Sumoylation enhances the activity of the TGF-β/SMAD and
HIF-1 signaling pathways in keloids
Xiaohu Lin, Yuming Wang, Yan Jiang, Mingyuan Xu, Qianqian
Pang, Jiaqi Sun, Yijia Yu, Zeren Shen, Rui Lei, Jinghong Xu
PII: S0024-3205(20)30609-3
Reference: LFS 117859
To appear in: Life Sciences
Received date: 18 March 2020
Revised date: 24 May 2020
Accepted date: 26 May 2020
Please cite this article as: X. Lin, Y. Wang, Y. Jiang, et al., Sumoylation enhances the
activity of the TGF-β/SMAD and HIF-1 signaling pathways in keloids, Life Sciences
(2020)
This is a PDF file of an article that has undergone enhancements after acceptance, such
as the addition of a cover page and metadata, and formatting for readability, but it is
not yet the definitive version of record. This version will undergo additional copyediting,
typesetting and review before it is published in its final form, but we are providing this
version to give early visibility of the article. Please note that, during the production
process, errors may be discovered which could affect the content, and all legal disclaimers
that apply to the journal pertain.
© 2020 Published by Elsevier.
Journal Pre-proof 1
Sumoylation Enhances the Activity of the TGF
-β/SMAD and HIF
-1 Signaling Pathway
s in Keloid
Xiaohu Lin
, Yuming Wang
, Yan Jiang
, Mingyuan Xu
, Qianqian Pang
, Jiaqi Sun
, Yijia Yu
, Zeren
Shen
, Rui Lei
, Jinghong Xu
a Department of Plastic Surgery, The First Affiliated Hospital, School of Medicine, Zhejiang University,
Hangzhou, 310003, China *Corresponding author at Department of Plastic Surgery, The First Affiliated Hospital, Zhejiang
University School of Medicine, 79Qingchun Road, Hangzhou Zhejiang, China
Emai
1address:
[email protected]
n, Tel. (0571)8723
-6307
Abstract
Excessive fibrosis and extracellular matrix deposition resulting from upregulation of target genes
expression mediated by transforming growth factor
-beta
(TGF
β)/SMAD and hypoxia inducible factor
-1
(HIF
-1) signaling pathway
s are the main mechanisms that drive keloid formation. Sumoylation is a
protein posttranslational modification that regulates the function of proteins in many biological processes.
In the present study, we aimed to investigate the mechanism underlying the effects of sumoylation on the
TGF
-β/SMAD and HIF
-1 signaling pathways in keloid
s. We used 2
-D08 to block sumoylation and
silenced the expression of sentrin sumo
-specific protease 1 (SENP1) to enhance sumoylation in human
foreskin fibroblasts (HFF
s) and human keloid fibroblasts (HKF
s). We also reduced and increased
intracellular SUMO1 levels by silencing SUMO1 and transfecting cells with a SUMO1 overexpression
lentivirus, respectively. Sumoylation has the ability to amplify TGF
-β/SMAD and HIF
-1 signals in
keloids, while SUMO1, especially the SUMO1
-RanGAP1 complex, is the key molecule affecting the
TGF
-β/SMAD and HIF
-1 signaling pathway
s. In addition, we also found that hypoxia promotes
sumoylation in keloid
s and that HIF
-1α is covalently modified by SUMO1 at Lys 391 and Ly
s 477 in
HKF
s. In summary, we elucidated the role and molecular mechanism of sumoylation in the formation of
keloid
s, providing a new perspective for
a potential therapeutic target of keloid
Keywords: Sumoylation, keloid, SUMO1, SUMO1
-RanGAP1 complex, HIF
-1α, hypoxia, TG
-β/SMAD
signaling pathway
1. Introduction
Keloids are formed from a pathological process of skin wound healing, and they have racial
predisposition and familial tendency [1
-3]. They usually occur after surgery, infection or trauma involving
the deep layers of the dermis
, and they do not degenerate on their own [4, 5]. Surgical resection is usually
ineffective due to high recurrence rates, external malformations and postoperative dysfunction [6, 7].
Therefore, keloid
s have become a popular research topic in skin wound healing, but their pathogenesis is
still not completely understood.
The pathological essence of keloid
s are overabundant wound healing caused by excessive fibrosis,
mainly manifested by increased proliferation of fibroblasts and high expression of factors in the
Journal Pre-proof
Journal Pre-proof 2
extracellular matrix, which is marked especially by the excessive synthesis and deposition of collagen [4,
-hypoxi
c state due to trauma
-induced subcutaneous vascular system
damage and hypermetabolism of cells during inflammation and wound repair [8]. Hypoxia can induce a
hypoxic stress response and stimulate HIF
-1α protein expression and nuclear translocation as well as
upregulate of TGF
β gene expression [10
-12]. Previous studies have reported significant enhancement of
TGF
-β/SMAD and HIF
-1 pathway activity in keloid
s, but the mechanism remains unclear [13]
Sumoylation is a posttranslational modification that has been extensively studied in recent years.
SUMOs, small ubiquitin
-like modifiers, are widely distributed in the human body. Similar to the process
of ubiquitination, SUMOs recognize and bind to specific lysine residues on target proteins modify
substrate proteins in eukaryotes [14, 15]. However, unlike protein degradation induced by ubiquitination,
sumoylation plays an important role in regulating the function or activity of target proteins, controlling
intracellular localizations, enhancing the stability of proteins and inducing cell proliferation or apoptosis
[16]. Sumoylation is also a dynamic and reversible process regulated by the human sentrin sumo
-specific
protease
(SENP
) family. SENPs remove SUMOs from their substrates by hydrolysis of the linkage, which
is a process known as desumoylation [17
-19]. Notably, recent studies have found that the protease activity
of SENPs in human cervical cancer cells can be inhibited by hypoxia, and that hypoxia can upregulate
-1α isn’t instantly degraded through the pVHL
-mediated ubiquitin
-proteasome pathway under
hypoxic conditions, and with Hif-1β, it forms a stable heterodimeric transcription factor,
upregulate TGF
-β expression and activate the TGF
-β/SMAD signaling pathway [22]. However, current
studies only shed light on elevated protein levels involved in the development of keloid formation and
disease progression. In this study, we hypothesized that sumoylation of HIF
-1α may increase its stability
or enhance its transcriptional activity to promote keloid occurrence and development of keloid
s by
amplifying the TGF
-β/SMAD and HIF
-1 signaling pathway
s, which has not yet been reported. Thus, this
study aims to investigate the molecular mechanism of keloid formation from the perspective of qualitative
changes in the activity or functional enhancement of related proteins or factors in keloid
s. The results of
this study can not only clarify the role and mechanism of sumoylation in the formation of keloid
s, but also
provide clues for the clinical treatment of keloid
s and drug development, which is highly innovative and
medically meaningful.
2. Materials and methods
2.1 Cell isolation and culture
Human foreskin fibroblasts were purchased from the American Type Culture Collection
(ATCC® SCRC
-1041). Human keloid fibroblasts were isolated from surgical specimens of patients
undergoing plastic surgery to repair auricle keloids. Tissues were digested by collagenase type I (Solarbio,
China) and trypsin (Gibco, USA) in succession at 37°C for 15 min. Cells were cultured in DMEM
(Dulbecco
’s modified Eagle
’s medium; BI, Israel) supplemented with 12% FBS (fetal bovine serum;
Gibco, USA), 100 U/mL penicillin and 100 mg/ml streptomycin (Gibco, USA), and they were maintained
at 5% CO
2 at 37°C
Fibroblasts at passages four to six were utilized in the following studies unless
otherwise indicated.
2.2 Patients and tissue specimens
Journal Pre-proof
Journal Pre-proof 3
Human specimens including normal skin tissue specimens and keloid specimens were obtained from
patients who underwent repair surgery which was performed at the Department of Plastic Surgery, the
First Affiliated Hospital of Zhejiang University from 2017 to 2019. All patients in this study provided
informed consent for surgery prior to surgery. All procedures used in this study were approved by the
hospital ethics committee. The patients’ information is displayed in Supplementary Table
2.3 Hypoxia treatment and a sumoylation inhibitor
Hypoxia treatment was induced by cell culture in a hypoxia incubator (Thermo Fisher
Scientific,
USA) maintained at 37°C in a humid atmosphere and a low oxygen environment (1% O2, 5% CO2, 94%
N2). The sumoylation inhibitor, 2
-D08 (Medchemexpress, China) was dissolved in DMSO (dimethyl
sulfoxide) to make a solution with a final working concentration of 20 μM
2.4 Immunohistochemistry
Tissue specimens resected from patients were immediately fixed in 10% formalin for 24 h. After
embedding in paraffin, tissue sections (3 mm thick) were cut. After dewaxing with xylene, slides were
rehydrated in ethanol and blocked with 5% bovine serum albumin solution. Tissue sections were
incubated at 4°C overnight with primary antibodies (anti
-SUMO1 (1:250, ab32058, Abcam, USA),
anti
-SUMO2/3 (1:500, ab3742, Abcam, USA), anti
-SENP1 (1:300, ab108981, Abcam, USA), anti
-SENP3
(1:300, ab124790, Abcam, USA) and anti
-SENP6 (1:300, ab77619, Abcam, USA). Horseradish
peroxidase
-conjugated goat anti
-rabbit/mouse IgG (DAKO, Denmark) was used as a secondary antibody
and was incubated with the sections at room temperature for 1 h. Tissue sections were then stained with
diaminobenzidine solution (DAKO, Denmark). Finally
, tissue sections were counterstained in
hematoxylin and visualized under the microscope (ZEISS, Germany). IHC score
were computed by
multiplying staining intensity grade (score
s of 0, 1, 2 or 3 represented no
-staining, yellow, tan or
yellowish brown, respectively) by the proportion of positive cells score (score
s of 0, 1, 2, 3 or 4 indicate
the percentage of positive cells to b
-5%, 6
-25%, 26
-50%, 51
-75%, 76
-100%, respectively)
2.5 Immunofluorescence
Cells were cultured on sterile glass slips in six
-well plates. After three immersions in PBS
(phosphate buffered saline), the cells were fixed with a 4% paraformaldehyde solution (Servicebio, China)
for 15 min and they were permeabilized in 0.1% Triton X
-100 for 20 min. Then the cells were blocked
with 5% BSA (bovine serum albumin, BBI, China) before being incubated with primary antibodies
(anti
-SUMO1 (1:250, ab32058, Abcam), anti
-SUMO2/3 (1:100, ab3742, Abcam), anti
-SENP1 (1:100,
ab108981, Abcam) and anti
-HIF1α (1:100, ab51608, Abcam) at 4°C overnight. Goat
anti
rabbit IgG H&L
Alexa Fluor® 488(1:500, ab150077, Abcam) was used as a secondary antibody and was incubated with
cells for 1 h in the dark. Nucle
i were stained with DAPI (dimercapto
-phenylindole; Servicebio, China)
by incubation in the dark for 5 min. Finally, cell images were obtained using an inverted fluorescence
microscope (Olympus, Japan) and protein localization was observed.
2.6 Protein extraction and
western blotting
Whole cell lysates were prepared using RIPA (
radio
immunoprecipitation
assay) lysis buffer
(Beyotime, China) supplemented with protease inhibitor cocktails (Servicebio, China). After
ultrasound
-mediated cytolysis, cell lysates were centrifuged at 14,000
g in a
4°C centrifuge for 10 min. A
BCA
protein
assay
kit (Thermo
Scientific, USA) was used to quantify the total protein content. The same
amount of protein for each sample was separated in 10% polyacrylamide gels for SDS
-PAGE, and the
Journal Pre-proof
Journal Pre-proof 4
separated protein mixtures were then transferred to PVDF (polyvinylidene difluoride) membane
(Millipore, USA). The PVDF membranes were blocked with 5% nonfat powdered milk (BBI, China) in
TBS (buffered saline) containing 0.1% Tween
-20 (TBST; BBI, China) at room temperature for 1 h. The
following primary antibodies were incubated with the membranes at
4°C overnight: anti
-SUMO1 (1:1000,
ab32058, Abcam)
; anti
-SUMO2/3(1:1000, ab3742, Abcam); anti
-HIF1α (1:500, ab51608, Abcam);
anti
SENP1 (1:1000, ab108981, Abcam);anti
-RanGAP1(1:1000, ab92360, Abcam); anti
-Smad2(1:1000,
ab33875, Abcam); anti
-Smad3 (1:1000, #9513, CST); anti
-tubulin (1:5000,
E021040, Earthox). After washing three times with TBST, the membranes were incubated with HRP conjugated goat anti-rabbit IgG (H+L) (1:5000, A21020, Abbkine) at room temperature for 1 h. The bands
were observed using a Super Lumia ECL plus HRP Substrate Kit (Abbkine, China). All images were
scanned and analyzed using Image Lab software (BioRad, USA). Protein expression levels were
determined by analyzing grayscale value
s and normalizing them to standard protein levels.
2.7 RNA extraction and quantitative
real
-time
polymerase
chain
reaction
Total mRNA was extracted using a Quick
-RNA Purification Kit (YISHAN, China). The cDNA was
reverse transcribed using the PrimeScriptTM RT reagent Kit (Takara, Japan) following the standard
procedure. Real time PCR was carried out in a CFX96TM Real
-Time System (BioRad, USA) using
SYBR® Premix Ex Taq™ II (Takara, Japan) according to the manufacturer’s instructions. Gene
-specific
primer sequences are listed in Supplementary Table
-2. The transcription level of the target gene was
normalized to the level of GAPDH and the relative expression of mRNA was measured using the 2
-ΔΔCt
method.
2.8 Lentiviral and RNA interference
The lentiviral vectors encoding SUMO1, HIF
-1α and the negative control were constructed,
packaged, purified and titrated at Bio link Co Ltd. HFF
s and HKF
s were seeded on a six
-well plate and
infected at MOI (multiplicity of infection) of 30 for 12 h, and the infections were supplemented with 3
mg/ml polybrene (Solarbio, China). After 72 h, the overexpression efficiency was evaluated by
fluorescence microscop
y and western blotting analysis. Small interfering RNAs (siRNAs) were designed
and synthesized by GenePharma Co Ltd. (Shanghai, China). Each gene was interfered with by three
effective siRNAs to reduce off-target effects. Cells were transfected with siRNAs (50 nM) using
Lipofectamine 2000 Reagent (Invitrogen, USA) according to the manufacturer’s instructions. The final
efficiencies of the siRNAs were detected by qRT
-PCR and western blotting. The siRNA sequence
s are
listed in Supplementary Table
2.9 Transwell migration assays
Transfected cells were digested with trypsin (Gibco, USA) and then were resuspended in serum
-free
medium. A cell suspension containing 8×10
cells was seeded into the bottom of a transwell chamber
(Millipore, USA), and immersed in culture medium containing 12% FBS (
fetal bovine serum). After 36 h
of culture, the cells were fixed with methanol for 30 min and stained in a 0.5% crystal violet solution for
20 min. The cells were photographed and the number of cells was counted under a microscop
e (ZEISS,
Germany).
2.10 Coimmunoprecipitation (Co
-IP)
The Co
-IP of the protein of interest was tested using a Dynabeads Co
-Immunoprecipitation Kit (Life
Journal Pre-proof
Journal Pre-proof 5
Technologies, USA). Briefly, cells were lysed in immunoprecipitation buffer supplemented with 100 mM
NaCl, 2 mM MgCl2, 1 mM DTT (
dithiothreitol) and protease inhibitors. Cell lysates were then incubated
with antibody
-coupled Dynabeads and were incubated on a roller at 4°C for 30 min. Subsequently, the
Dynabeads were washed with elution buffer and analyzed by SDS
-PAGE and western blotting.
2.11 Statistical analysis
Statistical analysis was performed using SPSS 22.0 (SPSS Inc., USA). Data are expressed as the
mean ± SD (
standard deviation) and are compared between groups using Student
’s t
-test to determine the
significance of the difference. All data were obtained from at least three independent experimental
replicates. A p
-value of 5% or lower is considered to be statistically significant.
3. Results
3.1 Difference
s in the expression levels of sumoylation proteins between normal skin and keloid
tissues
Twenty surgically resected specimens were collected from patients (ten normal skin tissues vs ten
keloid tissues). Immunohistochemical staining was carried out and normal skin tissues and keloid tissues
were stained with SUMO1, SUMO2/3
, SENP1, SENP3 and SENP6. After immunohistochemical staining,
we found that the expression levels of SUMOs were higher in keloid tissue than they were in normal skin
dermis
, as shown in Figure 1A
-1D, while the SENP
s expression levels were lower in keloid tissues than
they were in normal skin tissues (Figure 1E
-1F, Figure S1A
-S1D). This suggests that sumoylation
proteins might be correlated with the formation of scar
: IHC staining reveals correlative difference
s of sumoylation proteins in normal skin tissues and
keloid tissues. (A, C)The expression level
s of SUMO1 and SUMO2/3 are significantly higher in keloid
tissues than they are in normal skin tissues (magnification x200 and x400). (E)The expression level of
SENP1 is lower in keloid tissue keloid tissues than it is in normal skin tissues. (B, D, F) IHC scores of 10
Journal Pre-proof
Journal Pre-proof 6
pairs of normal and keloid tissues (**P<0.01).
3.2 Inhibition of sumoylation can lead to attenuation of the TGF
-β/SMAD and HIF
-1 signaling
pathway
It is widely known that the TGF
-β/SMAD signaling pathway is a classic signaling pathway in
collagen deposition and scar formation. To investigate whether protein sumoylation is associated with the
TGF
-β/SMAD signaling pathway, we used the inhibitor
-D08 inhibited sumoylation by preventing the transfer of a small ubiquitin
-like modifier from the
UBC9
-SUMO thioester to the substrate [23]
. After 48 h of treatment with 2
-D08 at a concentration of 20
μM, we found that the expression level of the SUMO1
-RanGAP1 complex in the experimental group was
significantly lower than it was in the DMSO group and blank control group (Figure 2A
-B). The relative
ratio of the SUMO1
-RanGAP1 complex to free SUMO1 was also significantly downregulated under
-D08 treatment conditions (Figure 2C). Both SUMO2/3 and SENP1 showed a changing trend
, but the
change was no significant (Figure 2A
-2B). The expression of related proteins Smad2, Smad3, Smad4, Col
collagen type III), and HIF
-1α in the TGF
-β/SMAD and HIF
1 signaling pathway
was shown to be
significantly inhibited in the group using 2
-D08 in HFF
s and HKF
s (Figures 2D
-2E).We knocked down
the expression of RanGAP1 in HFF
s and HKF
s and found that the intensity of the band near 80kDa(near
where SUMO1
-RanGAP1 complex would be expected) was significantly decreased, which indicated that
the band was the combination form of SUMO1 and RanGAP1(Figure S2A).The SUMO2/3
-RanGAP1
complex was also detected
, but the ratio was far lower(Figure S2B). The above results indicate that
inhibition of intracellular sumoylation can inhibit the TGF
-β/SMAD and HIF
-1 signaling pathway
Figure 2: Sumoylation inhibitor treatment attenuates the TGF
-β/SMAD and HIF
-1 signaling pathway
s in
HFF
s and HKF
s. Treatment with 2
-D08 at a concentration of 20 μM, DMSO, and blank. (A, B) The
expression level
s of sumoylation related proteins
, SUMO1, SUMO2/3, SENP1 and the
Journal Pre-proof
Journal Pre-proof 7
SUMO1
-RanGAP1 complex were measured by western blotting in HFF
s and HKF
s The histogram show
the band intensity ratio of sumoylation related protein to reference protein (*P<0.05). (C) The histogram
shows that the relative band intensity ratio of the SUMO1
-RanGAP1 complex to free SUMO1 was
significantly decreased in the
-D08 group (*P<0.05). (D, E) The expression level
s of TG
-β/SMAD and
HIF
-1 signaling pathway
-related proteins
, including Col III, Smad2, Smad3, Smad4 and HIF
-1α were
measured by western blotting in HFF
s and HKFs. The histogram shows that the band intensity ratio of the
TGF
-β/SMAD and HIF
-1 signaling related protein to reference protein was significantly decreased in the
-D08 group (*P<0.05).
3.3 Inhibition of desumoylation enhances collagen deposition and fibroblast migration. To investigate whether protein sumoylation is further related to collagen deposition and scar
formation, we enhanced the sumoylation of proteins in HFF
s and HKF
s by inhibiting SENP1
-mediated
desumoylation. The efficiency of siSENP1 was evaluated by qRT-PCR (Figure
A) and
western blotting
(Figure 3B). Figure3C shows that the expression level of Col III is apparently increased, which indirectly
indicates that protein sumoylation has a certain effect on the deposition of collagen. Meanwhile, the
increase in the expression level of the SUMO1
-RanGAP1 complex was accompanied by a decrease in
free SUMO1 expression levels. The significant increase in the relative ratio of the SUMO1
-RanGAP1
complex to SUMO1 indicated that the SUMO1
-RanGAP1 complex may be a key factor affecting the
expression of downstream proteins in the TGF
-β/SMAD and HIF
1 signaling pathway
s. Then we
conducted experiments in siSENP1 cells to inhibit sumoylation by 2
-D08. We found that the relative ratio
of the SUMO1
-RanGAP1 complex to free SUMO1 was inhibited in the siSENP1+2
-D08 group
(Figure3F). The expression level of Col III in the siSENP1+2
-D08 group was also significantly inhibited
compared to the siSENP1 group, and HIF
-1α, Smad2, Smad3 and Smad4 exhibited a similar pattern
(Figure 3G). A transwell assay was conducted to explore the relationship between sumoylation and the migration
capacity of fibroblasts. We found that siSUMO1 inhibited migration, which is contrary to the effect of
siSENP1. Furthermore, the migration capacity of fibroblasts was significantly inhibited after
cotransfection of siSUMO1 and siSENP1 (Figure
3H). In general, the above results indicate that the
inhibition of the desumoylation process by siSENP1 and the maintenance of high level
s of protein
sumoylation in the cells can effectively enhance the deposition of collagen and the migration capacity of
fibroblasts.
Journal Pre-proof
Journal Pre-proof 8
Figure
3: Inhibition of SENP1
-mediated desumoylation enhances the expression of collagen and cell
migration in HFF
s and HKF
s. (A, B) qRT-PCR and western blotting showed clear silencing efficiency of
siSENP1 in HFF
s and HKF
s. (C, D) The expression level of Col III was apparently increased after
silencing SENP1, with an increased ratio of the SUMO1
-RanGAP1 complex to free SUMO1 (*P<0.05).
The experiment of enhancement and inhibition of sumoylation was conducted by using siSENP1 and 2-D08. (E) The expression of the SUMO1-RanGAP1 complex was increased, and the expression of free
SUMO1 was decreased in the siSENP1 group, while the effect of siSENP1 was reduced by 2
-D08
treatment. (F) The histogram shows that the relative ratio of the SUMO1
-RanGAP1 complex to free
SUMO1 was significantly inhibited in the siSENP1+2
-D08 group (*P<0.05). (G) The expression level
s of
TGF
-β/SMAD and HIF
-1 signaling
-related proteins
, Col III, Smad2, Smad3, Smad4 and HIF
-1α
, were
downregulated in siSENP1+2
-D08 group compared with siSENP1 group. The histogram show
s the band
intensity ratio of TGF
-β/SMAD and HIF
-1 signaling
-related protein to reference protein (*P<0.05,
**P<0.01). (
H) Transwell assay
s revealed the relationship between sumoylation and the migration
capacity of fibroblasts. Inhibition of SUMO1 inhibited migration, while inhibition of SENP1 promoted
migration. Cosilencing of SUMO1 and SENP1 weakened the enhancement of migration caused by
Journal Pre-proof
Journal Pre-proof 9
knockdown of SENP1 (*P<0.05).
3.4 SUMO1 activates the TGF
-β/SMAD and HIF
-1 signaling pathway
SUMO1 is regarded as the major modification molecule in sumoylation.
To explore whether
SUMO1 acts as a key modifier affecting the TGF
-β/SMAD and HIF
-1 signaling pathway
s, we first used
small interfering RNA (siRNA) to knock down the expression of SUMO1 in HFF
s and HKF
s. The
silencing efficiency of siRNA was verified by qRT
-PCR (Figure 4A) and
western blotting (Figure 4B).
We found that the expression levels of Smad2, Smad3, Smad4 and Col III were significantly decreased in
HFF
s and HKF
s (Figure 4C
-4D), and the expression of HIF
-1α was also decreased compared with the
normal group. Next
, we upregulated SUMO1 levels in HKF
s by transfecting lentiviruses carrying
SUMO1 overexpression gene tagged with GFP.
Western blotting (Figure 4E) results showed that
lentiviral transduction was successful. Figure
4F showed that the expression levels of proteins in the
TGF
-β/SMAD and HIF
1 signaling pathway
s in the LV
-SUMO1 group were significantly higher than
they were in the negative control groups. These results indicate that SUMO1 positively activates the
TGF
-β/SMAD and HIF
-1 signaling pathway
Figure 4: SUMO1 regulates the TGF
-β/SMAD and HIF
-1 signaling pathway
s in fibroblasts. (A, B)
qRT-PCR and western blotting showed clear silencing efficiency of SUMO1 in HFF
s and HKF
s. (C, D)
The expression level
s of TGF
-β/SMAD and HIF
-1 signaling
-related proteins
, including Col III, Smad2,
Smad3, Smad4 and HIF
-1α
, were downregulated in the siSUMO1 group. The histogram show
s the band
intensity ratio of the TGF
-β/SMAD and HIF
-1 signaling
-related protein to reference protein (*P<0.05). (E)
Western blotting show
s efficient transfection of lentivirus containing the SUMO1. (F) Compared with the
control, the expression level
s of Col III, Smad2, Smad3, Smad4 and HIF
-1a were obviously upregulated
in the LV-SUMO1 group in HKF
3.5 Hypoxia increases the level
s of SUMOs and HIF
-1α in keloid
To further explore the possible sumoylation sites in the TGF
-β/SMAD and HIF
1 signaling pathway
we first silenced Smad3 and Smad4 with siRNAs (Figure 5A
-5B) and found that HIF
-1α and Col III were
Journal Pre-proof
Journal Pre-proof
10
significantly inhibited
, nevertheless
, there were no significant changes in sumoylation
-related proteins in
HFF
s and HFK
s (Figure 5C). In our previous studies, we found that HIF
-1α may be involved as a key
factor in the regulation of collagen deposition by the TGF
-β/SMAD signaling pathway [24]
. Therefore,
we induced enhanced expression of endogenous HIF
-1α by culturing HFF
s and HKF
s in a hypoxic
incubator for 8 h. We found that the expression of HIF
-1α and Col III were significantly improved, while
there was no difference in the expression of Smad2, Smad3 and Smad4 protein (Figure 5D). The
expression levels of SUMO1 and SUMO2/3 were significantly increased and the expression of SENP1
was decreased (Figure5E). Immunofluorescence staining indicated that SUMO1 was localized exclusively
to the nucleus, while SUMO2/3, SENP1 and HIF
-1α were mainly detected in the nucleus and cytoplasm
(Figure 5F). The expression of SUMO1 and SUMO2/3
was significantly increased and the nuclear
transfer of SUMO2/3 was obviously increased after hypoxia treatment.
Immunofluorescence double
staining for SUMO1 and HIF
-1α was performed and the results showed that SUMO1 and HIF
-1α were
colocalized in HKFs (Figure S
3). These results suggest that hypoxia can promote the expression of
SUMOs in vivo, and the enhancement of HIF
-1α sumoylation may promote its effect on TGF
-β/SMAD
signaling in HFF
s and HKF
5: Hypoxia increases the level of SUMO proteins as well as HIF
-1α
, and the SUMOs enhances its
effect on TGF
-β/SMAD signaling pathway. (A, B) qRT
-PCR showed clear knockdown of Smad3 and
Smad4 at the mRNA level following transfection with si
Smad3 and si
Smad4 transfection for 48 h. (C)
The expression level of sumoylation
-related proteins showed no significant changes after treatment with
siSmad3 and siSmad4. (D) Col III and HIF
-1α protein levels were upregulated by 8 h of hypoxia
treatment (1%
2), while the expression level
s of Smad2, Smad3 and Smad4 showed no obvious
difference after hypoxia treatment for 8 h. The histogram show
s the band intensity ratio of related protein
to reference protein (*P<0.05). (E) The expression level of sumoylation
-related proteins
, including
SUMO1 and SUMO2/3
, were significantly increased and the expression level of SENP1 was significantly
decreased after hypoxia treatment for 8 h. The histogram show
s the band intensity ratio of
sumoylation
-related protein to reference protein (*P<0.05). (F) The protein expression and intracellular
localization of SUMO1, SUMO2/3, SENP1 and HIF
-1α were detected by immunofluorescence staining
(magnification x200) under normoxia and hypoxia (1%
-1α is modified by SUMO1 and its modification sites
We next explore
d the mechanism by which HIF
-1α enhances the TGF
-β/SMAD signaling pathway.
HIF
-1α isn’t rapidly degraded by ubiquitination in anoxic environments (Figure 6A), therefore
, we
induced endogenous HIF
-1α expression by hypoxi
c culture for 8 h. SUMO1 was detected in HIF
immunoprecipitated products. The band position of this form of SUMO1 was consistent with the position
of SUMO1 in whole cell lysates (Figure 6B, upper middle), suggesting that HIF
-1α was indeed
sumoylated by SUMO1 in HKF
s. The band also showed an increasing binding of SUMO1 and HIF
-1α
after exposure to hypoxia (Figure 6B, upper right), which was consistent with the result of
immunoprecipitation with SUMO1 (Figure 6B, third). According to the prediction by SUMOsp 2.0, a tool
for the prediction of sumoylation sites, there are five potential sumoylated lysine binding sites in the
HIF
-1α amino acid sequence (360
-540), near the oxygen
-sensitive degradation domain (ODD). The
partial amino acid sequences of HIF
-1α are shown in Figure 6C. We designed two lysine (K) mutants by
replacing them with arginine to identify the SUMO modification sites in HIF
-1α (Figure 6C).
Subsequently, we transfected two mutant lentiviruses encoding the His
-tagged HIF
-1α protein and a
wild
-type lentivirus into the HKF
s. Anti
-SUMO1 immunoblotting revealed that wild
-type HIF
-1α was
obviously sumoylated, while there was a decrease in HIF
-1α sumoylation levels in the 391M and 477M
mutation groups. This indicates that Lys391 and Lys477 of HIF
-1α are sites of sumoylation in HKF
Figure 6E shows that the expression of Col III, Smad2, Smad3 and Smad4 was significantly reduced in
the 391M and 477M groups. In summary, HIF
-1α is modified by SUMO1 in keloid fibroblasts, and its
modification sites are Lys391 and Lys477. Furthermore, HIF
-1α sumoylation can promote the deposition
of collagen.
Figure 6: Hypoxia promotes the modification of HIF
1α by SUMO1 and its modification sites are
identified by amino acid substitutions. (A)The amount of ubiquitin proteins decreased
anoxic environments. (B) SUMO1 was immunoprecipitated with an anti
1α antibody in HKFs, and
hypoxia treatment promote
d the coimmunoprecipitation of SUMO1 and HIF
1α (upper, third lane). (C)
SUMOsp 2.0 was used to predict sumoylation sites near the ODD in the HIF
1α amino acid sequence.
Two mutants were designed by substituting lysine with arginine at amino acids 391 and Coimmunoprecipitation was performed using the wide type and two mutants. The intensity of
sumoylation bands was significantly decreased in the 391M and 477M groups (upper, second, third lane).
(E)The expression of Col III, Smad2, Smad3 and Smad4 was significantly reduced in the 391M and
477M groups, which indic
ated that Lys391 and Lys477 of HIF
1α have SUMO modification
s and that
mutations can affect its role in the TGF
-β/SMAD signaling pathway.
4. Discussion
The essential role of sumoylation has now been established in many biological processes [16]. SUMO
modifications play important roles in regulating pathways of cell differentiation, apoptosis, cell cycle and
stress response by modifying protein function which may include changes in activity, subcellular
localization or protection of substrates from ubiquitination [14, 16, 25]. Lysine residues not only serve as
binding sites for sumoylation, but also serve as binding sites for other ubiquitin
-like proteins [16]. Thus,
sumoylation regulates protein activity, stability and protein
-protein interactions in hundreds of signaling
pathways by direct site modification, or antagonism of binding to other posttranslational modifications.
The identification and analysis of endogenous SUMO targets in physiological processes and complex
organisms is the future prospect and research direction of sumoylation [25]. Trends indicate that a rising
number of diseases are associated with the perturbation of SUMO modification and deconjugation. Aberrant sumoylation may be a risk factor for experimental fibrosis in different preclinical models.
Previous studies have found that sumoylation of the promyelocytic leukemia (PML) protein in tumor
suppression is required for arsenic trioxide
-induced collagen synthesis in osteoblasts [26]. Meanwhile,
sumoylation plays a key role in the activation of stellate cells, leading to the initiation of alcohol
-induced
liver fibrosis, and the SUMO
-dependent pathway is a critical determinant in the development of
myocardial fibrosis [27, 28]
In the present study, we first investigated the effect of sumoylation on fibrosis in keloids. We found that
the expression levels of SUMOs were significantly increased in keloid tissues, while the expression level
of SENPs was lower in keloid tissues. The results suggest that the degree of sumoylation in keloid tissues
is higher than it is in normal tissues, and sumoylated protein might be correlated with scar formation.
Recent studies have shown that inhibiting the SUMO pathway could repress tumor
-related pathways [29].
In our study, after inhibition of sumoylation, the TGF
-β/SMAD and HIF
and HKFs were repressed to a certain extent. In addition, we enhanced sumoylation in fibroblasts by
silencing SENP1
-mediated desumoylation, and the results showed that the expression level of Col III was
significantly increased, indicating that sumoylation is associated with excessive fibrosis of keloid
s. The
results shown in Figure
3F further demonstrate the antagonistic effect of sumoylation and desumoylation
on the TGF
-β/SMAD and HIF
1 signaling pathway
s in HFF
s and HKF
s. A new study suggested that
sumoylation may be involved in the regulation of scar formation in astrocytes [30]. In summary, our
studies establish that sumoylation has a role in promoting excessive fibrosis and keloid formation.
Previous studies have shown that the SUMO pathway targets various components of the TGF
-β/SMAD
signaling pathway [31]. TGFβ receptor type
I (TβR I) has been shown to be sumoylated, resulting in
enhanced activation and regulation of the downstream Smad signaling pathway [32]. Smad nuclear
interacting protein
-1 (SNIP1), a transcription repressor of the TGF
-β/SMAD signaling pathway, is
modified by SUMOs, leading to attenuation of its inhibitory effect on signaling pathways [33]. Smad3
and Smad4 have also been studied as substrates for sumoylation. Imoto et al. suggested that Smad3 was
SUMO modified by SUMO E3 ligase in vivo sumoylation assays [34]. Lin et al. found that
overexpression of SUMO1 and Ubc9 promoted Smad4
-dependent TGFβ
-induced transcription in human
breast and colon cancer cells [35]. Interestingly, sumoylation has been reported to counteract the ability of
Smad4 to mediate TGFβ signaling in developing xenopus embryos [36, 37]. In general, the ability of
SUMOs to promote or inhibit SMAD protein
-mediated TGFβ signaling may be related to factors such as
the cell type or context [31]. We silenced the expression of SUMO1 with a small interfering RNA and
found that the expression of Smads were decreased in HFF
s and HKF
s. Then, we used lentivirus to
overexpress intracellular SUMO1
, and the expression of Smad
s in fibroblasts was significantly increased.
These results indicate that the content of SUMO1 is positively correlated with the expression of SMAD
protein in HFF
s and HKF
s, and changes in the expression of HIF
-1α and Col III indicate that SUMO1
promotes the TGF
-β/SMAD and HIF
1 signaling pathway
s to some extent.
We also detected the expression level of the SUMO1
-RanGAP1 complex. RanGAP1 is the first
documented substrate and is one of the most prominent SUMO substrates to date for conjugation with
SUMO1 [38]. The Ran family of proteins play
s an important role in nucleocytoplasmic transport. It is
known that after the formation of E1
-SUMO thioester adducts in eukaryotes, SUMOs are transferred to
UBC9
, and the complex is subsequently ligated to E3 ligases or directly conjugated to a protein target
[14].
Nucleoporin RanBP2 catalyzes E3 ligase activity and forms a stable complex with SUMO
-modified
RanGAP1 and UBC9 in the nuclear pore complex [39]. Previous studies have shown that RanGAP1 and
UBC9 interact more extensively through the surface outside the consensus motif than other known
substrates, resulting in increased binding and more efficient SUMO transfer [40]. Recent studies on the
RanBP2/RanGAP1/SUMO1/Ubc9 complex have not only shown that sumoylation can fine
-tune the
transport machinery, but also have found that the transport machinery can recruit proteins as substrates for
sumoylation [38]. Therefore, we chose the expression level of the SUMO1
-RanGAP1 complex as a
quantitative indicator of intracellular conjugated SUMO1 in keloid fibroblasts. Our results indicate that
after inhibition of SENP1
-mediat
ed desumoylation, the expression level of the SUMO1
-RanGAP1
complex was significantly increased
, accompanied by a decrease in free SUMO1 expression level. While
the expression of Col III was apparently increased, this was consistent with the increasing trend of the
ratio of SUMO1
-RanGAP1 complex to free SUMO1. In the experiments of sumoylation and
desumoylation, the changing trend of the ratio of SUMO1
-RanGAP1 complex to free SUMO1 was
consistent with the trend in Smads expression in HFF
s and HKF
s. Therefore, we propose that the ratio of
the SUMO1
-RanGAP1 complex to free SUMO1 is an indicator that reflect
s the content of conjugated
SUMO1, and to some extent reflect the degree of intracellular sumoylation. Our results further
demonstrate that the effects of sumoylation on proteins or factors are more related to qualitative changes
in activity or functions rather than quantitative changes.
In our previous study, we found that hypoxia promotes TGFβ/SMAD signaling and collagen deposition
through HIF
-1α [24]. Several studies have demonstrated that under hypoxic conditions, HIF
-1α does not
undergo rapid enzymatic degradation by ubiquitination and forms stable heterodimeric transcription
factors [10, 11]. Our results indicated that the expression of HIF
-1α was significantly increased after
hypoxia, and it is accompanied by an increase in SUMOs and a decrease in SENP1. Meanwhile, the
expression and nuclear transfer of SUMOs increased significantly after hypoxia treatment. These results
suggest that hypoxia may promote sumoylation in HFF
s and HKF
s. Given that keloid tissues are in a
e, we suggest that sumoylation of HIF
-1α may have increased stability
and an amplified effect on TGFβ/SMAD signaling. In recent years, several studies have shown that
HIF
-1α is modified by sumoylation, but its role is still uncertain [41]. Li et al. proposed that HIF
-1α
sumoylation enhanced HIF
-1α activity to promote angiogenesis in hepatocellular carcinoma [42], but
Kang et al. found sumoylation negatively regulated the stability and transcriptional activity of HIF
-1α and
downregulate
d angiogenesis in colorectal cancer [43]. Differences can be explained by the fact that
sumoylation is regulated by a cascade of enzymes that are unique based on their targets, and these
enzymes play a leading role requires further exploration [44]
We performed coimmunoprecipitation assays to verify HIF
-1α was modified by SUMO1 in HKF
s and
that hypoxia may promote the modification of HIF
-1α by SUMO1. SUMO1 has been shown to modify
HIF
-1α in various cells, such as He
La cells [45], SKov3 cells [46], and primary brain and cardiac cells of
C57BL/6 mice [47]. Subsequently, we investigated the sumoylation site in HIF
-1α. The tetrapeptide motif
-D/E is present in most SUMO
-modified targets, where ψ is a hydrophobic residue, K is the lysine
conjugated to SUMO, x is any amino acid, and D or E is an acidic residue [25]. It has been shown that
sumoylation act
s antagonistically toward ubiquitination by enhancing the stability of HIF
-1α [45, 48].
There are many potential sumoylated lysine binding sites in the amino acid sequence of HIF
-1α. We
designed two lysine mutants (K391 and K477) by replacing lysine with arginine and transfected these
HIF
-1α mutants into cells. The results indicated that K391 and K477 are
sumoylation sites of HIF
-1α in
fibroblasts, which is consistent with previous studies [45]. We also studied the effect of sumoylation sites
of HIF
-1α on TGFβ/SMAD signaling, and the results may offer new therapeutic strategies for keloid
5. Conclusion
This study shows the mechanism of sumoylation on the TGF
-β/SMAD and HIF
-1 signaling pathways
in keloid
s. Sumoylation positively regulates keloid formation by amplifying the effects of the
TGF
-β/SMAD and HIF
-1 signaling pathways. The level of small ubiquitin
-like modifiers, especially
SUMO1, increase significantly in keloid
s. SUMO1 maintains high levels of SMAD and HIF protein
sumoylation in keloids, enabling both TGF
-β/SMAD and HIF
-1 signaling pathways to maintain a long
and stable effect. Furthermore, hypoxia promotes sumoylation in keloid
s and HIF
-1α is covalently
modified by SUMO1 at Lys 391 and Ly
s 477, which competitively bind to ubiquitin to block
ubiquitination and improve protein stability. We elucidated the role and molecular mechanism of
sumoylation in the formation of keloid
s, providing a new perspective for the potential therapeutic target
of keloid
Acknowledgments
This work was supported by National Natural Science Foundation of China (81873937).
Declaration of competing interest
The authors declare that they have no conflicts of interest with the contents of this article.
References
1. PA P, JK B, KP Y (2012): Treatment outcomes for keloid scar management in the pediatric burn population.
Burns : journal of the International Society for Burn Injuries 38, 767
2. Wong VW, Gurtner GC, Longaker MT (2013): Wound healing: a paradigm for regeneration. Mayo Clinic
3. LM S, KH W, YC L (2014): Keloid incidence in Asian people and its comorbidity with other fibrosis
-related
diseases: a nationwide population
-based study. Archives of dermatological research 306, 803
-A, S M, JM T, CL K, SP D (2006): Keloid pathogenesis and treatment. Plastic and reconstructive surgery
5. Butler PD, Longaker MT, Yang GP (2008): Current progress in keloid research and treatment. J. Am. Coll.
Surg. 206, 731
6. SM C, A P, MA F, VM dB, RV L (2015): Assesment of the treatment of earlobe keloids with triamcinolone
injections, surgical resection, and local pressure. Revista do Colegio Brasileiro de Cirurgioes 42, 9
7. P J, R B, J D, M G, FA S, P N, J B, D D (2016): Perioperative Interstitial High
-Dose
-Rate Brachytherapy for the
Treatment of Recurrent Keloids: Feasibility and Early Results. International journal of radiation oncology,
biology, physics 94, 532
8. S S, RM S, L P, S K, M L, P B (2010): Oxygen in acute and chronic wound healing. The British journal of
dermatology 163, 257
9. GG G, HC K, T P, T R, MG J (2011): Hypertrophic scarring and keloids: pathomechanisms and current and
emerging treatment strategies. Molecular medicine (Cambridge, Mass.) 17, 113
10. J S, S O, J R, CW P, PJ R, DR M (2011): High
-resolution genome
-wide mapping of HIF
-binding sites by
ChIP
11. B D, JM Z, Y W, TT L, YB D, WM X, GT L, P B, XZ S (2013): Intratumor hypoxia promotes immune tolerance
by inducing regulatory T cells via TGF
-β1 in gastric cancer. PloS one 8, e63777
12. HS JM, AK T, PA M (2014): Nitric oxide and hypoxia signaling. Vitamins and
13. R T, K U, N K, M T (2016): Central regions of keloids are severely ischaemic. Journal of plastic,
reconstructive & aesthetic surgery : JPRAS 69, e35
-F, F M (2007): Concepts in sumoylation: a decade on. Nature reviews. Molecular cell biology 8, 947
15. ZJ H, YH F, BH G, YM L, H C (2018): The post
-translational modification, SUMOylation, and cancer (Review).
International journal of oncology 52, 1081
16. RT H (2005): SUMO: a history of modification. Molecular cell 18, 1
17. D M, M D (2007): Modification in reverse: the SUMO proteases. Trends in biochemical sciences 32, 286
18. ET Y (2009): SUMOylation and De
-SUMOylation: wrestling with life’s processes. The Journal of biological
19. H L, M V (2011): Emerging roles of the SUMO pathway in development. Cellular and molecular life
sciences : CMLS 68, 4045
20. F W, F C, R S, JN W, XT W (2016): Hypoxia regulates sumoylation pathways in intervertebral disc cells:
implications for hypoxic adaptations. Osteoarthritis and cartilage 24, 1113
21. K K, K W, L M, S H, N D, S M (2016): SUMO Signaling by Hypoxic Inactivation of SUMO
-Specific
Isopeptidases. Cell reports 16, 3075
22. GN M, W L (2015): HIF
-1α pathway: role, regulation and intervention for cancer therapy. Acta
pharmaceutica Sinica. B 5, 378
23. YS K, SG K, JS S (2014): Synthesis of 2′,3′,4′
-trihydroxyflavone (2
-D08), an inhibitor of protein sumoylation.
Bioorganic & medicinal chemistry letters 24, 1094
24. X M, P Q, X S, Y C, L R, S Y, X J (2018): Hypoxia
-inducible factor
1α activates transforming growth
factor
-β1/Smad signaling and increases collagen deposition in dermal fibroblasts. Oncotarget
25. A F, F M (2013): Sumoylation: a regulatory protein modification in health and disease. Annual review of
biochemistry 82, 357
26. WX X, SZ L, D W, GF Q, J Y (2015): Sumoylation of the Tumor Suppressor Promyelocytic Leukemia Protein
Regulates Arsenic Trioxide
-Induced Collagen Synthesis in Osteoblasts. Cellular physiology and
biochemistry : international journal of experimental cellular physiology, biochemistry, and pharmacology
27. Y L, D Z, F Q, LL Z, SK L, YY L, MT L, D W, JX W, XQ D, YX L, CJ D, XQ S, BF Y, WF C (2017): Manipulating PML
SUMOylation via Silencing UBC9 and RNF4 Regulates Cardiac Fibrosis. Molecular therapy : the journal of
the American Society of Gene Therapy 25, 666
28. J L, Z W, D H, C W, Y L, T J, L C, M S, Y C, F H, X G, J Y, Y S, H W, L F, D W, S F, S W, Y S (2019): MANF inhibits
liver cancer via SUMOylation
-related suppression of NF
-κB/Snail signaling pathway and 2-D08
epithelial
-mesenchymal transition. Hepatology (Baltimore, Md.) undefined, undefined
29. JS S, A D (2017): SUMO and the robustness of cancer. Nature reviews. Cancer 17,
30. Y N, K Z (2019): Clustering analysis to identify key genes associated with motor neuron excitability
following spinal cord injury. The International journal of neuroscience 129, 856
31. A C, A S, S B (2018): The SUMO System and TGFβ Signaling Interplay in Regulation of
Epithelial
-Mesenchymal Transition: Implications for Cancer Progression. Cancers 10, undefined
32. K M, Y K, K M (2008): SUMO amplifies TGF
-beta signalling. Nature cell biology 10, 635
33. S L, J L, B Y, M Z, M X, J X, X L, XH F (2016): SUMO Modification Reverses Inhibitory Effects of Smad Nuclear
Interacting Protein
-1 in TGF
β Responses. The Journal of biological chemistry 291, 24418
-24430
34. S I, K S, R M, N S, T Y, T M (2003): Regulation of transforming growth factor
-beta signaling by protein
inhibitor of activated STAT, PIASy through Smad3. The Journal of biological chemistry
35. X L, M L, YY L, FC B, F M, XH F (2003): Activation of transforming growth factor
-beta signaling by SUMO
modification of tumor suppressor Smad4/DPC4. The Journal of biological chemistry
36. J L, G W, D H, F L (2004): Repression of Smad4 transcriptional activity by SUMO modification. The
Biochemical journal 379, 23
37. PS L, C C, D L, R D (2003): Sumoylation of Smad4, the common Smad mediator of transforming growth
factor
-beta family signaling. The Journal of biological chemistry 278, 27853
38. A F, A W (2012): The RanBP2/RanGAP1*SUMO1/Ubc9 complex: a multisubunit E3 ligase at the
intersection of sumoylation and the RanGTPase cycle. Nucleus (Austin, Tex.) 3, 429
JR G, D R, CD L (2012): Determinants of small ubiquitin
-like modifier 1 (SUMO1) protein specificity, E3
ligase, and SUMO
-RanGAP1 binding activities of nucleoporin RanBP2. The Journal of biological chemistry
40. V B
-V, DA S, MJ M, CD L (2002): Structural basis for E2
-mediated SUMO conjugation revealed by a complex
between ubiquitin
-conjugating enzyme Ubc9 and RanGAP1. Cell 108, 345
41. M W, X J (2019): The significance of SUMOylation of angiogenic factors in cancer progression. Cancer
biology & therapy 20, 130
42. J L, Y X, XD L, W W, HK J, Z M, QQ Y, LN M, AW Z, LS W, M Y, Q X, GQ C (2014): Cbx4 governs HIF
potentiate angiogenesis of hepatocellular carcinoma by its SUMO E3 ligase activity. Cancer cell 25, 118
43. X K, J L, Y Z, J Y, H Z, M C, ET Y, J C (2010): PIASy stimulates HIF1α SUMOylation and negatively regulates
HIF1α activity in response to hypoxia. Oncogene 29, 5568
44. L W, T Z, M F, N S, D W, J T, B F, H X, Q H, H L (2015): Podocytes protect glomerular endothelial cells from
hypoxic injury via deSUMOylation of HIF
-1α signaling. The international journal of biochemistry & cell
biology 58, 17
45. SH B, JW J, JA P, SH K, MK B, SJ C, KW K (2004): Sumoylation increases HIF
-1alpha stability and its
transcriptional activity. Biochemical and biophysical research communications 324, 394
46. Q A, W S, S G, L C, L H (2015): SENP1 desensitizes hypoxic ovarian cancer cells to cisplatin by up
-regulating
HIF
-1α. Scientific reports 5, 16396
47. Y J, J W, H T, G L, H Z, L L, R H, A D (2015): Increased SUMO
-1 expression in response to hypoxia: Interaction
with HIF
-1α in hypoxic pulmonary hypertension. International journal of molecular medicine 36,
-OM, E B (2013): Deciphering the emerging role of SUMO conjugation in the hypoxia
-signaling cascade.
Biological chemistry 394, 459
Author Contribution
Xiaohu Lin and Yan Jiang conducted the experiments. Jinghong Xu and Mingyuan Xu conceived the
experiments. Xiaohu Lin
Qianqian Pang and Jiaqi Sun analysis the results. Yuming Wang
、YijiaYu and
Rui Lei collected samples. XiaohuLin and Zeren Shen wrote the paper. All authors reviewed the
manuscript.
Journal Pre-proof