Can the hyperthermia-mediated heat shock factor/heat shock protein 70 pathway dampen the cytokine storm during SARS-CoV-2 infection?
Abstract
Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is a major global public health problem. Infection by this virus involves many pathophysiological processes, such as a “cytokine storm,” that is, very aggressive inflammatory response that offers new perspectives for the management and treatment of patients. Here, we analyse relevant mechanism involved in the hyperthermia-mediated heat shock factors (HSFs)/heat shock proteins (HSP)70 pathway which may provide a possible treatment tool.
LINKED ARTICLES
This article is part of a themed issue on The Pharmacology of COVID-19. To view the other articles in this section visit http://onlinelibrary.wiley.com/doi/10.1111/bph.v177.21/issuetoc
Abbreviations
-
- ACE2
-
- angiotensin-converting enzyme 2
-
- C/EBP
-
- CCAAT enhancer binding protein
-
- GSDMD
-
- gasdermin D
-
- HIPEC
-
- hyperthermic intraperitoneal chemotherapy
-
- HSF
-
- heat shock factor
-
- HSP
-
- heat shock proteins
-
- IFN
-
- interferon
-
- IL
-
- interleukin
-
- LPS
-
- lipopolysaccharide
-
- MERS-CoV
-
- Middle East Respiratory Syndrome-CoV
-
- NF-κB
-
- nuclear factor-kappa B
-
- NLRP3
-
- NOD-leucine rich repeat and pyrin containing protein 3
-
- ORF
-
- open reading frame
-
- SARS-CoV-2
-
- Severe acute respiratory syndrome coronavirus 2
-
- TLR
-
- toll-like receptor
-
- TNF
-
- tumour necrosis factor
-
- TRAF6
-
- tumour necrosis factor receptor-associated factor 6
1 SARS-COV-2 AETIOLOGY
At the end of 2019, a new coronavirus family member named severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) emerged and is responsible for coronavirus disease, COVID-19. At the outset, the virus seemed to be less dangerous than previous epidemics of coronavirus, such as SARS-CoV in 2002 or Middle East Respiratory Syndrome-CoV (MERS-CoV) in 2014 (Coperchini, Chiovato, Croce, Magri, & Rotondi, 2020). However, in contrast to previous coronaviruses, SARS-CoV-2 is highly contagious, although harmless for 85% of the population, COVID-19 can be life-threatening in vulnerable (e.g. older, immunosuppressed, or comorbid) patients. It has already led to a high number of human deaths worldwide and represents a major global public health concern. Observations made by clinicians, researchers and epidemiologists have broadened our understanding of the different stages of SARS-CoV-2 infection, onset of clinical symptoms, and, sometimes, lethality. The infection phase seems to be dependent on the angiotensin-converting enzyme 2 (ACE2), notably expressed on airway epithelial cells and endothelial cells (Li, Liu, Yu, Tang, & Tang, 2020). Then, an asymptomatic or low-symptomatic phase begins (fever), followed by a mildly symptomatic phase (fever, shortness of breath and abnormal chest imaging) and, finally, a severe phase (acute respiratory distress syndrome [ARDS] and hyperinflammation), which can be potentially lethal for the patient (Nile et al., 2020). Further, SARS-CoV-2 can also damage tissues other than the lung and that this entails hyperinflammation in the form of a “cytokine storm”, which give new perspectives for the management and treatment of patients infected by this virus.
2 CYTOKINE STORM
The cytokine storm is a consequence of hyperactivation of the immune system, leading to increased release of various pro-inflammatory cytokines/chemokines. Accordingly, a large amount of cytokines have been measured in patients' blood. Among these cytokines, the most commonly described are interleukin (IL)-1β, IL-6, tumour necrosis factor (TNF)α, interferon (IFN)γ or IL-10 (Ciavarella, Motta, Valente, & Pasquinelli, 2020; Coperchini, Chiovato, Croce, Magri, & Rotondi, 2020; Jose & Manuel, 2020; Kuppalli & Rasmussen, 2020; Nile et al., 2020; Pedersen & Ho, 2020; Rothan & Byrareddy, 2020; Sun et al., 2020). The cytokine storm sustains an aberrant inflammatory response in the blood, driving the immune system to attack the body involving several organs such as lung. This, in turn, causes alveolo–capillary membrane injury, lung permeability, acute respiratory distress syndrome and multiple organ failure and, in the most severe cases, death.
There is a clear need to understand this cytokine storm and to find new treatments to mitigate it, in order to reduce clinical symptoms and prevent death. In this context, some clinical trials are in process, for example, testing the use of the anti-IL-6 tocilizumab, the anti-IL-1β, anakinra, or TNF blockers (Ye, Wang, & Mao, 2020). However, as mentioned above, the cytokine storm involves numerous cytokines and the majority of the pro-inflammatory ones must be neutralized in order to obtain the most efficient effect. Thus, clinicians need a general treatment targeting many cytokines at the same time.
3 HEAT SHOCK FACTORS/HEAT SHOCK PROTEINS
One potential strategy to target the cytokine storm could be hyperthermia. Hyperthermia or heat shock consists in elevating the body temperature to a range of 40–42°C (104–108°F) for 1 or 2 h. In the cells, it enables induction of heat shock factor (HSF)-1/2, transcription factors that lead to the expression of different genes, notably that of the heat shock proteins (HSPs). Among them, the HSP70 gene family (HSP1A1, HSPA1B, HSPA1L, HSPA4, HSPA6, HSPA8 and HSPA9) is consistently the most markedly increased after heat shock in human cells (Kovacs et al., 2019; Mahat, Salamanca, Duarte, Danko, & Lis, 2016; Richter, Haslbeck, & Buchner, 2010). Therefore, we focus our interest here on the possible impact of HSF-1/2 and HSP70 on cytokines involved in the cytokine storm.
4 POTENTIAL EFFECTS OF HEAT SHOCK FACTORS (HSFs) AND HEAT SHOCK PROTEIN FAMILY A (HSP70) ON THE CYTOKINE STORM
The production of cytokines can be orchestrated by transcriptional pathways and by a protease activation programme (Figure 1).
4.1 Effects on cytokine transcription
The transcription pathway leading to cytokine synthesis mostly relies on nuclear factor-kappa B (NF-κB), which can be activated by the engagement of ligands on their receptors localized at the plasma membrane, such as toll-like receptors (TLRs), tumour necrosis factor (TNF) receptors or IL-1R. HSFs and HSP70 have been reported to have the capacity to inhibit the production of some cytokines by acting at different levels in this pathway.
HSP70 increases the capacity of HSF-1 to bind to the TNFA promoter, which reduces lipopolysaccharide (LPS)-mediated production of TNFα by monocytes (Ferat-Osorio et al., 2014). On the contrary, Hsf1-deficient mice exposed to LPS or bacterial infection have increased production of TNFα, IFNγ, IL-1β, IL-6 and IL-10, as compared with wild type (WT) mice (Barber et al., 2014; Murapa, Ward, Gandhapudi, Woodward, & D'Orazio, 2011). In ulcerative colitis patients, the blood level of HSF-2 is correlated with IL-1β and TNFα levels, two main actors in the severity of the disease. Moreover, when HSF-2 expression is downregulated in colon cancer cells, the LPS-induced production of IL-1β and TNFα is enhanced (Miao et al., 2014).
Heat shock is able to decrease TNFα and IL-6 production in murine myeloid cells in a HSF-1-dependent but HSP70-independent manner (Mortaz et al., 2006). Heat shock and overexpression of HSP70 in human macrophages impair LPS and bacterial production of TNFα, IL-1, IL-10 and IL-12, without affecting IL-6 (Ding, Fernandez-Prada, Bhattacharjee, & Hoover, 2001). Consistent results were observed in murine macrophages, where HSP70 overexpression was shown to decrease LPS-induced production of TNFα, IL-1β and IL-6 (Muralidharan et al., 2014). These studies did not explore the mechanism involved. However, other studies implicated an effect of HSP70 on NF-κB. In rats infected with adenovirus allowing overexpression of HSP70 or exposed to hyperthermia or in murine macrophages submitted to a heat shock to overexpress HSP70, the inhibitor of NF-κB, IκB is not degraded, preventing the nuclear re-localization of NF-κB and, in turn, inhibiting LPS-induced TNFα, IL-6 or IL-1β expression (Chen, Kuo, Wang, Lu, & Yang, 2005; Dokladny, Lobb, Wharton, Ma, & Moseley, 2010; Shi et al., 2006). Another mechanism proposed involves the capacity of HSP70 to inhibit LPS-induced NF-κB activation by interacting directly with tumour necrosis factor receptor-associated factor 6 (TRAF6), a member of the TLRs and interleukin 1 receptor, type I transduction pathways (Chen et al., 2006). In a viral-infection context, heat shock was shown to be able to dampen IL-6, TNFα, IFNβ and IFNγ production in H5N1-infected mice (Xue et al., 2016).
HSF-1 is also able to interact with CCAAT enhancer binding protein (C/EBPβ) to repress the transcription of IL-1B in human monocytes treated with LPS (Xie, Chen, Stevenson, Auron, & Calderwood, 2002). Similarly, HSP70 is also able to decrease C/EBPβ and C/EBPδ levels in murine dendritic cells, leading to decreased production of TNFα, IFNγ and MCP-1 (CCL2) (Borges et al., 2013).
However, in CD4+ and CD8+ T-cells, HSP70 can have ambiguous effects. In CD8+ T-cells, HSP70 is able to decrease the expression of TLR4 (one of the receptors involved in NF-κB activation), leading to increased IL-10 and decreased IFNγ production (Ghosh, Sinha, Mukherjee, Biswas, & Biswas, 2015). Exposure of CD4+ Treg cells to HSP70 decreases the production of IFNγ and TNFα but increases the production of IL-10 and TGFβ (Wachstein et al., 2012).
4.2 Effects on protease-mediated cytokine production
All the cytokines described above are synthesized in an active form, except for IL-1β. Actually, the IL-1β gene (IL1B) is translated into an inactive form, pro-IL-1β. Pro-IL-1β needs to be cleaved to become active. The main components responsible for this maturation are multiprotein complexes, named inflammasomes. In response to several types of stimulation, they enable the activation of caspase-1, which, in turn, cleaves pro-IL-1β into IL-1β (Chevriaux et al., 2020). Inflammasomes may also be activated by viruses (Hayward, Mathur, Ngo, & Man, 2018). Thus, inhibiting inflammasomes would inhibit IL-1β production. The effects of HSPs on inflammasomes were recently reviewed (Martine & Rebe, 2019). We have shown that the absence of HSP70 in mice led to increased production of IL-1β in vivo and in vitro, while a heat shock or overexpression of HSP70 inhibited IL-1β production. This effect of HSP70 is due to its interaction with NOD-leucine rich repeat and pyrin containing protein 3 (NLRP3), a component of the inflammasome (Martine et al., 2019). In myeloid cells, HSF-1 is also able to activate β-catenin, a repressor of NLRP3 transcription, thereby leading to the inhibition of caspase-1 activation and IL-1β maturation (Yue et al., 2016).
Another possibility is that pro-IL-1β is cleaved extracellularly by proteases such as neutrophil elastase (Afonina, Muller, Martin, & Beyaert, 2015). A possible effect of HSP70 at such a level is conceivable, because HSP70 was shown to reduce neutrophil elastase-mediated cell injury in bronchial epithelial cells (Ito et al., 2006).
Finally, the effect of heat shock and HSP70 on the NLRP3 inflammasome can possibly explain their effect on the production of other cytokines. Actually, inflammasome activation by bacteria triggers caspase-1 activation (and sometime caspase-11/4/5 activation) and cell death by pyroptosis. This type of cell death is enabled by the cleavage of gasdermin D (GSDMD) by inflammatory caspases. Cleaved GSDMD oligomerizes into ring-shaped structures to form membrane pores (Evavold et al., 2018). These pores enable the exit of small molecules such as IL-1β (but not pro-IL-1β, which is too large). Other molecules, such as IL-6, monomeric TNFα, IFNγ or IL-10, which all are smaller than 20 kDa, could also exit. This was suggested by a recent study showing that Nlrp3-deficient mice infected by Streptococcus suis produced less IL-1β, TNFα and IFNγ (Lin et al., 2019).
5 THERAPEUTIC PERSPECTIVES
The level of expression of HSPs and HSFs may thus be important for patients suffering from SARS-CoV-2-associated cytokine storm. In humans, several single nucleotide polymorphisms exist for HSF-1 (Bridges et al., 2015) and HSP70 (Boiocchi et al., 2014; Klausz et al., 2005; Pablos et al., 1995), with possible effects on their expression and/or function. Some of these single nucleotide polymorphisms were shown to be associated with inflammatory diseases, such as multiple sclerosis, Crohn's disease, or systemic lupus erythematosus (Boiocchi et al., 2014; Klausz et al., 2005; Pablos et al., 1995). A link between these single nucleotide polymorphisms and the intensity of the cytokine storm in SARS-CoV-2 patients would be an interesting avenue for further investigation.
Hyperthermia as a means to induce HSFs/HSPs in the context of SARS-CoV-2 must be tested. We showed that whole body hyperthermia can be used on mice to inhibit IL-1β effects (Pilot et al., 2020). In humans, therapeutic hyperthermia has previously been described and can be applied locally, regionally, or on the whole body (Habash, Krewski, Bansal, & Alhafid, 2011). Whole-body hyperthermia seems difficult to consider. However, in cancer treatment, several protocols were tested locally, such as hyperthermic intraperitoneal chemotherapy (Gelli et al., 2018; Lavoue et al., 2019), or using deep regional apparatus (Franckena et al., 2009; Overgaard et al., 1996; Pennacchioli, Fiore, & Gronchi, 2009). Recently, a clinical study showed that warming blankets (38°C) can be used to decrease inflammatory cytokines in patients after surgical treatment of hip displacement (He, Liu, Wen, & Wu, 2020; Petta et al., 2016). In inflammatory or infectious pathologies, hyperthermia was shown to have promising effects in sepsis-induced acute lung injury in rats (Villar et al., 1994). Glutamine or the HSP90 inhibitor 17-demethoxygeldanamycin or geranylgeranylacetone, used to induce HSP70 expression, reduced lung injury and diaphragm impairment after sepsis in rats (Masuda, Sumita, Fujimura, & Namiki, 2003; Singleton, Serkova, Beckey, & Wischmeyer, 2005; Wang et al., 2016). This is in accordance with a study that described that aged-mice male deficient in HSP70 are more sensitive to sepsis than WT animals (McConnell et al., 2011). In a septic shock model, a febrile response resulted in better respiratory function, lower blood lactate concentration, and prolonged survival time of sheep (Su et al., 2005). In humans, several studies showed that fever can have beneficial effects while antipyretic have harmful effects on patients' outcome after virus infection or sepsis. However, hyperthermia should be tightly controlled (high temperature should not be maintained for a long duration) as beneficial effects can be lost and increase disease severity (Hasday et al., 2011). Further experiments are required to evaluate the impact of hyperthermia on the cytokine storm and recovery in mice, before considering such therapy in humans. To do this, several models can be used, such as poly (I:C), SARS-CoV-2 spike protein mice infection or transfection of cells with open reading frame (ORF) SARS-CoV-2 accessory proteins (Shi, Nabar, Huang, & Kehrl, 2019; Siu et al., 2019). Because HSP70 is released by the cells (free or within exosomes), measuring circulating levels of HSP70 may help to define the appropriate hyperthermia conditions for patients (Chalmin et al., 2010; Gobbo et al., 2016).
Among proposed medication, corticosteroids present a real benefits for patients (Group et al., 2020). HSP70 is implicated in the folding of glucocorticoid receptors to allow fixation of glucocorticoids (Petta et al., 2016). One can speculate that an increased expression of HSP70 will improve glucocorticoids-glucocorticoid receptor signalling. Because activation of glucocorticoid receptors blocks cytokine production through inhibiting NF-κB, like HSP70/HSF1, the association of hyperthermia with glucocorticoid may have no supplementary effect (Petta et al., 2016). However, glucocorticoids were shown to have ambivalent effects on NLRP3 inflammasome (Busillo, Azzam, & Cidlowski, 2011; Yang et al., 2020). Thus, further studies are required to understand the crosstalk between hyperthermia/HSP70 and glucocorticoids-glucocorticoid receptor, especially in SARS-CoV-2 infection.
6 CONCLUSIONS
In viral infections, HSPs are chaperones known to play a role at different levels. By interacting with the virus, they can regulate viral infections, including cell entry and nuclear import, viral replication and gene expression, folding/assembly of viral protein, apoptosis regulation and host immunity. They are also effective carrier molecules for cross-presentation by antigen presenting cells (Bolhassani & Agi, 2019). Thus, HSPs and particularly HSP70 can enhance the development of innate and adaptive immune responses against infecting agents (Oglesbee, Pratt, & Carsillo, 2002). In agreement with a recent publication (Heck, Ludwig, Frizzo, Rasia-Filho, & Homem de Bittencourt, 2020), we propose here a possible additional role for HSFs/HSP70 on the production of inflammatory cytokines. However, their role in SARS-CoV-2 infection and host response remains to be clarified.
6.1 Nomenclature of targets and ligands
Key protein targets and ligands in this article are hyperlinked to corresponding entries in the IUPHAR/BPS Guide to PHARMACOLOGY http://www.guidetopharmacology.org, and are permanently archived in the Concise Guide to PHARMACOLOGY (Harding et al., 2018), 2019/20 (Alexander et al., 2019).
ACKNOWLEDGEMENT
We thank Fiona Ecarnot, PhD (EA3920, University of Franche-Comté, Besancon, France) for careful reading of the manuscript.
CONFLICT OF INTEREST
The authors declare no conflict of interest.