Digestive and Liver Disease 52 (2020) 314–323

Contents lists available at ScienceDirect

Digestive and Liver Disease

journal homepage: www.elsevier.com/locate/dld

Liver, Pancreas and Biliary Tract

L-carnitine supplementation attenuates NAFLD progression and

cardiac dysfunction in a mouse model fed with methionine and
choline-deficient diet
Giulia Mollica a,b,1 , Pamela Senesi a,1 , Roberto Codella a,b , Fernanda Vacante b ,
Anna Montesano a , Livio Luzi a,b , Ileana Terruzzi a,b,∗
a
Department of Biomedical Sciences for Health, Università degli Studi di Milano, Milan, Italy
b
Metabolism Research Center, IRCCS Policlinico San Donato, Milan, Italy

a r t i c l e i n f o a b s t r a c t

Article history: Non-alcoholic fatty liver disease (NAFLD) is a common cause of chronic liver disorder. NAFLD, associated
Received 18 April 2019 lipotoxicity, fibrosis, oxidative stress, and altered mitochondrial metabolism, is responsible for systemic
Accepted 4 September 2019 inflammation, which contributes to organ dysfunction in extrahepatic tissues, including the heart.
Available online 10 October 2019
We investigated the ability of L-carnitine (LC) to oppose the pathogenic mechanisms underlying NAFLD
progression and associated heart dysfunction, in a mouse model of methionine-choline-deficient diet
Keywords:
(MCDD). Mice were divided into three groups: namely, the control group (CONTR) fed with a regular diet
Cardiac fibrosis
and two groups fed with MCDD for 6 weeks. In the last 3 weeks, one of the MCDD groups received LC
Lipid accumulation
Liver fibrosis
(200 mg/kg each day) through drinking water (MCDD + LC).
Oxidative stress The hepatic lipid accumulation and oxidative stress decreased after LC supplementation, which also
reduced hepatic fibrosis via modulation of ␣-smooth muscle actin (␣SMA), peroxisome-activated recep-
tor gamma (PPAR␥), and nuclear factor kappa B (NfB) expression. LC ameliorated systemic inflammation,
mitigated cardiac reactive oxygen species (ROS) production, and prevented fibrosis progression by acting
on signal transducer and activator of transcription 3 (STAT3), extracellular signal-regulated kinase 1-2
(ERK1-2), and ␣SMA.
This study confirms the existence of a relationship between fatty liver disease and cardiac abnormalities
and highlights the role of LC in controlling liver oxidative stress, steatosis, fibrosis, and NAFLD-associated
cardiac dysfunction.
© 2019 The Authors. Published by Elsevier Ltd on behalf of Editrice Gastroenterologica Italiana S.r.l.
This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-
nc-nd/4.0/).

1. Introduction lation in the liver results in the activation of multiple cellular stress
pathways and induces the generation of reactive oxygen species
The liver performs essential functions in the regulation of (ROS), which in turn causes the transdifferentiation of hepatic
glucose and lipid metabolism during feeding and fasting. Non- stellate cells into activated myofibroblasts expressing ␣-smooth
alcoholic fatty liver disease (NAFLD), also known as hepatic muscle actin (␣SMA) [3–5]. Some authors have recently proposed
steatosis, is characterized with an excessive fat deposition in the caspase-2 as a new NASH marker, as its expression increases with
liver and presents a histological spectrum of conditions ranging the worsening of fibrosis in humans and animal models of NASH
from simple steatosis to nonalcoholic steatohepatitis (NASH). Hep- [6–8].
atocellular damage, lobular necroinflammation and fibrogenesis Together these effects trigger the expression of nuclear factor
are the main features of NASH [1]. According to the “Two Hits kappa B (NF␬B), a crucial inflammatory signal that may exacer-
Hypothesis”, NAFLD development requires the double hit, steatosis bate NAFLD progression [9,10]. Furthermore, activated peroxisome
and oxidative stress, wherein the latter leads to an increase in lipid proliferator-activated receptors ␥ (PPAR␥) inhibits NF␬B transcrip-
peroxidation, inflammation, and fibrosis [2]. The initial fat accumu- tion factor [11,12].
Evidence suggests that NAFLD not only represents a potentially
progressive liver disease but also has systemic consequences [13].
∗ Corresponding author. In particular, recent reports from the Framingham Heart study
E-mail address: [email protected] (I. Terruzzi). revealed the significant association between NAFLD and subclinical
1
GM and PS contributed equally to this work.

https://doi.org/10.1016/j.dld.2019.09.002
1590-8658/© 2019 The Authors. Published by Elsevier Ltd on behalf of Editrice Gastroenterologica Italiana S.r.l. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/4.0/).
 G. Mollica et al. / Digestive and Liver Disease 52 (2020) 314–323 315

cardiovascular diseases (CVD) [14]. Patients with NAFLD showed (HT501128), and Bouin’s Solution (HT10132) were all procured
myocardial metabolism disorders with impaired heart functions from Sigma Chemical Co. (Saint Louis, MO, USA).
and structure, i.e., with cardiovascular insufficiency. Several stud-
ies have demonstrated the oxidative stress-induced activation
2.2. Experimental animal model for NAFLD
of extracellular signal-regulated kinase (ERK) pathways and the
increase in ␣SMA and collagen expression, which eventually con-
Eight-week-old male C57BL/6 mice (n = 30) were used in this
tribute to fibrosis [15]. Furthermore, recent data have indicated
study (Charles River Laboratories, Calco, Bergamo, Italy), which was
the influence of signal transducer and activator of transcription 3
conducted in compliance with the approved institutional animal
(STAT3) on the accumulation of fibrotic proteins in the heart [16].
care of the Università degli Studi di Milano.
Therefore, innovative NAFLD treatment strategies should focus
All animals were maintained in a 12 h light/dark cycle with
on both hepatic and cardiovascular damages. Studies have evalu-
unlimited access to standard rodent chow, food, and water under
ated the effects of nutraceuticals as a co-adjuvant treatment for
controlled temperature (22 ± 2 ◦ C). The study design is shown in
NAFLD and cardiovascular complications. In some clinical trials,
Fig. 1A.
nutraceuticals were found to ameliorate the liver function and his-
After two weeks of adaptation, 10-week-old mice were divided
tology [17]. L-carnitine (LC) is an essential nutrient that converts
into three groups (n = 10 each). One group was fed with 120 g/week
fats into energy in the mitochondrion and plays an important role
of regular diet (CONTR) and two groups were fed with 120 g/week of
in lipid metabolism. It acts as an essential cofactor for ␤-oxidation
MCDD (ssniff® EF R/M Induction of fatty liver — Ssniff Spezialdiäten
by facilitating the transport of long-chain fatty acids. The process
GmbH: 83 KJ% carbohydrate, 8 KJ% fat, 9 KJ% protein, ME calculated
of ␤-oxidation is one of the principal pathways underlying lipid
with Atwater factor) for 3 weeks. Thereafter, one of the MCDD
metabolism. The inhibition of ␤-oxidation leads to the accumula-
groups was provided with water-diluted 200 mg/kg of oral LC
tion of lipids within hepatocytes. LC administration was shown to
(MCDD + LC) daily for 3 weeks. Control and MCDD mice received
ameliorate or prevent the liver from insult through the augmenta-
water without LC. The LC dose was adjusted to net body weight.
tion of hepatic mitochondrial ␤-oxidation [18]. Recent studies have
We used CONTR group as an internal control of NAFLD progression
suggested the role of oxidative pathways as favourable targets for
induced by MCDD.
the treatment of NAFLD [19]. LC has been recognized as a nutri-
At the end of the 6 weeks, all animals were euthanized for fur-
tional supplement with anti-oxidative properties [20] and used as
ther analysis.
an adjunctive therapy in various heart conditions with promis-
ing results [21]. In particular, Strilakou et al. have described the
beneficial effects of LC on a cardiomyopathy rat model fed with a 2.3. Histopathological analysis
choline-deficient diet [22].
In the present study, we investigated the antioxidative effects Upon sacrifice, the liver and heart tissues were excised, weighed,
of LC on both liver and cardiac complications of steatohepatitisin a and sampled for histological analysis.
mouse model fed with methionine-choline-deficient diet (MCDD). Liver and heart cryosections were examined with differ-
MCDD was used to study the ability of LC to prevent liver fat ent staining techniques. Tissue sections were examined with
deposition, oxidative stress, and fibrosis progression and counter- haematoxylin and eosin staining according to the manufacturer’s
act the NAFLD-induced cardiac inflammation and fibrosis. instruction. Images were acquired with phase-contrast microscopy
and steatosis score was calculated with ImageJ program (http://
imagej.nih.gov/ij/).
2. Materials and methods Staining for intracellular lipid content was performed with Oil
Red O (ORO) technique according to the manufacturer’s instruction.
2.1. Materials Masson-Goldner staining involving a trichrome stain was used
to primarily image the connective tissue in organs (nuclei were
Reagents were purchased from Sigma Chemical Co. (Saint stained black or blue; cytoplasm and muscle were stained red;
Louis, MO, USA). LC was obtained from Sigma-Aldrich (C0283). collagen fibers were stained blue–green) [23,24].
Primary antibodies against glyceraldehyde 3-phosphate dehydro-
genase (GAPDH; sc-25778), NF␬B p65 (sc-109), PPAR␥ (sc-7196),
␣SMA (sc-53142), phosphor-Ca2+ /calmodulin-dependent protein 2.4. Western blot analysis
kinase II␣ (pCaMKII␣; 22B1), ERK 1-2 (K-23) and pERK 1-2 (E-4),
peroxidase-conjugated secondary antibodies for western blot anal- Liver and heart protein extracts were obtained from homoge-
ysis, and fluorescein isothiocyanate (FITC)/rhodamine-conjugated nized mouse tissues using radioimmunoprecipitation assay (RIPA)
antibodies for immunofluorescence analysis were supplied by buffer [25]. Protein contents were quantified using the Brad-
Santa Cruz Biotechnology (Santa Cruz, CA, USA). Primary anti- ford method. Aliquots of quantified 30 ␮g supernatant proteins
bodies against STAT3 (#9132), pSTAT3 (#9131), and CaMKII␣ were resolved on sodium dodecyl sulfate polyacrylamide gel elec-
(#50049) were obtained from Cell Signaling Technology (Dan- trophoresis (SDS-PAGE) gels and separated protein bands were
vers, MA, USA). Anti-caspase-2 (ab179520) and anti-Cytochrome transferred onto nitrocellulose membranes (Protran® , Whatman®
C (COX; ab110325) antibodies were procured from Abcam (Cam- Schleicher & Schuell). The membranes were incubated with spe-
bridge, UK). cific primary antibodies, followed by treatment with horseradish
Cell ROX® Oxidative Stress Reagent Kit (C10443) was purchased peroxidase-conjugated species-specific secondary antibodies. To
from Thermo Fisher Scientific, Life Technologies Italia (Monza- confirm equal protein loading per sample, anti-GAPDH was used.
Italy) and Masson-Goldner staining kit (100485), Weigert’s iron Quantitative measurement of immunoreactive bandintensities was
haematoxylin kit (115973), Entellan® Neo (107961) were obtained performed with the enhanced chemiluminescence method (Amer-
from Merck Millipore, Merck KGaA (Darmstadt, Germany). sham Pharmacia Biotech, Piscataway, NJ, USA) using densitometric
Mayer’s Haematoxylin Solution (MHS16), 0.5% aqueous eosin analysis with the Scion Image software (Scion Corporation, Freder-
Y solution (HT110216), Scott’s Tap Water Substitute Concentrate ick, MD, USA). Only caspase-2 and COX bands were visualized with
(S5134), 0.5% Oil Red O solution in isopropanol (O1391), glyc- the UVITEC Alliance LD9 gel imaging system (Uvitec, Cambridge,
erol gelatin aqueous slide mounting medium (GG1), 10% formalin UK). Data were converted into ratio relative to the control.
316 G. Mollica et al. / Digestive and Liver Disease 52 (2020) 314–323

Fig. 1. Experimental animal model fornon-alcoholic fatty liver disease (NAFLD). (A) Animal experiment scheme. (B) Mice fed with methionine-choline-deficient diet (MCDD)
showed a significant loss in body weight as compared with those from the control group (CONTR). This condition remained unchanged despite supplementation with L-
carnitine (LC) (n = 10/treatment group: control group, CONTR; methionine-choline deficient diet, MCDD; methionine-choline deficient diet supplemented with L-carnitine,
MCDD + LC).
Data are presented as mean ± SD. ***p < 0.001 as compared with the mice fed with the control diet.

2.5. Immunofluorescence analysis mary and secondary antibodies (rhodamine/FITC-conjugated) and

the nuclei were stained with 4 ,6-diamidino-2-phenylindole (DAPI)
Liver and heart cryosections were fixed with 4% paraformalde- [26].
hyde for 30 min at room temperature. The sections were washed Cell ROX® oxidative stress reagents are fluorogenic probes
with phosphate-buffered saline (PBS), permeabilized with 0.2% Tri- designed to reliably measure ROS levels in tissues. These cell-
ton X-100, and incubated for 30 min at room temperature with 10% permeable reagents are non-fluorescent or very weakly fluorescent
serum. The cryosections were immunostained with specific pri- but emit a strong fluorogenic signal under reduced state and upon

Table 11
Liver mass analysis and steatosis score. Relative liver mass is expressed as the percentage of body mass. Steatosis score: 0 (<5%), 1 (5–33%), 2 (33–66%), 3 (>66%). Groups:
control group-CONTR; methionine-choline deficient diet-MCDD; methionine-choline deficient diet supplemented with L-carnitine-MCDD + LC. n = 5/treatment group. Data
are presented as mean ± SD. *p < 0.05 as compared with the mice fed with the control diet.

CONTR MCDD MCDD + LC

Liver mass (g) 1.65 ± 0.21 0.90 ± 0.27 0.94 ± 0.13

Relative liver mass (as % of body mass) 5.46 ± 0.83 4.68 ± 1.19 4.75 ± 0.68
Percentage differences in relative liver mass MCDD and MCDD + LC vs. CONTR (%) 14.2 13.0
Percentage differences in relative liver mass MCDD vs. MCDD + LC (%) 1.4
Steatosis score 0 (0.81% ± 0.51) 1 (18.84% ± 4.23)* 1 (13.36% ± 3.16)*
 G. Mollica et al. / Digestive and Liver Disease 52 (2020) 314–323 317

Fig. 2. Effects of L-carnitine on liver morphology, histology, lipid accumulation, and oxidation. (A) No macroscopic differences were observed in the three mouse groups
(n = 10/treatment group: control group, CONTR; methionine-choline-deficient diet, MCDD; methionine-choline-deficient diet supplemented with L-carnitine, MCDD + LC).
(B) Haematoxylin and eosin-stained sections showed differences in steatosis and ballooning in the liver of mice fed with methionine-choline-deficient diet (MCDD).
n = 5/treatment group. (C) Oil Red O (ORO)-stained sections confirmed hepatic lipid accumulation in the mice from methionine-choline-deficient diet (MCDD) group. LC
reduced hepatic fat accumulation. n = 5/treatment group. (D) We used the Cell ROS assay to evaluate the effect of L-carnitine on the liver of mice. Methionine-choline-
deficient diet (MCDD) increased the production of reactive oxygen species (ROS). L-carnitine supplementation ameliorated this effect. n = 5/treatment group. (E) Western blot
analysis showed that MCDD significantly improved cytochrome C (COX) protein level as compared with control diet. This increase was significantly reduced after L-carnitine
supplementation.
Data are presented as mean ± SD. **p < 0.01 as compared with the mice fed with the control diet. # p < 0.05 as compared with the mice fed with methionine-choline-deficient
(MCD) diet.
318 G. Mollica et al. / Digestive and Liver Disease 52 (2020) 314–323

oxidation. Cell ROX® Orange Reagents are localized in the cyto- LC treatment counteracted this effect (ratio to CONTR: 1.27 ± 0.2,
plasm. [27] p ≤ 0.05 versus MCDD group).
Slides were mounted with Moviol. Tissue sections were
observed under a Nikon Eclipse 50I microscope and images were 3.4. Role of LC in hepatic fibrosis progression
captured using Nis-Elements D 4.00 software (Nikon Instruments
Europe BV, Netherlands). To investigate whether LC supplementation could delay fibrosis
development, Masson-Goldner staining was performed. Liver fibro-
2.6. Statistical analysis sis was higher in MCDD group than in the CONTR group. Blue–green
colour, specific for collagen-I, was widespread and corresponded to
All data were presented as the mean ± standard deviation (SD). a strong intensity. Mice supplemented with LC diet showed ame-
The Shapiro–Wilk test was used to ascertain normality of data dis- lioration in the progression of NAFLD associated with fibrosis as
tribution. Every mouse tissue was analyzed thrice. compared with MCDD-fed mice (Fig. 3A).
A two-way (treatment × time) analysis of variance (ANOVA) fol- We evaluated the level of ␣SMA protein to dermine the rate
lowed by Tukey’s multiple comparison test were used to analyse of hepatic fibrogenesis. As shown in Fig. 3B, 3 weeks of LC supple-
body weight changes throughout the experimental time course. mentation induced a decrease in ␣SMA level as compared to MCDD
For variables with Gaussian distribution, one-way ANOVAwas diet. Caspase-2 protein level also significantly increased in MCDD
used to compare the means among groups, followed by Tukey’s mice (ratio to CONTR: 1.52 ± 0.3, p ≤ 0.05 versus CONTR group) but
post-hoc test. The Kruskal–Wallis test followed by Dunn’s post-hoc remained low and was similar in CONTR and LC groups (ratio to
test were used for the variables with non-Gaussian distribution. CONTR: 1.1 ± 0.1) (Fig. 3C). No significant differences were detected
A value of p ≤ 0.05 was considered significant. Statistical analysis in procaspase-2 protein expression between the groups.
was performed with GraphPad Prism 7 software (San Diego, CA). The effect of LC on fibrosis progression was supported by west-
ern blot analysis result. LC supplementation caused a significant
increase in the level of hepatic PPAR␥ as compared to CONTR and
3. Results
MCDD treatment (ratio to CONTR: 1.5 ± 0.01, p ≤ 0.05 versus CONTR
and MCDD groups) (Fig. 3D).
3.1. Changes in body weight
We also evaluated NFkB level in inflamed livers and found it
to be significantly elevated in mice from MCDD group as compared
Animals fed with MCDD showed a loss in body weight, con-
with those from the CONTR group (ratio to CONTR: 5 ± 0.5, p ≤ 0.001
sistent with the histological features of human NASH [28]. MCD
versus CONTR group) (Fig. 3D). NFkB protein content significantly
diet is ideal for studying NASH development and, in particular, for
decreased after LC treatment (ratio to CONTR: 4.1 ± 0.2, p ≤ 0.05
the analysis of mechanisms underlying inflammation, fibrosis and,
versus MCDD group, p ≤ 0.001 versus CONTR group) (Fig. 3D).
oxidation [29].
The 6-week administration of MCDD resulted in a significant
3.5. Heart morphology and histology
decrease in the body weight of mice from MCDD group as com-
pared with those from the control group (18.8 ± 0.9 g vs 30.6 ± 0.8 g,
Among the three groups, cardiac tissues did not showed altered
respectively — p ≤ 0.001) [30]. This difference remained unchanged
morphology in haematoxylin and eosin staining (Fig. 4A), and dif-
despite supplementation with LC (19.9 ± 0.4 g — p ≤ 0.001 versus
ferences in the intracellular accumulation of lipids were observed
CONTR group) (Fig. 1B).
with ORO-staining (Fig. 4B).

3.2. Liver mass, morphology, histology and lipid accumulation 3.6. Cardiac antioxidant effects of LC

Based on the effect of MCDD on body weight, hepatic mass was Oxidant stress is an important feature of CVD. Under stress
calculated as the percentage of body mass (relative liver mass). condition, CaMKII activity (by phosphorylation) increases in the
Liver mass showed no significant differences among the three myocardium [33]. ROS level was higher in the heart samples of
groups. The livers from CONTR group were 14.2% and 13% heavier mice from MCDD group than in those of the mice from the CONTR
than those from MCDD and MCDD + LC groups, respectively. The liv- group (Fig. 4C), and LC supplementation ameliorated this effect.
ers from MCDD + LC group were 1.4% heavier than those from MCDD These data were verified with western blotting. CaMKII activa-
group (Table 1). Moreover, the samples from MCDD group appeared tion increased in the MCD-fed mice (ratio to CONTR: 1.26 ± 0.01,
paler than those from CONTR and MCDD + LC groups (Fig. 2A). p ≤ 0.05 versus CONTR group) while LC treatment reduced the
Steatosis was scored on HE-stained sections according to Brunt’s phosphorylated form of CaMKII (ratio to CONTR: 1 ± 0.1) (Fig. 4D).
criteria [31]. As shown in Fig. 2B, 6 weeks of MCDD treatment
resulted in the development of hepatocyte steatosis and balloon- 3.7. Role of LC in cardiac fibrosis progression
ing (steatosis score: 1–18.84%) while LC supplementation seemed
to mitigate this condition (steatosis score: 1–13.36%) (Table 1). The ratio of pERK 1-2/ERK 1-2 was investigated in cardiac tissues
As shown in Fig. 2C, LC treatment ameliorated hepatic fat accu- with western blot analysis. As shown in Fig. 5A, MCDD significantly
mulation, as evident from ORO staining results. increased ERK activation in mice fed with MCDD as compared with
those from the CONTR group (ratio to CONTR: 1.18 ± 0.02, p ≤ 0.05
3.3. Hepatic antioxidant effects of LC versus CONTR group). LC supplementation counteracted this effect.
pERK 1-2/ERK 1-2 ratio was similar between CONTR and MCDD + LC
Although LC was unable to restore ROS levels to those observed groups (ratio to CONTR: 1 ± 0.03, p ≤ 0.05 vs MCDD group).
in the CONTR group, it could limit ROS production in mice from We also investigated the role of STAT3, a mediator of oxidative
MCDD + LC group as compared to those from MCDD group (Fig. 2D). stress and/or metabolism in cardiomyocytes. [16] As reported in
Moreover, we evaluated the protein level of COX involved in ROS Fig. 5B, LC supplementation significantly decreased pSTAT3/STAT3
production in MCDD mice [32]. As reported in Fig. 2E, a a signif- ratio (ratio to CONTR: 0.7 ± 0.2, p ≤ 0.05 versus MCDD group), which
icant increase in COX level was observed in MCDD mice (ratio to was significantly higher in MCDD group than in CONTR group (ratio
CONTR: 1.64 ± 0.2, p ≤ 0.01 as compared with CONTR group), while to CONTR: 1.56 ± 0.1, p ≤ 0. versus CONTR group).
 G. Mollica et al. / Digestive and Liver Disease 52 (2020) 314–323 319

Fig. 3. Effect of L-carnitine on hepatic fibrosis. (A) Masson-Goldner (MG)-trichrome-stained sections showed fibrotic areas in the liver samples from methionine-choline-
deficient diet (MCDD) group (magnification 20 ). n = 5/treatment group. (B) Immunofluorescence assay revealed the increase in the level of ␣-smooth muscle actin (␣-SMA)
protein in the liver samples from mice fed with methionine-choline-deficient diet (MCDD). This increase was attenuated after L-carnitine supplementation. n = 5/treatment
group. (C) Caspase-2 protein level was significantly higher in mice from MCDD group. In the other two groups, the levels were similar. We failed to observe any differences
in procaspase-2 protein levels. n = 5/treatment group. (D) Western blot data indicated that L-carnitine (LC) supplementation significantly increased the hepatic peroxisome
proliferator-activated receptors ␥ (PPAR␥) protein level as compared with the methionine-choline deficient diet (MCDD), and decreased the activation of nuclear factor
kappa B (Nf␬B) p65. n = 5/treatment group.
Data are expressed as the ratio relative to the control group ± SD. *p < 0.05 as compared to the mice fed with the control diet. ***p < 0.001 as compared to the mice fed with
the control diet. # p < 0.05 as compared to the mice fed with methionine-choline-deficient diet (MCDD).

We evaluated fibrosis in heart tissues with Masson-Goldner 4. Discussion

staining, (Fig. 5C). LC treatment ameliorated myocardial fibrosis
and decreased cardiac ␣SMA signal (Fig. 5D) as compared with Accumulating evidence has shown that NAFLD not only affects
MCDD treatment, confirming the delay in the progression of fibro- the liver, but also the extrahepatic tissues such as the heart. Natu-
sis. ral molecules have been successfully used for the prevention and
320 G. Mollica et al. / Digestive and Liver Disease 52 (2020) 314–323

Fig. 4. Effects of L-carnitine on heart histology, lipid accumulation, and oxidation. (A) Haematoxylin and eosin-stained cardiac tissues showed no diffused vacuolar degener-
ation in the three groups (control group, CONTR; methionine-choline deficient diet, MCDD; methionine-choline-deficient diet supplemented with L-carnitine, MCDD + LC).
Alteration in cardiomyocyte size and morphology were not significant (magnification 20×). n = 5/treatment group. (B) Oil Red O (ORO)-stained sections showed the absence
of lipid accumulation in all three groups (magnification 20×). n = 5/treatment group. (C) Cell ROS assay was used to evaluate the effect of L-carnitine (LC) on the heart
tissue. Reactive oxygen species (ROS) production increased in the mice fed with methionine-choline-deficient diet (MCDD). L-Carnitine treatment attenuated this effect.
n = 5/treatment group. (D) Western blot data showed that methionine-cholinedeficient diet (MCDD) improved the phosphorylated form of Ca2+ /calmodulin-dependent pro-
tein kinase II (CaMKII). L-carnitine treatment reduced the level of this kinase. n = 5/treatment group.
Data are expressed as the ratio relative to the control group ± SD. *p < 0.05 as compared with the mice fed with the control diet.
 G. Mollica et al. / Digestive and Liver Disease 52 (2020) 314–323 321

Fig. 5. Role of LC in cardiac fibrosis progression. (A) Western blot analysis indicated that methionine-choline-deficient diet (MCDD) significantly increased the activation of
extracellular-signal-regulated kinases (ERKs) as compared with the control (CONTR) diet. LC supplementation counteracted this effect. n = 5/treatment group. (B) The ratio
of phosphorylated and non-phosphorylated signal transducer and activator of transcription 3 (pSTAT3/STAT3) was similar in the mice from methionine-choline-deficient
diet supplemented with L-carnitine (MCDD + LC) and control groups, while MCDD. significantly increased n = 5/treatment group. (C) Masson-Goldner-trichrome stained
sections showed fibrotic areas in the heart samples from the mice fed with MCDD. LC treatment ameliorated myocardial fibrosis (magnification 20×). n = 5/treatment group.
(D) Immunofluorescence assay for ␣-smooth muscle actin (␣SMA) in the mice fed with methionine-choline-deficient diet (MCDD. LC treatment ameliorated this effect.
n = 5/treatment group.
Data are expressed as ratio relative to the control group ± SD. *p < 0.05 as compared with the mice fed with control diet. # p < 0.05 as compared with the mice fed with
methionine-choline-deficient diet (MCDD).

treatment of NAFLD [34]. The results of the present study demon- At the cellular level, steatosis is one of the key histolog-
strate the favourable effects of LC on hepatocyte steatosis, oxidative ical features of NAFLD that evolve into steatohepatitis [35].
stress, and fibrosis development in the liver and myocardium of In our model, LC supplementation decreased the steatosis
C57BL/6 male mice fed with MCDD. score (Table 1, Fig. 2B, C), indicative of the recovery from
322 G. Mollica et al. / Digestive and Liver Disease 52 (2020) 314–323

MCD-induced NALFD and prevention of steatohepatitis progres- group but decreased in MCDD + LC group. Thus, LC supplementa-
sion. tion ameliorated hepatic liver condition by acting on NF␬B p65 in
Liver fat accumulation is associated with several pathogenic a PPAR␥-dependent manner (Fig. 3D). Future studies should elu-
mechanisms involved in NASH development, including oxidative cidate how the LC-activated PPAR/NFkB axis could regulate the
fat-induced injury [1,5], the major source of ROS [36]. Inhibition of expression of pro-inflammatory cytokines.
mitochondrial fatty acid oxidation is thought to be a major cause Hepatic steatosis induces systemic low-grade inflammation,
of intrahepatic lipid accumulation and ROS production exacerba- possibly leading to cardiac inflammation and fibrosis [58]. NAFLD
tion [1–3]. Considering the role of LC in ␤-oxidation, studies have could be ameliorated by LC administration, as evident from the
suggested the potential use of LC for the treatment of lipotoxic- lower inflammatory and oxidant cardiac states and decreased ROS
ity, under pathological conditions [37], including steatohepatitis production and CaMKII phosphorylation level (Fig. 4C–D).
[38]. LC has the ability to scavenge free radicals [39,40] and Fibrosis level decreased in the cardiac tissues of mice fed with
protect hepatocytes against oxidative damages caused under dif- MCDD + LC, as evident from ␣SMA staining result (Fig. 5D). Fibrosis
ferent conditions including ethanol intoxication [41], obesity [42], was also witnessed from the activation of STAT3 and depended on
and carcinoma [43]. Here, we observed that LC counteracts COX ERK phosphorylation [58]. STAT3 is activated in response to sev-
imbalance caused by MCDD (Fig. 2E). Romestaing et al. [32] have eral cardiac insults, such as oxidative damage, hypertrophy, and
demonstrated the increase in COX expression mediated by MCDD remodeling [59,16]. The ERKs/STAT3 pathway was inhibited in the
that alters mitochondrial metabolism, as demonstrated with ele- LC-supplemented heart tissues (Fig. 5A, B), supporting the hypoth-
vated ROS level. In addition, other studies have described the esis that LC could prevent cardiac damages observed in NAFLD.
capacity of LC to attenuate ROS accumulation and cardiac dys-
functions in ischemic or irradiated hearts [44,45] and to decrease 5. Conclusion
oxidative stress in normal and hyperglycemic cardiomyocytes
[46,21]. The present study confirmed that LC attenuates ROS pro- Our results demonstrate that LC supplementation could miti-
duction in the liver (Fig. 2D) and heart (Fig. 4C) of mice fed with gate MCDD-induced steatosis and fibrosis in mice and ameliorate
MCDD. Taken together, our data suggest that LC prevents NAFLD the mechanisms underlying NASH development. LC could act on
progression and consequently heart damage. both the “two hits” involved in NAFLD progression. It could regulate
The activation of hepatic stellate cells and the conversion of hepatic lipid accumulation and oxidative stress, the key phenom-
cardiac fibroblasts into active myofibroblasts are indicative of the ena involved in fibrosis. Thus, LC supplementation could serve as an
increase in the expression of ␣SMA in both hepatic and cardiac effective nutraceutical adjuvant for the prevention and treatment
tissues [47,48,5]. LC supplementation was found to reduce ␣SMA of NASH.
expression both in the liver (Fig. 3B) and heart (Fig. 5D), sugges- Our data suggest that LC may mitigate heart damage associ-
tive of its inhibitory effect against fibrogenic ␣SMA-producing cells. ated with NAFLD. Future studies should evaluate the applicability
Furthermore, at the hepatic level, the anti-fibrotic effect of LC was of long-term LC supplementation to control the pathophysiologic
confirmed from the level of caspase-2 protein, which is thought to evolution of this disease in humans.
be a marker of NASH [6,8]. Caspase-2 level increases based on the
stages of fibrosis both in animals and humans. Caspase-2-deficient
mice did not develop NASH even after being fed with MCDD [7]. Sources
As shown in Fig. 3C, the levels of pro-caspase-2 and caspase-2
were similar between controls and LC-fed mice, and MCDD signif- This study was partially supported by Ricerca Corrente funding
icantly elevated caspase-2 protein expression, thereby confirming from Italian Ministry of Health to IRCCS Policlinico San Donato.
the anti-fibrotic action of LC.
Fibrosis is a reversible process, and fibrotic tissues may be spon- Conflicts of interest
taneously resorbed upon retraction of the injurious stimulus [49]. None declared.
PPAR␥, mainly known for its central role in driving adipogenesis
and lipid metabolism [50], was proposed to perform a putative References
function in the reversal of liver fibrosis. In fact, the level of in vivo
fibrogenesis was found to be decreased in PPAR␥-depleted rat liv- [1] Brunt EM. Pathology of nonalcoholic fatty liver disease. Nat Rev Gastroenterol
ers and increased in PPAR␥-overexpressing rats fed with MCDD Hepatol 2010;7(4):195–203.
[2] Day CP, James OF. Steatohepatitis: a tale of two “hits”? Gastroenterology
[51]. Several studies with different nutraceuticals have revealed 1998;114(4):842–5.
the increase in hepatic fibrosis upon PPAR␥ upregulation [52]. The [3] Valenti L, Bugianesi E, Pajvani U, et al. Nonalcoholic fatty liver disease: cause
protective role of PPAR␥ against liver fibrosis has been widely or consequence of type 2 diabetes? Liver Int 2016;36(11):1563–79.
[4] Rotman Y, Sanyal AJ. Current and upcoming pharmacotherapy for non-
demonstrated [11,12]. In particular, Chen et al. [53], used MCDD-
alcoholic fatty liver disease. Gut 2017;66(1):180–90.
fed mice and found that although MCDD was unable to modify [5] Friedman SL. Mechanisms of hepatic fibrogenesis. Gastroenterology
PPARy level, an extract of Schisandra sphenanthera (medicinal herb) 2008;134(6):1655–69.
[6] Machado MV, Kruger L, Jewell ML, et al. Vitamin B5 and N-acetylcysteine in
could counteract NASH progression by increasing PPARy expres-
nonalcoholic steatohepatitis: a preclinical study in a dietary mouse model. Dig
sion. Similar to this study, our results showed that the progression Dis Sci 2016;61(1):137–48.
of hepatic fibrosis was affected upon LC administration and asso- [7] Machado MV, Michelotti GA, Pereira Tde A, et al. Reduced lipoapoptosis,
ciated with a significant increase in PPAR␥ level (Fig. 3D). Further hedgehog pathway activation and fibrosis in caspase-2 deficient mice with
non-alcoholic steatohepatitis. Gut 2015;64(7):1148–57.
studies are warranted to clarify the effect of PPAR␥ on NAFLD and [8] Kim JY, Garcia-Carbonell R, Yamachika S, et al. ER stress drives lipogenesis and
NASH progression. steatohepatitis via caspase-2 activation of S1P. Cell 2018;175(1), 133–145.e15.
PPAR-␥ activation blocks the NF␬B pathway through the repres- [9] Leclercq IA, Farrell GC, Sempoux C, et al. Curcumin inhibits NF-kappaB activa-
tion and reduces the severity of experimental steatohepatitis in mice. J Hepatol
sion of the translocation of p65 to the nucleus [54]. NF␬B is 2004;41(6):926–34.
known to control liver oxidative stress [9,10] and the transcrip- [10] Salamone F, Galvano F, Cappello F, et al. Silibinin modulates lipid homeostasis
tion of pro-inflammatory cytokines responsible for the activation and inhibits nuclear factor kappa B activation in experimental nonalcoholic
steatohepatitis. Transl Res 2012;159(6):477–86.
of hepatic stellate cells, collagen deposition, and fibrogenesis [55], [11] Xu J, Fu Y, Chen A. Activation of peroxisome proliferator-activated receptor-
thereby contributing to the progression of NAFLD and liver fibrosis gamma contributes to the inhibitory effects of curcumin on rat hepatic stellate
[56,57]. We found that NF␬B p65 protein level increased in MCDD cell growth. Am J Physiol Gastrointest Liver Physiol 2003;285(1):G20–30.
 G. Mollica et al. / Digestive and Liver Disease 52 (2020) 314–323 323

[12] Hazra S, Xiong S, Wang J, et al. Peroxisome proliferator-activated receptor [37] Benedini S, Perseghin G, Terruzzi I, et al. Effect of L-acetylcarnitine on body com-
gamma induces a phenotypic switch from activated to quiescent hepatic stel- position in HIV-related lipodystrophy. Horm Metab Res 2009;41(11):840–5.
late cells. J Biol Chem 2004;279(12):11392–401. [38] Jun DW, Cho WK, Jun JH, et al. Prevention of free fatty acid-induced hepatic
[13] Byrne CD, Targher G. NAFLD: a multisystem disease. J Hepatol 2015;62(1 lipotoxicity by carnitine via reversal of mitochondrial dysfunction. Liver Int
Suppl):S47–64. 2011;31(9):1315–24.
[14] Long MT, Fox CS. The framingham heart study—67 years of discovery in [39] Montesano A, Senesi P, Luzi L, et al. Potential therapeutic role of L-carnitine
metabolic disease. Nat Rev Endocrinol 2016;12(3):177–83. in skeletal muscle oxidative stress and atrophy conditions. Oxid Med Cell
[15] MacLean J, Pasumarthi KB. Signaling mechanisms regulating fibroblast acti- Longevity 2015;2015:646171.
vation, phenoconversion and fibrosis in the heart. Indian J Biochem Biophys [40] Terruzzi I, Montesano A, Senesi P, et al. L-carnitine reduces oxidative stress and
2014;51(6):476–82. promotes cells differentiation and bone matrix proteins expression in human
[16] Haghikia A, Ricke-Hoch M, Stapel B, et al. STAT3, a key regulator of cell-to-cell osteoblast-like cells. Biomed Res Int 2019;2019:5678548.
communication in the heart. Cardiovasc Res 2014;102(2):281–9. [41] Dobrzyńska I, Szachowicz-Petelska B, Skrzydlewska E, et al. Effect of L-carnitine
[17] Hung CK, Bodenheimer Jr HC. Current treatment of nonalcoholic fatty liver on liver cell membranes in ethanol-intoxicated rats. Chem Biol Interact
disease/nonalcoholic steatohepatitis. Clin Liver Dis 2018;22(1):175–87. 2010;188(1):44–51.
[18] Ishikawa H, Takaki A, Tsuzaki R, et al. L-carnitine prevents progression of non- [42] Wu T, Guo A, Shu Q, et al. L-carnitine intake prevents irregular feeding-induced
alcoholic steatohepatitis in a mouse model with upregulation of mitochondrial obesity and lipid metabolism disorder. Gene 2015;554(2):148–54.
pathway. PLoS One 2014;9(7):e100627. [43] Jiang F, Zhang Z, Zhang Y, et al. L-carnitine ameliorates the liver inflamma-
[19] Li W, Ma F, Zhang L, et al. S-propargyl-cysteine exerts a novel protective effect tory response by regulating carnitine palmitoyltransferase I-dependent PPAR␥
on methionine and choline deficient diet-induced fatty liver via akt/Nrf2/HO-1 signaling. Mol Med Rep 2016;13(2):1320–8.
pathway. Oxid Med Cell Longev 2016;2016:4690857. [44] Xue M, Chen X, Guo Z, et al. L-carnitine attenuates cardiac dysfunction by
[20] Ferraretto A, Bottani M, Villa I, et al. L-carnitine activates calcium signaling in ischemic insults through Akt signaling pathway. Toxicol Sci 2017;160(2):
human osteoblasts. J Funct Foods 2018;47:270–8. 341–50.
[21] Vacante F, Senesi P, Montesano A, et al. L-carnitine: an antioxidant remedy for [45] Fan Z, Han Y, Ye Y, et al. L-carnitine preserves cardiac function by activating
the survival of cardiomyocytes under hyperglycemic condition. J Diabetes Res p38 MAPK/Nrf2 signalling in hearts exposed to irradiation. Eur J Pharmacol
2018;2018:4028297. 2017;804:7–12.
[22] Strilakou AA, Lazaris AC, Perelas AI, et al. Heart dysfunction induced by choline- [46] Mao CY, Lu HB, Kong N, Li, et al. Levocarnitine protects H9c2 rat cardiomyocytes
deficiency in adult rats: the protective role of L-carnitine. Eur J Pharmacol from H2O2-induced mitochondrial dysfunction and apoptosis. Int J Med Sci
2013;709(1-3):20–7. 2014;11(11):1107–15.
[23] Krishna M. Role of special stains in diagnostic liver pathology. Clin Liver Dis [47] Wang J, Fan J, Laschinger C. Smooth muscle actin determines mechanical force-
2013;2(Suppl 1):S8–10. induced p38 activation. J Biol Chem 2005;280(8):7273–84.
[24] Clark I, Torbenson MS. Immunohistochemistry and special stains in medical [48] Friedman SL. Fibrogenic cell reversion underlies fibrosis regression in liver. Proc
liver pathology. Adv Anat Pathol 2017;24(2):99–109. Natl Acad Sci U S A 2012;109(24):9230–1.
[25] Senesi P, Montesano A, Luzi L, et al. Metformin treatment prevents sedentari- [49] Panebianco C, Oben JA, Vinciguerra M, et al. Senescence in hepatic stellate cells
ness related damages in mice. J Diabetes Res 2016;2016:8274689. as a mechanism of liver fibrosis reversal: a putative synergy between retinoic
[26] Codella R, Lanzoni G, Zoso A, et al. Moderate intensity training impact on acid and PPAR-gamma signalings. Clin Exp Med 2016;17(3):269–80.
the inflammatory status and glycemic profiles in NOD mice. J Diabetes Res [50] Wang Z, Xu JP, Zheng YC, et al. Peroxisome proliferator-activated recep-
2015;2015:737586. tor gamma inhibits hepatic fibrosis in rats. Hepatobiliary Pancreat Dis Int
[27] Terruzzi I, Montesano A, Senesi P, et al. Erratum to: ranolazine promotes mus- 2011;10(1):64–71.
cle differentiation and reduces oxidative stress in C2C12 skeletal muscle cells. [51] Wu CW, Chu ESH, Lam CNY, et al. PPAR␥ is essential for protection against
Endocrine 2017;58(1):46. nonalcoholic steatohepatitis. Gene Ther 2010;17:790–8.
[28] Hebbard L, George J. Animal models of nonalcoholic fatty liver disease. Nat Rev [52] Choi JH, Jin SW, Choi CY, et al. Capsaicin inhibits dimethylnitrosamine-
Gastroenterol Hepatol 2011;8(1):35–44. induced hepatic fibrosis by inhibiting the TGF-␤1/Smad pathway via
[29] Gao D, Wei C, Chen L, et al. Oxidative DNA damage and DNA repair enzyme peroxisome proliferator-activated receptor gamma activation. J Agric Food
expression are inversely related in murine models of fatty liver disease. Am J Chem 2017;65(2):317–26.
Physiol Gastrointest Liver Physiol 2004;287(5):G1070–7. [53] Chen Z, Liu F, Zheng N, et al. Wuzhi capsule (Schisandra sphenanthera extract)
[30] Kirsch R, Clarkson V, Shephard EG, et al. Rodent nutritional model of attenuates liver steatosis and inflammation during non-alcoholic fatty liver
non-alcoholic steatohepatitis: species, strain and sex difference studies. J Gas- disease development. Biomed Pharmacother 2019;110:285–93.
troenterol Hepatol 2003;18(11):1272–82. [54] Feng X, Weng D, Zhou F, et al. Activation of PPAR␥ by a natural flavonoid
[31] Brunt EM, Janney CG, Di Bisceglie AM, et al. Nonalcoholic steatohepatitis: a modulator, apigenin ameliorates obesity-related inflammation via regulation
proposal for grading and staging the histological lesions. Am J Gastroenterol of macrophage polarization. EBioMedicine 2016;9:61–76.
1999;94:2467–74. [55] Wobser H, Dorn C, Weiss TS, et al. Lipid accumulation in hepatocytes induces
[32] Romestaing C, Piquet MA, Letexier D, et al. Mitochondrial adaptations to steato- fibrogenic activation of hepatic stellate cells. Cell Res 2009;19(8):996–1005.
hepatitis induced by a methionine- and choline-deficient diet. Am J Physiol [56] Luedde T, Schwabe RF. NF-␬B in the liver—linking injury, fibrosis and hepato-
Endocrinol Metab 2008;294(1):E110–9. cellular carcinoma. Nat Rev Gastroenterol Hepatol 2011;8(2):108–18.
[33] Anderson ME. Oxidant stress promotes disease by activating CaMKII. J Mol Cell [57] Tian Y, Ma J, Wang W, et al. Resveratrol supplement inhibited the NF-␬B inflam-
Cardiol 2015;89(Pt B):160–7. mation pathway through activating AMPK␣-SIRT1 pathway in mice with fatty
[34] Del Ben M, Polimeni L, Baratta F, et al. The role of nutraceuticals for the treat- liver. Mol Cell Biochem 2016;422(1–2):75–84.
ment of non-alcoholic fatty liver disease. Br J Clin Pharmacol 2017;83(1):88–95. [58] Targher G, Day CP, Bonora E. Risk of cardiovascular disease in patients with
[35] Itoh M, Kato H, Suganami T, et al. Hepatic crown-like structure: a unique histo- nonalcoholic fatty liver disease. N Engl J Med 2010;363(14):1341–50.
logical feature in non-alcoholic steatohepatitis in mice and humans. PLoS One [59] Dai B, Cui M, Zhu M, et al. STAT1/3 and ERK1/2 synergistically regulate cardiac
2013;8(12):e82163. fibrosis induced by high glucose. Cell Physiol Biochem 2013;32(4):960–71.
[36] Wei Y, Rector RS, Thyfault JP, et al. Nonalcoholic fatty liver disease and mito-
chondrial dysfunction. World J Gastroenterol 2008;14(2):193–9.