Coronaviruses and Integrin αvβ3: Does Thyroid Hormone Modify the Relationship?

ABSTRACT

Background

Uptake of coronaviruses by target cells involves binding of the virus by cell ectoenzymes. For the etiologic agent of COVID-19 (SARS-CoV-2), a receptor has been identified as angiotensin-converting enzyme-2 (ACE2). Recently it has been suggested that plasma membrane integrins may be involved in the internalization and replication of clinically important coronaviruses. For example, integrin αvβ3 is involved in the cell uptake of a model porcine enteric α-coronavirus that causes human epidemics. ACE2 modulates the intracellular signaling generated by integrins.

Objective

We propose that the cellular internalization of αvβ3 applies to uptake of coronaviruses bound to the integrin, and we evaluate the possibility that clinical host T4 may contribute to target cell uptake of coronavirus and to the consequence of cell uptake of the virus.

Discussion and Conclusions

The viral binding domain of the integrin is near the Arg-Gly-Asp (RGD) peptide-binding site and RGD molecules can affect virus binding. In this same locale on integrin αvβ3 is the receptor for thyroid hormone analogues, particularly, L-thyroxine (T4). By binding to the integrin, T4 has been shown to modulate the affinity of the integrin for other proteins, to control internalization of αvβ3 and to regulate the expression of a panel of cytokine genes, some of which are components of the ‘cytokine storm’ of viral infections. If T4 does influence coronavirus uptake by target cells, other thyroid hormone analogues, such as deaminated T4 and deaminated 3,5,3ʹ-triiodo-L-thyronine (T3), are candidate agents to block the virus-relevant actions of T4 at integrin αvβ3 and possibly restrict virus uptake.

KEYWORDS:

• ACE2
• Arg-Gly-Asp (RGD) peptide
• clotting risk
• L-thyroxine (T4)
• porcine epidemic diarrhea α-coronavirus (PEDV)
• SARS-CoV-2
• serum reverse T3 (3,3ʹ5ʹ-triiodo-L-thyronine, rT3)
• nonthyroidal illness syndrome (NTIS)

Introduction

The 2019 coronavirus disease (COVID-19) pandemic that is current has fostered worldwide interest in the pathophysiology of and possible therapeutic interventions in the disease.^1–5 Experimental treatment targets include inhibition of cellular uptake of coronavirus, disruption of replication of the virus once it has gained access to the cell interior, blockade of the cytokine storm that may accompany systemic viral infection and activation of the host immune system to affect multiple facets of viral attack.

Uptake by host cells of β-coronavirus SARS-CoV-2 – the cause of COVID-19 – is thought primarily to involve an ectoenzyme, angiotensin-converting enzyme 2 (ACE2).^6,7 The host cellular uptake and the replication of another epidemiologically important coronavirus have recently been shown to require integrin αvβ3.⁸ This is the porcine epidemic diarrhea α-coronavirus (PEDV) that infects animals and human subjects.⁹ Sigrist and coworkers have also implicated integrins in host cell uptake of SARS-CoV-2.¹⁰ It may be noted that ACE2 via its Arg-Gly-Asp (RGD) sequence binds to integrins¹¹; thus, a distinction between integrins and ACE2 in terms of uptake of coronaviruses such as SARS-CoV-2 and intracellular signaling induced by viruses may be increasingly difficult to establish. If integrin αvβ3 is involved primarily or secondarily via ACE2 in SARS-CoV-2 uptake (as well as PEDV), then pharmacologic options exist to prevent internalization of the integrin and bound virus, as discussed below.

Integrins are heterodimeric structural proteins of the plasma membrane that are involved in cell-cell and cell-extracellular matrix protein interactions.^12,13 Of particular relevance to viruses, however, is the regular internalization of integrins such as αvβ3, a factor that underlies viral replication and has already been shown to be relevant to coronavirus PEDV uptake.⁸ Control of internalization of integrin αvβ3 is pharmacologically feasible, as discussed below.

Certain small molecules, such as thyroid hormone analogues, have specific receptors on integrin αvβ3 that regulate cellular uptake of the integrin.¹⁴ The thyroid hormone receptor also modulates expression of the genes for the monomeric proteins that make up αvβ3,^14,15 and coronavirus PEDV has also been shown to control expression of the integrin β3 monomer. L-thyroxine (T4) regulates from αvβ3 the activity of specific signal-transducing enzymes,^14,15 and coronavirus binding by the integrin similarly affects mitogen-activated protein kinase (MAPK) and phosphatidylinositol-3-kinase (PI3K/Akt) activities.¹⁶ T4 also regulates expression of genes coding for components of ‘cytokine storm’ that often are a component of systemic viral infections.^17,18 It is also of interest that ACE2 has been shown to interact with integrins to regulate downstream signal transduction.¹¹

What we review next are the features of a well-studied, clinically relevant α-coronavirus (PEDV) binding site on integrin αvβ3 and certain actions of thyroid hormone analogues on the integrin that are relevant to clinical behavior of the virus. While SARS-CoV-2, the cause of COVID-19, is a β-coronavirus reported to have its target binding site on ACE2, we note that ACE2 has important interactions with integrins and that an important case has very recently been made for integrins in terms of SARS-CoV-2 entry into host cells.¹⁰

A Receptor for Coronaviruses on Integrin αvβ3

Li and coworkers in 2019 reported in two types of epithelial cells that coronaviruses target the overexpression or suppression of plasma membrane αvβ3 and, respectively, facilitated or inhibited virus uptake.⁸ The model coronavirus studied was PEDV. Further, RGD peptides described above that are inhibitors of actions of αvβ3 blocked replication of the virus when tested in one of the cell lines. Earlier, the same group had shown that among the genes whose transcription was enhanced in epithelial cells infected with coronavirus were certain integrin monomers, such as β3¹⁶ and a panel of signal-transducing kinases that we have shown to be controlled from the extracellular head of integrin αvβ3.¹⁴

Integrin αvβ3 Binding Domains for Hormones and Arg-Gly-Asp (RGD) Peptides

The majority of the integrin heterodimeric protein is extracellular and is involved in intercellular binding and in binding to extracellular matrix proteins, such as vitronectin, fibronectin, and osteopontin.^12,13 Integrin αvβ3 has recently been shown to contain small-molecule receptors for nonpeptide hormones such as thyroid hormone,^19,20 testosterone,^21,22 and the stilbene, resveratrol.^22,23 Integrin αvβ3 also has, proximal to the hormone receptor domain, a binding site for RGD peptides and for proteins that contain the RGD sequence.²⁴

Cellular Uptake/Internalization of Integrin αvβ3

We have shown that the cellular internalization of αvβ3 is driven by the binding of T4 to the integrin.²⁵ Thus, the binding of the virus by the integrin in the presence of T4 may support cellular virus uptake. The internalized integrin breaks into the component αv and β3 monomers. The β3 monomer remains in cytoplasm, but αv targets the cell nucleus where it functions as a co-activator protein, affecting expression of a heterogeneous panel of genes, including that for cyclooxygenase-2 (COX-2). Of interest is that the coronaviral nucleocapsid activates COX-2 gene transcription as part of the virus-induced pro-inflammatory process.²⁶

Pharmacologic Manipulation of the Extracellular Domains and Functions of Integrin αvβ3

The head of the extracellular domain of heterodimeric αvβ3 is composed of contributions from both the αv and β3 monomers, and the receptor for thyroid hormone analogues is at the junction of the two monomers.^24,27 Noted above is that receptors for androgen and stilbenes exist in the same domain as receptors for thyroid hormone analogues and RGD peptides. Each of these sites offers opportunities for specifically modulating activities of the integrin. For example, the thyroid hormone analogue, tetraiodothyroacetic acid (tetrac), blocks T4 actions at the receptor, including the internalization of αvβ3.

Cytokine Storm

Another feature of the thyroid hormone-integrin αvβ3-coronavirus uptake we have discussed here is that thyroid hormone analogues, via αvβ3, also generate transcription of a number of cytokines and chemokines^18,28 consistent with the ‘cytokine storm’ of systemic viral infections. The abundance of specific interleukins and of CC and CXC chemokines in target cells is affected by PEDV²⁹ and a similar pattern of pro-inflammatory factors results from the action of T4 via αvβ3.¹⁸

Is the Nonthyroidal Illness Syndrome (NTIS) a Risk Factor for Cellular Coronavirus Uptake?

The nonthyroidal illness syndrome (NTIS) is a set of serum thyroid function test results that is often present in systemic illness that does not primarily involve the pituitary-thyroid axis.³⁰ The clinical laboratory findings may include elevated circulating free thyroxine (FT4), reduced total serum 3,5,3-triiodo-L-thyronine (T3), and normal thyrotropin (TSH). Serum reverse T3 (3,3,5-triiodo-L-thyronine, rT3) may also be elevated in the syndrome. NTIS is not seen to be an indication for therapeutic intervention, but elevations of FT4 and rT3 may be factors that promote tumor cell proliferation.^14,30,31 We have raised the possibility that clinical cancers complicated by NTIS and that are not responding to conventional therapies may reflect growth-promoting actions of T4 and rT3.³⁰

As noted above, coronaviruses gain access to target cells by binding to integrin αvβ3. The interaction of the virus is with both the αv and β3 monomer components of the extracellular head of the integrin.⁸ This location may place the virus proximal to the thyroid hormone receptor site, the possible significance of which is discussed below. We would emphasize first, however, that T4 is potentially stage-setting for virus uptake because the hormone (1) increases expression of the genes for the specific integrin monomers³² and (2) facilitates internalization (cellular uptake) of the integrin.²⁵ Thus, the number of binding sites for the virus on the target cell surface and the internalization of the virus may be in part subject to regulation by changes in circulating thyroid hormones – particularly, free T4 – that can be part of the NTIS. We suggest that when NTIS complicates cancer or lung or heart disease, it may thus be a risk factor for viral infection of the target epithelial cells of the airway. Were NTIS shown to be a risk factor for coronavirus infection, it is possible pharmacologically to arrest host thyroid gland production of T4 and eliminate the FT4 contribution to NTIS. In this setting, euthyroidism may be maintained with exogenous T3, as we have proposed when NTIS complicates cancer.³⁰ At physiologic concentrations, T3 does not affect functions of αvβ3.^14,20

Discussion and Conclusions

The model cited here of a well-studied coronavirus, PEDV, and its epithelial cell uptake via binding to integrin αvβ3, is proposed as a model for the current pandemic of COVID-19. PEDV is a porcine virus that has been examined as a typical α-coronavirus in preclinical studies of pathophysiology of infections by this family of viruses.⁸ SARS-CoV-2 is a β-coronavirus that has been shown to have ACE2 as a receptor protein that facilitates cell uptake of virus.⁶ But α-coronaviruses may also interact with ACE2⁶ and Sigrist and coworkers have made the case for integrin participation in SARS-CoV-2 viral uptake by integrins participating in SARS-CoV-2 internalization by host cells.¹⁰ However, evidence we have reviewed above indicates that ACE2 interacts with integrins. We would also point out that integrin αvβ3 appears to be involved in host cell uptake of other epidemiologically important infectious agents.³³ A critical feature of viral uptake involving αvβ3 is that the process appears to be driven by physiological concentrations of thyroid hormone as T4.

In contrast, Luan has suggested that integrins inhibit host cell internalization of SARS-CoV-2.³⁴ These authors identified an RGD motif that is integrin-binding in the spike (S) glycoprotein of SARS-CoV-2; they also found a Lys-Gly-Asp (KGD) integrin-binding motif in ACE2. They concluded that interactions of ACE2 and virus with motifs of the integrin would ‘shield’ the virus and receptor from forming a virus-receptor complex that is subject to uptake. What we would note – and what is consistent with PEDV-integrin and other integrin-coronavirus studies – is that in the presence of physiological concentrations of T4, αvβ3 will be internalized by host cells and both the virus and ACE2 will reach the cytoplasm of the target cell. This may explain the reports of ACE2 effects on integrin signaling cited above. Another critical feature of actions of thyroid hormone analogues on integrin αvβ3 is that the hormone analogues control the conformation of the external domain of the integrin and thus may regulate the accessibility of the binding sites on the integrin for RGD motif-containing proteins.³⁵ Thus, we see the contention of Luan to be of interest, but not relevant in the context of thyroid hormone actions on αvβ3 that may be relevant to cell uptake of viruses.

The rationale for examining in the present paper the role of thyroid hormone in the uptake of integrin αvβ3 is that the actions of thyroid hormone as T4 can be substantially and safely reduced. One intervention is the induction of euthyroid hypothyroxinemia, as described by Hercbergs and coworkers in patients with advanced cancers.³⁶ The actions of T4 at the integrin have been inhibited in preclinical studies by tetrac^14,20 or chemically modified tetrac.^28,37 Chemically modified forms of tetrac have been developed as anticancer agents; they require testing as antiviral agents where the viruses gain access to target cells by binding to cell surface αvβ3 that may be internalized under the direction of T4. If T4 significantly increases target cell virus uptake, a group of patients at particular risk are those with thyroid cancer on high-dose T4 to suppress endogenous TSH.

We also raise the possibility here that elevation of circulating free T4 occurring as a component of the NTIS that complicates a variety of systemic diseases, including infections, may enhance the cell surface abundance and uptake of αvβ3.^25,32 If this indeed is the case and the settings may include virus-bearing αvβ3, then we have a possible explanation for the increased frequency of COVID-19 in patients with serious systemic diseases.

It is also important to note that thromboembolic disease may critically complicate coronavirus infection.^38,39 T4 is known to activate human platelets⁴⁰ and we have postulated that elevated circulating levels of T4 may promote pathological clotting. In the context of systemic infection, NTIS with elevated free T4 may contribute to the increased clotting risk encountered in COVID-19.

Conflicts of Interest

Co-authors Davis and Mousa have developed tetrac derivatives for management of cancers. All other authors have no conflicts to declare.