Content uploaded by Isabel Guillen
Author content
All content in this area was uploaded by Isabel Guillen on Jan 03, 2014
Content may be subject to copyright.
Corresponding author "
Growth Hormone Releasing Peptide 6 (GHRP6) reduces
liver fibrosis in CCl4 chronically intoxicated rats
" Jorge Berlanga-Acosta1, Dania Vázquez-Blomquist2, Danay Cibrián1, Yssel Mendoza1,
María E Ochagavía3, Jamilet Miranda3, José Suárez4, Yolanda González-Ferrer7,
José M Vila7, Angel Abreu7, Dayana Ugarte-Moreno7, Yolanda Cruz7, Ivon Howland7,
Rosa Coro-Antich8, Olga S León9, Ricardo Bringas3, Diana García-del Barco1,
Karelia Cosme-Díaz4, Daniel Palenzuela2, Julio R Fernández2, Marcelo Nazabal2,
Isabel Guillén2, Alberto Cintado2, Lidia Inés2, Ernesto López-Mola5,
Gerardo E Guillén-Nieto6
1Departments of Tissue Repair and Cytoprotection,
2 Genomics, 3 Bioinformatics and 4 Animal Care;
5 Business Development Group and 6 Direction of Biomedical Research
Center for Genetic Engineering and Biotechnology, CIGB
Ave. 31 / 158 and 190, Playa, PO Box 6162, CP 10 600, Havana, Cuba
7 Center for Medical and Surgical Research
8 Institute of Neurology and Neurosurgery, INN
9 Center for Biological Studies, Institute of Pharmacy and Food, University of Havana, UH
E-mail: jorge.berlanga@cigb.edu.cu
ABSTRACT
Tissue fibrosis is a leading cause of morbidity and mortality. Current treatments for conditions such as hepatic fibrosis
have been unsuccessful. The growth hormone relasing peptide 6 (GHRP6) is endowed with cardioprotective actions
but its antifibrotic effect had not been anticipated. We examined the GHRP6 ability to prevent and revert liver cir-
rhosis after induction in Wistar rats by a subcutaneous administration of CCl4. GHRP6 effects were examined after
concomitant and delayed administration to toxic respectively. The percentages of hepatic fat, fibrosis, nodularity
and septae thickness were histologically and morphometrically determined. Ascitis and portal dilation were judged
by ultrasound and serum biochemical profile and oxidative stress parameters determined. Mechanistic involvement
of selective gene/proteins was assessed by RT-PCR and immunohistochemistry. Microarrays showed gene expression
profiles of GHRP6-treated liver samples on CapitalBio Rat Genome Oligo Array. GHRP6 concomitant intervention
prevented in more than 85% parenchymal fibrotic induration (p < 0.0001) and therapeutic administration for only 15
days allowed for 37% fibrotic clearance (p = 0.0004) with more than 30% reduction of septae thickness (p = 0.0011).
The 60 days GHRP6 administration scheme produced a 75% reduction of the fibrotic area with more than 60% re-
duction of nodularity. GHRP6 reduced oxidative damage enhancing the activity of antioxidant enzymes. Vimentin
and alpha smooth muscle actin immunodetection profile indicated GHRP6 reduced the number of activated stellate
cells. GHRP6 administration reduced fibrogenic factors as TGF-β and CTGF on Kupffer cells. Differentially expressed
genes in the microarray experiment indicated GHRP6 modulate the redox balance and parenchymal cells response
to injury. These evidences suggest GHRP6 may control the liver’s fibroplastic response.
Keywords: Secretagogue, GHRP6, liver, fibrosis, cirrhosis, carbon tetrachloride, cytoprotection
Biotecnología Aplicada 2012;29:60-72
RESUMEN
El péptido liberador de la Hormona de Crecimiento-6 (GHRP6) reduce la fibrosis del hígado en ratas
intoxicadas crónicamente con CCl4. La fibrosis es causa fundamental de morbilidad y mortalidad. Los tratamientos
actuales han fracasado. El péptido liberador de la hormona de crecimiento-6 (GHRP6) ejerce efectos cardioprotec-
tores, pero no se ha descrito su acción antifibrótica. Se examinó la propiedad del GHRP6 para prevenir y revertir
la cirrosis hepática, luego de su inducción en ratas mediante la administración de CCl4. Se evaluó el porcentaje de
grasa hepática, fibrosis, nodularidad y grosor septal mediante estudios histomorfométricos y la ascitis o dilatación
portal por ultrasonido. Se determinó el perfil bioquímico y los parámetros de estrés oxidativo en suero, así como
la participación de genes y proteínas, mediante reacción en cadena de la polimerasa con transcripción inversa e
inmunohistoquímica. Se utilizó un microarreglo de oligonucleótidos del genoma de rata para estudiar el perfil de
expresión de los genes inducidos por el GHRP6. La intervención concomitante con GHRP6 previno la induración
fibrótica en más del 85% (p < 0.0001). La administración terapéutica durante 15 días permitió la remoción fibrótica
del 37% (p = 0.0004), la reducción del grosor septal superó el 30% (p = 0.0011). La administración durante 60
días redujo las áreas fibróticas en 75%, y la reducción de la nodularidad fue de más del 60%. El GHRP6 redujo el
daño oxidativo porque aumentó la actividad de las enzimas antioxidantes y las células estrelladas activadas positivas
a vimentina y actina alfa de músculo liso. También redujo la expresión de factores fibrogénicos sobre las células
Kupffer. El perfil de expresión de genes indicó que el GHRP6 modula el balance redox y la respuesta al daño tisular
en las células parenquimales. Estas evidencias sugieren que el GHRP6 pudiera controlar la respuesta fibroplásica
del hígado.
Palabras clave: Secretagogo, GHRP6, hígado, fibrosis, cirrosis, tetracloruro de carbono, citoprotección
RESEARCH
Jorge Berlanga-Acosta et al. GHRP6 reduces liver fibrosis
61 Biotecnología Aplicada 2012; Vol.29, No.2
Introduction
Liver fi brosis is the fi nal common pathway of many
human hepatic diseases and represents a major source
of morbidity and mortality worldwide [1]. In contrast
to the traditional view that liver fi brosis is an irrevers-
ible disease, recent evidences obtained from animal
models and patients indicate that advanced cirrhosis
may be ameliorated [2-5].
Hepatic fi brosis is the wound-healing response of
the liver to chronic injury [6]. After an acute liver
injury, parenchymal cells regenerate and replace the
necrotic and/or apoptotic hepatocytes. This process is
associated with a controlled infl ammatory response
and a limited deposition of extracellular matrix
(ECM) proteins. If the injury persists, the liver regen-
eration process fails and the parenchymal hepatocytes
are replaced by abundant ECM proteins that disrupt
the hepatic architecture by forming cirrhotic nodules.
This induces hepatocellular dysfunction and increases
intrahepatic resistance to blood fl ow, which results in
liver insuffi ciency and portal hypertension, respec-
tively [7].
The growth hormone releasing peptide 6 (GHRP6)
is a six-amino acids synthetic peptide that belongs
to the growth hormone secretagogues (GHS) family.
Besides its fi rst described GH-releasing activity [8],
mounting evidences substantiate that GHRP6 and
their analogs exhibit ever expanding pharmacological
effects including cytoprotection [9-13]. We had previ-
ously demonstrated that a single GHRP6-prophylactic
administration prevented hepatocytes demise in a
stringent setting of hepatic ischemia [11]. Whether this
GHRP6-induced hepatoprotective effect had impact
in liver fi brosis remained unexplored. Nevertheless,
serendipitous observations inspired this study, when
we observed that rats affected by doxorubicin-induced
dilated myocardiopathy and treated with GHRP6 ex-
hibited far less fi brosis in their major parenchymal or-
gans than their saline-treated counterparts.
This work demonstrates for the fi rst time that
GHRP6 intervention substantially attenuates the onset
of a fi brotic process as well as triggers the regression
of cirrhosis in CCl4 chronically intoxicated rats. The
intervention appeared to amplify hepatic cells detoxi-
fi cation mechanisms which may ultimately attenuate
hepatic stellate cells (HSC) activation and the onset of
a fi brogenic program.
Materials and methods
Reagents
The GHRP6 (His-D-Trp-Ala-Trp-D-Phe-Lys-NH2)
was purchased from BCN-Peptides (Barcelona,
Spain). The product was certifi ed as a sterile, pyro-
gen-free white powder with 95% purity. For animal
administration, fresh solutions were always prepared
by diluting the peptide in sterile normal saline so-
lution. CCl4 and mineral oil were purchased from
Merck (Darmstadt, Germany). The antibodies anti-
Transforming growth factor beta (TGF-β), anti-p53,
anti-Cyclin D1 and anti-FasL were purchased from
Santa Cruz Biotechnology Inc. (USA). The anti- al-
pha smooth muscle actin (α-SMA) and anti-Vimentin
monoclonal antibodies were purchased from DakoCy-
tomation (Denmark).
Animals
A total of 75 male Wistar rats (250-270 g, 9-10 weeks)
were purchased from the National Center for the Pro-
duction of Laboratory Animals (Havana, Cuba). The
rats were maintained in a certifi ed room at the Ani-
mal Facility of the Center for Genetic Engineering
and Biotechnology (Havana, Cuba), under controlled
environmental conditions and with unrestricted ani-
mals’ access to food and water. The study protocol
was approved by the institutional board on laboratory
animals’ welfare.
Fibrosis induction protocol
Liver fi brosis was induced by the subcutaneous in-
jection of CCl4 twice a week (Monday and Friday),
for fi ve or seven months. The CCl4 selected dose was
1 mL/kg, diluted as 1:1 proportion with mineral oil
before injection [14].
Phases and experimental groups
Seven intact animals received mineral oil for seven
months and were terminated once the study was com-
pleted (Intact control group). The 68 remaining rats
were subjected to the fi brosis induction protocol as
mentioned above.
As this study aimed to examine the potential effect
of the GHRP6 intervention toward both fi brosis pre-
vention and regression, two experimental blocks were
established. The fi rst one, developed during the fi ve
initial months, included the concomitant administra-
tion of GHRP6 with CCl4 to assess hepatic fi brosis
prevention. The second was conducted for the sixth
and seventh months while GHRP6 was therapeuti-
cally administered to assess its potential in promoting
cirrhosis regression.
Experimental groups for the prevention trial
The groups named CCl4 + GHRP6 and CCl4 + Saline
received CCl4 as previously described while receiving
concomitantly two daily intraperitoneal (i.p.) doses of
GHRP6 (400 μg/kg) or normal saline injections, re-
spectively, for fi ve months. Both groups were of 12
rats each and all the animals were autopsied upon
concluding the fi fth month of the study.
Experimental groups for the regression trial
The 44 remaining rats also received CCl4 as previ-
ously described during the fi ve initial months of the
study. Afterwards, all these animals were individually
subjected to a diagnostic laparotomy (see below). This
allowed for allocating the animals into four balanced
groups according to their hepatic disease severity.
Two groups of 10 rats each (denominated GHRP6-
15d and Saline-15d) received two daily i.p. injections
of GHRP6 (dose 400 μg/kg) or normal saline, respec-
tively, for 15 days, starting at the beginning of the
sixth month. The CCl4 administration was interrupted
during the application of treatments. All these animals
were autopsied when the short time intervention was
completed.
The last two groups comprised 12 rats each, de-
nominated GHRP6-60d and Saline-60d, respectively.
These animals were treated either with GHRP6 or
normal saline as described for groups GHRP6-15d
and Saline-15d, but for 60 days (sixth and seventh
Lefton HB, Rosa A, Cohen M. Diagnosis 1.
and epidemiology of cirrhosis. Med Clin
North Am. 2009;93(4):787-99.
Hammel P, Couvelard A, O’Toole D, 2.
Ratouis A, Sauvanet A, Flejou JF, et al.
Regression of liver fibrosis after biliary dra-
inage in patients with chronic pancreatitis
and stenosis of the common bile duct. N
Engl J Med. 2001;344(6):418-23.
Arima M, Terao H, Kashima K, Arita T, 3.
Nasu M, Nishizono A. Regression of liver
fibrosis in cases of chronic liver disease
type C: quantitative evaluation by using
computed image analysis. Intern Med.
2004;43(10):902-10.
Issa R, Zhou X, Constandinou CM, 4.
Fallowfield J, Millward-Sadler H, Gaca
MD, et al. Spontaneous recovery from
micronodular cirrhosis: evidence for in-
complete resolution associated with matrix
cross-linking. Gastroenterology. 2004;
126(7):1795-808.
Bourliere M, Kahloun A, Gascou-5.
Tessonnier G. Analogs and fibrosis regres-
sion in hepatitis B. Gastroenterol Clin Biol.
2009;33(10-11):923-9.
Friedman SL. Liver fibrosis -- from 6.
bench to bedside. J Hepatol. 2003;38
Suppl 1:S38-53.
Gines P, Cardenas A, Arroyo V, Rodes 7.
J. Management of cirrhosis and ascites. N
Engl J Med. 2004;350(16):1646-54.
Bowers CY, Momany FA, Reynolds GA, 8.
Hong A. On the in vitro and in vivo activity
of a new synthetic hexapeptide that acts
on the pituitary to specifically release
growth hormone. Endocrinology. 1984;
114(5):1537-45.
Shen YT, Lynch JJ, Hargreaves RJ, 9.
Gould RJ. A growth hormone secretago-
gue prevents ischemic-induced mortality
independently of the growth hormone
pathway in dogs with chronic dilated car-
diomyopathy. J Pharmacol Exp Ther. 2003;
306(2):815-20.
Paneda C, Arroba AI, Frago LM, Holm 10.
AM, Romer J, Argente J, et al. Growth
hormone-releasing peptide-6 inhibits
cerebellar cell death in aged rats. Neuro-
report. 2003;14(12):1633-5.
Cibrian D, Ajamieh H, Berlanga J, Leon 11.
OS, Alba JS, Kim MJ, et al. Use of growth-
hormone-releasing peptide-6 (GHRP-6) for
the prevention of multiple organ failure.
Clin Sci (Lond). 2006;110(5):563-73.
Berlanga J, Cibrian D, Guevara L, 12.
Dominguez H, Alba JS, Seralena A, et
al. Growth-hormone-releasing peptide 6
(GHRP6) prevents oxidant cytotoxicity and
reduces myocardial necrosis in a model of
acute myocardial infarction. Clin Sci (Lond).
2007;112(4):241-50.
Granado M, Martin AI, Lopez-Men-13.
duina M, Lopez-Calderon A, Villanua MA.
GH-releasing peptide-2 administration
prevents liver inflammatory response in
endotoxemia. Am J Physiol Endocrinol
Metab. 2008;294(1):E131-41.
Abe W, Ikejima K, Lang T, Okumura 14.
K, Enomoto N, Kitamura T, et al. Low mo-
lecular weight heparin prevents hepatic fi-
brogenesis caused by carbon tetrachloride
in the rat. J Hepatol. 2007;46(2):286-94.
Jorge Berlanga-Acosta et al. GHRP6 reduces liver fibrosis
62 Biotecnología Aplicada 2012; Vol.29, No.2
experimental months).The CCl4 administration con-
tinued in both groups until the end of the seventh
month, when all these rats were fi nally autopsied.
Hepatic ultrasound
All the animals of the CCl4 + GHRP6 and CCl4 +
Saline groups were subjected to a comparative ultra-
sonic study after fi ve months of concomitant CCl4 and
treatments administrations. Animals from GHRP6-60d
and Saline-60d groups were comparatively evaluated
by ultrasound at the end of the seventh month. The rats
from the Intact control group were concurrently evalu-
ated in each case. Ultrasounds were performed with an
Aloka apparatus (Japan) connected to an 11 MHz trans-
ducer in previously anesthetized (Ketamine, 50 mg/
kg) animals. The studied parameters included: portal
diameter (mm), ascites, parenchymal nodularity, echo-
genicity increase and liver texture evaluation. Ascites
was scored according to the following criteria: 0- no
ascites, 1- small quantity of ascites that is only detect-
able by ultrasound, 2- clinically evident ascites. For
parenchymal nodularity a scale from 0 to 3 was used
to score the result: 0- no nodules, 1- one or two nod-
ules, 2- a faint multinodularity and 3- multinodular
images with large size nodules included. Echogenicity
increase was qualitatively graded according to the fol-
lowing scale: 0- no increase, 1- moderate increase and
2- signifi cant increase of parenchymal echogenicity.
Liver texture was graded as follows: 0- homogeneous
texture; 1- heterogeneous, faintly granulated; and
2- heterogeneous, grossly granulated. The ultrasonic
fi brosis index (UFI) was defi ned as the total sum of
the values obtained in these four gradation scales. To
avoid a biased judgment all the images from hepatic
ultrasound were blindly scored.
Diagnostic study by laparotomy
The 44 rats included in the fi brosis regression phase
were subjected to a diagnostic study by laparotomy
once concluding the fi ve initial months of CCl4 ad-
ministration. Briefl y, a small abdominal incision was
performed to previously anesthetized animals (Ket-
amine, 50 mg/kg), and by gentle manipulation the
whole liver mass was fully exposed for macroscopic
inspection, classifi cation and scoring according to
the World Health Organization score: macronodular,
micronodular, mixed and fat organ gross appearance
[15]. Finally, the liver was appropriately returned to
its cavity and the small wound was sutured. Balanced
groups were made up as judged by the hepatic mac-
roscopic aspect.
Liver histology
For histopathology, three liver fragments from all the
animals were harvested during the autopsy, each one
from a different hepatic lobe, and were 10% buffered
formalin fi xed and paraffi n embedded. Slides with
semi-thin sections (2-3 μm) were prepared and stained
with hematoxylin/eosin and Mallory trichrome. All the
histological evaluations and morphometric protocols
were conducted in a blinded manner. Mallory stained
slides were used to assess the cirrhotic nodularity for
each animal. For this purpose, the total number of cir-
rhotic nodules was counted in three equally randomly
selected microscopic fi elds (5 ×) per liver fragment (a
total examined area of 10 mm2 by hepatic lobe). The
fi nal result was presented as the average of the nod-
ules/mm2 among the three hepatic fragments.
Histomorphometric analysis was conducted using
the ImageJ program (NIH, USA). Digital images were
captured from the Mallory stained slides in RGB for-
mat, with 24 bit true colors and at 3072 × 2304 pixels
resolution, through a Carl Zeiss Axiotron microscope
(Germany) coupled to a Canon PC1089 camera (Can-
on, Japan). Fat and fi brosis percentages were assessed
for all the animals using images from three equally
randomly selected microscopic fi elds (10 ×) per liver
fragment. The fi nal values from the nine captured im-
ages were averaged to obtain the representative per-
centages of fat and fi brosis hepatic covered area for
each animal.
In order to determinate the averaged fi brotic sep-
tum thickness of each animal, a total of 100 randomly
selected microscopic fi elds (40 ×) was captured by an-
imal, and processed with the calibrated ImageJ soft-
ware. The thickness was always measured amidst the
whole septum length and was reported in microns.
Serum biochemical determinations
Blood samples were collected from the retro-orbital
plexus of previously anesthetized rats from the intact
control group, on the fi rst experimental day and from
all the animals of the CCl4 + GHRP6 and CCl4 + Sa-
line groups after two months of experiment initiation.
The fi nal blood samples were harvested by myocar-
dial punction during the autopsies. Serum samples
were aliquoted and kept at –20 °C until processing.
alanine aminotransferase (ALAT) and aspartate ami-
notransferase (ASAT) serum activities; as the serum
concentration of total proteins (TP), albumin, very
low density lipoprotein, cholesterol and triglycerides
were assessed in an automatic analyzer Hitachi 747
(Boehringer Mannheim, Germany). Commercial kits
and analytical procedures were conducted according
to the manufacturer’s instructions.
Hepatic oxidative stress assessment
Liver central lobe fragments collected during the
autopsies from all the rats were used to assess the
hepatic oxidative stress. The protocols for tissue
homogenates and superoxide dismutase (SOD) and
catalase enzymes activities were followed as previ-
ously described [11]. The lipid peroxidation potential
(LPP) and malondialdehyde (MDA) were measured
using the Bioxytech LPO-586 commercial kit; while
the total hydroperoxide content was assayed by the
Bioxytech H2O2-560 commercial kit, both according
to manufacturer’s instructions (Bio-Rad Laboratories,
Germany). The advanced oxidation protein products
(AOPP) content was assessed according to the Witko-
Sarsat described technique [16]. All the hepatic bio-
chemical data were adjusted to the total protein con-
centration determined in the tissue homogenates using
a commercial kit (Bio-Rad Laboratories, Germany).
Immunohistochemistry
For immunohistochemistry studies, liver sections
(2-3 μm) were mounted on sialinized slides (DAKO,
Denmark), heat-treated for antigen exposure, and pro-
cessed according to the manufacturer’s instructions
Anthony PP, Ishak KG, Nayak NC, 15.
Poulsen HE, Scheuer PJ, Sobin LH. The
morphology of cirrhosis: definition, no-
menclature, and classification. Bull World
Health Organ. 1977;55(4):521-40.
Witko-Sarsat V, Gausson V, Nguyen 16.
AT, Touam M, Drueke T, Santangelo F, et al.
AOPP-induced activation of human neutro-
phil and monocyte oxidative metabolism:
a potential target for N-acetylcysteine
treatment in dialysis patients. Kidney Int.
2003;64(1):82-91.
Jorge Berlanga-Acosta et al. GHRP6 reduces liver fibrosis
63 Biotecnología Aplicada 2012; Vol.29, No.2
from DakoCytomation LSABTM+ System-HRP com-
mercial kit. Tissue samples were incubated for 30 min
with: anti-αSMA (1:100), anti-TGF-β (1:250), anti-
p53 (1:200), anti-Cyclin D1 (1:100), anti-FasL (1:200)
and anti-Vimentin (1:100). Antibodies were diluted
in Dako background reducing solution. Immunohis-
tochemistry was accomplished on material retrieved
from three representative animals per group purposely
selected according to the histopathology judgment.
Healthy animals were also included. Slides were
counterstained with hematoxilin or light green and
were blindly analyzed by two different investigators.
The number of Cyclin D1 positively labeled hepa-
tocytes nuclei and FasL positively labeled Kupffer
cells within the hepatic parenchyma were quantifi ed
in 15 microscopic fi elds (20 ×), evenly distributed in
the three liver fragments collected from GHRP6-15d
and Saline-15d groups only. Internal controls included
liver sections from healthy intact rats and omission or
replacement of the commercial primary antibody by
pre-immune isospecies serum. Data are presented as
the averaged value of the 45 microscopic fi elds studied
by group (three rats in each one).
Gene expression analyses by semiquantitative
RT-PCR
Liver central lobe fragments collected from fi ve ran-
domly selected rats of the Saline-15d, GHRP6-15d
and Intact control group were processed to isolate to-
tal RNA using TRI Reagent (Sigma, St. Louis, USA).
Total RNA was digested with RNase-free DNase
I (Epicentre Technologies, USA) according to the
manufacturer’s instructions for DNA contaminant
removal. Afterward, one microgram of total RNA
was reverse transcribed using a commercial available
kit (GeneAmp® RNA PCR Core Kit. Applied Bio-
systems, USA) with an oligo-dT primer. PCR were
performed using specifi c primers and annealing tem-
peratures referred in table 1. Final PCR products were
detected in a 1% (w/v) agarose gel and were quanti-
fi ed using the Kodak ID 3.6 software package (Kodak
Inc, USA). β-actin was used as housekeeping gene for
normalization.
RNA extraction for microarray experiment
Rats in Groups GHRP6-60d and Saline-60d were
checked for their percentage of fi brosis at the end of the
seventh month and fi brosis reduction was then calcu-
lated. Total RNA isolation was carried out as for semi-
quantitative RT-PCR experiments and further purifi ed
using a NucleoSpin RNA clean-up kit (Macherey-
Nagel, Germany). The quality of the total RNA (i.e.,
the purity and integrity of the intact RNA) was as-
sessed by Nanodrop 1000 (ThermoScientifi c, USA)
and Bioanalizer Agilent 2100 (Agilent, USA), report-
ing the concentration, absorbance 260/280 nm ratio of
1.8 or higher, and RNA integrity number equal to or
higher than 7, respectively. Five paired samples from
GHRP6-60d and Saline-60d that met RNA quality re-
quirements and exhibited a fi brosis reduction superior
to 69% were used for the experiment.
Microarray experiment
We chose a reference design with fi ve samples per
groups GHRP6-60d and Saline-60d compared to a
reference, representing the seven pooled samples
from the Intact control group.
The amplifi cation and labeling of mRNA were per-
formed using the CapitalBio cRNA Amplifi cation and
Labeling kit according to manufacturing instructions
(CapitalBio, Beijing, China). The rat 27 K oligonu-
cleotide microarray comprises 26 962 oligo probes
of 70-mer (Capitol-Bio Corporation, Beijing, China)
from the Operon Company (Rat Genome Oligo Set,
Version 3.0.5). Dual channel microarray hybridiza-
tion was performed with 70-80 total pmol of Cy3-
labeled control sample and Cy5-labeled test samples
(from GPHR6-60d and Saline-60d) onto 25 × 75 mm
chips. Hybridization and washing of slides were car-
ried out according to the manufacturer instructions
(Capitol-Bio Corporation, Beijing, China).The slides
were scanned with a confocal LuxScan scanner (Cap-
italBio Corp, China) and the raw data were extracted
using LuxScan™ 3.0 software (CapitalBio Corp).
For dual-channels microarray data, the scanning set-
ting for Cy3 and Cy5 channels were balanced. The
signals detected from housekeeping and Hex genes,
and those detected from the exogenous controls were
used as positive controls. Negative controls were at
background levels.
Statistical analysis
All the experimental data were initially evaluated for
a normal distribution using the Kolmogorov-Smirnov
test (p < 0.05). When a normal distribution was estab-
lished, an unpaired Student’s test was used for compar-
isons between the groups of prevention phase; while
paired Student’s t test was used for the groups of the
regression phase. In case of multiple comparisons, the
one way ANOVA followed by the Student-Newman-
Table 1. Primers sequences and PCR conditions used for gene expression analyses
Primer sequences (5´-3´)
sense CCA TGT ACG TAG CCA TCC AGG
sense TGC CAG AAC CCC CAT TGC TG
sense AGA GCT GGG TGT GTG TCC TCC
sense GGA TGG AGT GGT AGA GCC TTT
sense AAA GGG GAT AAC AGC CAC TAC AAG G
antisense GAC AGT GAG GCC AGG ATA GAG C
antisense TCC ACC TTG GGC TTG CGA CC
antisense GCA GCA AAC ACT TCC TCG TGG
antisense TCT ACA CCC AAA TGC TGC ACA GG
antisense GAG GGA TTA ACA AAC ATG GTG GAG C
Product
length (bp)
660
700
547
557
470
No. of
Cycles
30
32
30
30
32
Tm
(ºC)
72
70
70
70
55
Gene
name
β-actin
TGF-β
CTGF
SODMn
MMP13
GenBank
accession #
BC063166
NM021578
NM022266
NM017051
M60616
Jorge Berlanga-Acosta et al. GHRP6 reduces liver fibrosis
64 Biotecnología Aplicada 2012; Vol.29, No.2
Keuls test was used. The percentage values were com-
pared using the Fisher’s exact test. A value of p < 0.05
was used to indicate a signifi cant difference.
The R Limma package (http://bioinf.wehi.edu.au/
limma) was used for preprocessing and differential
expression analysis of microarray data [17]. The me-
dian average intensity of foreground and background
were extracted from the lsr fi les. A quality criterion
[18] was applied to identify low intensity or high
background spots assigning weights 0 or 1, which
were later used as inputs for limma approach. The
normexp+offset method was selected for adaptively
adjusting the foreground for the background intensi-
ties. The within-arrays normalization was performed
with the print-tip loess method using between chan-
nels non-differentially expressed controls and for be-
tween-array normalization the scale method using all
control probes. The moderated paired t-test for each
gene was calculated. Genes with p < 0.001 [18] and
fold change greater than 1.5 were considered for bio-
informatics analysis and biological interpretation.
Bioinformatics analysis of differentially
expressed gene
Bioinformatics analysis was performed using, as in-
put, the list of differentially expressed genes in rat
(GHRP6-60d vs. Saline-60d) and also the list of their
human orthologous in order to take benefi t of the
functional annotation available on human genes
aiming to predict signifi cant biological process in-
volved in the putative GHRP6-mediated antifi brotic
mechanisms. Human orthologous for rat genes were
identifi ed by searching the Homologene database.
Whenever orthologous genes were not found in
Homologene, sequence similarity searches of rat’s
transcripts against the human RefSeq transcripts were
performed with Blastn. The most similar gene, for ev-
ery rat gene, was chosen as the best Blastn hit (gene
with lowest E-value, E-value < 1 × 10-7). To identify
putative protein-protein interactions, in which the
protein products of the input genes are involved, gene
networks were constructed using the Cytoscape’s [19]
plugin BisoGenet [20]. All molecular interaction data
sources available at BisoGenet were used to gener-
ate the gene networks. These networks were enriched
with transcription regulation data extracted from the
literature. The Cytoscape’s plugin BiNGO [21] was
used to identify the Gene Ontology (GO) [22] biologi-
cal processes enriched in the lists of input genes. The
statistically signifi cant GO processes were determined
by using the hypergeometric test and the Benjamini-
Hochberg False Discovery Rate (FDR) correction for
multiple testing [23] with a threshold of 0.05.
Results
Hepatic ultrasound exploration
At the fi fth month, the comparative ultrasonic study
between CCl4 + GHRP6 and CCl4 + Saline groups
demonstrated a signifi cant difference (p = 0.0009)
between the calculated UFI values, thus suggesting
that GHRP6 prevented fi brosis (Table 2). The animals
of the GHRP6-60d and Saline-60d groups were also
subjected to a comparative study by ultrasound at
the seventh month. At this time point, the Saline-60d
achieved the largest UFI value registered (Table 2),
which was signifi cantly different (p = 0.0005) to that
of GHRP6-60d group; thus indicating that a process
of fi brosis regression had been set forth.
The portal vein appeared dilated over 60% in the
animals from the CCl4 + Saline group at the fi fth
month, as compared to the Intact control group (p =
0.0052; table 2). In contrast, the averaged portal diam-
eter for the CCl4 + GHRP6 group did not differ from
the normal values (p = 0.2938). At the seventh month,
once the regression phase concluded, the GHRP6-60d
group showed a signifi cant reduction of portal dilation
as compared to the Saline-60d group (p = 0.0201).
As shown in table 2, a signifi cant reduction of the
animals bearing ascites was detected in the concomi-
tant (p = 0.0361) and the therapeutic GHRP6 for 60
days (p = 0.0373). No ascites was detected for the
GHRP6-15d and Saline-15d groups during the short
time of therapeutic intervention.
Liver histopathology
After fi ve months of continuous CCl4 administration,
the animals from the CCl4+Saline group showed dense
collagen bundles surrounding cirrhotic nodules (Fig-
ure 1A) which represented about 17% of hepatic area
covered by fi brosis (Table 3). However, the concomi-
tant intervention with GHRP6 prevented in more than
85% the fi brotic induration (p < 0.0001; fi gure 1B).
In line with this, the CCl4 + GHRP6 group showed
far less cirrhotic nodules and averaged lesser septum
thickness as compared to the CCl4+Saline group (both
p < 0.0001). These fi ndings sustain the ultrasound
evidences of GHRP6-mediated anti-fi brotic response.
Moreover, the percent of fi brosis, the number of cir-
rhotic nodules, and the averaged septum thickness of
the saline-treated animals within the reversion trial for
15 days were very close to those of the CCl4 + Saline
group (all p > 0.05); which indicated that no relevant
spontaneous fi brosis resolution took place during the
15 days in which the CCl4 was not injected (Figures
1C and D; table 3).
The therapeutic administration of GHRP6 for 15
days in the fi rst reversion protocol, allowed for 37%
of fi brosis clearance (p = 0.0004). It was mainly due
to reduction of more than 30% of septae thickness
(p = 0.0011). No differences were found in the number
of cirrhotic nodules (p = 0.0602) between the groups
(Table 3).
After seven months of CCl4 continuous administra-
tion, the CCl4 + Saline-60d group reached the largest
Table 2. Results from hepatic ultrasound studies†
Portal diameter
(mm)
Clinical ascites
(%)
Ultrasound
ascites (%)
0.71 ± 0.05 0 0
0.80 ± 0.06 1(8) 2(17)
1.12 ± 0.14* 5(42) 8(67)
0.96 ± 0.08 0 3(25)
1.34 ± 0.13* 5(42) 8(67)
N
7
12
12
12
12
Experimental
groups
Intact control
CCl4 + GHRP6
CCl4 + Saline
GHRP6-60d
Saline-60d
UFI
0
2.50 ± 0.30
5.93 ± 0.79***
4.80 ± 0.42
7.00 ± 0.29***
† Ultrasounds to CCl4 + GHRP6 and CCl4 + Saline groups were conducted at the fifth experimental month while
to GHRP6-60d and Saline-60d groups were at the seventh month. Portal diameter value reported for Intact
control group was obtained at seventh month. UFI: Ultrasonic fibrosis index. Data from UFI and portal diameter
are presented as average ± SEM by group. Ascites results are indicated as the total number of animals and
as percentage, by group. Ultrasound ascites data also include the clinical ascites. (*/***) indicate significant
differences between the GHRP6-treated animals and their counterpart Saline groups for at least p < 0.05.
Smyth GK. Linear models and empiri-17.
cal bayes methods for assessing differential
expression in microarray experiments. Stat
Appl Genet Mol Biol. 2004;3:Article 3.
Simon R, Korn E, McShane L, Rad-18.
macher M, Wright G, Zhao Y. Design and
Analysis of DNA Microarray Investigations.
New York: Springer-Verlag; 2004.
Shannon P, Markiel A, Ozier O, Baliga 19.
NS, Wang JT, Ramage D, et al. Cytoscape:
a software environment for integrated mo-
dels of biomolecular interaction networks.
Genome Res. 2003;13(11):2498-504.
Martin A, Ochagavia ME, Rabasa LC, 20.
Miranda J, Fernandez-de-Cossio J, Bringas
R. BisoGenet: a new tool for gene network
building, visualization and analysis. BMC
Bioinformatics. 2010;11:91.
Maere S, Heymans K, Kuiper M. BiN-21.
GO: a Cytoscape plugin to assess overre-
presentation of gene ontology categories
in biological networks. Bioinformatics.
2005;21(16):3448-9.
Ashburner M, Ball CA, Blake JA, 22.
Botstein D, Butler H, Cherry JM, et al.
Gene ontology: tool for the unification of
biology. The Gene Ontology Consortium.
Nat Genet. 2000;25(1):25-9.
Benjamini Y, Hochberg Y. Controlling 23.
the false discovery rate: a practical and
powerful approach to multiple testing. J R
Stat Soc Ser B. 1995;57(1):289-300.
Jorge Berlanga-Acosta et al. GHRP6 reduces liver fibrosis
65 Biotecnología Aplicada 2012; Vol.29, No.2
percent of fi brosis which was signifi cantly differ-
ent from the CCl4 + Saline group (p = 0.0139) at the
fi fth experimental month, indicating a progression in
the disease severity due to a substantial increase in
septal thickness. As shown in table 3, the therapeu-
tic administration of GHRP6 by 60 days achieved a
75% reduction of the fi brotic area when compared
to the Saline-60d group (p < 0.0001), even though
the CCl4 injections were not interrupted (Figure 1F).
Table 4 shows the reduction in fi brosis from each of
the twelve pairs of liver from GHRP6-60d and Saline-
60d groups. A reduction higher than 69% is observed
in pairs 3, 4, 7, 8, 10, 11, 12 and higher than 80% in
pairs 3, 7 and 8. Accordingly, the GHRP6-60d group
showed signifi cant reductions in the cirrhotic nodules/
mm2 and the septae thickness as compared to the Sa-
line-60d group (both p < 0.0001).
The CCl4 + GHRP6 group exhibited a two-fold
increase of hepatic fat percentage with respect to the
CCl4 + Saline group (p = 0.0002; table 3). At the sev-
enth month, the GHRP6-60d group showed a 38%
increase of fat deposition compared to the Saline-60d
group (p = 0.0067). In contrast, in the rats from the
GHRP6-15d group, where the CCl4 injections were
interrupted, a notorious reduction of fat was detected
as compared to the Saline-15d group (p < 0.0001).
Serum biochemistry
In the fi brosis prevention trial the serum biochemical
analysis was done after the second and the fi fth month
of continuous CCl4 and GHRP6/Saline administra-
tions. At the second month, the CCl4 + Saline group
showed the highest ASAT and ALAT levels detected as
compared to the Intact control group (both p < 0.001;
fi gures 2A and B). These values dropped by the fi fth
experimental month although remained signifi cantly
superior to the healthy animals (both p < 0.001). A
similar biphasic behavior for both transaminases was
observed for the CCl4 + GHRP6 group; which resulted
signifi cantly lower than those detected for the CCl4 +
Saline group at the second month (both p < 0.001). At
the fi fth month, the ASAT level of the CCl4 + GHRP6
group was also signifi cantly lower than the CCl4 +
Saline group (p < 0.001) while similar values for
ALAT were observed for both groups. The CCl4-in-
duced liver damage was associated to a reduction of
the organ’s biosynthetic function in both groups (Fig-
ures 2C to F). However, at the two evaluation time
points, the CCl4 + GHRP6 group exhibited a signifi -
cantly better liver function as compared to the CCl4 +
Saline group for all the evaluated parameters (at least
p < 0.05).
In the fi brosis regression trial, serum chemical pa-
rameters in the GHRP6-15d and Saline-15d groups
were evaluated once the short therapeutic scheme
was completed. As showed in table 5, signifi cant dif-
ferences for ALAT (p < 0.01) and ASAT (p < 0.05)
values were detected between these groups refl ecting
the GHRP6-induced hepatoprotective effect. Despite
this, no signifi cant differences were detected between
the GHRP6-15d and the Saline-15d groups in any of
the evaluated hepatic synthesis indicators, which re-
mained lower than those of the Intact control group
(at least p < 0.05). The serum biochemical parameters
from the GHRP6-60d and the Saline-60d groups were
A
C
E
B
D
F
Figure 1. Histopathological images from liver rats’ biopsy. Images correspond to Mallory staining slides
(10 ×) and are representative from the experimental groups: A) CCl4 + Saline; B) CCl4 + GHRP6; C)
Saline-15d; D) GHRP6-15d; E) Saline-60d; F) GHRP6-60d. A more intense degree of fibrosis (blue stai-
ned) is observed in all the Saline-groups in comparison with their counterpart GHRP6-treated rats.
Table 3. Histomorphometric results†
Nodules/mm2Fibrosis (%) Fat (%)
0.82 ± 0.23*** 1.99 ± 0.31*** 32.97 ± 2.23***
5.58 ± 0.52 16.80 ± 0.63 15.14 ± 2.85
3.80 ± 0.41 9.80 ± 1.15*** 7.01 ± 0.60***
5.00 ± 0.44 15.63 ± 0.67 13.21 ± 0.78
2.24 ± 0.42***
5.90 ± 0.43
4.93 ± 0.65***
19.37 ± 0.68
24.71 ± 1.94**
17.92 ± 1.23
N
12
12
10
10
12
12
Experimental
groups
CCl4 + GHRP6
CCl4 + Saline
GHRP6-15d
Saline-15d
GHRP6-60d
Saline-60d
Septae
thickness
10.97 ± 1.66***
72.13 ± 7.85
47.31 ± 4.62**
71.63 ± 4.67
34.62 ± 4.36***
102.73 ± 8.14
† The histomorphometric analyses were conducted using Mallory staining slides. Data are presented as average
± SEM by group. The septum thickness was determined in microns. (**/***) indicate significant differences
between GHRP6-treated animals and their respectively Saline groups for at least p < 0.01.
Table 4. Percentage of fibrosis reduction in twelve
pairs of livers*
Pair
1
2
3
7
4
8
5
9
11
6
10
12
Fibrosis reduction (%)
58.54
55.82
85.65
82.37
70.20
81.75
79.96
45.13
68.08
69.32
65.08
69.37
* It is shown the percentage of fibrosis reduction in paired animals
(1 to 12) after the treatment with GHRP6. Pairs 3, 4, 7, 8 and 10 to
12 showed the highest reductions in fibrosis (> 69%).
Jorge Berlanga-Acosta et al. GHRP6 reduces liver fibrosis
66 Biotecnología Aplicada 2012; Vol.29, No.2
also assessed following autopsy (Table 6). No signifi -
cant differences were detected between these groups
for the serum transaminases as for any of the organ’s
functional parameters (at least p < 0.05).
Hepatic oxidative stress
An etiopathogenic ingredient of the CCl4 hepatotoxic
mechanism is the oxidative damage to liver cells [24].
It was confi rmed by the detected increase of the eval-
uated oxidative stress markers (total hydroperoxide
content, MDA, LPP and AOPP) in all the experimen-
tal groups, as compared to the Intact control group
(at least p < 0.001; table 7). The GHRP6 intervention
signifi cantly attenuated all these oxidative markers as
compared with each respective saline group (at least
p < 0.01), in the three interventional approaches as-
sessed. Concurrently, the GHRP6-treated rats exhib-
ited a remarkable increase of Catalase and SOD ac-
tivities as compared to their counterpart saline groups
(at least p < 0.05).
Immunohistochemical results
The α-SMA protein is a molecular marker broadly
used to detect activated HSC [25, 26]. Although based
on qualitative judgment, animals intervened with
GHRP6 either under concomitant (data not shown)
or therapeutic approaches exhibited far less α-SMA
labeling than their saline counterparts (Figures 3A
c
Figure 2. Serum biochemical assessment of fibrosis prevention phase. Data are presented as the average ± SEM by group:
Intact control group, CCl4 + Saline and CCl4 + GHRP6. Statistical analyses were conducted at the second and fifth months,
respectively. Different letters indicate significant differences for at least p < 0.05. A) Aspartate aminotransferase (ASAT); B)
Alanine aminotransferase (ALAT); C) Cholesterol; D) Triglycerides; E) Very low density lipoprotein (VLDL); F) Total proteins.
Table 5. Serum biochemical assessments of the 15 days fibrosis regression scheme*
GHRP6-15d group
(n = 10)
Saline-15d group
(n = 10)
Intact control
group (n = 7)
264.42 ± 41.15 b 62.78 ± 6.44 c
396.36 ± 62.88 b 170.88 ± 19.94 c
1.64 ± 0.17 a 2.20 ± 0.10 b
0.79 ± 0.09 a 1.53 ± 0.16 b
35.13 ± 1.20 a
45.75 ± 1.3 a
47.45 ± 0.99 b
60.60 ± 1.01 b
Parameters
ALAT (U/L)
ASAT (U/L)
Cholesterol (mmol/L)
Triglycerides (mmol/L)
Albumin (g/L)
Total proteins (g/L)
150.60 ± 15.25 a
263.80 ± 27.25 a
1.87 ± 0.11 a
0.73 ± 0.09 a
34.92 ± 0.98 a
46.31 ± 1.59 a
*Serum biochemical parameters were assessed after 15 days of GHRP6 or Saline fibrosis regression treatments.
Data are presented as the average ± SEM by group. Different letters indicate significant differences for at
least p < 0.05. ASAT: aspartate aminotransferase; ALAT: alanine aminotransferase.
E
VLDL (mmol/L) (U/L)
0.20
0.30
0.40
0.50
0.60
0.70
0.80
b
c
a
c
a
b
c
Day 1 2 months 5 months
F
Total proteins (g/L)
30
40
50
60
70
c
b
a
a
b
Day 1 2 months 5 months
D
Triglycerides (mmol/L)
0.30
0.50
c
c
b
a
a
b
0.70
0.90
1.10
1.30
1.50
1.70
Day 1 2 months 5 months
C
Cholesterol (mmol/L)
0.50
1.00
1.50
2.00
2.50
3.00
b
a
c
a
b
Day 1 2 months 5 months
A
ASAT (U/L)
100
200
c
c
b
a
a
b
300
400
500
600
700
800
Day 1 2 months 5 months
Intact control
CCl4 + Saline
CCl4 + GHRP6
B
ASAT (U/L)
30
90
150
210
270
b
a
a
b
Day 1 2 months 5 months
330
Jorge Berlanga-Acosta et al. GHRP6 reduces liver fibrosis
67 Biotecnología Aplicada 2012; Vol.29, No.2
to D). The immuno-detection of the α-SMA antibody
appears particularly restricted to the external sides of
the fi brotic septae where it has been reported that acti-
vated HSC are confi ned [27]. It is worthy to highlight
that an appreciable reduction in the number of labeled
cells was a hallmark in the GHRP6-derived samples,
including to those fi elds in which the fi brotic septae
appeared similar in terms of thickening and cellular
density (Figures 3C and D). Similarly, TGF-β ap-
peared far less expressed in those samples derived
from GHRP6-treated rats from both preventive and
regression phases (Figures 3E to H). As discussed
below, our immunostaining for TGF-β also appeared
restricted to the fi brotic septae where it is anchored to
the ECM proteins [28].
Vimentin is also used as a molecular marker for
activated HSC and Kupffer cells [29, 30]. GHRP6-
treated animals within the prevention trial showed far
less Vimentin positive cells (in the fi brotic septae as
within the hepatic parenchyma) than the saline group
(Figures 4A and B). This result was also similar for
the samples of the fi brosis regression trial (data not
shown). As judged by cell morphology and topo-
graphic location (core of the fi brotic septae and he-
patic parenchyma), lineages immunolabeled with the
anti-p53 correspond to recruited round mononuclear
cells, Kupffer and HSC. Remarkably, the most intense
signal and amount of positive anti-p53 cells were de-
tected in the GHRP6 treated animals; in both preven-
tion (data not shown) and regression trials (Figures 4C
and D).
Although double immunohistochemistry was not
conducted, FasL expression specifi cally matched to
Kupffer cells, as suggested by its morphology and
location. A larger increase of FasL positively labeled
Kupffer cells was detected in the Saline-15d group
(14.22 ± 6.17 vs. 2.09 ± 2.12; p < 0.0001). Converse-
ly, the animals of the GHRP6-15d group showed a
remarkable increase of Cyclin D1 positively labeled
hepatocytes nuclei, as compared to Saline-15d group
(8.65 ± 2.36 vs. 2.22 ± 1.59; p < 0.0001), which is
likely an indicative of GHRP6-induced parenchymal
regeneration [31]. A similar result of FasL and Cyclin
D1 expression patterns was obtained in the remaining
experimental groups (data not shown).
RT-PCR analysis
The transcriptional profi le of fi brosis-committed tar-
get genes was studied in the GHRP6-15d and Saline-
15d groups (Figure 5). In comparison to intact rats,
the saline treated animals showed a signifi cant en-
hancement on the transcriptional expression of TGF-β
and connective tissue growth factor (CTGF; both p
< 0.001), two growth factors with well-characterized
role in the fi brogenic process [28]. A signifi cantly
sharp expression reduction for both fi brogenic cytok-
ines was readily observed in the GHRP6-15d group as
compared to the saline counterparts (at least p < 0.05).
The CCl4 also induced a signifi cant decrease of the
superoxide dismutase manganese enzyme (SODMn)
transcriptional expression in the Saline-15d group as
compared to the intact control group (p < 0.05; fi gu-re
5). However, the SODMn transcriptional expression
of the GHRP6-treated rats did not differ from the con-
stitutive levels found in the Intact control group and
was signifi cantly superior to the Saline-15d group (p
< 0.01). Matrix metalloprotease-13 (MMP13) is the
major interstitial collagenase that reduces liver fi bro-
sis by degrading the ECM proteins [32]. A dramatic
expression enhancement of MMP13 transcriptional
Table 6. Serum biochemical assessments of the 60 days fibrosis regression scheme*
GHRP6-15d group
(n = 10)
Saline-15d group
(n = 10)
Intact control group
(n = 7)
180.63 ± 16.20 a 66.29 ± 5.48 b
306.76 ± 20.91 a 184.00 ± 32.63 b
1.52 ± 0.19 a 2.07 ± 0.03 b
0.83 ± 0.17 a 1.99 ± 0.09 b
0.31 ± 0.04 a
26.44 ± 1.05 a
47.34 ± 1.09 a
0.90 ± 0.04 b
41.14 ± 1.11 b
62.55 ± 1.59 b
Parameters
ALAT (U/L)
ASAT (U/L)
Cholesterol (mmol/L)
Triglycerides (mmol/L)
VLDL (mmol/L)
Albumin (g/L)
Total proteins (g/L)
194.82 ± 17.51 a
315.54 ± 22.71 a
1.63 ± 0.10 a
0.85 ± 0.15 a
0.33 ± 0.05 a
27.32 ± 1.08 a
47.32 ± 1.33 a
* Serum biochemical parameters were assessed after 60 days of GHRP6/Saline fibrosis regression treatments. Data are presented as the
average ± SEM by group. Different letters indicate significant differences for at least p < 0.05. ASAT: aspartate aminotransferase; ALAT:
alanine aminotransferase; VLDL: very low density lipoprotein.
Table 7. Hepatic oxidative stress results†
MDA LPP Catalase SOD
0.22 ± 0.01 0.71 ± 0.04 73.5 ± 8.7 17027 ± 667
4.5 ± 0.4 3.7 ± 0.2 203.6 ± 16.7 21500 ± 1033
2.5 ± 0.2*** 2.2 ± 0.2*** 638.7 ± 38.8*** 52492 ± 3080***
3.0 ± 0.3 3.0 ± 0.1 304.0 ± 22.9 24361 ± 858
1.7 ± 0.1***
6.4 ± 0.2
3.0 ± 0.3***
2.2 ± 0.1***
2.8 ± 0.2
1.9 ± 0.2**
499.0 ± 38.8*** 32143 ± 909***
370.0 ± 35.3 33818 ± 1501
505.5 ± 35.8* 50606 ± 1648***
Groups
Intact control
CCl4+Saline
CCl4+GHRP6
Saline-15d
GHRP6-15d
Saline-60d
GHRP6-60d
AOPPTHP
3.9 ± 0.512.1 ± 1.0
24.2 ± 1.586.6 ± 6.8
14.9 ± 0.7***24.1 ± 2.7***
26.8 ± 2.086.6 ± 3.8
14.7 ± 0.4***37.2 ± 2.6***
31.8 ± 1.2153.5 ± 10.4
17.7 ± 1.4***81.2 ± 3.5***
† Data are presented as the average ± SEM of all the animals by group. THP: total hydroperoxide; AOPP: advanced oxidation protein
products; MDA: malondialdehyde; LPP: lipid peroxidation potential; SOD: superoxide dismutase. The values of THP, AOPP, MDA and LPP
are reported as nmoles/mg of total proteins. The enzymatic activities of catalase and SOD are reported as U/min per gram of tissue.
*/**/*** indicate significant differences between the GHRP6-treated animals and their counterpart Saline groups for p < 0.05, p < 0.01
and p < 0.001, respectively.
Biasi F, Albano E, Chiarpotto E, Co-24.
rongiu FP, Pronzato MA, Marinari UM, et
al. In vivo and in vitro evidence concer-
ning the role of lipid peroxidation in the
mechanism of hepatocyte death due to
carbon tetrachloride. Cell Biochem Funct.
1991;9(2):111-8.
Carpino G, Morini S, Ginanni Corra-25.
dini S, Franchitto A, Merli M, Siciliano M,
et al. Alpha-SMA expression in hepatic
stellate cells and quantitative analysis of
hepatic fibrosis in cirrhosis and in recurrent
chronic hepatitis after liver transplantation.
Dig Liver Dis. 2005;37(5):349-56.
Tomanovic N, Boricic I, Brasanac D. Im-26.
munohistochemical analysis of alpha-SMA
and GFAP expression in liver stellate cells.
Vojnosanit Pregl. 2006;63(6):553-7.
Ramm GA, Nair VG, Bridle KR, She-27.
pherd RW, Crawford DH. Contribution of
hepatic parenchymal and nonparenchymal
cells to hepatic fibrogenesis in biliary atre-
sia. Am J Pathol. 1998;153(2):527-35.
Gressner OA, Weiskirchen R, Gressner 28.
AM. Evolving concepts of liver fibrogenesis
provide new diagnostic and therapeutic
options. Comp Hepatol. 2007;6:7.
Wu HH, Tao LC, Cramer HM. Vimen-29.
tin-positive spider-shaped Kupffer cells. A
new clue to cytologic diagnosis of primary
and metastatic hepatocellular carcinoma
by fine-needle aspiration biopsy. Am J Clin
Pathol. 1996;106(4):517-21.
Jorge Berlanga-Acosta et al. GHRP6 reduces liver fibrosis
68 Biotecnología Aplicada 2012; Vol.29, No.2
levels appeared with the GHRP6 intervention as com-
pared to the saline group (p < 0.01).
Microarray and bioinformatic analyses
Microarray experiment was carried out comparing
paired samples 4, 7, 8, 11, 12 in Groups GHRP6-60d
and Saline-60d with the reference sample from the
Intact control group. The 1.5 -fold differentially-ex-
pressed rat genes and their homologous human genes
(ORT-Human), reported in Homologene or identifi ed
by Blastn are shown in table 8. The molecular inter-
action network that was generated with BisoGenet,
using differentially-expressed rat genes as input data,
contained only 41 genes and 25 molecular interac-
tions.
However, a similar network generated from ORT-
Human genes included 386 genes and 1883 interac-
tions. After a gene enrichment analysis performed
with BiNGO, the most biologically signifi cant pro-
cesses were underscored as consequence of the
GHRP6 intervention. These were: ‘oxidation-reduc-
tion’ and ‘response to wounding’ (Table 9). For the
former, members of the cytochrome P450 family as
CYP2A13, CYP2C18, CYP2C19 and CYP2C9; aldo-
ketoreductase (AKR1C1) and UDP glucuronosyl-
transferase, also committed in drugs and xenobiotic
metabolism pathways, were included. Cysteine dioxy-
genase type I (CDO1), participating in redox process,
and the NADP+-dependent isocitrate dehydrogenase 1
(IDH1)/Pipecolic acid oxidase (PIPOX), which par-
ticipate in peroxisome pathway, were also incorporat-
ed within the ‘oxidation-reduction’ process. Besides,
Cdo1, alpha-1-inhibitor 3 (A1i3), coagulation factor
X (F10), histidine-rich glycoprotein (Hrg), serine (or
cysteine) peptidase inhibitor, clade A, members 3N
(Serpina3n) and 3M (Serpina3m), transferrin (TF) and
hepcidin antimicrobial peptide (HAMP) genes were
related to ‘response to wounding’.
Discussion
CCl4 chronic administration induced an overt cirrhotic
disease to otherwise normal rats, which engendered
systemic disturbances for the animal homeostasis.
Herein, we provide the fi rst evidences suggesting
that a classic member of the Bower’s synthetic secre-
tagogue peptides, GHRP6, is not solely endowed with
cardioprotective actions but also with anti-fi brotic ef-
fect. The evidences derived from these experiments
provide the fundamentals to consider that the GHRP6
intervention prevented the progression of a liver fi bro-
genic process and also triggered its regression. This
result substantiates previous fi ndings of our group
[11] and others [13] in terms of GHRP6-mediated he-
patic tissue protection.
In this study three experimental settings were estab-
lished. The prophylactic approach attempted to recre-
ate a clinical condition in which a patient is threat-
ened to evolve to fi brosis following a triggering event.
Further, the therapeutic intervention trial was split in
two clinical schemes, one in which the hepatic chal-
lenge was interrupted and an alternative one related to
a chronic liver insult. This work seems to be the fi rst
preclinical study in which a fi brosis regression effect
is examined in homogeneously allocated groups ac-
cording to a scale of liver gross pathology [15].
AB
CD
EF
GH
Figure 3. Immunohistological detection of alpha smooth muscle actin (α-SMA) and transforming
growth factor beta (TGF-β) expression on fibrotic septae. Representative images of the alpha-SMA
expression on the experimental groups: A) GHRP6-60d; B) Saline-60d; C) GHRP6-15d; and D) Saline-
15d. Representative images of the TGF-β expression on the experimental groups: E) GHRP6-15d; F)
Saline-15 d; G) CCl4+GHRP6; and H) CCl4+Saline. Images A, B, G and H were captured with a 5 ×
magnification, and images C, D, E, and F with a 40 × magnification.
Figure 4. Immunohistological detection of Vimentin and p53 labeled cells. Representative images of
the Vimentin expression on the experimental groups: A) CCl4 + GHRP6 and B) CCl4 + Saline; 20 ×
magnification. Representative images of the p53 expression on the experimental groups: C) GHRP6-
15d and D) Saline-15d; 40 × magnification.
AB
CD
Jorge Berlanga-Acosta et al. GHRP6 reduces liver fibrosis
69 Biotecnología Aplicada 2012; Vol.29, No.2
Intact control Saline-15d GHRP6-15d
TGF-β
CTGF
SODMn
MMP13
β-actin
A
B
Expression ratio
CTGF
TGF-β
0.2
0.3
0.4
0.5
0.6
0.7
0.8
Intact Control Saline-15d
Groups
GHRP6-15d
a
a
b
b
a
c
SODMn
MMP13
C
Expression ratio
0.2
0.3
0.4
0.5
0.6
0.7
Intact Control Saline-15d
Groups
GHRP6-15d
a
a
b
a
a
b
Figure 5. RT-PCR analyses of the TGF-β, connective tissue growth factor (CTGF), superoxide dismutase manganese enzyme
(SODMn) and matrix metalloprotease-13 (MMP13) transcripts. RNA samples from the livers of five rats from the group Saline-
15d, GHRP6-15d and Intact control groups were reverse transcribed. A) PCR amplification products. The expression results
were normalized against β-actin and represented in the graphics (B and C) as the average ± SEM by group, n = 5. Different
letters indicate significant differences for at least p < 0.05.
The histomorphometric assessment as the ultra-
sound study concurrently indicated less fi brosis in all
the GHRP6 intervened animals. The most remarkable
result in terms of fi brosis material reduction was as-
sociated to the GHRP6 preventive scheme, where
the liver parenchyma appeared spared of nodular or-
ganization. Accordingly, the lowest UFI value was
calculated for this group. This fact highlights the
hepatoprotective effect induced by GHRP6 in order
to prevent parenchymal cells downfall and a subse-
quent fi brogenic response. The short term therapeutic
intervention (15 days) accounted for a signifi cant re-
gression of liver fi brosis, mainly expressed in fi brotic
septae involution. Importantly, this effect was attained
in a scenario where no spontaneous fi brotic resolution
took place in the saline group. Parenchymal nodules
clearance along with septae thickness reduction was
also confi rmed in the 60 days GHRP6-therapeutic
scheme. It is noteworthy that this effect appeared in
a scenario of continuous liver aggression as the CCl4
administration was intentionally maintained.
One of the methodological limitations of this study
is the lack of portal venous pressure/fl ow measure-
ments. Alternatively, the most reliable approach un-
dertaken was to assess portal dilation by ultrasound as
described elsewhere [33, 34]. Generally speaking, the
incidence of ascites and portal dilation was scarcely
registered within the GHRP6 groups, thus suggesting
a lesser amount of fi brotic accumulation and a more
physiologic haemodynamic performance. We do not
rule out however, the hypothetical involvement of the
GHRP6 induction of endothelial nitric oxide release
as a key factor controlling portal haemodynamic bal-
ance [35].
Fat quantifi cation results suggest that a magnifi ca-
tion of the fatty liver phenotype is associated to the
co-administration of CCl4 and GHRP6 and not to the
GHRP6 intervention alone. Long term clinical inter-
ventions with the cognate GHRP2 have proved to
be safe in dwarf children [36]. In line with this, our
long-term systemic administration toxicity studies in
healthy rats proved that GHRP6 does not harm the
hepatic tissue (Cosme-Díaz K; unpublished data). The
Cyclin D1 expression profi le incites to speculate that
GHRP6 stimulates hepatocytes mitosis. Whether this
mitogenic response is directly triggered by GHRP6
itself and/or through the GH/Insulin-like growth fac-
tor 1 axis remains to be examined [37]; moreover, re-
generation of the hepatic mass could hypothetically
explain the increase of fat storage in concomitantly
CCl4 + GHRP6-treated rats. It could also explain the
noticeable fat storage reduction in those animals sole-
ly exposed to GHRP6 without a correlate increase in
fat serum markers.
In general terms, ALAT and ASAT serum levels ex-
hibited a biphasic behavior. The acute damage phase,
histologically expressed as hepatic steatosis and non-
parenchymal cells reactivity (data not shown), cor-
related with the highest transaminases levels. From
this point onward, however, fi brosis severity in-
crease appeared associated to a drastic transaminases
drop-down, irrespective to the medication, which
has been previously reported [38, 39]. The fact that
ALAT levels from the GHRP6-preventively treated
Geerts A, Eliasson C, Niki T, Wielant A, 30.
Vaeyens F, Pekny M. Formation of normal
desmin intermediate filaments in mouse
hepatic stellate cells requires vimentin.
Hepatology. 2001;33(1):177-88.
Stacey DW. Cyclin D1 serves as a 31.
cell cycle regulatory switch in actively
proliferating cells. Curr Opin Cell Biol.
2003;15(2):158-63.
Watanabe T, Niioka M, Hozawa S, Ka-32.
meyama K, Hayashi T, Arai M, et al. Gene
expression of interstitial collagenase in
both progressive and recovery phase of rat
liver fibrosis induced by carbon tetrachlori-
de. J Hepatol. 2000;33(2):224-35.
Liu Y, Li L, Yu Z, Liu Q, Li Z, Wang Y, 33.
et al. Correlative study between portal
vein pressure and portal hemodyna-
mics in patients with portal hyperten-
sion. Zhonghua Gan Zang Bing Za Zhi.
2002;10(2):135-7.
Quintanilha LF, Mannheimer EG, Car-34.
valho AB, Paredes BD, Dias JV, Almeida AS,
et al. Bone marrow cell transplant does not
prevent or reverse murine liver cirrhosis.
Cell Transplant. 2008;17(8):943-53.
Fiorucci S, Antonelli E, Morelli A. Nitric 35.
oxide and portal hypertension: a nitric
oxide-releasing derivative of ursodeoxy-
cholic acid that selectively releases nitric
oxide in the liver. Dig Liver Dis. 2003;35
Suppl 2:S61-9.
Mericq V, Cassorla F, Salazar T, Avila A, 36.
Iniguez G, Bowers CY, et al. Effects of eight
months treatment with graded doses of a
growth hormone (GH)-releasing peptide
in GH-deficient children. J Clin Endocrinol
Metab. 1998;83(7):2355-60.
Jorge Berlanga-Acosta et al. GHRP6 reduces liver fibrosis
70 Biotecnología Aplicada 2012; Vol.29, No.2
Table 8. Differentially expressed rat genes and their human homologous*
Gene name (Rat)
hepcidin antimicrobial peptide
ribosomal protein L13
Vimentin
Paraoxonase 2
cytochrome P450, family 2, subfamily a,
polypeptide 1
cysteine dioxygenase, type I
amylase, alpha 1A (salivary)
serine (or cysteine) proteinase inhibitor,
clade A, member 3M
UDP glucuronosyltransferase 2
family; polypeptide B36
coagulation factor X
serine (or cysteine) peptidase inhibitor,
clade A, member3N
cytochrome P450, family 2, subfamily c,
polypeptide 7
pipecolic acid oxidase
selenoprotein P, plasma, 1
cytochrome P450, family 2, subfamily c,
polypeptide 12
cytochrome P450, family 2, subfamily c,
polypeptide 13
aldo-keto reductase family l, member
C14
late cornified envelope 1F
dermatopontin
transmembrane 7 superfamily member 2
alcohol dehydrogenase 1 (class I)
histidine rich glycoprotein
isocitrate dehydrogenase 1
(NADP+), soluble
glyceraldehyde-3-phosphate
dehydrogenase
transferrin
alpha-1-inhibitor III
Gene name (human)
hepcidin antimicrobial peptide
ribosomal protein L13
Vimentin
paraoxonase 2
cytochrome P450, family 2, subfamily A,
polypeptide 13
cysteine dioxygenase, type I
amylase, alpha 2A (pancreatic)
serpin peptidase inhibitor, clade A
(alpha-1 antiproteinase,
antitrypsin), member 3
UDP glucuronosyltransferase 2
farmily, polypeptide B15
coagulation factor X
serpin peptidase inhibitor, clade A
(alpha-1 antiproteinase,
antitrypsin), member 3
cytochrome P450, family 2, subfamily C,
polypeptide 9
pipecolic acid oxidase
selenoprotein P, plasma, 1
cytochrome P450, family 2, subfamily C,
polypeptide 18
cytochrome P450, family 2, subfamily C,
polypeptide 19
aldo-keto reductase family l, member
C1 (dihydrodiol dehydrogenase 1;
20-alpha (3-alpha) - hydroxysteroid
dehydrogenase)
late cornified envelope 1C
dermatopontin
transmembrane 7 superfamily member 2
alcohol dehydrogenase 1C
(class I) gamma polypeptide
histidine rich glycoprotein
isocitrate dehydrogenase 1
(NADP+), soluble
glyceraldehyde-3-phosphate
dehydrogenase
transferrin
alpha- 2- macroglobulin
Gene symbol
(Human)
HAMP
PPL13
VIM
PON2
CYP2A13
CDO1
AMY2A
SERPINA3
UGT2B15
F10
SERPINA3
CYP2C9
PIPOX
SEPP1
CYP2C18
CYP2C19
AKR1C1
LCE1C
DPT
TM7SF2
ADH1C
HRG
IDH1
GAPDH
TF
A2M
Probe naive
in chip
Hamp
Rpl13
Vim
Pori2_predicted
Cyp2a1
Cdo1
Amy1
Serpina3rn
Rn30001813
F10
Spin2c
Cyp2c7
Rn30007995
Sepp1
Cyp2c40
Cyp2c13
LOC191574
Rn30008613
Dpt predicted
Tm7 sf2_predicted
Adh1
Hrg
Idh1
Rn30002246
Tf
Rn30013602
Gene symbol
(Rat)
logFC
Hamp-0.90
Rpl13-0.63
Vim-0.63
Pon20.58
Cyp2a1
Cdo1
Amy1a
Serpina3m
Ugt2b36
0.60
0.66
0.62
0.69
0.85
F10
Serpina3n
Cyp2c7
Pipox
Sepp1
Cyp2c12
0.60
0.67
0.78
0.63
0.70
0.86
Cyp2c13
Akr1c14
Lce1f
Dpt
Tm7sf2
Adh1
Hrg
Idh1
Gapdh
Tf
Ali3
0.62
0.68
0.78
0.63
0.71
0.87
1.02
0.64
0.74
0.93
1.12
*Genes with p < 0.001 and fold change (FC) greater than 1.5 are shown. Negative (positive) values of logFC indicate down-expressed (over-expressed) genes.
group remained similar to the saline counterpart at
the fi fth month could speculatively be associated to
the increase of fat storage and/or to hepatocytes re-
generation. These two factors could be also related
to the fact that in the regression phase, transaminases
were only reduced within the GHRP6-short treatment
scheme where CCl4 injections were interrupted. The
liver synthesis function remained preserved with the
concomitant GHRP6 scheme, as indicative of its he-
patoprotective potential and/or to a hepatocytes’ re-
generation. However, the liver biosynthetic function
did not improve with any of the GHRP6 therapeutic
schemes. The CCl4-induced irreversible inhibition of
macromolecules exocytosis mechanism in the hepato-
cytes may explain this drawback [40, 41]. It is reason-
able to deem that a more prolonged treatment period,
associated to a liver-free insult, would be translated
into a superior GHRP6 therapeutic impact in terms of
hepatic synthesis function.
Aside from the limitations of this work to fully elu-
cidate the GHRP6 anti-fi brogenic mechanism, a major
contribution in this context is the fi nding that GHRP6
induces an enhancement of SOD activity, which ap-
pears mediated via the transcriptional activation of at
least the manganese-dependent isoform. In line with
this observation, the bioinformatics analysis of the mi-
croarray experiment results indicated that GHRP6 en-
hances the action of a series of genes encompassed in
the redox homeostasis as Cdo1. In addition, two other
enzymes whose expression appeared upregulated by
GHRP6 (PIPOX and IDH1) play instrumental roles
in peroxisomal processes as free radicals detoxifi ca-
tion. Furthermore, increases of CYP2A13, CYP2C18,
CYP2C19, CYP2C9, AKR1C1 and UDP glucurono-
syltransferase support the idea of an active detoxifi ca-
tion process, contributing to revert fi brogenesis [42].
This has extended previous fi ndings in which GHRP6
proved to ameliorate radicalary cytotoxicity and to
amplify antioxidant defenses [11]. As reactive oxygen
species are key mediators of the fi brogenic process by
directly activating HSC transdifferentiation [28], its
neutralization and/or pool reduction has been consid-
ered an appropriate strategy to mitigate the fi brogenic
responses [43-46].
Desbois-Mouthon C, Wendum D, 37.
Cadoret A, Rey C, Leneuve P, Blaise A, et
al. Hepatocyte proliferation during liver
regeneration is impaired in mice with
liver-specific IGF-1R knockout. FASEB J.
2006;20(6):773-5.
Poynard T, Munteanu M, Ngo Y, Mous-38.
salli J, Lebray P, Thabut D, et al. FibroTest is
effective in patients with normal transami-
nases, when accuracy is standardized on
fibrosis stage prevalence. J Viral Hepat.
2008;15(6):472-3.
Sebastiani G, Vario A, Guido M, Alberti 39.
A. Performance of noninvasive markers for
liver fibrosis is reduced in chronic hepatitis
C with normal transaminases. J Viral He-
pat. 2008;15(3):212-8.
Becker E, Messner B, Berndt J. Two me-40.
chanisms of CCl4-induced fatty liver: lipid
peroxidation or covalent binding studied
in cultured rat hepatocytes. Free Radic Res
Commun. 1987;3(1-5):299-308.
Jorge Berlanga-Acosta et al. GHRP6 reduces liver fibrosis
71 Biotecnología Aplicada 2012; Vol.29, No.2
Table 9. GO significant processes identified with BiNGO‡
A2M
ADH1C
AKR1C1
AMY2A
CDO1
CYP2A13
CYP2C18
CYP2C19
CYP2C9
F10
GAPDH
HAMP
HRG
IDH1
LCE1C
PIPOX
PON2
SEPP1
SERPINA3
TF
TM7SF2
UGT2B15
GOID GO term
Adjusted
p value
55114 oxidation reduction ** ***** * * * *
1.90 × 10-6
8202 steroid metabolic process *** **
4.98 × 10-3
7598 blood coagulation, extrinsic pathway **
4.98 × 10-3
16098 monoterpenoid metabolic process **
5.60 × 10-3
42738 exogenous drug catabolic process **
8.34 × 10-3
42737 drug catabolic process **
1.11 × 10-2
30194 positive regulation of blood coagulation **
1.73 × 10-2
9611 response to wounding ** ** **
1.73 × 10-2
50820 positive regulation of coagulation **
2.23 × 10-2
2526 acute inflammatory response ***
2.48 × 10-2
50817 coagulation ** *
2.89 × 10-2
17144 drug metabolic process **
2.89 × 10-2
7596 blood coagulation ** *
2.89 × 10-2
51897 positive regulation of protein kinase B **
2.89 × 10-2
7599 hemostasis ** *
2.99 × 10-2
6805 xenobiotic metabolic process * *
3.24 × 10-2
30193 regulation of blood coagulation **
3.24 × 10-2
9056 catabolic process ** ** * * *
3.24 × 10-2
44281 small molecule metabolic process ** * * * * * *
3.24 × 10-2
2543 activation of blood coagulation via clotting *
3.24 × 10-2
55072 iron ion homeostasis **
3.24 × 10-2
50878 regulation of body fluid levels ** *
3.24 × 10-2
6953 acute-phase response **
3.24 × 10-2
6954 inflammatory response ** * **
3.24 × 10-2
10037 response to carbon dioxide *
3.24 × 10-2
9410 response to xenobiotic stimulus **
3.24 × 10-2
42412 taurine biosynthetic process *
3.24 × 10-2
71466 cellular response to xenobiotic stimulus * *
3.24 × 10-2
6879 cellular iron ion homeostasis **
3.24 × 10-2
51896 regulation of protein kinase B signaling **
3.24 × 10-2
6721 terpenoid metabolic process **
3.24 × 10-2
22600 digestive system process **
3.24 × 10-2
32787 monocarboxylic acid metabolic process ****
3.24 × 10-2
43436 oxoacid metabolic process ** * **
3.51 × 10-2
19752 carboxylic acid metabolic process ** * **
3.51 × 10-2
50818 regulation of coagulation **
3.51 × 10-2
61041 regulation of wound healing **
3.51 × 10-2
42221 response to chemical stimulus *** ** ***
3.51 × 10-2
6082 organic acid metabolic process ** * **
3.54 × 10-2
10641 positive regulation of platelet-derived *
3.64 × 10-2
1868 regulation of complement activation, lectin *
3.64 × 10-2
42180 cellular ketone metabolic process ** * **
3.64 × 10-2
98 sulfur amino acid catabolic process *
3.64 × 10-2
46439 L-cysteine metabolic process *
3.64 × 10-2
6097 glyoxylate cycle *
3.64 × 10-2
1869 negative regulation of complement *
3.64 × 10-2
19448 L-cysteine catabolic process *
3.64 × 10-2
9093 cysteine catabolic process *
3.64 × 10-2
44273 sulfur compound catabolic process *
3.64 × 10-2
6952 defense response ** * **
3.97 × 10-2
32101 regulation of response to external stimulus ***
4.35 × 10-2
10667 negative regulation of cardiac muscle cell *
4.46 × 10-2
55098 response to low-density lipoprotein *
4.46 × 10-2
10662 regulation of striated muscle cell apoptosis *
4.46 × 10-2
10665 regulation of cardiac muscle cell apoptosis *
4.46 × 10-2
45916 negative regulation of complement *
4.46 × 10-2
6720 isoprenoid metabolic process **
4.46 × 10-2
10664 negative regulation of striated muscle cell *
4.46 × 10-2
19748 secondary metabolic process **
4.46 × 10-2
42060 wound healing ** *
4.46 × 10-2
‡ Significant GO biological processes are sorted by adjusted p-value obtained with a Benjamini-Hochberg False Discovery Rate (FDR) correction for multiple testing.
GOID: Gene Ontology term identifier. Genes involved in each process are marked with asterisks.
Jorge Berlanga-Acosta et al. GHRP6 reduces liver fibrosis
72 Biotecnología Aplicada 2012; Vol.29, No.2
The infl ammatory response associated to the he-
patocytes’ injury amplifi es the generation of reactive
oxygen species, the recruitment of infl ammatory cells,
as the local release of profi brogenic cytokines [47].
The fact that in the GHRP6 short term therapeutic in-
tervention, Kupffer cells expressed p53 in concomi-
tance to an obvious suppression of FasL expression,
may translate as an ongoing process of Kupffer cells
arrest or apoptotic induction. It would represent a pos-
sible disruption of hepatocytes’ demise and the ensued
activation cascade [48]. It is known that Kupffer cells
promote hepatocytes apoptosis via FasL pathway [49,
50]. Overall, these observations indicate that GHRP6
contributed to quench the liver infl ammatory response
as has been previously described [11, 13]. In vitro
models are in progress to delineate the actual potential
of GHRP6 to prevent HSC activation.
The notion that GHRP6 intervention reduced local
infl ammation and the consequent HSC activation-
perpetuation may be supported by the consistent
reduction of α-SMA and Vimentin expression, as
markers of an activation phenotype. In line with this,
GHRP6-treated animals exhibited far less expression
of the two major fi brogenic growth factors TGF-β and
CTGF which may anticipate a decrease in ECM pro-
tein synthesis and accumulation [28]. Furthermore,
GHRP6 increased MMP13 transcriptional expression
which may be obviously linked to fi brosis reduction
due to collagen degradation [32]. It obviously would
be complemented by a functional enzymatic assay.
Growth hormone secretagogues regulate the tran-
scriptional activation of the peroxisome prolifera-
tor-activated receptor gamma (PPARγ) through the
concerted interaction with the two types of receptors
identifi ed: CD36 and the ghrelin GHS-R1a. PPARγ
can also induce CD36 expression thus establishing a
mutually positive regulatory loop [51]. PPARγ agonis-
tic stimulation markedly inhibited HSC proliferation,
induced cells apoptosis and signifi cantly suppressed
TGF-β1-induced CTGF expression [52]; which is
in correspondence to our immunohistology and mi-
croarray data, suggesting an abortion of the HSC ac-
tivation program. Likewise, Milam and co-workers
recently confi rmed the PPARγ agonist ligands inhibit
the ability of TGF-β to promote myofi broblast trans-
differentiation as the ensuing collagen secretion and
fi brotic induration in a murine model of lung fi brosis
[53]. The above demonstrations render hypothetical
mechanistic involvement for the GHRP6-mediated
PPARγ activation via CD36 occupation. This may
represent a not redundant pathway for novel anti-
fi brotic agents.
In summary, the main contribution of this work
is the unprecedented demonstration that GHRP6
reduces the fi brotic induration of a parenchymal or-
gan in a renowned experimental model. The mecha-
nism whereby GHRP6 reduces fi broplasias seems to
be multifactorial involving a broad number of gene
products concerned to detoxifi cation pathways, redox
homeostasis, and response to injury. Distally, it ap-
pears to nurture hepatocytes and deactivate stroma-
associated cells. These preliminary fi ndings justify
further studies on the role of the GHS for the control
of fi brotic diseases.
Acknowledgements
Authors are indebted to Dr. Bienvenido Gra, Head of
the Pathology Department from the Institute of Gas-
troenterology, Havana, Cuba, for his valuable assis-
tance while examining the slides.
Poli G, Cottalasso D, Pronzato MA, 41.
Chiarpotto E, Biasi F, Corongiu FP, et al. Lipid
peroxidation and covalent binding in the early
functional impairment of liver Golgi apparatus
by carbon tetrachloride. Cell Biochem Funct.
1990;8(1):1-10.
Takahara Y, Takahashi M, Wagatsuma 42.
H, Yokoya F, Zhang QW, Yamaguchi M, et
al. Gene expression profiles of hepatic cell-
type specific marker genes in progression
of liver fibrosis. World J Gastroenterol.
2006;12(40):6473-99.
Shen XH, Cheng WF, Li XH, Xie LM, Sun 43.
J, Li F, et al. Effect of antioxidant on hepatic
stellate cell proliferation and apoptosis during
recovery from liver fibrosis. Wei Sheng Yan Jiu.
2005;34(2):194-6.
Di Sario A, Candelaresi C, Omenetti 44.
A, Benedetti A. Vitamin E in chronic liver
diseases and liver fibrosis. Vitam Horm.
2007;76:551-73.
El-Demerdash E, Salam OM, El-Batran 45.
SA, Abdallah HM, Shaffie NM. Inhibition of
the renin-angiotensin system attenuates the
development of liver fibrosis and oxidative
stress in rats. Clin Exp Pharmacol Physiol.
2008;35(2):159-67.
Ogeturk M, Kus I, Pekmez H, Yekeler H, 46.
Sahin S, Sarsilmaz M. Inhibition of carbon te-
trachloride-mediated apoptosis and oxidative
stress by melatonin in experimental liver fibro-
sis. Toxicol Ind Health. 2008;24(4):201-8.
Bataller R, Brenner DA. Liver fibrosis. J Clin 47.
Invest. 2005;115(2):209-18.
Laptenko O, Prives C. Transcriptional regu-48.
lation by p53: one protein, many possibilities.
Cell Death Differ. 2006;13(6):951-61.
Pianko S, Patella S, Ostapowicz G, Des-49.
mond P, Sievert W. Fas-mediated hepatocyte
apoptosis is increased by hepatitis C virus in-
fection and alcohol consumption, and may be
associated with hepatic fibrosis: mechanisms
of liver cell injury in chronic hepatitis C virus
infection. J Viral Hepat. 2001;8(6):406-13.
Canbay A, Higuchi H, Bronk SF, Taniai M, 50.
Sebo TJ, Gores GJ. Fas enhances fibrogenesis
in the bile duct ligated mouse: a link between
apoptosis and fibrosis. Gastroenterology.
2002;123(4):1323-30.
Demers A, Rodrigue-Way A, Tremblay A. 51.
Hexarelin Signaling to PPARgamma in Metabo-
lic Diseases. PPAR Res. 2008;2008:364784.
Sun K, Wang Q, Huang XH. PPAR ga-52.
mma inhibits growth of rat hepatic stellate
cells and TGF beta-induced connective tissue
growth factor expression. Acta Pharmacol Sin.
2006;27(6):715-23.
Milam JE, Keshamouni VG, Phan SH, Hu B, 53.
Gangireddy SR, Hogaboam CM, et al. PPAR-ga-
mma agonists inhibit profibrotic phenotypes in
human lung fibroblasts and bleomycin-induced
pulmonary fibrosis. Am J Physiol Lung Cell Mol
Physiol. 2008;294(5):L891-901.
Received in October, 2011. Accepted
Received in October, 2011. Accepted
for publication in February, 2012.
for publication in February, 2012.
