Over-expression of HSP70 induces apoptosis of intestinal epithelial cells in heat-stressed pigs: A proteomic approach

Heat stress (HS)-induced intestinal epithelial cell apoptosis may play a pivotal role in intestinal barrier function injury in animals. However, the underlying molecular mechanism by which HS induces apoptosis of intestinal epithelial cells is still poorly understood. The trial uses a prospective study. Firstly, a eukaryotic expression vector for the HSP70 gene was constructed and transfected into intestinal porcine epithelial cells (IPEC-J2) and then analyzed with a functional proteomics approach followed by liquid-chromatography-tandem mass spectrometry (LC-MS/MS) identication. Next, heat stress treatment of IPEC-J2 cells and pigs were performed. Immunoblotting and ow cytometry were used to examine the protein changes and cellular apoptosis.


Results
246 differentially expressed proteins (DEPs) were identi ed in HSP70-overexpressed IPEC-J2 cells, functional annotation suggested that most of the DEPs were related to ECM-receptor interaction, focal adhesion, and apoptosis. Moreover, apoptosis rate was increased in IPEC-J2 cells transfected with porcine HSP70 overexpressing plasmid accompanied with the expression of apoptosis-related proteins, Caspase-3, PARP, and Bax were up-regulated, whereas Bcl-2 protein expression was down-regulated.
Interestingly, increased Caspase-3, PARP, and Bax with decreased Bcl-2 were observed in IPEC-J2 cells under heat stress conditions. In an in vivo porcine model, HS-induced cellular apoptosis in the duodenum, cecum, and colon, as well the up-regulation of HSP70 in the intestinal tissues.

Conclusions
HSP70 may play a regulatory role in cellular apoptosis within IPEC-J2 cells. Taken together, our ndings provide new insights for understanding the molecular mechanisms of cellular apoptosis associated with HSP70.
The temperatures greater than human normal body temperature, (from 41.6 ℃ to 42 ℃), can result in cell death (from apoptosis) in a few hours [7]. The mild hyperthermia may induce apoptosis which is prevented in thermotolerant cells [8,9]. HS-induced human endothelial cells apoptosis via the mitochondrial pathway in form of mitochondrial p53 translocation in ROS dependent and ensure Ca 2+ dyshomeostasis, while Bax mitochondrial translocation, as the upstream events casing apoptosis taken place [10]. HS-induced cellular apoptosis and survival maintenance through the mitochondrial apoptotic pathway Bax/Caspase9/Caspase3 and apoptosis-inhibiting Bcl-2 in the trophoblast cells [11]. Moreover, HS induces HSP70 expression and the production of pro-in ammatory cytokines, which modulates the immune function in intestine of swine [12,13]. In addition, the expression of extracellular HSP70 induces the production of TNF-α and IL-6 through the activation of Toll-like receptor 4 (TLR4) signaling pathway in mouse bone marrow-derived mast cells [14]. In particular, when HSP70 fused with-or in complex withhuman with hantavirus nucleocapsid protein, it greatly enhances both Th1 and Th2 responses in mice [15]. Based on above evidences, the interaction mechanism of HSP70 induce apoptosis in intestinal epithelial cells of heat-stressed pigs requires further investigation.
In recent years, quantitative proteomic studies were used to investigate the underlying molecular response in porcine cells and tissues, including mesenchymal stem cells [16], intestines [17], pulmonary alveolar macrophages [18], and porcine circovirus type 2 (PCV2) infected PK-15 cells [19]. Nevertheless, no systematic study combined a proteomic analysis was performed to investigate the mechanism of HSP70 expression and cellular apoptosis in intestinal epithelial cells of heat-stressed pigs. Thus, in the present study, a functional proteomics approach was combined with LC-MS/MS to measure the interaction of HSP70 induced epithelial cell survival in heat-stressed pigs.
Methods IPEC-J2 cells were obtained from the Collection of Cell Lines in the College of Veterinary Medicine, Guangdong Ocean University, China.
The labeled peptides were fractionated by strong cation exchange (SCX) chromatography and the detailed steps was described by Han et al [21].

LC -ESI MS/MS and data analysis
The protocols of experiments were performed on a Q Exactive mass spectrometer coupled to Easy nLC (Thermo Fisher Scienti c, Shanghai, China). The steps are the same as mentioned by Wu et al [22]. Using the top 10 most abundant precursor ions as the MS/MS data modi cation. Subsequently, the detailed steps about identi cation and quanti cation of proteins as well as the determination of the DEPs in the IPEC-J2 cells transfected with pHSP70 and Empty DNA plasmids were same and presentation in our previous research [23]. Gene ontology (GO) and Kyoto encyclopedia of genes and genomes (KEGG) databases were used to facilitate the biological interpretation and pathway analysis of the identi ed proteins in this study using the online tool, DAVID (http://david.abcc.ncifcrf.gov).

Immunoblotting
Immunoblotting was performed as described previously by An et al. [24]. Brie y, total protein from IPEC-J2 cells transfected with pHSP70 and Empty DNA plasmid groups, or from intestine with heat stress treatment was extracted using a total protein extraction kit (Biochain, Hayward, CA, USA), and quanti ed using the BCA protein assay kit (Pierce, Rockford, IL, USA). Thereafter, proteins were separated on SDS-PAGE gels, and then transferred to polyvinylidene uoride membranes (Millipore, Darmstadt, Germany).
Blocking the membranes in 5% fat-free milk/TBST solution at room temperature for 2 h and then incubated overnight at 4℃ with primary antibodies, anti-PPM1B (1:1,000, Proteintech, China; loading control). Subsequently, the membranes were incubated with the corresponding secondary antibodies conjugated to horseradish-peroxidase-conjugated at room temperature for 2 h. A chemiluminescence method was used to visualize immunoreactive bands and the positive bands were visualized by enhanced chemo-luminescence (Tanon, Shanghai, China). The bands were analyzed semi-quantitatively via densitometry using Gel-Pro Analyzer v4.0 software (Meyer Instruments, Houston, TX, USA), and the relative protein expression levels were normalized to β-actin.
Flow cytometry assay IPEC-J2 cells grown into 6-well plates were transfected with pHSP70 and Empty DNA plasmid, respectively. Afterward, the cells were harvested for cellular apoptosis analysis with Annexin V-FITC apoptosis detection kit (BD Biosciences, San Jose, CA, USA), according to the manufacturer's instructions. The apoptotic results were analyzed using the same method as described previously by Ma et al [11].

Animals ethics statement and sample collection
Twenty-four crossbred pigs (Duroc × Local Luchuan pig ) weighing 15 ± 1 kg were obtained from the farm in the Guangdong province of China. The animals experiment and protocols were approved by the Guangdong Ocean University Animal Ethics Committee (Permit No. 206-1108). The pigs were randomly assigned to two groups (test and control, 12 barrows per groups). They were housed in three pens (1.5 m 2 per pig) with two females and two males in each pen, and habituated for 1 week. Pigs allotted to HS treatment were kept at 35±1℃ (carried out in an arti cial climate chamber) in a controlled climate room. Control pigs were kept at an ambient temperature of 20±2℃, and the relative humidity of the two groups was kept at approximately 85%. The experimental design was carried out as described previously [23]. All pigs were euthanized using a head-only electric stunning tong apparatus on the morning of sampling days 1, 7, 14, and 21 after the start of the experimental protocol, followed by manual exsanguination.
Duodenum, cecum, and colon were immediately removed after slaughter, and used for cellular apoptosis analysis.

Cellular apoptosis analysis
Serial 6-μm-thick sections were treated with DNase to fragment the DNA (positive control) and 50 μL of the prepared terminal deoxynucleotidyl transferase reaction mixture was added to each slide. The slides were incubated for 60 min at 37°C, rinsed with deionized water, and washed with 3% bovine serum albumin (BSA) + 0.1% Triton ® X-100 in phosphate-buffered saline (PBS) for 5 min. Following another rinse with 1 × PBS, 50 μL of the Click-iT ® Plus TUNEL reaction cocktail was added to each slide, ensuring that the solution was spread evenly over the surface. The slides were incubated for 30 min at 37°C in the dark. The Click-iT ® Plus TUNEL reaction cocktail was then removed, after which the slides were washed with 3% BSA in PBS for 5 min, and then rinsed with 1 ×PBS. Cellular apoptosis was examined under a uorescence microscope (Anjue Medical Equipment Co., Ningbo, China). The optical density of apoptotic cells per unit area was measured using Image Pro Plus 6.0.

Statistical analysis
All results in this study are represented as mean ± SEM. Student's t-test was used to compare differences of protein expression between the pHSP70 and pEmpty transfected IPEC-J2 cells groups and calculate P values. Proteins with at least two unique peptides and FDR of < 0.01 was quali ed for further quanti cation data analysis. Fold change of ≥1.3 or ≤0.77 was set as the threshold to identify differently expressed proteins. Differences in expression of HSP70 and apoptosis-related proteins, Caspase-3, PARP, Bax, and Bcl-2 in cells were analyzed using one-way ANOVA. *indicates p < 0.05 and ** p < 0.01.

Ectopic expression of porcine HSP70 in IPEC-J2 cells
We used a PCR technique to isolate cDNA fragments of the HSP70 gene from porcine peripheral blood mononuclear cells (PBMCs) and cDNA fragments were inserted into the PMD18-T vector (TaKaRa, Beijing, China). The pcDNA3.1/V5-His-TOPO vector with a His-tag at the C-terminus was ligated with porcine HSP70 (pHSP70) cDNA gene sequences, yielding pcDNA3.1/V5-HSP70-His (Fig. 1A, B). DNA plasmids of pHSP70 and the Empty vector were transfected into IPEC-J2 cells. Immunoblotting analysis showed that the HSP70 levels were signi cantly increased in transfected IPEC-J2 cells (p < 0.01). Thus, a eukaryotic expression vector for the HSP70 gene was successfully constructed (Fig. 1C, D).
iTRAQ proteomic analysis to identi ed proteins in IPEC-J2 cell transfected with pHSP70 To further investigate the underlying functions and features of HSP70 in IPEC-J2 cells, a quantitative proteomic method, iTRAQ analysis coupled with LC-MS/MS was used to identify and quantify the potential changes of the IPEC-J2 cells transfected with pHSP70 DNA plasmid. A total of 3546 proteins were identi ed, and a MS comparison of the protein expression identi ed 246 proteins that were up-or down-regulated (p < 0.05), as calculated by MaxQuant [25]. Hundred and forty-four proteins were downregulated, and hundred and two proteins were highly upregulated. Among the upregulated proteins, higher expression levels of HSP70 were detected (2.56-fold). A partial list of the up-and downregulated proteins, their respective peptides, corresponding P values, and relevant biological processes are shown in Tables 1 and 2. All the DEPs are provided in the Supplementary Table.  To understand the molecular/functional classes and subcellular annotations of the DEPs, the UniProtKB/Swiss-Prot and TrEMBL protein databases, as well as GO database were used to analyzed the underlying biological function of the 246 DEPs. Our results showed that both upregulated and downregulated proteins in IPEC-J2 cells transfected with pHSP70 and Empty DNA plasmid were localized in the same cellular organelles and extracellularly, however, at different ratios ( Fig. 2A, B).
We used the DAVID database to determine the genes associated with differentially abundant proteins and found that the DEPs can be categorized as: involved in biological processes, molecular functions, and cellular components. The biological processes of the upregulated and downregulated proteins were primarily associated with cellular processes, single-organism processes, metabolic processes, biological regulation, response to stimuli, cellular component organization or biogenesis, and localization (Fig. 3A). The cellular components of the DEPs include cells, organelles, membrane, macromolecular complex, extracellular regions, membrane-enclosed lumen, and cell junctions (Fig. 3B). The molecular functions of the proteins included binding, catalytic activity, structural/molecular activity, molecular function regulator, transporter activity, and molecular transducer activity (Fig. 3C). KEGG pathway enrichment analysis was performed to ascertain the biological signaling pathway of HSP70 in IPEC-J2 cells. The top fourteen most signi cantly enriched pathways identi ed by this analysis include arrhythmogenic right ventricular cardiomyopathy (ARVC), ECM-receptor interaction, dilated cardiomyopathy, focal adhesion, and apoptosis (Fig. 3D). A protein-protein interaction (PPI) network of the 246 DEPs was constructed by STRING (Fig. 4). The PPI network analysis revealed that most of the DEPs were highly connected. Two selected DEPs, PPM1B and CGNL-1, participated in various biological processes, such as in ammation and signal transduction, which may modulate signaling pathways and cellular function in IPEC-J2 cells.
Three candidate proteins, PPM1B, CGNL-1, and Caldesmon-1, were selected for validation of the proteomic results. The immunoblotting analysis showed that the expression of PPM1B and CGNL-1 were signi cantly upregulated in IPEC-J2 cells transfected with pHSP70 DNA plasmid compared with cells transfected with Empty DNA plasmid (p < 0.05 and 0.01, respectively), whereas the level of Caldesmon-1 was obviously downregulated (p < 0.05) (Fig. 5A, B). In sum, these results are consistent with data obtained from iTRAQ ndings.

Overexpression of HSP70 induces cellular apoptosis in in vitro and in vivo models
Proteomics results showed that HSP70 was related to cellular apoptosis in IPEC-J2 cells. To further investigate the interaction between HSP70 and cellular apoptosis, as shown in Fig. 6, we noticed that four cellular apoptosis-related proteins, including Caspase-3, PARP, Bcl-2, and Bax, were signi cantly changed in IPEC-J2 cells transfected with pHSP70 DNA plasmid compared with cells transfected with Empty DNA plasmid (p < 0.05 and 0.01, respectively). The expression of pro-apoptotic protein Caspase-3, PARP, and Bax were increased while the anti-apoptotic protein Bcl-2 was decreased in IPEC-J2 cells transfected with pHSP70 DNA plasmid compared to cells transfected with Empty DNA plasmid. Further, we detected the cellular apoptotic rate of IPEC-J2 cells by ow cytometry. The assay revealed that the percentage of apoptotic cells was much higher in IPEC-J2 cells transfected with pHSP70 DNA plasmid than that in control IPEC-J2 cells (Fig. 6C, D). Next, we examined the levels of HSP70 in IPEC-J2 cells challenged by HS for 2 h and allowed to recover for up to 1.5 h. The results clearly showed that the expression of HSP70 was signi cantly up-regulated (p < 0.01) in IPEC-J2 cells under HS and recovery condition (Fig. 7A, B). Moreover, immunoblotting analysis revealed that HS treatment promoted the expression levels of apoptosis-related proteins, Caspase-3, PARP, and Bax, and decreased the levels of Bcl-2 ( Fig. 7C-F).
We further elucidated the effect of HS treatment on HSP70 production and apoptosis in porcine in vivo intestinal cell model. The distribution of intestinal apoptotic cells after heat stress was detected by the TUNEL method. According to TUNEL assay, heat stress group had signi cantly higher ratio of apoptotic cells to total cells in duodenum and cecum on day 7 (p < 0.05) (Fig. 8A-D). On the other hand, HS markedly increased the ratio of apoptotic cells to total cells on days 7 and 14 in colon (p < 0.01) (Fig. 8E,  F). Furthermore, the expression of HSP70 in colon was signi cantly upregulated in HS pigs compared to that in control group (p < 0.05 and 0.01, respectively) (Fig. 8G, H).

Discussion
In recent years, there has been an increased interest in searching the potential effect of heat stress on host-apoptosis interaction. A study in baboon model revealed that HS-induced widespread apoptosis in spleen, thymus, and small bowel [26]. In in vitro studies, there were evidences showing that apoptosis can be promoted by application of heat stress [27,28]. In this study, we have found for the rst time that overexpression of HSP70 could induce apoptosis of intestinal epithelial cell in pigs. This nding reveals that HSP70 might be involved in the regulation of heat stress induced related disease, thus further developing the underlying functions of HSP70.
HSP70, one of the highly conserved molecular chaperone, plays a critical role in maintaining cellular protein homeostasis [29,30]. Intracellular HSP synthesis is very susceptible to abnormal environmental stressors, such as exposure to high temperature, chemicals, bacteria, viruses, and heavy metals. This synthesis plays a vital role in protecting cells against the abnormal environmental stress conditions. A previous study has shown that oxidative stress induced by benzene and its homologues can substantially induce HSPs expression in cells [31]. In this study, it was found that heat stress dramatically induced HSP70 expression in in vitro cultured IPEC-J2 cells and in the duodenum of pigs. Previous research has also demonstrated that the expression of HSPs were related to the phosphoinositide-3kinases (PI3K)/protein kinase B (Akt) signaling pathway like HSP27 can interact with PI3K/Akt and subsequently inhibit apoptosis [32]. In vitro cultured lung broblasts, exposure to high temperature treatment can activate the PI3K/Akt signaling pathway and leading to high expression of HSP70. Furthermore, when inhibited the activation of PI3K/Akt by siRNA the high HSP70 expression was also disappeared [33]. Therefore, further studies are needed to elucidate the role of PI3K/Akt signaling pathway in heat stress induced HSP70 expression.
Apoptosis e ciently contributes to intracellular pathogen removal by eliminating the favorable intracellular survival mechanisms. The relationship between invading pathogen and its host cells is a complex interaction, in which the pathogen strives to survive and replicate [34]. A former study has shown that HSP70 would suppress apoptosis by directly associating with Apaf-1 and blocking the assembly of a functional apoptosome [35]. Caspase-3, caspase-7, and Bax are the calpain-cleaved substrates. Calpains cleave and inactivate caspase-3, -7 in biochemical ex vivo apoptosis assays and during Ca 2+ ionophoreinduced apoptosis in vitro [36]. However, our results show that heat stress induces apoptosis in intestinal epithelia not only in in vivo experiment, but also in in vitro cultured cells that were exposed to heat stress for 2 h. Meanwhile, HSP70 overexpression induced the apoptosis of IPEC-J2 cells as well the upregulation of caspase-3, PARP, and the ratio of Bax to Bcl-2. These ndings suggesting that HSP70 may plays a vital role in heat stress induced apoptosis of intestinal epithelial cells.
KEGG enrichment analysis showed that apoptosis is one of the most signi cantly enriched pathway, our proteomic results showed that the caspase-7 and poly [ADP-ribose] polymerase (PARP) were downregulated in IPEC-J2 cells transfected with pHSP70 plasmid. Stable isotope labeling analysis revealed that apoptosis-related proteins, including caspase-3, caspase-7, and Bax-alpha protein were downregulated in PK-15 cells infected with PCV-2 [19]. These ndings are consistent with our proteomic results. The reason maybe is those apoptosis-related proteins (both pro-and anti-apoptotic proteins) are either activated or inactivated by calpains. In the proteomics results, calpain-cleaved products of apoptosis-related proteins are not identi ed by LC − MS/MS, resulting in their apparent reduced expression levels. Furthermore, calpains activate caspase-7 in B cell apoptosis [37]. Indeed, the function of calpains in apoptosis is puzzling, and appears to be highly dependent on the speci c cell type and apoptotic stimulus [38]. This phenomenon is linked to their seemingly different preferences for individual substrates among experimental systems.
Cell adhesion molecules are typically single-pass transmembrane receptors [39]. It reported that the cell adhesion molecules including the immunoglobulin super family of cell adhesion molecules (IgCAMs), cadherins, integrins, and the C-type lectin-like domain proteins (CTLDs) [40]. Integrins, as one of the major classes of receptors, mediate the interactions between the extracellular matrix (ECM) and collagen, brinogen, bronectin, and vitronectin [41]. Previous research revealed that integrins act essential links between the extracellular environment and the intracellular signaling pathway, which play a role in cell fate, such as apoptosis, differentiation, survival, and transcription [42]. Integrins can activate the RhoA/RhoKinase and MAPK signaling pathways and participate in wall remodeling changes by binding to extracellular ligands, such as bronectin, laminin, and matrix metalloproteinase-2 [43,44]. A previous study showed that GEGs in the aneurysm wall tended to take part in the process of extracellular matrix receptor interaction, focal adhesion and cellular communication [45]. In the proteomic analysis results, integrin beta and integrin subunit alpha 6 were upregulated approximately 1.2-fold in IPEC-J2 cells transfected with pHSP70 plasmid. Hence, we assume that alteration of these proteins might be involved in HSP70-mediated biological functions in IPEC-J2 cells.
Here, proteomics and bioinformatics were combined to identify proteins that are differentially expressed in IPEC-J2 cells transfected with pHSP70 DNA plasmid. A total of 246 DEPs were identi ed, including those involved in regulation of cellular proliferation, cellular assembly and organization, and signaling.
Furthermore, as cells were exposed to heat stress, overexpression of HSP70 ensued, further leading to increased Caspase-3, PARP, and ratio of Bax to Bcl-2. Moreover, in vivo studies revealed that heat stress signi cantly increased the ratio of apoptosis in intestinal epithelial cells and the expression of HSP70.
Taken together, these ndings suggest that HSP70 expression induces apoptosis in the intestine of heatstressed pigs. Thus, our study provides an important clue concerning the molecular mechanism underlying the development of apoptosis in the intestines due to heat stress.

Conclusions
Over-expression of HSP70 was observed in IPEC-J2 cells transfected with pHSP70 DNA plasmid and the iTRAQ-based quantitative proteomics approach showed that HSP70 may extensively in uence the expression of proteins associated with molecular response such as signaling pathways, cell adhesion, and apoptosis in IPEC-J2 cells. Apoptosis-related proteins, Caspase-3, PARP, Bax, and Bcl-2 expression were signi cantly activated in IPEC-J2 cells transfected with porcine HSP70 plasmid. Furthermore, heat stress-induced HSP70 overexpression and cellular apoptosis was observed in the intestine of pigs. Taken together, our expression pro les provide a novel insight into the molecular mechanisms of HSP70

Availability of data and material
The data that support the ndings of this study are available from the corresponding author upon reasonable request.
Ethics approval and consent to participate No ethics approval was required for this study that did not involve patients or patient data. Figure 1 The construction of HSP70 ectopic expression system and HSP70 overexpression in IPEC-J2 cells. A PCR product of HSP70 plasmid. B PCR product of pcDNA 3.1/V5-His-HSP70. C-D Densitometric analysis generated for western blots of HSP70 expression. **indicates p < 0.01.

Figure 2
Location of the proteins with differential expression (p < 0.05) between pHSP70-transfected and Emptytransfected IPEC-J2 cells. A Up-regulated proteins. B down-regulated proteins.