HepG2 cells were purchased from ATCC (HB-8065, USA) and cultured in Dulbecco's minimum essential media (DMEM) supplemented with 10% fetal bovine serum (FBS), 100 U/mL penicillin, and 100 μg/mL streptomycin in an atmosphere of 5% CO₂ at 37 °C. At 70–80% confluency, the cells were trypsinized and challenged by S. oralis with a multiplicity of infection (MOI)=50 for 24 h.

C57BL/6J mice were bought from the Beijing HFK Bioscience Co. Ltd (Beijing, China) and grouped randomly into either LD12:12 or jet-lagged. Homozygous BMAL1-deficiency (Bmal1^−/−) mice in the C57BL/6J background were produced by breeding heterozygous BMAL1-deficiency mating pairs (Bmal1^+/−), which was a kind gift from Dr. Ying Xu (Soochow University, Jiangsu, China). T BMAL1-deficiency was confirmed by Western blot as described (Bunger et al., 2000).

Total RNA was extracted on ice using TRIzol (Vazyme, China), according to the manufacturer's protocols. cDNA synthesis was performed with reverse transcription reagent (Vazyme). Real-time RT-PCR was performed using SYBR Green PCR protocol on an ABI 7300 real-time PCR system (Applied Biosystems, Carlsbad, CA, USA). Relative mRNA expression of the target genes was normalized against that of GAPDH using the 2^−ΔΔCt method, and presented as mean±SD of replicates. The primers used for amplification are listed in Supplementary Table 1.

Cells were lysed on ice by modified RIPA buffer (P0013B; Beyotime, Shanghai, China) to obtain total protein extract that was quantitated by BCA protein Assay kit (P0010; Beyotime, Shanghai, China), and resolved by 10% sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE). Subsequently, the proteins were transferred to polyvinylidene fluoride (PVDF) membranes that were then blocked in 5% milk diluted with TBST for 1 h at room temperature, probed with primary antibodies, such as BMAL1 (Abcam, 1:1000), CLOCK (Abcam, 1:1000), PER2 (Abcam, 1:1000), CRY2 (Abcam, 1:1000), GAPDH (Santa Cruz Biotechnology Inc., 1:10000), FVII (Santa Cruz Biotechnology Inc., 1:1000), and FXII (Santa Cruz Biotechnology Inc., 1:500), at 4°C overnight. Then, the membranes were incubated with a secondary goat anti-rabbit antibody (1:2000) on the following day for 1 h at room temperature. ECL enhanced chemiluminescence substrate kit (Millipore) was used for imaging and quantitation of the immunoreactive bands by the Image J software.

Liver tissues were fixed with 4% paraformaldehyde in phosphate-buffered saline (PBS) overnight and sliced into 5-μm-thick paraffin-embedded sections. After deparaffinization and rehydration, antigen retrieval was performed by boiling the sections in citrate antigen retrieval solution (P0081; Beyotime) for 15 min. Next, the endogenous peroxidase activity of the samples was blocked by 3% H₂O₂ for 15 min, followed by pre-blocking with 5% BSA (A1933; Sigma–Aldrich) for 1 h at room temperature. The levels of proteins were investigated in liver tissues using primary antibodies against BMAL1 (Abcam, 1:800), FVII (Santa Cruz Biotechnology Inc., 1:400), and FXII (Santa Cruz Biotechnology Inc., 1:400) using a Vectastain ABC kit (Vector Laboratories, Burlingame, CA, USA), followed by the DAB Substrate kit (Vector Laboratories). All protocols were followed according to the manufacturer's instructions.

Cells were plated at a density of 1 × 10⁵ cells/well in 24-well plates and then grown on glass coverslips for 12 h and infected with S. oralis, followed by staining with 5 mM SYTO® Green-Fluorescent Nucleic Acid Stains (Thermo Fisher, USA) for 30 min. After gentle washing for three times with PBS, the cells were fixed for 10 min in 4% paraformaldehyde (wt/vol). The cells were pre-blocked with 5% BSA (A1933; Sigma–Aldrich), labeled with TRITC Phalloidin (Yeasen, 1 μg/mL), and incubated with anti-rabbit Alexa Fluor 594 (Invitrogen) secondary antibodies (1:300) for 1 h at room temperature. The cell nuclei were stained with by DAPI dihydrochloride staining solution. Images were collected using a confocal microscope (SP2-AOBS Leica Microsystems).

The concentrations of mouse plasma FVII (Cusabio, China) and mouse plasma FXII (Cusabio) were measured using ELISA kits according to the manufacturer's instructions. HepG2 cells were seeded overnight in 1% FBS-DMEM in six-well plates (6 × 10⁶/well) and infected with S. oralis in 100 U/mL penicillin and 100 μg/mL streptomycin for 24 h (MOI=50). Then, cell-free-culture supernatants were collected and analyzed for FVII or FXII content by a specific ELISA (), according to the manufacturer's recommendations. The cells without infection are similar to infected cells. Briefly, the standard protein was solubilized and diluted to 20, 10, 5, 2.5, 1.25, 0.625, and 0.312 ng/mL for constructing a standard curve. The diluted samples were added to the reaction plate and incubated at 37°C for 2 h. The wells were washed five times, and the biotin primary antibody was added, and the reaction was incubated at 37°C for 1 h. Subsequently, horseradish peroxidase (HRP)-avidin was added and incubated at 37°C for 1 h, followed by the addition of a substrate chromogenic solution that was incubated at 37°C for 30 min. Finally, the reaction terminator and OD₄₅₀ were measured.

The following probe was used for FISH: S. oralis (probe sequences targeted 16S ribosomal-RNA (rRNA) genes; 6FAM-CTCCTACGGGAGGCAGCAGTAGGGA-BHQ-3; fluorescence emission maximum at 900 nM). As mentioned above, the liver sample was fixed in 4% paraformaldehyde for 2 h at room temperature and stored in 96% ethanol. Then, the paraffin-embedded sections (5-μm-thick) were incubated in hybridization buffer (100 mM Tris-HCl pH 7.2, 0.9 M NaCl, 0.1% SDS) containing 8 ng/mL of the S. oralis probe overnight at 37°C. Slides were washed 3 times for 15 min in prewarmed (37°C) hybridization buffer, following the labeling of the cell nucleus by DAPI. Images were collected using a confocal microscope (SP2-AOBS Leica Microsystems).

For in vivo imaging, mice (LD12:12 and jet-lagged) were injected 10⁹ fluorescence-labeled bacteria through the tail vein for 6 h prior to imaging, as described previously. Subsequently, the animals were euthanized with sodium pentobarbital, the hairs were wiped off, and the organs were removed and placed on plates. The fluorescence was detected at 490 nm (in-vivo FX PRO BRUKER). The images were analyzed using Bruker software. All procedures were performed using a standard protocol.

A volume of 1.8 mL heart blood was collected in a blood collection tube containing 0.2 mL of 3.2% sodium citrate anticoagulant (blood volume <1.8 mL was prepared according to the ratio of blood to anticoagulant 1:9) and mixed gently. The plasma was separated at 3,000 rpm for 15 min. The prothrombin time (PT), activated partial thromboplastin time (APTT), fibrinogen time, and fibrinogen concentration assays were conducted using a specific kit (Hanlisisw, China). All assays were carried out with an automatic coagulator (BE, Germany).

The strain used for research: S. oralis from ATCC. Brain Heart Infusion (BHI) medium (Becton, Dickinson and Company, USA) was used for bacterial growth under aerobic conditions for 24–48 h at 37°C in order to reach the exponential growth phase. Subsequently, 10-fold serial dilutions of each bacterium were performed in PBS to achieve concentrations of ~1 × 10⁸ CFU/mL. Specific primers were designed for S. oralis (forward primer: CAACGATACATAGCCGACCTGAG; reverse primer: TCCATTGCCGAAGATTCC).

Chromatin immunoprecipitation assays were conducted using chromatin immunoprecipitation kit (Millipore), according to the manufacturer's protocol. Briefly, HEPG2 cells were fixed in 1% formaldehyde, cross-linked with glycine, and lysed. Cell lysates were sonicated into fragments and precleared with Protein A/G Agarose beads for 1 h. After centrifugation (1,000 rpm × 1 min at 4°C), 10% of the supernatants were stored as input, and the remaining supernatants were divided into two parts, one for detection by BMAL1 antibody and the remaining for IgG reactivity (negative control). The DNA was purified in a final volume of 50 mL and subjected to PCR. Specific primers were designed for FVII promoter (forward primer: GACCGTCTACCCCAGTGTTT; reverse primer: CGGCAGTCCACGTCATTTC).

Bmal1-overexpression and Bmal1-knockout HepG2 cells were constructed and plated at a density of 5 × 10⁵ cells/well in 12-well plates. These cells were transfected with the wild-type FVII promoter, its deletion mutant, its mutation with mutated BMAL1-binding site, and empty vector (PGL3). Furthermore, firefly luciferase reporter vectors (100 ng) and Renilla luciferase reporter vectors pRL-SV40 (10 ng; Promega, USA) were transfected into cells using Lipofectamine™ 3,000 transfection kit (Thermo Fisher, USA) for 48 h. Subsequently, firefly and Renilla luciferase activity was measured using the Dual-Luciferase Reporter Assay System (Promega). The data were normalized by the firefly/Renilla ratio.

Briefly, mid-logarithmic phase cultures of S. oralis were diluted 1:20 in complete medium (Brain Heart Infusion, Becton, Dickinson, and Co.) containing 1 mM calcium chloride, 0.2% BSA (Sigma Aldrich Ltd, Dorset, UK) and 100 ng/mL competence-stimulating peptide 1 (CSP1; Mimotopes, Clayton, Victoria, Australia). The transformation reactions were incubated for 90 s at 42 °C and placed on ice for 2 min. Then, the sensory bacteria were transferred to 6-well plate with ampicillin for 16 h. The primers were designed for GFP (forward primer: GACCGTCTACCCCAGTGTTT; reverse primer: CGGCAGTCCACGTCATTTC).

All the statistical analyses were performed with GraphPad Prism Software and based on normally distributed datasets with equal variance (Bartlett's test). The investigators were not blinded during the experiments or outcome assessment. The data points were presented as mean ± SD values unless otherwise stated. Data were inferred as statistically significant if P-values were <0.05. Significance between two groups was determined by independent Student's t-test. For multiple comparisons, one-way analysis of variance (ANOVA) was used with Tukey's multiple comparisons post-hoc test unless stated otherwise.

α-hemolytic Streptococcus, such as S. oralis, exhibits strong pathopoiesia either by generating various enzymes and exotoxin indirectly or invading into a host directly. In order to elucidate how S. oralis interacts with hepatocytes, nucleic acid dye-treated bacteria had infected cells in MOI=10,000 for 30 min, which showed that S. oralis directly invaded into the cells (Figure 1A). Owning to the lack of elements initiating coagulation, the clotting factors were not consumed, and hence, it was found that FVII slightly decreased, and FXII increased significantly in the cellular supernatant of infected HepG2 (Figure 1B). To explore the correlation between S. oralis infection and circadian clock genes in hepatocytes, cells were treated with S. oralis, and the total protein and mRNA were extracted. Consequently, BMAL1 expression was significantly decreased as compared to the other circadian clock genes, and it was found that the coagulation factor FVII decreased slightly, while the FXII did not alter (Figures 1C,D). In another cohort of cells with the same treatment, the expression of BMAL1 and other circadian clock genes was lower in S. oralis infection group, and obvious upregulation of FVII and mild downregulation of FXII was detected at the transcriptional level (Figure 1E).

Correspondingly, to explore whether S. oralis modified the circadian clock genes and coagulation factor levels in vivo, another series of studies were carried out. The mice were injected with S. oralis into the tail vein. After the mice were sacrificed, the whole blood was collected for a routine examination, and it was found that the neutrophils increased significantly. In addition, the total DNA of the liver and spleen was extracted and amplified to detect the bacteria in these organs using S. oralis-specific primers (Figure 2A). To further consolidate the presence of bacteria in the liver, FISH was conducted, which showed green EGFP-labeled-bacterium in the liver, and many of them had invaded into the hepatocytes (Figure 2B). This phenomenon might suggest that S. oralis plays a functional role both directly and indirectly. The remaining of the whole blood was collected for coagulation arrays and ELISA. It was observed that the level of both plasma coagulation factors, FVII and FXII, was downregulated (Figure 2C). Additionally, it was found that PT, APTT, and fibrinogen reaction time were prolonged and plasma fibrinogen was declined (Figure 2D). In sepsis, bacteria can activate both intrinsic and extrinsic coagulation by the cytokine response. The serial sections of liver, spleen, kidney and lung stained with hematoxylin and eosin (H&E) indicated inflammation after infection (Figure 2E). Moreover, most of the coagulation factors, including FVII and FXII were generated in the liver. To investigate the correlation between S. oralis infection and coagulation factors, serial sections of the liver tissues were stained with H&E, and histochemistry was conducted (Figure 2F). S. oralis infection declined the expression of FVII expression, while that of FXII was elevated (Figure 2G). In addition, the livers of mice were collected and the total proteins were extracted. The results showed that BMAL1 expression was significantly decreased as compared to that of the other genes; also, an obvious decline was detected in FVII and a mild elevation of FXII as compared to the uninfected group (Figures 2H,I). On the other hand, BAML1 level was also decreased, while the mRNA expression of REV-ERBa and CRY1 was elevated, FVII was significantly downregulated, and FXII was markedly upregulated (Figure 2J). Therefore, we confirmed that S. oralis infection declined the level of FVII and BMAL1 while that of FXII was elevated at both protein and transcriptional levels in vivo.

To confirm the role of BMAL1 in coagulation factor biosynthesis, we constructed BMAL1-deficiency (Bmal1^−/−) mice, and the genotypes were verified by Western blot (Figure 2D). Interestingly, we found that BMAL1 had two-aspect functions in the coagulation process. The blood coagulation arrays showed that PT was slightly deduced while APTT was significantly prolonged with a huge reduction in these fibrinogen reaction time owning to elevated plasma fibrinogen in Bmal1^−/− mice (Figure 3A). FVII and FXII were the key enzymes that initiated the extrinsic and intrinsic coagulation cascade, respectively. Therefore, the serial sections of the liver tissues were stained with H&E, and histochemistry was performed (Figure 3B). The result showed that FVII approximately decreased to a third, and FXII increased nearly four-fold in Bmal1^−/− mice (Figure 3C). To further verify the correlation between BMAL1 and the biosynthesis of coagulation factors, the total protein of the liver tissues was extracted. This phenomenon was similar to that FVII decreased significantly while FXII increased by three-fold in Bmal1^−/− mice (Figures 3D,E). In addition, that the expression of the coagulation factor was altered at the mRNA level, which was synchronous with the changes in the protein level, indicating that BMAL1 was likely to act as a transcription factor to regulate the biosynthesis of coagulation factors (Figure 3F). Although coagulation factors FVII and FXII played major roles in coagulation cascades, other coagulation factors, such as vWF, also played a significant role. Therefore, we analyzed the expression of other coagulation factors (Figure 3G). To investigate the effect of BAML1 on the biosynthesis of coagulation factors in vitro, Bmal1-overexpression or BAML1-knockout was used (Figure 3H), which showed that FVII expression increased mildly and FXII significantly decreased in Bmal1-overexpressing cells. However, a contrast result was observed in BAML1-knockout cells (Figure 3I). It was also confirmed that the deficiency of BMAL1 negatively regulated FVII expression but positively regulated FXII.

Firstly, we found that FVII mRNA level showed 12-h circadian changes and reached nadir around ZT8 and ZT20, but FXII did not exhibit circadian changes; several circadian clock genes in the liver showed obvious rhythmicity, and BMAL1 mRNA expression was synchronous with FVII expression by hitting valleys around ZT8 and ZT20 (Figure 4A). Furthermore, the correlation between circadian clock genes and FVII showed that the expression of BMAL1 was strongly associated with that of FVII (Figure 4B). To further verify the rhythmicity of FVII expression, we measured the plasma coagulation factor at different time points. Strikingly, FVII expression reached nadir around ZT8 and ZT20, while that of FXII lacked circadian changes (Figure 4C). Typically, BMAL1 acts as a transcription factor that forms a heterodimer with CLOCK to activate the expression of the downstream genes, such as E-box-controlled genes PER1 and CRY1 (Duong et al., 2011). Therefore, it could be deduced that BMAL1 regulated FVII through transcription factor binding sites (Figures 4D,E).

Low activity of FVII promoter could be found in cells transfected with both the FVII promoter mutants with Bmal1 binding site and the FVII promoter deletion mutant. To confirm the role of BMAL1, firefly luciferase reporter vectors were transfected into Bmal1-overexpressing and Bmal1-knockout cells. Next, we found that the activity of FVII promoter was enhanced in Bmal1-overexpressing cells, while it was declined in Bmal1-knockout cells. Together, these results suggested that BMAL1 directly regulates the expression of FVII in a positive way (Figure 4F).

Since BMAL1 plays a key role in circadian rhythm, investigating the circadian rhythm disruption-driven coagulation is crucial for unveiling the role of BMAL1 in coagulation. To further explore the roles of circadian clock gene BMAL1 in coagulation, a jet-lagged mouse model was established to observe the impact of circadian rhythm disruption on the impairment of the liver ossification process. All mice were maintained under 12/12-h light/dark cycles with the light on from 8:00 a.m. (Zeitgeber time 0, ZT0) to 8:00 p.m. (ZT12) and fed antibiotic-free food and water ad libitum. In the experiment, C57BL/6J mice were placed on a previously described jet-lagged schedule with 8-h light advanced every 2–3 days for 12 weeks, which mimicked the circadian disturbance that humans undergo during shift work (Zhao et al., 2017). PT was mildly shortened, while APTT was significantly prolonged, and the fibrinogen reaction time was shortened owing to elevated plasma fibrinogen in jet-lagged mice (Figure 5A). To eliminate the impact of body weight and liver weight, measurement experiments were essential, which seemed to have no impact on the coagulation (Figures 5B,C). However, jet-lag seemed to induce a slight decline in BT, suggesting that circadian rhythm disruption-driven downregulation of BMAL1 might be conducive to thrombus formation (Figure 5D). In order to elucidate whether circadian rhythm disruption-driven coagulation prevents extravasation of S. oralis, EGFP-labeled S. oralis was constructed and confirmed by PCR amplification (Figure 5E). We found that the number of bacteria invaded into the liver was less than that in the LD12:12 group due to disruption-driven coagulation, which prevents the invasion of EGFP-labeled S. oralis. Intriguingly, a similar phenomenon was observed in the lung and spleen, and that the number of bacteria invaded into the kidneys was large in quantity in both the jet-lagged and LD12:12 groups because of abundant blood vessels (Figures 5F,G).