HSF1 Attenuates the Release of Inflammatory Cytokines Induced by Lipopolysaccharide through Transcriptional Regulation of Atg10

ABSTRACT

Autophagy plays an important role in endotoxemic mice, and heat shock factor 1 (HSF1) plays a crucial protective role in endotoxemic mice. However, the protective mechanisms of HSF1 are poorly understood. In this text, bioinformatics analysis, chromatin immunoprecipitation, and electrophoresis mobility shift assay were employed to investigate the underlying mechanisms. The results showed that the release of inflammatory cytokines increased and autophagy decreased significantly in Hsf1^−/− endotoxemic mice compared with those in Hsf1^+/+ endotoxemic mice. HSF1 could directly bind to the noncoding promoter region of the autophagy-related gene 10 (Atg10). The expression of ATG10 and the ratio of LC3-II/LC3-I were obviously decreased in LPS-treated Hsf1^−/− peritoneal macrophages (PM) versus those in LPS-treated Hsf1^+/+ PM. Overexpression of HSF1 increased the level of the ATG10 protein and enhanced the ratio of LC3-II/LC3-I in RAW264.7 cells. In contrast, silencing of HSF1 decreased the expression of ATG10 and markedly lowered the ratio of LC3-II/LC3-I. In a cotransfected cell experiment, the upregulation of autophagy by overexpression HSF1 was reversed by small interfering RNA (siRNA)-ATG10. Compared with the overexpression HSF1, the release of inflammatory cytokines induced by lipopolysaccharide (LPS) was decreased in pcDNA3.1-HSF1 with siRNA-ATG10 cotransfected RAW264.7 cells. On the other hand, the decrease of autophagy by siRNA-HSF1 was compensated by overexpression of ATG10. Compared with siRNA-HSF1, the release of inflammatory cytokines induced by LPS was increased in siRNA-HSF1 with pcDNA3.1-ATG10 cotransfected RAW264.7 cells. These results presented a novel mechanism that HSF1 attenuated the release of inflammatory cytokines induced by LPS through transcriptional regulation of Atg10. Targeting of HSF1-Atg10-autophagy might be an attractive strategy in endotoxemia therapeutics.

IMPORTANCE HSF1 plays an important protective role in endotoxemic mice. However, the protective mechanisms of HSF1 are poorly understood. In the present study, we demonstrated that HSF1 upregulated ATG10 through specifically binding Atg10 promoter’s noncoding region in LPS-treated PM and RAW264.7 cells. By depletion of HSF1, the expression of ATG10 was significantly decreased, leading to aggravate releasing of inflammatory cytokines in LPS-treated RAW264.7 cells. These findings provided a new mechanism of HSF1 in endotoxemic mice.

INTRODUCTION

Sepsis, a kind of inflammation caused by infection, can develop into life-threatening bacteremia and bacterial-driven endotoxemia (1). Most of the therapeutics targeting sepsis have failed in the clinics (2). We are in an urgent need for a multitargeting therapeutic approach for sepsis.

Lipopolysaccharide (LPS) is a component of bacteria that can induce immunogenic responses, which results in advanced sepsis in organs after the exposure of immune cells to such bacterial components (3, 4). Endotoxemia caused by LPS is marked by whole-body activation of the inflammatory responses, which can lead to multiple-organ failure and shock (5). A response to endotoxemia is the release of proinflammatory cytokines from its main source, visceral adipose tissue, which is associated with disease mortality (6, 7). Thus, attenuating the release of proinflammatory cytokines may be a crucial strategy in the treatment of endotoxemia.

As a transcription factor of heat shock proteins (HSPs), heat shock factor 1 (HSF1) plays a key role in combating endotoxemia, including inhibition of proinflammatory cytokine genes, such as tumor necrosis factor-α (TNF-α), interleukin 1 (IL-1), and IL-6 (8, 9). HSF1 can bind directly to the promoter of these inflammatory cytokines. Previous studies have shown that HSF1 inhibits the expression of TNF-α through direct binding to the TNF-α promoter (10). HSF1 represses the expression of IL-1β through physical interaction with the nuclear factor for IL-6 (NF-IL-6), which is an activator of IL-1β (11). HSF1 inhibits the expression of IL-6 by inducing the activation of transcription factor 3 (ATF3) (12), a negative regulator of inflammatory cytokines, including IL-6 (13).

HSF1 also reduces systemic markers of organ dysfunction, including alanine aminotransferase, aspartate aminotransferase, lactate dehydrogenase, and blood urea nitrogen (14). Studies indicate that HSF1 is essential for the prevention of systemic inflammation and tissue damage caused by bacterial endotoxins or sepsis. HSF1 could negatively regulate HMGB1 and be involved in regulating asthma (15). HSF1 plays an indispensable role in the initiation, promotion, and progression of cancer (16). Therefore, HSF1 operates in a multifaceted manner, expanding far beyond the heat shock response (HSR).

Autophagy is a dynamic process of a cell where biomacromolecules and damaged organelles are engulfed in an autophagic vesicle that merges with a lysosome and forms an autolysosome. Autolysosomes degrade the contents of the vesicle, such as organelles and proteins, through a variety of bioenzymes from the lysosome (17). As a cytoplasmic degradation pathway, autophagy protects cells from both exogenous hazards, such as infection and endogenous inflammation. In any type of cells, the cytoplasmic cleansing function of autophagy is by default anti-inflammatory and capable of activating cell-autonomous inflammatory responses (18, 19). Autophagy or individual autophagy factors target key intracellular combinations and platforms of proinflammatory signaling that inhibit or degrade to limit inflammation (20). In conclusion, autophagy plays an important role in inflammation.

In the previous study, HSF1 Thr120 phosphorylation promoted proteostasis and carboplatin-induced autophagy in breast tumor progression (21). Heat treatment resulted in an elevation in pAMPK (T172), Beclin-1, and LC3 II and an upregulation of HSF1 (22). However, the mechanisms of HSF1 in the process of autophagy remain to be fully characterized. In this paper, bioinformatics, chromatin immunoprecipitation assay, electrophoretic mobility shift assay, and other methods were used to explore the mechanisms of HSF1 in regulating autophagy. Our findings will benefit the development of new treatment strategies for endotoxemia.

RESULTS

HSF1 attenuated the release of inflammatory cytokines, which was associated with autophagy in endotoxemic mice.

First, Hsf1^+/+ and Hsf1^−/− endotoxemic mice were established by intraperitoneal injection of LPS, and the histomorphology of the liver and lung in mice was examined by hematoxylin and eosin (H&E) staining. As shown in Fig. 1A to C, visual damages of organs were more severe in Hsf1^−/− endotoxemic mice than those in Hsf1^+/+ endotoxemic mice. The symptoms included a large amount of inflammatory cells infiltration, substantial interstitial cellular degeneration and necrosis, and exudate blockage or rupture of capillaries in the lung and liver. Then, the levels of serum inflammatory cytokines were detected. As shown in Fig. 1D to F, compared with Hsf1^+/+ endotoxemic mice, the levels of serum IL-6, IL-1β, and TNF-α were significantly higher in Hsf1^−/− endotoxemic mice. These results suggested that HSF1 attenuated the release of inflammatory cytokines in endotoxemic mice. In order to examine whether the release of inflammatory cytokines in endotoxemic mice was associated with autophagy, autophagy markers LC3-II and LC3-I were examined by Western blot. As shown in Fig. 1G and H, the ratio of LC3-II/LC3-I decreased significantly in Hsf1^−/− endotoxemic mice liver and lung compared with those in Hsf1^+/+ endotoxemic mice. These results suggest that HSF1 decreases the releasing of inflammatory cytokines and enhances LPS-induced autophagy in endotoxemic mice.

FIG 1.

HSF1-mediated attenuating the release of inflammatory cytokines was associated with autophagy. (A and B) The histomorphology in the lung (scale bar, 50 μm) and liver (scale bar, 50 μm) of Hsf1^+/+ and Hsf1^−/− endotoxemic mice by H&E staining. (C) Semiquantitative analysis about the histopathological injury of lung and livers were evaluated. **, P < 0.01 versus Hsf1^+/+ sham mice (n = 6); ^##, P < 0.01 versus Hsf1^+/+ LPS mice (n = 6). (D to F) The concentrations of IL-6, TNF-α, and IL-1β were detected in Hsf1^+/+ and Hsf1^−/− endotoxemic mice serum. **, P < 0.01 versus Hsf1^+/+ sham mice (n = 6); ^#, P < 0.05; ^##, P < 0.01 versus Hsf1^+/+ LPS mice (n = 6). (G and H) Mice were treated with LPS for 12 h, and the expressions of LC3-II and LC3-I protein were measured with anti-LC3 antibody in the liver and lung; **, P < 0.01 versus Hsf1^+/+ sham mice (n = 6); ^# P < 0.05 versus Hsf1^+/+ LPS mice (n = 6).

mRNA levels of six autophagy-related genes were detected in Hsf1^+/+ and Hsf1^−/− peritoneal macrophages.

Six autophagy-related genes Atg10, LC3, Beclin1, Atg12, Atg16l1, and Atg13, which promoter region contained HSF1 binding elements (nGAAnnTTCn or nTTCnnGAAn), were selected (Table 1). mRNA expression levels of these genes were detected by real-time PCR (RT-PCR). The results showed that the mRNA levels of Atg10, LC3, and Beclin1 decreased significantly in Hsf1^−/− peritoneal macrophages compared with those in Hsf1^+/+ peritoneal macrophages (Fig. 2A to C). However, there was no obvious difference in the expression levels of Atg12, Atg16l1, and Atg13 between Hsf1^+/+ and Hsf1^−/− peritoneal macrophages (Fig. 2D to F). These results indicated that Atg10, LC3, and Beclin1 might be involved in Hsf1^+/+ peritoneal macrophage autophagy induced by LPS.

TABLE 1.

Autophagy-related genes which the promoter region containing HSE^a

GenBank	Symbol	Description	HSE	HSE sequence	Position (transcription starting point + 1)
NM_026217	ATG12	Autophagy-related protein 12	2	ATTCAAGAAG	−1764 to −1755
				GTTCTAGAAGAGAACTTTCA	−1265 to −1246
NM_145528	ATG13	Autophagy-related protein 13	1	TTTCCCGAAT	−1646 to −1637
NM_025770	ATG10	Autophagy-related protein 10	1	AGAATATTCA	−1758 to −1749
NM_029846	ATG16 L 1	Autophagy-related protein 16-1	3	GGAACATTCC	−1950 to −1941
				GGAACTTTCC	−953 to −994
				TTTCTGGAAA	−784 to −775
NM_019584	Beclin1	Coiled-coil myosin-like BCL2-interacting protein	2	AGAAGATTCC	−1294 to −1285
				GGAAGTTTCC	−1398 to −1389
NM_025735	LC3α	Microtubule-associated protein 1 light chain 3 alpha	2	GTTCCAGAAT	−366 to −357
				CTTCTCGAAT	−142 to −133

^aHSE, heat shock element (nGAAnnTTCn or nTTCnnGAAn).

FIG 2.

mRNA levels of six autophagy-related genes were detected in Hsf1^+/+ and Hsf1^−/− peritoneal macrophages. (A to F) mRNA of autophagy-related genes Atg10, LC3, Beclin1, Atg12, Atg16l1, and Atg13 was detected by RT-PCR. *, P < 0.05 versus Hsf1^+/+ PM treated with PBS (n = 3); ^##, P < 0.01 versus Hsf1^+/+ PM treated with LPS (n = 3).

HSF1 can bind to promoter regions of Atg10.

To explore the mechanism of HSF1 regulate autophagy, a chromatin immunoprecipitation (ChIP) assay and electrophoretic mobility shift assay (EMSA) were performed. Nuclear fractions were extracted from RAW264.7 cells, and the DNAs were immunoprecipitated by anti-HSF1 monoclonal antibody and analyzed by quantitative real-time PCR (qRT-PCR). As shown in Fig. 3A and B, Atg10, LC3, and Beclin1 existed in the precipitate, and the quantity of Atg10 was the highest. However, Atg12, Atg16l1, and Atg13 were not found in the precipitate. EMSA was used to detect whether HSF1 could bind to the promoter region of autophagy-related genes in vitro. As shown in Fig. 3C, HSF1 could bind to the Atg10 promoter region in RAW264.7 cells in vitro. The stronger binding band was observed between HSF1 and Atg10 when treated with LPS (Fig. 3C, lane 6 versus lane 1). Supershifted bands of a higher molecular weight were observed when an anti-HSF1 antibody was added (Fig. 3C, lane 2 and lane 7). The binding reaction did not occur when the cold and mutant probe were added (Fig. 3C, lanes 4 and 5 versus lanes 9 and 10). These results confirmed that HSF1 specifically bound the noncoding region of the Atg10 promoter in vitro. However, no obvious binding band or supershifted band was traced between HSF1 and LC3, Beclin1, Atg12, Atg16l1, and Atg13 (data not shown).

FIG 3.

HSF1 binds to autophagy-related gene promoter region. (A and B) ChIP assays were performed using chromatin isolated from RAW264.7 cells. Antibody against HSF1 was used for immunoprecipitation (HSF1-IP). Antibody against IgG was used as a negative control (IgG-IP). The total DNA (input) was used as the positive control. Precipitates were quantified by qRT-PCR using specific primers of Atg10, LC3, Beclin1, Atg12, Atg16l1, and Atg13. *, P < 0.05; **, P < 0.01 versus IgG-IP (n = 3). (C) EMSA was performed with the nuclear extracts from untreated or treated with LPS RAW264.7 cells using an Atg10 probe. The arrows indicated the specific binding bands (shift) and supershifted bands of higher molecular weight (supershift). Biotin-labeled free probe and mutation probe sequences were detailed in the Materials and Methods section. Cold probes were not labeled with biotin, more than 200 times as many as free probes. The experiment shown was representative of three similar experiments.

HSF1 upregulated the protein expression of ATG10 and increased the ratio of LC3-II/LC3-I in peritoneal macrophages and RAW264.7 cells.

The above results suggested that HSF1 might upregulate autophagy through transcriptional regulation of the expression of Atg10. Therefore, we selected Atg10 for further study. First, primary peritoneal macrophages (PM) were obtained from Hsf1^−/− and Hsf1^+/+ mice and treated with 1,000 ng/mL of LPS for 8 h. As shown in Fig. 4A to C, compared with Hsf1^+/+ PM+phosphate-buffered saline (PBS), the expression of ATG10 and the ratio of LC3-II/LC3-I were increased in Hsf1 ^+/+ PM+LPS. However, the expression of ATG10 and the ratio of LC3-II/LC3-I were obviously decreased in Hsf1^−/− PM+LPS versus Hsf1^+/+ PM+LPS. Overexpression of HSF1 increased the expression level of ATG10 and enhanced the ratio of LC3-II/LC3-I in RAW264.7 cells treated with LPS (Fig. 4D to F). In contrast, silencing of HSF1 decreased the expression of ATG10 and markedly lowered the ratio of LC3-II/LC3-I in RAW264.7 cells treated with LPS (Fig. 4G to I). These results indicated that HSF1 enhanced LPS-induced autophagy through transcriptional regulation of the expression of ATG10 in PM and RAW264.7 cells.

FIG 4.

HSF1 upregulated the expression of the ATG10 protein and increased the ratio of LC3-II/LC3-I in peritoneal macrophages and RAW264.7 cells. (A to C) Primary peritoneal macrophages (PM) were obtained from Hsf1^−/− and Hsf1^+/+ mice and treated with 1,000 ng/mL LPS for 8 h. The expression of ATG10, LC3-II, and LC3-I proteins were measured with anti-ATG10 and anti-LC3 antibody in Hsf1^−/− and Hsf1^+/+PM; **, P < 0.01 versus Hsf1^+/+PM +PBS (n = 3); ^##, P < 0.01 versus Hsf1^+/+PM +LPS (n = 3). (D to F) RAW264.7 cells were transfected with pcDNA3.1-HSF1 or pcDNA 3.1 for 48 h and treated with LPS (1,000 ng/mL) for another 8 h. The protein levels of ATG10 and LC3 were tested by Western blotting. *, P < 0.05; **, P < 0.01 versus pcDNA3.1+PBS (n = 3); ^##, P < 0.01 versus pcDNA3.1+LPS (n = 3). (G to I) RAW264.7 cells were transfected with siRNA-HSF1 or negative control for 48 h and treated with LPS (1,000 ng/mL) for another 8 h. Then the protein levels of ATG10 and LC3 were tested by Western blotting. *, P < 0.05; **, P < 0.01 versus negative control+PBS (n = 3); ^##, P < 0.01 versus negative control+LPS (n = 3).

siRNA-ATG10 suppressed HSF1-mediated autophagy.

To investigate the role of Atg10 in the process of HSF1-mediated autophagy, immunofluorescence assays of LC3 were performed in RAW264.7 cells. PcDNA3.1-HSF1 was transfected in RAW264.7 cells. The results found that overexpression of HSF1 increased the numbers of autophagosomes/cell. In the rescue experiment, when pcDNA3.1-HSF1 and siRNA-ATG10 were cotransfected, decreased numbers of autophagosomes/cell were detected compared with only pcDNA3.1-HSF1-transfected RAW264.7 cells (Fig. 5A and B). When cotransfected siRNA-HSF1 with pcDNA3.1-ATG10, increased numbers of autophagosomes/cell were detected compared with only siRNA-HSF1-transfected RAW264.7 cells (Fig. 5C and D).

FIG 5.

siRNA-ATG10 abrogates HSF1-mediated autophagy. LPS-induced autophagy was measured by immunofluorescence detection. The cells were incubated with anti-LC3 primary antibodies. The secondary fluorescence antibodies were goat anti-rabbit Alexafluor 488 (green fluorescence). Nuclei were stained with DAPI (blue fluorescence). (A and B) RAW264.7 cells were transfected with pcDNA3.1-HSF1 or pcDNA3.1-HSF1+siRNA-ATG10. Scale bar, 10 μm. Left, shows were representative photos; right, shows quantitative analysis; at least 30 cells were calculated. *, P < 0.05; **, P < 0.01 versus pcDNA3.1+PBS; n ≥ 5. ^#, P < 0.05; ^##, P < 0.01 versus pcDNA3.1+LPS; n ≥ 5. (C and D) RAW264.7 cells were transfected with siRNA-HSF1 or siRNA-HSF1+pcDNA3.1-ATG10; scale bar, 10 μm. Left, shows representative photos; right, shows quantitative analysis; at least 30 cells were calculated. *, P < 0.05 versus negative control + PBS; n ≥ 5. ^#, P < 0.05 versus negative control + LPS; n ≥ 5.

siRNA-ATG10 negated HSF1-mediated attenuation of the inflammatory cytokine release.

In order to determine whether HSF1-mediated attenuation of the release of inflammatory cytokines is related to Atg10, inflammatory cytokines were detected in RAW264.7 cells. As shown in Fig. 6A to C, compared with the HSF1 overexpression in RAW264.7 cells, the release of the cytokines IL-6, IL-1β, and TNF-α were significantly increased (P < 0.01, n ≥ 5) in pcDNA3.1-HSF1 with siRNA-ATG10-cotransfected RAW264.7 cells treated with LPS. In contrast, compared with the siRNA-HSF1 RAW264.7 cells, the release of IL-6, IL-1β, and TNF-α were significantly decreased (P < 0.01, n ≥ 5) in siRNA-HSF1 and pcDNA3.1-ATG10 cotransfected RAW264.7 cells (Fig. 6D to F). Thus, HSF1-induced protection from endotoxemia was largely Atg10 dependent.

FIG 6.

siRNA-ATG10 abrogated HSF1-mediated attenuation of the release of inflammatory cytokines. (A to C) LPS-induced IL-6, TNF-α, and IL-1β in the culture medium were measured by ELISA. **, P < 0.01 and ***, P < 0.001 versus pcDNA3.1+LPS, n ≥ 5; ^#, P < 0.05 and ^###, P < 0.001 versus pcDNA3.1-HSF1+LPS, n ≥ 5. (D to F). LPS-induced IL-6, TNF-α, and IL-1β in the culture medium were measured by ELISA. *, P < 0.05; **, P < 0.01; ***, P < 0.001 versus negative control+LPS; n ≥ 5. ^##, P < 0.01 and ^###, P < 0.001 versus siRNA-HSF1+LPS; n ≥ 5.

DISCUSSION

HSF1 is a transcription factor that regulates eukaryotic heat shock proteins (HSPs), which are highly conserved among different species (23). HSF1 also regulates non-heat shock genes, including autophagy-related genes. For instance, HSF1 plays a key role in chemotherapeutic agent-induced cytoprotective autophagy through the transcriptional upregulation of autophagy-related gene 7 (Atg7) (24). Another report has shown that HSF1 upregulates the expression of ATG4B in hepatoma carcinoma cells by binding to the promoter region of the gene and consequently enhancing the epirubicin (EPI)-induced protective autophagy, which in turn promotes hepatocellular carcinoma (HCC) cell survival (25). Samarasinghe et al. (26) have found that sequestosome 1 (p62/SQSTM1), a protein involved in the delivery of autophagic substrates and nucleation of autophagosomes, is an HSF1-regulated gene. Du et al. (27) found that silencing HSF1 correlated with a reduction in autophagy and thus enhanced cell death. Compared with the wild-type subgroup, the expressions of LC3II/LC3I and Beclin1 were shown to have an obvious increase, and increased autophagosome formation was observed on electron microscopy in the HSF1 transgene (HSF1-TG) subgroup. Opposite results were observed in the HSF1-KO subgroup. These results suggest that HSF1 promoted autophagosome formation.

Based on the above results, we speculated that HSF1 could participate in autophagy by transcriptionally regulating the expression of autophagy-related genes. Therefore, six autophagy-related genes (Atg10, LC3, Beclin1, Atg12, Atg16l1, and Atg13) were selected, which promoter region-containing HSF1 binding elements (nGAAnnTTCn or nTTCnnGAAn). The results showed that the mRNA levels of Atg10, LC3, and Beclin1 were increased significantly in Hsf1^+/+ peritoneal macrophages compared with those in Hsf1^−/− peritoneal macrophages by RT-PCR (Fig. 2). So, we speculated that Atg10, LC3, and Beclin1 may be the transcriptional target genes of HSF1. Then, a ChIP assay was used to detect the interaction relationship between HSF1 and these genes. The ChIP assay results found that Atg10, LC3, and Beclin1 were in the precipitate and that the quantity of Atg10 was the highest. It indicated that Atg10 was mainly a transcriptional target gene of the HSF1 (Fig. 3). EMSAs were employed to investigate the transcriptional target of HSF1. In the experiment of EMSA, a shift band and supershifted band were observed in Atg10 and HSF1. However, only a shift band was observed in LC3 or Beclin1 and HSF1, and no supershifted band of higher molecular weight was observed in LC3 or Beclin1. We considered that there may be two reasons for this finding, as follows: first, an antibody to HSF1 blocks the binding site, and second, there is no specific binding between HSF1 and LC3 or Beclin1. These ideas remain to be further studied.

Based on the above finding, we selected Atg10 for further study. Further studies found that the overexpression of HSF1 increased the expression level of Atg10 and enhanced the induction of autophagy in RAW264.7 cells. In turn, silencing of HSF1 inhibited the expression of Atg10 and autophagy (Fig. 4).

Atg10 is necessary for the occurrence of autophagy, it is a ubiquitin-binding (E2) enzyme mimic that conjugates Atg12 to Atg5, and it is essential for yeast autophagy (28, 29). There is also evidence that Atg10 has a significant effect on autophagy. In Crassostrea gigas, the mRNA expression of Atg10 in the mantle of oysters was significantly upregulated after poly (I·C) stimulation. The formation of autophagosome expression was inhibited in Atg10-knockdown oysters. Atg10 was involved in the formation of autophagosome and antivirus immune response of oysters (30). In the substantia nigra of a Parkinson’s disease (PD) rat model, p62, Atg5, Atg12, LC3, and Atg16l1 were expressed, while Atg10 was not expressed. This result indicates that Atg10 expression is necessary for the initiation of autophagy in the rat model of PD (31). ATG10 also plays an important role other than autophagy in diseases. Jo et al. (32) reported that the expression of ATG10 is associated with lymph node and lymphatic vessel metastasis in colorectal cancer. Moreover, ATG10 modulated epithelial mesenchymal-associated protein depletion in colorectal cancer cells (33). The c-Myc/miR-27b-3p/ATG10 signaling pathway was involved in regulating colorectal cancer chemoresistance (34). However, the role of ATG10 in endotoxemia has not been studied.

In order to determine whether HSF1-mediated protective autophagy and protective effect are related to Atg10 in endotoxemia, cotransfection of pcDNA3.1-HSF1 with siRNA-ATG10 or siRNA-HSF1 with pcDNA3.1-ATG10 was performed in RAW264.7 cells. Compared with the HSF1 overexpression group, the numbers of autophagosomes per cell was clearly lower in pcDNA3.1-HSF1 with siRNA-ATG10-cotransfected RAW264.7 cells (Fig. 5A and B). Compared with that of the siRNA-HSF1-transfected cells, the numbers of autophagosomes per cell were clearly increased in siRNA-HSF1 with pcDNA3.1-ATG10-cotransfected RAW264.7 cells (Fig. 5C and D). Similar trends are observable in the release of the cytokines IL-6, IL-1β, and TNF-α (Fig. 6).

In this study, we provided multiple lines of evidence that HSF1 attenuated the release of inflammatory cytokines in endotoxemia mice by upregulating Atg10. First, we found that the levels of Atg10 mRNA were elevated in LPS-treated Hsf1^+/+ peritoneal macrophages. Second, HSF1 can directly bind to the promoter region of Atg10 in RAW264.7 cells. Third, overexpression of HSF1 increased the expression of ATG10 and enhanced the ratio of LC3-II/LC3-I in RAW264.7 cells. In contrast, silencing of HSF1 decreased the expression of ATG10 and markedly lowered the ratio of LC3-II/LC3-I. Finally, silencing Atg10 with siRNA negated the protective effect of HSF1 in RAW264.7 cells treated with LPS.

Thus, HSF1-induced protection from endotoxemia was largely Atg10 dependent. However, the role of Atg10 in endotoxemia other than autophagy and the transcriptome change in HSF1-deficient cells have not been studied, which will be our interesting research avenue in the future.

MATERIALS AND METHODS

Mouse model of endotoxemia.

Hsf1 knockout (Hsf1^−/−) and wild-type (Hsf1^+/+) mice were provided by Ivor J. Benjamin (University of Utah, Salt Lake City, UT) and had been described elsewhere (35). Endotoxemia was induced in mice as described previously (8). Mice homozygous for the Hsf1 knockout allele were maintained along with age-matched control littermates. Hsf1^−/− and Hsf1^+/+ mice (n = 6 to 8 per group) were injected intraperitoneally with 10 mg/kg of body weight of LPS (Escherichia coli 0111: B4; Sigma-Aldrich) diluted in pyrogen-free normal saline (NS). Hsf1^−/− and Hsf1^+/+ mice (n = 6 to 8 per group) injected with NS were used as the sham. Lungs, livers, and sera were harvested 12 h after injections. Then the mice were euthanized. The investigation followed the Guide for the Animal Care and Use Committee of Central South University. The study protocol was approved by the Ethics Committee of Central South University, Medical Institution Animal Care and Research Advisory Committee (Changsha, China).

Isolation of peritoneal macrophages and RAW264.7 cell culture.

Peritoneal macrophages (PM) were collected from Hsf1^+/+ and Hsf1^−/− mice by peritoneal lavage with Dulbecco’s phosphate-buffered saline as described previously (8). Thioglycolate-elicited peritoneal cells were obtained from animals 3 days after intraperitoneal injection of 3 mL thioglycolate broth (Sigma-Aldrich, St. Louis, MO). Macrophages were further purified by adherence to culture dishes for 24 h and identified using anti-HSF1 antibodies (Santa Cruz Biotechnology, Santa Cruz, CA) before use. After 24 h of in vitro culture, peritoneal macrophages were treated with 1,000 ng/mL of LPS for 8 h. RAW264.7 cells (mouse leukemia cells of monocyte macrophage) were purchased from the Shanghai Cell Bank of the Chinese Academy of Sciences (Shanghai, China). The cells were cultured with Dulbecco’s modified Eagle medium (DMEM) plus 10% fetal bovine serum under 5% CO₂ conditions at 37°C. RAW264.7 cells were treated with 1,000 ng/mL of LPS for 8 h.

Histopathological examination.

The livers and lungs were fixed in 10% neutral buffered formalin at room temperature for 24 h. The samples were dehydrated in ethanol of increasing concentration gradient (75%, 85%, 95%, and 100%), cleared in xylene for 2 h, soaked in paraffin for 2 h, and then embedded. Sections were cut at 4-μm thickness and placed on glass slides. To deparaffinize samples, the slides were placed in xylene and then in ethanol of a decreasing concentration gradient. Slides were stained with hematoxylin for 5 min and then with eosin for 10 s, and finally they were dehydrated by a gradient of alcohol and xylene. Then the samples were sealed with neutral balsam on slides and examined by an experienced pathologist under a light microscope (Nikon, Tokyo, Japan). Semiquantitative analysis about the histopathological injury of livers was evaluated as described previously by Mohamed et al. (36). Briefly, liver injury was graded from 0 (normal) to 4 (severe) in four categories, as follows: hepatocellular necrosis, hemorrhage, hepatic parenchymal inflammatory infiltrate, and sinusoidal inflammatory infiltrate. Semiquantitative analysis about the histopathological injury of lungs was evaluated as described previously by Yousef et al. (37). Briefly, lung injuries were defined as neutrophil infiltration, interstitial edema, congestion, hemorrhage, hyaline membrane formation, and necrosis. The severity of microscopic injury was judged according to the following scoring system: 0, normal; 1, minimal (<25%); 2, mild (25% to 50%); 3, moderate (50% to 75%); and 4, severe (>75%). Tissue sections were examined in by an experienced pathologist who was blind to treatments.

Total RNA extraction and reverse transcription-PCR analysis.

Total RNA was isolated from RAW264.7 cells using TRIzol reagent (Thermo Fisher, USA). Total RNA was used to reverse transcribe a cDNA template using a PrimeScript RT reagent kit (RR037A; TaKaRa, Dalian, China). PCRs were performed using SYBR Premix Ex TaqII (RR820A; TaKaRa, Dalian, China). The PCR assays were conducted according to the manufacturer’s instructions. The sequences are listed for each of the following genes: Atg10 left primer, 5′-GGAGAACAGCCAAGGAAT-3′; Atg10 right primer, 5′-CTCGTCACTTCAGAATCATC-3′; Beclin1 left primer, 5′-TGGATGACGAACTCAAGAG-3′; Beclin1 right primer, 5′-GATGTGGAAGGTGGCATT-3′; LC3A left primer, 5′-GAGCGAGTTGGTCAAGAT-3′; LC3A right primer, 5′-TCATAGATGTCAGCGATGG-3′; Atg12 left primer, 5′-ATCCTGCTGAAGGCTGTA-3′; Atg12 right primer, 5′-TGATGAAGTCAATGAGTCCT-3′; Atg13 left primer, 5′-GGAGATTCTATGGAGTTGGA-3′, Atg13 right primer, 5′-TTCAGCAGCAGTGACAAT-3′; Atg16l1 left primer, 5′-CGCTCTGTCTCTTCCATC-3′; and Atg16l1 right primer, 5′-ACATACGAGGCAGTAGTTG-3′.

Immunofluorescence analysis.

First, RAW264.7 cells were seeded on glass covers lips in a 24-well culture plate. Second, cells were fixed in 4% paraformaldehyde for 10 min and washed three times with cold PBS. Then, cells were incubated with the primary antibody LC3 (MBL) at a dilution of 1:1,000 for 1 h after blocking with 10% fetal bovine serum or 5% bovine serum albumin (in PBS) for 30 min. Last, the cells were incubated with a second fluorescence antibody, Alexafluor 488 goat anti-rabbit antibody (Invitrogen, Carlsbad, A), at a dilution of 1:1,000. Nuclei were stained with 4′,6-diamidino-2-phenylindole (DAPI; C1006; Beyotime Biotechnology) for 5 min at room temperature. Cells were imaged with a confocal microscope (Leica Camera AG, Oskar-Barnack-StraQe, Germany) after the glass coverslip was mounted with glycerol and at least 30 cells were counted.

Enzyme-linked immunosorbent assay (ELISA).

Cytokines (IL-6, IL-1β, and TNF-α) in the serum and cell culture medium were measured in accordance with the manufacturer’s instructions from commercially available enzyme-linked immunosorbent assay kits (Boster Biological Technology, Wuhan, Hubei, China). The experiments were repeated at least five times, and the data shown are the means ± SD.

Protein preparation and Western blot.

Proteins from the liver and lung tissues, peritoneal macrophages, and RAW264.7 cells were extracted in radioimmunoprecipitation assay (RIPA) buffer (1% Triton X-100, 10 mM Tris-HCl, 5 mM EDTA, and 150 mM NaCl [pH 7.0]) containing a protease inhibitor cocktail. Protein extracts were subjected to centrifugation at 12,000 g for 20 min. The protein concentration was determined using a bicinchoninic acid (BCA) protein assay kit (Dingguo Changsheng Biotechnology, Beijing, China). Total protein was separated on a 12% or 15% sodium dodecyl sulfate-polyacrylamide gel and transferred to an polyvinylidene fluoride (FVDF) membrane (Millipore, Billerica, MA). The membranes were then incubated overnight at 4°C with a primary antibody against LC3 (MBL, Japan) at a dilution of 1:1,000, GAPDH (Sigma-Aldrich) at a dilution of 1:1,000, HSF1 (Santa Cruz Biotechnology) at a dilution of 1:500, ATG10 (Cell Signaling Technology) at a dilution of 1:1000, or β-actin (Sigma-Aldrich) at a dilution of 1:1,000. Subsequently, the membranes were washed three times with Tris-buffered saline with Tween 20 (TBST; pH 7.6) and incubated with horseradish peroxidase-conjugated goat or rabbit IgG antibody (diluted 1:2,000 in TBST buffer) for 1 h at room temperature. The membranes were then washed with TBST three times. The membranes were visualized by an ECL detection kit (Thermo Fisher, USA), and the experiments were repeated at least three times.

Plasmid constructs.

The plasmid construct pcDNA3.1-ATG10 was created as follows. Atg10 (GenBank NM_025770.3) was amplified by PCR with Pyrobest (TaKaRa, Dalian, China). The PCR products were cloned into the pcDNA3.1 vector at cloning sites NheI-BamHI. The putative pcDNA3.1-ATG10 was verified by DNA sequencing (Sangon Biotech, Shanghai, China). pcDNA3.1-HSF1 was constructed by other members of our research group and has been described elsewhere (38).

Transfect experiments.

After the cells reached approximately 80% confluence, RAW264.7 cells were transfected with pcDNA3.1-HSF1 plasmid or pcDNA3.1-ATG10 plasmid according to the manufacturer’s instructions (Lipofectamine 2000; Invitrogen, Carlsbad, CA). siRNA-HSF1 (Invitrogen, Carlsbad, CA) and siRNA-ATG10 (Invitrogen, Carlsbad, CA) were transfected by the HiPerFect transfection reagent according to the manufacturer’s instructions (Qiagen, Cambridge, MA).

Chromatin immunoprecipitation (ChIP) assay.

The ChIP assay was performed using a ChIP assay kit (RK20100; Abclonal, Wuhan, China) according to the instructions of the manufacturer. RAW264.7 cells were fixed with 1% formaldehyde for 10 min and lysed with lysis buffer. Then the chromatin fragmentation was performed by using a microtip probe sonicator to shear DNA and was immunoprecipitated with an HSF1 antibody (Santa Cruz Biotechnology); rabbit IgG was used as the negative control. The DNA amounts of Atg10, LC3A, Beclin1, Atg12, Atg13, and Atg16l1 in the precipitate were detected by quantitative real-time PCR (qRT-PCR). The primer sequences of the following Atg10, LC3A, Beclin1, Atg12, Atg13, and Atg16l1 promoter regions are listed: Atg10 left primer, TCTTGGCATTCTTCATGCTG; Atg10 right primer, GTTCCAACAGCGACCTATCC; Beclin1 left primer, CAGCTGCAGGGGTAAGAGAC; Beclin1 right primer, CCTGTGCTGGTTTCCAATTT; LC3A left primer, TTGTTCCTCTGTGGCCTCTT; LC3A right primer, AGAGTACATGCCTCCCATGC; Atg12 left primer, ATCCTGCTGAAGGCTGTA; Atg12 right primer, TGATGAAGTCAATGAGTCCT; Atg13 left primer, GGAGATTCTATGGAGTTGGA; Atg13 right primer, TTCAGCAGCAGTGACAAT; Atg16l1 left primer, CGCTCTGTCTCTTCCATC; and Atg16l1 right primer, ACATACGAGGCAGTAGTTG.

Electrophoretic mobility shift assay (EMSA).

Nuclear and cytosolic proteins were prepared from cells with NE-PER nuclear and cytoplasmic extraction reagents (Pierce Biotechnology, Rockford, USA). Biotin-labeling oligonucleotides for autophagy-related genes were synthesized by SongGon Biotech (Wuhan, China). The electrophoretic mobility shift assay (EMSA) reactions were prepared according to the manufacturer’s protocol (Light Shift Chemiluminescent EMSA kit; Thermo Fisher Scientific, Waltham, MA). The following are the probe, cold probe, and mutant probes of autophagy-related genes and their sequences: Atg10 probe, TGGGGGCTGAGGCAGAATATTCAAATATGACGG; Atg10 cold probe, TGGGGGCTGAGGCAGAATATTCAAATATGACGG; Atg10 mutant probe, TGGGGGCTGAGGCACAATAGGCAAATATGACGG; Beclin1 probe, TTGGAAGCAATGGAAGTTTCCATCAATAAAGAA; Beclin1cold probe, TTGGAAGCAATGGAAGTTTCCATCAATAAAGAA; Beclin1 mutant probe, TTGGAAGCAATGGCCGTTTGGATCAATAAAGAA; LC3A probe, CTTTCTATGGTTGTTCCAGAATTCTTTGAAAGCCTG; LC3A cold probe, CTTTCTATGGTTGTTCCAGAATTCTTTGAAAGCCTG; LC3A mutant probe, CTTTCTATGGTTGGGCCACAATTCTTTGAAAGCCTG; Atg12 probe, AAACAATGTTCACAGTTCTAGAAGAGAACTTTC; Atg12 cold probe, AAACAATGTTCACAGTTCTAGAAGAGAACTTTC; Atg12 mutant probe, AAACAATGTTCACAGGGCTAGCCGAGAACTTC; Atg13 probe, ATTTCTGGAAATAAGCCGCCTTTCTTGAAGAC; Atg13 cold probe, ATTTCTGGAAATAAGCCGCCTTTCTTGAAGAC; Atg13 mutant probe, ATGGCTGCAAATAAGCCGCCTGGCTTCAAGAC; Atg16l1probe, CCTGTGGGCTTTTCTGGAAAACCAGAAGTACC; Atg16l1cold probe, CCTGTGGGCTTTTCTGGAAAACCAGAAGTACC; and Atg16l1 mutant probe, CCTGTGGGCTTGGCTGCAAAACCAGAAGTACC.

Statistical analysis.

All quantitative data are presented as means ± SD. The differences between different groups were analyzed with one-way analysis of variance, followed by Student-Newman-Keuls or Dunnett’s test for post hoc comparisons. A P value of ≤ 0.05 was considered statistically significant.