Deconstructing SARS-CoV-2 neutralization: A modular molecular framework for computational design and comparison of antibodies and nanobodies targeting the spike RBD.

Since 2020 the COVID-19 pandemic has led scientists to search for strategies to predict the transmissibility and virulence of new severe acute respiratory syndrome coronavirus 2 variants based on the estimation of the affinity of the spike receptor binding domain (RBD) for the human angiotensin-converting enzyme 2 (ACE2) receptor and/or neutralizing antibodies. In this context, our lab developed a computational pipeline to quickly quantify the free energy of interaction at the spike RBD/ACE2 protein-protein interface, reflecting the incidence trend observed in the transmissibility/virulence of the investigated variants. In this new study, we used our pipeline to estimate the free energy of interaction between the RBD from 10 variants, and 14 antibodies (ab), or 5 nanobodies (nb), highlighting the RBD regions preferentially targeted by the investigated ab/nb. Our structural comparative analysis and interaction energy calculations allowed us to propose the most promising RBD regions to be targeted by future ab/nb to be designed by site-directed mutagenesis of existing high-affinity ab/nb, to increase their affinity for the target RBD region, for preventing spike-RBD/ACE2 interactions and virus entry in host cells. Furthermore, we evaluated the ability of the investigated ab/nb to simultaneously interact with the three RBD located on the surface of the trimeric spike protein, which can alternatively be in up- or down- (all-3-up-, all-3-down-, 1-up-/2-down-, 2-up-/1-down-) conformations.

Modeling SARS-CoV-2 spike/ACE2 protein-protein interactions for predicting the binding affinity of new spike variants for ACE2, and novel ACE2 structurally related human protein targets, for COVID-19 handling in the 3PM context.

Aims: The rapid spread of new SARS-CoV-2 variants has highlighted the crucial role played in the infection by mutations occurring at the SARS-CoV-2 spike receptor binding domain (RBD) in the interactions with the human ACE2 receptor. In this context, it urgently needs to develop new rapid tools for quickly predicting the affinity of ACE2 for the SARS-CoV-2 spike RBD protein variants to be used with the ongoing SARS-CoV-2 genomic sequencing activities in the clinics, aiming to gain clues about the transmissibility and virulence of new variants, to prevent new outbreaks and to quickly estimate the severity of the disease in the context of the 3PM. Methods: In our study, we used a computational pipeline for calculating the interaction energies at the SARS-CoV-2 spike RBD/ACE2 protein-protein interface for a selected group of characterized infectious variants of concern/interest (VoC/VoI). By using our pipeline, we built 3D comparative models of the SARS-CoV-2 spike RBD/ACE2 protein complexes for the VoC B.1.1.7-United Kingdom (carrying the mutations of concern/interest N501Y, S494P, E484K at the RBD), P.1-Japan/Brazil (RBD mutations: K417T, E484K, N501Y), B.1.351-South Africa (RBD mutations: K417N, E484K, N501Y), B.1.427/B.1.429-California (RBD mutations: L452R), the B.1.141 (RBD mutations: N439K), and the recent B.1.617.1-India (RBD mutations: L452R; E484Q) and the B.1.620 (RBD mutations: S477N; E484K). Then, we used the obtained 3D comparative models of the SARS-CoV-2 spike RBD/ACE2 protein complexes for predicting the interaction energies at the protein-protein interface. Results: Along SARS-CoV-2 mutation database screening and mutation localization analysis, it was ascertained that the most dangerous mutations at VoC/VoI spike proteins are located mainly at three regions of the SARS-CoV-2 spike "boat-shaped" receptor binding motif, on the RBD domain. Notably, the P.1 Japan/Brazil variant present three mutations, K417T, E484K, N501Y, located along the entire receptor binding motif, which apparently determines the highest interaction energy at the SARS-CoV-2 spike RBD/ACE2 protein-protein interface, among those calculated. Conversely, it was also observed that the replacement of a single acidic/hydrophilic residue with a basic residue (E484K or N439K) at the "stern" or "bow" regions, of the boat-shaped receptor binding motif on the RBD, appears to determine an interaction energy with ACE2 receptor higher than that observed with single mutations occurring at the "hull" region or with other multiple mutants. In addition, our pipeline allowed searching for ACE2 structurally related proteins, i.e., THOP1 and NLN, which deserve to be investigated for their possible involvement in interactions with the SARS-CoV-2 spike protein, in those tissues showing a low expression of ACE2, or as a novel receptor for future spike variants. A freely available web-tool for the in silico calculation of the interaction energy at the SARS-CoV-2 spike RBD/ACE2 protein-protein interface, starting from the sequences of the investigated spike and/or ACE2 variants, was made available for the scientific community at: https://www.mitoairm.it/covid19affinities. Conclusion: In the context of the PPPM/3PM, the employment of the described pipeline through the provided webservice, together with the ongoing SARS-CoV-2 genomic sequencing, would help to predict the transmissibility of new variants sequenced from future patients, depending on SARS-CoV-2 genomic sequencing activities and on the specific amino acid replacement and/or on its location on the SARS-CoV-2 spike RBD, to put in play all the possible counteractions for preventing the most deleterious scenarios of new outbreaks, taking into consideration that a greater transmissibility has not to be necessarily related to a more severe manifestation of the disease. Supplementary Information: The online version contains supplementary material available at 10.1007/s13167-021-00267-w.

New insights about the structural rearrangements required for substrate translocation in the bovine mitochondrial oxoglutarate carrier.

The oxoglutarate carrier (OGC) belongs to the mitochondrial carrier family and plays a key role in important metabolic pathways. Here, site-directed mutagenesis was used to conservatively replace lysine 122 by arginine, in order to investigate new structural rearrangements required for substrate translocation. K122R mutant was kinetically characterized, exhibiting a significant Vmax reduction with respect to the wild-type (WT) OGC, whereas Km value was unaffected, implying that this substitution does not interfere with 2-oxoglutarate binding site. Moreover, K122R mutant was more inhibited by several sulfhydryl reagents with respect to the WT OGC, suggesting that the reactivity of some cysteine residues towards these Cys-specific reagents is increased in this mutant. Different sulfhydryl reagents were employed in transport assays to test the effect of the cysteine modifications on single-cysteine OGC mutants named C184, C221, C224 (constructed in the WT background) and K122R/C184, K122R/C221, K122R/C224 (constructed in the K122R background). Cysteines 221 and 224 were more deeply influenced by some sulfhydryl reagents in the K122R background. Furthermore, the presence of 2-oxoglutarate significantly enhanced the degree of inhibition of K122R/C221, K122R/C224 and C224 activity by the sulfhydryl reagent 2-Aminoethyl methanethiosulfonate hydrobromide (MTSEA), suggesting that cysteines 221 and 224, together with K122, take part to structural rearrangements required for the transition from the c- to the m-state during substrate translocation. Our results are interpreted in the light of the homology model of BtOGC, built by using as a template the X-ray structure of the bovine ADP/ATP carrier isoform 1 (AAC1).

Identification of new highly selective inhibitors of the human ADP/ATP carriers by molecular docking and in vitro transport assays.

Mitochondrial carriers are proteins that shuttle a variety of metabolites, nucleotides and coenzymes across the inner mitochondrial membrane. The mitochondrial ADP/ATP carriers (AACs) specifically translocate the ATP synthesized within mitochondria to the cytosol in exchange for the cytosolic ADP, playing a key role in energy production, in promoting cell viability and regulating mitochondrial permeability transition pore opening. In Homo sapiens four genes code for AACs with different tissue distribution and expression patterns. Since AACs are dysregulated in several cancer types, the employment of known and new AAC inhibitors might be crucial for inducing mitochondrial-mediated apoptosis in cancer cells. Albeit carboxyatractyloside (CATR) and bongkrekic acid (BKA) are known to be powerful and highly selective AAC inhibitors, able to induce mitochondrial dysfunction at molecular level and poisoning at physiological level, we estimated here for the first time their affinity for the human recombinant AAC2 by in vitro transport assays. We found that the inhibition constants of CATR and BKA are 4 nM and 2.0 µM, respectively. For finding new AAC inhibitors we also performed a docking-based virtual screening of an in-house developed chemical library and we identified about 100 ligands showing high affinity for the AAC2 binding region. By testing 13 commercially available molecules, out of the 100 predicted candidates, we found that 2 of them, namely suramin and chebulinic acid, are competitive AAC2 inhibitors with inhibition constants 0.3 µM and 2.1 µM, respectively. We also demonstrated that chebulinic acid and suramin are "highly selective" AAC2 inhibitors, since they poorly inhibit other human mitochondrial carriers (namely ORC1, APC1 and AGC1).

Distance-dependent hydrophobic-hydrophobic contacts in protein folding simulations.

Successful prediction of protein folding from an amino acid sequence is a challenge in computational biology. In order to reveal the geometric constraints that drive protein folding, highlight those constraints kept or missed by distinct lattices and for establishing which class of intra- and inter-secondary structure element interactions is the most relevant for the correct folding of proteins, we have calculated inter-alpha carbon distances in a set of 42 crystal structures consisting of mainly helix, sheet or mixed conformations. The inter-alpha carbon distances were also calculated in several lattice "hydrophobic-polar" models built from the same protein set. We found that helix structures are more prone to form "hydrophobic-hydrophobic" contacts than beta-sheet structures. At a distance lower than or equal to 3.8 Å (very short-range interactions), "hydrophobic-hydrophobic" contacts are almost absent in the native structures, while they are frequent in all the analyzed lattice models. At distances in-between 3.8 and 9.5 Å (short-/medium-range interactions), the best performing lattice for reproducing mainly helix structures is the body-centered-cubic lattice. If protein structures contain sheet portions, lattice performances get worse, with few exceptions observed for double-tetrahedral and body-centered-cubic lattices. Finally, we can observe that ab initio protein folding algorithms, i.e. those based on the employment of lattices and Monte Carlo simulated annealings, can be improved simply and effectively by preventing the generation of "hydrophobic-hydrophobic" contacts shorter than 3.8 Å, by monitoring the "hydrophobic-hydrophobic/polar-polar" contact ratio in short-/medium distance ranges and by using preferentially a body-centered-cubic lattice.

Sirtuin 3 interacts with Lon protease and regulates its acetylation status.

Lon is a mitochondrial protease that degrades oxidized damaged proteins, assists protein folding and participates in maintaining mitochondrial DNA levels. Changes in Lon mRNA levels, protein levels and activity are not always directly correlated, suggesting that Lon could be regulated at post translational level. We found that Lon and SIRT3, the most important mitochondrial sirtuin, colocalize and coimmunoprecipitate in breast cancer cells, and silencing or inhibition of Lon did not alter SIRT3 levels. Silencing of SIRT3 increased the levels of Lon protein and of its acetylation, suggesting that Lon is a target of SIRT3, likely at K917.