Bovine herpesvirus 1 (BHV-1) has received increasing attention for its potential oncolytic applications. BHV-1 recognizes nectin-1 for cell entry via viral glycoprotein D (gD) but represents a low-affinity nectin-1 binding virus. The molecular basis underlying this low receptor-binding affinity, however, remains unknown. Here, the crystal structures of BHV-1 gD in the free and nectin-1-bound forms are presented. While showing an overall resembled nectin-1 binding mode to other alphaherpesvirus gDs, BHV-1 gD has a unique G-strand/α2-helix interloop that disturbs gD/nectin-1 interactions. Residue R188 residing in this loop is observed to otherwise cause strong steric hindrance with the bound receptor, making a large conformational change of the loop a prerequisite for nectin-1 engagement. Subsequently, substitution of R188 with glycine markedly enhances the affinity of the BHV-1-gD/nectin-1 interaction (by about fivefold). These structural and functional data delineate the receptor-recognition basis for BHV-1, which might facilitate BHV-1-based oncolytic design in the future.

Usutu virus (USUV) is a mosquito-born flavivirus that can infect multiple avian and mammalian species. The viral surface envelope (E) protein functions to initiate the viral infection by recognizing cellular receptors and mediating the subsequent membrane fusion, and is therefore a key virulence factor involved in the pathogenesis of USUV. The structural features of USUV-E, however, remains un-investigated thus far.

The DcCoV UAE-HKU23 coronavirus is a newly-found betacoronavirus (betaCoV) that can infect human cells. The viral spike protein plays pivotal roles in mediating receptor-recognition and membrane-fusion, and is therefore a key factor involved in viral pathogenesis and inter-species transmission. Here we reported the structural and functional characterization of the spike N-terminal domain (NTD) from DcCoV UAE-HKU23 (HKU23-NTD). Via mucin-binding assays, we showed that HKU23-NTD is able to bind sugars. We further solved the structure of HKU23-NTD, performed structure-guided mutagenesis and successfully located the potential sugar-binding pockets in the structure. Furthermore, via comparison of available betaCoV NTD structures, we demonstrated that betaCoV NTDs contain a conserved β-sandwich core, but exhibit variant folds in the peripheral elements located in the top-ceiling region and on the lateral side. While showing different compositions and structures, these peripheral elements are topologically equivalent β-sandwich-core insertions, highlighting a divergent evolution process for betaCoVs to form different lineages.

Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) and related sarbecoviruses enter host cells by receptor-recognition and membrane-fusion. An indispensable step in fusion is the formation of 6-helix bundle by viral spike heptad repeats 1 and 2 (HR1 and HR2). Here, we report the construction of 5-helix bundle (5HB) proteins for virus infection inhibition. The optimal construct inhibits SARS-CoV-2 pseudovirus entry with sub-micromolar IC50. Unlike HR2-based peptides that cannot bind spike in the pre-fusion conformation, 5HB features with the capability of binding to pre-fusion spike. Furthermore, 5HB binds viral HR2 at both serological- and endosomal-pH, highlighting its entry-inhibition capacity when SARS-CoV-2 enters via either cell membrane fusion or endosomal route. Finally, we show that 5HB could neutralize S-mediated entry of the predominant SARS-CoV-2 variants and a wide spectrum of sarbecoviruses. These data provide proof-of-concept evidence that 5HB might be developed for the prevention and treatment of SARS-CoV-2 and other emerging sarbecovirus infections.

During an analysis of the virome of bats from Myanmar, a large number of reads were annotated to orthohepadnaviruses. We present the full genome sequence and a morphological analysis of an orthohepadnavirus circulating in bats. This virus is substantially different from currently known members of the genus Orthohepadnavirus and represents a new species.

Pseudorabies virus (PRV) belongs to the Herpesviridae family, and is an important veterinary pathogen. Highly pathogenic PRV variants have caused severe epidemics in China since 2011, causing huge economic losses. To tackle the epidemics, we identified a panel of mouse monoclonal antibodies (mAbs) against PRV glycoprotein B (gB) that effectively block PRV infection. Among these 15 mAbs, fourteen of them block PRV entry in a complement-dependent manner. The remaining one, 1H1 mAb, however can directly neutralize the virus independent of complement and displays broad-spectrum neutralizing activities. We further determined the crystal structure of PRV gB and mapped the epitopes of these antibodies on the structure. Interestingly, all the complement-dependent neutralizing antibodies bind gB at the crown region (domain IV). In contrast, the epitope of 1H1 mAb is located at the bottom of domain I, which includes the fusion loops, indicating 1H1 mAb might neutralize the virus by interfering with the membrane fusion process. Our studies demonstrate that gB contains multiple B-cell epitopes in its crown and base regions and that antibodies targeting different epitopes block virus infection through different mechanisms. These findings would provide important clues for antiviral drug design and vaccine development.

The continuous spread of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) around the world has raised unprecedented challenges to the human society. Antibodies and nanobodies possessing neutralization activity represent promising drug candidates. In this study, we report the identification and characterization of a potent SARS-CoV-2 neutralizing nanobody that targets the viral spike receptor-binding domain (S-RBD). The nanobody, termed as Nb-007, engages SARS-CoV-2 S-RBD with the two-digit picomolar binding affinity and shows outstanding virus entry-inhibition activity. The complex structure of Nb-007 bound to SARS-CoV-2 S-RBD reveals an epitope that is partially overlapping with the binding site for the human receptor of angiotensin-converting enzyme 2 (ACE2). The nanobody therefore exerts neutralization by competing with ACE2 for S-RBD binding, which is further ascertained by our in-vitro biochemical analyses. Finally, we also show that Nb-007 reserves promising, though compromised, neutralization activity against the currently-circulating Delta variant and that fusion of the nanobody with Fc dramatically increases its entry-inhibition capacity. Taken together, these data have paved the way of developing Nb-007 as a drug-reserve for potential treatment of SARS-CoV-2 related diseases.

SARS-CoV-2 recognizes, via its spike receptor-binding domain (S-RBD), human angiotensin-converting enzyme 2 (ACE2) to initiate infection. Ecto-domain protein of ACE2 can therefore function as a decoy. Here we show that mutations of S19W, T27W, and N330Y in ACE2 could individually enhance SARS-CoV-2 S-RBD binding. Y330 could be synergistically combined with either W19 or W27, whereas W19 and W27 are mutually unbeneficial. The structures of SARS-CoV-2 S-RBD bound to the ACE2 mutants reveal that the enhanced binding is mainly contributed by the van der Waals interactions mediated by the aromatic side-chains from W19, W27, and Y330. While Y330 and W19/W27 are distantly located and devoid of any steric interference, W19 and W27 are shown to orient their side-chains toward each other and to cause steric conflicts, explaining their incompatibility. Finally, using pseudotyped SARS-CoV-2 viruses, we demonstrate that these residue substitutions are associated with dramatically improved entry-inhibition efficacy toward both wild-type and antibody-resistant viruses. Taken together, our biochemical and structural data have delineated the basis for the elevated S-RBD binding associated with S19W, T27W, and N330Y mutations in ACE2, paving the way for potential application of these mutants in clinical treatment of COVID-19.

Alongshan virus (ALSV), a newly discovered member of unclassified Flaviviridae family, is able to infect humans. ALSV has a multi-segmented genome organization and is evolutionarily distant from canonical mono-segmented flaviviruses. The virus-encoded methyltransferase (MTase) plays an important role in viral replication. Here we show that ALSV MTase readily binds S-adenosyl-L-methionine (SAM) and S-adenosyl-L-homocysteine (SAH) but exhibits significantly lower affinities than canonical flaviviral MTases. Structures of ALSV MTase in the free and SAM/SAH-bound forms reveal that the viral enzyme possesses a unique loop-element lining side-wall of the SAM/SAH-binding pocket. While the equivalent loop in flaviviral MTases half-covers SAM/SAH, contributing multiple hydrogen-bond interactions; the pocket-lining loop of ALSV MTase is of short-length and high-flexibility, devoid of any physical contacts with SAM/SAH. Subsequent mutagenesis data further corroborate such structural difference affecting SAM/SAH-binding. Finally, we also report the structure of ALSV MTase bound with sinefungin, an SAM-analogue MTase inhibitor. These data have delineated the basis for the low-affinity interaction between ALSV MTase and SAM/SAH and should inform on antiviral drug design.

Rabies virus (RABV), the etiological agent for the lethal disease of rabies, is a deadly zoonotic pathogen. The RABV glycoprotein (RABV-G) is a key factor mediating virus entry and the major target of neutralizing antibodies. Here, we report the crystal structures of RABV-G solved in the free form at ∼pH-8.0 and in the complex form with a neutralizing antibody 523-11 at ∼pH-6.5, respectively. RABV-G has three domains, and the basic-to-acidic pH change results in large domain re-orientations and concomitant domain-linker re-constructions, switching it from a bent hairpin conformation into an extended conformation. During such low-pH-induced structural transitions, residues located in the domain-linker are found to play important roles in glycoprotein-mediated membrane fusion. Finally, the antibody interacts with RABV-G mainly through its heavy chain and binds to a bipartite conformational epitope in the viral protein for neutralization. These structures provide valuable information for vaccine and drug design.

The receptor-binding domain (RBD) of spike recognizing the receptor angiotensin-converting enzyme 2 (ACE2) initiates membrane fusion between severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) and cell membrane. Although the structure of the RBD_ACE2 complex has been well studied, its functional mechanism in membrane fusion is still not fully understood. Here, using an in vitro cell-vesicle content-mixing assay, it is found that the cleavage at the S2' site by thrombin (Thr) protease strongly accelerates membrane fusion, compared to that of cleavage at the S1/S2 site by PreScission (3C) protease. Moreover, mutations at the RBD_ACE2 interface resulted in a positive correlation between binding affinity and fusion probability. In both the cell-vesicle and cell-cell fusion assays, by crosslinking two membranes via the neutravidin (NTV)_biotin interaction or complementary DNA strands, it is found that spike drives membrane fusion in the absence of ACE2, and a suitable distance between two membranes is critical for spike-mediated membrane fusion. Finally, unsuitable membrane crosslinkers significantly inhibited the fusion probability in the presence of ACE2. Taken together, the results suggest that the RBD_ACE2 complex may act as a crosslinker to bridge the viral and cell membranes at a suitable distance, which is critical, but also substitutable for spike-mediated SARS-CoV-2 entry.

Bats are reservoir animals harboring many important pathogenic viruses and with the capability of transmitting these to humans and other animals. To establish an effective surveillance to monitor transboundary spread of bat viruses between Myanmar and China, complete organs from the thorax and abdomen from 853 bats of six species from two Myanmar counties close to Yunnan province, China, were collected and tested for their virome through metagenomics by Solexa sequencing and bioinformatic analysis. In total, 3,742,314 reads of 114 bases were generated, and over 86% were assembled into 1,649,512 contigs with an average length of 114 bp, of which 26,698 (2%) contigs were recognizable viral sequences belonging to 24 viral families. Of the viral contigs 45% (12,086/26,698) were related to vertebrate viruses, 28% (7,443/26,698) to insect viruses, 27% (7,074/26,698) to phages and 95 contigs to plant viruses. The metagenomic results were confirmed by PCR of selected viruses in all bat samples followed by phylogenetic analysis, which has led to the discovery of some novel bat viruses of the genera Mamastrovirus, Bocavirus, Circovirus, Iflavirus and Orthohepadnavirus and to their prevalence rates in two bat species. In conclusion, the present study aims to present the bat virome in Myanmar, and the results obtained further expand the spectrum of viruses harbored by bats.

The T3SS chaperone CesT is recently shown to interact with the post-transcriptional regulator CsrA to modulate post-attachment signaling in enteropathogenic and enterohemorrhagic Escherichia coli. The molecular basis of the CesT/CsrA binding, however, remains elusive. Here, we show that CesT and CsrA both created two ligand binding sites in their homodimers, forming irregular multimeric complexes in solution. Through construction of a recombinant CsrA-dimer (Re-CsrA) that contains a single CesT binding site, the atomic binding features between CesT and CsrA are delineated via the structure of the CesT/Re-CsrA complex. In contrast to a previously reported N-terminally swapped dimer-form, CesT adopts a dimeric architecture with a swapped C-terminal helix for CsrA engagement. In CsrA, CesT binds to a surface patch that extensively overlaps with its mRNA binding site. The binding mode therefore justifies a mechanism of CsrA-modulation by CesT via competitive inhibition of the CsrA/mRNA interactions.

The emergence of SARS-CoV-2 infection has posed unprecedented threat to global public health. The virus-encoded non-structural protein 14 (nsp14) is a bi-functional enzyme consisting of an exoribonuclease (ExoN) domain and a methyltransferase (MTase) domain and plays a pivotal role in viral replication. Here, we report the structure of SARS-CoV-2 nsp14-ExoN domain bound to its co-factor nsp10 and show that, compared to the SARS-CoV nsp10/nsp14-full-length complex, SARS-CoV-2 nsp14-ExoN retains an integral exoribonuclease fold and preserves an active configuration in the catalytic center. Analysis of the nsp10/nsp14-ExoN interface reveals a footprint in nsp10 extensively overlapping with that observed in the nsp10/nsp16 structure. A marked difference in the co-factor when engaging nsp14 and nsp16 lies in helix-α1', which is further experimentally ascertained to be involved in nsp14-binding but not in nsp16-engagement. Finally, we also show that nsp10/nsp14-ExoN is enzymatically active despite the absence of nsp14-MTase domain. These data demonstrate that SARS-CoV-2 nsp10/nsp14-ExoN functions as an exoribonuclease with both structural and functional integrity.