To prepare graphene oxide (GO)-containing gelatin methacrylate anhydride (GelMA) hydrogel and to investigate the effects of in situ photopolymerized GO-GelMA composite hydrogel in wound vascularization of full-thickness skin defect in mice.

The experimental study method was used. The 50 μL of 0.2 mg/mL GO solution was evenly applied onto the conductive gel, and the structure and size of GO were observed under field emission scanning electron microscope after drying. Human skin fibroblasts (HSFs) were divided into 0 μg/mL GO (without GO solution, the same as below) group, 0.1 μg/mL GO group, 1.0 μg/mL GO group, 5.0 μg/mL GO group, and 10.0 μg/mL GO group treated with GO of the corresponding final mass concentration, and the absorbance value was detected using a microplate analyzer after 48 h of culture to reflect the proliferation activity of cells (n=6). HSFs and human umbilical vein vascular endothelial cells (HUVECs) were divided into 0 μg/mL GO group, 0.1 μg/mL GO group, 1.0 μg/mL GO group, and 5.0 μg/mL GO group treated with GO of the corresponding final mass concentration, and the migration rates of HSFs at 24 and 36 h after scratching (n=5) and HUVECs at 12 h after scratching (n=3) were detected by scratch test, and the level of vascular endothelial growth factor (VEGF) secreted by HSFs after 4, 6, and 8 h of culture was detected by enzyme-linked immunosorbent assay method (n=3). The prepared GO-GelMA composite hydrogels containing GO of the corresponding final mass concentration were set as 0 μg/mL GO composite hydrogel group, 0.1 μg/mL GO composite hydrogel group, 1.0 μg/mL GO composite hydrogel group, and 5.0 μg/mL GO composite hydrogel group to observe their properties before and after cross-linking, and to detect the release of GO after soaking with phosphate buffer solution for 3 and 7 d (n=3). The full-thickness skin defect wounds were made on the back of 16 6-week-old female C57BL/6 mice. The mice treated with in situ cross-linked GO-GelMA composite hydrogel containing GO of the corresponding final mass concentration were divided into 0 μg/mL GO composite hydrogel group, 0.1 μg/mL GO composite hydrogel group, 1.0 μg/mL GO composite hydrogel group, and 5.0 μg/mL GO composite hydrogel group according to the random number table, with 4 mice in each group. The general condition of wound was observed and the wound healing rate was calculated on 3, 7, and 14 d of treatment, the wound blood perfusion was detected by laser Doppler flowmetry on 3, 7, and 14 d of treatment and the mean perfusion unit (MPU) ratio was calculated, and the wound vascularization on 7 d of treatment was observed after hematoxylin-eosin staining and the vascular density was calculated (n=3). The wound tissue of mice in 0 μg/mL GO composite hydrogel group and 0.1 μg/mL GO composite hydrogel group on 7 d of treatment was collected to observe the relationship between the distribution of GO and neovascularization by hematoxylin-eosin staining (n=3) and the expression of VEGF by immunohistochemical staining. Data were statistically analyzed with analysis of variance for repeated measurement, one-way analysis of variance, and Tukey's method.

GO had a multilayered lamellar structure with the width of about 20 μm and the length of about 50 μm. The absorbance value of HSFs in 10.0 μg/mL GO group was significantly lower than that in 0 μg/mL GO group after 48 h of culture (q=7.64, P<0.01). At 24 h after scratching, the migration rates of HSFs were similar in the four groups (P>0.05); at 36 h after scratching, the migration rate of HSFs in 0.1 μg/mL GO group was significantly higher than that in 0 μg/mL GO group, 1.0 μg/mL GO group, and 5.0 μg/mL GO group (with q values of 7.48, 10.81, and 10.20, respectively, P<0.01). At 12 h after scratching, the migration rate of HUVECs in 0.1 μg/mL GO group was significantly higher than that in 0 μg/mL GO group, 1.0 μg/mL GO group, and 5.0 μg/mL GO group (with q values of 7.11, 8.99, and 14.92, respectively, P<0.01), and the migration rate of HUVECs in 5.0 μg/mL GO group was significantly lower than that in 0 μg/mL GO group and 1.0 μg/mL GO group (with q values of 7.81 and 5.33, respectively, P<0.05 or P<0.01 ). At 4 and 6 h of culture, the VEGF expressions of HSFs in the four groups were similar (P>0.05); at 8 h of culture, the VEGF expression of HSFs in 0.1 μg/mL GO group was significantly higher than that in 0 μg/mL GO group and 5.0 μg/mL GO group (with q values of 4.75 and 4.48, respectively, P<0.05). The GO-GelMA composite hydrogels in the four groups were all red liquid before cross-linking, which turned to light yellow gel after cross-linking, with no significant difference in fluidity. The GO in the GO-GelMA composite hydrogel of 0 μg/mL GO composite hydrogel group had no release of GO at all time points; the GO in the GO-GelMA composite hydrogels of the other 3 groups was partially released on 3 d of soaking, and all the GO was released on 7 d of soaking. From 3 to 14 d of treatment, the wounds of mice in the 4 groups were covered with hydrogel dressings, kept moist, and gradually healed. On 3, 7, and 14 d of treatment, the wound healing rates of mice in the four groups were similar (P>0.05). On 3 d of treatment, the MPU ratio of wound of mice in 0.1 μg/mL GO composite hydrogel group was significantly higher than that in 0 μg/mL GO composite hydrogel group, 1.0 μg/mL GO composite hydrogel group, and 5.0 μg/mL GO composite hydrogel group (with q values of 10.70, 11.83, and 10.65, respectively, P<0.05 or P<0.01). On 7 and 14 d of treatment, the MPU ratios of wound of mice in the four groups were similar (P>0.05). The MPU ratio of wound of mice in 0.1 μg/mL GO composite hydrogel group on 7 d of treatment was significantly lower than that on 3 d of treatment (q=14.38, P<0.05), and that on 14 d of treatment was significantly lower than that on 7 d of treatment (q=27.78, P<0.01). On 7 d of treatment, the neovascular density of wound of mice on 7 d of treatment was 120.7±4.1 per 200 times of visual field, which was significantly higher than 61.7±1.3, 77.7±10.2, and 99.0±7.9 per 200 times of visual field in 0 μg/mL GO composite hydrogel group, 1.0 μg/mL GO composite hydrogel group, and 5.0 μg/mL GO composite hydrogel group (with q values of 12.88, 7.79, and 6.70, respectively, P<0.01), and the neovascular density of wound of mice in 1.0 μg/mL GO composite hydrogel group and 5.0 μg/mL GO composite hydrogel group was significantly higher than that in 0 μg/mL GO composite hydrogel group (with q values of 5.10 and 6.19, respectively, P<0.05). On 7 d of treatment, cluster of new blood vessels in wound of mice in 0.1 μg/mL GO composite hydrogel group was significantly more than that in 0 μg/mL GO composite hydrogel group, and the new blood vessels were clustered near the GO; a large amount of VEGF was expressed in wound of mice in 0.1 μg/mL GO composite hydrogel group in the distribution area of GO and new blood vessels.

GO with mass concentration lower than 10.0 μg/mL had no adverse effect on proliferation activity of HSFs, and GO of 0.1 μg/mL can promote the migration of HSFs and HUVECs, and can promote the secretion of VEGF in HSFs. In situ photopolymerized of GO-GelMA composite hydrogel dressing can promote the wound neovascularization of full-thickness skin defect in mice and increase wound blood perfusion in the early stage, with GO showing an enrichment effect on angiogenesis, and the mechanism may be related to the role of GO in promoting the secretion of VEGF by wound cells.

To explore the heterogeneity and growth factor regulatory network of dermal fibroblasts (dFbs) in mouse full-thickness skin defect wounds based on single-cell RNA sequencing.

The experimental research methods were adopted. The normal skin tissue from 5 healthy 8-week-old male C57BL/6 mice (the same mouse age, sex, and strain below) was harvested, and the wound tissue of another 5 mice with full-thickness skin defect on the back was harvested on post injury day (PID) 7. The cell suspension was obtained by digesting the tissue with collagenase D and DNase Ⅰ, sequencing library was constructed using 10x Genomics platform, and single-cell RNA sequencing was performed by Illumina Novaseq6000 sequencer. The gene expression matrices of cells in the two kinds of tissue were obtained by analysis of Seurat 3.0 program of software R4.1.1, and two-dimensional tSNE plots classified by cell group, cell source, and gene labeling of major cells in skin were used for visual display. According to the existing literature and the CellMarker database searching, the expression of marker genes in the gene expression matrices of cells in the two kinds of tissue was analyzed, and each cell group was numbered and defined. The gene expression matrices and cell clustering information were introduced into CellChat 1.1.3 program of software R4.1.1 to analyze the intercellular communication in the two kinds of tissue and the intercellular communication involving vascular endothelial growth factor (VEGF), platelet-derived growth factor (PDGF), epidermal growth factor (EGF), and fibroblast growth factor (FGF) signal pathways in the wound tissue, the relative contribution of each pair of FGF subtypes and FGF receptor (FGFR) subtypes (hereinafter referred to as FGF ligand receptor pairs) to FGF signal network in the two kinds of tissue, and the intercellular communication in the signal pathway of FGF ligand receptor pairs with the top 2 relative contributions in the two kinds of tissue. The normal skin tissue from one healthy mouse was harvested, and the wound tissue of one mouse with full-thickness skin defect on the back was harvested on PID 7. The multiple immunofluorescence staining was performed to detect the expression and distribution of FGF7 protein and its co-localized expression with dipeptidyl peptidase 4 (DPP4), stem cell antigen 1 (SCA1), smooth muscle actin (SMA), and PDGF receptor α (PDGFRα) protein.

Both the normal skin tissue of healthy mice and the wound tissue of full-thickness skin defected mice on PID 7 contained 25 cell groups, but the numbers of cells in each cell group between the two kinds of tissue were different. Genes PDGFRα, platelet endothelial cell adhesion molecule 1, lymphatic endothelial hyaluronic acid receptor 1, receptor protein tyrosine phosphatase C, keratin 10, and keratin 79 all had distinct distributions on two-dimensional tSNE plots, indicating specific cell groups respectively. The 25 cell groups were numbered by C0-C24 and divided into 9 dFb subgroups and 16 non-dFb groups. dFb subgroups included C0 as interstitial progenitor cells, C5 as adipose precursor cells, and C13 as contractile muscle cells related fibroblasts, etc. Non-dFb group included C3 as neutrophils, C8 as T cells, and C18 as erythrocytes, etc. Compared with that of the normal skin tissue of healthy mice, the intercellular communication in the wound tissue of full-thickness skin defected mice on PID 7 was more and denser, and the top 3 cell groups in intercellular communication intensity were dFb subgroups C0, C1, and C2, of which all communicated with other cell groups in the wound tissue. In the wound tissue of full-thickness skin defected mice on PID 7, VEGF signals were mainly sent by the dFb subgroup C0 and received by vascular related cell groups C19 and C21, PDGF signals were mainly sent by peripheral cells C14 and received by multiple dFb subgroups, EGF signals were mainly sent by keratinocyte subgroups C9 and C11 and received by the dFb subgroup C0, and the main sender and receiver of FGF signals were the dFb subgroup C6. In the relative contribution rank of FGF ligand receptor pairs to FGF signal network in the normal skin tissue of healthy mice and the wound tissue of full-thickness skin defected mice on PID 7, FGF7-FGFR1 was the top 1, and FGF7-FGFR2 or FGF10-FGFR1 was in the second place, respectively; compared with those in the normal skin tissue, there was more intercellular communication in FGF7-FGFR1 signal pathway, while the intercellular communication in FGF7-FGFR2 and FGF10-FGFR1 signal pathways decreased slightly or did not change significantly in the wound tissue; the intercellular communication in FGF7-FGFR1 signal pathway in the wound tissue was stronger than that in FGF7-FGFR2 or FGF10-FGFR1 signal pathway; in the two kinds of tissue, FGF7 signal was mainly sent by dFb subgroups C0, C1, and C2, and received by dFb subgroups C6 and C7. Compared with that in the normal skin tissue of healthy mouse, the expression of FGF7 protein was higher in the wound tissue of full-thickness skin defected mouse on PID 7; in the normal skin tissue, FGF7 protein was mainly expressed in the skin interstitium and also expressed in the white adipose tissue near the dermis layer; in the two kinds of tissue, FGF7 protein was co-localized with DPP4 and SCA1 proteins and expressed in the skin interstitium, co-localized with PDGFRα protein and expressed in dFbs, but was not co-localized with SMA protein, with more co-localized expression of FGF7 in the wound tissue than that in the normal skin tissue.

In the process of wound healing of mouse full-thickness skin defect wound, dFbs are highly heterogeneous, act as potential major secretory or receiving cell populations of a variety of growth factors, and have a close and complex relationship with the growth factor signal pathways. FGF7-FGFR1 signal pathway is the main FGF signal pathway in the process of wound healing, which targets and regulates multiple dFb subgroups.

To investigate the effects of methacrylic anhydride gelatin (GelMA) hydrogel loaded with silver and recombinant human basic fibroblast growth factor (rh-bFGF) on deep partial-thickness burn wounds in rabbits.

The experimental research method was adopted. Low-concentration GelMA materials, medium-concentration GelMA materials and high-concentration GelMA materials containing different concentrations of methacrylic anhydride (MA) were prepared, after adding photoinitiator, low-concentration GelMA hydrogels, medium-concentration GelMA hydrogels, and high-concentration GelMA hydrogels were obtained, respectively. The nuclear magnetic resonance spectroscopy was performed to detect the hydrogen nuclear magnetic resonance spectra of the above-mentioned three concentrations of GelMA materials, and to calculate the degree of substitution according to the spectrum diagram. The three-dimensional microstructure and pore size of 3 types of above-mentioned GelMA hydrogels were detected by field emission scanning electron microscopy (FESEM), with 9 samples measured. According to the selected concentration of MA, ten kinds of solutions of GelMA with different concentration of silver (silver-containing GelMA) were synthesized, and the silver-containing GelMA solution of each concentration was divided into three parts, and then exposed to ultraviolet light lasting for 20, 25, and 35 s, respectively. After adding photoinitiator,the corresponding silver-containing GelMA hydrogels were obtained. The residual degradation rate of silver-containing GelMA hydrogel with different photocrosslinking times was detected by collagenase degradation method at degradation of 12, 24, 36, and 48 h; and the time required for complete degradation was detected, and the sample number was 5. The inhibition zone diameter of GelMA hydrogel under above screened photocrosslinking times containing 10 concentrations of silver against Staphylococcus aureus was measured to reflect its antibacterial ability, and the sample numbers were all 5. The silver-containing GelMA hydrogel with statistical significance compared with the antibacterial circle diameter of the silver-containing GelMA hydrogel containing the lowest concentration (no silver) was considered as having antibacterial activity. The three-dimensional microstructure and pore size of the silver-containing GelMA hydrogels with antibacterial activity and the lowest drug concentration selected were detected by FESEM, and the sample numbers were all 9. The freeze-dried alone GelMA hydrogel and the freeze-dried silver-containing GelMA hydrogel were soaked in phosphate buffer solution for 24 h, respectively, then the swelling rate of the two GelMA hydrogel were calculated and compared by weighing method, and the sample number was 5. GelMA hydrogel containing silver and rh-bFGF, namely compound hydrogel for short, was prepared according to the preliminary experiment and the above experimental results. The appearance of the composite hydrogel was observed in general, and its three-dimensional microstructure and pore size were detected by FESEM. The deep partial-thickness burn wound was made on the back of 30 rabbits (aged 4-6 months, female half and half). Meanwhile, with the rabbit head as the benchmark, the wounds on the left side of the spine were treated as composite hydrogel treatment group, and the wounds on the right side were treated as gauze control group, and which were treated accordingly. On post injury day (PID) 3, 7, 14, 21, and 28, the healing of wounds in the two groups was observed. On PID 7, 14, 21, and 28, the wound healing area was recorded and the healing rate was calculated, with a sample number of 30. Data were statistically analyzed with analysis of variance for repeated measurement, one-way analysis of variance, and independent sample t test.

The substitution degree among low-concentration GelMA materials, medium-concentration GelMA materials, and high-concentration GelMA materials was significantly different (F=1 628.00, P<0.01). The low-concentration GelMA hydrogel had a loose and irregular three-dimensional spatial network structure with a pore size of (60±17) μm; the medium-concentration GelMA hydrogel had a relatively uniform three-dimensional spatial network and pore size with a pore size of (45±13) μm; the high-concentration GelMA hydrogel had the dense and disordered three-dimensional spatial network with a pore size of (25±15) μm, the pore sizes of 3 types of GelMA hydrogels were significantly differences (F=12.20, P<0.01), and medium concentration of MA was selected for the concentration of subsequent materials. The degradability of silver-containing GelMA hydrogels with different concentrations of the same photocrosslinking time was basically same. The degradation residual rates of silver-containing GelMA hydrogels with 20, 25, and 35 s crosslinking time at 12 h were (74.2±1.7)%, (85.3±0.9)%, and (93.2±1.2)%, respectively; the residual rates of degradation at 24 h were (58.3±2.1)%, (65.2±1.8)%, and (81.4±2.6)%, respectively; the residual rates of degradation at 36 h were (22.4±1.9)%, (45.2±1.7)%, and (68.1±1.4)%, respectively; the residual rates of degradation at 48 h were (8.2±1.7)%, (32.4±1.3)%, and (54.3±2.2)%, respectively, and 20, 25, and 30 s photocrosslinking time required for complete degradation of silver-containing GelMA hydrogels were (50.2±2.4), (62.4±1.4), and (72.2±3.2) h, and the difference was statistically significant (F=182.40, P<0.01), 25 s were selected as the subsequent photocrosslinking time. The antibacterial diameters of 10 types of silver-containing GelMA hydrogels against Staphylococcus aureus from low to high concentrations were (2.6±0.4), (2.5±0.4), (3.2±0.4), (12.1±0.7), (14.8±0.7), (15.1±0.5), (16.2±0.6), (16.7±0.5), (16.7±0.4), and (16.7±0.6) mm, respectively, and which basically showed a concentration-dependent increasing trend, and the overall difference was statistically significant (F=428.70, P<0.01). Compared with the silver-containing GelMA hydrogel with the lowest concentration, the antibacterial circle diameters of other silver-containing GelMA hydrogels with antibacterial ability from low to high concentration were significantly increased (with t values of 26.35, 33.84, 43.65, 42.17, 49.24, 55.74, and 43.72, respectively, P<0.01). The silver-containing GelMA hydrogel with the antibacterial diameter of (12.1±0.7) mm had the lowest antibacterial activity against Staphylococcus aureus and the lowest drug loading concentration, and the concentration of silver was selected for the concentration of subsequent materials. The microscopic morphology of the silver-containing GelMA hydrogel containing silver element with a pore size of (45±13) μm had a regular and linear strip-like structure. After soaking for 24 h, the swelling ratio of silver-containing GelMA hydrogel was similar to that of alone GelMA hydrogel. The composite hydrogel was colorless, clear and transparent, and its three-dimensional microstructure was a regular and uniform grid, with a filament network structure inside, and the pore size of (40±21) μm. On PID 3, a large amount of necrotic tissue and exudate of rabbit wound in composite hydrogel group were observed, and scattered scabs, a small amount of necrotic tissue and exudate of rabbit wound in gauze control group were observed. On PID 7, the area of rabbit wound in composite hydrogel group was significantly reduced, and adhesion of rabbit wound and gauze in gauze control group was observed. On PID 14, In composite hydrogel group, the rabbit wound surface was ruddy, and the growth of granulation tissue was observed, and in gauze control group, the rabbit wound base was pale, and the blood supply was poor. On PID 21, the rabbit wounds in composite hydrogel group healed completely, and rabbit wound in gauze control group had healing trend. On PID 28, new hair could be seen on rabbit wound surface in composite hydrogel group; oval wound of rabbit in gauze control group still remained. On PID 7, 14, 21, and 28, the wound healing areas of rabbit in composite hydrogel group were significantly larger than those in gauze control group (with t values of 2.24, 4.43, 7.67, and 7.69, respectively, P<0.05 or P<0.01).

The medium-concentration GelMA hydrogel has good physical and chemical properties in terms of swelling and degradability. The screened silver-containing GelMA hydrogels had the lowest antibacterial activity and the lowest drug loading concentration. Composite hydrogel can significantly shorten the healing time of deep partial-thickness burn wounds in rabbits.

To investigate the effects and molecular mechanism of exogenous L-carnitine on hepatic pyroptosis mediated by excessive endoplasmic reticulum stress in severely scald rats.

The experimental research method was adopted. According to the random number table (the same group method below), fifteen female Sprague Dawley rats aged 6-8 weeks were divided into sham-injury group, scald alone group, and scald+carnitine group (with 5 rats in each group), and full-thickness scald of 30% total body surface area were made on the back of rats in scald alone group and scald+carnitine group, and rats in scald+carnitine group were additionally given intraperitoneal injection of L-carnitine. At post injury hour (PIH) 72, The levels of aspartate aminotransferase (AST) and alanine dehydrogenase (ALT) of biochemical indicators of liver injury were detected by automatic biochemical analyzer with the sample number of 5. At PIH 72, liver tissue damage was detected by hematoxylin-eosin staining. At PIH 72, The mRNA levels of nucleotide-binding oligomerization domain-containing protein-like receptor family pyrin domain containing 3 (NLRP3), cysteine aspartic acid specific protease 1 (caspase-1), gasderminD (GSDMD), and interleukin 1β(IL-1β) in liver tissue as pyroptosis-related markers and glucose regulatory protein 78 (GRP78) and CCAAT/enhancer binding protein homologous protein (CHOP) in liver tissue as endoplasmic reticulum stress-related markers were detected by real-time fluorescence quantitative reverse transcription polymerase chain reaction (RT-qPCR). Protein expression levels of GRP78, CHOP, NLRP3, caspase-1, caspase-1/p20, GSDMD-N, and cleaved IL-1β in liver tissue were detected by Western blotting, and the sample numbers were all 5. HepG2 cells as human liver cancer cells were divided into dimethyl sulfoxide (DMSO) group, 0.1 μmol/L tunicamycin (TM) group, 0.2 μmol/L TM group, 0.4 μmol/L TM group, and 0.8 μmol/L TM group and were treated accordingly. After 24 h of culture, cell viability was detected by cell counting kit 8, and the intervention concentration of TM was screened, and the sample number was 5. HepG2 cells were divided into DMSO group, TM alone group, and TM+carnitine group, and treated accordingly. After 24 h of culture, the protein expression levels of GRP78, CHOP, NLRP3, caspase-1, caspase-1/p20, GSDMD-N, and cleaved IL-1β in cells were detected by Western blotting, and the sample numbers were all 3. Data were statistically analyzed with one-way analysis of variance and least significant difference-t test.

At PIH 72, the AST and ALT levels of serum in scald alone group were (640±22) and (157±8) U/L, which were significantly higher than (106±13) and (42±6) U/L in sham-injury group, respectively, with t values of -46.78 and -25.98, respectively, P<0.01. The AST and ALT levels of serum in scald+carnitine group were (519±50) and (121±10) U/L, which were significantly lower than those in scald alone group, respectively, with t values of 4.93 and 6.06, respectively, P<0.01. At PIH 72, the morphology of liver tissue of rats in sham-injury group were basically normal with no obvious inflammatory cell infiltration; compared with those in sham-injury group, the liver tissue of rats in scald alone group showed a large number of inflammatory cell infiltration and disturbed cell arrangement; compared with that in scald alone group, the liver tissue of rats in scald+carnitine group showed a small amount of inflammatory cell infiltration. At PIH 72, the mRNA expression on levels of NLRP3, caspase-1, GSDMD, and IL-1β in liver tissue of rats in scald alone group were significantly higher than those in sham-injury group (with t values of 34.42, 41.93, 30.17, and 15.68, respectively, P<0.01); the mRNA levels of NLRP3, caspase-1, GSDMD, and IL-1β in liver tissue of rats in scald+carnitine group were significantly lower than those in scald alone group (with t values of 34.40, 37.20, 19.95, and 7.88, respectively, P<0.01). At PIH 72, the protein expression levels of NLRP3, caspase-1, caspase-1/p20, GSDMD-N, and cleaved IL-1β in liver tissue of rats in scald alone group were significantly higher than those in sham-injury group (with t values of 12.28, 26.92, 5.20, 10.02, and 24.78, respectively, P<0.01); compared with those in scald alone group, the protein expression levels of NLRP3, caspase-1, caspase-1/p20, GSDMD-N, and cleaved IL-1β in liver tissue of rats in scald+carnitine group were significantly decreased (with t values of 10.99, 27.96, 12.69, 8.96, and 12.27, respectively, P<0.01). At PIH 72, the mRNA levels of GRP78 and CHOP in liver tissue of rats in scald alone group were significantly higher than those in sham-injury group (with t values of 21.00 and 16.52, respectively, P<0.01), and the mRNA levels of GRP78 and CHOP in liver tissue of rats in scald+carnitine group were significantly lower than those in scald alone group (with t values of 8.92 and 8.21, respectively, P<0.01); the protein expression levels of GRP78 and CHOP in liver tissue of rats in scald alone group were significantly higher than those in sham-injury group (with t values of 22.50 and 14.29, respectively, P<0.01), and the protein expression levels of GRP78 and CHOP in liver tissue of rats in scald+carnitine group were significantly lower than those in scald alone group (with t values of 14.29 and 5.33 respectively, P<0.01). After 24 h of culture, the cell survival rates of 0.1 μmol/L TM group, 0.2 μmol/L TM group, 0.4 μmol/L TM group, and 0.8 μmol/L TM group were significantly decreased than that in DMSO group (with t values of 4.90, 9.35, 18.64, and 25.09, respectively, P<0.01). Then 0.8 μmol/L was selected as the intervention concentration of TM. After 24 h of culture, compared with that in DMSO group, the protein expression levels of GRP78 and CHOP in cells in TM alone group were significantly increased (with t values of 10.48 and 17.67, respectively, P<0.01), and the protein expression levels of GRP78 and CHOP in TM+carnitine group were significantly lower than those in TM alone group (with t values of 8.08 and 13.23, respectively, P<0.05 or P<0.01). After 24 h of culture, compared with those in DMSO group, the protein expression levels of NLRP3 and GSDMD-N in cells in TM alone group were significantly increased (with t values of 13.44 and 27.51, respectively, P<0.01), but the protein expression levels of caspase-1, caspase-1/p20, and cleaved IL-1β in cells were not significantly changed (P>0.05); compared with that in TM alone group, the protein expression levels of NLRP3 and GSDMD-N in cells in TM+carnitine group were significantly decreased (with t values of 20.49 and 21.95, respectively, P<0.01), but the protein expression levels of caspase-1, caspase-1/p20, and cleaved IL-1β in cells were not significantly changed (P>0.05).

In severely scald rats, exogenous L-carnitine may play a protective role against liver injury by inhibiting the pathways related to excessive endoplasmic reticulum stress-mediated pyroptosis.