Hyperthermia and protein homeostasis: Cytoprotection and cell death

Highlights

• Hyperthermia (HT) is a promising strategy for cancer therapy.

• HT-induced disturbance in protein homeostasis leads to thermotolerance.

• Heat shock factor 1 induced heat shock proteins synthesis prevent cell death and maintain pro-survival environment.

• Generation of reactive oxygen species activates autophagic process to maintain cell survival conditions.

Abstract

Protein homeostasis or proteostasis, the correct balance between production and degradation of proteins, is an essential pillar for proper cellular function. Among the several cellular mechanisms that disrupt homeostatic conditions in cancer cells, hyperthermia (HT) has shown promising anti-tumor effects. However, cancer cells are also capable of thermoresistance. Indeed, HT-induced protein denaturation and aggregation results in the up regulation of heat shock proteins, a group of molecular chaperones with cytoprotective and anti-apoptotic properties via stress-inducible transcription factor, heat shock factor 1(HSF1). Heat shock proteins assist in the refolding of misfolded proteins and aids in their elimination if they become irreversibly damaged by various stressors. Furthermore, HSF1 also initiates the unfolded protein response in the endoplasmic reticulum (ER) to assist in the protein folding capacity of ER and also promotes the translation of pro-survival proteins’ mRNA such as activating transcription factor 4 (ATF 4). Moreover, HT associated induction of microRNAs is also involved in thermal resistance of cancer cells via up-regulation of anti-apoptotic Bcl-2 proteins and down regulation of pro-apoptotic Bax and caspase 3 activities. Another cellular protection in response to stressors is Autophagy, which is regulated by the Mammalian target of rapamycin (mTOR) protein. Kinase activity in mTOR phosphorylates HSF1 and promotes its nuclear translocation for heat shock protein synthesis. Over-expression of heat shock proteins are reported to up-regulate Beclin-1, an autophagy initiator. Moreover, HT-induced reactive oxygen species (ROS) generation is sensitized by transcription factor NF-E2 related factor 2 (Nrf2) and activates the cellular expression of antioxidants and autophagy gene. Furthermore, ROS also potentiates autophagy via activation of Beclin-1. Inhibition of thermotolerance can potentiate HT-induced apoptosis.

Here, we outlined that heat stress alters cellular proteins which activates cellular homeostatic processes to promote cell survival and make cancer cells thermotolerant.

Introduction

Maintenance of homeostasis is the key feature in cellular physiology. It keeps key regulated variables within an acceptable range with the help of specialized sensors. These sensors are signaling proteins that detect alterations in protein folding, reactive oxygen species (ROS) levels, and nutrient availability (Chovatiya and Medzhitov, 2014). Any stressful stimulus that disrupts cellular homeostasis activates intracellular mechanisms to rebalance the biochemical processes within the cell.

Hyperthermia (HT) is a known modality for the treatment of cancer for centuries (Ahmed et al., 2015). Heat stress can cause many intracellular changes and can even lead to loss of cellular homeostasis (Richter et al., 2010). HT-induced protein denaturation and aggregation is the key event in the disruption of cellular homeostasis. At the molecular level, it targets a variety of proteins such as cytoskeletal structures, plasma membrane components, intracellular enzymes and signal transduction molecules (Kaur et al., 2016). Both prokaryotes and eukaryotes depend on the heat shock protein (Hsp) chaperone system, which is a house keeper of protein homeostasis. It folds and assembles proteins, as well as facilitates the breakdown of stress-induced irreversibly denatured proteins (Dokladny et al., 2013). Autophagy is another mechanism present only in eukaryotic cells to deal with cellular stress. In this process, misfolded proteins, large protein aggregates and whole damaged organelles inaccessible to smaller proteolytic systems in the cell are delivered to lysosome for degradation (Dokladny et al., 2013). It functions as a recycling program to provide biofuel to the cell from degraded macromolecules to maintain sufficient ATP production for survival (Dokladny et al., 2015). Therefore, the Hsp family and autophagy are the two controllers of cellular proteostasis. Under stressful cellular conditions, these two mechanisms are likely to complement each other.

In this short review, we focused on the effects of heat stress-induced alterations in protein homeostatic processes and on the initiation of cell death after inhibiting these processes.

The cellular response to stressful stimuli ranges from activation of pathways that promotes survival to elimination of damaged cells via programmed cell death. Initially, cells respond to stressful stimuli by helping them to defend against and recover from the insult. However, if the stimulus is strong enough, then cells activate death signaling pathways. The effectiveness of HT is dependent on the temperature and duration of treatment. Heat stress is a known inducer of apoptosis in cancer cells. Previously, we reviewed the detailed mechanism of HT-induced apoptosis (Ahmed et al., 2015); therefore, in the current review we will focus on other cellular responses to heat stress.

Despite the enormous effects of HT to induce apoptosis in cancer cells (Ahmed et al., 2015) it is also known to induce transient resistance in cells against subsequent heat stress called thermotolerance. This resistance can be induced by short exposures to higher temperatures (42–45 °C for 30 min), or by continuous heating at mild temperatures (39.5–41.5 °C for 3–24 h) (Glory et al., 2014). Thermotolerance is a physiological homeostatic phenomenon in cells to deal with stress but it is one of the biggest hurdles in cancer treatment by HT (Fig. 1). Hyperthermia induces heat shock response (HSR) mechanism with the help of molecular chaperones called Hsp. They are up-regulated via heat shock transcription factor 1 (HSF1), the main player in the cellular homeostatic process (Tabuchi and Kondo, 2013). Heat shock proteins play an important role in protein protection and homeostasis. In physiological states, HSF1 is present in the cytoplasm as an inactive monomer. Under cellular stress, such as heat stress the HSR mechanism is activated. The inactive HSF1 monomer is converted to a DNA-binding homotrimer that translocates from the cytoplasm to the nucleus where it binds with cis-acting DNA elements, termed heat shock elements (HSE), which are present in heat shock genes, and activate transcription of the Hsp genes such as Hsp27, Hsp70 and Hsp90 (Tabuchi and Kondo, 2013; Wang et al., 2014; Xia et al., 2012).The function of Hsp is to regulate protein folding, transport, translocation and assembly (Zhang and Calderwood, 2011). It assists in the refolding of misfolded proteins and aids the degradation of abnormal proteins through ubiquitin-proteasome system (UPS) (Kim et al., 2011). Thus, HSR and ubiquitin-proteasome degradation pathways are closely interconnected.

Stress-inducible Hsp such as Hsp27, Hsp70 and Hsp90 are the most extensively studied (Ahmed et al., 2015; Richter et al., 2010). Many different death stimuli further increase their expression (Wang et al., 2014) and repress apoptosis through inhibition of pro-apoptotic factors such as p53, Bax, Bid, Akt, Apaf-1 and other Bcl-2 family members (Ikwebue et al., 2018). Cancer cells have high HSR and proteasome activities because they are coping with elevated levels of constitutively misfolded proteins mainly due to the rapid rate of proliferation and specific intracellular conditions of cancer cells such as hypoxia or glycolysis-related acidification (Rybinski et al., 2013). A number of studies reported the resistance of cancer cells to apoptosis due to their high HSR and proteasome activity (Dudeja et al., 2011; Santagata et al., 2011; Ishiwata et al., 2012; Mendillo et al., 2012; Gabai et al., 2012; Jin et al., 2011; Fujimoto et al., 2012).Therefore, blocking HSR by targeted inhibition of HSF1 and Hsp is an effective way to reverse thermotolerance in cancer cells (Fig. 1) ( Grimmig et al., 2017; Ali et al., 2016). Silencing of HSF1 using small interfering RNA (siRNA) enhances the sensitivity of human oral squamous cell carcinoma at 42 °C and 44 °C HT for 90 min (Tabuchi et al., 2011). The melanoma cells silencing of Hsp27, also known as HSPB1, showed significant intolerance of HT treatment via reduced cell viability, increased percentage of apoptotic cells, and cell cycle arrest (Wang et al., 2016). Another study reported that co-inhibition of Hsp70 and Hsp90 synergistically enhanced apoptosis after thermotherapy in nasopharyngeal carcinoma (Cui et al., 2012). Hence, treatment of cancer cells with a combination of cytotoxic drugs and heat stress make them more thermo sensitive as compared to normal cells because combined treatment increases the level of misfolded proteins beyond the capacity of intracellular HSR mechanism to synthesize chaperones and ubiquitin for thermoresistance (Neznanov et al., 2011).

High proteasome activity has been detected in several different types of tumors (Arlt et al., 2009; Soave et al., 2017). HT has been reported to induce proteasomal degradation of anti-apoptosis regulator c-FLIP and augmented anticancer agent-induced apoptosis (Song et al., 2013). Furthermore, as compared to healthy cells, cancerous cells have acquired defects in their DNA damage repair genes which compromised their ability to adequately process DNA damage and therefore relied on the up regulation of mutagenic pathways for DNA damage repair (Trenner and Sartori, 2019). These DNA repair pathways can counteract the cytotoxic effects of anticancer agents and radiation. HT is reported to interfere with DNA repair pathway by degrading BRCA2 protein which is required for the repairing of DNA double strand break via homologous recombination (Tempel et al., 2017). Furthermore, another research demonstrated an alternative approach to reverse thermotolerance via simultaneous induction of protein misfolding with heat stress and inhibition of proteasome-mediated degradation by using proteasome inhibitor. This combination treatment enhances proteotoxic stress in cancer cells which leads to apoptosis in vitro and suppression of cell growth in vivo (Neznanov et al., 2011).

HT-induced alteration in protein homeostasis also affects centrosomes which is the main microtubule organizing center of the cell (Hut et al., 2005). Hsp 70 assists in the repair of centrosome after heat stress (Hut et al., 2005). Normal cells arrest their cell cycle until centrosomal damage is repaired whereas, tumor cells resume cell division before the centrosomes are repaired (Nakahata et al., 2002). Once mitosis starts, disassembled centrosomes form multiple poles which in turn cause multipolar cell division leading to mitotic catastrophe (MC). (Pawlik et al., 2013; Nakahata et al., 2002). Cells undergoing MC may subsequently die by apoptosis or other modes of cell death (Wilson and Bedford, 2010; Sorokina et al., 2017).

Some studies also examined the effect of heat stress on proteins such as RIPK1 and RIPK3 to investigate necroptosis (Thompson et al., 2014; Mouratidis et al., 2015). With increasing thermal dose, necrosis is the principal mechanism of heat stress-induced cell death. It has been reported that inhibition of RIPK1, the principal kinase for necroptosis, increased hepatocellular carcinoma (HCC) cell viability under both physiologic (37 °C) and hyperthermic (45–50 °C) temperatures in the N1S1 cell line. However, in the AS30D HCC cell line increase in cell viability was observed only under hyperthermic conditions (Thompson et al., 2014). Inhibition of RIPK1 has no significant effects on cell viability on HCT116 and HT 29 cells (Mouratidis et al., 2015). These studies suggest that heat stress can induce necroptosis or apoptosis in a cell type and temperature dependent manner. Further research is necessary to investigate the role of necroptosis in heat-induced cell death and as suggested by Thompson et al., more studies are needed to explore the signaling mechanisms induced by heat stress upstream of RIPK1.

The role of stress-induced microRNA (miR/miRNA) in HT-treated cells has been widely studied (Roufayel et al., 2014; Barna et al., 2018). MicroRNAs are short non-coding RNAs of 18–24 nucleotides long in length (Permenter et al., 2019). They originate from a longer molecule that is processed into the nucleus to a precursor miRNA (pre-miRNA) by a class 2 RNase III enzyme, Drosha. It is then transported to the cytoplasm where it is further processed into a mature miRNA by a Dicer, an RNase III type protein (Fig. 1) (Wahid et al., 2010). miRNA plays a vital regulatory role by targeting specific mRNA for degradation or translation repression (Permenter et al., 2019). HSF1 is reported to be a regulator of miRNA via a heat stress dependent and independent manner (Burnquell et al., 2017). In heat stressed cells, miR-216b abundance inhibits the expression of Bax and caspase-3, and promotes the expression of anti-apoptotic Bcl-2 protein. This results in the resistance of cells to HT-induced apoptosis. The inhibition of miR-216b increases caspase-3 and Bax activities and down regulates the anti-apoptotic Bcl-2 levels (Cai et al., 2018). Another study reported that HT decreases the synthesis of oncogenic miR23a. This miR-23a targets the pro-apoptotic BH3-only protein NOXA mRNA. NOXA inhibits the activities of anti-apoptotic Bcl-2 proteins and promotes apoptosis. However, HT-induced Hsp70 expression maintains miR-23a levels and prevents the accumulation of NOXA mRNA and protein leading to thermo-resistance (Roufayel et al., 2014). HT is also reported to increase the expression of miR34-a which induces p53-dependent apoptosis in HCT116 cells (Luo et al., 2017). Whereas, in heat stressed Sca-1⁺ stem cells HSF1 suppressed miR34-a and promotes thermo-tolerant effects of Hsp70 (Fujimoto et al., 2012). This suggest that the thermo-resistant effect of miR34-a is dependent on cell type. Another study in human oral squamous cell carcinoma HSC-4 cell line demonstrated that low expression level of miR-27a is involved in thermal resistance in HSC-4 cells and transfection of cells with a miR-27a mimic oligonucleotide enhanced HT-induced cell death (Kariya et al., 2014).

In addition to the cytoplasmic response, HSF1 is also reported to induce the unfolded protein response (UPR) initiated in the endoplasmic reticulum (ER) (Barna et al., 2018). It has been reported that 42–43 °C HT enhanced the transcription of UPR key proteins in HeLa cells (Bettaieb and Averill-Bates, 2015). It is an adaptive response to the disturbances in ER homeostasis due to disorder of calcium levels, accumulation of unfolded or misfolded proteins and hypoxia (Fig. 1). The process of UPR depends on the activation of three main transmembrane ER sensor proteins namely PRKR-like ER kinase (PERK), inositol-requiring enzyme 1 (IRE1α) and activating transcription factor 6 (ATF6) (Gardner et al., 2013). During ER stress, when unfolded polypeptides becomes abundant within the ER, causing the IRE1α and ATF6 to work synergistically in transactivating ER chaperone genes such as glucose related protein 94 (Grp94), Erp72, glucose-related protein 78 (GRP78/BiP) and CypB (Remondelli and Renna, 2017; Devasthanam and Tomasi, 2017). The ER chaperones function collectively to increase the folding capacity of the ER (Remondelli and Renna, 2017; Devasthanam and Tomasi, 2017). Furthermore, the presence of misfolded proteins in the ER activates PERK to phosphorylate alpha subunit of eukaryotic translation initiation factor 2 (eIF2α) (B'chir et al., 2013). The phosphorylation of eIF2α decreases the translation of most mRNAs. However, under stressful conditions, it also favors increased mRNA translation of a selected number of proteins. One of the pro-survival proteins for which translation is increased is activating transcription factor 4 (ATF4) (B'chir et al., 2013). ATF4 is a master regulator that plays a crucial role in the adaptation to stressors by regulating the transcription of many genes (Devasthanam and Tomasi, 2017). Pre-exposure of HeLa cells to non-lethal temperature, such as 40 °C before a 42–43 °C heat stress, leads to thermoresistance with inhibition of heat-induced cell death. The 40 °C exposure was shown to blunt the transcription of ER stress proteins and to stimulate transcription of anti-apoptotic Hsp72 protein (Bettaieb and Averill-Bates, 2015).

The cellular stress response to mild HT also reveals a role of dicer protein in thermotolerance (Devasthanam and Tomasi, 2017). Elevated dicer protein levels are associated with phosphorylation of eIF2α and an increase in ATF4 mRNA to favor pro-survival outcome (Devasthanam and Tomasi, 2017). However, under conditions of prolonged and strong ER stress, the adaptive responses mediated by UPR are not sufficient to restore normal cellular homeostasis and subsequently activate apoptosis signaling pathways, such as the pro-apoptotic pathway of C/EBP-homologous protein (CHOP), apoptosis signal-regulating kinase 1/c-Jun NH2-terminal kinase (JNK), and ER localized cysteine protease, caspase-12 (Devasthanam and Tomasi, 2017; B'chir et al., 2013).

Autophagy is another homeostatic response present in cells exposed to stressful conditions (Murrow and Debnath, 2013). Depending on the exact cell type and conditions, it either acts as a protagonist or an antagonist of apoptosis (Kasprowska-Liśkiewicz, 2017). Autophagy helps cancer cells survival despite a shortage of nutrients and hypoxic conditions (Cirone et al., 2019; Dokladny et al., 2013). During this catabolic process, formation of vesicles called autophagosome engulf damaged cellular components and fuse with lysosome for their breakdown (Gump and Thoraburn, 2011). Its primary function is to clean up the cell by recycling damaged proteins, organelles and aggregates and provide new building blocks to replace damaged or depleted cellular components (Yang and Klionsky, 2010). This process is initiated by induction of several autophagy genes, such as microtubule-associated protein light chain 3 (LC3), Beclin-1 and other autophagy-related (Atg) protein (Li et al., 2011). Infections, DNA damage, and HT are known to enhance autophagy (Nivon et al., 2009). It has been demonstrated that overexpression of HSP70 can suppress the basal level of LC3-II protein and inhibit the activation of autophagy (Dokladny K., et al., 2013). On the other hand, knockdown or deletion of HSF1can increase the basal level of autophagy (Dokladny K., et al., 2013) while constitutive activation of HSF1 results in decreased autophagy (Zhao et al., 2009). These studies suggest that autophagy is under regulatory control of HSR via increased chaperone activity. When HSR is failed then autophagy becomes the primary mean by which cell promotes proteostasis (Dokladny et al., 2013). This shows that the two systems coordinate with each other to maintain protein homeostasis in cells during heat stress.

There are a series of proteins to control the mechanism of autophagy (Fig. 2). Mammalian target of rapamycin (mTOR) belonging to the phosphatidylinositol kinase-related kinase (PIKK) family is a key regulator of autophagy (Jung et al., 2010). It has two complexes, mTOR1 and mTOR2, each of which exhibit distinct functions and localizations (Chou et al., 2012). mTOR2 regulates cell survival and cytoskeletal organization (Zhou and Huang, 2011) while mTOR1 senses a number of stress signals such as protein misfolding, oxygen and energy deficiencies, growth factors deprivation and amino acid depletion (K.H. Su and Dai, 2017). It has been reported that the mTOR pathway is complexly linked to HSF1 activity (Khadrawy et al., 2019). The kinase activity present in mTOR1 is responsible for HSF1 phosphorylation and its nuclear translocation for HSP synthesis (Chou et al., 2012). Activation of HSF1 is also reported to induce the expression of the protein called BIS (Bcl-2 interacting cell death suppressor) also known as BAG3 (Nguyen and Kim, 2017). This protein has a multi-modular domain structure that permits its interaction with a variety of proteins including anti-apoptotic Bcl-2 protein and the HSC/HSP70 chaperone (Doong et al., 2003; Stürner and Behl, 2017). Hsp70-BAG3 complex exhibits its anti-apoptotic and pro-survival functions by modulating the activity of transcription factors such as NF-κB, hypoxia-inducible factor1-α (HIF1-α) and cell cycle regulators such as p21 and survivin (Colvin et al., 2014; Rapino et al., 2015). Several reports strongly suggest the participation of BAG3 in promoting autophagy (Stürner and Behl, 2017; Carra et al., 2008; Gamerdinger et al., 2009). The linkage between different stressors-induced Hsp and autophagy has been investigated (Li et al., 2011; Jiang et al., 2011) but research is needed to explore the linkage between heat stress-induced Hsp and activation of autophagy as a protective pathway.

HT is a known inducer of oxidative stress in cells (Slimen et al., 2014; Kikisato et al., 2015; Li et al., 2017). Proteins are the major target of ROS due to their elevated quantities compared to other cell components (Lévy et al., 2019). HT is also a source of oxidative stress induced apoptosis (Ahmed et al., 2015) however, cellular adaptation response to stress can initially resist apoptosis by activating Hsps, antioxidants, anti-apoptotic proteins and endoplasmic reticulum (ER) defenses (Bettaieb and Averill-Bates, 2008, 2015; Pallepati and Averill-Bates, 2010; Glory et al., 2014). ROS can regulate autophagy in a cell type and time-dependent manner (Scherz-Shouval and Elazar, 2011; Gibson, 2013). It has been reported that ROS act on the complex of Beclin-1 and anti-apoptotic Bcl-2 homologs such as Bcl-2, Bcl-xL and Mcl-1 (Wu et al., 2016) This complex represses the pro-autophagic activity of Beclin-1 (Gibson, 2013; Wu et al., 2016) and its dissociation by ROS activates Beclin-1 and promotes autophagy (Gibson, 2013). ROS is also known to induce autophagy through the regulation of NF-κB activity which leads to the induction of Beclin-1expression (Jiang et al., 2011; Boyer, 2014). Moreover, ROS is also reported to up-regulate the activity of HIF1-α which promotes the transcription of autophagy key proteins such as BCL2 interacting protein 3 (BNIP3) and NIP-like protein X (NIX) expression which regulates autophagy through Beclin-1 (Ciccarone et al., 2019). Many studies reported that apoptosis and autophagy are dependent on ROS production and inhibition of autophagy potentiates ROS dependent apoptosis (Ganguli et al., 2014; Xu et al., 2019). Other studies are necessary to explore a more detailed mechanism between heat stress-induced ROS generation and autophagy.

Another defense molecule involved in the HT-induced oxidative stress is the transcription factor NF-E2 related factor 2 (Nrf2) (Bozaykut et al., 2016). It is a key sensor of oxidative stress that can induce transcription of cytoprotective genes and protects cells from excessive oxidative stress. In basal conditions and in the absence of major cellular stresses, Nrf2 is present in cytoplasm in an inactive form with Kelch-like ECH-associated protein (Keap1) (Bozaykut et al., 2016; Kavian et al., 2018). Nrf2 regulation is linked to autophagy via p62, an autophagy substrate and cargo receptor. p62, also known as SQSMT1, transfers target proteins to autophagosome for degradation (Liu et al., 2016). In response to oxidative stress, p62 phosphorylation increases its binding affinity to Keap1 (Kapuy et al., 2018). This binding dissociates the Nrf2-Keap1 complex resulting in the release of Nrf2 (Kapuy et al., 2018). Keap1 and p62 complex are rapidly degraded by autophagy while Nrf2 translocates into the nucleus and activates the expression of many antioxidants and autophagy genes such as Atg3, Atg5, Atg7 and SQSMT1 (Boazykut et al., 2016; Kavian et al., 2018; Kapuy et al., 2018).

Mitochondrial membrane potential (MMP) is important for maintaining the physiological function of respiratory chain complexes which are located in the inner mitochondrial membrane and are responsible to generate ATP. Mitochondria are the major site of intracellular ROS production, which occurs via electron leakage as a byproduct of ATP generation by oxidative phosphorylation. During mild heat stress, antioxidant enzymes detoxify ROS and induce thermotolerance (Tchouagué M, 2019), but excessive ROS generation can overwhelm the capacity of these defenses, leading to the damage of functionality of mitochondria by inducing mitochondrial DNA mutations, and lipid peroxidation ultimately causing loss of MMP (Lin and Beal, 2006; Scherz-Shouval and Elazar, 2007). ROS directly attack the complexes in the mitochondrial respiratory chain and leads to the depletion of ATP synthesis (Guo et al., 2013).

Reduction in heat-induced ATP levels in cells results in an increase in the AMP/ATP ratio (Ciccarone et al., 2019). This activates AMP-activated protein kinase (AMPK), an energy sensing regulator. AMPK is also activated in response to oxidative stress to enhance autophagy (Kapuy et al., 2018). Activation of AMPK inhibits mTORC1 activity and promotes autophagy via dephosphorylation of UNC-51 like autophagy activating kinase 1 (ULK1) and autophagy related 13 (ATG13), two key components of the autophagy initiating complex (Ciccarone et al., 2019; Kapuy et al., 2018; Jiang et al., 2019; Su and Dai, 2017). This dephosphorylation leads to the activation of Beclin 1/PI-3 kinase III complex (Zhang and Calderwood, 2011). It has been demonstrated that ER chaperone, GRP78, induces autophagy in lung cancer cells after heat stress through activation of the AMPK-mTOR pathway. Therefore, inhibiting autophagy or knocking down GRP78 enhances HT-induced apoptosis in these lung cancer cells (Xie et al., 2016). Another study also reported that inhibition of autophagy by chloroquine significantly suppressed hepatocellular carcinoma growth and enhanced HT-induced apoptosis in vivo (Jiang et al., 2019). There are reports that once apoptosis is activated, proteins that are involved in the promotion of autophagy are cleaved or inactivated by caspases leading to inhibition of autophagy and enhancement of apoptosis (Wirawan et al., 2010; Kang et al., 2011).