PI3Kgamma protects from myocardial ischemia and reperfusion injury through a kinase-independent pathway.

Abstract

PI3Kgamma functions in the immune compartment to promote inflammation in response to G-protein-coupled receptor (GPCR) agonists and PI3Kgamma also acts within the heart itself both as a negative regulator of cardiac contractility and as a pro-survival factor. Thus, PI3Kgamma has the potential to both promote and limit M I/R injury.
Complete PI3Kgamma-/- mutant mice, catalytically inactive PI3KgammaKD/KD (KD) knock-in mice, and control wild type (WT) mice were subjected to in vivo myocardial ischemia and reperfusion (M I/R) injury. Additionally, bone-marrow chimeric mice were constructed to elucidate the contribution of the inflammatory response to cardiac damage. PI3Kgamma-/- mice exhibited a significantly increased infarction size following reperfusion. Mechanistically, PI3Kgamma is required for activation of the Reperfusion Injury Salvage Kinase (RISK) pathway (AKT/ERK1/2) and regulates phospholamban phosphorylation in the acute injury response. Using bone marrow chimeras, the cardioprotective role of PI3Kgamma was mapped to non-haematopoietic cells. Importantly, this massive increase in M I/R injury in PI3Kgamma-/- mice was rescued in PI3Kgamma kinase-dead (PI3KgammaKD/KD) knock-in mice. However, PI3KgammaKD/KD mice exhibited a cardiac injury similar to wild type animals, suggesting that specific blockade of PI3Kgamma catalytic activity has no beneficial effects.
Our data show that PI3Kgamma is cardioprotective during M I/R injury independent of its catalytic kinase activity and that loss of PI3Kgamma function in the hematopoietic compartment does not affect disease outcome. Thus, clinical development of specific PI3Kgamma blockers should proceed with caution.

Full text

Page 1

PI3Kc Protects from Myocardial Ischemia and
Reperfusion Injury through a Kinase-Independent
Pathway
Bernhard J. Haubner1., G. Gregory Neely2., Jakob G. J. Voelkl1, Federico Damilano3, Keiji Kuba2,
Yumiko Imai2, Vukoslav Komnenovic2, Agnes Mayr1, Otmar Pachinger1, Emilio Hirsch3, Josef M.
Penninger2*, Bernhard Metzler1
1Department of Internal Medicine III (Cardiology), Innsbruck Medical University, Innsbruck, Austria, 2Institute of Molecular Biotechnology of the Austrian Academy of
Sciences, Vienna, Austria, 3Department of Genetics, Biology and Biochemistry, Molecular Biotechnology Center, University of Torino, Torino, Italy
Abstract
Background: PI3Kc functions in the immune compartment to promote inflammation in response to G-protein-coupled
receptor (GPCR) agonists and PI3Kc also acts within the heart itself both as a negative regulator of cardiac contractility and
as a pro-survival factor. Thus, PI3Kc has the potential to both promote and limit M I/R injury.
Methodology/Principal Findings: Complete PI3Kc2/2mutant mice, catalytically inactive PI3KcKD/KD(KD) knock-in mice, and
control wild type (WT) mice were subjected to in vivo myocardial ischemia and reperfusion (M I/R) injury. Additionally, bone-
marrow chimeric mice were constructed to elucidate the contribution of the inflammatory response to cardiac damage.
PI3Kc2/2mice exhibited a significantly increased infarction size following reperfusion. Mechanistically, PI3Kc is required for
activation of the Reperfusion Injury Salvage Kinase (RISK) pathway (AKT/ERK1/2) and regulates phospholamban
phosphorylation in the acute injury response. Using bone marrow chimeras, the cardioprotective role of PI3Kc was
mapped to non-haematopoietic cells. Importantly, this massive increase in M I/R injury in PI3Kc2/2mice was rescued in
PI3Kc kinase-dead (PI3KcKD/KD) knock-in mice. However, PI3KcKD/KDmice exhibited a cardiac injury similar to wild type
animals, suggesting that specific blockade of PI3Kc catalytic activity has no beneficial effects.
Conclusions/Significance: Our data show that PI3Kc is cardioprotective during M I/R injury independent of its catalytic
kinase activity and that loss of PI3Kc function in the hematopoietic compartment does not affect disease outcome. Thus,
clinical development of specific PI3Kc blockers should proceed with caution.
Citation: Haubner BJ, Neely GG, Voelkl JGJ, Damilano F, Kuba K, et al. (2010) PI3Kc Protects from Myocardial Ischemia and Reperfusion Injury through a Kinase-
Independent Pathway. PLoS ONE 5(2): e9350. doi:10.1371/journal.pone.0009350
Editor: Alicia J. Kowaltowski, Instituto de Quı ´mica, Universidade de Sa ˜o Paulo, Brazil
Received September 25, 2009; Accepted February 1, 2010; Published February 22, 2010
Copyright: ? 2010 Haubner et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits
unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: Supported by grants from IMBA and EuGeneHeart (EU grant) to J.M.P. and from the O¨sterreichische Kardiologische Gesellschaft and Medizinischer
Forschungsfond Tirol to B.M. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing Interests: The authors have declared that no competing interests exist.
* E-mail: josef.penninger@oeaw.ac.at
. These authors contributed equally to this work.
Introduction
Cardiac damage in acute myocardial infarction results from a
disruption of blood flow to the heart (ischemia) that may be
followed by either spontaneous or therapeutically induced
restoration of blood supply (reperfusion). Myocardial ischemia/
reperfusion (M I/R) results in infarction due to intrinsic
cardiomyocyte death [1,2] and damage via excessive immune
activation [3,4,5]. Many inflammatory factors that build up during
injury exert their effects through G-protein coupled receptors
(GPCRs) [6] suggesting that mediators of GPCRs might be
critically involved in the M I/R damage response. One key
signaling pathway downstream of GPCRs is mediated via
phosphoinositide 3-kinase c (PI3Kc).
Phosphatidylinositol-3-OH kinases (PI3K) are lipid kinases that
convert phosphatidylinositol 4,5-bisphosphate (PIP2) into phos-
phatidylinositol (3,4,5)-trisphosphate (PIP3). In mammals, there
are four class I PI3K, three of which (a, b, and d) are primarily
activated by tyrosine kinase signaling pathways, whereas PI3Kc is
the principal PI3K that relays signals via GPCRs to downstream
pathways such as AKT and ERK/1/2 [7]. PI3Kc2/2mice
(lacking the catalytic subunit p110) are viable and in certain
experimental settings exhibit reduced inflammation [8,9] as the
PI3Kc pathway controls migration of inflammatory cells [10]. In
addition, we have previously shown that PI3Kc acts as an intrinsic
negative regulator of cardiac contractility [11,12]. Given its dual
roles in cardiomyocyte function and inflammation, it is not clear if
the net effect of PI3Kc loss is cardio-protective or cardio-
destructive during M I/R. For example, we have previously
shown that PI3Kc2/2mice are resistant to the detrimental effects
of chronic b-adrenergic receptor signaling [13]. Conversely,
PI3Kc2/2mice showed a worse outcome both in a model of
chronic myocardial ischemia without reperfusion [14] and chronic
pressure overload [12], while overexpression of catalytically
PLoS ONE | www.plosone.org1 February 2010 | Volume 5 | Issue 2 | e9350

Page 2

inactive PI3Kc protected from heart failure in this model [15].
Moreover, it has been reported that a PI3Kc blocker might
protect from M I/R injury due to inhibition of the inflammatory
response during reperfusion [16]. However, since such blockers
are not specific for PI3Kc and PI3Kc inhibitors are currently
developed as a novel treatment option for M I/R in humans, it is
paramount to genetically define the role of PI3Kc in cardiac
ischemia/reperfusion injury.
We therefore assessed the role of PI3Kc in M I/R injury using
PI3Kc2/2mice. We report that PI3Kc mutant mice develop
massive early and late stage cardiac M I/R injury. PI3Kc seems to
be the key kinase that activates the Reperfusion Injury Salvage
Kinase (RISK) pathway (AKT/ERK1/2) in the acute setting of M
I/R. Using bone marrow chimeras these effects were mechanis-
tically mapped to non-haematopoetic cells suggesting that the
effects are intrinsic to cardiac tissue. To assess whether these effects
are dependent on catalytic activity, we performed M I/R injury in
mice with a knock-in mutation in the PI3Kc kinase domain
(PI3KcKD/KD). PI3KcKD/KDmice developed cardiac injury
similar to wild type mice. Thus, PI3Kc protects the heart from
M I/R injury through a kinase-independent mechanism in vivo.
Results
Genetic Inactivation of PI3Kc Results in Severe Cardiac
Damage to M I/R Injury
To assess the role of PI3Kc in M I/R injury, we performed in
vivo reversible LAD ligation in sex and age matched PI3Kc2/2
(KO) mice and PI3Kc expressing wild type mice (WT). The extent
of injury and heart functions were assayed 3 hours, 24 hours, 1
week, and 3 weeks after initiation of reperfusion. Intriguingly, in
PI3Kc2/2mice serum Troponin T levels, a marker of cardiac
damage, were significantly increased after 3 hours of reperfusion
(Figure 1A), without a significant change in the number of
inflammatory cells infiltrating the damaged cardiac tissue
(Figure 1B). In parallel, we observed a massive increase of the
infarction size within the area at risk in PI3Kc2/2hearts at the
early stage of M I/R injury (assayed at 24 hours, Figure 1C–E).
Thus, immediately after M I/R the loss of PI3Kc expression
results in massive early heart damage.
It was possible that in the long term the early cardioprotective
effects of PI3Kc would not outweigh secondary pro-inflammatory
effects due to impaired migration of inflammatory cells into the
damaged heart; therefore, the net result of disrupting PI3Kc
expression could be cardioprotective. To address this possibility we
performed M I/R injury on wild type (WT) and PI3Kc2/2(KO)
mice and followed these mice for up to three weeks. One week
after M I/R injury, PI3Kc2/2mice still exhibited a massive
increase in infarction size as compared to the control cohort
(Figure 2A,B). Again we observed no significant differences in the
numbers of inflammatory cells infiltrating the cardiac tissue
between control and PI3Kc2/2mice (Figure 2C), suggesting that
PI3Kc is not required for recruitment of inflammatory cells to the
infarction site during the early (3 hours) as well as later stages (1
week) stages of M I/R injury.
Following the inflammatory phase of M I/R injury, scar
formation occurs at the infarction site. In both wild type and
PI3Kc2/2mice the primary damage site was replaced by scar
tissue 3 weeks after the initial M I/R injury (Figure 2D). In line
with the increased injury, collagen deposition and scar formation
was markedly increased in PI3Kc2/2mice (Figure 2D, E). We
next assessed cardiac function 3 weeks after M I/R injury.
Functionally, the excessive early damage and increased scar
formation in M I/R injured PI3Kc2/2mice culminated in a
severe reduction in cardiac contractility as defined by changed
fractional shortening from the baseline (Figure 2F). Thus, PI3Kc is
cardio-protective during M I/R injury and genetic disruption of
PI3Kc results in severe cardiac damage that is sustained over time
and severely compromises cardiac function.
PI3Kc Affects Cardiac M I/R Injury Independent of the
Haematopoietic System
PI3Kc controls multiple activation pathways in haematopoetic
cells including cell migration [7] and it has been proposed that
PI3Kc might be a key drug target to block inflammation in various
pathologic conditions including M I/R injury in heart attacks [17].
To directly test whether the protective effects of PI3Kc are due to
its function in the immune system, we generated bone marrow
chimeras. Interestingly, although PI3Kc has clearly been impli-
cated in inflammation in many other systems [7], wild type mice
given either wild type or PI3Kc2/2bone marrow developed
comparable M I/R injuries (Figure 3A, C) and showed equal
Troponin T release (Figure 3D). Moreover, PI3Kc2/2and wild
type immune cells accumulated to a similar extent at the site of M
I/R injury (Figure 3E).
In the reciprocal experiment, wild type or PI3Kc2/2bone
marrow was transferred into PI3Kc2/2recipients; in these bone
marrow chimeras we again observed comparable infarction sizes
(Figure 3B, C) and Troponin T release (Figure 3D). Irrespective of
the genotype of the transferred bone marrow cells, PI3Kc2/2
recipient mice exhibited similar inflammation at the site of
infarction (Figure 3E). Of note, in all bone marrow transplantation
experiments, PI3Kc2/2recipient mice developed markedly more
severe cardiac damage following M I/R injury as compared to
wild type host mice (Figure 3B–D). These data indicate that the
excessive damage seen in PI3Kc2/2mice following M I/R injury
occurs independent of its role in haematopoietic cells.
Loss of PI3Kc Results in Impaired RISK Signaling
Activation of the PI3K pathway leads to cell survival through
activation of the Reperfusion Injury Salvage Kinase (RISK)
pathway (AKT/ERK1/2) [2,10]. Similar to the growth factor
coupled class IA PI3Ks a,b, and d, PI3Kc can relay GPCR signals
to activate AKT and ERK1/2 in cardiomyocytes [11,12,13]. To
assess AKT and ERK activation in M I/R injury, we isolated
hearts from wild type and PI3Kc2/2mice prior to ischemia, after
30 minutes of ischemia, and 40 minutes as well as 24 hours after
initiation of reperfusion. Protein extracts were isolated from the
area at risk and immunoblotted for phosphorylated AKT (T308
and S473, Figure 4A) and phosphorylated ERK1/2 (T202/Y204,
Figure 4B), both indicative of activate AKT and ERK1/2.
As previously reported [11,13], wild type and PI3Kc2/2mice
exhibit comparable baseline phosphorylation of both AKT and
ERK1/2. Following 30 minutes of ischemia due to in vivo LAD
ligation, AKT and ERK1/2 activation were comparably reduced
in both genotypes. Ischemia-induced decrease in AKT and
ERK1/2 phosphorylation was rapidly restored in wild type
animals within 40 minutes of reperfusion. By contrast, in
PI3Kc2/2mice AKT and ERK1/2 phosphorylation were still
markedly reduced 40 minutes after reperfusion (Figure 4A, B).
These marked differences in AKT and ERK1/2 activation were
not apparent after 24 hours of reperfusion, indicating a role of
PI3Kc in the early initiation of pro-survival signaling through
AKT and ERK1/2. Immunohistochemical staining to detect
phosphorylated AKT (S473) showed that active AKT is localized
to the area at risk of wild type hearts (Figure 4C). Of note, in vivo M
I/R did not alter the PI3Kc protein content in the early phase of
reperfusion (Figure 4D). Thus, PI3Kc2/2mice fail to activate
PI3Kc in Ischemia/Reperfusion
PLoS ONE | www.plosone.org2February 2010 | Volume 5 | Issue 2 | e9350

Page 3

AKT and ERK1/2 immediately following M I/R injury, and this
lack of activation can be localized to the area at risk.
In addition to AKT and ERK1/2 activation, PI3Kc is a
regulator of myocardial contractility [11] via regulation of
phosphodiesterases and subsequent phospholamban (PLB) phos-
phorylation in the sarcoplasmatic reticulum [12,18]. Phosphory-
lation of PLB results in increased contractility [11]. Confirming
previous results, PI3Kc2/2hearts exhibit a much higher basal
level of phosphorylated PLB (Figure 4E). Whereas in wild type
hearts phosphorylated PLB was decreased following ischemia and
40 minutes of reperfusion, such decrease in phosphorylated PLB
was less pronounced in ischemic PI3Kc2/2hearts. Moreover, as
compared to wild type control mice, in PI3Kc2/2
phosphorylation of PLB was restored to basal levels during the
mice
vulnerable early stage (40 minutes) of reperfusion. Thus, loss of
PI3Kc results in marked dysregulation of the RISK pathway and
phospholamban.
Kinase Dependency of PI3Kc during M I/R
PI3Kc exerts kinase-dependent and kinase-independent func-
tions [12]. To address the issue of kinase dependency we subjected
wild type, PI3Kc2/2mice, and PI3Kc kinase-dead knock-in mice
(PI3KcKD/KD) to M I/R injury. Whereas PI3Kc2/2mice again
exhibited severe infarction following M I/R injury, PI3KcKD/KD
mice displayed cardiac damage that was comparable to wild type
control mice (Figure 5A–E). These differences between PI3Kc2/2
mice and PI3KcKD/KDas well as control mice were already
evident at 3 and 24 hours of reperfusion (Fig. 5A–C). PI3Kc2/2
Figure 1. Loss of PI3Kc results in massive myocardial ischemia/reperfusion induced heart damage. (A) Following M I/R injury, serum
TroponinT is significantly increased in PI3Kc2/2(KO) mice compared to wild type (WT) controls. Data are from 3 hours after reperfusion. n=13 per
group. (B) No significant difference in the numbers of inflammatory cells in the injured hearts of PI3Kc2/2and WT mice. Inflammatory cell infiltrates
were assessed 3 hours after reperfusion. n=7 per group. (C) No significant difference in the area at risk (AAR)/left ventricle (LV) in PI3Kc2/2and WT
mice. (D) PI3Kc KO mice display a markedly increased area of infarction/area at risk (AAR) 24 hours after reperfusion. n=6 per group for (C) and (D).
(E) Representative left ventricular TTC-stained sections from the basis (left) to the apex (right) of wild type (WT; upper panels) and PI3Kc2/2(KO;
lower panels) hearts after 24 hours of reperfusion. Pale areas represent necrotic regions (arrows). Bars indicate 1 mm. In all bar graphs mean values
+/2 SEM are shown. *p,0.05. **p,0.01. NS=not significant.
doi:10.1371/journal.pone.0009350.g001
PI3Kc in Ischemia/Reperfusion
PLoS ONE | www.plosone.org3 February 2010 | Volume 5 | Issue 2 | e9350

Page 4

mice again showed severely compromised cardiac contractility and
fractional shorting following M I/R injury. By contrast, the
observed reduction in MI/R-induced infarction size in PI3KcKD/KD
mice translated into impaired cardiac contractility that was again
comparable to that of wild type mice (Figure 5F, G). Thus, the
cardioprotective function of PI3Kc during M I/R injury appears to
be independent of its kinase activity. However, PI3KcKD/KDmice
exhibited a cardiac injury similar to wild type animals.
Discussion
In the current study we surprisingly observed little require-
ment for PI3Kc during inflammation associated with M I/R
injury. Genetic evidence has implicated PI3Kc as a pro-
inflammatory intracellular signaling component important for
neutrophil chemotaxis [8,9]. In vivo, PI3Kc2/2mice have been
shown defective in diverse models of inflammation, including
acute peritonitis [8,9] and delayed type hypersensitivity [19]. In
addition, PI3Kc2/2mice develop attenuated atherosclerosis
[20]. While loss of PI3Kc alone may slow neutrophil migration
[21], this loss does not appear to affect inflammation even in the
early stages of our M I/R injury model. It has been shown that
in vivo PI3Kc cooperates with PI3Kd to promote efficient
neutrophil migration in response to chemokines and pro-
inflammatory cytokines such as TNF-a [22]. Most importantly,
our bone marrow transplantation experiments exclude a
requirement for PI3Kc in cardiac inflammation following M
I/R injury.
Enhanced contractility by increased PLB phosphorylation
[23,24] and/or reduced RISK signaling via AKT and ERK1/2
[2] could contribute to the severe injury phenotype we observed in
PI3Kc2/2mice. We found that knock-in mice lacking the PI3Kc
kinase activity (PI3KcKD/KD) display an M I/R injury phenotype
that is similar to wild type animals. Therefore, the detrimental
outcome observed in PI3Kc2/2animals appears to be due to
kinase-independent events, possibly via phospholamban hyper-
phospholryation and hypercontractility. PI3Kc blockers have
entered clinical development for chronic inflammation and heart
attacks in humans [16,25]. Our data on PI3Kc in M I/R injury
suggest that specific inhibition of PI3Kc will not result in any
overall improvement concerning cardiac function and the extent
of injury. Moreover, we observed a loss of RISK activation in
PI3Kc2/2mice, suggesting that treatment with PI3Kc blockers
may interfere with these early pro-survival signals in some
contexts. Of note, the observed loss of immediate RISK activation
itself does not appear to compromise M I/R outcome in our
model. These data are consistent with other recent studies
questioning the cardioprotective role of RISK [26,27]. Similar
to our first in vivo evidence on M I/R injury using mutant mice
and kinase-dead knock-in mice, previous work has shown that
PI3Kc is an important signaling pathway that controls ischemic
preconditioning in vitro [28], arguing that if anything, strategies to
activate, not inhibit, PI3Kc may improve outcome for patients
with ischemic heart disease.
The present study provides evidence that PI3Kc is cardiopro-
tective in vivo in a mouse model of myocardial ischemia/
reperfusion. Surprisingly, the severe cardiac damage observed
in PI3Kc2/2mice occurs very rapidly and is not influenced by
PI3Kc expression within the hematopoietic compartment,
suggesting that PI3Kc exerts a cell autonomous protective
function in cardiomyocytes. Using kinase dead knock-in mice,
these detrimental effects were mapped to be independent of
PI3Kc lipid kinase activity. However, loss of PI3Kc catalytic
activity does not translate into a beneficial outcome but could
negatively influence RISK signaling during the acute phase of
reperfusion. In conclusion, our genetic data indicate that specific
inhibition of PI3Kc is not cardioprotective. Therefore, treatment
of acute infarctions with PI3Kc inhibitors should proceed with
caution.
Figure 2. Loss of PI3Kc results in long term myocardial
ischemia/reperfusion induced heart damage. (A) Areas of
infarction (arrows) in representative left mid-ventricular sections 1
week after M I/R injury. (H&E staining). Bars indicate 1 mm. (B) PI3Kc2/2
(KO) hearts show a markedly greater infarction size in the left ventricle
(LV) at 1 week after M I/R. n=7 per group. (C) No significant difference
between PI3Kc2/2and WT mice in the numbers of infiltrating
inflammatory cells 1 week after M I/R. n=7 per group. (D)
Representative left mid-ventricular sections stained with Trichrome
and Aniline blue to visualize collagen deposits (arrows) in hearts of
PI3Kc2/2(KO) and wild type (WT) mice 3 weeks after M I/R.
Magnifications 620. (E) Markedly increased scar tissue in PI3Kc2/23
weeks after M I/R injury. n=7 per group. (F) PI3Kc2/2mice exhibit a
significant greater loss in percentage fractional shortening (FS) relative
to their respective baseline FS after 3 weeks of M I/R compared to WT
controls. n=9 per group. In all bar graphs mean values +/2 SEM are
shown. *p,0.05. NS=not significant.
doi:10.1371/journal.pone.0009350.g002
PI3Kc in Ischemia/Reperfusion
PLoS ONE | www.plosone.org4 February 2010 | Volume 5 | Issue 2 | e9350

Page 5

Methods
Mice/Ethics Statement
PI3Kc2/2(p110c2/2) and kinase-dead knock-in PI3KcKD/KD
mice have been previously described [9,12] and were backcrossed
onto a C57BL/6 background for at least 8 generations. Only male
animals were used for the experiments. Animals were handled in
accordance with institutional guidelines all experiments were
approved by the Austrian Federal Ministry of Science and
Research (GZ: BMBWK 66.011/0061BrGT/2006), and in
accordance with the Guide for the Care and Use of Laboratory
Animals published by the US National Institutes of Health (NIH
Publication No. 85–23, revised 1996).
Surgical Procedure and Experimental Protocol
PI3Kc2/2, PI3KcKD/KD, and control mice were randomised
into a three-hour, 24-hour, one-week, and three-week reperfusion
groups and underwent in vivo reversible LAD ligation as described
[29,30]. Briefly, following sedation, tracheotomy, ventilation, and
thoracotomy, the LAD was reversibly ligated aided by a PE tube.
After 30 minutes of ischemia, the PE tube was removed and
reperfusion was visually confirmed. The overall peri- and post-
procedural mortality was below 10% (control mice: 5,7%;
PI3Kc2/2: 6,1%; PI3KcKD/KD: 9,7%).
Infarct Evaluation
Heparinized blood was collected from the retrobulbar vein
plexus. Plasma levels of troponinT were evaluated as a biomarker
for cardiac damage using a quantitative assay (Roche Diagnostics,
Austria). Moreover, hearts were fixed in 4% formaldehyde
overnight and embedded in paraffin. Three 5-mm sections were
cut from defined levels of the left ventricle, stained with
hematoxylin and eosin (H&E). Infarct sizes were analyzed using
Axio Imager M1 (Zeiss, Axiovert, Jena, Germany) and Tissue-
FAXS software package (TissueGnostics, Vienna, Austria). In
addition, formalin-fixed heart sections were stained with Tri-
chrome and Aniline Blue to detect differences in collagen
deposition and scar size. Using a light microscope (Axioskop2
equipped with AxioCam MRc5, Zeiss), Axio Vision 4.0 (Zeiss),
and Adobe Photoshop CS3, the mean values of inflammatory
infiltrate were determined within 3640 visual fields of each area of
infarction. To evaluate the size of the infarction in relation to the
Figure 3. PI3Kc functions in non-haematopoietic cells in myocardial ischemia/reperfusion injury. (A) Representative left mid-ventricular
histological sections (H&E staining) from WT mice receiving WT bone marrow (WTWT) and WT mice receiving PI3Kc2/2bone marrow (KOWT). Hearts
were isolated and analysed from mice 1 week after M I/R injury. Arrows indicate areas of infarction. (B) Representative left mid-ventricular histological
sections (H&E staining) from PI3Kc2/2mice receiving WT bone marrow (WTKO) and PI3Kc2/2mice receiving PI3Kc2/2bone marrow (KOKO) after 1
week of M I/R injury. Arrows point at areas of infarction. (C) No significant difference between WTWT and KOWT chimeras and no difference between
WTKO and KOKO chimeras in the size of infarction determined 1 week after M I/R injury. n=5 per group. (D) No significant difference between WTWT
and KOWT chimeras and no difference between WTKO and KOKO chimeras in TroponinT levels assayed 3 hours after reperfusion. n=7 per group. (E)
No significant difference between all 4 groups of chimeric mice in the numbers of infiltrating inflammatory cells 1 week after M I/R. n=5 per group. In
all histological pictures the bar indicates 1 mm. All bar graphs show mean values +/2 SEM. *p,0.05. NS=no significant difference.
doi:10.1371/journal.pone.0009350.g003
PI3Kc in Ischemia/Reperfusion
PLoS ONE | www.plosone.org5 February 2010 | Volume 5 | Issue 2 | e9350