A comprehensive review of the interaction between COVID-19 spike proteins with mammalian small and major heat shock proteins

Spike glycoproteins

Spike glycoproteins/spike proteins (SARS-CoV-2 spike proteins), as one of the SARS-CoV-2 proteins, potentially serve as hotspots for antiviral drug development based on their structure. SARS-CoV-2 spike proteins are transmembrane glycoproteins that exist as trimeric proteins that protrude from the virus’s surface. The SARS-CoV-2 spike proteins mediate SARS-CoV-2 entry into the host cells and subsequent pathogenesis; hence, it is a potential antiviral target [3]. SARS-CoV-2 spike proteins consist of two functional subunits, S1 and S2 (Figure 2). The S1 subunit contains a receptor-binding domain (RBD) and an N-terminal domain. The pivotal role of the S1 subunit in COVID-19 viral entry and subsequent pathogenesis is to identify and bind to the host cell receptors (viral attachment) [3,5]. The S2 subunits comprise fusion peptides, including heptad repeat 1 (HR1) and heptad repeat 2 (HR2) hence the trimeric protein. The role of the S2 subunit is to facilitate the in-between fusion of the host cell membrane and the viral-enveloped membrane [5].

Figure 2

Trimeric structure of the SARS-CoV-2 spike protein (created by Biorender.com).

Spike proteins are essential for viral entry into the host cell. According to a study by Hu et al. [6], coronavirus exhibits spike proteins (SARS-CoV-2 spike protein) that cover the virus’s surface. The spike proteins mediate viral entrance by attaching to angiotensin-converting enzyme 2 (ACE2), which acts as the host’s cell receptor [6]. Spike proteins exist in a metastable prefusion conformation. When SARS-CoV-2 interacts with a host cell, external structural rearrangement of the spike proteins happens, which permits the virus to combine with the host’s cell membrane [7].

According to recent studies, SARS-CoV-2 utilizes trimetric spike proteins to attach to the host’s ACE2 (receptor) and combine with the cell membrane to gain cell entry [8]. The fusion of spike proteins and the host cell membrane process and gaining cell entrance is a multi-step process involving three different and separate spike protein cleavage events that prime the SARS-CoV-2 for interaction with ACE2, membrane fusion, and cell entry [8]. Huang et al. [7] conducted a study on SARS-CoV-2 spike protein proteins’ structural and functional capabilities in developing potential antiviral drugs for COVID-19. The research study highlighted the current research advances in the arrangement, utility, and progress of antivirus drugs aiming at the coronavirus spike proteins [7].

No paucity of research investigates the establishment of binding sites of the SARS-CoV-2 spike protein that mediate the viral entry with interactions with human HSPA8 and HSP27 or heat shock proteins (HSPs) for that matter toward the synthesis of COVID-19 treatment that will target those binding sites and the spike proteins and prevent viral entry in the first place. With SARS-CoV-2 virus still posing imminent threats due to lack of treatment, drug development needs more attention, specifically working on drugs that treat the virus [9]. Therefore, it seemed potent to carry out a research study that primarily investigates the establishment of the binding sites of the spike proteins from SARS-CoV-2 through the interaction with human HSPA8 and HSP27 through sequence analysis and alignment, homology modeling, and protein–protein docking toward the synthesis of COVID-19 treatment.

COVID-19 virus, like other viruses and infectious agents, calls into action the chaperonin system, which under normal circumstances defends the host organism being infected. Still, however, in the event of infection, they may assist the attacker. This scenario is called chaperonopathy by mistake, in which the infectious agent (COVID-19 virus) commands the chaperoning systems of the host organism, redirecting the usual activities to favor infection [10].

Different viruses exploit different strategies for viral attachment and endocytosis, viral penetration, and uncoating, viral assembly and budding, and lastly, viral transcription and replication within the host [3]. Despite all the differences, there is one emerging common principle that is viruses co-opt the host’s molecular chaperons to proselytize distinct entry steps. In the case of the SARS-CoV-2 virus, the role of chaperones and the mechanism by which the virus hijacks the host’s molecular chaperones are still poorly known [1]. From recent studies, SAR-CoV and SARS-CoV-2 have been found to enter the host cells using the same receptor which is ACE2 via the RBD located on spike proteins; hence, it is presumed that the two viruses might share a similar mechanism in hijacking the host’s chaperoning system [1,3].

The binding of the spike proteins on the receptor initiates endocytosis of the viral particles, targeting the endosome and resulting in the subsequent release of the nucleocapsid into the cytosol [11]. Within the cytosol, the endoplasmic reticulum (ER) then assumes a vital role during the replication cycle of the virus, with the virus expropriating the activities of the ER-associated chaperones that are generated via viral-ER-induced stress [1,11]. In fact, there have been speculations that ER molecular chaperones facilitate hCoVs’ entry into host cells. For example, both bCoV-HKU9 beta coronavirus and MERS-CoV’s spike proteins are known to engross the chaperone BiP, which normally mediates protein degradation and folding pathways in the ER, but in this case, it assists in cellular uptake of the viruses [1,11]. The viral RNA is translated and transcribed via the interaction of ER molecular chaperones (for example, calnexin) ensuring correct and proper folding and then the nucleocapsid proteins and the genomic RNA combine forming new viral particles [1].

From recent studies, it is presumed that the viral load is impacted by the level of expression of chaperones. Due to viral replication, there is an increase in the accumulation of unfolded, nascent viral polypeptide in the ER exceeding its folding capacity which is the result of the ER stress [1,11]. This ER stress activates cell-signaling pathways that regulate gene expression at both translational and transcriptional levels, which raises the level of chaperones. If gene-encoding chaperones are activated and the level of chaperone expression in ER increases, the maturation, folding, and degradation of proteins resumes and this includes the viral ones [1]. This in turn results in the restoration of homeostasis, thereby assuring viral survival.

Small HSPs

sHSPs are a unique family of molecular chaperones that are devoid of the ATPase domain. These are classified as tiny proteins, ranging in size from 10 to 40 kDa, with a-crystallin domain at their core and varying N- and C-terminal domains on each side. The crystalline domain, which is present in sHSPs, has subunit molecular weights of 10–40 kDa. Both HSP27, alpha-crystalline and beta-crystallin, are sHSP family members that form sizable oligomeric species. The focus of this review is mostly on mammalian sHSPs. According to sequence homology, the human genome contains ten sHSPs (designated HSPB1 to HSPB10). For ease of identification, HSPB1 was referred to as HSP27, HSPB4 as alpha-crystallin, HSPB5 as beta-crystallin, HSPB6 as HSP20, and HSPB8 which is their previous nomenclature [17]. The sHSPs have been divided into Classes I or II [18]. Class I sHSPs include HSP27, beta-crystallin, HSP20, and HSP22. These are extensively distributed, present in a variety of tissues (such as eye lenses, muscles, and heart), and predominately heat-inducible (HSP22 heat-inducibility is cell type dependent, for example, in the heart the functionality of HSP22 is dependent on BCL-2-associated athanogene 3 [BAG3]) [19] and crucial for cell survival in high-stress situations [20]. The Class II sHSPs consist of a tissue-specific pattern of expression (for example, alpha-crystallin is expressed in genomic stability, chaperone activity, and refractive index of the eye lens; the chaperons HSPB3, HSPB7, and HSPB2 are expressed in the maintenance of myofibrillar integrity) [17,19]. Members of the class II sHSPs include beta-crystallin, HSPB2, HSPB7, HSPB3, HSPB9, and HSPB10. It is thought that Class II sHSPs have a significant impact on development, differentiation, and specialized tissue-specific functions [18].

In addition to their molecular chaperone-like function in the prevention of peptide and protein aggregation, sHSPs like B-crystallin and HSP27 are involved in a variety of cellular processes including cytoskeletal integrity maintenance, protein folding, degradation of proteins, stress tolerance, cell death, differentiation, cell cycle, signal transduction, and growth [21]. Proteins belonging to the HSP family have interactions with multiple substrates and display cardio and neuroprotection, pro-angiogenic property, powerful anti-apoptotic activity, and anti-inflammatory function. Uncertainty surrounds the underlying molecular mechanism driving the promiscuous interactions and pleiotropic activities of sHSPs. These functions of sHSPs have significant effects on both general health and illness situations. Therapeutic targets have been suggested to include HSP27 and alpha- beta-crystallins [22].

HSP family 70

HSP 70 (HSP70) family is the most relevant HSP of all the HSP families comprising stress-inducible HSP70 proteins (for example, 72-kDa protein also known as HSP72), nuclear and cytosolic constitutive HSP70 proteins (for example, HSC70/HSP73/HSPA8), mitochondrial mtHSP70 (for example, Grp75), and the endoplasmic BiP (for example, GRP78) [16]. HSP70s are found in all surviving domains of life. Structurally and at a molecular level, HSP70s contain three functional domains, a nucleotide-binding domain (NBD) connected to a substrate-binding domain (SBD) by a flexible and long hydrophobic linker, and a C-terminal peptide-binding domain [32,33]. Additionally, cytosolic eukaryotic HSP70s consist of a particular Glu-Glu-Val-Asp motif at the C-terminal extremes that is vital for interaction with the co-chaperones for HSP that regulates the HSP70 ATPase activity and its capabilities of binding to a substrate [34,35].

HSP70 molecular chaperones are involved in a variety of processes that include maintaining cell homeostasis, folding newly synthesized proteins, subcellular transport of proteins, formation and dissociation of complexes, stabilization of protein substrates against denaturation, or aggregation under adverse conditions, such as viral infection, heat, or acid. They rarely perform these processes on their own that is they require assistance from co-chaperones that speed up ATP hydrolysis and stabilize HSP70-substrate interactions. Some co-chaperones include J-domain containing HSP40, HSP90, eukaryotic HSP110, and nucleotide exchange factors [36]. HSP70s have been a hot spot for extensive study due to their different functions according to their location, for example, extracellular HSP70s carry out immunomodulatory tasks that activate tolerance responses and immunological responses, whereas intracellular HSP70s carry out cytoprotective functions as molecular chaperon proteins [16].

Apart from the diverse cellular functions performed by HSP70s, they are hijacked by viruses and included in their life cycle mediating viral attachment and endocytosis, viral penetration and uncoating, viral assembly and budding, and lastly viral transcription and replication within the host. Because of these extensive roles of HSP70, it has been the antiviral target to inhibit virus infection [3].

Interactions of human HSP27 with SARS-CoV-2 spike protein

The multiple sequence alignment result of the human HSP27 from different mammals Homo sapiens, Mus musculus, Sus scrofa, Bos taurus, Macaca mulatta, Equus caballus, Gorrilla gorilla, Felis catus, Canis lupus, Poeciliopsis lucida, Cricetulus longicaudatus, and Rattus norvegicus that indicate that the amino acid residues of HSP27 are highly conserved (in black) which indicates that they have the same biochemical functions and similar structures (Figures 3 and 4) [37].

Figure 3

Sequence alignment of mammalian HSP27 from 12 different mammalian species (amino acid residues 1–215); Homo sapiens (human), Mus musculus (Mouse), Sus scrofa (pig), Bos taurus (cow), Macaca mulatta (Rhesus macaque), Equus caballus (Horse), Gorrilla gorilla (Gorilla), Felis catus (Cat), Canis lupus (Dog), Poeciliopsis lucida (Desert topminnow), Cricetulus longicaudatus (Chinese Hamster), and Rattus norvegicus (laboratory rat) to show 161 the similarities between the protein sequences [37].

Figure 4

The image shows a 3D model showing the interaction between SARS-CoV-2 spike proteins and the human HSP27 [40,39].

Table S1 shows the types of intermolecular bonds involved in the three chains of SARS-CoV-2 spike protein with the HSP27 and the binding energy (kcal/mol) involved in the interaction of HSP27 with each chain of SARS-CoV-2 spike protein in which the interactions were the same and results were obtained after docking the two proteins and visualized by Discovery Studio [40].

Amino acids of SARS-CoV-2 spike protein with multiple interactions were identified interacting with amino acids of HSP27 and visualized the segments of amino acids residue of SARS-CoV-2 spike protein with amino acids interacting with HSP27 showing the type of intermolecular bonds involved shown in Table S2 [40].

The 2D diagrams, as shown above in Figure 5a, show interactions between amino acids from the SARS-CoV-2 spike protein forming intermolecular bonds with amino acids from human HSP27. Part of the amino acids from SARS-CoV-2 spike protein forming bonds with HSP27 was selected, consisting of amino acids from positions 710–714 (Arg20, Tyr23, Trp22) that formed conventional hydrogen bonds, carbon hydrogen bond, and electrostatic interaction with the amino acids from the human HSP27. Amino acid Ser711 formed carbon–hydrogen bonds with Trp22, amino acid Ile712 formed hydrophobic (Alkyl) interactions with Arg20, and amino acid Ala713 formed conventional hydrogen bonds with Trp22, Typ23, and Arg20 from the human HSP27. This contributes to the formation of a stable complex between the two proteins [40].

Figure 5

(a) The intermolecular bonds involved around amino acids 710, 711, 712, 713, and 714 of SARS-CoV-2 spike protein interacting with amino acids (Arg20, Tyr23, Trp22) humans HSP27. (b) The intermolecular bonds involved around amino acids 1,037, 1,038, 1,039, 1,040, and 1,041 of SARS-CoV-2 spike protein interacting with amino acids (Pro39, Gln44, Glu41, Trp45, Leu38, Gly48) human HSP27. (c) The intermolecular bonds involved around amino acids 703, 704, 705, 706, and 707 of SARS-CoV-2 spike protein interacting with amino acids (Asp21, Trp22, Tyr23) human HSP27 [40].

The interactions are also shown in Figure 5b using amino acids from the SARS-CoV-2 spike protein from position 1,037 to1,040 (Pro39, Glu41, Leu38, and Gly48) interacting with amino acids from human HSP27 where amino acid Lys1038 on the spike protein formed conventional hydrogen bonds with Gln44, carbon–hydrogen bond with Gly48, and electrostatic (attractive charge) interactions with Trp45, and Arg1039 formed electrostatic (attractive charge) interactions with Glu41 and formed conventional hydrogen bonds with Pro39. VAL1040 formed hydrophobic (Alkyl) interactions with Leu38, and lastly, Phe1042 formed conventional hydrogen bonds with Pro39 from human HSP27 [40].

Amino acids from position 703–707 (Asp21, Trp22, Tyr23) from spike protein were used to show other interactions between the two proteins. The amino acid residues Ser704 and Ala706 formed conventional hydrogen bonds with Tyr23 from human HSP27, Val705 formed hydrophobic (pi-sigma) interactions with Trp22 and Tyr3 of HSP27, and Tyr707 formed conventional hydrogen bonds with Asp21 [40].

Interactions between human HSPA8 and SARS-CoV-2 spike protein

Sequence analysis of human HSPA8 (with accession number AAH19816.1) indicated that the molecular chaperon is located on chromosome 11 and has 646 amino acid residues (Table S3). A heat shock 70 kDa protein domain was depicted at positions 1–613 and the chaperon has an N-terminal NBD at positions 2–386, and an SBD at positions 349–509 (Table S3) [41] that makes part of a class of proteins associated with membrane protein complexes for signal transduction or transport.

From previous preliminary studies, HSPA8 has been known to be localized in the cytoplasm and nucleus and shuttles between the cytoplasm and nucleus to perform various functions [3]. However, the results from sequence analysis of the HSPA8 postulated that the human HSPA8 is localized in the melanosomes and the cell surface membrane (Table S3) where it acts as a membrane-anchored protein expressed in the cytoplasmic membranes of distinctive cell linings that include smooth muscle cells, respiratory tract cells, and the epithelia of the small intestine [1]. The location of the human HSPA8 on the cell surface membrane is convenient for the molecular chaperon to act as a receptor that binds with the SARS-CoV-2 spike protein, thereby mediating viral entry into the host cell [42].

Sequence analysis of the SARS-CoV-2 spike protein (accession number YP_009724390.1) (Table S4) indicated that the surface glycoprotein consists of 1273 amino acids [43,8] with the RBD at positions 319–541, an N-terminal domain at positions 14–303, and a C-terminal domain at positions 334–527 [32,33,34,35,36,37,38,39,40,41,42,43] (Figure 6). The previously mentioned domains make part of the S1 subunit of the spike protein [7] that major functions in viral attachment during the initial stages of viral infection.

Figure 6

3D structures of the human HSPA8 and SARS-CoV-2 spike protein. (a) 3D structure of human HSPA8 obtained from Phyre 2 as visualized by PyMOL indicating the N-terminal NBD and SBD. (b) Ramachandran plot for human HSPA8. (c) 3D structure of the SARS-CoV-2 spike protein obtained from Phyre 2 as visualized by PyMOL indicating RBD, C-terminal domain, N-terminal domain, fusion peptides (FP1 and FP2), and HR units (HR1 and HR2). (d) Ramachandran plot for SARS-CoV-2 spike protein obtained from PDBsum [44,39].

Other essential segments or regions within the SARS-CoV-2 spike protein predicted from sequence analysis included the integrin-binding motif that is essential for cell-to-cell adhesion [45], a receptor-binding motif that binds to human host receptors, [46], fusion peptides, and HR units. These regions make part of the S2 membrane-fusion subunit and assist in viral fusion to the host cell membrane allowing viral genomes to penetrate the host cells. The presence of the S1 and S2 subunits substantiates that the spike protein is essential for viral attachment and fusion, the initial stages of viral infection since it has the necessary commodities required to carry out the function [47].

In the study, Bioedit software was utilized to perform multiple sequence alignment of the SARS-CoV-2 spike proteins with spike proteins from other known hCoVs and multiple sequence alignment of human HSPA8 with other closely related mammalian species attained from BLAST. The purpose of multiple sequence alignment was to align all sequences in the query set to determine and identify conserved domains across a group of evolutionarily related sequences [48].

Sequences of HSPA8 were attained from nine mammalian species that are closely related to the HSPA8 from H. sapiens using BLAST and the max score of the results. Figure 7 shows a section of multiple sequence alignment of the 10 sequences of mammalian HSPA8 (including one from H. sapiens), from amino residues 403–545 as they are the amino acid residues involved in the interaction with the amino acid residues from SARS-CoV-2 spike protein. The multiple sequence alignment (Figure 7) indicates that the amino acid residues of the functional domain in the human HSPA8, SBD at positions 349–509 (Table S3), are highly conserved throughout all the 10 different mammalian species of the same HSP from the selected amino acid residue region (from position 403 to 545) indicating that the SBD is more likely to be essential for the function of the HSPA8 [49].

Figure 7

Sequence alignment of HSPA8 from 10 different mammalian species (amino residues 403–545); Homo sapiens, Fukomys damarensis, Manis javanica, Marmota monax, Equus caballus, Mus musculus, Bos taurus, and Microtus ochrogaster [37].

Multiple sequence alignment for the spike proteins (Figure 8) was performed from sequences of the SARS-CoV-2 spike proteins and the spike proteins from known hCoVs [50] using Bioedit. Figure 8 indicates the sequence alignment of the spike proteins from the coronaviruses from amino acid residues 1–245. The residues highlighted in red are the residues involved in the protein–protein interaction between the spike protein and human HSPA8 in Figure 5. For most of the sequence alignment (Figure 8), the amino acids are non-conserved with other regions being conserved and semi-conserved which are essential in viral entry and interaction with the host cell receptor.

Figure 8

Sequence alignment of spike proteins from the six known hCoV, middle east respiratory syndrome-related coronavirus spike protein, human coronavirus HKU1 spike protein, human coronavirus 229E spike protein, human coronavirus NL63 spike protein, and human coronavirus OC43 spike protein. Amino residues in red are the residues involved in intermolecular interaction with human HSPA8 (amino acid residues 27–245) [37].

SARS-CoV-2 spike protein consists of three distinct chains, chains A, B, and C. These chains were docked separately with the human HSPA8 to predict whether the human HSPA8 interacts with each individual chain [51]. The amino acid residues from human HSPA8 (from positions 403 to 545) interacted with each individual chain of the spike protein forming hydrogen bonds, hydrophobic interactions, and electrostatic interactions. The binding energy from the interaction between human HSPA8 and chain A for the SARS-CoV-2 spike was −828.3 kcal/mol, and that between human HSPA8 and chain B, chain C for the SARS-CoV-2 spike was −827.9 kcal/mol (Table S5).

The negative binding energy value indicated that the interaction between the human HSPA8 and the SARS-CoV-2 spike protein occurs spontaneously without consuming energy. The stability of a protein complex is related to its binding energy that is the higher the binding energy the more stable the protein complex. Because the binding energy values of the docked molecules were high, it can be postulated and concluded that the complex formed from the interaction between the HSPA8 and SARS-CoV-2 spike protein is stable [52].

To better comprehend and visualize the chemical interactions between SARS-CoV-2 spike proteins and human HSPA8, 2D diagrams were generated for each individual chain of the spike protein (Figure 9). The 2-dimensional structure of Figure 9a was constructed using amino acids from the SARS-CoV-2 spike protein with the most interactions with amino acids from human HSPA8 [33]. A small chain of amino acids from SARS-CoV-2 spike protein chain A was selected, which consisted of amino acids from positions 136–140 (Cys136, Asn137, Asp138, Pro139, and Phe140) that formed conventional hydrogen bonds and electrostatic interaction with the amino acids from the human HSPA8. Amino acid Asn137 formed conventional hydrogen bonds with Arg469, and amino acid Asp138 formed electrostatic interactions with Arg469 and conventional hydrogen bonds with Arg469 and Asp433 from the human HSPA8. This indicates that the amino acids in the spike protein form extensive bonds with those from the human HSPA8, thereby contributing to a stable complex from the interactions [33].

Figure 9

2D diagrams indicating the chemical interactions between SARS-CoV-2 spike proteins and human HSPA8. (a) Intermolecular interactions between selected amino acid residues of the SARS-CoV-2 spike proteins in chain A (Cys136, Asn137, Asp138, Pro139, and Phe140) and human HSPA8. (b) Intermolecular interactions between selected amino acid residues of the SARS-CoV-2 spike proteins in chain B (Phe65, His66, Ala67 Ile68, and His69) and human HSPA8. (c) Intermolecular interactions between selected amino acid residues of the SARS-CoV-2 spike proteins in chain C (Leu212, Val213, Arg214, Asp215, and Leu216) and human HSPA8 [33].

The 2-dimensional structure of Figure 9b was constructed using amino acids from the SARS-CoV-2 spike protein from positions 65–69 (Phe65, His66, Ala67 Ile68, and His69) with the most interactions with amino acids from human HSPA8 [33]. Multiple extensive bonds were depicted from the analysis showing multiple conventional hydrogen bonds and hydrophobic interactions. Amino acid HIS69 on the spike protein formed multiple conventional hydrogen bonds with Glu404, Gln441, and Tyr545 from human HSPA8 also forming hydrophobic (pi-alkyl and alkyl) interactions with amino acids Ile403 and Ile440. Other amino acid residues interacted with the human HSPA8 that is His66 from chain B interacted with Gly407 using hydrogen bonds, and Ala67 from chain B interacted with Leu542 and Tyr545 via hydrophobic interactions [33]. These interactions are highly extensive therefore more energy will be required to break these bonds making the protein complex formed from the interaction more stable. The interaction between human HSPA8 and chain B of the SARS-CoV-2 spike protein yields the most stable complex due to these multiple extensive bonds.

Chains A and C showed roughly similar interactions; hence, the complex formed from the interactions of the two chains and the human HSPA8 may be deemed roughly the same in terms of stability. From Figure 9c, amino acids from positions 212–216 (Leu212, Val213, Arg214, Asp215, and Leu216) from chain C from spike proteins were used to construct the 2D diagram showing the interactions between the two proteins of interest forming conventional hydrogen bonds, hydrophobic interactions, and electrostatic interactions[33]. The amino acid residues Val213 and Arg214 formed conventional hydrogen bonds with Lys451 from human HSPA8, Val213 interacts with Phe547 via hydrophobic interactions, Arg214 interacts with Asp452 via electrostatic interactions, and Asp215 interacts with Lys451 via hydrophobic interactions.