Temperature‐Gradient Transcriptomic Atlas Reveals the PPAR Signaling Pathway Dysfunction in Heat Stroke Induced Liver Injury

ABSTRACT

Heat stroke (HS) is the most severe form of hyperthermia, with mortality exceeding 50% in severe cases. The liver is highly vulnerable to HS‐induced injury, often triggering multi‐organ failure. Although rapid cooling remains the primary treatment, the molecular mechanisms underlying hepatic damage remain elusive, highlighting an urgent need for mechanistic insights, especially given global extreme heat events. We established a HS model by gradually increasing the core temperature of mice from 40°C to 43°C. Mice were sacrificed at each target temperature to collect blood and liver tissues for hematological, biochemical, and histopathological analyses. Transcriptomic profiling was conducted on murine livers, and differentially expressed genes (DEGs) were identified and analyzed. The peroxisome proliferator‐activated receptor (PPAR) signaling pathway was identified as a significantly enriched pathway and 12 key DEGs were validated by reverse transcription quantitative PCR (RT‐qPCR) to assess temperature‐dependent metabolic reprogramming. The expression of CD36, ACOX3, and PPARα was validated by immunohistochemistry at the protein level to investigate their response to heat stress. A graded murine HS model was established and histopathology analysis showed significant liver injury with core temperatures ≥ 42°C, manifesting as weight loss, elevated serum levels of alanine aminotransferase (ALT) and aspartate aminotransferase (AST), neutrophilia, thrombocytopenia, hepatocyte necrosis, and sinusoidal congestion. Transcriptomic profiling revealed temperature‐dependent DEGs from 41°C onward mainly involved in inflammatory/immune, lipid metabolism, apoptosis, and stress response pathways. DEGs consistently dysregulated across different temperatures were enriched in PPAR, insulin signaling, and endoplasmic reticulum (ER) stress‐related pathways. RT‐qPCR analysis revealed the altered expression of PPAR‐related key genes, indicating functional disruption in lipid metabolism. Immunohistochemistry further confirmed these transcriptomic findings at the protein level, suggesting that heat stress induced reprogramming of the PPAR signaling pathway. Together, these findings suggest that HS‐induced liver injury is closely associated with progressive metabolic reprogramming, with dysregulated lipid metabolism playing a central pathogenic role. By combining a murine stepwise HS model and transcriptomic analysis, we identified dysregulated PPAR signaling as a key temperature‐dependent feature of liver injury, suggesting its potential role as a temperature‐sensing node and therapeutic target. This work provides a framework with precise temporal windows and molecular candidates for the development of mechanism‐directed intervention strategies for HS.

Gradient‐ Temperature Atlas Uncovers Hepatic PPAR Metabolic Reprogramming. We present a murine heat‐stroke atlas (40°C–43°C) and identify PPAR‐linked lipid reprogramming as the primary mechanistic signature of temperature‐dependent liver injury, providing time‐resolved molecular targets for mechanism‐based therapies.

Abbreviations

AMPK

adenosine monophosphate‐activated protein kinase

ANOVA

analysis of variance

DAB

diaminobenzidine

DAMPs

damage‐associated molecular patterns

ER

endoplasmic reticulum

EVs

extracellular vesicles

GO

gene ontology

H&E

hematoxylin and eosin

HMGB1

high mobility group box 1

HRP

horseradish peroxidase

HS

heat stroke

IACUC

institutional animal care and use committee

IHC

immunohistochemistry

IL‐1α

interleukin‐1α

IL‐1β

interleukin‐1β

KEGG

Kyoto encyclopedia of genes and genomes

LSECs

liver sinusoidal endothelial cells

Lym

lymphocytes

MODS

multiple organ dysfunction syndrome

MPO

myeloperoxidase

mtROS

mitochondrial reactive oxygen species

NASH

non‐alcoholic steatohepatitis

NCPSB

National Center for Protein Sciences (Beijing)

NETs

neutrophil extracellular traps

Neu

neutrophils

NF‐κB

nuclear factor kappa B

NLRP3

NLR family pyrin domain‐containing protein 3

PLT

platelets

PPAR

peroxisome proliferator‐activated receptor

PT

prothrombin time

RBC

red blood cells

RET

reverse electron transport

ROS

reactive oxygen species

RT

room temperature

RT‐qPCR

reverse transcription quantitative PCR

SD

standard deviation

SREBP‐1c

sterol regulatory element‐binding protein‐1c

TCA

tricarboxylic acid

TLR4/6

Toll‐like receptor 4/6

VLCFAs

very long‐chain fatty acids

WBC

white blood cells

1. Introduction

Heat stroke (HS) is a life‐threatening condition caused by the failure of the body's thermoregulatory system and represents the most severe form of heat illness [1, 2]. The mortality rate among severe cases can exceed 50% [3], and survivors often suffer from significant neurological impairments [4, 5]. This results in a substantial burden on patients' families and public health systems. Although early recognition and rapid cooling remain the cornerstone of treatment, the complex pathophysiological mechanisms of HS are not yet fully understood, severely hindering the development of specific pharmacological therapies and precise intervention strategies. More alarmingly, in the context of global warming, extreme weather events such as heatwaves are becoming more intense and frequent, leading to a significant increase in the incidence of HS [6, 7, 8]. Therefore, elucidating the underlying mechanisms of HS has become a critical and urgent scientific challenge in the fields of emergency and critical care medicine.

The hallmark clinical manifestations of HS include a core body temperature exceeding 40°C and central nervous system dysfunction (e.g., confusion, agitation, seizures) [9]. The onset of HS is characterized by a dynamic process in which various physiological stressors accumulate progressively until they surpass the body's capacity for compensatory thermoregulation. Once thermoregulatory homeostasis is disrupted, the condition rapidly progresses to life‐threatening multiple organ dysfunction syndrome (MODS) [10, 11]. Although some patients may exhibit precursor symptoms such as heat cramps, heat syncope, or heat exhaustion, many severe cases of HS present acutely. Of greater concern, the emergence of clinical symptoms often lags behind early pathological changes at the organ and even molecular levels [12, 13], leading to missed opportunities for timely and effective intervention. Therefore, elucidating this covert early dynamic process is crucial for achieving early warning and halting disease progression.

During the progression from HS to MODS, the liver, owing to its high metabolic activity, rich blood supply, and abundant Kupffer cells, is among the earliest and most vulnerable target organs [14]. Evidence indicates that heat stress can trigger a cytokine storm, endothelial injury, and coagulation abnormalities, leading to hepatic microcirculatory dysfunction and parenchymal damage [15]. As the core barrier of the hepatic microcirculation, liver sinusoidal endothelial cells (LSECs) may, upon activation or injury, aberrantly express and release von Willebrand factor (vWF) multimers. Elevated vWF levels have been closely associated with intrahepatic microthrombus formation [16, 17, 18]. This exacerbates hepatic microcirculatory dysfunction and represents a critical early event in the progression to MODS. Furthermore, aberrant hepatocyte death (including apoptosis and pyroptosis) [19] and Kupffer cell dysfunction [14] play central roles in liver injury. Evidence further suggests that mechanisms such as high mobility group box 1 (HMGB1)‐mediated NLR family pyrin domain‐containing protein 3 (NLRP3) inflammasome activation [20], reactive oxygen species (ROS) burst [21], and impaired mitophagy [22] contribute significantly to liver injury. Notably, recent studies have shown that heat stress induces mitochondrial dysfunction and endoplasmic reticulum stress, leading to the release of large numbers of extracellular vesicles (EVs) from hepatocytes. These EVs can carry various damage‐associated molecular patterns (DAMPs) and may mediate or amplify apoptotic and necroptotic pathways via intercellular communication, creating a self‐perpetuating cycle of injury propagation that exacerbates tissue damage [23, 24, 25].

Although the aforementioned studies have elucidated the downstream mechanisms of HS‐induced liver injury in terms of inflammation, coagulation, cell death, and intercellular communication, current research on this condition remains limited. Existing studies predominantly focus on downstream phenotypic manifestations such as cell death and coagulation disorders, as well as associated molecular changes (e.g., inflammatory factors), while the dynamics of upstream protein networks remain insufficiently characterized [26]. In particular, how this upstream network responds to progressively elevated core temperatures and undergoes step‐wise rewiring during the transition from reversible heat stress to irreversible hepatic damage has yet to be systematically dissected within a temperature‐gradient framework. The failure to clarify the temporal sequence of these molecular events across increasing temperatures has hindered the identification of early warning signs and potential intervention targets. Therefore, establishing a high‐resolution temperature‐gradient model to chart a dynamic temperature‐to‐molecule map from initial heat stress to organ failure represents a key step toward overcoming this bottleneck.

To systematically address these questions, we established a murine HS model featuring precise control over core temperature gradients. By creating a defined temperature gradient, we recapitulated the continuous dynamic changes in liver injury during HS progression. We first integrated histopathological and physiological assessments to systematically characterize liver injury across different stages of heat stress. Subsequently, we performed transcriptomic analysis to construct a high‐resolution temperature‐molecular evolution map, aiming to delineate the critical temperature thresholds, key molecules, and core signaling pathways responsible for the transition from compensatory stress to decompensated injury. This study aimed to uncover the mechanisms of HS‐induced liver injury and to provide a theoretical basis for diagnosis and the development of therapeutic targets.

2. Materials and Methods

2.1. Experimental Animals

All animal experiments were conducted in accordance with established standards and were approved by the Institutional Animal Care and Use Committee (IACUC) of the National Center for Protein Sciences (Beijing) (NCPSB) (Approval No. NCPSB‐20231018‐59 MB). A total of 40 male C57BL/6 mice, aged 8–10 weeks, were purchased from Beijing Huafukang Bioscience Co., Ltd. (Beijing, China). All mice were housed in a temperature‐controlled environment under a 12‐h light/dark cycle with free access to food and water.

2.2. Mouse Model of Heat Stroke

After one week of acclimation feeding, mice were randomly divided into the control (CTRL) group and the HS model group. Model mice were subdivided into 40°C, 41°C, 42°C, and 43°C groups according to the target core temperature during heat exposure. Food and water were withheld throughout the experiment. CTRL mice were maintained at 25.0°C and 65% relative humidity without heat exposure. Model mice were placed in a controlled climate chamber (PEAKS, Shanghai, China); temperature and humidity were initially set at 25°C and 65%, then gradually increased to 39.5°C over 1 h and held there until the end of heat exposure. Core temperature was monitored in real time with a rectal thermometer (KEWBASIS, Nanjing, China); when it reached 40°C, 41°C, 42°C, or 43°C, heat exposure was immediately terminated and the mouse was transferred to a 25°C room for recovery, defining successful modeling. Animals that died or exhibited severe instability in vital signs during heat exposure were excluded from further analysis. In rodent heat‐stroke models, peak parenchymal organ injury occurs between 12 and 24 h; earlier intervals chiefly reflect acute stress, whereas beyond 24 h secondary multi‐organ failure confounds the signature [27]. Thus, mouse liver samples were collected at 16 h post‐heat stress, allowing stable, liver‐injury‐related transcriptional profiles to be captured before late systemic complications emerge.

2.3. Histology Evaluation

Liver tissue samples were collected at 16 h after successful modeling. Tissues were fixed in 10% neutral buffered formalin, embedded in paraffin, and sectioned at a thickness of 4 μm for histological examination. Sections were stained with hematoxylin and eosin (H&E) and evaluated using a blinded, semi‐quantitative scoring system to assess hepatic injury. Briefly, two pathologists, blinded to the experimental group assignments, independently scored the slides according to the Suzuki criteria [28], which evaluate the extent of necrosis, inflammatory infiltration, and ballooning degeneration. The final score for each sample was calculated as the mean of the two independent scores. In cases of disagreement by more than one point, consensus was reached through joint re‐evaluation.

2.4. Biochemical Indicators and Blood Cell Monitoring

At 16 h after successful modeling, mice were anesthetized with Avertin, and 0.2–0.8 mL of blood was collected from the retro‐orbital venous plexus for subsequent analysis. The blood specimen was aliquoted into a clot‐activator tube, followed by 15‐min centrifugation at 350 × g to isolate the supernatant. After a tenfold dilution with 1× PBS, the supernatant was analyzed using a fully automated biochemical analyzer (HITACHI, Japan) to measure serum biochemical markers, including alanine aminotransferase (ALT) and aspartate aminotransferase (AST). The other portion was transferred to an EDTA‐K2 anticoagulant tube, mixed immediately, and analyzed using a fully automated five‐part hematology analyzer (Mindray, Shenzhen, China) to determine the absolute counts of neutrophils (Neu), white blood cells (WBC), platelets (PLT), lymphocytes (Lym), and red blood cells (RBC).

2.5. RNA Sequencing

To extract total RNA, mouse livers were snap‐frozen in liquid nitrogen, powdered, and processed with the TRNzol Universal Reagent Kit (TIANGEN Biotech, Beijing, China) following the manufacturer's instructions. RNA concentration and purity were measured using a NanoDrop One/OneC spectrophotometer (Thermo Fisher, USA). Samples with an A260/A280 ratio between 1.8 and 2.1 were considered qualified. RNA integrity was assessed using the Qsep 400 system (BiOptic, Taiwan, China), and only samples with an RNA Integrity Number (RIN) ≥ 7.0 were used for library construction.

The transcriptome sequencing library was constructed using the VAHTS Universal V8 RNA‐seq Library Prep Kit for Illumina (Vazyme, Nanjing, China). Starting with 1 μg of total RNA, mRNA with a poly(A) tail was captured using magnetic beads coated with oligo(dT). The captured mRNA was then chemically fragmented. Subsequent steps included reverse transcription, cDNA purification, end repair, A‐tailing, adapter ligation, and PCR amplification.

Library concentration was quantified using a Qubit 4.0 Fluorometer (Invitrogen, USA) with the Qubit dsDNA HS Assay Kit, and fragment size distribution was assessed using the Agilent Bioanalyzer 2100 system (Agilent Technologies, USA) with the High Sensitivity DNA chip, requiring a main peak between 300 and 350 bp. Qualified libraries were pooled in equimolar amounts and subjected to 2 × 150 bp paired‐end sequencing on the Illumina NovaSeq X Plus platform (Illumina, USA), with a target output of ≥ 6 Gb raw reads per sample.

2.6. RNA Extraction and Quantitative Real‐Time PCR (qPCR)

Mouse liver tissues were rapidly frozen in liquid nitrogen and homogenized prior to total RNA extraction with the SteadyPure Universal RNA Extraction Kit (Accurate Biology, Hunan, China), following the supplier's instructions. The concentration and purity of the extracted RNA were evaluated by measuring absorbance on the NanoDrop spectrophotometer, and only samples exhibiting an A260/A280 ratio within the range of 1.8–2.1 were selected for further experiments. Reverse transcription was performed according to the kit manufacturer's guidelines in 20 μL reaction mixtures. The reaction conditions were 37°C for 15 min, followed by 85°C for 5 s to inactivate the reverse transcriptase.

Reverse transcription quantitative PCR (RT‐qPCR) was performed on a qTOWER 2.2 Real‐Time PCR System (Analytik Jena, Germany). Each 20 μL reaction mixture contained 10 μL of 2× SYBR Green SupTaq HS Premix (Low Rox Plus) (Accurate Biology, Hunan, China), 1.6 μL each of forward and reverse primers (10 μM), 2 μL of cDNA template, and nuclease‐free water to a final volume of 20 μL. The thermal cycling protocol was as follows: initial denaturation at 95°C for 30 s; 40 cycles of 95°C for 5 s and 60°C for 30 s. A melt curve analysis (60°C–95°C) was conducted to confirm the specificity of amplification. All experiments were performed with three biological replicates. The relative gene expression levels were calculated using the 2^−ΔΔCT method and normalized to the endogenous control Gapdh. All primers (listed in Table S1) were synthesized by Suzhou Junji Biotechnology Co., Ltd. (Suzhou, China).

2.7. Immunohistochemistry

To further investigate the effects of HS on the expression of hepatic lipid metabolism–related proteins and to integrate these findings with the mRNA changes observed in our prior RT‐qPCR analysis, immunohistochemistry (IHC) was performed to assess the expression levels of CD36, ACOX3, and peroxisome proliferator‐activated receptor alpha (PPARα) proteins.

Mouse liver tissues were fixed in 4% paraformaldehyde, embedded in paraffin, and serially sectioned at a thickness of 4 μm. Sections were deparaffinized and rehydrated through a graded ethanol series, followed by microwave‐mediated antigen retrieval using sodium citrate buffer (10 mM, pH 6.0). Endogenous peroxidase activity was blocked by incubation with the peroxidase‐blocking reagent provided in the kit (ZSGB‐BIO, Beijing, China) at room temperature (RT) for 10 min. Sections were then incubated overnight at 4°C with the following primary antibodies: rabbit anti‐mouse CD36 antibody (1:80, Affinity, DF13262, Jiangsu, China), mouse anti‐mouse ACOX3 antibody (1:60, Santa Cruz Biotechnology, sc‐373,977, USA), and mouse anti‐mouse PPARα antibody (1:60, Santa Cruz Biotechnology, sc‐398,394, USA). The next day, the CD36 staining group was processed using the ZSGB‐BIO Rabbit Two‐Step Detection Kit, while the ACOX3 and PPARα staining groups were processed using the Mouse Two‐Step Detection Kit (ZSGB‐BIO, Beijing, China), with all remaining steps performed strictly according to the manufacturer's instructions.

Sections were sequentially incubated with biotinylated secondary antibodies and horseradish peroxidase (HRP)‐conjugated streptavidin, followed by diaminobenzidine (DAB) chromogenic development and hematoxylin counterstaining. Finally, slides were dehydrated through a graded ethanol series, cleared in xylene, and mounted with neutral balsam. All images were acquired under identical exposure parameters, and quantitative analysis was performed using ImageJ: the percentage of positive staining area was measured for CD36 and ACOX3, whereas the number of PPARα‐positive nuclei was counted.

2.8. Data Analysis

Transcriptomic sequencing data were analyzed for differential expression using the DESeq2 package (v1.36.0) in R (v4.2.0). The raw gene count matrix was used as input, and normalization, negative binomial modeling, and hypothesis testing were performed following the standard DESeq2 pipeline. Differentially expressed genes (DEGs) were defined by the criteria: |log₂FoldChange| > 0.585 (approximately corresponding to a 1.5‐fold change) and an adjusted p‐value < 0.01 using the Benjamini‐Hochberg (B‐H) method. Functional enrichment analysis was performed on the identified DEGs. Gene Ontology (GO) analysis was conducted using the online platform DAVID (https://david.ncifcrf.gov/) to annotate biological processes, and the Kyoto Encyclopedia of Genes and Genomes (KEGG) database was used to identify enriched biological pathways. Significance of enrichment results was determined using a threshold of Benjamini‐Hochberg adjusted p < 0.05.

All experimental data are presented as mean ± standard deviation (SD). Group comparisons were performed using one‐way analysis of variance (ANOVA) in GraphPad Prism 9.0 (GraphPad Software, USA). Tukey's or Sidak's multiple comparisons test was applied for post hoc correction. The significance level was set at α = 0.05. p < 0.05 (*), 0.01 (**), or 0.001 (***) was considered statistically significant.

3. Results

3.1. Phenotypic Alterations in HS Mice

To characterize HS‐induced liver injury under controlled hyperthermia, a graded core temperature model (40°C, 41°C, 42°C, and 43°C) was established in mice. In accordance with an NCPSB‐IACUC‐approved protocol, mice were maintained under deep anesthesia and then euthanized immediately following the attainment of the target core temperature. Blood and liver samples were collected for subsequent analyses (Figure 1A). Compared with controls, HS mice exhibited a significant body weight loss ranging from 3% to 13%, and the loss rate increased with core temperature (Figure 1B). Gross examination of the liver revealed no visible changes in the 40°C and 41°C groups. At 42°C, prominent vascular markings appeared on some lobes. Livers from the 43°C group displayed diffuse pathological changes, including dark‐red vascular tracks, mild hepatomegaly, and pallor (Figure 1C).

FIGURE 1.

Phenotypic characteristics of the liver in HS mice. (A) Schematic illustration of the experimental procedure for mouse model establishment and sample collection. HS mouse models with graded core temperatures (40°C, 41°C, 42°C, and 43°C groups) were established using a controlled temperature‐ and humidity‐ controlled environmental climate chamber. Tissue and blood samples were collected at 16 h after model induction. Continuous dynamic changes in liver injury during HS progression were monitored through biochemical indicators, blood cell analysis, histopathology, and transcriptomics. (B) Percentage of body weight loss during HS modeling. Data are presented as mean ± SD, n = 8 per group. Significance between experimental groups and the control group was evaluated by one‐way ANOVA followed by Tukey's post hoc test: ***p < 0.001. (C) Representative images of liver sections from control and HS mice (scale bar: 5 mm). No obvious macroscopic differences were observed in the 40°C and 41°C groups compared to controls. At 42°C, prominent vascular markings were evident on several lobes. At 43°C, livers displayed diffuse injury manifested by dark‐red congestion, mild hepatomegaly, and pallor.

These results indicate that HS induces gross structural abnormalities in the mouse liver at core temperatures ≥ 42°C. Moreover, the severity of these lesions progressed with increasing temperature, suggesting that HS‐induced macroscopic hepatic alterations are temperature‐dependent.

3.2. HS‐Induced Liver Injury in Mice

To further evaluate the impact of HS on liver injury, we collected peripheral blood and liver tissue to assess blood cell counts, serum liver enzyme levels, and histopathological changes. Hematological analysis showed that neutrophils began to increase at 41°C and were significantly higher at 42°C (p < 0.05) and 43°C (p < 0.01, Figure 2A). Total white blood cell and lymphocyte counts started to decline at 42°C and were significantly lower than control values at 43°C (p < 0.01, Figure 2B,C). Platelet counts were significantly reduced at 42°C (p < 0.05) and 43°C (p < 0.001, Figure 2D). Red blood cell counts decreased stepwise with rising core temperature (Figure 2E). Serum biochemistry analysis revealed significant elevations of AST and ALT at 42°C (p < 0.001 and p < 0.05, respectively), indicative of hepatocellular injury. Levels of both enzymes remained highly elevated at 43°C (p < 0.001, Figure 2F,G). Hematoxylin and eosin (H&E) staining revealed no obvious differences between the control and 40°C groups. In the 41°C group, however, hepatocellular edema and erythrocyte aggregation in sinusoids and central veins were observed. The 42°C group exhibited edema, spotty necrosis, and sinusoidal congestion; these lesions were exacerbated at 43°C, accompanied by erythrocyte stasis in central veins and loss of hepatic lobular architecture (Figure 2H). The overall histological injury score increased significantly with temperature (Figure 2I).

FIGURE 2.

HS‐induced liver injury in mice. Peripheral blood cell counts in HS mice, including (A) neutrophils (Neu), (B) white blood cells (WBC), (C) lymphocytes (Lym), (D) platelets (PLT), and (E) red blood cells (RBC). Serum biochemical analysis of (F) aspartate aminotransferase (AST) and (G) alanine aminotransferase (ALT). (H) Representative H&E‐stained liver sections. Scale bars: 100 μm (× 20), 50 μm (× 40). (I) Liver injury score quantification based on histopathological assessment of liver tissue sections. Data are presented as mean ± SD. Group differences were analyzed by one‐way ANOVA. ***p < 0.001.

Collectively, core temperatures ≥ 42°C elicit multifaceted abnormalities, including neutrophilia, thrombocytopenia, elevated transaminases, and architectural disruption, whose severity escalates progressively with increasing temperature.

3.3. Transcriptomic Analysis Reveals Molecular Signatures of HS‐Induced Liver Injury in Mice

To investigate the molecular mechanisms underlying HS‐induced liver injury, we performed transcriptomic analysis on liver tissues from mice exposed to graded core temperatures. Compared to the control group, 233, 559, 390, and 3058 DEGs were identified in the 40°C, 41°C, 42°C, and 43°C groups, respectively (Figure 3A). The 40°C group exhibited relatively few DEGs. Upregulated genes were primarily enriched in biological processes such as protein degradation and stress response, whereas downregulated genes were enriched in pathways related to signal transduction, metabolic process, and ion homeostasis (Figure 3B). These results suggest that at 40°C, the liver has initiated early stress resistance responses under heat exposure.

FIGURE 3.

Transcriptomic profiling of HS‐induced liver injury in mice. (A) Volcano plots of DEGs identified in mouse livers from comparisons between control (n = 5) and 40°C, 41°C, 42°C or 43°C treatment groups (n = 5 for each group). Red: Up‐regulated; blue: Down‐regulated; gray: Non‐significant. Threshold: |log2FC| > 0.585 and B‐H adjusted p < 0.01. Heatmaps displaying the expression of DEGs identified in mouse livers by comparisons between the control group with the 40°C (B), 41°C (C), 42°C (D), and 43°C (E) groups, respectively (n = 5 per group). Right: Top GO biological‐process terms enriched among up‐regulated (red bars) and down‐regulated (blue bars) DEGs.

At 41°C, upregulated genes were mainly enriched in transcriptional regulation and canonical signaling pathways. Downregulated genes were enriched in inflammation, immune responses, cytokine regulation, lipid metabolism, energy metabolism, and apoptosis regulation (Figure 3C). Notably, lipid metabolism‐related functions exhibited sustained downregulation starting at 41°C. These findings indicate that transcriptional regulation is enhanced at this stage, while immune responses, metabolic processes, and inflammatory pathways begin to be suppressed at the transcriptional level.

In the 42°C group, upregulated genes were enriched in apoptosis, insulin regulation, protein degradation, inflammation, and stress response. Downregulated genes were involved in lipid metabolism, immunity, and cellular stress response (Figure 3D). These results suggest that under sustained hyperthermia, hepatocytes suffer severe damage and cell death pathways are activated.

The number of DEGs increased significantly in the 43°C group. Upregulated genes were associated with ion homeostasis, stress response, regulation of angiogenesis, apoptosis, and cytoskeletal dynamics. Downregulated genes were involved in innate immunity, adaptive immunity, inflammatory response, apoptosis regulation, lipid metabolism, and glucose metabolism (Figure 3E). This indicates that under extreme hyperthermia, the liver enters a state of severe functional imbalance and structural disruption.

Collectively, as core body temperature increased from 40°C to 43°C, the hepatic molecular response followed a distinct evolutionary pattern: transitioning from early transcriptional activation and stress regulation to progressive suppression of inflammatory and immune functions, and ultimately manifesting as significant dysregulation of apoptosis, angiogenesis, and cytoskeletal remodeling. Notably, lipid metabolism, inflammation, and immune responses exhibited sustained dysregulation starting at 41°C, a change that closely correlates with the progressively worsening liver pathology. This suggests that these processes may represent key molecular events in HS‐induced liver injury.

3.4. Identification and Functional Enrichment Analysis of Persistent DEGs Under Varying Heat Stress Conditions

Based on differential expression analysis, we identified the shared DEGs from each group (40°C, 41°C, 42°C, and 43°C) compared to the control group, aiming to characterize core transcriptional signatures consistently altered under different thermal stresses. A total of 74 DEGs were commonly dysregulated across all HS groups relative to controls (Figure 4A). In‐depth analysis of these core genes may reveal stable molecular events underlying HS‐induced liver injury. Specifically, the commonly upregulated genes included Ddit3 (a key mediator of endoplasmic reticulum (ER) stress and the unfolded protein response) and Irs1 (a key insulin‐receptor substrate) (Figure 4B). The commonly downregulated genes included Me1, which encodes an enzyme that generates NADPH required for de novo lipogenesis; its downregulation may contribute to impaired redox balance and disrupted metabolic homeostasis in hepatocytes (Figure 4B).

FIGURE 4.

Identification of persistently dysregulated genes across graded core temperatures and their functional enrichment. (A) Venn diagram illustrating the overlap of DEGs between the control group (CTRL) and each HS group at 40°C, 41°C, 42°C, and 43°C. The intersection represents 74 shared DEGs. (B) Heatmap depicting the expression levels of the 74 shared DEGs across the CTRL and HS groups (40°C–43°C). Red indicates upregulation; blue indicates downregulation. (C) Functional enrichment analysis of the 74 DEGs using GO and KEGG databases. Bar length represents the enrichment significance, expressed as –log₁₀ (p‐value); higher values indicate greater statistical significance. Bubble size corresponds to the number of genes enriched in each term. Enriched pathways are color‐coded: KEGG pathways (green), GO biological process (BP, red), GO molecular function (MF, purple), and GO cellular component (CC, blue).

To further explore the potential biological significance of these persistently dysregulated genes in heat stress‐induced liver injury, we performed GO annotation and KEGG pathway enrichment analysis on the 74 core DEGs. These genes were significantly enriched in biological processes of fatty acid metabolism (Ankrd23, Acot11, and Lpl) and regulation of cell proliferation (Irs1, Sox9, and Syne1). Their molecular functions and cellular components were primarily associated with DNA binding, transcription factor activity, and the nucleus (Bmal1, Npas2, and Sox9). KEGG analysis identified the PPAR (Fabp5, Me1, and Lpl) and insulin (Irs1, Ppp1r3c, and Slc2a4) signaling pathways as the top significantly enriched pathways. Notably, some genes within these pathways exhibited opposing regulatory trends: Me1, essential for fatty acid synthesis, was downregulated; whereas genes such as Chka (a rate‐limiting enzyme for phosphatidylcholine synthesis), Irs1 (a core component of the insulin signaling pathway), Ppp1r3c (participating in glycogen synthesis), and Ddit3 (a key transcription factor inducing apoptosis under endoplasmic reticulum stress) were persistently upregulated compared to the control group and reached the highest expression levels at 43°C.

Collectively, under graded HS conditions, coordinated dysregulation of core metabolic and stress‐response pathways persists, which may represent stable molecular hallmarks of HS‐induced liver injury. Sustained heat stress may contribute to hepatic damage by inducing systemic imbalance in core metabolic pathways.

3.5. Expression Dynamics of Genes and Key Proteins in the PPAR Signaling Pathway

To further investigate the role of metabolic dysregulation in HS‐induced liver injury, we selected the PPAR signaling pathway for detailed analysis based on KEGG enrichment results, as it was the most significantly enriched pathway. Then, we combined all DEGs across all HS groups and mapped them to the PPAR pathway. Ultimately, 12 DEGs were identified and subsequently validated by RT‐qPCR, including Me1, Scd1, Slc27a1, Fabp3, Lpl, Cd36, Perilipin, Fabp4, Acadm, Acox3, Aqp7, and Slc27a6 (Figure 5A). To further validate the role of the PPAR signaling pathway in HS‐induced hepatic metabolic dysfunction at the protein level, we examined the expression changes of key proteins in the PPAR pathway, including CD36, ACOX3, and PPARα in the liver tissues of HS mice by IHC (Figure 5B).

FIGURE 5.

RT‐qPCR and immunohistochemical staining validation of key components of PPAR signaling pathway in HS mouse livers. (A) Relative expression levels of PPAR signaling pathway‐related genes measured by RT‐qPCR. Data are presented as mean ± SD (n = 3 per group). Statistical differences among groups were analyzed by one‐way ANOVA; ***p < 0.001 vs. control group. (B) Immunohistochemical staining of CD36, ACOX3, and PPARα in mouse livers with graded core temperatures. Scale bars: 25 μm (× 80, CD36), 50 μm (× 40, ACOX3 and PPARα). (C) Quantitative analysis of CD36‐positive area. (D) Quantitative analysis of ACOX3‐positive area. (E) Quantitative analysis of PPARα‐positive nuclei. Data are presented as mean ± SD (n = 3 per group). Statistical differences among groups were analyzed by one‐way ANOVA; ***p < 0.001 vs. control group.

RT‐qPCR results revealed a pronounced temperature‐dependent pattern in the expression of PPAR pathway‐related genes. Specifically, genes involved in fatty acid synthesis, such as Me1 and Scd1, were significantly downregulated across the temperature gradient. Genes associated with fatty acid uptake exhibited various patterns in which Slc27a6 was significantly downregulated after 42°C, while Slc27a1 was significantly upregulated at 43°C; Fabp3 and Lpl displayed a biphasic pattern characterized by downregulation at the early stage, followed by upregulation with increasing temperature; Cd36 showed a non‐significant increasing trend with temperatures ≤ 42°C and was significantly upregulated at 43°C (Figure 5A). IHC revealed a significant upregulation of CD36 protein at 42°C, which further increased at 43°C (Figure 5B,C). This finding aligns with the RT‐qPCR data, confirming its temperature‐dependent activation. Importantly, the earlier detection of CD36 protein elevation compared to its mRNA suggests potential post‐transcriptional regulation during early heat stress.

Genes related to lipid droplet formation and fat storage, including Perilipin and Fabp4, were upregulated in a temperature‐dependent manner. The biphasic expression pattern of Fabp3 and Lpl suggests a potential shift in hepatic lipid management strategies across different stages of heat stress. A key gene for peroxisomal β‐oxidation, Acox3, was continuously and significantly downregulated (Figure 5A). IHC results showed that ACOX3 protein levels were persistently downregulated as core temperature increased, showing significant suppression as early as 41°C and remaining inhibited at 42°C and 43°C (Figure 5B,D). The consistent downregulation of Acox3 at both the mRNA and protein levels indicates that this gene is subject to coordinated transcriptional and translational repression within the PPAR signaling pathway. In contrast, Acadm, the rate‐limiting enzyme for mitochondrial β‐oxidation, showed a compensatory upregulation trend at 43°C. The glycerol permeability gene Aqp7 was significantly upregulated at 43°C, which may promote intracellular glycerol utilization in the liver (Figure 5A).

In addition, PPARα, the core transcription factor, exhibited a progressive decline across the entire temperature gradient, with a significant reduction observed starting at 41°C and reaching its lowest level at 43°C (Figure 5B,E). The progressive decline in PPARα may directly impair the transcription of downstream oxidation‐related genes, leading to compromised overall fatty acid oxidation capacity.

Therefore, the expression of PPAR pathway‐related genes exhibited a distinct dynamic pattern with rising core body temperature. At 40°C–41°C, the downregulation of fatty acid synthesis genes such as Me1 and Scd1, along with partial suppression of uptake genes, suggested an early reduction in anabolic metabolism. When core body temperature reached 42°C, increased expression of lipid storage genes such as Perilipin and Fabp4 indicated a shift toward lipid accumulation. Furthermore, when the temperature reached 43°C, fatty acid uptake genes including Cd36 were significantly upregulated, while peroxisomal β‐oxidation, represented by Acox3, remained suppressed. Despite the compensatory upregulation of Acadm, overall fatty acid oxidation capacity appeared to decline (Figure 6).

FIGURE 6.

Schematic of gene expression changes in the PPAR signaling pathway in the liver under graded heat stress conditions. Gene expression changes shown in boxes are based on the average values from RT‐qPCR experiments across treatment groups. Green indicates downregulation; red indicates upregulation.

In summary, under HS conditions, key molecules of the PPAR signaling pathway exhibit functional imbalance, characterized by significantly enhanced CD36‐mediated fatty acid uptake and concurrently sustained suppression of the ACOX3‐ and PPARα‐dependent β‐oxidation pathway. This metabolic reprogramming may lead to net hepatic lipid accumulation, thereby exacerbating metabolic dysfunction and paralleling the severity of temperature‐dependent liver injury.

4. Discussion

This study establishes a transcriptomic‐based molecular map of HS‐induced liver injury, revealing that the PPAR signaling pathway is progressively suppressed as core temperature rises. Our findings suggest that this metabolic dysfunction represents an early and sustained event that precedes and may contribute to the subsequent inflammatory and coagulatory phases of hepatic damage.

We identified a temperature‐dependent molecular pathological trajectory. At 40°C, transcriptomic changes were limited, primarily involving protein degradation and ion homeostasis, indicating that the liver was in an early adaptive stress state [29]. At 41°C, hepatic physiological dysfunction emerged [30, 31], opening a molecular warning window before overt organ damage. This process escalated at 42°C and culminated at 43°C with hematologic abnormalities and histologic disruption (Figure 2).

The core feature of this progression was metabolic reprogramming following suppression of the PPAR signaling pathway. This was characterized by inhibited fatty acid synthesis, enhanced uptake and storage, and impaired oxidation (Figure 6). The expression of fatty acid synthesis genes Me1 and Scd1 was downregulated (Figure 5A), suggesting suppression of the sterol regulatory element‐binding protein‐1c (SREBP‐1c)‐mediated lipogenesis program [32]. Concurrently and more critically, our IHC results showed that PPARα protein was significantly downregulated under heat stress (Figure 5E). PPARα is a master transcriptional regulator of hepatic fatty acid β‐oxidation, and its dysfunction is closely linked to metabolic disorders. For example, reduced PPARα expression was observed in liver diseases such as non‐alcoholic steatohepatitis (NASH) [33], and Ppara‐knockout mice exhibited severe defects in fatty acid oxidation and pronounced hepatic lipid accumulation [34]. Clinically, the PPARα agonist fenofibrate lowers serum triglycerides by activating this pathway [35]. Thus, the PPARα downregulation observed in our model likely contributes directly to the impaired fatty acid oxidation. Although the downregulation of lipogenic genes may represent a compensatory adaptation to limit de novo lipid synthesis, this protective response is likely overwhelmed by the combined effects of enhanced fatty acid uptake (via CD36) and suppressed β‐oxidation, ultimately promoting hepatic steatosis during severe heat stress.

Under sustained heat stress, genes central to liver metabolic functions exhibited progressive suppression. This combination of enhanced uptake and storage with impaired clearance capacity could lead to lipid metabolic imbalance and promote lipid accumulation. Metabolic dysfunction may disrupt cellular homeostasis, thereby activating the NLRP3 inflammasome [36]. Moreover, elevated levels of free fatty acids resulting from lipid accumulation can directly induce the formation of neutrophil extracellular traps (NETs) [37]. Importantly, studies have confirmed that NETs and the NLRP3 inflammasome form a positive feedback loop that exacerbates liver injury [38]. Therefore, based on the current correlative data, we propose that the CD36–NLRP3–NETs axis contributes to liver injury as a working hypothesis. To formally establish causality, future work should employ genetic ablation (e.g., Cd36 or Nlrp3) and/or pharmacological inhibitors targeting CD36, NLRP3, or PAD4‐mediated NET formation in both in vivo and in vitro settings. Our findings suggest that the intrahepatic self‐amplifying loop may constitute a critical trigger for systemic complications. Synchronous hepatocyte membrane rupture and NET formation during heat stress predispose to the discharge of substantial liver‐derived DAMPs (e.g., HMGB1, histones, mtDNA) into the circulation [39]. These molecules may subsequently travel via the bloodstream to the pulmonary and renal microvasculature, where they can activate endothelial cells through the toll‐like receptor 4 (TLR4)/NLRP3‐tissue factor axis, potentially instigating distal immunothrombosis and tissue injury [40]. Consequently, this pathway may functionally link localized hepatic damage to systemic multiple organ dysfunction, thereby transforming the liver from a passive target into an active danger‐signal amplifier in HS.

As core body temperature further increased to 43°C, expression of the fatty acid transporter gene Cd36 was upregulated. However, insufficient activation of PPARα/δ prevented timely transfer of incoming fatty acids into β‐oxidation, potentially leading to an imbalance between uptake and oxidation. Meanwhile, key enzymes in peroxisomal β‐oxidation such as Acox3 were downregulated, a direct consequence of reduced PPARα transcriptional activity, ultimately resulting in further abnormal lipid accumulation within hepatocytes [41]. Besides, Acox3 was continuously and significantly downregulated, forming an oxidative bottleneck at the peroxisomal level, which impeded β‐oxidation of very long‐chain fatty acids (VLCFAs). In contrast, the mitochondrial β‐oxidation rate‐limiting gene Acadm was significantly upregulated at 43°C in a compensatory manner (Figure 5A), indicating that VLCFAs must be shortened before entering the mitochondria and consequently causing a surge of long‐chain fatty acids into the organelle. However, constrained tricarboxylic acid cycle (TCA cycle) flux elevated succinate and depleted coenzyme Q (CoQ), provoking reverse electron transport (RET) and mitochondrial reactive oxygen species (mtROS) to trigger NLRP3‐mediated pyroptosis [42, 43]. This adaptive attempt to balance energy and ROS manifested as a selective restriction of fatty acid oxidation [44, 45]. Notably, Fabp3 and Lpl showed an early suppression followed by a late rebound. We speculate that this biphasic pattern reflects an initial, adaptive down‐regulation to limit lipid influx and prevent lipotoxicity. However, when the hepatic fatty‐acid handling capacity is chronically overwhelmed, this response switches to a maladaptive overshoot. The resulting accumulation of lipid intermediates and ROS may then prime neutrophils for NET formation, as previously described for lipotoxic environments [46, 47, 48]. Functional studies are required to test this hypothesis.

Furthermore, NETs, enriched in histones and myeloperoxidase (MPO), directly damage hepatocyte membranes. They also serve as scaffolds that trap platelets and activate the coagulation cascade, promoting immunothrombosis in the hepatic sinusoids [49]. In this study, the observed hematologic alterations constitute a characteristic hematologic triad (Figure 2A,B). Thus, lipid accumulation, NLRP3 activation, and NET release engage in a self‐amplifying loop, with this triad reflecting the systemic manifestation. Targeting NET formation or NLRP3 activation may therefore break this cycle and mitigate liver injury.

Beyond lipid uptake, human CD36 modulates inflammation, apoptosis, and platelet activation [50, 51], suggesting a multifaceted role in HS. Since significant liver injury emerges at 42°C, this temporal expression shift suggests that CD36 may transition from a protective silenced state to an active role in injury amplification. Mechanistically, the CD36‐TLR4/6‐nuclear factor kappa B (NF‐κB)/ROS axis provides a priming signal for NLRP3, while lysosomal damage provides the activation signal, triggering caspase‐1‐dependent interleukin‐1β (IL‐1β) and interleukin‐1α (IL‐1α) release [52]. Based on this, we propose a model in which metabolic stress, together with the observed upregulation of Cd36, could provide a priming signal (potentially via TLR4/6 and ROS) for NLRP3 inflammasome activation, culminating in caspase‐1‐dependent IL‐1β release. As a potent procoagulant, IL‐1β can drive immunothrombosis. Therefore, we hypothesize that the 43°C‐specific surge in Cd36 expression may act as a critical amplifier, converting hepatic lipotoxic stress into local microcirculatory dysfunction and systemic inflammation via a putative CD36‐NLRP3‐IL‐1β axis. This positions CD36 as a candidate central node linking metabolic dysregulation to coagulopathy in HS. Consequently, targeting this node represents a theoretically promising strategy that could simultaneously address multiple facets of HS‐induced liver injury.

5. Conclusion

Using a murine model with a core‐temperature gradient of 40°C–43°C, this study systematically maps the temperature‐to‐molecule trajectory of HS‐induced liver injury and reveals the phased suppression of PPAR signaling as a defining, temperature‐dependent transcriptional signature. This signature precedes downstream metabolic, inflammatory, and coagulatory dysregulation, suggesting that PPAR signaling acts as a central regulatory node and a candidate mediator of the hepatic response to heat stress. These findings are integrated into a testable framework with defined temporal windows and prioritized molecular targets, establishing a rationale for future investigation into PPAR‐directed, mechanism‐based metabolic immunomodulation.

Author Contributions

Yunping Zhu and Jie Ma: conceptualization; Ying Zhu, Chunyuan Yang, Teng Ma, and Jinxu Zhang: data curation; Ying Zhu, Wanlin Liu, and Mingfei Han: formal analysis; Yunping Zhu and Jie Ma: funding acquisition; Ying Zhu and Jie Ma: investigation; Ying Zhu and Jie Ma: methodology; Yunping Zhu and Jie Ma: project administration; Yunping Zhu and Jie Ma: resources; Ying Zhu, Wanlin Liu, and Mingfei Han: software; Yunping Zhu and Jie Ma: supervision; Ying Zhu, Wanlin Liu, and Jie Ma: validation; Ying Zhu and Wanlin Liu: visualization; Ying Zhu and Wanlin Liu: writing – original draft; Yunping Zhu and Jie Ma: writing – review and editing.

Funding

This work was funded by the National Natural Science Foundation of China (32270697), the National Key Research Program of China (2021YFA1301603), and the Open Project Program of the State Key Laboratory of Medical Proteomics (SKLP‐O202204).

Ethics Statement

The study was conducted in accordance with established standards and was approved by the Institutional Animal Care and Use Committee of the National Center for Protein Sciences (Beijing) (Approval No.: NCPSB‐20231018‐59 MB).

Consent

All data in this study are anonymized and no individual identifiers are included.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Table S1: Real time‐qPCR primer base sequence.

Data Availability Statement

Stored in repository.