key: cord-0329167-4i0t6xy6
authors: Xiong, Qing; Cao, Lei; Ma, Chengbao; Liu, Chen; Si, Junyu; Liu, Peng; Gu, Mengxue; Wang, Chunli; Shi, Lulu; Tong, Fei; Huang, Meiling; Li, Jing; Zhao, Chufeng; Shen, Chao; Chen, Yu; Zhao, Huabin; Lan, Ke; Wang, Xiangxi; Yan, Huan
title: Close relatives of MERS-CoV in bats use ACE2 as their functional receptors
date: 2022-01-25
journal: bioRxiv
DOI: 10.1101/2022.01.24.477490
sha: 449660a1eb37bdf90eb98aa0ee2a2afb57cd1f87
doc_id: 329167
cord_uid: 4i0t6xy6

Middle East Respiratory Syndrome coronavirus (MERS-CoV) and several bat coronaviruses employ Dipeptidyl peptidase-4 (DPP4) as their functional receptors1–4. However, the receptor for NeoCoV, the closest MERS-CoV relative yet discovered in bats, remains enigmatic5. In this study, we unexpectedly found that NeoCoV and its close relative, PDF-2180-CoV, can efficiently use some types of bat Angiotensin-converting enzyme 2 (ACE2) and, less favorably, human ACE2 for entry. The two viruses use their spikes’ S1 subunit carboxyl-terminal domains (S1-CTD) for high-affinity and species-specific ACE2 binding. Cryo-electron microscopy analysis revealed a novel coronavirus-ACE2 binding interface and a protein-glycan interaction, distinct from other known ACE2-using viruses. We identified a molecular determinant close to the viral binding interface that restricts human ACE2 from supporting NeoCoV infection, especially around residue Asp338. Conversely, NeoCoV efficiently infects human ACE2 expressing cells after a T510F mutation on the receptor-binding motif (RBM). Notably, the infection could not be cross-neutralized by antibodies targeting SARS-CoV-2 or MERS-CoV. Our study demonstrates the first case of ACE2 usage in MERS-related viruses, shedding light on a potential bio-safety threat of the human emergence of an ACE2 using “MERS-CoV-2” with both high fatality and transmission rate.

coronavirus-ACE2 binding interface and a protein-glycan interaction, distinct from other known 23 ACE2-using viruses. We identified a molecular determinant close to the viral binding interface that 24 restricts human ACE2 from supporting NeoCoV infection, especially around residue Asp338. 25 Conversely, NeoCoV efficiently infects human ACE2 expressing cells after a T510F mutation on the 26 receptor-binding motif (RBM). Notably, the infection could not be cross-neutralized by antibodies 27 targeting SARS-CoV-2 or MERS-CoV. Our study demonstrates the first case of ACE2 usage in 28 MERS-related viruses, shedding light on a potential bio-safety threat of the human emergence of an 29 Introduction 34 Coronaviruses (CoVs) are a large family of enveloped positive-strand RNA viruses classified into 35 four genera: Alpha-, Beta-, Gamma-and Delta-CoV. Generally, Alpha and Beta-CoV can infect 36 mammals such as bats and humans, while Gamma-and Delta-CoV mainly infect birds, occasionally 37 mammals 6-8 . It is thought that the origins of most coronaviruses infecting humans can be traced back 38 to their close relatives in bats, the most important animal reservoir of mammalian coronaviruses Specific receptor recognition of coronaviruses is usually determined by the receptor-binding 48 domains (RBDs) on the carboxyl-terminus of the S1 subunit (S1-CTD) of the spike proteins 17 . 49 Among the four well-characterized coronavirus receptors, three are S1-CTD binding ectopeptidases, 50 including ACE2, DPP4, and aminopeptidase N (APN) 1,18,19 . By contrast, the fourth receptor, 51 antigen-related cell adhesion molecule 1(CEACAM1a), interacts with the amino-terminal domain 52 (NTD) of the spike S1 subunit of the murine hepatitis virus 20,21 . Interestingly, the same receptor can 53 be shared by distantly related coronaviruses with structurally distinct RBDs. For example, the 54 NL63-CoV (an alpha-CoV) uses ACE2 as an entry receptor widely used by many sarbecoviruses 55 (beta-CoV linage B) 22 . A similar phenotype of cross-genera receptor usage has also been found in 56 APN, which is shared by many alpha-CoVs and a delta-CoV (PDCoV) 7 . In comparison, DPP4 usage 57 has only been found in merbecoviruses (beta-CoV linage C) such as HKU4, HKU25, and related 58 strains 2-4 . 59 4 rendering them taxonomically the same viral species 27, 29 . However, their S1 subunits are highly 67 divergent compared with MERS-CoV (around 43-45% amino acid similarity), in agreement with 68 their different receptor preference 23 . 69 In this study, we unexpectedly found that NeoCoV and PDF-2180-CoV use bat ACE2 as their 70 functional receptor. The cryo-EM structure of NeoCoV RBD bound with the ACE2 protein 71 from Pipistrellus pipistrellus revealed a novel ACE2 interaction mode that is distinct from how 

Evidence of ACE2 usage 79 To shed light on the relationship between merbecoviruses, especially NeoCoV and PDF-2180-CoV, 80 we conducted a phylogenetic analysis of the sequences of a list of human and animal coronaviruses. 81 Maximum likelihood phylogenetic reconstructions based on complete genome sequences showed 82 that NeoCoV and PDF-2180-CoV formed sister clade with MERS-CoV (Fig. 1a) . In comparison, the 83 phylogenetic tree based on amino acid sequences of the S1 subunit demonstrated that NeoCoV and 84 PDF-2180-CoV showed a divergent relationship with MERS-CoV but are closely related to the 85 hedgehog coronaviruses (EriCoVs) (Fig. 1b) . A sequence similarity plot analysis (Simplot) queried 86 by MERS-CoV highlighted a more divergent region encoding S1 for NeoCoV and PDF-2180-CoV 87 compared with HKU4-CoV (Fig. 1c) . We first tested whether human DPP4 (hDPP4) could support 88 the infection of several merbecoviruses through a pseudovirus entry assay 30 . The result revealed that 89 only MERS-CoV and HKU4-CoV showed significantly enhanced infection of 293T-hDPP4. 90 Unexpectedly, we detected a significant increase of entry of NeoCoV and PDF-2180-CoV in 91 293T-hACE2 but not 293T-hAPN, both of which are initially set up as negative controls (Fig. 1d Bat37ACE2 triggers more efficient cell-cell membrane fusion than hACE2 in the presence of 103 NeoCoV spike protein expression ( Fig. 1g-h) . Notably, the failure of the human or hedgehog ACE2 104 to support entry of EriCoV-HKU31 indicates that these viruses have a different receptor 105 usage (Extended Data Fig.6 ). In agreement with a previous study 23,28 , our results against the 106 possibility that bat DPP4 act as a receptor for NeoCoV and PDF-2180-CoV, as none of the tested 107 DPP4 orthologs, from the vesper bats whose ACE2 are highly efficient in supporting vial entry, 108 could support a detectable entry of NeoCoV and PDF-2180-CoV (Fig. 1i, Extended Data Fig.7) . 109 Infection assays were also conducted using several other cell types from different species, including 110 a bat cell line Tb 1 Lu, ectopically expressing ACE2 or DPP4 from Bat40 (Antrozous pallidus), and 111 each test yielded similar results (Extended Data Fig.8 ).

112 113 S1-CTD mediated species-specific binding 114 The inability of NeoCoV and PDF-2180-CoV to use DPP4 is consistent with their highly divergent 115 S1-CTD sequence compared with the MERS-CoV and HKU4-CoV. We produced S1-CTD-hFc 116 proteins (putative RBD fused to human IgG Fc domain) to verify whether their S1-CTDs are 117 responsible for ACE2 receptor binding. The live-cell binding assay based on cells expressing various 118 bat ACE2 showed a species-specific utilization pattern in agreement with the results of the 119 pseudovirus entry assays (Fig. 2a) . The specific binding of several representative bat ACE2 was also 120 verified by flow-cytometry (Fig. 2b) . We further determined the binding affinity by Bio-Layer 121 Interferometry (BLI) analysis. The results indicated that both viruses bind to the ACE2 122 from Pipistrellus pipistrellus (Bat37) with the highest affinity (KD=1.98nM for NeoCoV and 1.29 123 nM for PDF-2180-CoV). In contrast, their affinities for hACE2 were below the detection limit of our 124 BLI analysis (Fig. 2c , Extended Data Fig.9 ). An enzyme-linked immunosorbent assay (ELISA) also 125 7 infections of cells expressing Bat37ACE2 (Fig. 2g) . We further demonstrated the pivotal role of 133 S1-CTD in receptor usage by constructing chimeric viruses and testing them for altered receptor 134 usage. As expected, batACE2 usage was changed to hDPP4 usage for a chimeric NeoCoV with CTD, 135 but not NTD, sequences replaced by its MERS-CoV counterpart (Fig. 2h) . These results confirmed 136 that S1-CTD of NeoCoV and PDF-2180-CoV are RBDs for their species-specific interaction with 137 ACE2. Fig. 14) . 151 Like other structures of homologs, the NeoCoV RBD structure comprises a core subdomain 152 located far away from the engaging ACE2 and an external subdomain recognizing the receptor 153 ( Fig. 3c and Extended Data Fig. 15 ). The external subdomain is a strand-enriched structure with 8 four anti-parallel β strands (β6-β9) and exposes a flat four-stranded sheet-tip for ACE2 155 engagement (Fig. 3c ). By contrast, the MERS-CoV RBD recognizes the side surface of the DPP4 156 β-propeller via its four-stranded sheet-blade (Fig. 3c) . The structural basis for the differences in Table. 2). Notably, the methyl group from residues A509 and T510 of the NeoCoV 173 RBM are partially involved in forming a hydrophobic pocket with residues F308, W328, L333, 174 and I358 from Bat37ACE2 at the interface. A substitution of T510 with F in the PDF-2180-CoV 175 RBM further improves hydrophobic interactions, which is consistent with an increased binding 9 affinity observed for this point mutation (Figs. 3d, Extended Data Table. 2). Apart from 177 protein-protein contacts, the glycans of bat ACE2 at positions N54 and N329 sandwich the strands 178 (β8-β9), forming π-π interactions with W540 and hydrogen bonds with N532, G545, and R550 179 from the NeoCoV RBD, underpinning virus-receptor associations ( Fig. 3d and Extended Data 180 Table. 2). 181 The critical residues were verified by introducing mutations and testing their effect on receptor Evaluation of zoonotic potential 192 A major concern is whether NeoCoV and PDF-2180-CoV could jump the species barrier and 193 infect humans. As mentioned above, NeoCoV and PDF-2180-CoV cannot efficiently interact with 194 human ACE2. Here we first examined the molecular determinants restricting hACE2 from 195 supporting the entry of these viruses. By comparing the binding interface of the other three 196 hACE2-using coronaviruses, we found that the SARS-CoV, SARS-COV-2, and NL63 share similar 197 interaction regions that barely overlapped with the region engaged by NeoCoV (Fig. 4a) . Analysis of 198 the overlapped binding interfaces reveals a commonly used hot spot around residues 329-330 199 (Fig.4b) . Through sequences alignment and structural analysis of hACE2 and Bat37ACE2, we 200 predicted that the inefficient use of the hACE2 for entry by the viruses could be attributed to 201 incompatible residues located around the binding interfaces, especially the difference in sequences 202 between residues 337-342 (Fig.4c) . We replaced these residues of hACE2 with those from the 203 Bat37ACE2 counterparts to test this hypothesis ( Fig.4c- 

The positive ratio was indicated based on the threshold (dash line). c, BLI assays 604 analyzing the binding kinetics between NeoCoV-S1-CTD-hFc with selected ACE2-ecto proteins. d, 605 ELISA assay showing the binding efficiency of NeoCoV and PDF-2180-CoV S1-CTD to human and 606

The inhibitory activity of soluble ACE2-ecto proteins against NeoCoV 607 infection in 293T-Bat37ACE2. f, Dose-dependent competition of NeoCoV infection by 608

Bat37ACE2-ecto proteins in 293T-Bat37ACE2 cells. g, The inhibitory effect of NeoCoV

PDF-2180-CoV S1-CTD-hFc and MERS-CoV RBD-hFC proteins on NeoCoV infection in

Receptor preference of chimeric viruses with S1-CTD or S1-NTD swap 611 mutations in cells expressing the indicated receptors

PDF-2018CoV RBD-Bat37ACE2 complex (b). The NeoCoV RBD

PDF-2180-CoV RBD, and Bat37ACE2 were colored by red, yellow, and cyan, respectively. c, 618 Structure comparison between NeoCoV RBD-Bat37ACE2 complex (left) and MERS-CoV

The NeoCoV RBD, MERS-CoV RBD

light yellow, gray, cyan, and blue, 621 respectively. d, Details of the NeoCOV RBD-Bat37ACE2 complex interface. All structures are 622 shown as ribbon with the key residues shown with sticks. The salt bridges and hydrogen bonds are 623 presented as red and yellow dashed lines

NeoCoV RBD affecting viral binding (e), and entry efficiency (f) in 293T-Bat37ACE2 cells. g-h, 625

Verification of the critical residues on Bat37ACE2 affecting NeoCoV RBD binding (g), and viral 626 entry efficiency(h)

NeoCoV RBD were colored in purple, light purple, green, and red, respectively. b, A common 632 virus-binding hot spot on ACE2 for the four viruses

The expression level of the hACE2 mutants by Western blot (d) 635 and immunofluorescence (e). f-g, Receptor function of hACE2 mutants evaluated by virus RBD 636 binding assay (f) and pseudovirus entry assay (g). h, Molecular dynamics (MD) analysis of the effect 637 of critical residue variations on the interaction between NeoCoV

The NeoCoV 639 RBD and hACE2 were colored in red and sky blue, respectively. Details of the NeoCoV RBD key 640 mutation T510F was shown. All structures are presented as ribbon with the key residues shown with 641 sticks. j-k, The effect of NeoCoV and PDF-2180-CoV RBM mutations on hACE2 fitness as 642 demonstrated by binding (j) and entry efficiency (k) on 293T-hACE2 and 293T-Bat37ACE2 cells. l, 643 hACE2 dependent entry of NeoCoV-T510F

Neutralizing activity of SARS-CoV-2 vaccinated sera against the 645 infection by SARS-CoV-2, NeoCoV, and PDF-2180-CoV. n, Neutralizing activity of MERS-RBD 646 targeting nanobodies against the infections by MERS-CoV, NeoCoV, and PDF-2180-CoV. Mean± 647 SD for g,k-n

bat-to-human transmission of MERS coronavirus

Receptor Usage of a Novel Bat Lineage C Betacoronavirus Reveals Evolution 762 of Middle East Respiratory Syndrome-Related Coronavirus Spike Proteins for Human 763

Dipeptidyl Peptidase 4 Binding

Discovery of Novel Bat Coronaviruses in South China That Use the Same 765 Receptor as Middle East Respiratory Syndrome Coronavirus

Close relative of human Middle East respiratory syndrome coronavirus in 768 bat

Origin and evolution of pathogenic coronaviruses

Broad receptor engagement of an emerging global coronavirus may potentiate 772 its diverse cross-species transmissibility

Independent infections of porcine deltacoronavirus among Haitian 775 children

Origin and cross-species transmission of bat coronaviruses in China

Global Epidemiology of Bat Coronaviruses. 779 Viruses

A novel coronavirus associated with severe acute respiratory syndrome

Isolation 783 of a novel coronavirus from a man with pneumonia in Saudi Arabia

A pneumonia outbreak associated with a new coronavirus of probable bat 786 origin

RNA based mNGS approach identifies a novel human coronavirus from two 788 individual pneumonia cases in 2019 Wuhan outbreak

the carcinoembryonic antigen family of glycoproteins

The N-terminal domain of the murine 805 coronavirus spike glycoprotein determines the CEACAM1 receptor specificity of the virus 806 strain

Human coronavirus NL63 employs the severe acute respiratory syndrome 808 coronavirus receptor for cellular entry

Further Evidence for Bats as the Evolutionary Source of Middle East 811 Respiratory Syndrome Coronavirus

Characterization of a novel betacoronavirus related to middle East 813 respiratory syndrome coronavirus in European hedgehogs

Identification of a Novel Betacoronavirus (Merbecovirus) in Amur 816 Hedgehogs from China

Rooting the phylogenetic tree of middle East respiratory syndrome 818 coronavirus by characterization of a conspecific virus from an African bat

A metagenomic viral discovery approach identifies potential zoonotic 821 and novel mammalian viruses in Neoromicia bats within South Africa

Trypsin Treatment Unlocks Barrier for Zoonotic Bat Coronavirus 824 Infection

Aerosol stability of infectious 826 and potentially infectious reovirus particles

Generation of VSV pseudotypes using recombinant DeltaG-VSV for studies on 829 virus entry, identification of entry inhibitors, and immune responses to vaccines

ACE2 receptor usage reveals variation in susceptibility to SARS-CoV and 832 SARS-CoV-2 infection among bat species

higher compared to SARS coronavirus

The reproductive number of the Delta variant of SARS-CoV-2 is far higher 848 compared to the ancestral SARS-CoV-2 virus

Relative Reproduction Number of SARS-CoV-2 Omicron (B.1.1.529)

Compared with Delta Variant in South Africa

Establishing a genetic recombination map 854 for murine coronavirus strain A59 complementation groups

A clustering of RNA recombination sites adjacent to a 857 hypervariable region of the peplomer gene of murine coronavirus

MERS coronaviruses in dromedary camels

SARS-CoV-2 variants, spike mutations and immune escape

Heavily mutated Omicron variant puts scientists on alert

Comparison of vesicular stomatitis virus pseudotyped with 866 the S proteins from a porcine and a human coronavirus

envelope-mediated membrane fusion using modified split green fluorescent proteins

RELION: implementation of a Bayesian approach to cryo-EM structure 872 determination

Tools for macromolecular model building and refinement into electron 874 cryo-microscopy reconstructions

Towards automated crystallographic structure refinement with 877 phenix.refine