Severe heat stress modulated nuclear factor erythroid 2-related factor 2 and macrophage migration inhibitory factor pathway in rat liver

Abstract

Heat stress impairs physiology and overall functionality of the body at tissue and organ level in animals. Liver being a vital organ performs more than hundreds regulatory functions of the body. Present study investigates the modulation of molecular pathways that are responsible for liver damage triggered by heat stress. Male Sprague dawley rats were exposed to heat stress (45 °C) in heat simulation chamber till core temperature reaches 40 °C and 42 °C in 25 and 42 min respectively. For in-depth evaluation of liver functions during severe heat stress, hepatic transcriptome and proteome were analysed by microarray and two dimensional gel electrophoresis respectively. Results revealed major alterations in redox status, inflammation, mitochondrial dysfunction and proteostasis related pathways. Data of molecular pathway analysis demonstrate that nuclear factor erythroid 2-related factor 2 (NRF-2) mediated oxidative stress response and macrophage migration inhibitory factor (MIF) regulated inflammatory pathways were upregulated in severe heat stressed liver. Expression levels of downstream molecules of above pathways such as heat shock protein 90AB 1, peroxiredoxin 5, Jun N-terminal kinases 1/2, heme-oxygenase 1, apolipoprotein 1 and interleukin 10 were examined and result suggested the upregulation of these genes modulates the NRF-2 and MIF regulated pathways in heat stressed liver. Irregularity in molecular signalling networks lead to mitochondrial dysfunction indicated by upregulation of ATP synthase β and peroxiredoxin 1 along with decreased levels of glucose-6-phosphate dehydrogenase and enhanced activity of cytochrome c in liver mitochondria. Thus, current study demonstrated heat induced alterations in key liver functions were regulated by NRF-2 and MIF pathways.

Supplementary Information

The online version contains supplementary material available at 10.1007/s12079-022-00668-0.

Introduction

Fluctuating environmental changes results in high ambient temperature (T_a) which leads to heat related illnesses and eventually increased morbidity and mortality (Poumadère et al. 2005). Several heat related mortalities and uncounted severe heat stress (SHS) conditions has been reported in the era of 1971–2020 and future estimation also suggest that there will be rapid increase in heat related cases globally (Y. Guo et al. 2018). According to the world meteorological organization approximately 220 million peoples were exposed to severe heat waves in the year 2016 (Haustein et al. 2017). Severity of heat stress (HS) has resulted more than 7,00 and 1,25,000 deaths in last two decades in United States of America and Europe respectively (Barriopedro et al. 2011; Robine et al. 2008; Semenza et al. 1996). Russia alone had a mortality count of 55,000 in the year 2010 (Barriopedro et al. 2011; Hoag 2014). In India, during the year 2001–2005 mortality counts was about 3,863 which further increased upto 6,429 between the years 2011–2015 (Kumar & Singh 2021; Mazdiyasni et al. 2017). In the year 2010, heat waves in Gujarat state hits total mortality counts of 1,344 (Azhar et al. 2014). Exposure to high T_a results in heat cramp, heat exhaustion and heat stroke followed by a sequential organ damage leading to multiple organ failure. Heat stroke is characterized by an increased core body temperature (T_c ≥ 40.5 °C) which elicits alterations in physiological, biochemical and cellular levels including changes in transcriptional and translational levels (Epstein & Roberts 2011). In addition to this, higher T_a results in dehydration with lack of fluid replacement in the body leading to loss of blood volume resulting in improper distribution of nutrients and oxygen supply to vital organs of the body. Continuous exposure to high environmental temperature leads to cardiovascular collapse and eventually multi-organ dysfunction due to heat stroke (Leon & Helwig 2010a, b).

Clinical investigations on humans were performed in which heat stressed patients showed 100% neurological impairment, 30% gastrointestinal dysfunction, along with involvement of hepatic (34%) and kidney (57%) damage (Kalaiselvan et al. 2015; Weigand et al. 2007). Severe aberrations and dysfunctionalities have been observed in humans exposed to various HS conditions such as systemic inflammatory response syndrome (SIRS) which is the primary cause of organ damage resulting in acute respiratory distress syndrome, rhabdomyolysis, hepatocellular injury, hemorrhagic complications, acute renal failure, intestinal ischemia and pancreatic injury. Results of a case study suggest that 30% of heat stroke survivors experienced neurological dysfunction and disability even after cooling of the whole body (Argaud et al. 2007; Dematte 1998). In hyper-thermic conditions, SIRS is more complicated with endotoxemia and disseminated intravascular coagulation (DIC) due to leakage in gut epithelium (Leon & Helwig 2010a, b) and vascular endothelium (Grogan & Hopkins 2002). HS induces expression of heat shock proteins (HSPs) to protect structural integrity of cellular proteins by their chaperonic activity with regulated expression of cell cycle, DNA replication and repair related genes (Jastrebski et al. 2017). Thus, HS results in multi-organ dysfunction in humans (Broessner et al. 2005; Mozzini et al. 2017).

In animal studies, HS causes decline in performance with elevated production of reactive oxygen species (ROS). Reports suggest that elevated levels of oxidative stress parameters were observed in chickens on exposure to acute HS. Recently, it has been observed that heat stress also induces increased mitochondrial ROS production in rats. Free radicals play an important roles in cellular functions such as cell signaling and detoxification of intruders but elevated levels causes mitochondrial and cellular damage in hepatocytes (Slimen et al. 2014; Mujahid et al. 2007). In addition to the above reports, our recent investigations demonstrate that heat triggered oxidative damage, alterations in liver functions and inflammatory cytokine levels in rats; these alterations need to be further investigated in detail using bioinformatic tools and high throughput molecular technologies (Gupta et al. 2018, 2021).

In-depth study of molecular mechanism of liver pathophysiology in humans exposed to SHS is not possible because HS is lethal for life of individuals along with several ethical issues hence, in continuation to our previous investigations which establish heat induced aberrations and liver injury in rats (Gupta et al. 2018); we are reutilizing the same animals in present study to understand the detailed pathophysiology of heat induced hepatic injury in rats to further strengthen our findings. Here also, HS was categorized into two states; moderate heat stress (MHS) and severe heat stress (SHS) to study the HS induced alterations in liver transcriptome and proteome leading to major aberrations in key cellular processes that effects overall functionality of the liver in rats. The differential gene and protein expressions were analysed by microarray and 2-dimensional gel electrophoresis (2-DGE) respectively in heat stressed liver. The transcriptional and translational changes were regulated by three major pathways; nuclear factor erythroid 2-related factor 2 (NRF-2), macrophage migration inhibitory factor (MIF) and mitochondrial dysfunction pathways. The above data validation and further functional characterization were attempted to explore regulatory functions of molecular signalling pathways in hepatic parenchyma of heat exposed rats.

Materials and methods

Animal housing and guidelines

Rats were received from our in-house experimental animal facility and accommodated in room maintained at temperature 23 ± 2 °C and relative humidity (RH) of 50–55% with alternating 12 h day/night cycle in which rats had access to water and food ad libitum. The diet follows the National Institute of Nutrition (NIN), Hyderabad, India standards for inbred rat and mice maintenance of their ration. All institutional and international guidelines for the care and use of laboratory animals were followed during course of the study. The study protocols including heat exposure were approved by the Institutional Animal Ethical Committee (IAEC/DIPAS/2015–17) of Defence Institute of Physiology and Allied Sciences, Delhi-110054, India.

Experimental design and stress exposure

Eighteen male Sprague dawley rats weighing 250 ± 20 g and 12 week old were divided into three groups in which six (N = 6) rats in each group. In an unexposed control (UC) group rats were kept in heat simulation chamber with T_a 25 ± 0.1 °C and RH 30 ± 10%. Rats of MHS and SHS conditioned groups were exposed to T_a 45 ± 0.1 °C with 30 ± 10% RH and their T_c reaches 40 °C in 25 min and 42 °C in 42 min respectively. After achieving a stabilized core temperature for about 30–60 s, animals were further used for sample collection (Gupta et al. 2018). For microarray and 2-DGE from each group liver tissue samples of three rats were pooled as first sample and another three rats were pooled as second sample. Thus, pooled biological duplicate samples were used to perform differential gene and protein expression studies by microarray and 2-DGE respectively. All six animals in each group was examined for biochemical parameters and protein expression in liver lysate by immunoblotting (IB). For immunohistochemistry analysis, a parallel set of eighteen animals having all three groups (UC, MHS and SHS) (N = 6) in which MHS and SHS groups of rats were exposed to HS in heat simulation chamber.

Sample collection

Rats were fasted overnight before stress exposure to MHS and SHS states. After heat exposure rats were moved out from the heat simulation chamber and physiological parameters such as NIBP (non-invasive blood pressure), HR (heart rate), skin temperature, core temperatures and weight loss were measured of all the animals (Gupta et al. 2018, 2021). Then, rats were anesthetized with sodium pentobarbital (70 mg/kg body mass) administered through intraperitoneal route. Blood was collected by cardiac puncture in left ventricle from which serum was extracted and later used for liver function test and blood electrolyte measurement (Gupta et al. 2018, 2021). Also liver tissue of non-perfused animals from each group was collected followed by snap-freeze in the liquid nitrogen for microarray, 2-DGE, biochemical and molecular analysis. After stress exposure, six animals in each group were perfused transcardially with ice cold 0.1 M phosphate buffer saline (PBS, pH 7.4; S Table 1) followed by fixation in 4% paraformaldehyde solution for analysis of protein expression, morphological and structural alterations in hepatic tissue. Similarly, after habituation with heat simulation chamber, NIBP and temperature probes, samples from unexposed control group of rats were collected as mentioned above.

Differential gene expression studies of heat exposed liver by microarray analysis

Total RNA was isolated from UC, MHS and SHS conditioned rat liver as per manufacturer instructions using RNeasy mini kit; (Cat. No. 74104, Qiagen) followed by quantification and quality check were done by bioanalyzer. RNA samples were hybridized on Affymetrix array of 100 midi formats against 30,000 probe set and then incubated at 48 °C with 60 rpm for 16 h, subsequently chip was analyzed with microarray scanner 3000 7G (Affymetrix Inc). Expression analysis was performed using the Gene Chip® Rat 2.0 ST array on Affymetrix platform. The raw files created at Affymetrix platform were processed through GC-RMA normalization using Transcription Analysis Console (TAC 4.0) software with inbuilt system at I-Life Discovery, Gurugram, India. Then, data were extracted in the form of excel sheet containing gene ID, fold change and p-value along with other additional information of respective gene ID’s. The detailed analysis was performed in accordance to ingenuity pathway analysis (Cat. No. IPA, 83,001, Qiagen) software in our laboratory. For statistical comparison, multivariate simulation was used to detect differentially expressed genes with a probability value (p-value) ≤ 0.05 and fold change cut off value ≥  ± 2. Differentially expressed candidate genes were subsequently evaluated for their biological significance through IPA. The data were also analysed by freely available ‘cytoscape’ software (Version: 3.8.0) for further cross validation of the data. On cytoscape platform ‘GeneMania’ app was used to build network of commonly expressed genes in all the canonical pathways.

Differential protein expression studies of heat exposed liver by two-dimensional gel electrophoresis

Protein was isolated and quantified as per standard protocol from liver tissue (Gupta et al. 2018). Total 400 μg protein with protease inhibitors and nuclease mix were incubated at room temperature for 1 h followed by centrifugation at 15,000 rpm for 10 min at 4 °C. Supernatant was precipitated in 100% ice-cold acetone (Cat. No. 0122129, SRL) and trichloroacetic acid (Cat. No. 90544, SRL) in ratio of 8:1:1 (8 Acetone: 1 Trichloroacetic acid: 1 Protein) at -20 °C overnight. Next day, samples were centrifuged at 12,000 rpm for 15 min at 4 °C and supernatant was discarded and then pellet washed twice with 1 ml of ice-cold acetone for complete removal of trichloroacetic acid. Pellet was allowed to air dry at room temperature for removal of acetone completely. Then pellet was dissolved in 350 μl rehydration buffer (S Table 2) containing 0.4% ampholyte (Cat. No. 163–1112, Bio-Rad), which was further used for isoelectric focusing.

First dimension (Isoelectric Focusing): 350 µl of sample was loaded in rehydration tray by pipetting the samples onto the walls of the tray. Then strips (Cat. No. 163–2007, Bio-Rad) of pH range 3–10 were overlaid onto sample and incubated at 20 °C for 1 h then mineral oil (Cat. No. 163–2129, Bio-Rad) was added to prevent evaporation of the samples and incubated at 20 °C for 24 h (for passive rehydration). Next day, strips were transferred onto IEF focusing setup (Protean i-12 IEF Cell, Bio-Rad) with gel side downward and electrodes were placed accordingly after placing the wicks soaked with 20 µl of proteomic grade water (Cat. No. 163–2091, Bio-Rad) on either side of the focusing tray. Then strips were allowed to run for 19 h as per steps described in S Table 3. After focusing, strips were washed twice with 2 ml of equilibration buffer-I (S Table 4) and further with equilibration buffer-II (S Table 5) with shaking for 10 min at room temperature.

Second dimension (SDS-PAGE): After polymerization of 10% acrylamide gel (17 × 17 cm), strips were loaded onto gel and fixed with protean plus overlay agarose containing bromophenol blue (Cat. No. 163–2092, Bio-Rad). Polyacrylamide gels were allowed to run for 5 h at 200 V for proper resolution of proteins (Fig. 2a). The gel was removed from 2-DGE apparatus and stained with standard coomassie staining method (Fig. 2b–d).

Fig. 2.

Two-dimensional gel electrophoresis data analysis of liver exposed to HS: a Schematic representation of 2D-PAGE comprising of differential protein expression by isoelectric focusing of proteins followed by SDS-PAGE, b Displays hepatic protein profile of UC conditions, c Hepatic protein profile of MHS, d Hepatic protein profile of SHS, e Differentially expressed proteins were spotted and extracted for identification by mass spectrometry (MALDI-TOF), f) List of differentially expressed proteins in MHS and SHS state with respect to UC along with mascot score ≥ 300 cut off value, g–i Analysis of genes corresponding to differentially expressed proteins; densitometric analysis of mRNA levels were shown as graphs, g ATP synthase β, h Regucalcin, and i PRDX 1

Spot Cutting: The protein spots that showed higher expressions in heat exposed as compared to unexposed control samples were analyzed by ImageJ software. Then, spots were excised and de-stained as per standard protocol and identified with mass spectrometry (MALDI-TOF) (Fig. 2e) with the help of inbuilt Mascot software.

Analysis: On cytoscape platform ‘Gene Mania’ app was used to build network of identified differentially expressed proteins in heat exposed liver tissue samples (Fig. 3).

Fig. 3.

Schematic diagram of common outcome in the form of interactomes of microarray and 2-DGE data analysis resulting in three major cellular processes/pathways e.g. NRF-2 mediated oxidative stress response, MIF regulated immune response and mitochondrial dysfunction pathways in liver exposed to MHS and SHS state with respect to UC condition

RNA isolation, cDNA synthesis and quantitative real time-polymerase chain reaction (qRT-PCR)

100 mg liver tissue of UC and heat stressed rats were stored in RNA later (Cat. No. R0901, SIGMA) for stabilization of RNA and further used to isolate total RNA by TRIzol (Cat. No. 15596018, Life Technologies) reagent as per manufacturer instructions. The quantification and integrity of RNA was measured using nanodrop (Cat. No. 2000C, Thermo Scientific) and agarose gel electrophoresis respectively. The cDNA was prepared using kit (AB1453A, Thermo Scientific) as instructed by manufacture. The oligonucleotides were designed for selected genes using Integrated DNA Technology and Beacon designer web tool followed by primer synthesis using standard methods (S Table 6). qRT-PCR was performed using sybr green (Cat. No. F416L, Thermo Scientific) mix on Bio-Rad CFX Connect real time PCR system as per manufacture instructions. In brief, initial denaturation at 95 °C for 10 min, denaturation at 95 °C for 10 s, annealing at 46–58 °C for 15 s and extension at 72 °C for 15 s with 35 repeated thermal cycles of gene expression signals in the form of green fluorescence. The PCR reactions were performed in duplicates (N = 6) of each group and specificity of PCR products were confirmed using melting curve analysis. Data analysis were performed using CFX3.0, Bio-Rad software along with normalization with internal control (18S RNA). qRT-PCR was used to measure the mRNA expression levels of unexposed control with respect to heat exposed liver tissue samples. Relative expression levels of experimental groups were calculated using 2^−∆∆CT method.

Immunofluorescence (IF)

Protein expression in tissue sections were performed using immunofluorescence as described in Edith.et.al 2015 (Honvo-Houéto & Truchet 2015). In brief, liver tissue sections of 25 μm were prepared using cryostat (CM1520, Leica) and stored in 0.1 M PBS containing 0.02% sodium azide at 4 °C. Paraformaldehyde-fixed liver sections were washed with phosphate buffer saline tween-20 (PBST). Antigen retrieval were done through heat activation using sodium citrate buffer followed by permeabilization in 1% goat serum along with 0.4% triton X-100 and blocking in 5% goat serum. Then, sections were incubated with primary antibodies (S Table 7) for overnight at 4 °C. Tissue sections were washed with PBST and incubated with fluorescein isothiocyanate (FITC) labelled (Cat. No. ab150077; Alexa flour 488, Thermo Fischer Scientific) secondary antibody at room temperature. The subsequent washing with PBST and sections were incubated with nuclear stain 4′6-diamidino-2-phenylindole (DAPI; Cat. No. D21490, Thermo Fischer Scientific) for 5 min then washing with PBS. Then, sections were mounted with antifade mounting medium and covered with cover slip and dried up at -20 °C for overnight. At the end, tissue sections were analysed with inbuilt software of fully automated inverted fluorescence microscope (Eclipse Ti-2E- Nikon, Japan) using FITC and DAPI filters.

Immuno-blotting (IB)

Total protein was isolated in the form of liver tissue lysate and then quantification were done by Bradford method (Cat. No. 5000006, Bio-Rad). SDS-PAGE and western blotting were performed as per standard protocol (Gupta et al. 2018). In brief, equal amount of protein samples were resolved on a vertical SDS–polyacrylamide gel at a constant potential gradient of 100 V for 2–4 h. Proteins were transferred on PVDF membrane (Cat. No. IPVH00010, Millipore) followed by blocking with bovine serum albumin than incubated with primary antibodies (S Table 7) for overnight at 4 °C. Blots were washed with tris-base saline (7.5) buffer with tween-20 before incubating with alkaline phosphatase conjugated (Cat. No. EAB1047, Elabscience) secondary antibody for 2 h at room temperature and again washed with similar buffer and finally blots were developed using 5-Bromo-4-chloro-3-indolyl phosphate (BCIP; Cat. No PCT1201, Himedia) and nitro-blue tetrazolium chloride (NBT; Cat. No 484235, Calbiochem) as substrate. All protein bands on blots were analyzed for densitometry by ‘ImageJ’ software. Respective graphs were plotted for MHS and SHS against UC in each protein.

Assessment of cytochrome c activity and glucose-6-phosphate dehydrogenase levels in liver mitochondria

Liver mitochondria were isolated using Qproteome mitochondria isolation kit (Cat. No. 37612, QIAGEN). Cytochrome c activity of mitochondria was examined using cytochrome c oxidase assay kit (Cat. No. CYTOCOX1, Sigma). In brief, vial containing 950 µl assay buffer, 50 µl enzyme dilution buffer, 50 µl ferrocytochrome c substrate solution and 50 µl of samples or standards were mixed thoroughly. Immediately after mixing the solution transferred into cuvette and reading was measured at 550 nm per min for five min using spectrophotometer (Multiskan GO, Thermo Scientific). Levels of glucose-6-phosphate dehydrogenase was also measured by kit method (Cat. No. 700300, Cayman). In brief, as per kit description 10 µl of G6PDH substrate was added in mixture of 150 µl assay buffer, 10 µl co-factor, 10 µl enzyme mixture and 10 µl of samples/standards in 96 well plate followed by incubation for 20 min at 37 °C. Then, fluorescence was measured with excitation of 540 nm and emission of 595 nm in fluorometer (Infinite M 200 Pro, Tecan Infinite M Plex).

Bioinformatic and statistical analysis of microarray data

Transcriptome analysis console (TAC) software was used to extract and normalize the microarray data in the excel file format from Affymetrix set-up. Normalized data were further analyzed by ingenuity pathways analysis (IPA) software and compared as UC vs MHS, UC vs SHS and SHS vs MHS. IPA identifies top differentially expressed pathways on the basis of two score system that focuses two independent aspects of the inference problem: a) ‘Enrichment’ score [Fisher’s exact test (FET) p-value] which measures overlap of observed and predicted regulated gene sets, b) Z-score assessing the match of observed and predicted up/down regulated patterns of respective pathways. The Z-score {= -log (p-value)} is particularly used to solve this type of problem since it serves as both significance measure and predictor for activation of pathways. Top canonical pathways are well characterized accounting for metabolism and cell signalling mechanisms which are curated on the basis of information contained in the pathways coming from specific scientific research and review articles, text books and KEGG pathways in IPA. With the help of IPA, network interactome and two-parametric graphs were prepared in each comparison considering statistical constraints such as p-value ≤ 0.05 and fold change cut off value ≥  ± 2.

Statistical analysis of immunological and biochemical assays

Statistically significant difference between experimental (MHS and SHS) and UC groups were considered as p-value ≤ 0.05 for IF, IB and biochemical assays (cytochrome c oxidase and glucose-6-phosphate dehydrogenase). The standard deviation was taken for each condition (N = 6 rats) and level of significance is denoted with eight-tear drop spoked propeller asterisk for UC vs MHS/SHS comparison with p-value ≤ 0.05 = “❊”, p ≤ 0.01 = “❊❊”, p ≤ 0.001 = “❊❊❊”. Similarly, looped square was used to denote comparison between MHS vs SHS with p-value ≤ 0.05 = “⌘”, p ≤ 0.01 = “⌘⌘”, p ≤ 0.001 = “⌘⌘⌘”.

Results

Our previously published results on heat induced liver injury (Gupta et al. 2018) suggest that HS severely altered structural and biochemical functions of rat liver. In the present study, the same animals were reutilised to perform detailed transcriptional and translational investigations to understand the relationship between structural, biochemical and molecular alterations in heat exposed animals which further strengthen our earlier findings of liver pathophysiology. The current results exhibit putative molecular signatures for alterations in parenchymal structure and biochemical functions in moderate and severe heat stressed rat liver.

Analysis of differential gene and protein expression in liver exposed to MHS and SHS

Before proceeding towards rigorous analysis of liver transcriptome, quantification and integrity of total RNA samples were examined by bioanalyzer. Microarray data were analysed by TAC and IPA software’s that showed 63 genes to be differentially expressed in MHS with  − 6.7 to + 7.3 fold change in which 8 genes were downregulated and 55 genes were upregulated as compared with UC (Fig. 1a). In SHS state 1559 genes were differentially expressed with  − 28 to + 39 fold change in which 1239 genes were upregulated and 320 genes were downregulated with respect to UC (Fig. 1a). In MHS and SHS state, changes in gene expression reflects the effect of HS when T_c was increased from 37 °C to 40 °C and 42 °C that showed in the form of hierarchal clustering (Fig. 1c). Upon analysis through IPA software, top 14 significant canonical pathways (S Fig. 1a) were predicted in MHS which showed neutral activity (Z score = 0); whereas in SHS conditioned rats out of 25 significant pathways, 11 pathways were upregulated (Z score > 0) and 1 was downregulated (Z score < 0) and in rest of 13 pathways showed no activity patterns (Fig. 1d). It was also observed that the dysregulated pathways in SHS state are cross talking with each other by sharing their common genes in between that shown in the form of interactome (Fig. 1b). Total 03 and 12 significant toxological pathways were also identified in MHS (S Fig. 1b) and SHS conditioned (Fig. 1e) rats respectively which describes pathological aberrations in heat stressed liver. Majority of deferentially regulated pathways under SHS state are NRF-2 mediated oxidative stress response (Fig. 1d, S Fig. 2a), MIF regulated immune response (Fig. 1d, S Fig. 2f, g), oxidative phosphorylation (Fig. 1d, S Fig. 2c), mitochondrial dysfunction (Fig. 1d, S Fig. 2b), glutathione redox reactions I and superoxide radicals degradation I pathways (Fig. 1d, S Fig. 2d) and IL-10 pathway (Fig. 1d, S Fig. 2e). All the top regulated pathways are comprises of genes participating in redox imbalance, inflammation and irregular proteostasis.

Fig. 1.

Microarray data analysis of liver exposed to HS: a Comparison between UC vs MHS vs SHS state represent total number of upregulated (shown in blue) and downregulated (shown in orange) genes, b Interactome showing cross-talk between various canonical pathways in SHS vs UC, c Heat map in the form of hierarchal clustering of differentially expressed genes in MHS (red) and SHS (violate) compared with UC (blue) shown with the help of colour key defining upregulation of gene expression in fold change from red to green, d Canonical pathways, and e Toxological pathways in SHS vs UC is shown by right-sided x-axis signifying ratio of differentially expressed genes to total number of genes in respective pathways. Left-sided y-axis signify negative log p-value of each pathway

Next, we analysed the differential protein expression in same set of heat exposed rat liver in comparison to UC by two dimensional PAGE as described in methodology section. Analysis of proteins on the basis of iso-electric point and molecular weight (Fig. 2a) that revealed 13 differential expressed protein as spots on SDS-PAGE gel of MHS (Fig. 2c) and SHS (Fig. 2d) conditioned rats with respect to UC rats (Fig. 2b). Upon identification of these protein spots by MALDI-TOF (Fig. 2e), 7 proteins in MHS and 6 proteins in SHS state were identified (Fig. 2f). To verify this, differential protein expression levels, corresponding genes were also scrutinised in liver exposed to HS. Gene expression levels of major mitochondrial proteins i.e. ATP 5 Synthase β (Fig. 2g), Regucalcin (Fig. 2h) and PRDX 1 (Fig. 2i) were examined and found to be elevated in both MHS and SHS state.

Comparative analysis of microarray and 2-DGE data

Interactomes were generated using cytoscape for differentially expressed genes and proteins involved in the top canonical pathways shown in Fig. 3. List of important functions (S Table 8) of pathways and genes resulted from cytoscape analysis of microarray and 2-DGE data were complementing each other. Upon comparison of both the interactomes, three major pathways i.e. NRF-2 mediated oxidative stress, MIF regulated immune response and mitochondrial dysfunction pathways were identified that are playing crucial roles in heat triggered hepatic injury in rats (Fig. 3).

Severe heat stress triggered NRF-2 mediated oxidative damage in liver

Heat induced oxidative stress increased ROS production and reactive intermediates such as free radicals and peroxidases that have damaging effect on cellular lipids, proteins and nucleic acids. Results of differential gene and protein expression analysis indicates that the impact of HS generated severe oxidative stress through NRF-2 mediated oxidative stress response pathway (S Fig. 2a) and major biomolecular player involved in this pathway are tabulated in form of Table 1. Functional characterization of pathway showed alteration in protein and gene expression levels of key genes (Table 1) which were analysed by IF in hepatic tissue sections, IB in liver lysate (Fig. 4) and qRT-PCR (Fig. 5) respectively. NRF-2 functions as primary regulator in oxidative stress which was found to be upregulated in liver analysed by IF and IB with rise in T_c (Fig. 4a). JNK1/2 mediated signalling pathway cross talk (S Fig. 2a) with NRF-2 and found to be activated with MAPKAP2 and ATF4; expression levels of JNK1/2 was found to be neutral in MHS and SHS state but Phospho-JNK1/2 was elevated in both the HS states shown by densitometric ratio of (P-JNK1/2)/(JNK1/2) which was analysed by IF and IB (Fig. 4c). Analysis of gene expression by qRT-PCR of MAPKAP2 (Fig. 5a, e) and ATF4 (Fig. 5a, g) were elevated in heat exposed hepatic parenchyma. Heat induced Hsp 90AB1 expression level was also elevated in heat stressed rats as analysed by IF and IB (Fig. 4b). Gene expression levels of PRDX5 (Fig. 5a, c), DNAJB4 (Fig. 5a, d) and MGST3 (Fig. 5a, f) were also found to be elevated in both the HS conditioned rats which resulted oxidative burst in parenchymal tissue. Protein expression of PRDX5 (Fig. 4d) and DNAJB4 (Fig. 4e) were significantly elevated in MHS and SHS state of rats which was analysed by IF and IB. In both qRT-PCR and microarray, all the genes were followed the same patterns of expression levels in both MHS and SHS state. Protein expression of antioxidant marker superoxide dismutase 1 (SOD 1) was analysed by IB and found to be upregulated with increasing T_c and in qRT-PCR also it was found to be upregulated (Fig. 5a, b) in HS (Gupta et al. 2021). In immunohistochemistry experimentation; protein expression studies were performed through IF in liver sections and a representative image showing FITC (S Fig. 3a), counter stained with DAPI staining (S Fig. 3b) and a merged picture of FITC and DAPI (S Fig. 3c) is shown as supplementary figure (S Fig. 3). NRF-2 mediated signalling and oxidative stress related key protein and gene expression levels were altered significantly in liver exposed to HS in comparison to UC are listed in Table 2 and 3 respectively.

Table 1.

Genes involved in NRF-2 mediated oxidative stress response pathway

S. No	Gene Symbol	Fold Change	p Value
1	AKR7A3	4.67	0.0287
2	ATF4	3.6	0.0499
3	DNAJA4	4.31	0.0122
4	DNAJB1	7.54	0.0204
5	DNAJB4	20.51	0.0070
6	FOS	5.35	0.0087
7	FTH1	5.74	0.0002
8	FTL	3.27	0.0009
9	GSTA5	5	0.0012
10	HERPUD1	3.83	0.0285
11	HMOX1	11.3	0.0018
12	HSPB8	3.62	0.0252
13	JUN	11.04	0.0242
14	JUND	3.59	0.0073
15	KEAP1	3.15	0.0170
16	MAF	2.28	0.0222
17	MAP2K3	2.94	0.0030
18	MGST3	2.42	0.0373
19	NQO1	4.7	0.0145
20	PRKCQ	-2.07	0.0435
21	RRAS	5.16	0.0307
22	SCARB1	3.8	0.0269
23	SOD1	2.67	0.0355
24	SOD3	2.61	0.0185
25	TXN	2.09	0.0366

Fig. 4.

Assessment of molecular markers involved in NRF-2 mediated oxidative stress response pathway under HS by IF and IB: Protein expression levels of a NRF-2, b Hsp 90AB1, c Phospho-JNK 1/2, d PRDX 5, e DNAJB 4 are shown by IF and IB. Densitometric analysis of IF and IB were done for n = 6 rats in each condition, a representative figure of IF and IB bands are shown in respective sub sections. For IB data, liver lysates were normalized with loading control as β-actin for analysis of protein expression of NRF-2, Hsp 90AB 1, DNAJB 4, PRDX 5 and JNK 1/2 for Phospho-JNK 1/2. Standard deviation of each condition was taken for statistical significant of n = 6.Level of significance is denoted with eight tear drop spoked propeller asterisk for UC vs MHS/SHS comparison with p-value ≤ 0.05 = “❊”, p ≤ 0.01 = “❊❊”, p ≤ 0.001 = “❊❊❊”. Similarly, looped square was used to denote comparison of MHS vs SHS with p-value ≤ 0.05 = “⌘”, p ≤ 0.01 = “⌘⌘”, p ≤ 0.001 = “⌘⌘⌘”

Fig. 5.

Transcription levels were analysed by qRT-PCR in comparison with microarray data: a Fold change of molecular markers in the form of heatmap which are involved in NRF-2 mediated oxidative stress pathway in rats exposed to HS. Transcript levels of (b) SOD 1, c PRDX 5, d DNAJB 4, e MAPKAP 2, f MGST 3, g) ATF 4. Standard deviation of each condition was taken for statistical significant of n = 6. Level of significance is denoted with eight tear drop spoked propeller asterisk for UC vs MHS/SHS in comparison with p-value ≤ 0.05 = “❊”, p ≤ 0.01 = “❊❊”, p ≤ 0.001 = “❊❊❊”. Similarly, looped square was used to denote comparison of MHS vs SHS with p-value ≤ 0.05 = “⌘”, p ≤ 0.01 = “⌘⌘”, p ≤ 0.001 = “⌘⌘⌘”

Table 2.

Summary: Analysis of protein expression by IF and IB

S. No	Proteins	IF & IB	UC	MHS	SHS
1	NRF-2	IF	711 ± 12	1273 ± 24❊❊❊	1598 ± 35❊❊❊, ⌘⌘
		IB	0.75 ± 0.03	0.87 ± 0.02❊	1.23 ± 0.05❊❊, ⌘⌘
2	P-JNK	IF	776 ± 9	973 ± 12❊❊	1270 ± 13❊❊❊, ⌘⌘⌘
		IB	0.25 ± 0.02	0.57 ± 0.002❊❊	0.75 ± 0.04❊❊, ⌘
3	HSP90AB 1	IF	817 ± 14	1358 ± 45❊❊	1733 ± 30❊❊❊, ⌘⌘
		IB	0.89 ± 0.03	1.07 ± 0.01❊❊	1.57 ± 0.03❊❊, ⌘⌘⌘
4	DNAJB 4	IF	742 ± 39	830 ± 9❊	1060 ± 52❊, ⌘
		IB	0.0008 ± 0.03	0.0081 ± 0❊❊	0.018 ± 0❊❊❊, ⌘⌘
5	PRDX 5	IF	742 ± 39	908 ± 14❊❊	1045 ± 4❊❊, ⌘⌘⌘
		IB	0.64 ± 0.02	0.94 ± 0.09❊❊	1.45 ± 0.0❊❊❊, ⌘⌘
6	MIF	IF	640 ± 8	1071 ± 13❊❊❊	1387 ± 12❊❊❊, ⌘⌘
		IB	0.88 ± 0.01	1 ± 0.04❊❊	1.25 ± 0.02❊❊, ⌘⌘
7	c-FOS	IF	569 ± 26	774 ± 16❊❊❊	858 ± 40❊❊
		IB	0.57 ± 0.02	0.65 ± 0.04❊❊	0.86 ± 0.03❊❊, ⌘
8	HMOX 1	IF	593 ± 12	754 ± 34❊❊	1158 ± 46❊❊, ⌘⌘
		IB	0.25 ± 0.02	0.38 ± 0.01❊	0.58 ± 0.01❊❊❊, ⌘⌘
9	IL-10	IF	393 ± 12	554 ± 34❊❊❊	758 ± 46❊❊❊, ⌘⌘⌘
		IB	0.21 ± 0.02	0.31 ± 0.01❊	0.51 ± 0.01❊❊, ⌘⌘
10	NF-κB	IB	0.16 ± 0.01	0.56 ± 0.03❊❊	0.66 ± 0.02❊❊❊, ⌘⌘

Table 3.

Summary: Analysis of gene expression by qRT-PCR

S. No	Genes	MHS	SHS
a	Mitochondrial dysfunction pathway
1	ATP5A1	1.88 ± 0.29❊❊	2.86 ± 0.64❊❊❊, ⌘
2	ATP5L	0.86 ± 0.11❊	2.29 ± 0.87❊
3	COX5B	1.26 ± 0.41	2.17 ± 0.61❊
4	COX6A1	1.37 ± 0.9❊	2.87 ± 0.46❊❊, ⌘
5	NDUFA11	1.76 ± 0.55❊❊	2.96 ± 0.17❊❊❊, ⌘⌘
6	NDUFB4	0.87 ± 0.66❊❊	3.26 ± 0.48❊❊, ⌘⌘⌘
7	NDUSF6	0.79 ± 0.78❊	1.28 ± 1.13❊❊, ⌘⌘
8	CYC1	1.08 ± 0.2❊	1.96 ± 0.83❊❊, ⌘
9	SDHD	1.08 ± 0.76❊	2.89 ± 0.48❊❊❊, ⌘
10	UQCR11	1.28 ± 0.24❊❊	5.41 ± 0.39❊❊❊, ⌘
b	Oxidative stress response pathway
1	PRDX5	1.2 ± 0.57❊	1.45 ± 0.36❊❊❊, ⌘
2	SOD1	1.84 ± 0.7❊❊❊	3.14 ± 0.92❊❊❊, ⌘
3	MGST3	2.2 ± 1.24❊❊	5.09 ± 0.15❊❊❊, ⌘
4	ALDH2	1.47 ± 0.62❊❊	3.96 ± 1.75❊❊
c	Inflammatory response pathway
1	NF-κB	1.02 ± 0.33	1.35 ± 0.32❊❊, ⌘⌘
2	PSMB5	4.99 ± 0.66	6.12 ± 0.67❊
3	TLR4	1.04 ± 0.34❊	3.34 ± 1.33❊❊, ⌘⌘
4	CD74	1.56 ± 1.13❊❊❊	2.33 ± 0.46❊❊, ⌘⌘⌘
5	IL10	0.65 ± 0.28❊❊	1.95 ± 1.12❊❊, ⌘⌘
6	IL1A	2.33 ± 0.66❊❊	3.44 ± 1.2❊❊❊, ⌘⌘
7	C1QA	0.78 ± 0.35	3.3 ± 0.73❊
d	Heat shock proteins
1	DNAJB4	1.04 ± 0.31❊❊	4.14 ± 0.64❊❊❊, ⌘
2	DNAJB1	2.03 ± 0.32❊❊❊	4.48 ± 0.9❊❊❊, ⌘⌘
3	HSP27	2.34 ± 0.58❊❊❊	9.47 ± 1.29❊❊❊, ⌘⌘
4	HSP70	1.62 ± 0.17❊❊❊	3.87 ± 0.36❊❊❊, ⌘⌘⌘
5	HSP90	2.14 ± 0.73❊❊❊	4.29 ± 0.68❊❊❊, ⌘⌘
e	Protein synthesis related genes
1	UPF1	0.86 ± 0.09❊❊	4.21 ± 2.36❊❊, ⌘⌘
2	ACACB	1.41 ± 0.42❊❊	4.31 ± 1.34❊❊, ⌘⌘
3	POLG	1.2 ± 0.44❊❊	5.55 ± 1.13❊❊❊
4	EIF3F	2.17 ± 0.4❊❊	5.29 ± 3.54❊❊, ⌘
5	RPS24	0.86 ± 0.09❊❊	4.21 ± 2.36❊❊, ⌘
6	HOGA1	0.95 ± 0.54❊❊	1.54 ± 0.29❊, ⌘
7	PPP2RA1	1.01 ± 0.33❊❊	1.98 ± 1.12❊❊, ⌘⌘
8	UBE2V1	1.4 ± 0.73❊	2.66 ± 1.65❊❊, ⌘⌘
9	UBE2Z	1.27 ± 0.36❊	2.97 ± 1.41❊❊
10	GOT	0.82 ± 0.44❊❊	2.9 ± 0.24❊❊, ⌘⌘
11	IDNK	1.95 ± 0.99❊	3.08 ± 2.29❊❊, ⌘⌘
12	GCN1L1	1.13 ± 0.42❊❊	2.29 ± 0.87❊❊, ⌘⌘
13	RAB30	0.95 ± 0.74❊	1.08 ± 0.85❊❊, ⌘⌘
14	ATF4	1.09 ± 0.5❊❊	1.75 ± 1.22❊❊❊, ⌘⌘
f	Signal transduction pathways
1	MAPKABK2	1.49 ± 0.48❊❊	3.98 ± 2.38❊❊❊, ⌘⌘
2	AQP8	1.15 ± 0.09❊	3.37 ± 1.27❊, ⌘⌘
3	F11R	1.38 ± 0.64❊	4.59 ± 1.98❊❊, ⌘⌘
4	NOTCH2	1.08 ± 0.76❊	5.01 ± 2.31❊❊, ⌘⌘
5	Eif4	2.8 ± 1.56	5.89 ± 4.33❊
6	SLC22A23	0.62 ± 0.32	0.96 ± 0.53❊
7	VDAC2	1.24 ± 0.67❊❊	1.99 ± 1.12❊❊❊, ⌘
8	UBC	1.7 ± 0.29❊	13.92 ± 1.71❊❊, ⌘
g	Cell death related genes
1	ATG2A	1.41 ± 0.42❊	4.31 ± 1.34❊❊❊, ⌘
2	HMOX1	1.2 ± 0.44❊❊	5.55 ± 1.13❊❊, ⌘⌘
3	XIAP	0.7 ± 0.35❊	4.69 ± 0.66❊❊, ⌘
4	DUSP1	2.64 ± 1.64❊❊	4.84 ± 0.54❊❊❊, ⌘⌘
5	FOXO3	1.76 ± 0.48	4.79 ± 0.41❊❊, ⌘

Activation of MIF mediated immune response in liver during HS

MIF is a cytokine and an integral mediator of the innate immune system in mammals. It exerts pro-inflammatory activities and play important roles in tissue repairing during septic shock. Differential gene expression studies reveals that severe inflammation occurs in liver through MIF mediated inflammatory response pathway (S Fig. 2f, g) and molecular markers involved are listed in Table 4. MIF protein expression levels (Fig. 6a) were found to be upregulated in IF and IB in tissue sections and lysate respectively with rise in T_c. The MIF binds with CD 74 on different cells of immune system to trigger an acute immune response. After MIF activation, CD74 gene expression levels were measured by qRT-PCR and found to be elevated in both the HS state (Fig. 7a, e). MIF and TLR4 are tightly regulates the NF-κB activity in MIF signaling pathway (S Fig. 2f, g). The C-FOS (Fig. 6b) and NF-κB (Fig. 6d, 7a, d) were found to be increased on elevation of HS shown by IF, IB and qRT-PCR assays. Subsequently, IL-10 (S Fig. 2e) is also stimulated in HS shown by IF, IB and qRT-PCR (Fig. 7a, c) (Gupta et al. 2021). As per previous literature and well known fact that IL-10 also regulates HMOX1, AP1 and IL-1A signalling pathway in pathological and stress conditions in animals. In the present study, protein and gene expression levels of HMOX 1 (Fig. 6c, 7a, b), AP 1 (Fig. 7a, g) and IL-1A (Fig. 7a, f) which are inter-connected in inflammatory pathways were examined and found to be elevated in MHS and SHS conditioned rats. MIF also activates a cascade of events consisting of the induction of cytoplasmic JNK activity (Fig. 4c). In addition to the above analysis, it was also found that MIF regulated innate immunity (S Fig. 2f) and MIF mediated glucocorticoid regulation (S Fig. 2g) related key gene and protein expression levels were altered significantly in rats exposed to MHS and SHS state (Table 2, 3). Thus, all these differentially expressed bio-markers in liver signifies the persistence of hepatic inflammation under HS.

Table 4.

Genes involved in MIF related immune response pathway

S. No	Symbol	Fold Change	p-Value
1	CD14	3.61	0.0086
2	CD74	5.01	0.0294
3	FOS	5.35	0.0087
4	JUN	11.04	0.0242
5	MIF	4.66	0.0215
6	NFKBIB	3.44	0.0458
7	PLA2G12B	2.84	0.0063
8	PLA2G1B	-2.24	0.0319
9	TLR4	2.13	0.0028

Fig. 6.

Protein expression levels were analysed by IF and IB of molecular markers which are involved in MIF regulated immune response pathway in MHS and SHS compared with UC: Expression level of a MIF, b C-FOS, c HMOX 1 were shown by IF and IB and d) NF-kB was shown by IB only. Densitometric analysis of IF and IB were done for n = 6 rats in each condition, a representative figure of IF and IB bands are shown in respective sub sections. For IB data, liver lysates were normalized with loading control as β-actin for analysis of protein expression levels of MIF, C-FOS, HMOX 1 and NF-kB. Standard deviation of each condition was taken for statistical significance of n = 6. Level of significance is denoted with eight tear drop spoked propeller asterisk for UC vs MHS/SHS comparison with p-value ≤ 0.05 = “❊”, p ≤ 0.01 = “❊❊”, p ≤ 0.001 = “❊❊❊”. Similarly, looped square was used to denote comparison of MHS vs SHS with p-value ≤ 0.05 = “⌘”, p ≤ 0.01 = “⌘⌘”, p ≤ 0.001 = “⌘⌘⌘”

Fig. 7.

Transcription level were analysed by qRT-PCR in comparison with microarray data: a Fold change of molecular markers in the form of heatmap which are involved in MIF regulated immune response pathway in rats exposed to HS. Transcript levels of (b) HMOX 1, c IL-10, d NF-kB, e CD 74, f IL-1A, g AP-1. Standard deviation of each condition was taken in statistical significant of n = 6. Level of significance is denoted with eight tear drop spoked propeller asterisk for UC vs MHS/SHS in comparison with p-value ≤ 0.05 = “❊”, p ≤ 0.01 = “❊❊”, p ≤ 0.001 = “❊❊❊”. Similarly, looped square was used to denote comparison of MHS vs SHS with p-value ≤ 0.05 = “⌘”, p ≤ 0.01 = “⌘⌘”, p ≤ 0.001 = “⌘⌘⌘”

Redox imbalance leads to mitochondrial dysfunction in liver during HS

After analysis of NRF-2 and MIF regulated pathways, identification of an another most important pathway which is related to mitochondrial dysfunction. Mitochondria are the primary consumers of oxygen in the cells and contain a multitude of redox carriers that are capable of transferring single electrons to oxygen resulting in formation of ROS. Mitochondria also contains an extensive antioxidant defence system to detoxify different types of superoxide’s and oxidative damage to cellular components. Mitochondrial dysfunction (S Fig. 2b) occurs during ROS-mediated oxidative stress overpowers the antioxidant defence system. To verify this, differential gene and protein expression levels in liver exposed to HS were studied by microarray and 2D-PAGE respectively which are listed in the Table 5. Gene expression levels of major mitochondrial inter-membrane matrix biomolecules such as NDUSF6 (Fig. 8a, b), NDUFA11 (Fig. 8a, c), VDAC2 (Fig. 8a, d), SDHD (Fig. 8a, e), ATP5A1 (Fig. 8a, f), NDUFB4 (Fig. 8a, g), COX5B (Fig. 8a, h), CYC1 (Fig. 8a, i), COX6A1 (Fig. 8a, j) and UQCR11 (Fig. 8a, k) were found to be elevated during HS. Results suggest that liver mitochondrial functions are disrupted during HS and further leads to mitochondrial dysfunction of rat liver. Therefore, it is imperative to analyse the levels of glucose-6-phosphate dehydrogenase (Fig. 8l) and determine the activity of cytochrome c oxidase (Fig. 8m) in liver mitochondria which were downregulated and upregulated in rats exposed to severe HS respectively. Upregulation in oxidative phosphorylation pathway (S Fig. 2c) also supports the mitochondrial dysfunction in liver exposed to HS. List of upregulated and downregulated genes of mitochondrial dysfunction pathway are shown in the Table 5.

Table 5.

Genes involved in mitochondrial dysfunction pathway

S. No	Symbol	Fold Change	p Value
1	ATP5F1A	2.19	0.0400
2	ATP5MC1	5.69	0.0002
3	ATP5MC3	4.75	0.0020
4	ATP5MF	2.5	0.0013
5	ATP5MG	3.66	0.0026
6	ATP5PF	3.86	0.0143
7	COX4I1	2.83	0.0398
8	COX5A	2.29	0.0065
9	COX6A1	6.01	0.0040
10	Cox6c	2.1	0.0256
11	COX8A	4.08	0.0009
12	CYC1	6.99	0.0132
13	FURIN	2.66	0.0259
14	GPX4	2.7	0.0253
15	HSD17B10	2.21	0.0008
16	HTRA2	3.45	0.0022
17	NDUFA1	2.2	0.0005
18	NDUFA2	3.87	0.0363
19	NDUFA3	8.18	0.0029
20	NDUFA6	2.05	0.0204
21	NDUFA7	3.73	0.0180
22	NDUFA8	2.39	0.0402
23	NDUFB3	4.07	0.0114
24	NDUFB4	8.29	0.0122
25	NDUFS8	4.38	0.0082
26	PARK7	2.13	0.0312
27	PINK1	3.09	0.0049
28	PRDX5	4.18	0.0285
29	SDHD	2.47	0.0332
30	SURF1	2.62	0.0102
31	UCP2	3.78	0.0440
32	UQCR11	12.16	0.0478
33	UQCRC1	2.05	0.0164
34	VDAC2	3.07	0.0415

Fig. 8.

Analysis of mitochondrial genes, levels of glucose-6-phosphate dehydrogenase and activity of cytochrome C oxidase in rats exposed to HS: a Comparison between qRT-PCR and microarray in the form of heatmap. Densitometric analysis of mRNA levels were shown as graph b NDUSF6, c NDUFA11, d VDAC2, e SDHD, f ATP5A1, g NDUFB4, h COX5B, i CYC1, j COX6A1, k UQCR11, l Levels of Glucose-6-phosphate dehydrogenase, and m Activity of Cytochrome C oxidase. Standard deviation of each condition was taken for statistical significant of n = 6. Level of significance is denoted with eight tear drop spoked propeller asterisk for UC vs MHS/SHS in comparison with p-value ≤ 0.05 = “❊”, p ≤ 0.01 = “❊❊”, p ≤ 0.001 = “❊❊❊”. Similarly, looped square was used to denote comparison of MHS vs SHS with p-value ≤ 0.05 = “⌘”, p ≤ 0.01 = “⌘⌘”, p ≤ 0.001 = “⌘⌘⌘”

Discussion

Liver executing innumerable functions that regulates cellular physiology of hepatic parenchyma in healthy human and animals. Liver physiology and biochemistry is affected by various internal and external factors including high T_a. Recent findings from our laboratory suggest that HS induces perturbation in physiological parameters (weight loss, elevated HR and blood pressure) along with oxidative stress. Also, we had established that upon heat stress, neuroinflammation in hypothalamus and memory loss occurs in hippocampus (Chauhan et al. 2017, 2021).

Findings from biochemical and microscopic studies conclude that there is severe functional and structural alterations in rat liver exposed to HS. At biochemical level alterations were found under HS in compared with unexposed rat liver. There were non-significantly small alterations in salts or ions level which help in water retention capacity in the body during HS. Our previous observation of biochemical study has revealed that HS trigger redox imbalance along with LFT parameters in plasma and tissue illustrates the heat triggered damage with rise in T_c. Along with biochemical alterations HS also elicits inflammation supported by results of histological examination which reveal a significant loss of parenchymal tissue with distortion in lobular orientation. Thus, increased levels of LFT, oxidative stress parameters, and inflammatory cytokines is in equivalent to HE stained tissue sections resulting perturbation in rat liver parenchyma which results in thermo-cytotoxic effect leading to cell death either by necrosis or apoptosis. The above observations suggest that heat induced redox imbalance leads to liver dysfunction shown by altered liver function tests and structural distortion in hepatic tissue (Gupta et al. 2018, 2021). In the present study, we further investigated the effect of moderate and severe heat stress on liver pathophysiology with specific emphasis on identification of molecular determinants that are responsible for alterations in stress response functions of rat liver. To find out the detailed molecular mechanism of liver pathophysiology of MHS and SHS; we studied the differential gene and protein expression of rat liver exposed to MHS and SHS state with respect to unexposed control by microarray and two-dimension SDS-PAGE respectively. Further, functional characterization was done by wet lab experimentation using biochemical and molecular techniques. Top fourteen and twenty-five pathways were found to be dysregulated in moderate and severe HS respectively as compared to UC which were statistical significant. Analysis suggest that alterations in canonical pathways further induced toxicity in the liver parenchyma which are displaying through toxological pathways during thermal insult as observed in our previous histological experiments in Gupta et al 2018. Results demonstrate that many common genes among various pathways got affected in SHS state of animals. Results of comparative analysis of SHS with respect to MHS showed, a total 1082 dysregulated genes were found in similar canonical pathways. Data of proteomic study were analysed by cytoscape which were also complemented the similar observations as shown in microarray results.

Liver transcriptomic study accomplished by IPA analysis which was explored the key pathways comprising the redox imbalance, oxidative phosphorylation, inflammation and protein synthesis. Appearance of most prominent and upregulated anti-oxidative pathways such as NRF-2-mediated oxidative stress response (S Fig. 2), glutathione redox reactions I (S Table 10) and superoxide radicals degradation (S Fig. 2d, S Table 11) demonstrated severe redox imbalance in hepatic parenchyma upon encountering thermal insult (Cho et al. 2013; Shin et al. 2013). Liver being a metabolically active organ that produces excessive ROS as by-product for detoxification and activation of important metabolic pathways. To further counter the elevation of ROS, activation of antioxidant response element in the cells during exposure to HS is primarily regulated by NRF-2. Result suggest that expression levels of NRF-2, SOD 1, MGST 3 and peroxiredoxin 5 were found to be elevated (S Fig. 2a) which demonstrated the important roles of NRF-2 in regulation of oxidative stress response pathways in liver during HS. On exposure to HS, cells activates stress response functions through synthesizing HSPs which is a well-known conserved mechanism for protection against various harsh environmental stress conditions (Zuo et al. 2016). Under control of NRF-2, HSP90AB 1 and DNAJB 4 (HSP 40) helps in performing chaperonic activity to regulate protein aggregation, unfolding and/or mis-folding to maintain the structural integrity during exposure to HS. Generally, three major mitogen kinase kinase kinases (MKKKs) are controlling the basic signal transduction regulation; we scrutinized all three signalling network MKKKs of p38, ERK1/2 and JNK1/2. Data shown increased expression levels of phospho-JNK1/2 with respect to JNK1/2 (Fig. 4c), however there is no significant alteration in protein expression levels of phospho-p38, p-38, phospho-ERK1/2 and ERK1/2 (data not shown) were observed. Thus, results demonstrated the NRF-2 mediated oxidative stress response pathways in liver during exposure to HS are controlling by MAPKKKs which mediated through JNK1/2 signalling.

Study indicates that oxidative stress acts as pathogenic mediator to stimulate inflammation (Hussain et al. 2016). Recent studies indicate that multiple factors are influencing the pathophysiology of inflammation during heat stress in mammals (Chen et al. 2018; Punder & Pruimboom 2015; Lim & Suzuki 2017). NRF-2 participated in hepatic inflammation; our results also showed activation and regulation of inflammatory signalling through NRF-2 mediated regulation in liver under HS (Cho et al. 2013; Shin et al. 2013). Liver being a target organ for non-infectious and infectious inflammatory pathologies such as non-alcoholic fatty liver disease and viral hepatitis leads to high morbidity and mortality. MIF being a cytokine modulates host response on encountering different types of infection and stress conditions (Assis et al. 2014). Data suggest that upregulated pathways like MIF regulation of innate immunity and MIF regulated glucocorticoid pathways showed macrophage mediated inflammatory sepsis, shock, cellular apoptosis and cell growth arrest along with anti-inflammatory affects (S Fig. 2f). Result shows that macrophage migration inhibitory factor was found to be elevated in both MHS and SHS state of rats. This was explained by elevated expression of CD 74, NF-kB and AP 1 (Fig. 7). Results were also complemented by proteomic data analysis, since AP 1 was upregulated in similar conditions. MIF pathways are interconnected with stimulated IL-10 pathway (S Fig. 2e, S Table 12). Data suggested, the upregulation of IL-10 pathway in liver tissue during HS, which was further indicated by elevated expression levels of IL-10, IL-1A and C-Fos. Upregulated pathways like IL-10 signaling, FXR/RXR activation and toll-like receptor signalling activates different cytokine and chemokine production for activation of downstream signalling cascade during heat exposure.

Mitochondrion is a primary oxygen consumer in the cells and containing various redox carriers for electron transfer. The redox mediated electron transport leading to ATP production in five proteins-lipid enzyme complexes (Complex I-V) which are located on the mitochondrial inter membrane space. Mitochondria is also playing crucial roles in activation and regulation of hepatic apoptosis in various pathological conditions in liver. Results of microarray (Fig. 1, S Fig. 2), 2-DGE data analysis (Fig. 2, S Table 8), activity of cytochrome c (Fig. 8m) oxidase and levels of glucose-6-phosphate dehydrogenase (Fig. 8l) data indicates that mitochondria is not functioning properly during HS and leads to mitochondrial dysfunction. Many key genes that are participating in energy metabolism and oxido-reductive reactions in mitochondria were also dysregulated as displayed in hepatic transcriptome and proteome analysis (Table 1, 2, 3, 4, 5). Results also demonstrate that a pool of genes which are differentially expressed and playing critical roles in mitochondrial dysfunction (S Fig. 2b) and oxidative phosphorylation (S Fig. 2c, S Table 9) (Topf et al. 2019; Turkseven et al. 2020). These pathway are known to be dysregulated in various pathological and environmental stress conditions in animals (Bouchama et al. 2017; X. Guo et al. 2020; Ippolito et al. 2014; Metzger et al. 2020; Sies & Jones 2020). Results of oxidative burst directly distorted the mitochondrial structure and functions by oxidizing various superoxide’s and lipid peroxidation in mitochondria (Enns et al. 2014; Sies & Jones 2020; Walheim et al. 2018). Our present data also showed similar alterations in mitochondrial functions and structural integrity in liver exposed to HS. Earlier, we have also found alterations in the levels of ROS, protein oxidation, lipid peroxidation and anti-oxidant enzymes in liver lysate during HS which further supports our current findings.

The data of comparative analysis between moderate and severe HS suggests that heat induced modulation in the expression levels of major regulators such as NRF-2 (Fig. 4, 5), MIF (Fig. 6, 7) and mitochondrial dysfunction (Fig. 8) pathways in a significant manner as mentioned in result sections. Results demonstrates that NRF-2 and MIF regulated pathways were dysregulated with rise in T_c; data suggested the liver parenchymal damage is more in SHS conditioned rats as compared to MHS state. Thus, comparative data of MHS and SHS complements our study i.e. HS hampers liver transcriptome by modulation in NRF-2 and MIF functions leading to mitochondrial dysfunction. In addition to above pathways, we identified various other pathways which are involved in protein synthesis and its integrity like EIF2 signalling, mTOR signalling, regulation of eIF4 and p70S6K signalling, 4-hydroxyproline degradation I and aryl hydrocarbon receptor signalling were upregulated in SHS conditioned rats in comparison with UC (Thompson et al. 2016). The major functions of these pathways are ribosomal protein synthesis and activation of chaperonic activity of HSPs (Beere 2005; Li et al. 2000; Hawle et al., 2013; Zuo et al. 2016).

In summary, HS is a global environmental problem due to rise in mean T_a. In the present report, the effect of HS on rat liver was examined to explore the pathophysiology of severe heat stress. The study reveals that redox imbalance and activation of pathways involved in oxidative stress i.e. NRF-2 mediated oxidative stress response, dysregulated redox state of liver in HS elicits inflammation through MIF regulated pathways. In-depth analysis of knockout animal models used for characterization of NRF-2 and MIF pathways also showed the suppression of NRF-2 antioxidant defence plays an essential roles in the development of salt-induced oxidant stress (Priestley et al. 2016). Absence of NRF-2 also affects the brain functions (Muramatsu et al. 2013). Thus, NRF-2 helps in protection against environmental stress conditions (Kensler et al. 2007). MIF helps in protection against inflammation, its absence in knock-out animal models resulting in severe infection (Koebernick et al. 2002; Fingerle-Rowson et al. 2009). The lack of MIF has shown low mitochondria number (Herbst et. al., 2019) and earlier studies established that NRF-2 and MIF are playing crucial roles against different types of stress conditions and our data also suggest that both the marker along with mitochondrial dysfunction are responsible for alterations in liver pathophysiology in animals exposed to MHS and SHS conditions. SHS also effects on liver parenchyma and major cellular functions of hepatic tissue. Thus, the findings of the present study may help to identify new drugs or therapeutic targets for combating against heat induced liver pathology in rats.

Limitation of the study

Main limitation of present study is that the similar experimental design and severe heat exposure cannot be done in human volunteers, due to ethical concerns and risk to human life. Since, the investigation uses whole rat liver tissue however, liver comprises of various cell types such as hepatocytes, hepatic stellate, kupffer, and liver sinusoidal endothelial cells having different cellular and molecular changes during HS hence, this cannot be investigated in single study.

Supplementary Information

Below is the link to the electronic supplementary material.

Abbreviations

2-DGE

2-Dimensional gel electrophoresis

HS

Heat stress

HSP

Heat shock protein

IL

Interleukin

MIF

Macrophage migration inhibitory factor

MHS

Moderate HS

MOD

Multi organ dysfunction

MALDI-TOF

Matrix assisted laser desorption/ionization-time of flight

NRF-2

Nuclear factor erythroid 2-related factor

RH

Relative humidity

ROS

Reactive oxygen species

SEM

Standard error mean

SHS

Severe HS

SDS-PAGE

Sodium dodecyl sulfate-poly acrylamide gel electrophoresis

SOD 1

Superoxide dismutase 1

T_a

Ambient temperature

T_c

Core temperature

Authors' contributions

Experimental plan and study design: RCM and AG. Funding obtained: RCM and LG. Methodology: AG, DS, HG and RCM. Data analysis: RCM, AG, AS, DC. Manuscript writing: RCM and AG. All authors read and approved the final manuscript.

Funding

This work was supported by the Defence Institute of Physiology and Allied Sciences, Defence Research and Development Organisation, Government of India, Ministry of Defence, Lucknow Road Timarpur, Delhi-110054, India under project/grant number: DIPAS/TASK-184 and DIP-265.

Declarations

Conflicts of interest

Author(s) declares no conflict of interest.

Ethical approval

The study protocols including heat exposure was approved by the Institutional Animal Ethical Committee (IAEC/DIPAS/2015–17) of Defence Institute of Physiology and Allied Sciences, Delhi-110054, India.

Consent to participate

Not applicable'.

Consent for publication

Not applicable'.

Data availability

All raw data is available and part of them are provided as supplementary figures and data.

Code availability

Not applicable'.