Content uploaded by Chenju Yi
Author content
All content in this area was uploaded by Chenju Yi on Mar 22, 2021
Content may be subject to copyright.
Targeting Nestin
+
hepatic stellate cells ameliorates liver
fibrosis by facilitating TbRI degradation
Graphical abstract
Activated HSCs Quiescent HSCs
Smad2/3
SBE Nestin
Smad2/3
degradation
TβRI
degradation
TβRI
Ligand
Caveolin1
Nestin
P
TβRII
TβRI
Increased Nestin expression Decreased Nestin expression
Nestin
SBE
TSS TSS
Smad binding elemnt, SBE Transcription start site, TSS
Highlights
Nestin was upregulated in activated HSCs with the progression of
liver fibrosis in both patient specimens and mouse models.
Activated TGFb-Smad2/3 signalling induced Nestin expression.
Knocking down Nestin hampered TGFb-Smad2/3 signalling and
fibrosis marker expression in HSCs by promoting Caveolin1-
mediated TbRI degradation.
Targeting Nestin using AAV6 alleviated liver fibrosis in mouse
models.
Authors
Huaxin Chen, Jianye Cai,
Jiancheng Wang, .,YanXu,Andy
Peng Xiang, Qi Zhang
Correspondence
yysysu@163.com (Y. Yang),
xuyan55@mail.sysu.edu.cn (Y. Xu),
xiangp@mail.sysu.edu.cn (A.P. -
Xiang), zhangq27@mail.sysu.edu.cn
(Q. Zhang).
Lay summary
Liver fibrosis has various aetiologies
but represents a common process
in chronic liver diseases that is
associated with high morbidity and
mortality. Herein, we demonstrate
that the intermediate filament
protein Nestin plays an essential
profibrogenic role in liver fibrosis
by forming a positive feedback loop
with the TGFb-Smad2/3 pathway,
providing a potential therapeutic
target for the treatment of liver
fibrosis.
https://doi.org/10.1016/j.jhep.2020.11.016
© 2020 European Association for the Study of the Liver. Published by Elsevier B.V. All rights reserved. J. Hepatol. 2021, -,1–12
Research Article
Experimental and Translational Hepatology
Targeting Nestin
+
hepatic stellate cells ameliorates liver fibrosis by
facilitating TbRI degradation
Huaxin Chen
1,2,†
, Jianye Cai
3,4,†
, Jiancheng Wang
3,6,†
, Yuan Qiu
3
, Chenhao Jiang
3
, Yi Wang
3
,
Yiqin Wang
3
, Chenju Yi
6
, Guo lv
4
, Lijie Pan
1,2
, Yuanjun Guan
7
, Jun Zheng
4
, Dongbo Qiu
1,2,5
,
Cong Du
1,2,5
, Qiuli Liu
1
, Guihua Chen
4
, Yang Yang
4,
*,YanXu
1,
*, Andy Peng Xiang
3,
*,
Qi Zhang
1,2,5,
*
1
Biotherapy Centre, The Third Affiliated Hospital, Sun Yat-sen University, Guangzhou, China;
2
Cell-gene Therapy Translational Medicine
Research Centre, The Third Affiliated Hospital, Sun Yat-sen University, Guangzhou, China;
3
Centre for Stem Cell Biology and Tissue
Engineering, Key Laboratory for Stem Cells and Tissue Engineering, Ministry of Education, Sun Yat-sen University, Guangzhou, China;
4
Department of Hepatic Surgery and Liver Transplantation Centre, The Third Affiliated Hospital, Sun Yat-sen University, Guangzhou, China;
5
Guangdong Provincial Key Laboratory of Liver Disease Research, The Third Affiliated Hospital, Sun Yat-sen University, Guangzhou, China;
6
Scientific Research Centre, The Seventh Affiliated Hospital, Sun Yat-sen University, Shenzhen, China;
7
Core Facility Centre, Zhongshan School
of Medicine, Sun Yat-sen University, Guangzhou, China
Background & Aims: Liver fibrosis is a wound healing response
that arises from various aetiologies. The intermediate filament
protein Nestin has been reported to participate in maintaining
tissue homeostasis during wound healing responses. However,
little is known about the role Nestin plays in liver fibrosis. This
study investigated the function and precise regulatory network
of Nestin during liver fibrosis.
Methods: Nestin expression was assessed via immunostaining
and quantitative real-time PCR (qPCR) in fibrotic/cirrhotic sam-
ples. The induction of Nestin expression by transforming growth
factor beta (TGFb)-Smad2/3 signalling was investigated through
luciferase reporter assays. The functional role of Nestin in hepatic
stellate cells (HSCs) was investigated by examining the pathway
activity of profibrogenic TGFb-Smad2/3 signalling and degrada-
tion of TGFbreceptor I (TbRI) after interfering with Nestin. The
in vivo effects of knocking down Nestin were examined with an
adeno-associated virus vector (serotype 6, AAV6) carrying short-
hairpin RNA targeting Nestin in fibrotic mouse models.
Results: Nestin was mainly expressed in activated HSCs and
increased with the progression of liver fibrosis. The profibrogenic
pathway TGFb-Smad2/3 induced Nestin expression directly.
Knocking down Nestin promoted caveolin 1-mediated TbRI
degradation, resulting in TGFb-Smad2/3 pathway impairment
and reduced fibrosis marker expression in HSCs. In AAV6-treated
murine fibrotic models, knocking down Nestin resulted in
decreased levels of inflammatory infiltration, hepatocellular
damage, and a reduced degree of fibrosis.
Conclusion: The expression of Nestin in HSCs was induced by
TGFband positively correlated with the degree of liver fibrosis.
Knockdown of Nestin decreased activation of the TGFbpathway
and alleviated liver fibrosis both in vitro and in vivo. Our data
demonstrate a novel role of Nestin in controlling HSC activation
in liver fibrosis.
Lay summary: Liver fibrosis has various aetiologies but repre-
sents a common process in chronic liver diseases that is associ-
ated with high morbidity and mortality. Herein, we demonstrate
that the intermediate filament protein Nestin plays an essential
profibrogenic role in liver fibrosis by forming a positive feedback
loop with the TGFb-Smad2/3 pathway, providing a potential
therapeutic target for the treatment of liver fibrosis.
© 2020 European Association for the Study of the Liver. Published by
Elsevier B.V. All rights reserved.
Introduction
Liver fibrosis is a wound healing process that develops in
response to various aetiologies, including hepatitis virus infec-
tion, alcohol-related liver disease (ALD), cholestatic disorders,
and non-alcoholic steatohepatitis.
1
It is characterised by exces-
sive deposition of extracellular matrix mainly secreted by he-
patic stellate cells (HSCs).
2
In a fibrotic microenvironment,
extracellular signals –including soluble cytokines and chemo-
kines –promote the activation of HSCs and their trans-
differentiation into myofibroblasts. Additionally, activated HSCs
secrete proinflammatory and profibrogenic factors; these include
transforming growth factor-b(TGFb), which plays a pivotal role
in the progression of fibrosis.
3
The positive feedback between
HSCs and their microenvironment contributes to the persistent
activation of HSCs. This deregulates the self-sustaining process
and, consequently, results in irreversible liver damage.
3
Given
the mechanisms involved, targeting and modulating activated
HSCs during liver fibrosis could be a promising antifibrotic
strategy.
Keywords: Liver fibrosis; Nestin; Cav-1; TbRI degradation; AAV6.
Received 16 January 2020; received in revised form 2 November 2020; accepted 12
November 2020; available online xxx
*Corresponding authors. Addresses: Biotherapy Centre, Third Affiliated Hospital of
Sun Yat-sen University, 600# Tianhe Road, Guangzhou, 510630, China. Tel.: 86-20-
85253106, Fax: 86-20-85253305, (Q. Zhang), or Key Laboratory for Stem Cells and
Tissue Engineering, Ministry of Education, Sun Yat-sen University, 74# Zhongshan
2nd Road, Guangzhou, 510080, China. Tel.: 86-20-87335822, Fax: 86-20-87335858,
(A.P. Xiang), or Biotherapy Centre, Third Affiliated Hospital of Sun Yat-sen University,
600# Tianhe Road, Guangzhou, 510630, China. Tel.: 86-20-82179071, Fax: 86-20-
85253305, (Y. Xu), or Department of Hepatic Surgery and Liver Transplantation
Centre, Third Affiliated Hospital of Sun Yat-sen University, 600# Tianhe Road,
Guangzhou, 510630, China. Tel.: 86-20-85252276, Fax: 86-20-85252276, (Y. Yang).
E-mail addresses: yysysu@163.com (Y. Yang), xuyan55@mail.sysu.edu.cn (Y. Xu),
xiangp@mail.sysu.edu.cn (A.P. Xiang), zhangq27@mail.sysu.edu.cn (Q. Zhang).
†
Huaxin Chen, Jianye Cai, and Jiancheng Wang contributed equally to this study.
https://doi.org/10.1016/j.jhep.2020.11.016
Journal of Hepatology 2021 vol. -j1–12
Research Article
Experimental and Translational Hepatology
Nestin is a class VI intermediate filament that is primarily
found in the central nervous system.
4
Recently, it has been re-
ported that Nestin participates in the maintenance of tissue
homeostasis during wound healing responses. For instance,
Nestin is induced in reactive astrocytes that contribute to glial
scar formation in traumatic central nervous system injuries.
5
Another study showed that the expression of Nestin correlates
with the degree of tubulointerstitial fibrosis.
6
Nestin is barely
expressed in normal healthy adult liver but is upregulated upon
either acute or chronic liver injuries.
7–10
This indicates the po-
tential role of Nestin in liver regeneration. Although the role of
Nestin has been extensively studied in malignant settings in the
liver,
8–10
little is known about the function and precise regula-
tion of Nestin during liver fibrosis.
During the progression of liver fibrosis, the profibrotic TGFb-
Smad2/3 pathway is induced in the activated HSCs. This results
in upregulated expression of profibrogenic genes.
3
In the pro-
totypic TGFb-Smad2/3 pathway, ligands induce the assembly of
the TGFbreceptor I (TbRI) and TGFbreceptor II (TbRII) hetero-
meric complex and lead to Smad2/3 phosphorylation and nu-
clear translocation. Consequently, phosphorylated Smad2/3
binds to specific Smad-binding elements (SBEs) in the gene
promoter and activates/represses target genes. TbRs link extra-
cellular stimuli to intracellular responses, so their distribution
and stability are critical for TGFbsignal transduction.
11
For
instance, receptor levels on the cell surface increase in response
to TGFbstimulation, which subsequently enhances TGFbsignal-
ling.
12
In contrast, the degradation of TbRs mediated by ubiq-
uitination decreases TbR stability, resulting in the attenuation of
TGFbresponsiveness.
Our study sought to define the role of Nestin in HSCs during
liver fibrosis. We found that Nestin expression was positively
correlated with the progression of liver fibrosis in both patient
specimens and animal models. Mechanistically, the TGFb-
Smad2/3 pathway induced Nestin expression during HSC acti-
vation. Knocking down Nestin in HSCs promoted caveolin 1 (Cav-
1)−mediated TbRI degradation and downregulated TGFbsignal-
ling. Notably, we demonstrated that in vivo treatment with an
adeno-associated virus vector (serotype 6, AAV6) carrying a
short-hairpin (sh)RNA targeting Nestin (AAV6-shNestin) in HSCs
alleviated liver fibrosis in mouse models. Together, these findings
highlight a previously undiscovered profibrotic role of Nestin in
liver fibrosis and provide insights into the molecular events
underpinning fibrogenesis.
Materials and methods
Animals and liver fibrosis models
Eight- to 14-week male C57BL/6 mice (purchased from the
Nanjing University Model Animals Institute) and Nestin-GFP
mice (expressing Nestin promoter-driven GFP) in the C57BL/6
background, kindly gifted from Dr Masahrio Yamaguchi) were
used in this study. They were provided with free access to food
and water at the Sun Yat-sen University Animal Centre. Liver
fibrosis was induced by intraperitoneal injection of 20% carbon
tetrachloride (CCl
4
; Sigma-Aldrich, 289116) in corn oil (Sigma-
Aldrich, 23-0230) at 5
l
L$g
−1
of body weight, twice per week, or
a diet containing 0.1% 3,5-diethoxycarbonyl-1,4-dihydrocollidine
(DDC; Sigma-Aldrich, 137030) for an indicated time. In the AAV6-
shRNA-treated liver fibrosis model, mice were pre-treated with
CCl
4
/DDC administration for 2 weeks. Following this, they were
administered with 100
l
l of AAV6-shControl (non-targeting
control) or AAV6-shNestin (1.5 × 10
12
viral genomes$ml
−1
; Hanbio
Biotechnology, Shanghai, China) through tail vein injection. After
an additional 4 weeks’administration of CCl
4
/DDC, the livers
were harvested and analysed. All procedures were conducted
under the Sun Yat-sen University Institutional Animal Care and
Use Committee guidelines.
Isolation of non-parenchymal liver cells from mouse liver
Mouse primary HSCs were isolated from mouse livers using a
previously reported protocol.
13
Briefly, livers were digested with
retrograde stepwise perfusion with solutions containing pro-
nase (Sigma-Aldrich, 11459643001) and collagenase (Sigma-
Aldrich, 11213865001). The cell suspension was centrifuged at
50 g and supernatants containing non-parenchymal hepatic
cells were collected. HSCs were purified from non-parenchymal
hepatic cells with density gradient centrifugation and cultured
in DMEM containing 10% foetal bovine serum (FBS; GIBCO,
10099).
Patient specimens
One hundred and fourteen human liver fibrosis/cirrhosis spec-
imens were obtained from patients undergoing transplantation
based on the University of California, San Francisco (UCSF) or
Hangzhou criteria (for patients with hepatocellular carcinoma,
HCC) or Child-Pugh classification of grade B and C with
decompensated symptoms (such as ascites or hepatic enceph-
alopathy) (for patients with end-stage benign liver disease) in
the Third Affiliated Hospital of Sun Yat-sen University. These
included 67 individuals with HBV, 12 with HCV, 13 with ALD, 14
with biliary atresia (BA) and 8 with primary biliary cholangitis
(PBC). Patients with evidence of other liver diseases (including
Wilson's disease, alpha-1 antitrypsin deficiency, or other
inherited liver diseases) based on standard clinical, laboratory
and histological assessments were excluded in this study. Ten
normal liver specimens were obtained from donation after
cardiac death (DCD) donors during liver transplantations.
Informed consent was obtained from all individuals and recor-
ded in the electronic database. Research Board protocols were
conducted under the guidelines set by the medical ethical
committees of the Third Affiliated Hospital of Sun Yat-sen Uni-
versity (#2018-020182-01). Histological scoring was performed
by experienced pathologists according to the Ishak score
criteria.
14
Basic information of the patients examined in this
study is summarised in Table S1.
Statistical analysis
Results are expressed as the mean ± SEM. The Student’sttest was
used to compare 2 groups of data. One-way analysis of variance
followed by Dunnett’s test was used to compare data with 3 or
more groups. p<0.05 was considered statistically significant.
Results
Nestin expression levels correlate with the progression of
liver fibrosis
To determine whether Nestin was associated with liver fibrosis,
we used 2 well-established liver fibrosis models, the CCl
4
and
DDC models.
2
Fibrotic progression was determined by Picrosirius
red staining and
a
-SMA immunohistochemistry staining in liver
tissues collected at different time points (Fig. 1A, B). Notably,
fibrotic liver samples showed much more Nestin expression than
normal liver samples, and higher levels of Nestin were
2 Journal of Hepatology 2021 vol. -j1–12
Research Article Experimental and Translational Hepatology
profoundly associated with more advanced liver fibrosis (Fig. 1A,
B). Furthermore, quantitative real-time PCR (qPCR) analysis of
liver homogenate demonstrated that the enhancement of Nestin
expression was consistent with an increase of fibrotic genes
(including Ctgf, Timp1,Mmp9,Acta2 and Col1a1) upon progres-
sion of liver fibrosis (Fig. 1C). Next, we used a Nestin-GFP reporter
mouse model
15
to monitor Nestin expression during liver
fibrosis. Two-photon fluorescence analysis showed that the GFP
fluorescence intensity of the liver increased when fibrosis was
aggravated in both CCl
4
and DDC models (Fig. S1A).
Next, we compared human fibrotic/cirrhotic hepatic speci-
mens with those of healthy donors. Both mRNA and protein
levels of NESTIN were significantly upregulated in fibrotic/
cirrhotic livers from patients with HBV/HCV infection, ALD, BA
and PBC (Fig. 1D, E). According to the semi-quantitative scoring
(Ishak score) of virus-associated fibrosis/cirrhosis specimens,
14
we found a positive correlation between Nestin expression and
the degree of liver fibrosis (Fig. 1F). Furthermore, increased
NESTIN expression was observed in progressive stages of HBV-
induced liver fibrosis (Fig. S1B−D). It was also positively corre-
lated with the expression of profibrotic genes (Fig. S1E) and in-
flammatory scores (Fig. S1F). Taken together, these results
suggested that Nestin expression was positively correlated with
the severity of liver fibrosis.
C
D
E
HCVHBV
Normal
ALD BA PBC
F
A
PSR
OilCCl4 4w CCl4 2w
α-SMA Nestin
***
***
PSR
positive area (%)
2
4
6
8
0
***
***
Nestin positive
area (%)
5
10
15
0
Oil
CCl
4
2w
CCl
4
4w
B
DDC 4w DDC 1w
Normal diet
PSR α-SMA
Nestin
***
***
5
10
15
20
PSR
positive area (%)
0
***
***
2
4
6
Nestin positive
area (%)
0
Normal diet
DDC 1w
DDC 4w
Nestin Ctgf Timp1 Mmp9 Acta2 Col1a1
**** ***
***
***
**
***
***
***
***
***
*** ***
***
***
**
***
***
***
***
***
*** ***
***
0
5
10
15
20
25
30
Relative mRNA expression
(normalised to 18s)
HBV
R = 0.5981
p <0.0001
Ishak score
NESTIN IHC score
1.0
0.2
0.0
0.4
0.6
0.8
HCV
R = 0.8495
p = 0.0025
Ishak score
1.0
0.2
0.0
0.4
0.6
0.8
Log2 (fold change)
0 5.02 2.06 2.26 3.97 1.68
HBV
Normal
HCV ALD BA PBC
(n = 67)
(n = 10)
(n = 12) (n = 13) (n = 14) (n = 8)
***
*
***
***
**
-5
0
5
10
15
Relative mRNA expression of NESTIN
log2 (fibrosis/normal liver)
Normal diet
Oil
DDC 1w
CCl4 2w
DDC 4w
CCl4 4w
0123401
2
34
Fig. 1. Nestin expression levels correlate with the progression of liver fibrosis. (A&B) Left: representative photographs of PSR staining, IHC for
a
-SMA, IHC for
Nestin and IF for Nestin (green) in indicated groups of (A) CCl
4
- and (B) DDC-induced fibrosis models. Right: quantification of PSR staining (upper) and Nestin IHC
(lower) in (Left). n = 6/group. Scale bars for PSR staining and IHC photographs, 200
l
m. Scale bars for IF photographs, 50
l
m. Nuclei were counterstained with
DAPI in IF (also hereafter in similar experiments). Magnified images of the boxed areawere shown (also hereafter in similar experiments). (C) qPCR for Nestin and
fibrotic genes of control and fibrotic liver tissues of CCl
4
and DDC models collected at different time points. The values were normalised based on 18s values and
referred to as oil control (for the CCl
4
model) or normal diet control (for the DDC model) (and hereafter for similar experiments). n = 6/group. (D) Log
2
-fold change
of NESTIN mRNA levels in human healthy control and fibrotic livers of various aetiologies (HBV, HCV, ALD, BA and PBC) were analysed by qPCR. Data were
normalised based on GAPDH values and compared to the healthy control (Normal). (E) Representative photographs of IHC for NESTIN and PSR staining of human
healthy control and fibrotic livers of various aetiologies. Scale bar for IHC staining, 50
l
m. Scale bar for PSR staining, 100
l
m. (F) Pearson correlation analysis of
Ishak scores with NESTIN IHC scores in HBV- or HCV-induced human fibrotic/cirrhotic livers. Data in (A-C) are reported as the mean ± SEM with indicated
significance (*p<0.05, **p<0.01, ***p<0.0 01, n.s.: not significant; Student’sttest). ALD, alcohol-related liver disease; BA, biliary atresia; CCl
4
, carbon tetrachloride;
DDC, 3,5-diethoxycarbonyl-1,4-dihydrocollidine; IF, immunofluorescence; IHC, immunohistochemistry; PBC, primary biliary cholangitis; PSR, Picosirius red;
qPCR, quantitative real-time PCR.
Journal of Hepatology 2021 vol. -j1–12 3
Increased Nestin expression in HSCs after liver injury
In fibrotic tissues, we noticed that Nestin was mainly localised in
fibrotic septa (Fig. 1A, B, E, and Fig. S1A, B). This suggested that
Nestin might be upregulated in myofibroblasts activated in
fibrosis. To characterise the main cell types expressing Nestin in
fibrosis, we co-stained Nestin with specific hepatic parenchymal
cell markers: Albumin for hepatocytes and cytokeratin 19 for bile
duct epithelium; and non-parenchymal cell markers: CD31 for
liver sinusoidal endothelial cells (LSECs), F4/80 for Kupffer cells
(KCs), Desmin for HSCs, CD90 (Thy1) or Fibulin2 for portal fi-
broblasts and
a
-SMA for activated myofibroblasts in mouse
fibrotic liver tissues. We found that Nestin expression was largely
co-localised with
a
-SMA and Desmin, suggesting that Nestin was
mainly increased in activated HSCs (Fig. 2A, Fig. S2A). Similar
findings were observed in patient fibrotic/cirrhotic liver tissues
(Fig. 2B, Fig. S2B).
Next, we used Nestin-GFP murine models to further delineate
Nestin expression in hepatic non-parenchymal cells. Previous re-
ports have demonstrated that HSCs contain vitamin A (Vit.A)
droplets and can be detectedby flow cytometry through their auto-
fluorescent violet signals.
13,16
A selective increase of Nestin
expression was observed in Vit.A
+
HSCs after CCl
4
or DDC admin-
istration (Fig. 2C−E). To quantify the contribution of Nestin
+
HSCs
during liver fibrosis, we used Nestin-GFP mice to trace the fate of
these cells (Fig. S2C). FACS results indicated that both CCl
4
-and
DDC-treated Nestin-GFP mice had increased numbers of HSCs
(Vit.A
+
), and that around half of the HSCs were Nestin-GFP
+
(Fig. S2C, D). Furthermore, the GFP
+
Vit.A
+
cells showed higher
mRNA levels of fibrogenic genes than GFP
−
Vit.A
+
populations
(Fig. S2E, F). This suggested that Nestin could serve as a functional
marker of activated HSCs. The GFP
+
Vit.A
+
cells in fibrotic liver also
showed a typicalphenotype of activated HSCs(Fig. 2F, G). Increased
expressionof Nestin in activated HSCs wasfurther validated during
the spontaneous activation of HSCs in vitro (Fig. S3A−D). Taken
together, these data demonstrated the enhanced expression of
Nestin in HSC activation during liver fibrosis.
Induction of Nestin expression upon TGFbstimulation in HSCs
The enhancement of Nestin expression upon activation of HSCs
prompted us to further explore the mechanism of Nestin
F
HSCs from CCl4 treated
Nestin-GFP mice
α-SMA
Merge
GFP
GFP
Desmin
Merge
G
HSCs from DDC treated
Nestin-GFP mice
α-SMA
Merge
Desmin
GFP
GFP
Merge
NormalBAHBV
α-SMA NESTIN
BA
CCl
4
4w OilNormal dietDDC 4w
α-SMA/Nestin CD31/NestinF4/80/Nestin
Alb/Nestin CK19/Nestin
D
n.s.
n.s.
***
*
Vit.A CD31 F4/80
0
0.5
1
1.5
2
2.5
Relative GFP
expression (MFI)
***
*
*
Vit.A CD31 F4/80
0
1
2
3
4
E
***
**
Vit.A CD31 F4/80
0
2
4
6
Relative mRNA
expression of Nestin
(normalised to 18s)
***
*
Vit.A CD31 F4/80
0
2
4
6
8
C
Count
Nestin-GFP
Vit.A
100
80
60
40
20
0
-103010
3104105
CD31
100
80
60
40
20
0
-103010³10
4105
F4/80
100
80
60
40
20
0
-103010³10
4105
Vit.A F4/80
100
80
60
40
20
0
-103010³10
4105
CD31
50
40
30
20
10
0
-103010³10
4105
50
40
30
20
10
0
-103010³10
4105
Normal diet DDC 4wOil
CCl4 4w
DDC 4w
Oil
Normal diet
CCl4 4w
Fig. 2. Increased Nestin expression in HSCs after liver injury. (A) Representative fluorescent photographs for co-staining of Nestin (green) with
a
-SMA (red),
Alb (red), CK19 (red), F4/80 (red) and CD31 (red) in mouse liver tissues in indicated groups. Scale bars, 50
l
m. Nuclei were counterstained with DAPI. (B)
Representative fluorescent photographs for co-staining of NESTIN (green) with
a
-SMA (red) in normal and HBV- or BA-induced human fibrotic livers. Scale bars,
50
l
m. (C−E) (C) Representative histogram plots and (D) MFI quantification of FACS for Nestin-GFP and (E) qPCR for Nestin in Vit.A
+
, CD31
+
and F4/80
+
cells of
control and fibrotic livers from CCl
4
- or DDC-treated Nestin-GFP mice. n = 5/group. (F&G) Representative fluorescent photographs of GFP (green) and IF for
a
-SMA
or Desmin (red) of GFP
+
Vit.A
+
cells sorted with FACS from (F) CCl
4
- or (G) DDC-treated Nestin-GFP mice. Scale bars, 25
l
m. Data in (D) and (E) are reported as the
mean ± SEM with indicated significance (*p<0.05, **p<0.01, ***p<0.0 01, n.s.: not significant; Student’sttest). Alb, albumin; CK19, cytokeratin-19; CCl
4
, carbon
tetrachloride; DDC, 3,5-diethoxycarbonyl-1,4-dihydrocollidine; HSCs, hepatic stellate cells; MFI, mean fluorescence intensity; qPCR, quantitative real-time PCR;
Vit.A, Vitamin A.
4 Journal of Hepatology 2021 vol. -j1–12
Research Article Experimental and Translational Hepatology
induction in the liver fibrosis microenvironment. Recent studies
have revealed several transcription factors responsible for the
regulation of Nestin expression in tissue repair.
9,17
During liver
fibrosis, the TGFb-Smad2/3 pathway was persistently activated
and the transcription factor binding site prediction showed
multiple SBEs on the promoter of both human and murine Nestin
(Table S2). Thus, we tested whether the TGFbpathway could
regulate Nestin expression. TGFbstrongly induced NESTIN in
both the human HSC cell line LX2 and in mouse primary HSCs,
but not in human hepatic L02 cells (Fig. 3A, B). TbRI inhibitor
SB431542 decreased Nestin expression induced by TGFbin HSCs
(Fig. 3C), indicating that Nestin was a TGFb-responsive gene.
Similarly, in a 3D spheroid culture model of LX2, cell spheroids
treated with SB431542 showed weakened NESTIN and
a
-SMA
expression (Fig. 3D).
To further explore the detailed mechanism by which TGFb
induces NESTIN expression, we constructed different sections of
the promoter of NESTIN for the luciferase reporter assay. The
results showed that functional SBEs on the NESTIN promoter
involved in TGFbresponsiveness resided in the −1805 base pair
to −2067 base pair range, and mutation of these SBEs abolished
TGFb-induced luciferase activity (Fig. 3E, F). Together, these re-
sults demonstrated that NESTIN could be induced by TGFb
through functional SBEs on the promoter of NESTIN in HSCs.
Knockdown of Nestin expression hampers TGFb-Smad2/3
activity
To investigate the functional impact of Nestin on liver fibrosis,
we knocked down Nestin using shRNA in primary mouse HSCs
and LX2 cells. Knocking down Nestin in both systems markedly
decreased expression of the TGFbpathway downstream targets
in both basal and TGFb-stimulated conditions (Fig. 4A, B).
Consistently, we observed a decreased phosphorylation of
Smad2/3 in Nestin-knockdown HSCs (Fig. 4C, D). This suggested
that reduced expression of Nestin weakened TGFb-Smad2/3 ac-
tivity. Furthermore, luciferase reporter assays showed that the
knockdown of Nestin significantly decreased the activity of SBEs
induced by TGFb(Fig. 4E). Furthermore, in a 3D spheroid culture
model of LX2, spheroids with NESTIN-knockdown showed
reduced expression of
a
-SMA compared with the control in both
basal and TGFb-stimulated conditions (Fig. 4F). In summary, we
demonstrated that the knockdown of Nestin within HSCs
downregulated the TGFb-Smad2/3 pathway and thus antagon-
ised profibrotic gene expression.
Knockdown of Nestin promotes TbRI degradation through the
ubiquitin-proteasome pathway
TbRs can determine TGFbsignal transduction and subsequent
cell responses.
18
As protein binding affinity prediction indicated
B
D
NESTIN α-SMAMerge
TGFβ
SB431542
_+_+
__++
A-TGFβ +TGFβ
***
***
L02 LX2 Mouse
primary HSCs
0
2
4
6
Relative mRNA expression
of NESTIN
(normalised to GAPDH)
F
SBEs
The promoter regions of NESTIN (human)
*** ***
***
***
***
pGL3-basic
pGL3-A
pGL3-B
pGL3-C
pGL3-D
0
20
40
60
80
100
pGL3-basic
pGL3-E mut
pGL3-E
Relative luciferase intensity
pGL3-D
pGL3-E
pGL3-E mut
pGL3-C
pGL3-B
pGL3-A
-2,067 -1,895
,
-2,067 -1,895
Mutation
X
0-566-1,317-1,895-2,067
E
***
*** ***
0
2.5
5
20
40
60
90
120
150
Relative luciferase intensity
pGL3-basic
0 to -2,000
-2,000 to -3,500
-3,500 to -5,000
TGFβ +
_+
_+
_
L02 LX2
Mouse
primary HSCs
α-SMA
GAPDH
NESTIN
C
TGFβ
SB431542
α-SMA
GAPDH
LX2 Mouse primary HSCs
_+_+
__++
NESTIN
_+_+
__++
-TGFβ +TGFβ
n.s.
n.s.
Fig. 3. Induction of Nestin expression upon TGFbstimulation in HSCs. (A) qPCR and (B) representative western blot for NESTIN in L02, LX2 and mouse primary
HSCs treated with or without TGFbfor 48 h. GAPDH was used as a loading control for western blot (also hereafter in similar experiments). (C) Representative
western blot for NESTIN and
a
-SMA in LX2 cells and mouse primary HSCs in indicated conditions. Cells were treated with SB431542 (10 mM), TGFb(5 ng$ml
−1
)or
SB431542 plus TGFbfor 48 h. (D) Representative fluorescent photographs of IF staining for
a
-SMA (red) and NESTIN (green) in LX2 3D spheroids in indicated
groups. Cells were treated as in (C). (E) Luciferase reporter assay of HEK293T cells co-transfected with Renilla and different pGL3-based constructs of the NESTIN
promoter. Twenty-four hours after transfection, cells were treated with or without TGFb(5 ng$ml
−1
) for 48 h. Luciferase activity was measured and values were
referred to Renilla values and normalised to untreated pGL3-basic. (F) Schematic representation of pGL3-based constructs of the NESTIN promoter and mutant
constructs (left) and luciferase reporter assay of HEK293T cells co-transfected with Renilla and indicated constructs (right). Red blocks represent SBEs. Twenty-
four hours after transfection, cells were treated with TGFb(5 ng$ml
−1
) for 48 h. Values were normalised to pGL3-basic. Data in (A), (E) and (F) are reported as the
mean ± SEM of 3 independent experiments with an indicated significance (*p<0.05, **p<0.01, ***p<0.001, n.s.: not significant; Student’sttest). HSCs, hepatic
stellate cells; qPCR, quantitative real-time PCR; SBE, Smad-binding element.
Journal of Hepatology 2021 vol. -j1–12 5
that Nestin may interact with TbRs (http://dcv.uhnres.utoronto.
ca/FPCLASS/), we tested whether Nestin could regulate TbRs
during liver fibrosis. We found that knockdown of Nestin
significantly decreased the protein levels of TbRI but not TbRII, in
both mouse primary HSCs and LX2 cells (Fig. 5A). Meanwhile, the
mRNA levels of both TbRI and TbRII were barely affected (Fig. 5B).
These results indicated that Nestin may regulate TbRI post-
translationally.
The post-translational modification and degradation of TbRI is
tightly controlled and essential for TGFbsignalling activity.
19
Therefore, we next detected the stability of TbRI in control and
NESTIN-knockdown LX2 cells after treatment with the trans-
lation inhibitor cycloheximide (CHX). Western blot revealed that
TbRI was highly degraded in NESTIN-knockdown cells (Fig. 5C).
Statistical analysis further showed that TbRI had a shorter half-
life in NESTIN-knockdown cells relative to controls (Fig. 5D).
These data suggested that Nestin could protect TbRI from
degradation. We then pre-treated the cells with the broad-
spectrum proteasome inhibitor MG132 and found that it could
mitigate the decrease of TbRI in NESTIN-knockdown LX2 cells
shNes#1
shControl + TGFβ shNes#1 + TGFβ
shControl
Nestinp-Smad2/3Merge
D
B
Nestin
α-SMA
Gapdh
CollagenI
Ctgf
TGFβ
shControl shNes#1
_
shNes#2
Mouse primary HSCs
+_+_+_
LX2
shControl shNES#1 shNES#2
+_+_+
A
shControl
shControl + TGFβ
shNes#1
shNes#1 + TGFβ
shNes#2
shNes#2 + TGFβ
LX2
***
***
***
**
*
***
***
***
***
**
**
**
***
**
**
NESTIN CTGF ACTA2 COL1A3
0
2
4
6
8
Mouse primary HSCs
**
**
**
**
***
*** *
*
***
***
***
*** ***
**
***
***
Nestin Ctgf Acta2 Col1a1
0
5
10
15
Relative mRNA expression
(normalised to Gapdh)
F
NESTIN
Merge α-SMA
shControl + TGFβ shNES#1 shNES#1 + TGFβshControl
C
GAPDH
NESTIN
p-Smad2/3
Smad2/3
shNES#1 shNES#2
TGFβ (min)
shControl
015300 15 30 0 15 30
ELX2
***
***
***
**
0
5
10
15
20
Relative luciferase intensity
Mouse primary HSCs
**
***
***
*
Relative luciferase intensity
5
10
15
20
25
30
35
40
0
**
shControl
shControl + TGFβ
shNes#1
shNes#1 + TGFβ
shNes#2
shNes#2 + TGFβ
Fig. 4. Knockdown of Nestin expression hampers TGFb-Smad2/3 activity. (A) qPCR and (B) representative western blot for Nestin and fibrotic genes in control
and Nestin-knockdown mouse primary HSCs (left) and LX2 cells (right) treated with or without TGFb(5 ng$ml
−1
) for 48 h. shControl represents a non-target
scramble shRNA and was used as a negative control. Values were normalised to untreated shControl (also hereafter in similar experiments). (C) Representa-
tive western blot for NESTIN, phosphorylated and total Smad2/3 in the control and NESTIN-knockdown LX2 cells treated with TGFb(5 ng$ml
−1
) for 0, 15 and 30
min. (D) Representative fluorescent photographs of IF staining for Nestin (green) and phosphorylated Smad2/3 (red) in mouse primary HSCs in indicated groups.
Cells were treated with or without TGFb(5 ng$ml
−1
) for 6 h before analysis. Scale bars, 25
l
m. (E) Luciferase reporter assay for control and Nestin-knockdown
mouse primary HSCs (left) and LX2 (right) transfected with pGL3-SBE
9
-luciferase (SBE
9
, 9 tandem repeats of SBEs) construct. Twenty-four hours after trans-
fection, cells were treated with or without TGFb(5 ng$ml
−1
) for 48 h. Values were referred to Renilla values and normalised to the untreated shControl. (F)
Representative fluorescent photographs of IF for
a
-SMA (red) and NESTIN (green) in control and NESTIN-knockdown LX2 3D spheroids treated with or without
TGFb(5 ng$ml
−1
) for 48 h. Data in (A) and (E) are reported as the mean ± SEM of 3 independent experiments with an indicated significance (*p<0.05, **p<0.01,
***p<0.001, n.s.: not significant; Student’sttest). IF, immunofluorescence; HSCs, hepatic stellate cells; qPCR, quantitative real-time PCR; SBE, Smad-binding
element; sh, short-hairpin RNA; shNes,shNestin
6 Journal of Hepatology 2021 vol. -j1–12
Research Article Experimental and Translational Hepatology
(Fig. 5E). This indicated that NESTIN deficiency increased TbRI
degradation through proteasome-related pathways.
The ubiquitination of TbRI is required for its degradation.
20
Therefore, we performed a ubiquitination assay in LX2 cells
and found that knockdown of NESTIN increased the ubiquitina-
tion of TbRI (Fig. 5F). As previously reported, TbRI ubiquitination
followed by degradation mainly occurs on lipid rafts.
18,21
Accordingly, we used methyl-b-cyclodextrin (MbCD), an amphi-
pathic polysaccharide known to disrupt lipid rafts, to investigate
whether destroying the degradation machinery could restore the
loss of TbRI induced by the decrease in Nestin. As expected,
MbCD ameliorated the degradation of TbRI in NESTIN-
knockdown LX2 cells (Fig. 5G). Taken together, these findings
suggested that Nestin protected TbRI from ubiquitination and
proteasome degradation.
Nestin prevents TbRI from Cav-1-mediated degradation
Previous reports have indicated that Cav-1, an important
component of lipid rafts, interacts with TbRs and mediates their
endocytosis and degradation.
22,23
It has also been reported that
intermediate filaments can interact with Cav-1 and participate in
the regulation of vesicular trafficking.
24
Therefore, we speculated
that Nestin regulates TbRI degradation by modulating the
interaction between Cav-1 and TbRI.
To test our hypothesis, we first explored the interaction be-
tween NESTIN, CAV-1, and TbRI in LX2. Co-immunoprecipitation
results showed that NESTIN, CAV-1 and TbRI were within a
complex (Fig. 6A). Immunofluorescence co-staining also showed
the co-localisation of NESTIN with CAV-1 and TbRI in LX2 cells
(Fig. 6B). We knocked down NESTIN in LX2 cells and found that
this had little effect on CAV-1 expression in terms of both mRNA
and protein levels (Fig. 6C). However, the interaction between
endogenous CAV-1 and TbRI was significantly increased in
NESTIN-knockdown cells with protein degradation inhibited by
MG132 (Fig. 6D, E). Receptor endocytosis assays also showed that
TbRI-CAV-1 co-localisation was dramatically increased in
NESTIN-knockdown LX2 cells, indicating increased endocytosis
of TbRI after NESTIN knockdown (Fig. 6F). Furthermore, we
observed restoration of TbRI in NESTIN-knockdown LX2 cells co-
transduced with shCAV-1 (Fig. 6G). Knocking down CAV-1 in
NESTIN-knockdown LX2 cells not only recovered the activity of
the TGFb-Smad2/3 pathway but also restored the expression of
TGFb-responsive genes (Fig. 6H, I). Taken together, these findings
indicated that knocking down NESTIN could promote TbRI
degradation mediated by CAV-1 in HSCs.
These data demonstrated a positive feedback loop between
the TGFb-Smad2/3 pathway and Nestin during liver fibrosis.
Therefore, we hypothesised that inhibition of TGFbcould disrupt
AMouse primary HSCs
shControl
shNes
#1
shNes
#2
shControl
shNES
#1
shNES
#2
Gapdh
TβRI
TβRII
LX2
Nestin
B
Mouse primary HSCs
TβRI TβRII
0.0
0.5
1.0
1.5
Relative mRNA expression
(normalised to Gapdh)
LX2
TβRI
TβRII
0.0
0.5
1.0
1.5
C
NESTIN
TβRI
GAPDH
CHX (hr) 0 4 8 12 0 4 8 120 4 8 12
shNES#1 shNES#2shControl
D
Time (hr)
***
***
shNES
#2
shControl
shNES
#1
0
20
40
60
80
100
120
Relative remaining
TβRI levels (%)
04812
G
shNES#2
shControl
MβCD
shNES#1
GAPDH
TβRI
NESTIN
+
__
++
___
_
____
+
+
___
++
FIP:
IB:
t
up
nI
TβRI
GAPDH
MG132
shControl
shNES
#1
shNES
#2
TβRI
Ub
shControl shNes#1 shNes#2
n.s.
n.s.
n.s.
n.s.
n.s.
n.s.
n.s.
n.s.
NESTIN TβRI
*** ***
0.0
0.5
1.0
1.5
Optical density ratio
(normalised to GAPDH)
E
GAPDH
TβRI
NESTIN
shNES#2
shControl
MG132
shNES#1
++
___
____
+
___
++
+
___+
shControl
shNES#1
shNES#1 + MG132
shNES#2 + MG132
shNES#2
Fig. 5. Knockdown of Nestin promotes TbRI degradation through the ubiquitin-proteasome pathway. (A) Representative western blot and (B) qPCR for TbRs
in control and Nestin-knockdown mouse primary HSCs (left) and LX2 cells (right). (C) Representative western blot for NESTIN and TbRI in control and NESTIN-
knockdown LX2 cells treated with CHX (50
l
g$ml
−1
) for the indicated time. (D) Densitometry analysis of the remaining TbRI levels relative to GADPH in (C). (E)
Representative western blot (top) and densitometry quantification (bottom) for NESTIN and TbRI in control and NESTIN-knockdown LX2 cells treated with or
without MG132 (20
l
M) for 6 h. (F) Representative ubiquitination assays for TbRI ubiquitination after NESTIN knockdown in LX2 cells. Cells were treated with
MG132 (20
l
M) for 6 h before harvest. (G) Representative western blot for NESTIN and TbRI in control and NESTIN-knockdown LX2 cells treated with or without
MbCD (1.5 mM) for 4 h. Data in (B), (D), and (E) are reported as the mean ± SEM of 3 independent experiments with indicated significance (*p<0.05, **p<0.01, ***p
<0.001, n.s.: not significant; Student’sttest). CHX, cycloheximide; HSCs, hepatic stellate cells; MbCD, methyl-b-cyclodextrin; qPCR, quantitative real-time PCR; Ub,
ubiquitin.
Journal of Hepatology 2021 vol. -j1–12 7
this loop and alleviate liver fibrosis. As expected, Nestin and
a
SMA expression were downregulated in the CCl
4
-induced
fibrosis models treated with TGFbpathway inhibitor SB431542.
The histological analysis also showed that TGFbinhibition
slowed the progression of liver fibrosis (Fig. S4).
Knockdown of Nestin alleviates mouse liver fibrosis
To explore whether targeting Nestin could prevent further pro-
gression of established fibrosis in vivo, we used AAV6-shNestin to
treat CCl
4
- and DDC-induced mouse fibrosis models (Figs. 7 and
8). Recent studies have indicated that AAV6 exhibits organ
AInput IgG TβRI
IP
TβRI
CAV-1
NESTIN
IP
Input IgG Flag
TβRI
CAV-1
Flag
CAV-1
NESTIN
TβRI
Input IgG CAV-1
IP
NESTIN/CAV-1 NESTIN/TβRI
B
NESTIN
CAV-1
GAPDH
shControl
shNES#1
shNES#2
Relative mRNA expression of
CAV-1 (normalised to GAPDH)
1.5
1.0
0.5
0.0
shControl
shNES#1
shNES#2
C
shControl
shNES#1
TβRI TβRI CAV-1 Merge
4° C 37 °C, 40 min
37 °C, 40 min
shControl
shNES#1
**
0
10
20
30
40
50
Co-localisation spots of
CAV-1 with TβR1 (%)
F
shControl
shNES#1
- TGFβ+ TGFβ
shCAV-1
+
+
+
_
_
_
_
+
_
p-Smad2/3
ICOL1A3
**
***
0
1
2
3
***
***
0
1
2
3
4
Relative mRNA expression of
CAV-1 (normalised to GAPDH)
ACTA2
n.s.
n.s.
***
***
***
***
Relative mean p-Smad2/3
intensity/nucleus
0
20
40
80
100
60
***
***
G
shControl
shNES#1
+__
+
_+
NESTIN
CAV-1
TβRI
GAPDH
shCAV-1 +
__
***
***
***
***
NESTIN CAV-1
Optical density ratio
(normalised to GAPDH)
1.5
1.0
0.5
0.0
TβR1
DIP: TβRI MG132
CAV-1
TβRI
CAV-1
GAPDH
TβRI
Input IB
EIP: CAV-1 MG132
shControl
shNES#1
shNES#2
shControl
shNES#1
shNES#2
TβRI
GAPDH
CAV-1
TβRI
CAV-1
Input IB
H
n.s.
n.s.
shControl
shNES#1
shNES#1 + shCAV-1
shControl
shNES#1
shNES#1 + shCAV-1
shControl + TGFβ
shNES#1 + TGFβ
shNES#1 + shCAV-1 + TGFβ
Fig. 6. Nestin protects TbRI from Cav-1 mediated degradation. (A) Representative western blot for Co-IP of TbRI with CAV-1 and NESTIN (top), CAV-1 with TbRI
and NESTIN (middle), and Flag-NESTIN with TbRI and CAV-1 (bottom) in LX2 cells. Cells were transfected with 3*Flag-tagged NESTIN (Flag-NESTIN) and lysed 72 h
later. IgG was used as a negative control. (B) Representative fluorescent photographs of co-staining for NESTIN (green) with CAV-1 (red) (left) and NESTIN (green)
with TbRI (red) (right) in LX2 cells. Scale bar, 5
l
m. (C) Representative western blot (top) and qPCR (bottom) for indicated genes in control and NESTIN-
knockdown LX2 cells. (D&E) Representative Co-IP of TbRI and CAV-1 in control and NESTIN-knockdown LX2 cells treated with MG132 (20
l
M) for 6 h. (F)
Representative fluorescent photographs (left) and quantification (right) of receptor endocytosis assays for TbRI (green) and CAV-1 (red) in control and NESTIN-
knockdown LX2 cells. Scale bar, 5
l
m. (G) Representative western blot for NESTIN, CAV-1 and TbRI (left) and densitometry quantification (right) in LX2 cells with
indicated treatments. (H) Representative fluorescent photographs (top) and quantification of MFI (bottom) of IF for phosphorylated Smad2/3 (red) in LX2 cells in
indicated groups. Values were normalised to the untreated shControl. Cells were treated with or without TGFb(5 ng$ml
−1
) for 6 h before harvest. (I) qPCR for
fibrotic genes of LX2 cells in indicated groups. Cells were treated with or without TGFb(5 ng$ml
−1
) for 48 h before harvest. Data in (C), (F)−(I) are reported as mean
± SEM of 3 independent experiments with an indicated significance (*p<0.05, **p<0.01, ***p<0.0 01, n.s.: not significant; Student’sttest). Co-IP, co-immuno-
precipitation; IF, immunofluorescence; MFI, mean fluorescence intensity; qPCR, quantitative real-time PCR; sh, short-hairpin RNA.
8 Journal of Hepatology 2021 vol. -j1–12
Research Article Experimental and Translational Hepatology
tropism for activated HSCs in the liver during fibrosis.
25
The
administration of AAV6 (a non-targeting control vector
expressing GFP, AAV6-shControl) barely affected the liver his-
tology, body weight, liver weight or liver function of the animals
(Fig. S5A−D). We assessed the transduction efficiency of AAV6 to
HSCs upon CCl
4
administration using AAV6-shControl. Statistical
analysis of GFP
+
cells showed that the efficacy of AAV6 trans-
duction in Nestin
+
HSCs was about 17.04 ± 1.55% (Fig. S5E). AAV6
showed a clear liver tropism compared to other organs (e.g.
heart, lungs; Fig. S5F).
Following this, we used AAV6-shNestin in liver fibrotic mice to
investigate the function of Nestin in CCl
4
-induced liver fibrosis
in vivo (Fig. 7A). Histological examination, hydroxyproline con-
tent analyses and qPCR results indicated that the levels of in-
flammatory infiltration, hepatocellular damage and the degree of
fibrosis were alleviated upon AAV6-shNestin treatment (Fig. 7B
−H). A similar recovery of liver fibrosis was observed in the
DDC model after AAV6-shNestin treatment (Fig. 8). Additionally,
decreased Smad2/3 phosphorylation was observed in the AAV6-
shNestin group (Fig. S5G, H). AAV6-shNestin administration had
minimal influence on histology and Nestin expression in other
organs, including the lungs and heart (Fig. S5I, J). These results
showed that targeting Nestin with AAV6 alleviated liver fibrosis
in vivo.
Discussion
This study investigated the function and precise regulation of
Nestin during liver fibrosis. The expression of Nestin was upre-
gulated in activated HSCs and was correlated with the severity of
fibrosis in both patient specimens and mouse models. Mecha-
nistically, we demonstrated that TGFbdirectly induced Nestin
expression and aggravated fibrogenesis. Also, the expression of
Nestin facilitated the TGFb-Smad2/3 pathway through allevia-
tion of the Cav-1-mediated degradation of TbRI. The knockdown
of Nestin expression within HSCs attenuated the profibrogenic
phenotype in vitro. Most importantly, we provided evidence to
20% CCl4 i.p. (twice per week) Harvest
C57BL/6 Mice
AAV6 shControl/shNes#1/shNes#2 i.v.
0w 1w 2w 3w 4w 5w 6w
Nestin α-SMA Merge
AAV6-shControl AAV6-shControl
AAV6-shNes#1
AAV6-shNes#2
CCl
4
Oil
AAV6-shControl AAV6-shControl
AAV6-shNes#1
AAV6 -s h Nes#2
Oil
CCl
4
HE PSR α-SMA Nestin
***
***
0
2
4
6
8
PSR
positive area (%)
***
***
0
5
10
15
Nestin positive area (%)
F
***
***
0
100
200
300
Hyp (μg.100 mg-1)
E
D
A
BC
G
**
*
0
100
200
300
ALT (U.L-1)
***
***
0
100
200
300
AST (U.L-1)
*
*
0
20
40
60
80
ALP (U.L-1)
0
2
4
6
8
10
TBIL (μmol.L-1)
H
**
**
***
*** ***
***
***
*** **
***
**
**
Relative mRNA level
Nestin Acta2
0
5
10
15
Col1a1 Il-1β Tnfα Ccl5
Oil + AAV6-shControl
CCl4 + AAV6-shNes#1
CCl4 + AAV-shControl
CCl4 + AAV6-shNes#2
n.s.
n.s.
Fig. 7. Knockdown of Nestin alleviates CCl
4
-induced mouse liver fibrosis. (A) Schematic overview of the experimental design. (B) Representative fluorescent
photographs of IF staining for Nestin (red) and
a
-SMA (white) in liver tissues of indicated groups. Scale bars, 100
l
m. (C) Representative H&E, PSR staining and IHC
for
a
-SMA, and IHC for Nestin. Scale bars, 100
l
m. Quantification for (D) PSR staining; (E) Nestin IHC; and (F) Hyp content in liver tissues of indicated groups. N =
6/group. (G) Levels of ALT, AST, ALP and TBIL in serum from mice in indicated groups at the time of sacrifice. n = 6/group. (H) qPCR for Nestin, profibrotic genes and
inflammatory genes of mouse liver tissues in indicated groups. n = 6/group. Data in (D)−(H) are reported as the mean ± SEM with an indicated significance (*p
<0.05, **p<0.01, ***p<0.001, n.s.: not significant; Student’sttest). ALP, alkaline phosphatase; ALT, alanine aminotransferase; AST, aspartate aminotransferase; CCl
4
,
carbon tetrachloride; Hyp, hydroxyproline; IF, immunofluorescence; IHC, immunohistochemistry; PSR, Picrosirius red; qPCR, quantitative real-time PCR; TBIL,
total bilirubin
Journal of Hepatology 2021 vol. -j1–12 9
show that using an AAV6 vector to reduce Nestin expression
within activated HSCs could alleviate liver fibrogenesis in mouse
models, providing a promising antifibrotic target.
Nestin, an intermediate filament, forms extensive and elab-
orate networks with other cytoskeleton components. It plays an
essential role in stem cell maintenance, proliferation, differen-
tiation and migration.
26,27
Nestin expression is highly dynamic
and precisely regulated under physiological and pathological
conditions.
28,29
In the foetal liver, Nestin is expressed in peri-
vascular cells around arterial portal vessels, where it provides a
niche for promoting haematopoietic stem cell expansion. Nestin
levels decrease in perivascular cells when portal vessels transit
into vein phenotypes.
27
However, Nestin has rarely been
observed in the healthy adult liver and the recurrence of Nestin
has been observed after either acute or chronic liver injury.
7–10
This indicates that Nestin may be involved in wound healing or
regeneration of the adult liver. Indeed, evidence from rat liver
regeneration models suggests that Nestin
+
populations partici-
pate in progenitor formation and epithelial regeneration.
30
In
malignant settings, evidence has shown that the upregulation of
Nestin contributes to tumorigenesis of HCC and intrahepatic
cholangiocarcinoma.
8–10
It also serves as a biomarker for poor
prognosis of liver cancers.
10
However, the exact role of Nestin in
chronic liver injury remains elusive.
Our study showed that Nestin was increased in HSCs in liver
fibrosis/cirrhosis samples and positively correlated with fibrosis
progression. It is well-known that fibrosis in multiple organs
(including the liver, lungs, kidney, skin, etc.) is a self-limiting
and homeostatic process that occurs in response to tissue
injury. However, it becomes a major risk factor for carcino-
genesis once the process becomes out of control.
8–10
Thus,
targeting Nestin in the early stages of liver fibrosis may serve as
a potential antifibrotic approach and prevent progression to
HCC.
0.1% DDC diet Harvest
C57BL/6 Mice
AAV6 shControl/shNes#1/shNes#2 i.v.
0w 1w 2w 3w 4w 5w 6w
B
GH
A
CD
F
E
AAV6-shControl AAV6-shControlAAV6-shNes#1AAV6-shNes#2
DDC Normal diet
Nestin α-SMA Merge
PSRHE α-SMA Nestin
AAV6-shControlAAV6-shNes#1AAV6-shNes#2
DDC
AAV6-shControl
Normal diet
***
***
0
5
10
15
20
PSR
positive area (%)
***
***
0
2
4
6
8
Nestin positive area (%)
***
***
0
100
200
300
400
Hyp (μg.100 mg-1)
**
***
0
500
1,000
1,500
2,000
ALT (U.L-1)
***
**
0
200
400
600
800
ALP (U.L-1)
***
***
0
50
100
150
TBIL (μmol.L-1)
200
400
600
800
AST (U.L-1)
0
*
*
Relative mRNA level
***
*** ***
*** ***
***
***
***
**
***
**
0
5
10
15
20
25
Normal diet + AAV6-shControl
DDC + AAV6-shNes#1
DDC + AAV-shControl
DDC + AAV6-shNes#2
Nestin Acta2 Col1a1 Il-1β Tnfα Ccl5
Fig. 8. Knockdown of Nestin alleviates DDC-induced mouse liver fibrosis. (A) Schematic overview of the experimental design. (B) Representative fluorescent
photographs of IF staining for Nestin (red) and
a
-SMA (white) in liver tissues of indicated groups. Scale bars, 100
l
m. (C) Representative HE, PSR staining and IHC
for
a
-SMA, and IHC for Nestin. Scale bars, 100
l
m. Quantification for (D) PSR staining; (E) Nestin IHC; and (F) Hyp content in liver tissues of indicated groups. N =
6/group. (G) Levels of ALT, AST, ALP and TBIL in serum from mice in indicated groups at the time of sacrifice. n = 6/group. (H) qPCR for Nestin, profibrotic genes and
inflammatory genes of mouse liver tissues in indicated groups. N = 6/group. Data in (D)−(H) are reported as the mean ± SEM with an indicated significance (*p
<0.05, **p<0.01, ***p<0.001, n.s.: not significant; Student’sttest). ALP, alkaline phosphatase; ALT, alanine aminotransferase; AST, aspartate aminotransferase; DDC,
3,5-diethoxycarbonyl-1,4-dihydrocollidine; Hyp, hydroxyproline; IF, immunofluorescence; IHC, immunohistochemistry; PSR, Picrosirius red; qPCR, quantitative
real-time PCR; TBIL, total bilirubin.
10 Journal of Hepatology 2021 vol. -j1–12
Research Article Experimental and Translational Hepatology
Nestin is classically regarded as a component of the cyto-
skeleton. Its function in cellular responses has been extensively
studied over the past few decades. Sahlgren et al. found that the
downregulation of Nestin sensitised neuronal progenitors to
exogenous reactive oxygen species-induced cell death.
31
Furthermore, our previous studies showed that Nestin regu-
lated cellular redox homeostasis,
32
and Nestin knockdown
resulted in mitochondrial respiratory dysfunction and thus
attenuated self-renewal of neural stem/progenitor cells.
33
Also,
Nestin was found to interact with the inner nuclear membrane
protein lamin A/C and participate in protecting tumour cells from
senescence.
34
In this study, we demonstrated that Nestin directly bound to
TbRI and Cav-1 in HSCs and protected TbRI from Cav-1-mediated
degradation. It thereby sustained the TGFb-Smad2/3 signalling
within HSCs, directly integrating extracellular signals to intra-
cellular events. We also demonstrated that Nestin was a direct
target gene of TGFb-Smad2/3 signalling, forming a positive
feedback loop to aggravate liver fibrosis. As TGFb-Smad2/3 sig-
nalling is a general profibrogenic stimulus, it will be interesting
to see whether this mechanism is conserved in fibrosis of other
organs.
Recent studies have demonstrated that HSCs are a very het-
erogeneous population that is of functional relevance for liver
homeostasis.
16
In both human fibrotic/cirrhotic specimens and
murine liver fibrotic models, we identified that Nestin
+
cells
largely overlapped with
a
-SMA
+
or Desmin
+
HSCs. This was
consistent with previous reports in rats.
30
Interestingly, not all
a
-
SMA
+
or Desmin
+
HSCs were stained for Nestin (Fig. S2A, B). Also,
Nestin
+
HSCs showed a more fibrogenic phenotype than Nestin
−
populations; this suggested that Nestin could be a functional
marker for activated HSCs, distinct from conventional markers
like Desmin, platelet-derived growth factor receptor b(Pdgfrb)
and
a
-SMA. However, previous reports have also shown that
Nestin marks a small population of stem cell−like cells in the
adult liver.
35
Additionally, the migration of bone marrow
mesenchymal cells (BM-MSCs) contributes to fibrosis in multiple
organs, including the liver,
36
and Nestin is regarded as a marker
for BM-MSCs.
37
Thus, it will be interesting to further characterise
the heterogeneity of Nestin
+
cells in liver fibrosis with single-cell
RNA sequencing and recently developed delicate lineage tracing
tools.
Although HSCs are regarded as the central driver of liver
fibrosis, targeting HSCs (a small population in the liver) has been
a tremendous challenge for many years. This is at least partially
because of the abundant receptor distribution and strong
phagocytosis of hepatocytes, LSECs and KCs. Recently, Rezvani
et al. demonstrated that an AAV6 vector carrying transcription
factors could successfully reprogramme activated HSCs into he-
patocytes and promote mouse liver fibrosis recovery.
25
AAV
vectors have been regarded as a safe and effective gene delivery
tool in clinical trials since they have weak immunogenicity and
low oncogenicity.
38,39
Therefore, we used AAV6 vectors carrying
shNestin to target activated HSCs and observed an apparent re-
covery from liver fibrosis (Fig. 7, 8). Consistent with previous
reports,
40,41
both in vitro and in vivo results indicated that tar-
geting Nestin reverted HSCs to a quiescent state and inhibited
cell proliferation. However, there was negligible effect on cellular
apoptosis. This was evidenced by an increased quiescent HSC
marker glial fibrillary acidic protein (GFAP) accompanying
decreased activation marker a-SMA expression, as well as
decreased Ki67 positive cells in targeted cells but unchanged
TUNEL-positive cells (Fig. S6). However, only a small proportion
of Nestin
+
HSCs were targeted after AAV6 administration
(Fig. S5E). Besides AAV vectors, liposomes and other novel
nanomaterial-based targeted delivery tools have been developed
in recent years. These have displayed better targeting efficiency
and great promise for liver fibrosis therapy.
42–44
Thus, it is of
great interest to further explore other therapeutic approaches by
targeting Nestin, which might provide novel insights into the
development of antifibrosis strategies.
Abbreviations
AAV6, adeno-associated virus vector serotype 6; ACTA2 or
a
-
SMA,
a
-smooth muscle actin; BA, biliary atresia; Cav-1, caveolin
1; CCl
4
, carbon tetrachloride; DDC, 3,5-diethoxycarbonyl-1,4-
dihydrocollidine; HCC, hepatocellular carcinoma; HSCs, hepatic
stellate cells; KCs, Kupffer cells; LSECs, liver sinusoidal endo-
thelial cells; MbCD, methyl-b-cyclodextrin; PBC, primary biliary
cholangitis; qPCR, quantitative real-time PCR; SBEs, Smad-
binding elements; shRNA, short-hairpin RNA; TBIL, total bili-
rubin; TGFb, transforming growth factor b;TbR, TGFbreceptor;
TbRI, TGFbreceptor I; TbRII, TGFbreceptor II.
Financial support
This work was supported by the National Key Research and
Development Program of China (2018YFA0107203 and
2017YFA0106100), Guangdong Province Universities and Col-
leges Pearl River Scholar Funded Scheme (2017), Key Technology
Innovation of Guangdong Province (2015B020226004), Key
Project Fund of Guangdong Natural Science Foundation
(2017A030311034), National Keypoint Research and Invention
Program of the Thirteenth (2018ZX10723203), Science and
Technology Program of Guangdong Province (2019B020236003),
the Fundamental Research Funds for the Central Universities
(19ykpy158), the National Natural Science Foundation of China
(81670601, 81971372, 81802402, 81770648, 81970537, 31601184,
3177616 and 81730005), Key Scientific and Technological Pro-
gram of Guangzhou City (201803040011 and 201802020023),
Guangdong Basic and Applied Basic Research Foundation
(2020A1515011385), Research Start-up Fund of the Seventh
Affiliated Hospital, Sun Yat-sen University (393011), and the
Ph.D. Start-up Fund of Natural Science Foundation of Guangdong
Province (2018A030310323).
Conflict of interest
All authors declare that they do not have anything to disclose
regarding funding or conflict of interest.
Please refer to the accompanying ICMJE disclosure forms for
further details.
Authors’contributions
HXC, JYC, and JCW performed experiments, analysed data, and
wrote the paper. YQ, CHJ, YW, YQW, CJY, QYL, LJP, GL, JZ, and YJG
performed experiments and analysed data. DBQ, CD, and QLL
analysed data and edited the paper. GHC, YY, YX, APX, and QZ
designed experiments, analysed data, and edited the paper.
Data availability statement
All data, models, and materials generated or used during the
study are available from the corresponding authors upon
reasonable request.
Journal of Hepatology 2021 vol. -j1–12 11
Acknowledgements
We thank all members of the Andy Peng Xiang’s lab and Qi
Zhang’s lab for their support and technical assistance.
Supplementary data
Supplementary data to this article can be found at https://doi.
org/10.1016/j.jhep.2020.11.016.
References
Author names in bold designate shared co-first authorship
[1] Bataller R, Brenner DA. Liver fibrosis. J Clin Invest 2005;115(2):209–218.
[2] Mederacke I,Hsu CC,Troeger JS, Huebener P, Mu X, Dapito DH, et al. Fate
tracing reveals hepatic stellate cells as dominant contributors to liver
fibrosis independent of its aetiology. Nat Commun 2013;4(1).
[3] Dewidar, Meyer, Dooley, Meindl B. TGF-bin hepatic stellate cell activation
and liver fibrogenesis—updated 2019. Cells 2019;8(11).
[4] Hockfield S, McKay RD. Identification of major cell classes in the devel-
oping mammalian nervous system. J Neurosci 1985;5(12):3310–3328.
[5] Frisen J, Johansson CB, Torok C, Risling M, Lendahl U. Rapid, widespread,
and longlasting induction of nestin contributes to the generation of glial
scar tissue after CNS injury. J Cel Biol 1995;131(2):453–464.
[6] Sakairi T, Hiromura K, Yamashita S, Takeuchi S, Tomioka M, Ideura H, et al.
Nestin expression in the kidney with an obstructed ureter. Kidney Int
2007;72(3):307–318.
[7] Niki T, Pekny M, Hellemans K, Bleser PD, Berg KV, Vaeyens F, et al. Class VI
intermediate filament protein nestin is induced during activation of rat
hepatic stellate cells. Hepatology 1999;29(2):520–527.
[8] Yang XR,Xu Y,Yu B,Zhou J, Qiu SJ, Shi GM, et al. High expression levels of
putative hepatic stem/progenitor cell biomarkers related to tumour
angiogenesis and poor prognosis of hepatocellular carcinoma. Gut
2010;59(7):953–962.
[9] Tschaharganeh Darjus F, Xue W, Calvisi Diego F, Evert M, Michurina
Tatyana V, Dow Lukas E, et al. p53-Dependent nestin regulation links
tumor suppression to cellular plasticity in liver cancer. Cell
2014;158(3):579–592.
[10] Xue R,Chen L,Zhang C,Fujita M, Li R, Yan SM, et al. Genomic and
transcriptomic profiling of combined hepatocellular and intrahepatic
cholangiocarcinoma reveals distinct molecular subtypes. Cancer Cell
2019;35(6):932–947. e938.
[11] Budi EH, Duan D, Derynck R. Transforming growth factor-beta receptors
and smads: regulatory complexity and functional versatility. Trends Cel
Biol 2017;27(9):658–672.
[12] Schmierer B, Hill CS. TGFbeta-SMAD signal transduction: molecular
specificity and functional flexibility. Nat Rev Mol Cel Biol 2007;8(12):970–
982.
[13] Mederacke I,Dapito DH, Affo S, Uchinami H, Schwabe RF. High-yield and
high-purity isolation of hepatic stellate cells from normal and fibrotic
mouse livers. Nat Protoc 2015;10(2):305–315.
[14] Ishak K, Baptista A, Bianchi L, Callea F, De Groote J, Gudat F, et al. Histo-
logical grading and staging of chronic hepatitis. J Hepatol
1995;22(6):696–699.
[15] Yamaguchi M. Visualization of neurogenesis in the central nervous system
using nestin promoter-GFP transgenic mice. Developmental Neuroscience
2000.
[16] Iwaisako K,Jiang C,Zhang M, Cong M, Moore-Morris TJ, Park TJ, et al.
Origin of myofibroblasts in the fibrotic liver in mice. Proc Natl Acad Sci U S
A 2014;111(32):E3297–E3305.
[17] Han DW, Do JT, Araúzo-Bravo MJ, Lee SH, Meissner A, Lee HT, et al.
Epigenetic hierarchy governing Nestin expression. Stem Cells
2009;27(5):1088–1097.
[18] Di Guglielmo GM,Le Roy C, Goodfellow AF, Wrana JL. Distinct endocytic
pathways regulate TGF-breceptor signalling and turnover. Nat Cel Biol
2003;5(5):410–421.
[19] Yan X, Chen Y. Posttranslational modifications of TGF-breceptors.
Methods Mol Biol (Clifton, NJ) 2016;1344:49–61.
[20] Heldin CH, Moustakas A. Signaling receptors for TGF-beta family mem-
bers. Cold Spring Harb Perspect Biol 2016;8(8).
[21] Lonn P, Moren A, Raja E, Dahl M, Moustakas A. Regulating the stability of
TGFbeta receptors and Smads. Cell Res 2009;19(1):21–35.
[22] Zhao B, Wang Q, Du J, Luo S, Xia J, Chen YG. PICK1 promotes caveolin-
dependent degradation of TGF-beta type I receptor. Cel Res
2012;22(10):1467–1478.
[23] Zuo W,Huang F, Chiang YJ, Li M, Du J, Ding Y, et al. c-Cbl-Mediated
neddylation antagonizes ubiquitination and degradation of the TGF-b
type II receptor. Mol Cel 2013;49(3):499–510.
[24] Margiotta A, Bucci C. Role of intermediate filaments in vesicular traffic.
Cells 2016;5(2).
[25] Rezvani M,Español-Suñer R,Malato Y,Dumont L, Grimm Andrew A,
Kienle E, et al. In Vivo hepatic reprogramming of myofibroblasts with AAV
vectors as a therapeutic strategy for liver fibrosis. Cell Stem Cell
2016;18(6):809–816.
[26] Kunisaki Y, Bruns I, Scheiermann C, Ahmed J, Pinho S, Zhang D, et al.
Arteriolar niches maintain haematopoietic stem cell quiescence. Nature
2013;502(7473):637–643.
[27] Khan JA, Mendelson A, Kunisaki Y, Birbrair A, Kou Y, Arnal-Estape A, et al.
Fetal liver hematopoietic stem cell niches associate with portal vessels.
Science 2016;351(6269):176–180 .
[28] Cheng F, Eriksson JE. Intermediate filaments and the regulation of cell
motility during regeneration and wound healing. Cold Spring Harb Per-
spect Biol 2017;9(9).
[29] Sahlgren CM, Mikhailov A, Vaittinen S, Pallari H-M, Kalimo H, Pant HC,
et al. Cdk5 regulates the organization of nestin and its association with
p35. Mol Cell Biol 2003;23(14):5090–5106 .
[30] Kordes C, Sawitza I, Gotze S, Herebian D, Haussinger D. Hepatic stellate
cells contribute to progenitor cells and liver regeneration. J Clin Invest
2014;124(12):5503–5515.
[31] Sahlgren CM,Pallari HM, He T, Chou YH, Goldman RD, Eriksson JE.
A nestin scaffold links Cdk5/p35 signaling to oxidant-induced cell death.
EMBO J 2006;25(20):4808–4819.
[32] Wang J,Lu Q,Cai J, Wang Y, Lai X, Qiu Y, et al. Nestin regulates cellular
redox homeostasis in lung cancer through the Keap1-Nrf2 feedback loop.
Nat Commun 2019;10(1):5043.
[33] Wang J,Huang Y,Cai J, Ke Q, Xiao J, Huang W, et al. A nestin-cyclin-
dependent kinase 5-dynamin-related protein 1 Axis regulates neural
stem/progenitor cell stemness via a metabolic shift. Stem Cells
2018;36(4):589–601.
[34] Zhang Y,Wang J,Huang W, Cai J, Ba J, Wang Y, et al. Nuclear Nestin
deficiency drives tumor senescence via lamin A/C-dependent nuclear
deformation. Nat Commun 2018;9(1):3613.
[35] Gleiberman AS, Encinas JM, Mignone JL, Michurina T, Rosenfeld MG,
Enikolopov G. Expression of nestin-green fluorescent protein transgene
marks oval cells in the adult liver. Dev Dyn 2005;234(2):413–421.
[36] Schneider RK, Mullally A, Dugourd A, Peisker F, Hoogenboezem R, Van
Strien PMH, et al. Gli1(+) mesenchymal stromal cells are a Key driver of
bone marrow fibrosis and an important cellular therapeutic target. Cell
Stem Cell 2017;20(6):785–800. e788.
[37] Mendez-Ferrer S, Michurina TV, Ferraro F, Mazloom AR, Macarthur BD,
Lira SA, et al. Mesenchymal and haematopoietic stem cells form a unique
bone marrow niche. Nature 2010;466(7308):829–834.
[38] Crystal RG. Adenovirus: the first effective in vivo gene delivery vector.
Hum Gene Ther 2014;25(1):3–11.
[39] Li H, Malani N, Hamilton SR, Schlachterman A, Bussadori G, Edmonson SE,
et al. Assessing the potential for AAV vector genotoxicity in a murine
model. Blood 2011;117(12):3311–3319.
[40] Troeger JS,Mederacke I, Gwak GY, Dapito DH, Mu X, Hsu CC, et al.
Deactivation of hepatic stellate cells during liver fibrosis resolution in
mice. Gastroenterology 2012;143(4):1073–1083. e1022.
[41] Kisseleva T, Cong M, Paik Y, Scholten D, Jiang C, Benner C, et al. Myofi-
broblasts revert to an inactive phenotype during regression of liver
fibrosis. Proc Natl Acad Sci U S A 2012;109(24):9448–9453.
[42] Bangen JM, Hammerich L, Sonntag R, Baues M, Haas U, Lambertz D, et al.
Targeting CCl4 -induced liver fibrosis by RNA interference-mediated in-
hibition of cyclin E1 in mice. Hepatology 2017;66(4):1242–1257.
[43] Li Y,Pu S, Liu Q, Li R, Zhang J, Wu T, et al. An integrin-based nanoparticle
that targets activated hepatic stellate cells and alleviates liver fibrosis.
J Control Release 2019;303:77–90.
[44] Chen Z,Xiao L, Liu W, Liu D, Xiao YY, Chen J, et al. Novel materials which
possess the ability to target liver cells. Expert Opin Drug Deliv
2012;9(6):649–656.
12 Journal of Hepatology 2021 vol. -j1–12
Research Article Experimental and Translational Hepatology
