key: cord-0789711-5r2hcovx
authors: Garcia-Dorival, Isabel; Ángel Cuesta-Geijo, Miguel; Barrado-Gil, Lucía; Galindo, Inmaculada; Urquiza, Jesús; del Puerto, Ana; Gil, Carmen; Campillo, Nuria; Martínez, Ana; Alonso, Covadonga
title: Identification of NPC1 as a novel SARS-CoV-2 intracellular target
date: 2020-12-20
journal: bioRxiv
DOI: 10.1101/2020.12.19.423584
sha: 59c31e384f5e41c9c6940df71a7d637333bfcadb
doc_id: 789711
cord_uid: 5r2hcovx

Niemann-Pick type C1 (NPC1) receptor is an endosomal membrane protein that regulates intracellular cholesterol trafficking, which is crucial in the Ebola virus (EBOV) cycle. The severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) enters the cell by binding of the viral spike (S) protein to the ACE2 receptor. This requires S-protein processing either by the surface transmembrane serine protease TMPRSS2 for plasma membrane fusion or cathepsin L for endosomal entry. Additional host factors are required for viral fusion at endosomes. Here, we report a novel interaction of the SARS-CoV-2 nucleoprotein (N) with the cholesterol transporter NPC1. Moreover, small molecules interfering with NPC1 that inhibit EBOV entry, also inhibited human coronavirus. Our findings suggest an important role for NPC1 in SARS-CoV-2 infection, a common strategy shared with EBOV, and a potential therapeutic target to fight against COVID-19.

To date, the COVID-19 pandemic has caused over one million of deaths and 32 6 membranes using the Trans-Blot Turbo Transfect Pack (Bio-Rad 1704159) and 144 the Trans-Blot Turbo system (Bio-Rad) . Following this, the transferred 145 membranes were then blocked in 10% skimmed milk powder dissolved in TBS-146 0.1% Tween (TBS-T) (50mM Tris-HCl (pH8.3), 150mM NaCl and 0.5% (v/v) 147

Tween-20) buffer for one hour at room temperature. Primary antibody was diluted 148 1:1000 in blocking buffer and then incubated at 4C overnight. After three 149 washes, blots were incubated with appropriate HRP secondary antibody diluted 150 in blocking buffer at a 1:5000 for 1 hour at room temperature. Blots then were 151 developed using enhanced chemiluminescence reagent (Bio-Rad) and detected 152

with ChemiDoc™ XRS Gel Imaging System using Image Lab™ software . 

The production of SARS-CoV-2 N protein in insect pupae (Tricoplusia ni; T. ni) 172 was performed as previously described (38). Briefly, pupae were allocated in the 173 inoculation robot that dispensed a maximum of 5 μl with the baculovirus titers 174 protein in 5 days pupae incubation time in constant temperature and humidity 175 chambers. After that period, pupae were collected and stored frozen, before 176 downstream processing. T.ni pupae containing the recombinant protein were 7 homogenized in extraction buffer. Then, subsequent steps of clarification, 178 diafiltration and His-tag purification were carried, out in order to obtain purified 179 SARS-CoV-2 N protein. Protein concentration, yield and level of purity were 180 determined by SDS-PAGE analysis using 4-20 % or 12 % Mini-Protean TGX 181 precast gels from Bio-Rad. Gels were stained with QC Colloidal stain (3 ng 182 sensitivity) in the case of concentration and yield evaluation and with SYPRO 183

Ruby (1 ng sensitivity) in the case of level purity analysis, both from Bio-Rad. 184

Recombinant SARS-CoV-2 N protein produced in pupae was measured by band 185 densitometry with the ChemiDoc™ XRS Gel Imaging System using Image Lab™ 186 software (Bio-Rad). A BSA standard curve was used for quantification. 187

High-binding 96-well ELISA plates (Nunc) were coated with 0.5 µg/well of purified 190 SARS-CoV-2 N protein in carbonate/bicarbonate buffer 0.05 M pH 9.6 and 191 allowed to bind over night at 4ºC. Then, endogenous human NPC1 and HSP90 192 were purified using immobilized Recombinant Protein G Resin (Generon) and 4 193 µg of specific antibodies against NPC1 (Abcam, ab108921) or HSP90 (Enzo Life 194 Sciences, ADI-SPA-835) respectively. All steps were performed as described in 195

Co-IP assays. Serial dilutions of these endogenous NPC1 and HSP90 were 196 added to the plate and capture was allowed to proceed for 1 hour at 37ºC. After 197 that, plates were washed with PBST (PBS 0.1%Tween20) and the binding of 198 NPC1 to SARS-CoV-2 N protein was detected with a rabbit anti-NPC1 antibody 199 

All the compounds tested in this work have a purity ≥95% by HPLC. SC 207 compounds were synthesized at Centro de Investigaciones Biológicas (CIB-208 CSIC) following described procedures. All these molecules were included in the 209 MBC chemical library and some of them were previously characterized as 210 potential inhibitors of the protein-protein interaction between NPC1 and EBOV-211 GP (39, 40). The compounds tested in this study are shown in Figure 2 and were 212 resuspended in DMSO at 50 mM. Sulfides SC198 and SC073, and carbazole 213 SC816, were used at working concentrations of 5, 50 and 50 µM; 214 benzothiazepines SC397, SC593, SC567, at working concentrations of 75 µM 215 and SC338 at 100 µM respectively. The first three compounds were shown 216 previously to be active against EBOV while the others were inactive (40). 217

Compounds MBX2254 and MBX2270 were used as gold standards as they have 218 been reported to inhibit EBOV-GP/NPC1 interaction with high selectivity (41) . Absorbance was measured at 490 nm using an ELISA plate reader. 232

Cell viability was reported as the percentage of absorbance in treated cells 233 relative to DMSO-treated cells ( Figure S3 ). The 50% cytotoxic concentration 234 (CC50) was calculated and non-toxic working concentrations (over 80% cell 235 viability) used to test the activities of these compounds on CoV infection. To further validate a specific interaction between EGFP-N and NPC1, two cellular 287 proteins were selected as negative controls. In this case, HSP90 chaperone and 288 endosomal protein EEA1 were used as controls given the abundance of these 289 proteins in cells ( Figure 1C) . 290 291

Co-immunoprecipitations against NPC1 (or reverse pull down) were performed 293 to confirm and further validate the interaction between SARS-CoV-2 N and 294 NPC1. SARS-CoV-2 N was overexpressed in HEK 293T cells and then cellular 295 extracts were analysed by co-immunoprecipitation using protein G-beads and 296 specific monoclonal antibodies against NPC1 ( Figure 1B) . Bound samples 297 obtained from the co-immunoprecipitations were then analysed by western blot, 298 which confirmed the presence of SARS-CoV-2 N ( Figure 1D ). As a result of this 299 interaction, we hypothesized that NPC1 might have an important function in virus 300 biology. 

For an orthogonal characterization of the interaction, we used NPC1 inhibitor 

Inhibitors MBX2254 and MBX2270 ( Figure 2 ) were selected as they target NPC1 325 with high selectivity and both have been described to inhibit HIV-pseudotyped-326 EBOV-GP binding to NPC1 (41) . MBX2254 and MBX2270 were used at 75 and 327 25 µM, respectively. We also used imipramine, a Food and Drug Administration 328 (FDA)-approved drug, that inhibits EBOV and other viruses due to its ability to 329 induce a phenotype similar to NPC1 deficiency (46). 330

Finally, we assayed a set of compounds initially selected by virtual screening of 331 the MBC chemical library in the EBOV-GP/NPC1 interaction (40). These 332 compounds were previously found to inhibit infection with EBOV pseudotyped 333 retrovirus and some of them -sulfides and carbazoles -were able to disturb the 334 NPC1-GP interaction in an ELISA assay. Compounds were classified in three 335 chemical classes, sulfides SC198 and SC073, and carbazole SC816 used at 5, 336 50 and 50 µM respectively; and benzothiazepines SC397, SC593, SC567 ( Figure  337 2), that were used at 75 µM, or 100 µM of SC338. Noteworthy, sulfides and 338 carbazoles were found to potentially act through inhibition of NPC1-GP 339 interaction, while benzothiazepines do not affect this interaction (40). Based on 13 these previous results, the three classes were included in this study for 341 comparative purposes. As a reference, we used U18666A compound (10 µM), 342 known to inhibit cholesterol transport function of NPC1 and the infectious entry of 343 several viruses including EBOV and ASFV (23, 26, 42) . 344

We detected that MBX2270 derivative potently inhibited HCoV infection (50% 345 inhibitory concentration IC50= 3.26 µM, selectivity index 28.36; Figure 3 ). In 346 general, our results yielded significant inhibition >99% of HCoV infection with the 347 U18666A compound and imipramine treatment and with sulfides from the library 348 compounds ( Figure 3A) . Others yielded over 80% of infectivity inhibition (except 349 for SC397 and SC338; Figure 3B ). IC50 was <1 µM in several sulfide compounds 350 

In addition to these functional experiments, we also tested the ability of these 364 compounds to disrupt the NPC1/SARS-CoV-2 N protein interaction in an ELISA 365 assay, as described in Materials and Methods. First, we tested increasing 366 concentrations of NPC1 and control protein HSP90 in plates coated with SARS-367

CoV-2 N protein. We detected a positive reaction with increasing concentrations 368 of NPC1, while negative control HSP90 remained unaltered ( Figure 4A ). Then, 369

we analysed the inhibition of NPC1/SARS-CoV-2 N binding with a sample of the 370 compounds previously described in this study. We obtained a significant inhibition 371 of this specific binding in those samples tested with one inhibitor compound from 372 each class 100 µM SC073 and 50 or 100 µM of MBX2270 ( Figure 4B) . infection by SARS- 11, [31] [32] [33] [34] 48) . 433

A recent study discovered that SARS-CoV-2 non-structural protein 7 (nsp7) 434 strongly interacts with Rab7a, and its depletion causes retention of ACE2 435 receptor inside late endosomes (45). Other reports highlighted the relevance of 436 a variety of proteins involved in cholesterol biosynthesis, including NPC1 infection 437 (32). Also, the cholesterol biosynthesis pathway is downregulated during SARS-438

CoV-2 infection and, according to that, drug treatments that regulate this pathway 439 impact the infection (49). 440

Here, we described in this study an interaction between SARS-CoV-2 N protein 441 and NPC1. This interaction unveiled a novel host-based target for antivirals and 442 a potential host factor for SARS-CoV-2 infectivity. As in other viruses, this 443 interaction could possibly regulate and modify cholesterol efflux from late 444 endosomes and alter the lipid composition in cellular membranes in its own 445 benefit (50). Besides, we presented data on how several compounds that block 446 NPC1 function severely impact 229E HCoV infection in a functional assay, which 447

suggests an essential role for NPC1 in HCoV infectivity. 448

Small molecule inhibitors were crucial to determine that NPC1 was essential for 449 EBOV infection (22). Compounds MBX2254, an aminoacetamide sulfonamide, 450 and MBX2270, a triazole thioether were reported to inhibit EBOV infection with 451 high selectivity (41) . All those compounds have NPC1 as a target and we found 452 that those chemicals strongly inhibited HCoV 229E infection. Also, compounds 453 found using the NPC1/EBOV-GP interaction for the screening of a library of 454 compounds, namely sulfides, carbazoles and benzothiazepines (shown in Figure  455 2), were tested for HCoV inhibition. We have shown that compounds that 456 inhibited EBOV-GP/NPC1 binding, namely sulfides SC073 and SC198 together 457 with carbazole SC816, and not others, presented a potent inhibition of 229E-CoV 458 infection. These chemicals have been shown to inhibit EBOV binding to NPC1 459 and the infection of EBOV-GP pseudovirions elsewhere (40). We have shown 460

here that those compounds that were able to inhibit EBOV-GP/NPC1 binding 461

were also capable to inhibit SARS-CoV-2 N protein/NPC1 binding in an ELISA 462 assay. 

Cell entry mechanisms of SARS-CoV-2

Structural basis of receptor recognition by SARS-CoV-2. 581

SARS coronavirus entry into host cells through a novel clathrin-556 and caveolae-independent endocytic pathway

Cryo-EM Structures of SARS-CoV-2 Spike without and with ACE2

Reveal a pH-Dependent Switch to Mediate Endosomal Positioning of Receptor-559 Binding Domains

Clathrin-dependent entry of severe acute respiratory syndrome 561 coronavirus into target cells expressing ACE2 with the cytoplasmic tail deleted

SARS-CoV-2 cell entry depends on ACE2 and TMPRSS2 564 and is blocked by a clinically proven protease inhibitor

Characterization of spike glycoprotein of SARS-CoV-2 on virus entry 566 and its immune cross-reactivity with SARS-CoV

The spike glycoprotein of the new coronavirus 2019-nCoV 568 contains a furin-like cleavage site absent in CoV of the same clade

Targeting Crucial Host 571 Factors of SARS-CoV-2

SARS-CoV-2 structure and replication characterized by in situ 573 cryo-electron tomography

SARS-coronavirus replication is supported by a reticulovesicular 575 network of modified endoplasmic reticulum

A molecular pore spans the double membrane of the coronavirus 577 replication organelle

Endosomal proteolysis of the Ebola virus glycoprotein is necessary for infection

Calcium ions directly interact with the Ebola virus fusion peptide 582 to promote structure-function changes that enhance infection

HIV-1 evades antibody-mediated neutralization through 584 conformational masking of receptor-binding sites

Structures and mechanisms of viral membrane fusion proteins: multiple 587 variations on a common theme

Ebola virus entry requires the cholesterol transporter 589 Niemann-Pick C1

Small molecule inhibitors reveal Niemann-Pick C1 is essential for 591 Ebola virus infection

593 Deficiency of niemann-pick type C-1 protein impairs release of human 594 immunodeficiency virus type 1 and results in Gag accumulation in late 595 endosomal/lysosomal compartments

Identification of the Niemann-Pick C1-like 1 cholesterol 597 absorption receptor as a new hepatitis C virus entry factor

Hepatitis C virus replication depends on endosomal cholesterol 599 homeostasis

Imipramine inhibits chikungunya virus replication in human skin 601 fibroblasts through interference with intracellular cholesterol trafficking

Aedes aegypti ML 604 and Niemann-Pick type C family members are agonists of dengue virus infection

U18666A, an intra-cellular cholesterol transport inhibitor, inhibits 607 dengue virus entry and replication

Bafilomycin A1 and U18666A efficiently impair ZIKV infection

The role of host cholesterol during flavivirus infection

The lysosome: 613 A potential juncture between SARS-CoV-2 infectivity and Niemann-Pick disease 614 type C, with therapeutic implications

Identification of required host factors for SARS-CoV-2 infection 616 in human cells

Potential COVID-19 therapeutics from a rare disease

Weaponizing lipid dysregulation to combat viral infectivity. jlr

NPC1 as a Modulator of Disease 621 Severity and Viral Entry of SARSCoV-2

Antiviral drugs targeting endosomal membrane proteins inhibit 623 distant animal and human pathogenic viruses

Dendritic cell-specific antigen delivery by 625 coronavirus vaccine vectors induces long-lasting protective antiviral and 626 antitumor immunity

Elucidation of the cellular interactome of Ebola virus 628 nucleoprotein and identification of therapeutic targets

Chrysalises as natural production units for recombinant 630 subunit vaccines

Medicinal and biological chemistry (MBC) library: an 632 efficient source of new hits

Identification of Putative inhibitors of protein-protein Interaction 634 useful to figth against Ebola and other highly pathogenic viruses

Novel small molecule entry inhibitors of Ebola virus

Identification of NPC1 as the target of U18666A, an inhibitor of 639 lysosomal cholesterol export and Ebola infection

Niemann-pick C1 is essential for ebolavirus replication and 641 pathogenesis in vivo

Elucidation of the Ebola virus VP24 cellular interactome 643 and disruption of virus biology through targeted inhibition of host-cell protein 644 function

Comparative host-coronavirus protein interaction networks 646 reveal pan-viral disease mechanisms

Abnormal cholesterol metabolism in imipramine-648 treated fibroblast cultures. Similarities with Niemann-Pick type C disease

Ebola virus entry requires the host-programmed recognition of 651 an intracellular receptor

Coronavirus 653 membrane fusion mechanism offers as a potential target for antiviral 654 development

Modulating the transcriptional landscape of SARS-CoV-2 656 as an effective method for developing antiviral compounds

Kozlov, m. biology, Mechanics of membrane 658 fusion

We are thankful to V. Thiel from the University of Bern, Switzerland for CoV 229E-475 GFP and T. Pietschman, Twincore, Germany for Huh-7 Lunet C3 cells. 476BioRender.com was used to created icons in Figures