Insufficient SIRT1 in macrophages promotes oxidative stress and inflammation during scarring

Macrophage is a critical regulator in wound healing and scar formation, and SIRT1 is related to macrophage activation and polarization, while the specific mechanism is still unclear. To explore the specific effects of SIRT1 in scarring, we established a skin incision mouse model and LPS-induced inflammation cell model. The expression of SIRT1 in tissue and macrophage was detected, and the level of SIRT1 was changed to observe the downstream effects. LPS-induced macrophages with or without SIRT1 deficiency were used for TMT-based quantitative proteomic analysis. SIRT1 was suppressed in scar while increased in macrophages of scar tissue. And macrophages were proven to be necessary for wound healing. In the early stage of wound healing, knockout of SIRT1 in macrophage could greatly strengthen inflammation and finally promote scarring. NADH-related activities and oxidoreductase activities were differentially expressed in TMT-based quantitative proteomic analysis. We confirmed that ROS production and NOX2 level were elevated after LPS stimulation while the Nrf2 pathway and the downstream proteins, such as Nqo-1 and HO-1, were suppressed. In contrast, the suppression of SIRT1 strengthened this trend. The NF-κB pathway was remarkably activated compared with the control group. Insufficient increase of SIRT1 in macrophage leads to over activated oxidative stress and activates NF-κB pathways, which then promotes inflammation in wound healing and scarring. Further increasing SIRT1 in macrophages could be a promising method to alleviate scarring. SIRT1 was suppressed in scar while increased in macrophages of scar tissue. Inhibition of SIRT1 in macrophage leads to further activated oxidative stress. SIRT1 is negatively related to oxidative stress in macrophage. The elevation of SIRT1 in macrophage is insufficient during scarring. SIRT1 was suppressed in scar while increased in macrophages of scar tissue. Inhibition of SIRT1 in macrophage leads to further activated oxidative stress. SIRT1 is negatively related to oxidative stress in macrophage. The elevation of SIRT1 in macrophage is insufficient during scarring.


Introduction
Hypertrophic scar (HS) is a fibro-proliferative disorder that presents after deep cutaneous injury caused by trauma, burns, surgery, and other injuries [1,2].Approximately 100 Ting He and Xiaozhi Bai have contributed equally to this work and share the first authorship.million people suffer from hypertrophic scar-related issues in developed countries each year [3].Numerous methods, including surgery, silicone gel, laser, and pressure therapy, have been used to treat hypertrophic scar, but they have not been satisfying.It is important to accelerate wound healing to lessen the formation of hypertrophic scars [4][5][6].Wound healing is a complex dynamic process involving three successive stages: inflammation, cell proliferation, and matrix remodeling [7,8].Disruption of these physiological processes can result in delayed wound closure and subsequently pathological scar formation [9].The inflammation process starts immediately after injury and affects the whole process of wound healing and scar formation [9].Approximately 3 days after injury, monocytes are recruited to the injury site and differentiate into macrophages to support wound healing.Research has shown that during the early stages of wound healing, M1 is the dominant phenotype in tissue, while 5-7 days after injury, only 15-20% of macrophages have an M1 phenotype, and the wound is primarily populated by cells with an M2 phenotype [10,11], along with a decreased level of proinflammatory cytokines such as TNFα, IL-1β and IL-6 [12,13].Our former studies also confirmed that macrophages played a crucial role in the early stage of wound healing and affected scar formation, while a change in SIRT1 level was keenly related to this process [14,15].SIRT1 is a NAD + -dependent deacetylase.It has also been reported to be one of the key factors during the inflammation and infection in several inflammatory-related diseases [16][17][18].SIRT1 is also a key regulator of macrophage self-renewal, related to macrophage polarization and inflammation [19].Therefore, we are curious about the relationship between SIRT1, inflammation, and scar formation.

Scar tissue and normal skin tissue acquirement
Hypertrophic scar (HS) and paired normal skin (NS) tissues contiguous to scar were obtained from adult patients who suffered from hypertrophic scar and planned to receive scar resection in our department.All protocols have been approved by the Ethics Committee of Xijing Hospital, and written content was acquired from patients or their legal representatives.(See "Declarations").
The collected samples were divided into three parts.One piece was preserved in a 4% paraformaldehyde solution for histopathological study.Two pieces were soaked in liquid nitrogen for other experiments.

Animal models establish
Eight-week-old myeloid-specific sirt1 knockout mice (sirt1 −/− mice) and littermate wild-type (WT) mice weighing 20-5 g were purchased from the Experimental Animal Center of Air Force Medical University.Mice were genotyped by performing PCR on tail-derived DNA.All mice were housed under pathogen-free conditions and provided access to standard mouse food and water ad libitum.
To establish an animal incision model, mice were anesthetized with isoflurane inhalation and shaved hair on the back.Then, a 1 × 1 cm skin in the middle of the back was removed.
WT mice were divided into two groups.In the CEL group, clodronate liposomes (CELs; liposoma) were used to deplete macrophages.According to the manufacturer's instructions, mice were injected with CELs (100 mL) or saline (100 mL) through the tail vein three times (24 h before surgery, 24 h, and 72 h after surgery).Wounds were recorded at 0 day, 1 day, 3 days, 5 days, 7 days, 9 days, and 11 days after surgery, and wound tissue was acquired at 11 days after surgery.
Seventy-two hours after surgery, half of sirt1 −/− mice and the same amount of WT mice were sacrificed.Blood and tissue around the wound were acquired.Other mice were sacrificed 14 days after surgery, and peritoneal macrophages and tissue after wound healing were collected.

Immunohistochemical staining and immunofluorescence staining
Immunohistochemical staining was performed as previously reported.In brief, human skin, hypertrophic scar tissues, and mouse wound tissues fixed in 10% buffered formalin were embedded in paraffin blocks and cut into 4-μm-thick tissue sections.Then, the processed tissue sections were dewaxed and treated with 3% hydrogen peroxide for 15 min, followed by blocking with goat serum for 30 min, and incubated at 4 °C overnight with primary monoclonal antibodies against SIRT1 (1:100, ab110304, Abcam) and Col 1 (1:100, ab88147, Abcam), immunostaining using an SP-9000 HistostainTM kit (Beijing Zhongshan Golden Bridge Biotechnology Co.) according to the manufacturer's instructions.Briefly, tissue sections were incubated with a biotinylated secondary antibody, treated with streptavidin-biotin-HRP for signal amplification, and then stained with diaminobenzidine (DAB).Finally, tissue sections were counterstained with hematoxylin.
For immunofluorescence staining, tissue sections were deparaffinized with xylene, and heat-mediated Ag retrieval was performed.All the sections were incubated overnight at 4 °C with anti-CD68 (ab125212, Abcam) and anti-SIRT1 (ab110304, Abcam) primary antibodies.The next day, slides were incubated with goat anti-mouse Alexa Fluor 488 and goat anti-rabbit Alexa Fluor 555 (Life) secondary antibody at 37 °C for 1 h.DAPI was used for nuclear staining.Images were analyzed by Image-Pro Plus 6.0 system.
For macrophage immunofluorescence staining, macrophages were seeded in 24-well plates and treated with siRNA and LPS as described before.Then, macrophages were fixed by 4% paraformaldehyde for 10 min, permeabilized by 1% Triton-X, and blocked by 1% BSA for 30 min.Then, macrophages were incubated with anti-NF-κB p65 antibody (Abcam, ab16502) overnight at 4 °C and goat antirabbit Alexa Fluor 555 (Life, Eugene, OR) for 1 h at room temperature.DAPI was used for nuclear staining.Images were analyzed by Image-Pro Plus 6.0 system.

ROS measurement
OS were measured in RAW264.7 cells using 2′,7′-dichlorofluorescein diacetate (DCF).Briefly, cells were seeded in RPMI1640 and treated with si-SIRT1 or negative control for 24 h, and 100 ng/ml LPS was added for 12 h.Fresh medium containing 10 μg/mL DCF was added to the cells, incubated for 30 min, and then observed immediately with a fluorescence microscope.

Masson staining
Human skin tissue, hypertrophic scar tissue, and mouse wound tissue were used for the Masson staining using the Masson trichrome staining kit (Nanjing Biotech, Nanjing, Jiangsu, China).Paraffin-embedded tissue sections were examined for the expression and arrangement of collagen under an FSX100 microscope (OLYMPUS, Tokyo, Japan).Images were recorded digitally onto a computer and analyzed with Image-Pro Plus 6.0 system.

Real-time PCR
Total RNA was isolated from tissues and cells using RNA isolation kit (Takara, following the manufacturer's protocol.Five hundred nanograms of total RNA was revisedtranscribed by Prime ScriptTM RT reagent kit (Takara).Then, obtained cDNA was amplified by the CFX Connect real-time system (BIO-RAD), using SYBR™ Premix Ex Taq™ kit (Takara, Japan) with specific primers.The PCR conditions were set as 95 °C for 30 s, 40 cycles of 95 °C for 30 s, 60 °C for 10 s, and elongation at 72 °C for 15 s.The relative quantification of the target gene was conducted using the 2 − ∆∆Ct method.GAPDH was used to normalize cDNA input levels.The relative quantification ΔΔCt method was used for comparisons between groups.The primers specific for SIRT1, TGF-β1, collagen 1 (Col 1), collagen 3 (Col 3), TNF-α, IL-1β, IL-6, MCP1, and GAPDH were shown in Supplementary material Table 1.The fidelity of PCR was determined by melting temperature analysis.Each analyzed sample was performed in three biological replicates and repeated three times to calculate expression levels.

Macrophage treatment with siRNA
SIRT1 siRNAs and negative control siRNA were purchased from GenePharma Co., Ltd.RAW264.7 cells were seeded in 6-well dishes (1 × 10 6 per dish) in RPMI1640 with 10% fetal bovine serum (FBS).When reaching 60% confluence, RPMI1640 without FBS were transiently transfected with siRNA (final concentration 33 nM) per dish for 6 h.Then, the medium was changed to RPMI 1640 medium with 10% FBS for another 24 h.

Subcellular localization and enrichment of gene ontology analysis
Macrophages treated with SIRT1 siRNAs and negative control siRNA were used for TMT-based quantitative proteomic analysis.Two groups of macrophages were grown to 1 × 10 6 and treated with 100 ng/ml LPS for 12 h.Then, protein from macrophages was extracted and determined with BCA kit according to the manufacturer's instructions.The protein solution was reduced with 5 mM DTT for 30 min at 56 °C and alkylated with 11 mM iodoacetamide for 15 min at room temperature for trypsin digestion.Then, trypsin-digested peptides were desalted with Strata X C18 (Phenomenex) and vacuum-dried.The peptides were dissolved in 0.5 M TEAB and labeled according to the TMT kit's instructions; the labeled reagent was dissolved in acetonitrile, mixed with the peptides, and then incubated at room temperature for 2 h.The labeled peptides were mixed, desalted, and vacuum-dried.Wolfpsort, a subcellular localization prediction soft, was used to predict subcellular localization.Differentially expressed proteins (DEPs) were then annotated for cellular components, biological activity, and molecular function using Gene Ontology (GO) analysis.For each category, a two-tailed Fisher's exact test was employed to test the enrichment of DEPs against all identified proteins.The GO with a corrected P-value < 0.05 is considered significant.

Statistical analysis
Data analysis was performed using the SPSS16.0statistical software.The comparison between two groups was analyzed by independent t-test, while multiple groups were analyzed by ANOVA.P < 0.05 was considered statistically significant.

SIRT1 level decreased in hypertrophic scar tissues
Hypertrophic scar and normal skin tissue were collected from patients.The Masson staining showed collagen contents were much higher in hypertrophic scar than in normal skin (Fig. 1a).The immunohistochemistry staining showed that SIRT1 expression decreased in hypertrophic scars compared with normal skin tissues (Fig. 1a).Western blot and RT-PCR confirmed the decrease of SIRT1 in HS in protein and mRNA levels (Fig. 1b-d).

SIRT1 in hypertrophic scar macrophages was increased
When we focused on macrophages, it was found that the expression of SIRT1 was higher in macrophages in hypertrophic scar than in normal tissue.As shown in Fig. 2, CD68 was stained red, while SIRT1 was stained green.When we overlapped the images, it was clear that there were more yellow spots in hypertrophic scar tissue.The ratio of macrophages was also increased in hypertrophic scar.It seemed that the positive ratio of SIRT1 in macrophages also increased (Supplementary Fig. 1).

Depletion of macrophage led to delayed wound healing, while knock out of SIRT1 in macrophages led to increased inflammation and collagen sediment after wound healing
Since macrophages were increased in hypertrophic scar, it reminded us that the depletion of macrophages might affect scar formation.Then, CEL was injected through a tail vein to deplete monocytes/macrophages specifically.However, the wound healing was significantly delayed because of CEL application (Fig. 3a).Wound tissue was collected on day 11 after surgery, and western blot showed lower levels of Col 1 and Col 3 in CEL injection mice than in control mice (Fig. 3b, c).The Masson staining confirmed that collagen sediment was also decreased after CEL injection (Fig. 3d).
Since the knockout of macrophages was harmful, the myeloid-specific sirt1 −/− mice and WT mice were used.It was found that the collagen sediment in sirt1 −/− mice was higher than in WT mice after wound healing (Fig. 4a-c).RT-PCR also showed that the mRNA levels of TGF-β1, Col 1, and Col 3 were also higher in myeloid-specific sirt1 −/− mice skin than in WT mice (Fig. 4d).The wound tissue of sirt1 −/− mice showed more significant inflammation than WT mice In sirt1 −/− mice and WT mice skin incision model, the infiltration of inflammatory cells was more significant in the wound of sirt1 −/− mice than in WT mice (Fig. 5a and 3 days after surgery).The expressions of inflammatory cytokines such as TNF-α, IL-1β, and IL-6 also increased in sirt1 −/− mice tissue.The level of MCP1, one of the critical cytokines that regulated migration and infiltration of macrophage, also increased in sirt1 −/− mice wound (Fig. 5b).

Analysis of DEPs by TMT-based quantitative proteomics
A total of 242 DEPs, including 127 upregulated proteins and 115 downregulated proteins, were identified by integrated, with the threshold of fold change rations > 1.2 or < 0.83, t-test P-value < 0.05.Half of the DEPs were in the cytoplasm (25%) and mitochondrion (22.79%) (Fig. 6a).The functional annotated biological processes (BP), cellular components (CC), and molecular function (MF) were shown in Fig. 6b.Significantly enriched MF terms showed that a large number of proteins involved in NADH-related activities and oxidoreductase activity were differentially expressed after the interference of SIRT1 in LPS-induced macrophages (Fig. 6c).

SIRT1 suppression led to enhanced oxidative stress, activated NF-κB signaling, and inflammation
Macrophages treated with SIRT1 siRNA and negative control siRNA were stimulated with 100 ng/ml LPS.It was found that using SIRT1 siRNA increased ROS production (Fig. 7a), while the expression of NRF2, HO-1, Nqo1, and NOX2 decreased (Fig. 7b, c).
As shown in Fig. 8a, b, using SIRT1 siRNAs resulted in an elevation of the optical density of the immunocytochemical staining for p65 compared with negative control.The mRNA levels of TNF-α, IL-1β, IL-6, and MCP1 were also increased with siRNA SIRT1 (Fig. 8c).

Discussion
Hypertrophic scar is the most commonly seen pathological scar in clinics.The process of inflammation during the early stage of wound healing relates to the formation of HS.There are three stages of inflammation in this process.In the very beginning, the innate immune response initiates the recruitment of key inflammatory cells to local tissue [20].An influx of macrophages plays a crucial role during this process, and then they switch to a reparative phenotype and gradually exit or eliminate [9,21].The orderly regulation of macrophages affected inflammation in wound healing and scar formation.
SIRT1, a member of the sirtuin family, is a NAD +dependent deacetylase.It is important in regulating a broad variety of cellular processes such as inflammation, mitochondria homeostasis, autophagy, DNA repair, apoptosis, oxidative stress, macrophage self-renewal, and senescence [19,[22][23][24].SIRT1 activated by resveratrol could attenuate isoproterenol-induced cardiac fibrosis by regulating endothelial-to-mesenchymal transition [25].It has also been reported that activation of the SIRT1 pathway could attenuate hyperglycemia-induced oxidative stress and fibrosis [26].Our former studies found that in fibroblasts of HS tissue, the level of SIRT1 decreased significantly [14].However, the elevation of SIRT1 could improve skin fibrosis [15].In inflammation, SIRT1 is also reported to inhibit TNFαdependent transactivation of NF-κB, limiting the expression of several proinflammatory genes such as IL-1β, iNos, and IL-6.SIRT1 is also widely reported to be related to inflammation during chronic wound healing [14][15][16][17].Our former study has revealed that SIRT1 is a promising drug target for hypertrophic scar formation, while the possible mechanism of SIRT1 in scarring is still unclear.
To clarify the possible mechanism, we first detected the expression of SIRT1 in human hypertrophic scar tissue.As shown in Fig. 1, the expression of SIRT1 decreased in hypertrophic scar tissue, which was in accordance with our former study [15].SIRT1 has been reported to decrease in hepatic Fig. 4 The myeloid-specific sirt1 knockout lead to increased collagen sediment as well as TGF-β1 expression after wound healing in mice.Sirt1 −/− mice and wild-type littermates (WT mice) were used to establish the mice skin incision model.Tissues were acquired after wound healing.a-c Representative Masson staining and immunohistochemical staining for Col 1 in sirt1 −/− mice and WT mice skin tissue after wound healing.The density of collagen was analyzed.d The mRNA expressions of TGF-β1, Col 1, and Col 3 were analyzed by RT-PCR in sirt1 −/− mice and WT mice skin tissue after wound healing (n = 4).*P < 0.05, **P < 0.01.Scale bars = 50 μm tissue in middle-aged mice compared to young mice, while restoration of SIRT1 in the liver ameliorated liver fibrosis in middle-aged mice [27].In epithelial cells, the expression of SIRT1 is high.Although we do not focus on the epidermis, other researchers have reported that SIRT1 is important in skin barrier function and protecting the skin against allergen challenge.SIRT1 also promotes keratinocyte differentiation in vitro [28,29].Since SIRT1 is a critical mediator of selfrenewal and differentiation of macrophages [19], we were also curious about the expression changes of SIRT1 in macrophages in scar tissue.We analyzed the level of SIRT1 in macrophages in HS tissue.SIRT1 is supposed to be a critical factor in attenuating scarring, so it was reasonable to assume that SIRT1 may also decrease macrophages in hypertrophic scar tissue.However, immunofluorescence staining showed that SIRT1 expression was elevated in macrophages (Fig. 2).
Macrophages contribute to immunity, homeostasis, and repair [12,[30][31][32].It has been reported that M1 macrophage was related to the amplification of inflammation.SIRT1 has been reported as a key regulator of macrophage self-renewal that integrates cell cycle and longevity pathways [19].There are plenty of researches that have confirmed that SIRT1 is related to macrophage-initiated inflammation.For example, the regulation of SIRT1/NF-κB signal may affect macrophage polarization in ulcerative colitis [33].SIRT1 is also reported to ameliorate septic associated-lung injury and macrophage apoptosis via inhibiting endoplasmic reticulum stress [34].The polarization direction of macrophages could reflect the prognosis of lung fibrosis [35,36].In scleroderma, the inhibition of macrophages improved the collagen sediment [35,37].It has been reported that the depletion of macrophages in the early stages of the repair process might be beneficial in reducing scar formation [21].We also tried to diminish macrophages in the wound.However, it was found that the lack of macrophage led to delayed wound healing.Eleven days after injury, wounds were almost healed in the control group mice, while in the CEL group mice, the unhealed area of the wound was significant (Fig. 3a).
It was clear that the elimination of macrophages harmed wound healing.Then, we tried to suppress the expression of SIRT1 in macrophages which may change the polarization of macrophages during inflammation.Myeloid-specific sirt1 knockout mice (sirt1 −/− mice) were used.It showed that compared with WT mice, the collagen sediment was increased in sirt1 −/− mice after wound healing without prolonging the healing time.The expression of TGF-β1 was also increased (Fig. 4d).TGF-β1 plays a predominant role in the onset and progression of fibrotic disease and hypertrophic scar.The elevation of TGF-β1 is a key characteristic of hypertrophic scar [38].In the early stage of wound healing, the lack of SIRT1 in macrophage led to increased infiltration and pro-inflammation cytokines, including TNFα, IL-1β and IL-6 expression in wound tissue (Fig. 5b), which suggested that the transition from M1 to M2 macrophages was partially blocked.The lack of SIRT1 also led to increasing of MCP1.MCP1 plays a key role in the migration and infiltration of inflammatory cells like macrophages and other cytokines at the site of inflammation.This may explain the lack of SIRT1 led to enhanced infiltration of macrophages in tissue (Fig. 5a).That was to say, the lack of SIRT1 in macrophages could lead to increased inflammation in the early stage of wound healing and increased collagen sediment and fibrotic after wound healing.Other researchers also have reported that activation of SIRT1 in macrophages effectively countered the pro-inflammation induced by LPS in human neural stem cells [39].To further explain the regulating mechanism of SIRT1 in macrophages, TMT-based quantitative proteomics was used.LPS stimulation was used to mimic the inflammation environment.It was found that a lack of SIRT1 changed the expression of cytoplasm proteins and mitochondrial proteins during inflammation.MF analysis showed that mitochondrial oxidoreductases changed significantly.During this, NADHrelated enzyme activities were broadly involved (Fig. 6).NAD + and NADH are coenzymes that provide oxidoreductive power for the generation of ATP by mitochondria.NAD + is the precursor for NADP and NADPH, which preserve cells from reactive oxygen species (ROS).While NADPH oxidase 2 (NOX2) can catalyze molecular oxygen to superoxide, which is the main source of cellular ROS production.Excessive production of ROS leads to intracellular oxidative stress and inflammation [40,41].Oxidative stress plays an essential role in an inflammatory response, and mitochondrial-related genes were greatly affected in sirt1 −/− mice [42].
To confirm the results of proteomics analysis, macrophages were stimulated with LPS.The interference of SIRT1 by siRNA in macrophages led to much more ROS production than those treated with negative control (Fig. 7a).
It was found that compared with macrophages treated with negative control, NOX2 expression was significantly increased in SIRT1-suppressed macrophages, which could explain the elevation of ROS in macrophages.NRF2, Nqo-1, and HO-1 were also down-regulated with the suppression of SIRT1 (Fig. 7b, c).Nrf2 is an important antioxidant protein in the cellular defense system.Under oxidative stress, Nrf2 could translocate to the nucleus, bind to DNA promoters, and then activate the transcription of many antioxidant genes and cellular protective proteins, including Nqo1 and heme oxygenase (HO-1) [43,44].HO-1 can exert protective effects by reducing ROS production and polarizing macrophages to an anti-inflammatory M2 phenotype.Increased HO-1 could interact with TNF-α and NF-κB signaling pathways and exert anti-inflammatory effects [45,46].This study found that the interference of SIRT1 in LPS-induced macrophage led to increased ROS and NOX2 levels.The anti-oxidative pathway of Nrf2 was suppressed.As shown in Figs.7 and 8, with the interference of SIRT1, the protein level of Nrf2, HO-1, and Nqo1 decreased significantly.The nuclear translocation of p65 in LPS-induced macrophage was also enhanced with the suppression of SIRT1, and

Conclusions
SIRT1 in macrophages was upregulated in hypertrophic scar tissue, while insufficient elevation of SIRT1 in macrophages promoted scarring after wound healing.This process was related to increased intracellular oxidative stress and suppressed Nrf2 pathway.The over-active oxidative stress leads to the production of ROS and increased M1 phenotype in tissue.Further up-regulation of SIRT1 in macrophages could be a promising method to reduce oxidative stress and inhibit scarring.

Fig. 2 Fig. 3
Fig. 2 The expression of SIRT1 in macrophages of HS and NS tissue.Representative immunocytochemical staining for CD68 and SIRT1 in NS and HS.(n = 4)

Fig. 8
Fig. 8 SIRT1 suppression in LPS-induced macrophage led to increased p65 expression as well as inflammatory-related cytokines.a, b Macrophages were treated with SIRT1 siRNA or negative control and then stimulated by 100 ng/ml LPS or control (NT).Then,