key: cord-1027525-ri2o86es
authors: Hassan, Yasser
title: Post-Infection Entry Mechanism of Ricin A Chain-Pokeweed Antiviral Proteins (RTA-PAPs) Chimeras is Mediated by Viroporins
date: 2021-06-17
journal: bioRxiv
DOI: 10.1101/2021.06.17.448882
sha: 64bebe8d40d2fda2e24feb0a8175f8df1004980c
doc_id: 1027525
cord_uid: ri2o86es

The limitations of virus-specific antiviral drugs became apparent during the current COVID-19 pandemic. The search for broad range antiviral proteins of a new kind to answer current and future pandemics has become an even more pressing matter. Here, the author further describes the expected anti-SARS-CoV-2 mechanisms of a novel broad range antiviral chimeric protein constructed between ricin A chain and pokeweed antiviral proteins. The latest in protein-ligand docking software were used to determine binding affinity of RTA-PAPs to SARS-CoV-2 frameshift stimulation element and elucidate the preferential post-infection entry mechanisms of RTA-PAPs into virus infected cells over non-infected ones, by doing a comparative analysis between in vitro and in silico results on numerous viruses. The results obtained strongly suggest that the post-infection preferential entry of RTA-PAPs into infected cells is mediated by the presence of viroporins integrated into the host cell membrane. The discovery of this mechanism revealed RTA-PAPs, and proteins like them, to be a new class of broad range antivirals that target with high specificity viroporin producing viruses, and with gain of functions in antiviral activities, post-infection.

Scientists around the globe are actively searching for potent antiviral drugs to treat 26 and cure all kind of diseases caused by various viral infections. The main strategies to 27 develop such antivirals have been focused on small virus-specific molecules that hinder a 28 precise mechanism in the viral cycle, over time, and on neutralizing antibodies, i.e. 29 vaccines or monoclonal antibodies. Those strategies have been successful so far, however, 30 they are not without limitations. Indeed, the small virus-specific molecules lead to 31 mutated forms of the viruses, over time, that are resistant to treatments, and the 32 development of neutralizing antibodies is a lengthy and costly process. In order to 33 develop virus-specific vaccines or monoclonal antibodies, an average of five to ten years 34 is required [1] . Additionally, those strategies are reactive as opposed to proactive, i.e. 35 those antivirals are only developed after the viral infections happened. 36 The current COVID-19 pandemic, caused by the severe acute respiratory syndrome 37 coronavirus 2 (SARS-CoV-2) infection, brought to light those limitations. Indeed, there are 38 still no efficient antivirals against SARS-CoV-2, and the existing vaccines, while efficient, 39 need to be renewed due to the emergence of new variants of the virus, capable of evading 40 the immune system. The need for broad range antiviral drugs capable to treat numerous 41 variants of the same virus and viruses of the same family became apparent with the 42 ongoing COVID-19 pandemic. 43 There are many potential strategies to develop broad range antivirals, and one of 44 those strategies of interest is to look into plants defensive mechanisms. Plants have 45 evolved numerous proteins capable of fighting off not only viral infections but also 46 bacterial and fungal ones, using direct and indirect mechanisms. Ribosome inactivating 47 proteins (RIPs) are such proteins, and can irreversibly inactive protein synthesis, leading 48 to cell death, upon infection by a foreign pathogen [2] . Interestingly, certain RIPs have 49 been found to have very potent direct activities against various animal and human viruses 50 by inhibiting viral entry, hindering viral cell machinery and depurinating viral 51 RNA/DNA. Furthermore, certain RIPs gain entry into infected cells preferentially over 52 non-infected ones by an unclear mechanism. For those reasons, RIPs have been the subject 53 of extensive investigations in order to circumvent the limitations of virus-specific 54 antivirals. 55 Among the most studied RIPs over the last forty years are pokeweed antiviral 56 proteins (PAPs) and ricin. PAPs and ricin were found to be very potent against numerous 57 human viruses such as Japanese encephalitis virus, human immunodeficiency virus-1 58 (HIV-1), human T-cell leukemia virus-1, herpes simplex virus, influenza, hepatitis B virus 59 (HBV), and poliovirus in vitro and in clinical trials [3] . However, RIPs suffer dose limiting 60 toxicity and unfavorable pharmacokinetics (PK) profiles when administered alone. RIPs 61 are mostly used today as immunotoxins or conjugated toxins. Nonetheless, the author 62 designed chimeras constructed with PAPs and ricin-A chain (RTA) (RTA-PAPs) with gain 63 of function in protein synthesis inhibition, and in antiviral activity in vitro and in silico 64 against numerous human viruses, such as HBV and SARS-CoV-2, at subtoxic dosages 65 with a better PK profile than either moieties alone [3] [4] . 66 Given the importance of developing broad range antivirals, the author hypothesized 67 that based on the previous results published [3] [4] [5] , and using state of the arts in silico tools, 68 it would be possible to elucidate RTA-PAPs preferential entry mechanism into infected 69 cells over non-infected ones, and further our understanding of the expected anti-SARS-70 CoV-2 mechanisms discussed previously [4] . Here, the author reports the latest results 71 regarding the novel binding mechanisms of RTA-PAPs against the frameshift stimulation 72 element (FSE) of SARS-CoV-2 genome, required for balanced expression of essential viral 73 proteins [6] , the novel associated potential depurination potential, and the logically 74 deduced preferential entry mechanism into infected cells over non-infected ones based on 75 all the accumulated in vitro, and newly generated in silico results. 76

The three-dimensional (3D) structure of SARS-CoV-2 FSE was made available 78 recently [6] , and was acquired from the Research Collaboratory for Structural 79 Bioinformatics (RCSB) website (https://www.rcsb.org/). The 3D structures of RTA, PAP1 80 and PAPS1 were also retrieved from RCSB. The predicted 3D structures of the fusion 81 protein between RTA and pokeweed antiviral protein from seed isoform 1 (PAPS1) (RTA-82 PAPS1), and the chimera between RTA mutant (RTAM) and pokeweed antiviral protein 83 isolated from leaf isoform 1 (PAP1) (RTAM-PAP1) were obtained as previously described 84 [3] . An additional stable conformation of RTAM-PAP1 (RTAM-PAP1 2 nd ) was discovered 85 in the process of double checking all the constructs, and was generated using the same 86 methodology used for RTAM-PAP1 described in our previous publications [3 -4] . The new 87 conformation is more similar in structure to RTA-PAPS1 than RTAM-PAP1, in the 88 orientation of the PAP1 moiety (figure 1.a-c). Furthermore, an analysis of RTAM-PAP1 2 nd 89 by I-TASSER showed that this particular conformation might have the ability to bind and 90 depurinate Uracil (Appendix A. 105 moieties are colored in beige, the PAPs moieties in blue to red, N to C, and the FSE in RNA 106 colors (green, red, blue and purple). All the models were viewed in Jena3D. FSE is 107 "neatly" in the main catalytic cleft for RTAM-PAP1 2nd-FSE supported by RTA. 108 109 A hybrid algorithm of template-based and template-free docking was performed for 110 the three compounds, RTA-PAPS1, RTAM-PAP1 and RTAM-PAP1 2 nd , against SARS-111 CoV-2 FSE (Protein-RNA docking) using the HDOCK webserver [7] . The results (table 1) 112 clearly show the gain of function on binding energies for the three chimeras in comparison 113 with the moieties alone. All of the fusion proteins bind SARS-CoV-2 FSE with both 114 moieties, but RTAM-PAP1 2 nd was found to have the highest binding affinity (lowest 115 binding energy) for FSE, which was achieved within the catalytic cleft of the PAP1 moiety 116 and on the putative Uracil biding site. In order to confirm the probability of purine or pyrimidine removal, a protein-126 ligand docking was run using Z-DOCK for the three chimeras against SARS-CoV-2 FSE 127 (figure 1.d-f). The binding conformation of RTAM-PAP1 2 nd to SARS-CoV-2 FSE was again 128 found to lie just within the catalytic cleft of PAP1 moiety. This conformation was found to 129 be extremely stable, with the lowest binding energy (-426 kcal/mol) among the three 130 fusion proteins, and is associated with depurinating activities [3] [4] . 131 The different chimeras have comparable binding energies, but with a noticeable 132 difference for RTAM-PAP1 2 nd . This doesn't come as a surprise since it is a known fact 133 that protein 3D structure determines function, and, thus, any conformational change will 134 incur most of the time a functional change. It is also interesting to note that PAPS1 has a 135 higher binding affinity (-326 kcal/mol) to SARS-CoV-2 FSE than PAP1 (-311 kcal/mol), 136 and, yet, RTA-PAPS1 has the lowest binding affinity (-354 kcal/mol) to SARS-CoV-2 FSE 137 of the chimeras. Given these new results, and the fact that RTA-PAPs were already found 138 to have very high binding affinity towards SARS-CoV-2 key proteins, comparable to that 139 of convalescent COVID-19 patient-origin B38 antibody to SARS-CoV-2 Spike protein [4] , 140 it is safe to assume that RTA-PAPs will have both direct and indirect anti-SARS-CoV-2 141 activities. 142 However, one question remains, will RTA-PAPs be able to enter preferentially 143 infected cells over non-infected ones to exert their antiviral activities, and, thus, have a 144 high Therapeutic Index (TI)? The understood mechanism for the preferential entry into 145 infected cells is during the virus adsorption phase, virions modify their conformation 146 upon receptor binding allowing the insertion of virion components into the cellular 147 membrane, inducing changes in membrane permeability. The viral components open 148 pores inducing a membrane potential drop allowing macromolecules to enter through 149 translocation, pushed through the cell membrane by a proton motive force [8] . 150 Given the structure and function similarities between the different RTA-PAPs, a 151 comparative study of in vitro results and in silico results was performed in order to 152 determine common factors to preferential entry into infected cells and antiviral activity. 153 Results of RTA-PAPS1 antiviral activities in vitro were previously published [3, 5], and 154 RTA-PAPS1 was found to have antiviral activity post-infection in cytoprotection assays 155 against five out of nine viruses tested with a TI greater than 1. (table 2) . The most obvious common structure between those viruses with a strong positive 165 correlation with TI > 1 are viroporins that get integrated into the host cell membrane 166 (Pearson's r(9) = 0.8803, p < .01, calculations in Appendix B.). Viroporins are, for the most 167 part, small viral proteins that oligomerize in the membrane of host cells and modify 168 several cellular functions, including Ca 2+ homeostasis and membrane permeability, i.e. 169 form ion channels. They have been found to be directly linked to viral virulence [9] . 170 Indeed, HBV [10] [11] , HCV [12] , Zika virus [13] , HIV-1 [14] and HCoV-229 [15] all express 171 viroporins. On the other hand, HIV-2 [14] , YFV17D [16] , Poliovirus [17] and DENV2 [18] 172 do not express viroporins, or the strain tested is attenuated with at least one mutation in 173 the viroporin gene (i.e. less virulent strain), or the viroporin does not integrate host cell 174 membrane (i.e. Polioviruses). The YFV-17D, albeit similar to Zika virus, is actually a live 175 attenuated virus used for vaccines with multiple mutations in its viroporin gene among 176 others, for example. Zika MR766 strain does not have mutations in its viroporin gene. 177 These results are very much in line with the current understanding of the preferential 178 entry mechanism, and, actually, confirm the correlation between the presence of 179 viroporins for RTA-PAPS1 to gain preferential entry into infected cells over non-infected 180 cells, post infection. 181 In order to determine if there is a correlation between viral proteins integrated into 182 host cell membranes during and after viral infection and TI of RTA-PAPS1, a knowledge-183 based scoring docking prediction was performed for all the compounds against the viral 184 membrane proteins and viroporins of HBV, HIV-1, HCoV-229 and SARS-CoV-2 (table 3) 185 using CoDockPP global docking. The results show that all the chimeras have comparable high binding affinity toward 195 all the viroporins and increasing binding affinity toward viral membrane proteins with 196 increasing TI against respective viruses. The only exception is the very high binding 197 affinity of RTA-PAPS1 toward wild type HCoV-229E M and E proteins with an in vitro TI 198 of 1.7. Again, this can be explained by many factors, but the fact that the strain tested in 199 vitro is a less virulent one, i.e. mutations in both the M and E proteins, it is conceivable 200 that RTA-PAPS1 is actually very active against wild type HCoV-229E. Albeit there are not 201 enough data to determine a clear correlation between antiviral activity and binding 202 affinity to envelope proteins, those results make sense. Indeed, the high binding affinity 203 towards viral proteins that get integrated into the cell membrane means that RTA-PAPs 204 spend more time interacting with said proteins and, thus, are more inclined to enter cells 205 infected by those particular viruses. The higher rate of entry into infected cells over non-206 infected ones by RTA-PAPs should correlate with higher TI. There are, of course, many 207 other factors that come into play, such as type of antiviral activity (i.e. depurination 208 activity of viral RNA, for example). This mechanism should nonetheless shed light on 209 why RTA-PAPS1 was so active against HBV in vitro (TI of 3333), and, also, help explain 210 the preferential entry mechanisms into infected cells over non infected ones of certain type 211 1 RIPs such as PAPs. Based on the data presented, it would be safe to assume that RTA-212 PAPs will be very active against SARS-CoV-2, its virulent variants, coronaviruses in 213 general, and, any viroporin producing viruses with key proteins RTA-PAPs have high 214 binding affinity to. Indeed, SARS-CoV-2 FSE is highly conserved across a wide range of 215 coronaviruses, a membrane M and E protein are expressed by most coronaviruses, and 216 any viroporin producing virus, that integrates the host cell membrane, should allow 217 preferential entry of RTA-PAPs into infected cells, opening the door to induction of cell 218 apoptosis at the very least. for HBV X, HBVsAG, HCoV-229e M and HCoV-229e E proteins were generated using 245 Phyre2 prediction server [19] only. The 3D structures weren't refined further as it was 246 not needed for the experiments to have a more accurate prediction to be meaningful, 247 and are available in the supplementary files (PDB file S1). The 3D structure of SARS-248 CoV-2 FSE was retrieved from RCSB in PDB format with the PDB ID: 6XRZ. The structures of the Protein-RNA bound complexes were generated by a hybrid 255 algorithm of template-based and template-free docking using HDOCK webserver [7] . 256 Additional models of the RTA-PAPS1, RTAM-PAP1, RTAM-PAP1 2 nd , RTA-PAPS1-FSE, 257 RTAM-PAP1-FSE and RTAM-PAP1 2nd-FSE complexes were generated using ZDOCK 258 [23] without inputting the active residues. All models were viewed using Jena3D 259 (http://jena3d.leibniz-fli.de/). 

Broad-Spectrum Antiviral Strategies and Nucleoside Analogues

Antiviral Activity of Ribosome-Inactivating Proteins

Expression of novel fusion antiviral proteins ricin a chain-pokeweed antiviral 323

RTA-PAPs) in Escherichia coli and their inhibition of protein synthesis and of hepatitis B virus

Novel Binding Mechanisms of Fusion Broad Range Anti-Infective Protein Ricin A 326

Chain Mutant-Pokeweed Antiviral Protein 1 (RTAM-PAP1) against SARS-CoV-2 Key Proteins in Silico

The World Intellectual Property Organization (WIPO)

electron Microscopy and Exploratory Antisense Targeting of the 28-kDa Frameshift Stimulation Element 333 from the SARS-CoV-2 RNA Genome

Modeling Protein-Protein or Protein-DNA/RNA Complexes Using the HDOCK Webserver

Entry of animal viruses and macromolecules into cells

Viroporins: structure and biological functions

A direct role for hepatitis B virus X protein in 342 inducing mitochondrial membrane permeabilization

Hepatitis B Virus X Protein Upregulates Intracellular Calcium Signaling by 345

Binding C-terminal of Orail Protein

Structure and Function of Hepatitis C Virus NS3 Helicase

The Hepatitis C Viruses. Current Topics in Microbiology and Immunology

Inhibitors of the Small Membrane (M) Protein Viroporin Prevent Zika 351

Viroporins: structure and biological functions

Coronavirus envelope protein: current knowledge

The phylogeny of yellow fever virus 17D 357 vaccines

Targeting the Channel Activity of Viroporins

Dengue Virus Serotype 2 and Its Non-Structural

Proteins 2A and 2B Activate NLRP3 Inflammasome. Front Immunol

The Phyre2 web portal for protein modelling, 364 prediction and analysis

interactive server for docking small molecules, peptides and antibodies against potential tar-gets of COVID-367

Specific Protein-Protein Docking

AutoDock Vina: Improving the speed and accu-racy of docking with a new scoring 371 function, efficient optimization and multithreading

Interactive docking 373 prediction of protein-protein complexes and symmetric multimers