A total of 64 male C57/BL6 mice weighing 20–25 g and aged 6–8 weeks were purchased from the Model Animal Research Center of Nanjing University (Nanjing, China) (stock no. J003752). The mice were maintained in a specific pathogen-free environment, with a humidity of 50%, constant temperature of 26°C, and a 12-h light/dark cycle. The mice had access to food and water ad libitum. All animal care and experimental procedures conformed strictly to the regulations of the Institutional Animal Care and Use Committee of Central South University (Changsha, China). The study protocol was reviewed and approved by the Medical Ethical Committee of the Second Xiangya Hospital of Central South University (Changsha, China). A total of 64 mice were divided randomly into four groups (n=16/group; 8 for observation of survival rate, 8 for heart sample analysis): The wild-type (WT)-sham, WT-CLP, anti-MyD88-CLP and anti-TRIF-CLP groups.

Severe sepsis was generated by CLP. Briefly, following anesthetization with 1.5% pentobarbital sodium (40 mg/kg; Sigma-Aldrich; Merck KGaA, Darmstadt, Germany) injected intraperitoneally, a 1.0-cm long incision was made on the abdomen of each mouse. The cecum was exposed, ligated below the ileocecal valve with a silk 4–0 suture, and punctured twice with a 20-gauge needle. Anti-MyD88 (5 µl/g; ab2068) and anti-TRIF antibodies (5 µl/g; ab13810) (both from Abcam, Cambridge, UK) were administered to the respective mice through the tail veins 2 h before CLP. The sham group underwent laparotomy without CLP. Following surgery, pre-warmed normal saline (37°C; 0.1 ml/g body weight) was administered subcutaneously to maintain fluid resuscitation and avoid iatrogenic hypothermia. The survival rate of each group was recorded from 12–24 h after sham operation or CLP.

Echocardiography was performed at 6 and 12 h after CLP under light anesthesia by pentobarbital sodium [40 mg/kg, intraperitoneal (IP)] using an ultrasound scanner (Acuson S3000) coupled with an 18.0-MHz linear transducer (both from Siemens Healthineers, Erlangen, Germany). A single experienced operator who was blinded to the experimental design acquired all images. LV systolic function was assessed by fractional shortening (FS) using the M-mode method at the mid-papillary level in the parasternal short-axis view. LV internal diameters at end diastole and end systole were measured. Strain was measured using a speckle tracking approach for the middle of the posterior wall on short-axis views during at least three consecutive heartbeats, and analyzed online using Velocity Vector Imaging software v. 3.5 (Siemens Healthineers). This approach enables rapid and accurate quantitative assessment of global and regional myocardial function (18). Strain was defined as relative deformation of the myocardium (19), which represents myocardial contraction sensitivity, and was expressed as a percentage.

Under anesthesia with pentobarbital sodium (40 mg/kg, IP), blood samples were collected via puncture from the inferior vena cava by opening the abdomen in all groups 12 h after CLP. Mice were then sacrificed via cervical dislocation. Blood samples were centrifuged at 589 × g/min for 10 min at 4°C. The serum cTnI level was measured using an ELISA Quantikine mouse kit (M6000B, MTA00B and MLB00C; R&D Systems Europe, Ltd., Abingdon, UK) according to the manufacturer's recommendations.

Following euthanasia, the hearts were harvested and stored at −80°C after being quickly immersed in liquid nitrogen for reverse transcription-polymerase chain reaction (RT-PCR) assays. Total RNA was purified from heart tissue using TRIzol reagent (Gibco; Thermo Fisher Scientific, Inc., Waltham, MA, USA) according to the manufacturer's protocol. RT-PCR was performed to amplify mouse interleukin-1 (IL-1), IL-6, tumor necrosis factor-α (TNF-α) and β-actin mRNA. Using 2 µl reverse transcriptase (Promega Corp., Madison, WI, USA), reactions were performed in a final volume of 20 µl with the gene-specific primers listed in Table I. The expression of the genes examined was normalized to that of β-actin as an internal control. Amplified PCR products were evaluated by 1.7% agarose gel electrophoresis and ethidium bromide staining. The integral optical density (IOD) of the electrophoretic bands was quantified. The data presented in the figures was the ratio of the IOD of the target gene to the IOD of the reference gene. Results were interpreted using Image-Pro Plus 6.0 (Media Cybernetics, Inc., Rockville, MD, USA).

Samples of hearts were dissected and fixed in 10% buffered formalin at room temperature for >48 h, and subsequently embedded in paraffin, cut into 5-µm sections, and stained with hematoxylin and eosin at 37°C for 15 min. Formalin-fixed, paraffin-embedded heart tissue was deparaffinized and blocked with 2% bovine serum albumin (Sigma-Aldrich; Merck KGaA) for 30 min at room temperature. To assess neutrophil accumulation in heart tissue, the sections were incubated with rabbit polyclonal anti-Gr-1 antibody (1:50; ab25377; Abcam). Macrophages in the myocardium were examined with rabbit polyclonal anti-cluster of differentiation (CD)45 antibody (1:200; ab3638; Abcam). These antibodies were applied for 12 h at 4°C. Following this, the secondary antibody (horseradish peroxidase goat anti-rabbit immunoglobulin G; 1:200; G23303; Jackson ImmunoResearch Laboratories, Inc., West Grove, PA, USA) was applied for 50 min at room temperature. Subsequently, the slides were treated with a 3,3′-diaminobenzidine staining system (Dako; Agilent Technologies, Inc., Santa Clara, CA, USA) according to the manufacturer's protocol. Hematoxylin was used to counterstain the nuclei for 3 min at room temperature. The slides were observed under a light microscope (Carl Zeiss AG, Oberkochen, Germany) at magnifications of ×200 and ×400.

Tissue MPO activity was determined in the heart. Heart samples were harvested, excised, homogenized, rinsed and dissolved in 60 ml phosphate-buffered saline (pH 6). Subsequently, 0.9 ml tissue homogenate was mixed well with 0.1 ml myocardial MPO (Nanjing Jiancheng Bioengineering Institute, Nanjing, China) and activity was assayed using a MPO Colorimetric activity assay kit (Nanjing Jiancheng Bioengineering Institute). After resting at 37°C in a water bath for 15 min, the mixture was added to color carrier and H₂O₂. Following this, the mixture was incubated at 60°C for 10 min in a water bath, and the rate of change in absorbance at 460 nm was measured using a spectrophotometer (CE 9000; Cecil Instruments, Ltd., Cambridge, UK). Myocardial MPO activity was expressed as U/l protein in wet tissue.

Apoptosis assays, including those conducted to detect caspase-3 and Fas/Fas ligand (FasL) activity, were performed using the RT-PCR method described above. The primers used are listed in Table II.

Statistical analysis was performed using GraphPad Prism 5.0 software (GraphPad Software, Inc., La Jolla, CA, USA). All data were presented as the mean ± standard deviation. Comparison among groups was performed using one-way analysis of variance with Bonferroni post hoc tests for statistical significance. Survival rates were estimated using the log-rank test. P<0.05 was considered to indicate a statistically significant difference.

A total of 12 h after CLP, WT-CLP mice displayed septic symptoms, including ruffled hair, slow physical actions, shivering and low temperature. The survival rate at 16 h was ~20%. Mice in the anti-MyD88-CLP and anti-TRIF-CLP groups presented signs of moderate health, such as shiny hair, and normal physical actions and temperature throughout the observation period. The survival rates at 16 h were 100% and at 24 h in these groups were 50 and 37%, respectively, which was significantly higher than that in the WT-CLP group (P<0.05). These results indicate that MyD88 and TRIF signaling contributed to the severe sepsis-linked high mortality rate observed in WT-CLP mice (Fig. 1).

M-mode studies indicated that baseline LV contractile function was similar in the four experimental groups. CLP resulted in significantly deteriorated LV function, as evidenced by 20 and 40% reductions in strain at 6 and 12 h after CLP, respectively, compared with the baseline measurements (WT-CLP vs. WT-sham: 6 h post-CLP, 16.6 vs. 20.3%, P<0.05; 12 h post-CLP, 12.5 vs. 19.3%, P<0.01) (Table III). In addition, compared with the baseline, the WT-CLP group demonstrated reductions in FS of 21% at 12 h after CLP (41 vs. 52%, P<0.05) and 50% at 24 h after CLP (27 vs. 52%, P<0.01) (Table III). The anti-MyD88- and anti-TRIF-CLP groups conserved LV contractile function, as demonstrated by significantly higher global longitudinal strain compared with that in the WT-CLP group at each time-point (all P<0.05) (Table III), as well as improved FS at 24 h after CLP (anti-MyD88- and anti-TRIF-CLP vs. WT-CLP: 38 and 42 vs. 27%, P<0.05) and smaller LV internal diameters at end systole at 24 h (anti-MyD88-and anti-TRIF-CLP vs. WT-CLP: 1.8 and 1.73 vs. 2.3 mm, P<0.05) (Table III).

The circulating level of cTnI in WT-CLP mice was six times higher than that in WT-sham mice (6.12±0.22 vs. 0.84±0.06 µg/l, respectively, P<0.05), and approximately two times higher than those in anti-MyD88-CLP and anti-TRIF-CLP mice (6.12±0.22 vs. 3.34±0.11 and 3.19±0.14 µg/l, respectively, P<0.05), with no significant difference between the latter two groups. These results further demonstrate that mice treated with anti-MyD88 and anti-TRIF antibodies exhibited less severe sepsis-induced cardiac injury than mice in the WT-CLP group (Fig. 2).

After 24 h, CLP induced a significant increase in myocardial IL-1, IL-6 and TNF-α mRNA production compared with the levels observed in the WT-sham group (P<0.05) (Fig. 3). These levels were similar in the anti-MyD88-CLP, anti-TRIF-CLP and WT-sham groups, and significantly lower than the levels in the WT-CLP group (P<0.05). Production was slightly, but not significantly, higher in the anti-MyD88 CLP group than in the anti-TRIF CLP group.

Following CLP, the myocardial MPO activity level was significantly higher in mice of the WT-CLP group than the level in WT-sham mice (73 vs. 49 µ/l, respectively, P<0.05) (Fig. 4). In contrast, MPO activity levels were significantly reduced in the anti-MyD88-CLP and anti-TRIF-CLP groups compared to the level in the WT-CLP group (62 and 59 vs. 73 µ/l, respectively, P<0.05) (Fig. 4). No significant difference was observed between the anti-MyD88 and anti-TRIF antibody groups, indicating that these antibodies had equivalent effects on alleviating neutrophil infiltration.

Increased expression of adherent molecules in heart tissue, including vascular cell adhesion molecule-1 (VCAM-1) and intracellular adhesion molecule-1 (ICAM-1), was observed in the WT-CLP group compared with the WT-sham group (P<0.05) (Fig. 5). The levels of VCAM-1 and ICAM-1 mRNA were significantly lower in the anti-MyD88-CLP and anti-TRIF-CLP groups than the levels in the WT-CLP group (P<0.05) (Fig. 5).

In the WT-sham group, myocardial fibers were arranged regularly, with distinct striation and no apparent degeneration or necrosis. In the WT-CLP group, edema, local swelling and some myocardial necrosis with disorganized myocardial fibers were evident in the center of reperfusion (Fig. 6).

Neutrophil and macrophage infiltration, demonstrated by a pervasive brown-yellow color in cardiac muscle cells, was observed in myocardial tissue samples from the WT-CLP, anti-MyD88-CLP and anti-TRIF-CLP groups. Consistent with the myocardial MPO results, the areas of neutrophil and macrophage infiltration in myocardial tissue samples from mice in the WT-CLP group were larger than in those from WT-sham mice. This expression in the anti-MyD88-CLP and anti-TRIF-CLP groups did not differ from that in the WT-sham group, and was notably lower than that in the WT-CLP group (Fig. 6).

Fas/FasL mRNA levels were significantly increased in hearts from mice of the WT-CLP group compared with the levels in the WT-sham group (P<0.05). Furthermore, compared to WT-CLP mice, the levels were significantly lower in the hearts of mice in the anti-TRIF-and anti-MyD88-CLP groups (P<0.05) (Fig. 7). Notably, the level of caspase-3 activity was slightly, yet significantly, higher in WT-CLP mice than in mice of the anti-TRIF- and anti-MyD88-CLP groups (P<0.05) (Fig. 7).

MyD88, TRIF, NF-κB and IRF-3 expression was significantly higher in the WT-CLP group than in the WT-sham group (P<0.05) (Fig. 8). These levels were also significantly higher in the WT-CLP group than in the anti-MyD88- and anti-TRIF-CLP groups (P<0.05) (Fig. 8). These results suggest that: i) MyD88 and TRIF promote cardiac injury during severe sepsis; ii) MyD88 is a key adaptor protein in the TLR-mediated MyD88-dependent NF-κB activation pathway, and TRIF is a key adaptor protein in the TLR-mediated MyD88-independent pathway; and iii) TLR domain alignment of the MyD88-dependent and TRIF-dependent pathways serves an essential role in impairment of cardiac function during severe sepsis.