seen by us and others using
Ad (Maurice et al., 1999).
In vivo effects of cardiac b2-adrenoceptors
overexpression on LV remodelling and
cardiac contractility
At 1 month post-MI and before gene delivery, all HF groups
had significantly impaired cardiac function compared with
sham rats. As assessed by echocardiography, LV fractional
shortening (FS) and internal diameter at diastole were comparable
in all the HF groups, indicating a similar degree of
cardiac dysfunction (Table 1).
At 2 weeks post-gene delivery, as assessed by in vivo invasive
haemodynamic analysis, HF control groups (HF-Saline
and HF-GFP) showed impaired basal and isoprenaline (ISO)-
stimulated cardiac contractility, and increased LV enddiastolic
pressure compared with sham, as expected
(Figure 2). Two weeks of b2-adrenoceptor overexpression
resulted in enhanced basal cardiac contractility and reduced
LV end-diastolic pressure compared with HF-saline and -GFP
rats (Figure 2 and Table S1).
At 4 weeks from gene transfer, a time point when
adenoviral-mediated b2-adrenoceptor overexpression was
minimal, the beneficial effects of b2-adrenoceptor gene
delivery on basal cardiac contractility disappeared, but
b-adrenoceptor mediated contractility induced by ISO was
significantly increased compared with HF controls (Table 2).
Interestingly, NA was able to increase b1-adrenoceptordependent
contractility in cardiomyocytes extracted from
AdV-b2-adrenoceptor infected hearts compared with HF
control. This result indirectly shows there is no obliteration
of b1-adrenoceptor responses in isolated cells (Figure S1). At
the end of the study-period, HF-saline and HF-GFP rats
showed further deterioration of cardiac function and progression
of LV maladaptive remodelling (LV dilatation and
reduced LVW/BW) compared with the echocardiographic
Table 1
Physical and echocardiographic data in sham-operated and HF rats at 4 weeks after myocardial infarction
Sham HF saline HF Ad-GFP HF Ad-b2-adrenoceptor
Physical data
Body wt (kg) 0.408  0.017 0.372  0.012* 0.360  0.018* 0.366  0.010*
Heart wt (g) 1.15  0.09 2.22  0.21* 2.30  0.19* 2.16  0.26*
Heart wt/body wt (gEkg-1) 2.86  0.13 5.94  0.63* 6.40  0.56* 5.89  0.60*
Lung wt (g) 1.68  0.18 2.87  0.22* 3.0  0.38* 2.92  0.45*
Lung wt/body wt (gEkg-1) 4.06  0.34 7.70  0.85* 8.3  0.92* 7.88  0.67*
Echocardiographic data
Heart rate (beats min-1) 366  22 355  16 348  32 352  28
LV internal diameter (mm)
Diastolic 5.72  0.45 8.86  0.74* 9.12  0.82* 9.36  1.0*
Systolic 3.05  0.28 6.35  0.62* 6.22  0.74* 6.40  0.82*
LV FS (%) 46.4  6.0 27.8  5.2* 30  4.7* 29.2  5.6*
Interventricular septum (mm)
Diastolic 1.49  0.9 1.03  0.11* 1.00  0.8* 1.04  0.9*
Systolic 2.64  0.32 1.06  0.12* 1.03  0.6* 1.08  0.10*
LV posterior wall (mm)
Diastolic 1.53  0.17 1.88  0.18* 1.86  0.13* 1.85  0.17*
Systolic 2.49  0.30 2.57  0.26 2.55  0.15 2.60  0.18
Sham, n = 10; HF-saline, n = 12; HF Ad-GFP, n = 11; HF Ad-b2-adrenoceptor, n = 12.
*P < 0.05 vs Sham.
HF, heart failure; LV, left ventricle; FS, fractional shortening.
b2-adrenoceptor regulates angiogenesis in heart failure BJP
British Journal of Pharmacology (2012) 166 2348.2361 2351
measurements performed at 1-month post-MI, as expected
(Table 1, Table 2 and Figure 3). Importantly, b2-adrenoceptor
overexpression resulted in improved FS, reduced ventricular
systolic and diastolic diameters, and increased LVW/BW compared
with HF control groups (Table 2 and Figure 3). Finally,
at the post-study analysis of LV infarct size, we found similar
infarct sizes among all study groups (Table 2). In this regard,
the lack of a reduction in infarct size in b2-adrenoceptor
overexpressing HF rats should be ascribed to the fact that
gene delivery was performed late after MI when infarct
healing was complete.
Taken together, these data indicate that b2-adrenoceptor
overexpression exerts a favourable effect on post-MI LV contractility
and remodelling. Interestingly, the positive effects
on isoprenaline-stimulated LV contractility and on adaptive
hypertrophic remodelling also persisted when the transgene
expression was exhausted.
Effects of b2-adrenoceptors overexpression
on cardiac angiogenesis and myocardial
blood flow
At 2 weeks after gene delivery, capillary density, but not
arteriolar length, was significantly improved in HF-b2-
adrenoceptor rats compared with the HF-saline and HF-GFP
groups (Supplemental Table 1).
Figure 4 displays histological and blood flow data in all
study groups at 4 weeks after gene delivery. All HF groups
showed a significant reduction in both capillary density and
arteriolar length compared with sham animals. Interestingly,
b2-adrenoceptor overexpression induced a significant reactivation
of the angiogenic mechanims in HF hearts. In fact,
capillary and arteriolar density in both border and remote
zones of the LV were significantly increased in the HF Ad-b2-
adrenoceptor group compared with HF-saline and HF-GFP
20 000
18 000
16 000
14 000
12 000
10 000
LV . dP/dt (mmHgEs.1) LV + dP/dt (mmHgEs.1)
.10 000
C 0.1
* *
* *
* õ
ISO (ugEkg.1Emin.1)
C 0.1 0.5
Sham HF
HF saline
Figure 2
(A) Average LV +dP/dt and LV -dP/dt values (B) at 2 weeks post-gene therapy in the four experimental groups evaluated under basal conditions
and after isoprenaline stimulation. (C) Average left ventricle end diastolic pressure (LVEDP) in the four experimental groups. Sham, n = 11;
HF-saline, n = 13; HF Ad-GFP, n = 12; HF Ad- b2-adrenoceptor (AR), n = 12. ANOVA analysis and Bonferroni test were used among all groups. Data
are presented as mean  SEM. *P < 0.05 vs sham at basal or at each respective dose of isoprenaline; õP < 0.05 vs HF-saline and HF Ad-GFP at basal
or at each respective dose of isoprenaline.
BJP G Rengo et al.
2352 British Journal of Pharmacology (2012) 166 2348.2361
hearts. Importantly, in HF Ad-b2-adrenoceptor rats, we
observed a further increase in capillary density between 2 and
4 weeks post-gene delivery (Figure 4 and Supplemental
Table 1). Moreover, although myocardial blood flow and
coronary reserve were significantly reduced in all HF groups
compared with sham, b2-adrenoceptor overexpression
resulted in a significant improvement in both perfusion
parameters compared with HF-controls (Figure 4).
Myocardial b-adrenoceptor status after in
vivo gene delivery
Two weeks after gene therapy, cardiac b-adrenoceptor density
was significantly reduced in HF controls compared with
sham, while the HF Ad-b2-adrenoceptor group showed a
15-fold increase in total membrane b-adrenoceptors compared
with HF controls, proving the efficacy of the viralmediated
gene delivery technique (Figure 1 and Table 3).
b2-Adrenoceptor overexpression was accompanied by higher
basal cardiac cAMP levels compared with HF-saline and
HF-GFP groups. As expected, in HF control groups, cAMP
cardiac levels were lower compared with sham (Table 3). ISO
induced roughly 2- to 2.5-fold increases in adenylate cyclase
activity in all HF groups. Myocardial GRK2 levels, measured
by Western blotting, were significantly up-regulated in
HF-saline and GFP groups compared with sham. In
b2-adrenoceptor-infected hearts, cardiac GRK2 expression was
decreased almost to the sham levels.
Interestingly, at 4 weeks post-gene delivery, a time point
when adenoviral-mediated overexpression of transgenes was
minimal, total b-adrenoceptor density and cAMP production
remained still higher in Ad-b2-adrenoceptor-infected hearts
compared with HF controls (Table 3). Accordingly, cardiac
GRK2 protein expression was still significantly lower in
Table 2
Physical, haemodynamic and echocardiographic parameters at 4 weeks after cardiac gene delivery
Sham-operated HF Saline HF Ad-GFP HF Ad-b2-adrenoceptor
Physical data
Body wt (kg) 0.426  0.024 0.388  0.022* 0.390  0.019* 0.382  0.028*
Heart wt (g) 1.17  0.07 2.58  0.28* 2.42  0.19* 2.82  0.23*õ
Heart wt/body wt (gEkg-1) 2.75  0.14 6.66  0.37* 6.12  0.23* 7.36  0.34*õ
Lung wt (g) 1.70  0.09 3.15  0.30* 3.22  0.16* 2.14  0.19*õ
Lung wt/body wt (gEkg-1) 3.98  0.32 8.22  0.46* 8.27  0.38* 5.60  0.36*õ
Hemodynamic and echo data
Heart rate (beats min-1) 355  32 342  28 340  29 365  26
LVSP (mmHg) 136  12 111  10* 109  8* 123  9*õ
LVEDP (mmHg) 4.5  0.8 21  5* 20  6** 10  3*õ
LV +dP/dt (mmHgEs-1)
Baseline 8842  457 4053  268* 4238  224* 4427  313*
ISO (1 mgEkg-1Emin-1) 16 724  966 6558  236* 6886  287* 9569  461*õ
LV .dP/dt (mmHgEs-1) -6 488  440 -3153  235* -2923  321* -3366  357*
Baseline -8 421  654 -4115  586* -3934  443* -5340  523*õ
ISO (1 mgEkg-1Emin-1)
LV internal diameter (mm)
Diastolic 5.69  0.37 9.52  0.48* 9.43  0.56* 7.10  0.32*õ
Systolic 3.10  0.22 7.22  0.39* 7.29  0.44* 5.10  0.36*õ
LV fractional shortening (%) 45.5  8 24.1  4.6* 22.8  5.9* 34.6  7.2*õ
Interventricular septum (mm)
Diastolic 1.47  0.8 1.02  0.9* 0.90  0.8* 1.00  0.7*
Systolic 2.58  0.28 1.03  0.8* 0.97  0.6* 1.03  0.13*
LV posterior wall (mm)
Diastolic 1.50  0.15 1.94  0.17* 1.97  0.13* 2.32  0.22*õ
Systolic 2.51  0.27 2.56  0.28 2.52  0.16 2.88  0.18*õ
Infarct size (%) . 48.8  5.6 51.3  6.7 47.3  7.5
Sham, n = 11; HF-saline, n = 13; HF Ad-GFP, n = 14; HF Ad-b2-adrenoceptor, n = 13.
*P < 0.01 vs. sham.
õP < 0.01 vs. HF saline and HF GFP.
HF, heart failure; LV, left ventricle; LVSP, left ventricular systolic pressure; LVEDP, left ventricular end diastolic pressure; ISO, isoprenaline.
b2-adrenoceptor regulates angiogenesis in heart failure BJP
British Journal of Pharmacology (2012) 166 2348.2361 2353
Ad-b2-adrenoceptor-infected hearts compared with other HF
Cardiac b2-adrenoceptors overexpression and
VEGF/PKB/eNOS pathway
Next, we explored the effects of b2-adrenoceptor overexpression
on the cardiac VEGF/PKB/eNOS pathway, which is
known to have a critical role in the control of the angiogenic
mechanisms in the heart (Lymperopoulos et al.,
2008; 2010).
At 2 weeks after gene transfer, all HF groups showed
higher cardiac phospho (p)-PKB/total (t)-PKB ratio compared
with sham animals (Figure 5A). Noteworthy, in the HF
b2-adrenoceptor group, PKB activation was associated with
enhanced cardiac VEGF and p-eNOS/eNOS protein levels
compared with HF control groups. In contrast, HF-saline and
HF-GFP groups did not show any increase in VEGF and
p-eNOS protein expression, with levels comparable to those
observed in sham hearts. These results agree with previous
data, indicating that PKB activation is not always associated
with enhanced activity of its downstream effectors, and, most
importantly, that dissociation between PKB and eNOS/VEGF
leads to negative cardiac remodelling (Shiojima et al., 2005;
Shiojima and Walsh, 2006). Overall, these data strongly
suggest that cardiac b2-adrenoceptor overexpression is able to
restore the integrity of the pro-angiogenic pathway and to
promote coordinated cardiac and vascular growth in the
post-MI failing myocardium.
Interestingly, 4 weeks post-gene therapy, the activation of
the VEGF/PKB/eNOS pathway was blunted in the HF-b2-
adrenoceptor group, whereas, in the HF control groups the
p-PKB/t-PKB ratio was still higher compared with both HF-the
Figure 3
(A) Representative LV cross sections and (B) echocardiographic M-mode recordings from of all study groups at the end of the study period. (C)
LV internal diameter at diastole (LVIDd) (left), LVID at systole (LVIDs) (middle) and fractional shortening (FS) (right) as measured by echocardiography
4 weeks after gene delivery. Sham, n = 11; HF-saline, n = 14; HF Ad-GFP, n = 11; HF Ad-b2-adrenoceptor (AR), n = 12. Bar = 10 mm. ANOVA
analysis and Bonferroni test among all groups. All data are expressed as mean  SEM. *P < 0.05 vs. sham at basal or at each respective dose of
isoprenaline; õP < 0.05 vs. HF-saline and HF Ad-GFP at basal or at each respective dose of isoprenaline.
BJP G Rengo et al.
2354 British Journal of Pharmacology (2012) 166 2348.2361
PKB activation was not associated with enhanced VEGF and
eNOS protein levels.
In order to link directly the effects of b2-adrenoceptor
overexpression to the activation of the VEGF/PKB/eNOS
pathway, an additional group of Ad.b2-adrenoceptor HF
infected rats was treated with the specific b2-adrenoceptor
antagonist ICI118 551 (0.2 mgEkg-1day-1) beginning 5 days
before gene delivery. As illustrated in Figure S2, ICI118551
was able to completely prevent b2-adrenoceptor
overexpression-dependent activation of the pro-angiogenic
signalling in the hearts. This result suggests a direct effect of
b2-adrenoceptors on the activation of VEGF/PKB signalling.
Interestingly, in b2-adrenoceptor-infected hearts, the
VEGF/PKB/eNOS pathway, evaluated separately in cardiomyocytes
and endothelial cells, was activated to a similar
extent in both cell subtypes at 2 weeks after gene delivery (see
supplemental Figure S3). This indicates that the overall effect
on angiogenesis activation in b2-adrenoceptor-overexpressing
hearts could be mediated by pro-angiogenic stimuli originating
from both myocytes and EC, thus, confirming previous
observations suggesting the existence of reciprocal cross-talk
mechanisms between the vasculature and cardiac myocytes
that regulate coronary angiogenesis and contractile function
(Shiojima et al., 2005; Shiojima and Walsh, 2006).
Impairment of cardiac angiogenesis and LV
contractility in post-ischaemic HF
b2-adrenoceptor-/- mice
To further support the notion of a crucial role for the
b2-adrenoceptor in the regulation of cardiac function and
angiogenesis during HF, homozygous b2-adrenoceptor-/-
(b2KO) or b2-adrenoceptor+/+ male mice littermates (control
group) underwent MI (to induce HF) or sham operation
(Figure 6). One month after surgically induced MI, the mortality
rate was threefold higher in b2-adrenoceptor-/-
compared with b2-adrenoceptor+/+ mice (Figure 6A).
Echocardiography performed at this time point showed
reduced FS in HF b2-adrenoceptor+/+ compared with sham
mice. The lack of b2-adrenoceptors was associated with
further decrease in FS compared with HF b2-adrenoceptor+/+
mice (Figure 6B and Supplemental Table 2).
Interestingly, cardiac capillary density was significantly
decreased in sham b2-adrenoceptor-/- compared with sham
b2-adrenoceptor+/+. Moreover, 1 month post-MI, both HF
Arterioles (mm mm.3)
Fold increase in conductance
Capillaries ƒÊm.2
mLEmin.1E100 g.1
HF Ad-ƒÀ2AR HF-GFP HF saline Sham
Max dilation
* *
* *
Figure 4
(A) Representative images of (left) Lectin Bandeiraea simplicifolia I staining of capillaries in LV sections and (right) of arterioles stained with
antibodies against smooth muscle a-actin obtained from all study groups at 4 weeks post-gene therapy in the lateral wall far from the infarcted
area (remote). Magnification \40. Scale bar: 50 mm. (B) Histograms show data on capillary counts, and (C) arteriolar length density in either LV
border anterior and lateral, and remote zones in all study groups at 4 weeks after gene therapy (n = 5 for each group). (D) Average of myocardial
blood flow at basal condition and after maximal coronary dilation by dipyridamole and of coronary reserve measured in all study groups at the
end of the study period (n = 8 rats for each group). ANOVA analysis and Bonferroni test among all groups. All data are expressed as mean  SEM.
*P < 0.05 vs. sham; õP < 0.05 vs. HF-saline and HF Ad-GFP.
b2-adrenoceptor regulates angiogenesis in heart failure BJP
British Journal of Pharmacology (2012) 166 2348.2361 2355
mouse lines showed a decrease in capillary density compared
with the respective sham groups. However, HF
b2-adrenoceptor-/- mice had significantly lower capillary
density compared with HF b2-adrenoceptor+/+ (Figure 6C). At
the molecular level, we investigated the VEGF/PKB/eNOS
pathway by Western blotting in both b2-adrenoceptor-/- and
b2-adrenoceptor+/+ mice 1 month post-sham operation or
surgically induced MI (Figure 6D). In HF b2-adrenoceptor+/+
mice, we found an enhanced p-PKB/t-PKB ratio compared
with both sham mouse lines. Interestingly, PKB activation
was even more pronounced in HF b2-adrenoceptor-/- compared
with HF b2-adrenoceptor+/+. However, VEGF and
p-eNOS/eNOS protein levels in both HF lines were not significantly
different from those observed in sham
b2-adrenoceptor-/- and b2-adrenoceptor+/+ mice.
In the present study, the b2-adrenoceptor has been shown to
be involved in the regulation of cardiac angiogenesis in the
context of HF. In fact, in a rat model of post-ischaemic HF, we
have reported for the first time that adenoviral-mediated
cardiac overexpression improves post-MI LV function and
remodelling, and that these favourable effects are associated
with increased in vivo myocardial angiogenesis and blood
perfusion, significant activation of the pro-angiogenic
VEGF/PKB/eNOS pathway, and a reduction in cardiac
b-adrenoceptor down-regulation. These data are consistent
with the impaired myocardial capillarization observed in
b2-adrenoceptor-/- mice, that also showed a blunted angiogenic
response to ischaemia, more severe post-MI LV dysfunction
and higher 30-day post-MI mortality rate compared
with control mice.
After MI, adequate growth of new capillaries and arterioles
is needed to allow compensatory hypertrophic responses
and favourable LV remodelling (Anversa et al., 1986).
However, neoangiogenesis is often inadequate in the post-MI
heart (Karam et al., 1990; Leosco et al., 2008), and the lack of
a valid readaptation of coronary conductance and reserve
contributes to infarct expansion and transition from adaptive
cardiac hypertrophy to LV dilatation and dysfunction.
Interestingly, several lines of evidence have shown that
the b2-adrenoceptor signals and functions in a substantially
different way compared with the b1-adrenoceptor. In fact,
whereas b1-adrenoceptor activation results in increased cardiomyocyte
apoptosis, b2-adrenoceptor stimulation protects
cardiomyocytes against apoptotic stimuli (Communal et al.,
1999). Moreover, in vivo studies also support the notion that
b2-adrenoceptor activation and signalling in the heart may
have positive implications that might be beneficial in HF
(Liggett et al., 2000). In particular, the aim of the present
investigation was to demonstrate that the replacement of
dysfunctioning b2-adrenoceptors, which is a well-described
phenomenon occurring in HF, could improve cardiac angiogenesis
in the failing myocardium. In this context, mechanistic
studies have shown that b2-adrenoceptors control EC
function (Ahmet et al., 2004) and promote cell survival in
other tissues (Morisco et al., 2000; Zhu et al., 2001). Iaccarino
et al. have shown a selective down-regulation of b2-
adrenoceptors in the skeletal muscle following ischaemia,
and neoangiogenesis improvement by b2-adrenoceptor overexpression
in the ischaemic muscle (Iaccarino et al., 2005).
In our study, we demonstrated that the b2-adrenoceptor
controls angiogenesis in the heart; in fact, cardiac adenoviralmediated
b2-adrenoceptor overexpression was associated with
reactivation of this mechanism, which is impaired in the
failing myocardium. The recovery of cardiac angiogenesis
Table 3
Cardiac b-adrenoceptor receptor signalling at 12 and 30 days from gene delivery
Sham HF Saline HF Ad-GFP HF Ad-b2-adrenoceptor
b-AR density (fmol mg-1 membrane protein)
12 day 67.8  22.1 23.2  6.6* 23.0  6.8* 345.5  82.2*õ
30 day 66.5  11.9 21.0  4.3* 24.1  5.6* 47.8  12.1*õö
Basal ISO 10-4 Basal ISO 10-4 Basal ISO 10-4 Basal ISO 10-4
Adenylyl-cyclase activity (pmol cAMPEmg-1Emin-1)
12 days 5.2  1.6 12.1  2.8 2.3  0.8* 5.4  1.0* 2.9  0.9* 6.1  1.2* 4.4  1.0õ 10.3  2.5õ
30 days 5.6  1.1 13.0  3.1 1.6  0.4*ö 4.2  0.9*ö 1.7  0.7*ö 4.8  0.9*ö 3.6  1.1*õö 7.2  2.0*õö
GRK2 protein expression (GRK2/GAPDH D.U.)
12 days 0.22  0.04 0.78  0.09* 0.80  0.12* 0.40  0.07*õ
30 days 0.18  0.03 1.26  0.24*ö 1.33  0.36*ö 0.75  0.08*õö
Sham, n = 7; HF-saline, n = 6; HF Ad-GFP, n = 7; HF Ad-b2-adrenoceptor, n = 8.
*P < 0.01 vs. sham operated.
õP < 0.01 vs. HF saline and HF-GFP.
öP < 0.005 vs. 12-day values in the same group.
HF = heart failure; GRK2 = G-protein coupled receptor kinase 2.
BJP G Rengo et al.
2356 British Journal of Pharmacology (2012) 166 2348.2361
properties prevents maladaptive LV remodelling and the
progression of cardiac dysfunction. Our data, in the context
of a clinically relevant model of HF, extend to the failing
myocardium what has been reported on the ability of
b2-adrenoceptor overexpression to enhance cardiac contractility
in vivo (Maurice et al., 1999; Liggett et al., 2000). Using
a transgenic mouse model of HF, it has been shown
that modest (30-fold) myocardial overexpression of
b2-adrenoceptors can prevent the development of hypertrophy
and ventricular dysfunction (Dorn et al., 1999). Furthermore,
selective b2-adrenoceptor stimulation reduced
apoptosis and increased contractility in a post-MI HF rat
model (Ahmet et al., 2004) and adenoviral-mediated
b2-adrenoceptor gene transfer enhanced cardiac performance
in healthy (Maurice et al., 1999) and unloaded rabbit failing
hearts after heterotopic cardiac transplantation (Tevaearai
et al., 2002). In our study, the overexpression of the receptor
was achieved by direct intramyocardial injection of Ad-b2-
adrenoceptors, a technique that allowed us to obtain a ~15-
fold increase in membrane b-adrenoceptors. Importantly,
these levels lie in the therapeutic range reported by other
authors who showed that the effects of b2-adrenoceptor overexpression
on cardiac function are dose-dependent and only
a moderate overexpression results in enhanced in vivo and in
vitro signalling and cardiac contractility (Liggett et al., 2000).
Furthermore, we explored the effects of b2-adrenoceptor
overexpression on cardiac b-adrenoceptor signalling. We
found improved b-adrenoceptor function at the receptor level
2 weeks post-gene therapy and, importantly, this positive
effect was still evident 4 weeks post-gene delivery.
To further support the notion that the b2-adrenoceptor
plays a crucial role in the regulation of the angiogenic mecha-
Figure 5
Shown is cardiac protein expression of VEGF, PKB, Ser473-phospho(p)-PKB, eNOS and Ser1177-phospho(p)-eNOS in all study groups at 2 (A) and
4 (B) weeks after gene therapy. Data between HF-saline and Ad-GFP were not statistically different and were pooled together and indicated as
HF-control. The expression of GAPDH was used as an internal control to normalize VEGF protein levels. p-PKB/PKB and p-eNOS/eNOS ratio
indicated respectively the levels of PKB and eNOS phosphorylation in the heart. Data are expressed as mean  SEM. *P < 0.05 vs. sham (n = 6
rats per each group, for each time point).
b2-adrenoceptor regulates angiogenesis in heart failure BJP
British Journal of Pharmacology (2012) 166 2348.2361 2357
nism in the heart, we studied b2-adrenoceptor-/- mice undergoing
MI. Previous observations indicated that total
elimination of both cardiac b1- and b2-adrenoceptors in mice
has little impact on chronotropy/inotropy (mainly controlled
by b1-adrenoceptor) or myocardial basal metabolism (controlled
by b2-adrenoceptor), although functional deficits are
clearly revealed by stress conditions, such as stimulation by
b-adrenoceptor agonists or maximal exercise (Rohrer et al.,
1999). In b2-adrenoceptor-/- mice, the reduction of antiapoptotic
defenses and the loss of preconditioning mechanisms
have been recognized as the main determinants of the
overtly exaggerated cardiomyopathy and dramatic mortality
in HF models (Bernstein et al., 2005). In the present study,
consistent with the results obtained in HF rats, we found for
the first time that the lack of b2-adrenoceptor signalling was
associated with a severe impairment of cardiac capillarization,
higher 30-day mortality and cardiac dysfunction compared
with control mice. Furthermore, we also showed that
cardiac capillary density is significantly reduced in the noninfarcted
heart of b2-adrenoceptor-/- mice with respect to
controls. The alteration in capillary density observed in
b2-adrenoceptor-/- mice is even more pronounced in the
post-MI failing heart lacking b2-adrenoceptors, also indicating
a defect in ischaemia-induced cardiac angiogenesis. Accordingly,
the more severe LV dysfunction and the higher mortality
after MI observed in KO mice strongly support the
protective role of cardiac b2-adrenoceptors against ischaemia.
In the present study, we explored the VEGF/PKB/eNOS
pathway, which is strongly implicated in cardiac growth and
angiogenesis. The ability of b2-adrenoceptors to stimulate PKB
Figure 6
(A) Post-MI 30-day mortality rate in HF b2-adrenoceptor (AR)+/+ and b2-adrenoceptor-/- mice (n = 35 mice for each group). (B) LV fractional
shortening FS (%) assessed by echocardiography at 4 weeks post-MI or sham operation in b2-adrenoceptor+/+ and b2-adrenoceptor-/- mice
(n = 15 for each group). (C) Capillary counts (expressed as total capillary density mm-2) in LV remote zone in all study groups (n = 8 for each group).
(D) Shown is the cardiac protein expression of Ser473p-PKB/PKB (left), Ser1177-p-eNOS/eNOS and (middle) and VEGF (right) in sham or HF
b2-adrenoceptor+/+ and b2-adrenoceptor-/- at the end of the study (n = 8 for each group). Data are expressed as mean  SEM. *P < 0.05 vs. sham
b2-adrenoceptor+/+; õP < 0.05 vs. HF b2-adrenoceptor+/+.
BJP G Rengo et al.
2358 British Journal of Pharmacology (2012) 166 2348.2361
has been previously demonstrated in neonatal and adult cardiomyocytes
(Morisco et al., 2000; Zhu et al., 2001), as well as
in ECs (Iaccarino et al., 2005). However, all these studies
focused on the role of b2-adrenoceptors in promoting
cardioprotection through activation of PKB, but none
of them examined the pro-angiogenic implications of
b2-adrenoceptor-dependent PKB activation in the heart. Most
importantly, PKB activation is able to induce coordinated
cardiac growth and angiogenesis only when accompanied by
a significant VEGF production and release. In fact, VEGF
inhibition early after PKB activation results in impaired coronary
angiogenesis and transition from adaptive to maladaptive
hypertrophy (Shiojima et al., 2005). Our group has
previously demonstrated that PKB is strongly activated immediately
after MI with a concurrent VEGF overproduction
(Leosco et al., 2008). In the post-acute phase of MI, PKB
activation is not accompanied by adequate VEGF production
leading to pathological remodelling, characterized by severe
LV dilatation and dysfunction (Leosco et al., 2008). Herein,
we showed that in HF, the overexpression of b2-adrenoceptors
induces a sustained and coordinated PKB and VEGF activation
that is completely inhibited by ICI 118551, a specific
b2-adrenoceptor antagonist. In this regard, in vitro studies
have demonstrated that overexpression of b2-adrenoceptors
in ECs induces VEGF production and EC proliferation (Iaccarino
et al., 2005). All these findings demonstrating the ability
of b2-adrenoceptors to activate PKB (Morisco et al., 2000; Zhu
et al., 2001; Iaccarino et al., 2005), as well as to induce in vitro
and in vivo VEGF production (Iaccarino et al., 2005; Leosco
et al., 2008), strongly suggest that the activation of the
pro-angiogenic pathway in the heart is mechanistically
related to increased b2-adrenoceptor signalling following
gene therapy. Interestingly, in b2-adrenoceptor-/- HF mice at
1 month post-MI, we found low VEGF protein levels, as
expected, but more pronounced PKB activation compared
with control.We speculate that the impaired angiogenic phenotype
observed in b2-adrenoceptor KO mice requires
enhanced PKB activation, although this is not accompanied
by VEGF production. As regards the explanation for the
reduced PKB and VEGF levels observed 4 weeks post gene
therapy, it is important to note that transgene expression
is almost exhausted at that time, thus lacking the
b2-adrenoceptor-mediated PKB and VEGF activation. It has
been recently proposed that b2-adrenoceptors can be cardioprotective
in acute myocardial injury via PKB/eNOS activation
and that GRK2 is a key regulator of this pathway in the
heart (Rohrer et al., 1999; Bernstein et al., 2005; Tong et al.,
2005). Interestingly, GRK2, which is up-regulated in several
animal models of cardiac diseases, has recently been shown
to bind and inhibit PKB with consequent reduced activation
of the downstream effectors, such as eNOS (Liu et al., 2005).
Moreover, GRK2 inhibition in the heart leads to enhanced
PKB/eNOS signalling and cardioprotection through activation
of NO-dependent anti-apoptotic/survival mechanisms
(Liu et al., 2005; Brinks et al., 2010). Of note, in our study,
cardiac b2-adrenoceptor overexpression induced an overall
improvement in cardiac b-adrenoceptor signalling and a significant
reduction in cardiac GRK2 protein levels. This might
represent an additional mechanism to explain cardiac
PKB/eNOS activation observed in b2-adrenoceptor infected
A potential limitation of the present study could be
the lack of a sustained transgenes expression; in fact,
b2-adrenoceptor overexpression was almost exhausted at 1
month from infection, consistent with previous studies utilizing
adenovirus. However, only 2 weeks of transgene expression
were adequate to activate angiogenesis in HF hearts.
Importantly, neovessel formation and maturation also continued
after the loss of transgene expression, as demonstrated
by the enhanced vascular network at 4 weeks from infection.
Importantly, capillary and arteriolar density at the end of the
study was even more pronounced than that reported at
2 weeks post-infection, the time of highest transgene
In summary, the present study demonstrates that replacing
dysfunctioning b2-adrenoceptors in the post-ischaemic
failing heart results in improved LV remodelling/function
and cardiac angiogenesis, which is inadequate in the failing
myocardium. This study adds novel additional mechanisms
for the beneficial effects of b2-adrenoceptors in HF and offers
potential insights about future therapeutic strategies for HF
treatment. In particular, we propose that an intervention
aimed at the stimulation of cardiac angiogenesis might have
valuable therapeutic consequences in the failing heart.
This work was supported in part by the Italian Ministry of the
University and Scientific Research, PRIN (Progetto di Ricerca
di Interesse Nazionale) to NF, DL, GR and CZ, and by postdoctoral
fellowships to GR from the American Heart Association



Myocardial ß2-adrenoceptor gene delivery promotes coordinated cardiac adaptive remodelling and angiogenesis in heart failure