Omicron SARS-CoV-2 Spike-1 Protein’s Decreased Binding Affinity to 7nAChr: Implications for Autonomic Dysregulation of the Parasympathetic Nervous System and the Cholinergic Anti-Inflammatory Pathway—An In Silico Analysis

Omicron is the dominant strain of COVID-19 in the United States and worldwide. Although this variant is highly transmissible and may evade natural immunity, vaccines, and therapeutic antibodies, preclinical results in animal models and clinical data in humans suggest omicron causes a less severe form of infection. The molecular basis for the attenuation of virulence when compared to previous variants is currently not well understood. Using protein–ligand docking simulations to evaluate and compare the capacity of SARS-CoV-2 spike-1 proteins with the different COVID-19 variants to bind to the human α7nAChr (i.e., the core receptor under the control of the vagus nerve regulating the parasympathetic nervous system and the cholinergic anti-inflammatory pathway), we found that 10 out of the 14 mutated residues on the RBD of the B.1.1.529 (Omicron) spike, compared to between 0 and 2 in all previous variants, were present at the interaction interface of the α7nAChr. We also demonstrated, through protein–protein docking simulations, that these genetic alterations cause a dramatic decrease in the ability of the Omicron SARS-CoV-2 spike-1 protein to bind to the α7nAChr. These results suggest, for the first time, that the attenuated nature of Omicron infection in humans and animals compared to previous variants may be attributable to a particular set of genetic alterations, specifically affecting the binding site of the SARS-CoV-2 spike-1 protein to the α7nAChr.


Introduction
The rapid spread of the B.1.1.529 (Omicron) SARS-CoV-2 variant, first identified in Botswana and South Africa in November 2021, raised major concerns regarding the prospect of a new global wave of lethal COVID-19 infections. Moreover, data indicating as many as 37 mutations in the Omicron spike protein relative to the Wuhan-Hu-1 strain [1] raised further concerns that this variant could evade natural immunity, vaccines, and therapeutic antibodies. Omicron has also been found to be more transmissible than previous strains (i.e., Omicron may infect and multiply 70 times faster in the upper airways than Delta) [2]. Early clinical data, however, suggest a less severe, more indolent form of infection by Omicron in humans [3][4][5][6]. Consistent with this human clinical data, in multiple recent reports, Omicron has been shown to cause milder respiratory infection in animal models and to be attenuated in both mice and hamsters for causing infection and/or disease, compared to previous variants [7,8]. Nevertheless, the molecular basis for this attenuation, with respect to previous variants in humans and rodents, is not yet well understood.
Biomedinformatics 2022, 2 554 We have recently performed protein-ligand docking simulations to evaluate the potential binding sites of multiple therapeutic compounds against the SARS-CoV-2 spike-1 protein [9]. In this in silico study, in agreement with other reports [10,11], we demonstrated a potential direct binding of the SARS-CoV-2 spike-1 protein to the Alpha-7 nicotinic acetylcholine receptor (α7nAChr), which is the core receptor under the control of the vagus nerve, regulating the parasympathetic nervous system and the cholinergic anti-inflammatory pathway [12]. This study, and other studies [13], demonstrated that the spike-1 Y674-R685 region penetrates deeply into the binding pocket in which it forms interactions with the residues lining the aromatic box, namely with TrpB, TyrC1, and TyrC2 [10,11]. This motif contains a four-residue, polybasic PRRA insertion that is homologous to several neurotoxins known to target the α7nAChr, as well as several known nAChR antagonists, further suggesting that this region of the SARS-CoV-2 spike protein may bind to the α7nAChr [10,11], as well as to other receptors, including neuropilins [14]. Consistent with this view, the abrogation of the PRRA motif has been experimentally demonstrated to affects virus entry into host cells [15,16]. Importantly, the α7nAChr-regulated vagus nerve-controlled anti-inflammatory pathway is known to play a crucial role in balancing the body's response to inflammation and sepsis [12,17], connecting the involuntary parasympathetic nervous system innervating all major organs to cytokine-producing cells, such as TNF, IL-1, and IL-6-secreting macrophages, lymphocytes, and mast cells [12,[17][18][19], which in turn play a major role during the inflammatory phase of COVID-19 infection (i.e., the cytokine storm) [20]. Because the α7nAChr is the known port of entry to the human body for other neurotropic viruses, such as the rabies virus [21], as well as the specific binding site to human neurons of multiple lethal snake venom toxins [10,11], it stands to reason that the ability of the SARS-CoV-2 spike-1 protein to bind and hijack the α7nAChr-regulated response of the autonomous nervous system to inflammation may represent one of the crucial characteristics, explaining its high lethality in both humans and animals. Clinical data correlating hypercytokinemia and hyperinflammation with the dysregulation of the Ach/α7nAChR pathway in severe COVID-19 patients further support this hypothesis [22].
Accordingly, in this study, we used protein-ligand docking simulations to evaluate and compare the capacity of SARS-CoV-2 spike-1 proteins with the different COVID-19 variants (i.e., Alpha, Beta, Gamma, Delta, and Omicron) to bind to the human α7nAChr. Notably, we found that 10 out of the 14 mutated residues on the RBD of the B.1.1.529 (Omicron) spike, compared to between 0 and 2 in all previous variants, were present at the interaction interface with the α7nAChr. We also demonstrated through protein-protein docking simulations that these genetic alterations cause a dramatic decrease in the ability of the Omicron SARS-CoV-2 spike-1 protein to bind to the α7nAChr. Clinically speaking, these results suggest that the attenuated nature of Omicron infection in humans and animals, compared to previous variants, may be attributable to a particular set of genetic alterations specifically affecting the binding site of the SARS-CoV-2 spike-1 protein to the α7nAChr.

Proteins Preparation
The structure of the complex between the spike (PDB: 6VSB) and α7nAChr was obtained from Aminpour et al. [9]. The atomic coordinates that activated the α7nAChr (PDB: 7KOX) were obtained from the PDB [23]. Meanwhile, the Omicron spike (B.1.1.529) crystallography structure was taken from PDB (7QO7) [24]. The Delta spike structure, finally, was obtained from PDB (7W92) [25]. All proteins were prepared in the Molecular Operating Environment (MOE, Chemical Computing Group; https://www.chemcomp. com/Products.htm, accessed on 5 September 2022) software package, with their protonation states adjusted and a physiological pH of 7 maintained. The structure was minimized also using MOE software.
Amino acid mutations of the Alpha, Beta, Gamma, Delta, and Omicron variants of SARS CoV-2, with a focus on NTD (aa 18-292) and RBD (aa 318-513) segments, were reproduced from Aminpour et al. [9] as listed in Table 1.

Contact Analysis
Considering the complex between the main spike protein and α7nAChr, all of the interactions involved in the binding of the two structures were analyzed using the MOE software and particularly its "Contacts" analysis function, as this tool provides information about the bond type and the interacting residues involved.
The interaction between the two proteins was evaluated using six types of contacts: hydrogen bonds (H) relative to hydrogen bond contacts; metal, relative to metal interactions which are bonded, or are close enough to be within bonded distance; ionic, for ionic bonds; arene, for arene interactions, including π:π, π-H, and π:cation contacts; covalent, relative to covalent bonds; and distance (D) interactions, relative to van der Waals distance interactions. Once these features had been extracted, we compared all of the residues involved in this bond to those involved in other variants' mutations in order to ascertain which amino acid will mutate and what the corresponding variant was. This method allowed us to determine whether the contact between the spike protein and α7nAChr modified the interaction, as well as which variant was driving these changes and which residues were involved.

Protein-Protein Docking
PatchDock software was used to perform protein-protein docking simulations [26,27]. PatchDock, it should be noted, is a geometry-based molecular docking algorithm using shape complementarity principles. The algorithm first computed the molecular surface of the proteins, then performed a segmentation to detect geometric patches (concave, convex, and flat surface pieces). Then, in the surface-match-patching step, the complementary patches were selected in order to generate candidate transformations. The candidate transformations from the previous step were evaluated using a scoring function, taking into account both geometric fit and solvation energy. The geometrically unacceptable presentations of the atoms of complex structures were filtered. Finally, an RMSD (root mean square deviation) clustering method was applied to the remaining candidates to discard similar and redundant structures. Considering the size of the proteins, we used a clustering RMSD of a 4.0 Å arm for all of the protein-protein docking analyses. The results obtained from PatchDock were further refined with the associated server, FireDock, which delivers a refinement of both the score function and the complex geometries.
The final structures were visualized and validated before the final representation was created, with the candidate complex in terms of lipophilicity and attachment to the α7nAChr, as well as with minimal binding energy, being selected in order to ensure that the states with the optimal features for every single complex were reflected in the final results. All references to the energy states are as specified in Table 2.

Align Complexes
After we obtained the different final structures, we proceeded to align these different poses. An alignment, or 'sequence alignment', is a way of arranging the primary sequences of our proteins to identify regions of similarity that may be a consequence of functional or structural relationships between the sequences. We used this instrument using MOE software to be able to align α7nAChr residues for all the 3 complexes, and this enabled us to investigate the interaction interface of all 3 variants of the COVID spike protein, demonstrating how they attach themselves individually to an α7nAChr protein. Previous simulations were performed with the same α7nAChr structure; hence, the present alignment procedure was expected to provide a consistent match between α7 and the 3 variants of the spike protein with the same position and orientation in their unique complexes.

Results
We first analyzed the interacting residues between the main spike protein (PDB:6VSB) and α7nAChr (PDB:7KOX), using contact analysis of MOE software. The summary of the interacting residues is presented in Table 3. Table 3. All residues involved in the bond between the α7nAChr (PDB:7KOX) and the original SARS-CoV-2 spike protein (PDB:6VSB), with bond type interactions: distance (D), hydrogen bond (H), and ionic (I). Set A refers to α7nAChr, set B refers to spike protein. The explanation of each feature in the bond type column is described in computational details (contact analysis).

Bond Type Set
First, we compared the interacting residues between the main spike protein (Table 3) and the mutated residues in different spike variants (Table 1). In this manner it was determined that only two residues in the Beta (E484K, N501Y), two in the Gamma (E484K, N501Y), one in the Alpha (N501Y), and none in the Delta variant were mutated. On the other hand, 33 mutations were found in the binding area between the spike and the α7nAChr in the Omicron variant. The mutated residues are summarized in Table 4 and represented in Figure 1.
Second, we performed the ulterior protein-protein docking analysis to determine the energy states of different complex 'spike variant-α7nAChr' cases. After searching configurations, scoring, filtering, and refining with FireDock software, the candidate complex was selected. Any structure, it should be noted, has a characteristic energy state that provides details about the bond and, in particular, its strength. The best docked structure was selected on the basis of its global energy. This value is considered to be related to binding free energy (Kcal/mol), and a higher negative value represents stronger binding. The details of the interaction energies between different complex 'spike variant-α7nAChr' cases are summarized in Table 5. These final predicted structures for each variant of spike protein in complex with the α7nAChr are represented in Figure 2A-C, respectively.
The α7-Delta's spike was found to have the lowest global energy (the strongest binding affinity), whereas α7-Omicron's spike was found to have the highest global energy (the weakest binding affinity). A visual inspection of the ACE feature confirmed the trend in which the impairment of the bond involving α7-Omicron, compared to α7-Delta, resulted in a lower binding energy (a stronger bond), compared to a higher one (a weaker bond).
In the last instance, the Align tool in MOE provided a graphical representation of the interaction's interface regarding the three different spikes and the α7nAChr, as shown in Figure 3. We noticed that Omicron and Delta shared a large portion of the interaction interface with α7. Table 4. Interacting residues involved in the bond between the α7nAChr and the main SARS-CoV-2 spike protein that are mutated in the Omicron variant. Set A refers to the α7nAChr, set B refers to the original spike protein and set B-mutated refers to the mutated residues in the Omicron variant.
Gly446 G446S BioMedInformatics 2022, 2, FOR PEER REVIEW 6 α7nAChr in the Omicron variant. The mutated residues are summarized in Table 4 and represented in Figure 1. Table 4. Interacting residues involved in the bond between the α7nAChr and the main SARS-CoV-2 spike protein that are mutated in the Omicron variant. Set A refers to the α7nAChr, set B refers to the original spike protein and set B-mutated refers to the mutated residues in the Omicron variant.

Set A Set B Set B-Mutated Set A Set B Set B-Mutated
Gly446 G446S Figure 1. The residues involved in the interaction between the mutated spike protein and the α7nA-Chr protein in the Omicron variant. To focus on the mutations, the main spike protein is isolated in the figure. The three different isomers are presented in red, green, and blue, and we highlight how the residues mutate in the Omicron variant.
Second, we performed the ulterior protein-protein docking analysis to determine the energy states of different complex 'spike variant-α7nAChr' cases. After searching configurations, scoring, filtering, and refining with FireDock software, the candidate complex was selected. Any structure, it should be noted, has a characteristic energy state that provides details about the bond and, in particular, its strength. The best docked structure was selected on the basis of its global energy. This value is considered to be related to binding free energy (Kcal/mol), and a higher negative value represents stronger binding. The details of the interaction energies between different complex 'spike variant-α7nAChr' cases  are summarized in Table 5. These final predicted structures for each variant of spike protein in complex with the α7nAChr are represented in Figure 2A-C, respectively.  The α7-Delta's spike was found to have the lowest global energy (the strongest binding affinity), whereas α7-Omicron's spike was found to have the highest global energy (the weakest binding affinity). A visual inspection of the ACE feature confirmed the trend in which the impairment of the bond involving α7-Omicron, compared to α7-Delta, resulted in a lower binding energy (a stronger bond), compared to a higher one (a weaker bond).
In the last instance, the Align tool in MOE provided a graphical representation of the interaction's interface regarding the three different spikes and the α7nAChr, as shown in Figure 3. We noticed that Omicron and Delta shared a large portion of the interaction interface with α7.

Discussion
Omicron, the latest SARS-CoV-2 variant, is spreading rapidly across the world, causing an unprecedented number of infections. The more rapid replication, when compared to the previous COVID-19 variants, and the higher number of novel mutations affecting the RBD and NTD of the spike protein (i.e., the primary sites targeted by naturally induced or vaccine-elicited neutralizing antibodies) allow Omicron to evade natural immunity, vaccines, and therapeutic antibodies. Regardless of the origin of these concerning genetic characteristics, both the clinical data from humans and the experimental data from animal models suggest that Omicron represents a less severe, more indolent form of infection [3][4][5][6]. Nevertheless, the biological factors underlying the decreased virulence of Omicron remain poorly understood. Strategies and methods performed in this study could potentially help predict how future variants will interact with different biological structures, depending on the amino-acids-mutation characteristics of the variant of interest.
In a previous study, through in silico analysis, we demonstrated the potential direct binding of the SARS-CoV-2 spike-1 protein to the Alpha-7 nicotinic acetylcholine receptor (α7nAChr), which is the core receptor under the control of the vagus nerve regulating the parasympathetic nervous system and the cholinergic anti-inflammatory pathway [12]. We also demonstrated strong binding by the macrocyclic lactone ivermectin and related compounds to the α7nAChr, while other studies have demonstrated that ivermectin acts as a positive allosteric effector of the α7nAChr [28,29]. This anti-inflammatory effect was identified as contributing to a significant reduction in the severity of clinical signs (p < 0.001) by concurrent ivermectin administration in golden hamsters intranasally inoculated with SARS-CoV-2 [30]. It has also recently been demonstrated that antibodies, such as human

Discussion
Omicron, the latest SARS-CoV-2 variant, is spreading rapidly across the world, causing an unprecedented number of infections. The more rapid replication, when compared to the previous COVID-19 variants, and the higher number of novel mutations affecting the RBD and NTD of the spike protein (i.e., the primary sites targeted by naturally induced or vaccine-elicited neutralizing antibodies) allow Omicron to evade natural immunity, vaccines, and therapeutic antibodies. Regardless of the origin of these concerning genetic characteristics, both the clinical data from humans and the experimental data from animal models suggest that Omicron represents a less severe, more indolent form of infection [3][4][5][6]. Nevertheless, the biological factors underlying the decreased virulence of Omicron remain poorly understood. Strategies and methods performed in this study could potentially help predict how future variants will interact with different biological structures, depending on the amino-acids-mutation characteristics of the variant of interest.
In a previous study, through in silico analysis, we demonstrated the potential direct binding of the SARS-CoV-2 spike-1 protein to the Alpha-7 nicotinic acetylcholine receptor (α7nAChr), which is the core receptor under the control of the vagus nerve regulating the parasympathetic nervous system and the cholinergic anti-inflammatory pathway [12]. We also demonstrated strong binding by the macrocyclic lactone ivermectin and related compounds to the α7nAChr, while other studies have demonstrated that ivermectin acts as a positive allosteric effector of the α7nAChr [28,29]. This anti-inflammatory effect was identified as contributing to a significant reduction in the severity of clinical signs (p < 0.001) by concurrent ivermectin administration in golden hamsters intranasally inoculated with SARS-CoV-2 [30]. It has also recently been demonstrated that antibodies, such as human mAb CR3022 and COVA1-16 mAb, which binds the SARS-CoV-2 spike glycoprotein specifically to the region targeting the conserved cryptic epitope on the spike glycoprotein that coincides with the toxin-like sequence interacting with the α7nAChr, have strong neutralizing properties against both SARS1 and COVID-19 [31,32]. One cannot overemphasize the importance of this binding in the pathophysiology of severe COVID infection, as dysautonomia-related symptoms (e.g., alterations in the respiratory rate, heartbeat, blood pressure, vessel tone, hormone secretion, intestinal peristalsis, digestion, etc.), uncontrolled inflammatory response (i.e., cytokine storm, which is mediated by α7nAChr-expressing alveolar macrophages and other cytokine-secreting cells, including histamine-secreting mast cells), and thromboembolic complications (i.e., the major causes of morbidity and mortality in infections caused by all previous COVID-19 variants) may all be attributable to the dysregulation of the nicotinic cholinergic system (NCS), as previously hypothesized [9][10][11].
If this hypothesis is correct, the attenuated nature of the Omicron infection in humans and animals, relative to previous variants, may be related to the acquisition of a set of genetic alterations specifically affecting the binding site of the SARS-CoV-2 spike-1 protein to the α7nAChr. To evaluate this possibility, we performed experiments using proteinligand docking simulations to evaluate and compare the capacity of SARS-CoV-2 spike-1 proteins from the different COVID-19 variants (i.e., Alpha, Beta, Gamma, Delta, and Omicron) to bind to the human α7nAChr. As a notable contribution of the present work, we found that 10 of the 14 mutated residues on the RBD of the B.1.1.529 (Omicron) spike were located at the interface of the interaction with the α7nAChr, whereas no more than two mutated residues were present at this interface for any of the previous COVID-19 variants. More importantly, we found that these genetic alterations predict a dramatic decrease in the ability of Omicron SARS-CoV-2 spike-1 protein to bind to the α7nAChr, which, as noted above, is the core receptor regulating the parasympathetic nervous system and the cholinergic anti-inflammatory pathway.
Several clinical features of the Omicron infection are consistent with a decreased COVID-19 binding to the α7nAChr. For example, while the loss of smell (anosmia) and/or of taste (ageusia) have been considered hallmarks of all previous COVID-19 infection variants, and bearing in mind that the α7nAChr is highly expressed in human nerves, including the olfactory and gustatory nerves [33], only 0.17 percent of patients infected with Omicron have been reported to have developed anosmia [34]. Furthermore, the life-threatening cytokine storm (i.e., the severe immune reaction seen after 7-8 days from the onset of symptoms in the subset of patients developing severe COVID-19 infection) has been shown to be less prevalent in Omicron-infected patients, with a recent case series of 125 hospitalized Omicron-infected patients reporting 0 such cases [35]. These clinical data, combined with our results coming from protein-protein docking and relative to the energy states of the complexes, provide further support to the notion that, in addition to the experimentally proven inability of Omicron to reproduce in the lower airways [1,2], this variant may also have lost the ability to infect/inflame the α7nAChr overexpressing alveolar macrophages.
In conclusion, empirical results suggest that the Omicron spike-1 protein has acquired RBD mutations that, while they allow it to evade immunity from vaccines and the natural immunity afforded by previous infection, also confer a reduced capacity to bind to the α7nAChr, i.e., the main regulators of the brain cholinergic anti-inflammatory pathway [36]. Because the uncontrolled inflammatory response and other clinical symptoms of COVID-19 (e.g., anosmia and thromboembolic complications) may be caused by NCS dysregulation (9)(10)(11), this unique feature of Omicron may explain its intrinsically reduced virulence and account for the decreased risk of severe hospitalization or death associated with Omicron, compared to previous variants [37].

Conclusions
We used protein-ligand docking simulations to evaluate and compare the capacity of SARS-CoV-2 spike-1 proteins from the different COVID-19 variants to bind to the human α7nAChr. Our in silico results demonstrated that 10 out of the 14 mutated residues on the RBD of the B.1.1.529 (Omicron) spike, compared to between 0 and 2 in all previous variants, were present at the interface of the interaction with the α7nAChr. Importantly, we found that these genetic alterations caused a dramatic decrease in the ability of the Omicron SARS-CoV-2 spike-1 protein to bind to the α7nAChr. Because the α7nAChr is the core receptor under the control of the vagus nerve regulating the parasympathetic nervous system and the cholinergic anti-inflammatory pathway, and this anti-inflammatory pathway is known to play a crucial role in balancing the body's response to inflammation and sepsis, connecting the involuntary parasympathetic nervous system innervating all major organs to cytokine-producing cells, which in turn play a major role during the inflammatory phase of COVID-19 infection (i.e., the cytokine storm), we suggest that the attenuated nature of the Omicron infection in humans and animals, compared to previous variants, may be attributable to a particular set of genetic alterations, specifically affecting the binding site of the SARS-CoV-2 spike-1 protein to the α7nAChr.

Data Availability Statement:
The datasets used and analyzed in the current study are available from the corresponding author upon request.

Conflicts of Interest:
A.D.S. reports grants from PUMA, grants from IMMUNOMEDICS, grants from GILEAD, grants from SYNTHON, grants and personal fees from MERCK, grants from BOEHINGER-INGELHEIM, grants from GENENTECH, grants and personal fees from TESARO and grants and personal fees from EISAI. The other authors declare no conflict of interest.

Abbreviations
The following abbreviations are used in this manuscript: