Cardiovascular diseases

4.1 Introduction

Cardiovascular diseases which are a common complication of diabetes, account for
about 80% of diabetes mortalities. Coronary artery disease is the main cause of
increased cardiovascular mortality in diabetes. However, clinical and experimental
studies demonstrate that heart structure and function are directly affected by diabetes
even in the absence of coronary artery atheroma and hypertension. Diabetic
cardiomyopathy is suggested as a specific type of cardiomyopathy in diabetes, which
is independent of coronary artery disease and characterized by asymptomatic diastolic
dysfunction, fibrosis, and metabolic disturbance.

Ghrelin is a recently identified peptide hormone in stomach. Acylated ghrelin has a
post-translational modification with O n-octanoyl acid at serine 3 position. This
acylation for ghrelin is necessary to enable its binding with GHSR and for its several
endocrine effects such as the stimulation of growth hormone release. Des-acyl
ghrelin, another form of ghrelin, accounts for about 80%- 90% of circulating ghrelin.
Des-acyl ghrelin lacks the ability to stimulate the release of growth hormone due to
the lack of post-translational modification through acylation, but it has been reported
to play an important role in the cardiovascular system as well as in glucose and lipid
metabolism independent of acylated ghrelin. Acylated ghrelin administration
increases blood glucose level and triglyceride content in liver and white adipose
tissue, whereas there are several reasons to propose des-acyl ghrelin as a therapeutic
agent for diabetic cardiomyopathy in type 2 diabetic patients. Firstly, des-acyl ghrelin
inhibits glucose release and improves insulin sensitivity. Secondly, des-acyl ghrelin
promoted pancreatic β-cells and prevented streptozotocin-induced type 1 diabetes in
rats. Indeed, an analog of des-acyl ghrelin is in development by AlizéPharma for the

treatment of Type 2 diabetes and PraderWilli Syndrome.

Acylated ghrelin and des-acyl ghrelin have both been found to stimulate inhibition of
cardiac apoptosis through activation of PI3K/ Akt signaling in H9c2 cardiomyocytes.
There are ample of studies that show protective effects of acylated ghrelin on cardiac
ischemic injury, heart failure, hypertension, and myocardial infarction through
improved cardiac contractility and remodeling. However, there is limited
experimental evidence showing protective effects of des-acyl ghrelin on heart disease,
particularly on diabetic cardiomyopathy. Therefore, this study examined the effect of
peripheral administration of des-acyl ghrelin on Type 2 diabetic cardiomyopathy and
the underlying cellular and molecular mechanisms.

4.2.2 Experimental Protocol

Mice in non-diabetes group (db/+) and diabetes group (db/db) were further randomly
assigned to the following groups: db/+-saline (n = 7), db/+-DAG (DAG, des-acylated
ghrelin; n = 7), db/db-saline (n=7), and db/db-DAG (n=7). Mice assigned to treatment
groups (db/+-DAG and db/db-DAG) were exposed to intraperitoneal injection of des-
acyl ghrelin (Des-acyl ghrelin, Tocris Bioscience, USA) for ten consecutive days.
Mice in sham groups (db/+-saline and db/db-saline) were intraperitoneally injected
with the same volume of saline instead of des-acyl ghrelin. The previously reported
administered dosage of 100 µg/kg bodyweight of ghrelin injected twice daily was
adopted. After the experimental period of ten days, mice were euthanized by overdose
of ketamine and xylazine. The heart was immediately removed and washed with cold
phosphate buffered saline (PBS). Subsequently, the left ventricle was quickly
dissected and frozen in liquid nitrogen and stored at -80ºC for later analysis.

4.2.3 Masson’s Trichome staining
Collagen deposition in the left ventricle of mice was detected by Masson’s Trichome
staining. The procedure was described in chapter 2.

4.2.4 Protein Fraction Preparation
The protein fraction of cardiac muscles was prepared by adopting the previously
described protocol in chapter 2.

4.2.5 RNA Extraction and Real Time Quantitative PCR
Myocardial fibrosis regulatory factors (MMP-8 and Adiponectin) were examined in
cardiac tissues by quantitative RT-PCR analysis. Total RNA extraction and real time
PCR analysis was followed using the procedure described in chapter 2.

4.2.6 Western Blot Analysis
The protein expression of autophagic factors (Beclin1 and Atg5-Atg12 conjugation)
and prosurvival ERK-Akt signaling, AMPK and GSK3α/β signaling markers
(phospho-ERK1/2, ERK1/2, phospho-Akt, Akt, phospho-AMPK, AMPK, phospho-
GSK3α/β, and GSK3α/β) were evaluated in cardiac tissues by Western immunoblot
given in chapter 2.

4.2.7 Statistical Analysis
Data were expressed as mean ± standard error of mean. Statistical analysis was
performed using Statistics Package for Social Science (SPSS) version 11.0.
Differences among groups were evaluated using ANOVA followed by Turkey’s HSD
post hoc test. Statistical significance was set at P < 0.05.

4.3 Results
4.3.1 Cardiac function by echocardiography

The cardiac dysfunction in diabetic cardiomyopathy was indicated by the decreased
fractional shortening in diabetic mice (by 25%, P< 0.05) compared to non-diabetic
mice at the pre-intervention level. But this decreased cardiac function was
considerably improved by des-acyl ghrelin treatment from 53% to 65% in the left
ventricle fractional shortening (Fig 4.1).

4.3.2 Cardiac fibrosis and fibrotic regulatory factors

Collagen deposition in left ventricles of mice was shown in Fig 4.2. Excessive
accumulation of collagen shown in blue color was observed in the heart of diabetic
db/db mice compared to non-diabetic db/+ group, and alleviation of this cardiac
fibrosis induction was achieved by des-acyl ghrelin treatment. The anti-inflammatory
adiponectin expression was decreased in diabetic group (by 30%, P< 0.05), but this
decrease was not found by the treatment of des-acyl ghrelin (Fig 4.3). The transcript
expression of collagenase-2 (MMP-8) was significantly up-regulated in diabetic db/db
mice heart (by 15 fold, P<0.001) relative to non-diabetic db/+ mice group. However,
this up-regulation of MMP-8 transcript expression was not affected by des-acyl
ghrelin treatment (Fig 4.4). The MMP-13 transcript expression was suppressed in the
heart of diabetic db/db mice group (by 80%, P<0.001) compared to non-diabetic
control group. A reversal of MMP-13 transcript expression suppression was not
observed after des-acyl ghrelin treatment (Fig 4.5).

4.3.3 Cardiac inflammatory markers

The inflammatory process was identified in the development of diabetic
cardiomyopathy. Cardiac inflammation markers were measured in gene expression
levels by real time PCR. Apelin gene expression in heart was not significantly
affected among four groups (Fig 4.6, P>0.05). In diabetic db/db mice, there was an
increase in Wnt5a mRNA concentration (P<0.05 vs. non-diabetic control mice), des-
acyl ghrelin administration did not affect the increase of Wnt5a gene expression (Fig
4.7). TLR4 transcript expression levels in diabetic mice heart were significantly lower
than in non-diabetic control group (by 29.8%, P<0.05) as shown in Fig 4.8.

4.3.3 Cardiac autophagy markers: Beclin1 and Atg5-
Atg12 conjugation

The levels of autophagic markers in mice heart including Beclin1 and Atg5-Atg12
conjugation were measured in four groups. The expression of Beclin1 in heart tissue
was reduced (by 29%, P<0.05) in diabetic db/db mice group, but this reduction was
reversed by des-acyl ghrelin treatment (Fig 4.9). There was no significant change
observed in expression levels of Atg5-Atg12 conjugation in diabetic db/db mice heart
compared to non-diabetic db/+ control group. However, des-acyl ghrelin treatment
enhanced conjugation of Atg5 and Atg12 in diabetic db/db mice (by 149%, P< 0.05)
but not in non-diabetic db/+ mice (Fig 4.10).

4.3.5 Other factors: Endothelial dysfunction (Edn3),
cardiomyocyte differentiation (p21 and GATA6), and
FK506 binding protein (FKBP5 and FKBP10)

No significant difference was observed in cardiac FKBP5 mRNA expression levels
between diabetic db/db mice and non-diabetic control mice and no effects followed
the treatment of des-acyl ghrelin (Fig 4.14). The expression levels of Edn3, GATA6,
and FKBP10 genes in the heart were all increased in type 2 diabetes group compared
to control group, but remained unchanged after des-acyl ghrelin administration (Fig
4.11, 4.13, and 4.15). There was a reduction in p21 transcript expression levels in the
heart during diabetic cardiomyopathy, and the expression levels remained unchanged
by des-acyl ghrelin administration (Fig 4.12).

4.3.6 AMPK, Akt, ERK1/2 and GSK3α/β signaling

AMPK activity in the heart was indicated by the ratio of phospho-AMPK to total
AMPK. However, there was no significant inhibition of AMPK activity in diabetic
db/db mice group compared to non-diabetic control group. After ten days of des-acyl
ghrelin administration (experimental period) there were enhanced levels of AMPK
phosphorylation in diabetic db/db mice group (by 150%), but not in non-diabetic db/+
mice group (Fig 4.16). Pro-survival Akt signaling was significantly suppressed in type
2 diabetes, indicated by a significant decrease in the ratio of phospho-Akt to total Akt
(by 72%, P<0.01) in diabetic db/db mice group relative to non-diabetic db/+ mice
group. However, this decrease was not found in db/db mice treated with des-acyl
ghrelin (Fig 4.17).

Although the significant decrease of the ratio of phospho-ERK1/2 to total-ERK1/2
was not seen in type 2 diabetic mice hearts, des-acyl ghrelin administration
significantly activated ERK1/2 signaling in type 2 diabetic db/db mice group (by
327%, P< 0.01) compared to diabetic control (Fig 4.18). As a target of Akt signaling,
a similar change pattern with Akt phosphorylation was found in GSK3α/β signaling.
The ratio of phospho-GSK3α/β to total GSK3α/β was significantly reduced (by 40%,
P<0.05) in type 2 diabetic db/db mice group relative to the non-diabetic db/+ mice
group. However, des-acyl ghrelin treatment led to reversal of GSK3α/β
phosphorylation levels’ reduction in diabetic db/db mice (Fig 4.19).

4.4 Discussion
4.4.1 Des-acyl ghrelin improves cardiac dysfunction in
type 2 diabetic cardiomyopathy
Pathophysiology of Type 2 diabetic cardiomyopathy is multifactorial, and factors
such as accumulated collagen deposition, altered cardiac autophagy, metabolism, and
pro-survival signaling all contribute to the development of diabetic cardiomyopathy.
The major finding of the present study is that des-acyl ghrelin had protective effect to
the heart in db/db mouse model of type 2 diabetes from the progression of adverse
functional and structural changes by reversing the inhibition of pro-survival signaling.

At the age of 14 weeks, cardiac systolic function (fractional shortening) was
decreased in db/db mice. This reduction of fractional shortening was evidence of the
presence of type 2 diabetic cardiomyopathy consistent with previous findings. Current
information concerning beneficial effects of des-acyl ghrelin in experimental models

of cardiovascular diseases including diabetic cardiomyopathy is very limited. The
focus of many previously conducted in vivo studies has been on the effect of acylated
ghrelin on cardiovascular diseases. However, in this study we have for the first time
demonstrated that, administration of des-acyl ghrelin can alleviate cardiac dysfunction
induced by diabetic cardiomyopathy. This improvement was indicated by the
measurement of cardiac contractility.

4.4.2 Des-acyl ghrelin alleviate cardiac fibrosis in type 2
diabetic cardiomyopathy
Myocardial fibrosis is a critical structural change in diabetic cardiomyopathy. Cardiac
collagen deposition was obviously increased in untreated diabetic db/db mice, and this
finding was consistent with previous studies. Cardiac dysfunction may be an outcome
caused by fibrosis-induced myocardial stiffness. Des-acyl ghrelin administration
showed inhibitory effect on this increased collagen deposition in type 2 diabetic mice.

Cardiac fibrotic regulators including adiponectin, MMP-8, MMP-13 and Wnt5a were
investigated at gene expression level. Adiponectin is a multi-functional adipokine
secreted by the adipose tissue. It was suggested that adiponectin was involved in
regulating insulin sensitivity and energy metabolism. Recent findings showed the
protective role of adiponectin in fibrosis-related diseases. For example, adiponectin
exerts an inhibitory role in hepatic stellate cell proliferation and liver fibrosis.
Moreover, adiponectin was shown to prevent myocardial hypertrophy and fibrosis.
Consistent with previous findings, a reduction in adiponectin expression was observed
in diabetic db/db mice heart. This may partly contribute to the development of cardiac
fibrosis in type 2 diabetic cardiomyopathy. The observed attenuation of cardiac
fibrosis by des-acyl ghrelin may have resulted from the induction of adiponectin

expression in the heart following des-acyl ghrelin treatment in diabetic db/db mice.
Two of the fibrosis factors considered in this study belongs to a group of enzymes
known as matrix metalloproteinase (MMP) such as MMP-8 and MMP-13 which are
mostly involved in collagen cleavage in many connective tissues. However, MMP-8
and MMP-13 fibrosis factors are encoded by MMP-8 gene and MMP-13 gene
respectively. Expression and activity of collagenase-2 (MMP-8) are strongly
associated with chronic inflammation and fibrosis. MMP-8 acts as a profibrotic factor
and promotes fibrosis via TGF-β-independent pathway.

Wilson and co-workers reported that expression of MMP-8 gene was up-regulated by
type 2 diabetes in hearts of db/db mice. In concurrence with findings of this previous
study, increased expression of MMP-8 in the hearts of untreated db/db mice was
observed in this study. Alternatively, there was down regulation of MMP-13 in type 2
diabetic heart. MMP-13 is the primary collagenase found in rodent myocardial
samples and cleaves type II collagen more efficiently than types I and III. However,
changes in the expression of MMP-8 and MMP-13 were not affected by des-acyl
ghrelin administration. Our data implicate that significant accumulation of collagen in
diabetic heart partly resulted from up-regulation of MMP-8 and down-regulation of
MMP-13 due to lack of substantial collagenase-mediated cleavage.

4.4.3 Genes involved in the type 2 diabetic
cardiomyopathy but not affected by des-acyl ghrelin

A wide range of genes are involved in induction and development of type 2 diabetic
cardiomyopathy, but some are not yet affected by des-acyl ghrelin. Our data suggest
that induction and development of type 2 diabetic cardiomyopathy is influenced by
expression products of various genes including: Wnt5a, TLR4, Edn3, p21, GATA6,
and FKBP10. However, administration of des-acyl ghrelin shows no effect on the
expression of these genes. These findings indicated that improvement of cardiac
dysfunction in type 2 diabetes which is induced by des-acyl ghrelin treatment is not
associated with cardiac inflammation (Wnt5a), endothelial dysfunction (Edn3),
cardiomyocyte differentiation (p21 and GATA6) and FK506 binding proteins

Cardiac inflammation is an important mechanism underlying the development of type
2 diabetic cardiomyopathy. Therefore, there are several inflammatory markers that
have been used to evaluate cardiac inflammation including: Apelin, Wnt5a, and
TLR4. Apelin is a kind of adipocytokine secreted by white adipose tissue. Aplein and
its receptor are also expressed in cardiovascular system. However, our data indicate
that apelin signaling is not involved in type 2 diabetic cardiomyopathy and des-acyl
ghrelin effects.

However, a significant increase in the expression of Wnt5a was found in type 2
diabetes induced cardiomyopathy. Activation of Wnt5a/β-catenin-independent
signaling contributes to the stimulation of inflammation. However, a human study
reported that plasma Wnt5a levels were decreased in type 2 diabetes. The role of
Wnt5a/β-catenin-independent signaling in diabetic cardiomyopathy is not clear. In the

present study, we observed that there was a significant increase in the expression of
Wnt5a gene in diabetic mice heart, suggesting that Wnt5a may have contributory
effects to the progression of cardiac inflammation in type 2 diabetic cardiomyopathy.

Toll-like receptors (TLRs) are crucial components of the innate immune system.
TLR4 signaling leads to inflammation-induced cardiac injury and it is thought to be a
key mediator of the development of inflammation-related cardiovascular diseases.
TLR-4 deficient mice showed increased obesity, but with prevention of insulin
resistance induced by high-fat diet, our data gave evidence that expression of TLR4
gene in the heart was down-regulated by type 2 diabetes and did not respond to des-
acyl ghrelin treatment. However, the other components of TLR4 signaling such as
CD14, MyD88, and its downstream signaling (NF-kB pathway) were worthy further

4.4.4 Enhanced cardiac autophagy by des-acyl ghrelin in
type 2 diabetic cardiomyopathy
Autophagy is the process of controlling degradation and recycling of proteins and
cytoplasmic organelles. Beclin1 (mammalian Atg6) initiate the formation of the
autophagosomal membrane, whereas Atg5-Atg12 is responsible for the membrane
elongation. In the present study, cardiac autophagy was suppressed by type 2 diabetes
as indicated by reduced expression of Beclin1 and Atg5-Atg12 conjugation proteins.
The suppression of autophagy in diabetic heart results to accumulation of abnormal
proteins and organelles, which further leads to cardiac dysfunction. For the first time
we demonstrated that, the up-regulatory effect of des-acyl ghrelin on cardiac Beclin1
level and Atg5-Atg12 conjugation in vivo. The improved cardiac autophagy may
contribute to the protective effect of des-acyl ghrelin on type 2 diabetic

cardiomyopathy. Regulatory signaling including mTOR signaling needs to be
investigated further in the present study.

4.4.5 Activation of AMPK, ERK1/2, Akt and
GSK3α/βsignaling by des-acyl ghrelin in type 2 diabetic
We investigated the possible molecular mechanisms and intracellular pathways
involved in the protective effects of des-acyl ghrelin. Des-acyl ghrelin induced
enzyme 5’-AMP-activated protein kinase (AMPK) phosphorylation and improved the
impaired pro-survival signaling such as Akt, ERK1/2 and GSK3α/βin in diabetic
db/db mice heart tissue.

AMPK is a key regulator of cellular energy metabolism and it is activated by the
increased ratio of AMP to ATP. Apart from being the cellular energy sensor, AMPK
also regulates cardiac autophagy and it is thought to be important in the heart
signaling. An in vivo study conducted earlier using type 1 diabetic OVE26 mouse
model suggested that the suppression of AMPK activity might be related to the
reduction of cardiac autophagy and dysfunction. It is well known that an anti-diabetes
drug (metformin) exerts its cardio-protective effect in myocardial complications via
the stimulation of AMPK activity.
Des-acyl ghrelin did not show the effect of AMPK stimulation in murine HL-1 adult
cardiomyocyte. However, our extended in vivo data showed the effect of des-acyl
ghrelin on AMPK activation.

Des-acyl ghrelin showed cyto-protective effect on H9c2 cardiomyocytes, human
pancreatic islet microendothelial cells and rat visceral adipocytes through activation

of Akt and ERK1/2 signaling. Additionally, our data confirmed that des-acyl ghrelin
protected the heart against cardiac dysfunction by improving the suppressed Akt and
ERK1/2 signaling in Type 2 diabetic cardiomyopathy. Furthermore, the present study
also shows that des-acyl ghrelin inactivated GSK3α/β by enhancing its
phosphorylation. GSK3α/β is a down-stream target molecule for several pro-survival
protein kinases including Akt and ERK1/2. Our findings suggest that administration
of des-acyl ghrelin partly alleviated cardiac dysfunction via Akt/ ERK1/2/ GSK3α/β
signaling pathway in Type 2 diabetic cardiomyopathy.

5.1 Introduction
Diabetes and cancer are currently the leading causes of death in the world according
to World Health Organization report. There is a strong connection which exists
between type 2 diabetes mellitus and certain cancers. Type 2 diabetes increases the
risk and mortality of breast, liver, colorectal and pancreatic cancer. Cardiovascular
diseases are the common complications of Type 2 diabetes and are responsible for
50%- 80% of deaths in diabetic patients. Recent studies suggest the presence of
diabetic cardiomyopathy in Type 2 diabetes since this specific cardiomyopathy in
patients with diabetes was firstly reported by Rubler and co-workers in 1972. Diabetic
cardiomyopathy is characterized by myocardial hypertrophy and decreased systolic
and diastolic function of the left ventricle independent of coronary arteries or

Cancer patients with long standing diabetic cardiomyopathy may present a difficult
handing of cancer chemotherapy. This is attributable to the fact that most
chemotherapeutic agents used to treat cancer have severe side effect such as

doxorubicin induced cardiotoxicity. Doxorubicin is widely used for treatment of
several cancers including breast, stomach, lung, and bladder cancer. However, in most
cases doxorubicin induces a life-threatening and irreversible cardiac damage.
Therefore, a cancer patient with diabetic cardiomyopathy must be more sensitive to
doxorubicin induced cardiotoxicity because of the altered cardiac metabolism and
cellular function in diabetic heart.
Presently, there is no study that has reported the precise mechanism of diabetic heart
response to doxorubicin. Thus, no ideal therapeutic strategies that have been designed
to address this complicated clinical situation. This study mainly focused on the
investigation of doxorubicin cardiotoxicity in type 2 diabetic cardiomyopathy using a
mice model.

5.2 Methods

5.2.2 Experimental protocol

There were two experimental time points (four days and six days) in this study. The
survival of the mice 14 days after doxorubicin administration was analyzed using
Kaplan- Meier method.

Experiment 1:

Mice in non-diabetes group (db/+) and diabetes group (db/db) were randomly
assigned to the following groups: db/+-saline (n = 8), db/+-DOX (DOX, doxorubicin;

n = 8), db/db-saline (n=8) and db/db-DOX (n=8). Mice assigned to DOX group were
exposed to a intraperitoneal (i.p.) injection of DOX (Pharmacia & Upjohn SpA,
Milan, Italy) at a dose of 15 mg/kg to induce cardiotoxicity as shown in a previous
study (Suliman et al., 2007) . Mice in sham groups (Saline) were i.p. injected with the
same volume of saline instead of DOX. After the experimental period of four days,
mice were euthanized using an overdose of ketamine and xylazine. The heart was
immediately removed and washed with cold phosphate buffered saline (PBS)
followed by quick dissection of the left ventricle which was then frozen in liquid
nitrogen and stored at -80ºC for later analysis.

Experiment 2

The group design and intervention protocol was the same with experiment 1. The
experimental period was six days. Mice were sacrificed on day 7 and heart tissue was
collected using the same methods and procedure used in experiment 1.

5.2.3 Measurement of cardiac function by
The detailed procedure was described in chapter 2.

5.2.4 Protein Fraction Preparation
The method of protein fraction preparation from mice heart was described in chapter

5.2.5 Apoptotic Cell Death Enzyme-linked
Immunosorbent Assay (ELISA)
The procedure of Cell Death ELISA was described in chapter 2.

5.2.6 RNA Extraction and Microarray analysis
The method of total RNA extraction was showed in chapter 2. To understand the
whole transcriptional profile of doxorubicin effect on type 2 diabetic heart, microarray
technique was used in this study. RNA samples from mice in db/db-saline and db/db-
DOX groups in experiment 1 were performed using microarray analysis. RNA
samples (3μg) from the hearts of two mice in each group were pooled, and generated
four biological replicates in db/db-saline and db/db-DOX groups. Before performing
the cDNA array, the RNA quantity and quality was assessed. RNA quantity was
detected using NanoDrop 1000 Spectrophotometer. The purity of RNA was assured
by examining the OD260/280 ratio. The RNA integrity was assessed by Agilent 2100
bioanalyzer following the manufacturer’s instructions. RNA samples could be further
applied for array hybridization only if samples showed intact bands corresponding to
18S and 28S ribosomal RNA and RNA Integrity Number (RIN) was more than seven.

Microarray experiments were performed at the Agilent Service Platform with Agilent
two-color mouse 4X44k microarray slides. Doxorubicin treatment in diabetes group
(db/db-DOX) was labeled by Cy5 dye (red channel) while diabetic control group
(db/db-saline) was labeled by Cy3 dye (green channel). 500ng of Cy3-labeled and
Cy5-labeled cRNA were mixed and incubated with the Agilent microarray slide
(G2519F) for 17 hours at 65°C in the dark. The slide was then washed and scanned
using an Agilent DNA microarray scanner. Raw data was obtained using Agilent’s

Feature Extraction Software. Further analysis of the raw data was used by
comprehensive R- and Bioconductor-based web service for microarray data analysis.
Several normalizations were performed for raw data preprocessing including
background correction using subtract method, removal of dye bias by lowess
normalization and multiple testing correction by BH adjusted p-values for the
Benjamini & Hochberg (Benjamini and Hochberg, 1995) step-up FDR controlling
procedure. Gene expression values were calculated using log base 2 ratio of red
channel intensity (mean) and green channel intensity (mean). Functional classification
of highly regulated genes was analyzed by the GeneOntology database.

5.2.7 Real Time Quantitative PCR analysis
Selected genes from microarray analysis (S100A8, S100A9, MMP-8, and collagen Ⅰ)
was confirmed by real time quantitative PCR. The protocol and condition of PCR was
given in chapter 2.

5.2.8 Western Blot Analysis
The protein expression of apoptotic markers (Bcl-2 and Bax) and autophagic factors
(Beclin1 and LC3ⅠⅡ) in mice heart was measured by Western Blot. The details were
described in chapter 2.

5.3 Results
5.3.1 Survival analysis
Survival of db/+-DOX and db/db-DOX mice was analyzed using the Kaplan–Meier
approach ( Fig . 5.1). Fourteen days after doxorubicin treatment, survival rate was
significantly lower in type 2 diabetes mice with DOX treatment compared with non-
diabetes mice with DOX mice (60% vs. 10%, P < 0.001). The survival rate was also

significantly decreased in doxorubicin-treated mice compared with untreated controls
(db+-DOX vs. db/+-saline, 60% vs. 100%, P < 0.01; db/db-DOX vs. db/db-saline,
10% vs. 100%, P < 0.001).

5.3.2 Fractional shortening of left ventricle
There was doxorubicin induced cardiac dysfunction in non-diabetic db/+ mice on day
5, which was indicated by a significant reduction from 69.3% to 52.1% in cardiac
fractional shortening (Fig 5.2). The cardiac dysfunction resulting from type 2 diabetes
was evident because of a significant decrease of fractional shortening from 69.5% in
non-diabetic db/+ mice to 55% in diabetic db/db mice at pre-intervention level.
However, further decrease of fractional shortening by doxorubicin was not found in
diabetic db/db mice on day 5. On the second experiment, doxorubicin significantly
induced further damage of cardiac dysfunction with the presence of diabetic
cardiomyopathy in db/db mice on day 7 in comparison to non-diabetic db/+ mice with
doxorubicin treatment, which was indicated by a decrease of fractional shortening
from 53.1% to 39.9% (Fig 5.3).

5.3.3 Cardiac apoptosis and apoptotic regulatory factor
in experiment 1
The level of apoptotic DNA fragmentation was increased (by 396%; P< 0.05, Fig 5.4)
in non-diabetic db/+ mice with doxorubicin group relative to non-diabetic control
group on day 5. However, doxorubicin did not significantly induce the increase of
apoptotic fragmentation in type 2 diabetic db/db mice. The protein content of anti-
apoptotic Bcl-2 in mice heart was slightly regulated by doxorubicin in both non-
diabetic db/+ mice and diabetic db/db mice on day 5, but the change was not
significant (Fig 5.5).

However, protein content of pro-apoptotic Bax in heart was significantly increased by
doxorubicin compared to control group in non-diabetic mice group (by 163%, P<
0.05) and diabetic mice group (by 177%, P< 0.05) (Fig 5.6). The protein expression
ratio of Bcl-2 to Bax was further calculated and was found to be slightly lower in non-
diabetic mice with doxorubicin group compared to control group, but the change is
not statistically significant (P> 0.05, Fig 5.7). No significant change was found in the
ratio of Bcl-2 to Bax in the heart of diabetic mice with doxorubicin group compared
to the diabetic control group (Fig 5.7).

5.3.4 Cardiac Autophagic Factor: Beclin1 in experiment
The protein content of autophagy maker Beclin1 in heart was significantly decreased
(by 46%, P< 0.05) in type 2 diabetic db/db mice group relative to non-diabetic db/+
mice group (Fig 5.8). The regulation of Beclin1 protein content in the heart was not
found in non-diabetic and diabetic mice treated with doxorubicin.

5.3.5 Gene profiling analysis of doxorubicin effect on
type 2 diabetic heart by microarray
Gene expression profiles in left ventricle of type 2 diabetic mice with doxorubicin
treatment and without treatment were compared by microarray analysis. Only genes
whose transcriptional level was regulated 2-fold or higher and identified as a
significantly regulated were screened as differentially expressed genes. According to
the criteria, there are 709 genes in total that were significantly affected by doxorubicin
administration in db/db mice. Of these, 408 genes were up-regulated and 301 genes
were down-regulated by doxorubicin administration in type 2 diabetic db/db mice.
Selected genes expression with significant alteration by doxorubicin administration in

type 2 diabetic heart was shown in Table 5.1. Functional enrichment analysis by
GeneOntology demonstrated that sets of highly regulated genes were involved in a
spectrum of cellular component, biological process and molecular function (Table 5.2,
5.3, and 5.4).

5.3.6 Confirmation of selected genes in experiment 1
and 2 by real time PCR

The transcript expression of target S100A8 and S100A9 from microarray analysis was
confirmed by real time PCR extended to two experimental time points, that is, day 5
and day 7 in non-diabetic mice group and diabetic mice group. The expression of
S100A8 gene was only significantly elevated by administration of doxorubicin in
diabetic mice on day 5 (by 2.8 fold) and day 7 (by 23 fold) relative to diabetic control
(Fig 5.9). This suggested that S100A8 was not involved in doxorubicin-induced
cardiotoxicity in non-diabetic mice.

The transcript expression of S100A9 in heart was up-regulated by doxorubicin in non-
diabetic db/+ mice on day 5 (by 24.1 fold, P< 0.001) and no further change which was
observable on day 7 (Fig 5.10). Similarly, the transcript expression of S100A9 in
heart was considerably increased by doxorubicin in diabetic db/db mice on day 5 (by
6.7 fold, P< 0.01), but dramatically elevated by doxorubicin on day 7 (by 46.3 fold,
P< 0.001). Moreover, type 2 diabetes (diabetic db/db mice group) also induced the
up-regulation of S100A9 gene expression (by 20.2 fold, P< 0.001) compared to non-
diabetic group (Fig 5.10).

5.4 Discussion

5.4.1 Further decrease of cardiac function on day 7 in
diabetic heart after doxorubicin treatment
It is well known that doxorubicin induced acute cardiac toxicity in vivo. Our data
showed doxorubicin induced cardiac dysfunction in non-diabetic mice on day 5 after
doxorubicin treatment. The db/db mice developed type 2 diabetes with increased body
weight, hyperinsulinemia and hyperglycemia from the age of 12 weeks accompanied
with decreased cardiac contractility. Consistent with the findings of previous studies,
cardiac functional data in this study showed that there was suppressed fractional
shortening in diabetic db/db mice when compared to non-diabetic db/+ mice at pre-
intervention level.

Up to now, there is no study which has investigated doxorubicin effect on type 2
diabetic heart. We hypothesized that the diabetic heart was more sensitive to
doxorubicin and showed further damage of cardiac function. Interestingly, our data
indicated that there was further depression of fractional shortening observed on day 7
after doxorubicin administration, but not day 5 in type 2 diabetic db/db mice. It was
suggested that doxorubicin indeed caused more severe cardiac damage with the
background of Type 2 diabetic cardiomyopathy than in non-diabetic heart, but there
was a delay of the time of cardiac dysfunction induced by doxorubicin to seven days.

5.4.2 Underlying mechanism of doxorubicin induced
cardiotoxicity in type 2 diabetic heart

Extensive evidence proposes that cardiac apoptosis contributed to the development of
cardiomyopathy by doxorubicin. Myocardial apoptosis is thought to be an essential
determinant of cardiac pathogenesis because it results to a loss of contractile units,
conduction disturbances, compensatory hypertrophy of myocardial cells and fibrosis.
Inhibition of cardiac apoptosis has become a potential therapeutic target in treatment
of cardiomyopathy induced by doxorubicin. In agreement with the findings of
previous studies, cardiac apoptosis which occurred after administration of doxorubicin
was obviously induced by doxorubicin in non-diabetic db/+ mice on day 5. However,
the induction of apoptosis was not observed in diabetic db/db mice. This may partly
explain that cardiac dysfunction in diabetic mice was not further suppressed by
doxorubicin on day 5.

It was proposed that cardiac autophagy was involved in cardiac physiological and
pathological conditions such as myocardial hypertrophy, ischemia reperfusion injury
and heart failure. Although autophagy activation or inhibition in diabetic heart was
not consistent in different experimental conditions, the suppressed autophagy shown
by down-regulation of Beclin1 expression was observed in diabetic heart in the
present study. The inhibition of autophagy may further induce accumulation of
abnormal proteins and organelles resulting to cardiac dysfunction. However, this
inhibition of cardiac autophagy was not affected by doxorubicin in diabetic heart on
day 5. In type 1 diabetic heart, the pharmacokinetics and acute cardiotoxicity of
doxorubicin were altered by hyperglycemia. Thus, we hypothesized that the molecular
mechanisms underlying diabetic heart response to doxorubicin at early stage have
already been totally distinct from doxorubicin induced cardiotoxicity and type 2
diabetes induced cardiomyopathy. We next investigated the whole transcriptional

profiling analysis to understand how doxorubicin affect type 2 diabetic hearts at gene
transcription level by using microarray analysis.

Our microarray data first revealed the unique genes that were involved in exerting
doxorubicin effects on the hearts with type 2 diabetic backgrounds. Combined with
Gene Ontology analysis, it was found that these genes that are significantly regulated
were mainly responsible for several cellular mechanisms such as cardiac remodeling
and matrix, inflammatory or immune response, DNA/RNA stability and repair,
oxidative stress, metabolism and specific signal transduction. The early transcriptional
response of doxorubicin-treated diabetic hearts showed several novel and potential
targets such as S100 calcium binding protein A8 (S100A8), S100 calcium binding
protein A9 (S100A9) and MMP-8. S100A8 (A8) and S100A9 (A9), the member of
the S100 family, contains two calcium binding sites and it is involved in the
inflammatory process and immune diseases.

A recent human study reported that S100A8/A9 complex was a useful biomarker for
the prediction of one year mortality in elderly patients with severe heart failure.
Moreover, an in vivo study showed that activation of S100A8/A9 was crucial for the
development of post-ischemic heart failure via activation of the receptor of advanced
glycation end products (RAGE) and Toll-like receptors. Therefore, it may suggest
that S100A8/A9 played an important role in the pathophysiology of heart diseases.
Our data showed that the expression of S100A8 and A9 genes was significantly up-
regulated by doxorubicin in mice heart with Type 2 diabetic cardiomyopathy on day 5
using microarray analysis and subsequent confirmation using real time PCR analysis.
Corresponding with the further damaged cardiac function by doxorubicin on day 7,

the transcriptional levels of S100A8 and A9 in mice heart were dramatically increased
by 23 fold and 46 fold respectively compared to diabetic db/db control. These
findings strongly suggested that S100A8 and A9 were crucial for the development of
cardiac dysfunction in diabetic heart in response to administration of doxorubicin. The
precise mechanisms involved are worthy further investigations to enable the design of
therapeutic agents targeting this specific cardiac injury.

6.1 Main findings
6.1.1 Aims of the thesis

The general aims of this thesis were (1) to identify the protective effects of des-acyl
ghrelin on doxorubicin- induced cardiomyopathy in a mouse model, (2) to investigate
the beneficial roles of des-acyl ghrelin in type 2 diabetic cardiomyopathy using a
murine model, (3) to further elucidate the potential mechanisms behind the toxic
effects of doxorubicin on type 2 diabetic hearts.

6.1.2 Summary of main findings

The work described in chapter 3 clearly demonstrated that des-acyl ghrelin protects
cardiomyocytes against doxorubicin-induced cardiomyopathy by preventing the
activation of cardiac fibrosis and apoptosis and the effects are mediated through
GHSR-independent mechanism. The study in chapter 4 firstly provided the in vivo
evidence that des-acyl ghrelin can protect the heart against type 2 diabetic
cardiomyopathy through improvement of cardiac fibrosis, abnormal autophagy, and
cardiac inflammation. The work described in chapter 5 focused on doxorubicin-

induced toxic effects on type 2 diabetic heart. In this study, microarray analysis was
performed for investigation of potential mechanisms.
The main findings of three studies were summarized in Fig 6.1.

6.2 Cardiac dysfunction of cardiomyopathy
(doxorubicin- and type 2 diabetes-induced)

In cardiomyopathy, weakening of the heart muscle is structurally and functionally
abnormal and results to fatal heart failure. In order to maintain the systolic and
diastolic function, the heart can develop adaptive responses to several stimuli. Cardiac
morphology change and dysfunction occurred when compensatory mechanisms to
heart failure occurred in order to keep the heart’s normal contractile and relax ability.
Experiments in present thesis focuses on the investigation of potential therapeutic
targets and mechanisms of the two acquired cardiomyopathy induced by Type 2
diabetes and doxorubicin. The high prevalence of these two forms of cardiomyopathy
result from their specific causes: cancer chemotherapy and Type 2 diabetes.

Numerous human studies have demonstrated that doxorubicin leads to acute and
chronic dilated cardiomyopathy indicated by the decreased ejection fraction in cancer
patients. Cancer patients under doxorubicin treatment showed low ejection fraction,
which may have resulted from a reduction of T-cell leukemia/lymphoma 1A (TCL1A)
levels further leading to increased apoptotic sensitivity. Our data also confirmed that
doxorubicin results to acute cardiomyopathy on day 5 indicated by the reduced
fractional shortening and ejection fraction in a mice model. The dosage of
doxorubicin used in chapter 3 was well-established and adopted from previous animal

studies on doxorubicin-induced cardiomyopathy. The work described in chapter 4
focused on cardiovascular complications of type 2 diabetes: diabetic cardiomyopathy.
Patients with type 2 diabetic cardiomyopathy demonstrated the presence of systolic
functional disorders indicated by the decrease in ejection fraction, fractional
shortening, and cardiac output. The cardiac dysfunction was also observed in several
animal models of type 2 diabetes including db/db mice and zucker diabetic rats.
Db/db mice, leptin receptor deficient mice, were used in chapter 4 of the study and
showed cardiac systolic dysfunction from the age of 12 weeks after they were
measured using echocardiography.

There is limited information about doxorubicin-induced cardiac toxicity with the
background of type 2 diabetic cardiomyopathy. An in vivo study showed that the
pharmacokinetics of doxorubicin was affected by hyperglycemia in streptozotocin-
induced type 1 diabetic rats, and the accumulation of doxorubicin in diabetic heart
was increased. Therefore, we hypothesized that doxorubicin will further impair
cardiac function with diabetic cardiomyopathy in type 2 diabetic db/db mice.

However, our functional data showed that db/db mice maintain the same level of
cardiac dysfunction on day 5 after doxorubicin administration, further decrease of
cardiac systolic function was suddenly observed in day 7 after doxorubicin treatment
in db/db mice. We predicted that different cellular mechanisms were responsible for
the toxic effects of doxorubicin in type 2 diabetic heart. It is worthy further
investigations in order to determine the precise mechanisms and provide more
effective therapeutic targets.

6.3 Potential mechanisms of cardiomyopathy induced
by doxorubicin and Type 2 diabetes

Different forms of cardiomyopathy exhibit several common responses including
cardiac apoptosis, autophagy, fibrosis, and altered metabolism via signaling pathways.

6.3.1 Cardiac cell death

Adult heart cell has a limited ability of regeneration and repair; therefore it is sensitive
to intrinsic and extrinsic stimuli. Two types of cardiac cell death, apoptosis and
autophagy, are the key cellular events in the progression of heart diseases including
cardiomyopathy, heart failure, myocardial infarction, and ischemia/reperfusion. Cardiomyocyte apoptosis

Experimental studies have shown that apoptosis played important roles in
cardiomyopathy such as dilated and hypertensive cardiomyopathy. Cardiac apoptosis
has been well demonstrated in cardiomyocyte treated with doxorubicin and the animal
model of doxorubicin-induced acute and chronic cardiomyopathy. In chapter 3, a
single treatment of doxorubicin resulted in the activation of apoptosis evidenced by
the increased TUNEL index, apoptotic DNA fragmentation and caspase-3 activity.
This is in agreement with the findings of previous studies that adopted a similar
doxorubicin treatment dosage.

Apoptosis is regulated by extrinsic and intrinsic pathways, which are both
investigated in cardiomyocyte. It has been suggested that mitochondrial-mediated
intrinsic apoptotic pathway was involved in doxorubicin-induced cardiotoxicity in
vivo and in vitro. Bcl-2 and Bax are two crucial regulators in the intrinsic apoptotic
pathway. Permeability of mitochondrial membrane was stabilized by homodimers of
Bcl-2 against the release of cytochrome C and further caspase-3 activation. However,
this protective effect of Bcl-2 was diminished when the Bax-Bcl-2 heterodimer
formatted. Moreover, homodimers of Bax can promote the activation of caspase-3
independently via creating pores in the outer membrane and inducing the efflux of
cytochrome C. Consistently, our data showed that activation of apoptosis induced by
doxorubicin was associated with elevated abundance of Bax and the ratio of anti-
apoptotic Bcl-2 to pro-apoptotic Bax in the mice heart. Our finding together with the
findings of previous studies suggested that mitochondrial-mediated apoptosis was
responsible for the doxorubicin-induced cardiac dysfunction. Cardiomyocyte autophagy

Autophagy regulates the abundance and quality of intracellular components through
delivery of organelles and proteins. Autophagy also provides cells with nutrients and
energy under the condition of starvation. Therefore, autophagy is a survival process
compared to apoptosis. Dysfunctional autophagy has been observed in Type 1 and
Type 2 diabetic cardiomyopathy. Six month-old OVE26 mice, type 1 diabetic mouse
model, showed the suppressed autophagic activity in heart characterized by a decrease
in expression of lipidated microtubule-associated protein1 light chain 3 (LC3-Ⅱ).
Moreover, autophagy in cardiomyocytes isolated from db/db mice was impaired as

shown by the lowered level of Beclin-1 compared to non-diabetic cardiomyocytes.
Consistently, our in vivo data confirmed that db/db mice exhibited decreased cardiac
Beclin1 expression measured by Western Blot compared to non-diabetic db/+ mice.
Beclin1 is crucial for the early stages of autophagosome formation, and its interaction
with class III type phosphoinositide 3-kinase (PI3KC3)/Vps34 is responsible for the
localization of other autophagic proteins to the autophagosome. The down-regulation
of Beclin1 in heart induced by type 2 diabetes indicated that autophagy was under the
basal level and induced the accumulation of abnormal organelles and proteins in the
mice heart.

Furthermore, the modification of Beclin1 is also involved in the regulation of
autophagy. Beclin1 can be phosphorylated by death-associated protein kinase
(DAPK) on Thr 119 at its BH3 domain, which promotes the dissociation of Beclin1
from Bcl-XL and subsequently inducing autophagy. In addition, ubiquitination of
Beclin1 by the tumor necrosis factor receptor-associated factor 6 (TRAF6) initiates
the formation of autophagosomes via activation of PI3KC3. Thus, it is necessary for
future investigation of the interaction of Beclin1 and Bcl-XL, phosphorylation and
ubiquitination of Beclin1 in type 2 diabetes induced cardiac autophagy.

6.3.2 Cardiac fibrosis

The fibrotic process is important in pathologies of different cardiovascular diseases.
Myocardial fibrosis contributes to ventricular stiffening and impairs the systolic and
diastolic function. Cardiac fibrosis was detected by T1 magnetic resonance imaging in
patients with doxorubicin-induced cardiomyopathy and Type 2 diabetic

cardiomyopathy. This fibrotic change can be reproduced in experimental models of
two kinds of cardiomyopathy. Unsurprisingly, excessive collagen deposition was
observed in mice heart treated with doxorubicin described in chapter 3 and heart of
type 2 diabetic mice described in chapter 4 by Masson’s Trichome staining. However,
the mechanisms of myocardial fibrosis in doxorubicin-induced and type 2 diabetic
cardiomyopathy were different. In the acute doxorubicin-induced cardiotoxicity,
myocardial fibrosis occurred as a response to cardiomyocyte loss, whereas cardiac
fibrosis is thought to be an adaptive process to the chronic pathophysiological stimuli
such as inflammation in type 2 diabetic heart. In addition, the cross-link of collagen
with advanced glycation end products (AGEs) has also been reported to aggravate
ventricular stiffness.

Transforming growth factor- β1 (TGF-β1), a potent pro-fibrotic stimulator of ECM
production and deposition, is thought to partially contribute to the fibrosis in heart
disease. Inhibition of TGF-β1 or TGF-β1 signaling appears to be the potential
therapeutic target for hepatic and cardiac fibrosis. However, the regulation of TGF-β1
was not involved in the model of doxorubicin induced cardiotoxicity.
Correspondingly, there was no significant effect of doxorubicin on the expression of
TGF-β1 gene in the heart shown in the current model. In contrast to doxorubicin-
induced cardiotoxicity, an increase in the expression of TGF-β1 protein was
demonstrated in the myocardium of rodents with diabetic cardiomyopathy. TGF-β1
activity and signaling in type 2 diabetic heart needs to be investigated in the future.

Another profibrotic cytokine, connective tissue growth factor (CTGF), acts as a
downstream and cooperative mediator of TGF-β1 in the fibrogenic process. We

observed that doxorubicin induced the up-regulation of CTGF expression in mice
heart. This is in agreement with the notion that fibrosis in renal and heart may not be
fully regulated by TGF-β, but greatly depended on CTGF. Doxorubicin also
substantially increased the ratio of CTGF to brain natriuretic peptide (BNP), which is
an antifibrotic factor through guanylyl cyclase-A activation or interacting directly
with CTGF. The balance between CTGF and BNP in cardiomyocyte was suggested to
be a central determinant of cardiac fibrosis. These findings in the work presented in
chapter 3 suggested that CTGF and its interaction with BNP played an important role
in doxorubicin induced fibrotic injury through a TGF-β-independent pathway.

Adiponectin is an adipocyte-derived cytokine that plays the anti-diabetic and anti-
fibrotic roles. Plasma adiponectin levels are suppressed in type 2 diabetes patients. In
cardiovascular system, adiponectin levels in circulation have been negatively
associated with the risk of myocardial infarction. Adiponectin has been demonstrated
to be expressed in cardiac myocyte and rodent heart. Deregulation of cardiac
adiponectin plays a role in the pathogenesis of dilated cardiomyopathy. Gene
expression of adiponectin was found to be significantly decreased in leptin-deficient
db/db mice heart with severe inflammatory damage. Similarly, our current data
showed that cardiac adiponectin mRNA level was significantly down-regulated in
type 2 diabetes db/db mice, implicating the role of myocardial adiponectin system in
the pathophysiology of type 2 diabetic cardiomyopathy. Our findings confirmed the
findings of a previous study that adiponectin gene expression was decreased in db/db
mice heart by 4.59 fold compared to db/+ mice using microarray analysis. Moreover,
severe cardiac fibrosis was observed in angiotensin II-infused adiponectin knockout
mice via activation of PPAR-alpha. Taken together, myocardial fibrosis in type 2

diabetic mice may partly result from the reduced adiponectin expression.

6.4 Protective effect of des-acyl ghrelin on

In the present work, des-acyl ghrelin exhibited protective effects on doxorubicin-
induced and type 2 diabetic cardiomyopathy including improvement of cardiac
dysfunction, cardiomyocyte apoptosis, abnormal autophagy, and myocardial fibrosis
via activation of Akt/ERK1/2 signaling pathway in the mice model.

Des-acyl ghrelin is thought to be a degradation product of acylated ghrelin through
removal of the acylation and it lacks the endocrine activity such as stimulation of
growth hormone release and food intake. However, findings from many previous
studies show that des-acyl ghrelin has multiple biological activities. Recent studies
demonstrated the beneficial effects of des-acyl ghrelin on cardiovascular system. For
example, des-acyl ghrelin inhibits apoptosis induced by doxorubicin in H9c2
cardiomyocytes and endothelial cells, protects heart against isoproterenol-induced
myocardial injury in a rat model, and improves vascular remodeling in patients with
type 2 diabetes.

Ghrelin mRNA expression was found in HL-1 cardiomyocyte, human atrium and
myocardium. Our experimental study successfully reproduced the cell culture model,
observed the anti-apoptotic effects of des-acyl ghrelin on doxorubicin-induced
cardiomyopathy with the evidence of inhibited TUNEL index, caspase-3 activity, and
increased ratio of Bcl-2 to Bax. The work described in chapter 4 firstly demonstrated

the protective effects of des-acyl ghrelin on cardiac dysfunction and impaired
autophagy in type 2 diabetic mice model. Administration of des-acyl ghrelin for ten
days effectively up-regulated Beclin1 expression in type 2 diabetic heart, which
sufficiently increased cardiac autophagy as suggested by findings from previous
studies. Moreover, we found that des-acyl ghrelin enhanced the conjugation of Atg5-
Atg12 in diabetic db/db mice heart. It promotes the elongation of pre-autophagosomal
membrane, implying the initiation of autophagic process.

Cardiac fibrosis is a common feature involved in the development of doxorubicin-
induced and diabetic cardiomyopathy as shown in our work. Previous studies showed
that acylated ghrelin had therapeutic effect on different diseases including cardiac
injury, systemic sclerosis, bleomycin-induced acute lung injury, liver injury because
of its anti-fibrotic effects. However, there are limited publications related to the role
of des-acyl ghrelin on fibrosis. Li and co-workers found that des-acyl ghrelin exerted
the anti-fibrotic effects on isoproterenol-induced myocardial injury via growth
hormone-independent pathway. Consistently, des-acyl ghrelin significantly reduced
collagen accumulation in the heart of mice treated with doxorubicin and diabetic
heart. This anti-fibrotic effect of des-acyl ghrelin was not diminished by GHSR
inhibitor [D-Lys3]-GHRP-6 in doxorubicin-induced cardiomyopathy, which was in
agreement with the findings of the previous study. The precise mechanisms of des-
acyl ghrelin effect on fibrosis in type 2 diabetic hearts need to be further investigated.

Des-acyl ghrelin exerts its multiple biological functions through the activation of Akt
and ERK1/2 signaling. Des-acyl ghrelin stimulates the proliferation of small intestinal
IEC-6 cells through activation of ERK1/2, inhibits doxorubicin-induced apoptosis in

H9c2 cardiomyocytes, as well as glucose-induced apoptosis in human pancreatic islet
microendothelial cells via activation of Akt and ERK1/2. In our present work, four
days treatment for doxorubicin-induced cardiotoxicity and ten days treatment for
diabetic cardiomyopathy significantly increased the phosphorylation of Akt and
ERK1/2 in heart, which was suppressed by doxorubicin and Type 2 diabetes.
Furthermore, this activation of Akt signaling in doxorubicin-induced cardiomyopathy
was not affected by GHSR antagonist [D-Lys3]-GHRP-6. This suggested that
exogenous des-acyl ghrelin administration activated Akt signaling through a GHSR-
independent pathway without acylated ghrelin conversion.

6.5 Limitations and future direction

Acylated ghrelin plays its central and peripheral biological roles via binding to
GHSR-1a. Although the binding studies of des-acyl ghrelin are explored, the specific
receptor of des-acyl ghrelin is still unknown. Thus, the limitation of the present work
is the investigation of the direct mechanisms responsible for the protective effects of
des-acyl ghrelin on cardiomyopathy.

Our finding showed that des-acyl ghrelin exerts its protective effects on doxorubicin-
induced and diabetic cardiomyopathy via activation of prosurvival Akt and ERK1/2
signaling. Inhibitors of Akt or ERK1/2 should be applied to animals with des-acyl
ghrelin treatment, to investigate if the activation which resulted from des-acyl ghrelin
will be blocked and whether the beneficial effects of des-acyl ghrelin on cardiac
dysfunction, apoptosis, abnormal autophagy, and fibrosis will be affected. In addition,
des-acyl ghrelin treatment enhanced the expression of adiponectin gene in type 2

diabetic heart. It’s worth to explore the level of adiponectin protein expression,
modification and the interactions of adiponectin as well as its receptor response to
des-acyl ghrelin treatment in the future.

In the work presented in chapter 5, S100A8&A9 were identified as crucial genes by
microarray analysis, responsible for the doxorubicin-induced cardiotoxicity in type 2
diabetic heart. The up regulation of S100A8&A9 expression in diabetic heart
following doxorubicin treatment was corresponding to the cardiac dysfunction. We
hypothesize that the blockage of complex of S100A8&A9 in the heart will be a
potential therapeutic target for doxorubicin induced toxicity under the condition of
type 2 diabetic cardiomyopathy. Next we will investigate the precise mechanisms
according to the pathway network of S100A8&A9.

6.6 Clinical applications

Des-acyl ghrelin accounts for 80-90% of circulating ghrelin. Des-acyl ghrelin is
devoid of endocrine activity such as stimulation of growth hormone release and it
lacks the posttranslational modification of acylation. However, given a variety of the
beneficial effects of des-acyl ghrelin on cardiovascular system as well as glucose and
lipid metabolism, its potential clinical applications have been assessed in preclinical
and clinical studies. Here, our current experimental data demonstrated that des-acyl
ghrelin improved cardiac contractility, apoptosis, and fibrosis in mice with
doxorubicin treatment. This suggested that des-acyl ghrelin has the potential for
application in the treatment of doxorubicin-induced cardiomyopathy. Moreover, side
effects of growth hormone on cancer cells can be avoided through des-acyl ghrelin

treatment in cancer patients undergoing chemotherapy.

In contrast to acylated ghrelin, des-acyl ghrelin exerts beneficial effects on
metabolism. Des-acyl ghrelin promotes glucose and free fatty acid uptake by
cardiomyocyte; des-acyl ghrelin administration diminished STZ-induced
hyperglycemia, decreased insulin levels; improved a high-fat diet induced insulin
sensitivity and prevented increased fat mass. Furthermore, des-acyl ghrelin analogs
prevent diabetes in STZ-treated rats and high-fat diet induced metabolic syndrome in
mice. In clinical study, des-acyl ghrelin administration reversed the hyperglycemic
effect of acylated ghrelin in healthy people. Our in vivo data firstly give the evidence
that des-acyl ghrelin administration alleviates the impaired cardiac function in type 2
diabetic cardiomyopathy. Together with its anti-diabetic effect, des-acyl ghrelin is
thought to be a competitive candidate for the treatment of diabetic cardiomyopathy.