SENSING SWEET SPOTS: A comprehensive study of SARS-CoV-2-binding lectins

doi:10.21203/rs.3.rs-5022883/v1

Download PDF

Article

SENSING SWEET SPOTS: A comprehensive study of SARS-CoV-2-binding lectins

https://doi.org/10.21203/rs.3.rs-5022883/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

The Spike glycoprotein of SARS-CoV-2 enables receptor binding, membrane fusion, and viral entry, and is the major target for vaccines and therapeutics. Spike is heavily glycosylated, and N-glycosylation is critical for the open conformation required for ACE2 binding. ACE2 binding and subsequent viral fusion are blocked by inhibitors of N-glycan biosynthesis and by a few lectins. Here, we performed a comprehensive study of lectins that bind to recombinant SARS-CoV-2 Spike glycoprotein (expressed and purified from HEK293F cells) and to SARS-CoV-2 virus (cultured from Vero E6 cells) by employing an array of 95 lectins with different carbohydrate specificities, 68 of which we characterized using publically available glycan array data and the MotifFinder software. Spike glycoprotein was bound by N-glycan core binding lectins, AMA, ASA, BANLEC, BC2LA, CA, ConA, GNA, GS-II, HHA, LcHA, MNA-M, NPA, ORYSATA, PHA-L, and TL, and O-glycan binding lectins, ACL, SBA, ABL, LSN-L, and RCA-60. Spike glycoprotein was also bound by L-fucose-specific lectins, PSA, RS-Fuc, and AAL (in an L-fucose-dependent manner). A subset of the N-glycan core binding lectins and other lectins that bound to recombinant Spike - AMA, ASA, ConA, GNA, NPA, BANLEC, and TL, and AAL, RCA-120, SBA, and PHA-P, respectively, also bound well to cultured SARS-CoV-2 virus. The lectins, CPA, MAA, PA-IIL, and STL bound with greater intensity to the cultured SARS-CoV-2 virus than the recombinant Spike glycoprotein, suggestive of more accessible complex glycans on the cultured virus. The SARS-CoV-2-specific lectins identified in our study may be assessed for their antiviral potential in future studies.

Biological sciences/Biochemistry/Glycobiology

Biological sciences/Biochemistry/Glycomics

SARS-CoV-2

COVID-19

lectins

Spike

glycosylation

lectin array

O-glycans

The Spike glycoprotein of the SARS-CoV-2 virus, the infectious agent of the coronavirus disease 2019 (COVID-19) pandemic, has emerged as a pivotal focal point in the scientific understanding of the virus and its pathogenicity. This trimeric glycoprotein, abundantly displayed on the viral envelope, plays a critical role in the initial stages of infection by facilitating viral entry into host cells by binding to the ACE2 receptor [1-4].

Structurally, the monomer Spike protein comprises two distinct subunits, the S1 subunit (14-685 aa) and the S2 subunit (686-1273 aa), linked with a furin cleavage site and a transmembrane protein serine protease 2 (TMPRSS2) site [5-7]. The S1 subunit, housing the N-terminal domain and a receptor-binding domain (RBD), is responsible for recognizing and binding to the host cell receptor ACE2, a crucial step in viral attachment. The S2 subunit, on the other hand, contains a fusion peptide, central helix, connected domain, heptad repeat domain, transmembrane domain, and a cytoplasmic tail which orchestrates membrane fusion, enabling the virus to enter the host cell [2,7]. The trimeric Spike protein is heavily glycosylated and harbors 66 N-linked glycosylation sites (22 sites per monomer) and variable O-glycan sites [8,9].

Besides providing structural stability, glycosylation provides viruses a shield to escape the host's immune response and maintain infection [10-12], Spike glycosylation of SARS-CoV-2 accounts for 17% of molecular weight and 40% of the protein surface [13]. Unlike HIV, which encompasses more high-mannose-type N glycans than complex-type N-glycans, the SARS-CoV-2 and SARS-CoV viruses possess a higher percentage of complex-type N glycans (with sialylation and fucosylation) than oligomannose glycans [11,14-16]. Around 79% of glycans are complex-type glycans having bi-antennary, tri-antennary, and tetra-antennary with or without bisecting GalNAc, and around 21% of N-glycosylation sites have high mannose or hybrid content in the S1 subunit of Spike protein isolated from SARS-CoV-2 infected Calu3 cells [17]. N-glycosylation sites are conserved in all variants of concern except for the addition of the N188 site in the Gamma variant and the removal of the N17 site in the Delta variant [18] reflecting the role of glycosylation in viral binding, infectivity, and transmission. The glycosylation sites, N331 and N343, are present in the RBD region and hence crucial for viral infectivity, whereas the N165 and N234 glycosylation sites present adjacent to the RBD region are responsible for the conformational change of RBD from the closed to the open conformation, which facilitates the receptor engagement to the ACE2 [19,20]. Thus, the extensive glycosylation of Spike protein influences its structural stability, immune recognition, and receptor binding affinity, which are fundamental to understanding SARS-CoV-2 infectivity, immune evasion, and the development of therapeutic interventions [21-23]. While the glycan shield presents a formidable challenge in vaccine and therapeutic development by concealing critical epitopes, it also offers opportunities [24,25]. By dissecting the shield's composition and function, researchers can design strategies to target conserved regions that are less shielded.

Lectins, non-immune glycoproteins, function as natural immune molecules that can bind to specific glycans on the surface of pathogens, including viruses, and can hinder viral infection either by blocking the viral attachment or interfering with viral entry at an early stage of infection. Their ability to target essential steps in the viral life cycle makes them promising candidates for prophylactic and therapeutic interventions against viral infections [26,27]. Several lectins such as Griffithsin [28], banana lectin [29] Wheat Germ Agglutinin (WGA) [30] Pholiota squarrosa lectin (PhoSL) [31] Lectin FRIL [32] UDA [33], and lentil lectin [34] have been shown to have anti-viral activity. Here, we have used a commercial lectin array to screen 95 lectins from different sources of origin and with different glycan binding profiles for their potency to bind to recombinant Spike and cultured SARS-CoV-2 virus (Figure 1a). We identify several lectins that may be explored for their diagnostic and therapeutic potential in future studies.

Analysis of glycan-specificities of lectins using glycan array data:

Building Lectin Models: We analyzed 68 lectins (out of the 95 lectins on the lectin array, for which glycan array datasets were available publicly) for their glycan-binding specificities. Glycan array datasheets of each lectin at different concentrations, from different protein sources, and from different data sources were put together by arranging the files “AllBindingData.csv, AllDataInfo.csv, and AllGlycanInfo.csv” downloaded from the API of the CarboGrove database (https://carbogrove.org/api/home.php). Downloaded datasheets were named Lectinname_Proteinsource_concentration_datasource_DATA and used to build an automated lectin model using MotifFinder v2 2.5 [35-37] at default settings with the default glycan motif list. The lectin model provided as output, the specificities and fine-specificities of the lectin to different glycan motifs.

Generation of glycans: Four different glycan lists containing N-core glycans, O-core glycans, N-extension glycans, and O-extension glycan, were generated as follows. The N-core glycan list was generated manually from N-glycan biosynthesis principles [38] and contained 24 glycans, which included high mannose glycans, complex glycans, and hybrid glycans, with options of core-fucose, bisecting GlcNAc, and GlcNAc-substitution on Mannose. The O-core glycan list contained 10 glycans, i.e., core 1 to 8, Tn antigen, and sTn antigen, and was generated manually [39,40]. The N-extension glycan list contained 144 glycans with different extensions such as LeA, LeB, LeX, LeY, and others, and was generated using MotifFinder on just the biantennary glycan scaffold. The O-extension glycan list contained 121 glycans and was generated manually.

Predicting the propensity of lectins to bind to the generated glycans: The lectin models (built using MotifFinder) were used as input to predict each lectin's propensity to bind the list of generated glycans (N-core glycan, O-core glycans, N-glycan extension sequences, O-glycan extension sequences) using the “Predict Model” tool of MotifFinder software. The output obtained was binding intensities of each lectin to N-core glycan, O-core glycans, N-glycan extension sequences, and O-glycan extension sequences, and this was used to generate a binding heat map (http://www.heatmapper.ca/expression/) by using Average linkage clustering and Euclidean distance measurement methods. Clustering was applied to rows as well as columns.

Chemicals and Reagents:

The Lectin array-95 kit was purchased from Ray Biotech. Glycoprofile β-Elimination kit, monosaccharides, and bovine Serum albumin (BSA) were purchased from Sigma. Zeba spin desalting columns were purchased from Thermo Scientific, Plasmid Modified pαH Vector Containing the SARS-Related Coronavirus 2, Wuhan-Hu-1 HexaPro Spike Glycoprotein Ectodomain (NR-53587) from BEI resources. PAA-Biotin, PAA-Biotin-Mannose, PAA-Biotin-L-Fucose, and PAA-Biotin-Sialic acid were purchased from Glycotech.

Expression and purification of recombinant Spike glycoprotein:

The SARS-CoV-2 Wuhan-Hu-1 HexaPro Spike Glycoprotein Ectodomain was produced in Expi293F™ cells. The cells were transiently transfected with Modified pαH Vector Containing the SARS-Related Coronavirus 2, Wuhan-Hu-1 HexaPro Spike Glycoprotein Ectodomain, NR-53587 (BEI) by polyethyleneimine at the density of ~3 x 10⁶cells/ml and the culture supernatant was harvested after four days post-transfection. The supernatant was spun at 4000 rpm followed by filtration through 0.22 µm filter to remove cell debris. The culture supernatant was passed through the nickel ion affinity resin (Thermo Scientific) at 4 °C and bound protein was eluted from the Ni-NTA column using 500 mM imidazole in PBS buffer at pH 7.4. The eluted protein was extensively dialyzed against phosphate-buffered saline using a 10 kDa (MWCO) dialysis membrane. The purity of purified protein samples was assessed by SDS-PAGE.

Beta-elimination of Spike glycoprotein:

Spike glycoprotein (30 μg) was dialyzed against water using a Zeba spin desalting column and then denatured at 70 °C for 30 min. The β-elimination reagent mixture (mixture of sodium hydroxide solution and β-elimination reagent in the ratio provided with the Glycoprofile β-Elimination kit) was added to the protein sample in a volume equal to 20% of the protein sample and incubated at 4 °C for 18 hours. The Spike protein was separated from the released O-glycans following β-elimination by passing the reaction mixture through centrifugal membrane filters (MWCO 10 kDa) after neutralizing the mixture (pH 6-8). The Spike protein was then lyophilized, biotinylated, and used for the lectin array binding study as described above.

Cell Culture and SARS-CoV-2 Propagation:

Vero E6 cells were obtained from the National Centre for Cell Science (NCCS) at passage number 12 and cultured in Dulbecco’s minimum essential medium (DMEM) supplemented with 5% fetal bovine serum (FBS) and 0.1 mg/ml Penicillin-Streptomycin. Cells were maintained in T25 cm² flasks at 37 °C with 5% CO₂ until reaching 90% confluency. SARS-CoV-2, isolated from a patient swab sample (Genbank number EPI_ISL_11450498) [41] and propagated through three passages in Vero E6 cells, was stored at -80 °C in DMEM supplemented with 5% FBS. This virus stock was used in all the experiments.

Virus Infection:

Vero E6 cells at 90% confluency were infected with SARS-CoV-2 at a volume of 250 µl for one hour at 37 °C. Control cells were treated with an equivalent volume of media. The infection was confirmed by Plaque assay (Figure S1a,b). Following infection, cells were maintained in DMEM without FBS at 37 °C for 48 hours, with daily observation for cytopathic effects such as cell aggregation, rounding of cells, and typically complete detachment of cells from the monolayer on the second day.

Virus harvest and enrichment:

At 48 hours post-infection, the virus-containing supernatant was harvested by centrifugation at 400xg for 10 minutes. The resultant supernatant was collected and stored at -80 °C until further analysis. Quantitative analysis was performed using reverse transcription-polymerase chain reaction (RT-PCR) with a GCC Biotech kit.

The harvested supernatant was initially concentrated using an Amicon centrifugal membrane filter with MWCO 30 kDa [42]. Subsequently, virus enrichment was performed using either polyethylene glycol (PEG) precipitation or a Dynabead kit.

PEG Enrichment Method:

PEG-8000 (80 g), sourced from Sigma Aldrich (Catalog no. 25322-68-3), was dissolved in MilliQ water of type 1. To this solution, 14 g of NaCl were added, followed by the addition of 20 ml of 10X PBS (1X PBS: 137 mM NaCl, 10 mM phosphate, 2.7 mM KCl; pH 7.4) to achieve a final volume of 200 ml. As a result, the resulting solution had a stock concentration of 40% (w/v) PEG and 1.2 M NaCl, forming the PEG concentrator. The concentrated supernatant was mixed with the PEG concentrator at a ratio of 1:3 (v/v) and then incubated on a shaker at 60 rpm at 4 °C overnight. Subsequently, the mixture underwent centrifugation at 1600xg for 60 min at 4 °C. Following centrifugation, the supernatant was decanted, and the pellet was reconstituted in 1X PBS for RT-PCR analysis and further use [43].

Dynabead Intact Virus Enrichment Kit:

The concentrated supernatant was enriched using the Dynabeads™ Intact Virus Enrichment kit (Invitrogen Catalog no. 10700D) as per the instructions [44]. Briefly, the viral supernatant was mixed with Dyna magnetic beads (20 µl) and subjected to a 20-minute incubation on a roller. Following this, the mixture was placed on a magnetic stand for 1 min to remove the supernatant. The beads underwent three washes with 1X PBS. Subsequently, a releasing buffer (50 mM citric acid, 50 mM Na-phosphate, pH 4) was added to liberate the virus from the beads, followed by another 20-min incubation on a roller. The resulting supernatant was collected and further processed through a 30K MWCO centrifugal filter (Amicon) for buffer exchange, ultimately yielding the virus in 1X PBS, which was subjected to RT-PCR analysis.

RT-PCR:

RNA isolation from samples was carried out using a GCC Biotech RNA Isolation kit, followed by reverse transcription with a DiAGSURE nCoV-19 Detection Assay kit. The resulting cDNA was amplified using primers targeting the E-gene (HEX) and ORF1ab gene (FAM). A cut-off value of Ct ≤36 was used for detection.

Virus Inactivation:

To enable lectin array analysis outside of biosafety level 3 (BSL3) facilities, virus aliquots underwent UVC treatment for 10 minutes followed by heat treatment at 55 °C for 20 minutes. This treatment has been previously shown to preserve viral glycosylation while ensuring inactivation [45].

Biotinylation of Spike glycoprotein / SARS-CoV-2 virus:

Recombinant Spike glycoprotein was biotinylated before application to the lectin array slides according to the manufacturer’s instructions. Briefly, 30-40 µg spike glycoprotein (in PBS) was labeled with biotin-NHS using the 1x labeling reagent provided with the kit and incubated for 30 min at room temperature with gentle rocking. The reaction was then stopped using a 3 µl stop solution, the protein was dialyzed extensively against PBS, and the concentration was estimated using the absorbance at 280 nm. SARS-CoV-2 virus, obtained after enrichment and inactivation, was biotinylated by 3.3 µl 1X labeling reagent as per the kit instruction for labeling cells. Like the Spike protein, it was incubated for 30 min, and the reaction was stopped using the stop solution and further dialyzed with PBS.

Lectin Array Assays:

The lectin binding profiles of recombinant Spike glycoprotein and cultured SARS-CoV-2 virus were detected using the label-based approach provided with the Ray Biotech Lectin Array 95 kit (cat# GA-Lectin-95). Briefly, the lectin array slide was first blocked by incubation with sample diluent at room temperature for 30 min, then 10 µg of biotinylated Spike glycoprotein or 100 µl SARS-CoV-2 virus was incubated on the lectin slide at 4 °C overnight and washed multiple times (10-12 times for 5 min each) using the provided wash buffers. The reaction was then incubated with a Cy3 equivalent dye-streptavidin conjugate in the dark at RT for 1 hour and washed multiple times. The slide was then dried, and the signal was visualized using the Cy3 detection settings on a Typhoon Molecular Imager (Cytiva Life Sciences). 1 µg of PAA-Biotin, PAA-Biotin-Mannose, PAA-Biotin-L-Fucose, and PAA-Biotin-Sialic acid were incubated on blocked lectin array slides in the same manner as mentioned above for the initial characterization of the lectin array. For competitive sugar inhibition assays set up to analyze the effect of different sugars on the lectin binding to Spike glycoprotein, biotinylated Spike was pre-incubated with 500 mM D-mannose, 100 mM L-fucose, or 100 mM sialic acid (pH 7) at 4 °C for one hour before application to the blocked lectin array slide. The rest of the steps were the same as described above.

Densitometry and quantification of Lectin Array results:

The spot intensity of each lectin in the fluorescence scan image obtained from the Typhoon Molecular Imager was determined by using the Protein Array Analyser toolset [46] under the Macros plugin of the Image J software [47]. Spot intensities were calculated with linear background subtraction (radius of rolling 2D ball: 25). In the case of visible artifacts or misalignment of the spot intensities with the grid in the Protein Array Analyser toolset, individual spot intensities were quantified manually using the same radius of measurement. Spot intensities thus calculated were imported into the GA-Lectin-95 Excel sheet provided with the lectin array kit to sort and map spot intensities to the respective lectins. These mapped intensities were then normalized with respect to the positive and negative controls and, hence, represented as normalized percent relative fluorescence units (RFU). In the case of Spike subjected to β-elimination or competition with monosaccharides, differences in lectin binding with respect to the untreated Spike were analyzed by calculating the fold change and performing a paired t-test.

Glycan binding profiles of lectins on the lectin array:

We used a commercial lectin array comprising 95 lectins for this study (Figure 1b). The 95 different lectins are spotted in duplicates, and the array also contains six positive control spots (indicative of binding to streptavidin) and three negative control spots (that do not show any signal upon incubation with streptavidin). The 95 lectins include 5 microbial lectins, 12 animal lectins, 65 plant lectins, 10 fungal lectins, 2 slime mold lectins, and 1 red algal lectin (Figure 1b), and the broad specificities of these lectins are indicated by the manufacturer, as gleaned from the published literature. We validated this by using biotinylated polyacrylamide probes with or without derivatization with the monosaccharides, mannose, fucose and sialic acid (PAA-Biotin, PAA-Biotin-Mannose, PAA-Biotin-L-Fucose, and PAA-Biotin-Neu5Ac), and saw that some levels of non-specific binding to PAA-Biotin was observed for the lectins, AAL, VVA, RS-Fuc, GHA and BC2LCN (Figure 1c). Binding of PA-IIL, BANLEC, PTL-2, UEA-I, and increased binding of AAL and RS-Fuc were observed when incubating with PAA-biotin-L-fucose (Figure 1c). BANLEC, AMA, Lentil, GRFT, ConA, BC2L-A, MNA-M, PTL-2, PA-IIL, and PSA showed binding to PAA-Biotin-Mannose (Figure 1c). However, other mannose-binding lectins such as NPA, GNA, Orysata, and Calsepa did not show any mannose-specific binding to PAA-Biotin-Mannose (Figure 1c). No specific binding by sialic acid-specific lectins to PAA-Biotin-Sialic acid was observed albeit binding by RCA-120 and non-specific binding by AAL, VVA, RS-Fuc, GHA and BC2LCN was observed (Figure 1c); this might be due to the low affinity of the sialic acid-specific lectins such as SNA and MAA for the monosaccharide Neu5Ac.

We sought to more thoroughly characterize the lectins on this array so that we could have reasonable expectations as to the binding of recombinant Spike glycoprotein or cultured SARS-CoV-2 virus to lectins on this array. We did this by employing glycan array data. Publically available glycan array datasets existed for 68 of these 95 lectins, and the glycan binding specificities of these lectins were also available on Carbogrove [37]. To have a unified analysis of all these lectins, we employed the MotifFinder tool using all available glycan array data for each of the lectins and the same settings of MotifFinder, and generated a lectin model for each of these 68 lectins (Supplementary Data 1). Following this, we generated a list of core and extension sequences of N- and O-glycans expected to be found on mammalian cells (Supplementary Data 2) and used the lectin models obtained in the previous step to generate binding propensities, i.e., predict the binding of each of these 68 lectins to these N- and O-glycans (Supplementary Data 2).

Out of the 68 characterized lectins, 17 lectins showed binding towards the O-glycan core, 32 lectins bound to the N-glycan core, and 18 lectins bound to extension sequences (N-glycan extensions and O-glycan extensions) (Figures S2a, b, c). MPA did not show significant binding to any of the generated glycans (Figure S2a).

Lectin binding to N-core and N-extension sequences: The generated heat map of the binding propensities for N-glycan core sequences (portrayed on the rows) showed that some lectins (portrayed on the columns) exhibit high binding towards N-glycans such as MNA-M, MNA-G, Banlec, AAL, CAA, VFA, DSA, AMA, ABA, and UDA whereas other lectins like Jacalin, SBA, SJA, WFA, EEL, GAL-7, and MOA exhibit less binding towards N-glycans (Figure 2a). Lectins binding to the N-glycan core were grouped in three clusters - lectins predicted to bind to the high mannose-containing glycans, lectins predicted to bind to complex type N-glycans, and lectins predicted to bind to both mannose-containing glycans as well as complex type glycans. Using a binding propensity cutoff of 0.1, we could classify the lectins into three discrete groups (Figure 2b). ASA, Banlec, BC2L-A, UDA, VFA, SNA-II, PA-IIL, LEA, and LTL bound to only high/oligo-mannose N-glycan cores (Man 3-5 tGlcNAc, Man 5-6 unsubstituted, and Man 7-9 unsubstituted binding lectins); lectins like PHA-L, DSA, RPA, TL, and PHA-E bound to only complex-type N-glycan cores (bi-antennary N-glycans with tGlcNAc, tri-antennary N-glycans with tGlcNAc, and tetra antennary N-glycans with tGlcNAc); the lectins, LcHA, BC2LCN, ConA, GNA, NPA, ORYSATA, AMA, HHA, MNA-G, MNA-M, LBA, CA, CAA, ABA, BSL/GS-II, and WGA bound to both high/oligo-mannose and complex-type N-glycan cores (Figures 2b, 2c). Using the binding propensities, we also determined the effect of branching and core-fucosylation. PHA-E was predicted to bind only to complex-type N-glycans with or without bisecting GlcNAc branch, besides binding to extension sequences (with higher binding propensities) (Figure 2c, Figure S2a). RPA displayed increased binding propensity for N-glycan cores with a bisecting GlcNAc branch. LcHA, ConA, ORYSATA, NPA, HHA, LBA, ABA, GS-II, and TL showed decreased binding propensity for N-glycan cores with a bisecting GlcNAc branch (Figure 2c). PSA and BC2LCN were predicted to bind only to core-fucosylated complex-type N-glycan cores, and were not predicted to bind to any extension sequences (Figure 2c, Figure S2a). AAL and LcHA displayed increased binding propensities for some core-fucosylated N-glycans. Most lectins, with the exception of BC2LCN, DSA, GNA, LcHA, LEA, PHA-L, PSA, RPA, SNA-II, and VFA, were also predicted to bind to N-extension sequences (Figures S2a,b).

Lectin binding to O-core and O-extensions:The generated heat map of the binding propensities for O-glycan core sequences identified Jacalin, SBA, ACG, HAA, WFA, DBA, BPA, PNA, PA-IL, CNL, ACL, AAA, and WGA as lectins with high binding propensities for O-glycan cores (Figure 3a). Many of these lectins were predicted to bind more than one type of O-core. The exceptions were CSA and VVA, which were predicted to bind only the Tn antigen; UEA-II, predicted to only bind the O-glycan core-4; and MOA, predicted to bind only the O-glycan core-8. Conversely, all cores were predicted to be bound by more than one lectin, with the exception of sialylated Tn antigen which was predicted to be bound only by WGA. The predicted binding of lectins to the various types of O-glycan cores is shown in Figure 3b. With the exception of CSA and VVA, all these lectins were also predicted to bind to O-extension sequences (Figures S2a,c).

Lectins binding to Spike glycoprotein of SARS-CoV-2:

We performed lectin array analysis of recombinant SARS-Related Coronavirus 2, Wuhan-Hu-1 HexaPro Spike Glycoprotein Ectodomain expressed and purified from BEI NR-53587 pαH Vector transfected HEK-293F cultures (Figures 4a, b, c). As controls, we used BSA (which is not glycosylated) and PBS, and we noted a low-level non-specific signal only for the VVA lectin spots in the former array (Figure 4c). High-intensity fluorescence signals were observed for many of the lectin spots in the arrays incubated with varying amounts (5 to 50 µg) of Spike glycoprotein (Figure 4c), indicative of the presence of glycosylation on Spike and optimal assay conditions.

Figure 4d shows an individual dot plot containing the average values of four replicates of 10 µg Spike glycoprotein assayed on the lectin array. We found that Banlec, PHA-L, PTL-2, PHA-E, PSA, TL, Lentil, MNA-M, NPA, HHA, GS-II, and GNA were the top lectins with the highest binding intensities to Spike glycoprotein, and there were many lectins with binding intensities exceeding that of the positive control in the lectin array (Figure 4e, Supplementary Data 2). These lectins were also found to display high intensities (more than positive control on average and >100% at least in two replicates) in the assays performed with different concentrations of Spike glycoprotein (Figure 4c, Supplementary Data 2). The top glycan motif (among the different N- and O-glycans expected in mammalian cells) predicted to be bound by each of these lectins is represented in Figure 5, and includes high/oligo-mannose or hybrid N-glycans (predicted to be bound by lectins, ConA, AMA, Banlec, BC2LA, GNA, HHA, NPA, and MNA-M) or complex-type N-glycans (predicted to be bound by lectins such as LcHA, ASA, RS-Fuc, GS-II, ORYSATA, PHA-E, AMA, PSA, PHA-L), as well as O-glycans and extension sequences (predicted to be bound by lectins such as ACL and SBA). The binding propensities of these lectins to the comprehensive list of N- and O-glycans in mammalian cells, as well as the lectin models with all predicted glycan motifs, are provided (Supplementary Data 1, 2). Lectins that did not bind to the Spike glycoprotein included PPL, BPA, PA-IL, SNA-II, SJA, UEA-II, Jacalin, SSA, and VFA, and the lectins, PNA, ECA, GAL1, UEA I, PWA, VVA-M, DISCOIDIN I, LAL, GAL2, PSL1A, SHA, VRA, LBA bound with low intensity (Supplementary Data 2).

Lectins binding to N-core glycans:We analyzed the binding of the Spike glycoprotein by the lectins predicted to bind to N-glycan core sequences (Figure 4f). We found that the lectins, AAL, AMA, ASA, BANLEC, BC2L-A, CA, CAA, ConA, DSA, GNA, GS-II, HHA, LcHA, UDA, MNA-G, MNA-M, NPA, ORYSATA, PHA-E, PHA-L, PSA, TL, and WGA showed >50% normalized percent fluorescence intensity to Spike glycoprotein (Figure 4f). This included lectins with high binding propensities for high/oligo-mannose N-glycan cores such as ASA, BANLEC, BC2L-A, and UDA, as well as lectins with high binding propensities for complex-type N-glycan cores such as PHA-L, DSA, TL, and PHA-E, and lectins with high binding propensities for both high/oligo-mannose and complex-type N-glycan cores such as ConA, GNA, NPA, ORYSATA, AMA, HHA, MNA-G, MNA-M, CA, CAA, GS-II, LcHA, and WGA (Figure. 2c, 4f). The lectins ABA, BC2LCN, LBA, LEA, LTL/Lotus, RPA, SNA-II, and VFA showed low or negligible binding to Spike glycoprotein (Figure 4f). Considering the predicted binding of PSA only to the core fucosylated complex-type N-glycan (Figure 2c, Figure S2a), our lectin array results indicate the presence of core-fucosylation on the Spike glycoprotein (Figure 4f). Further, the absence of binding to RPA (Figure 4f), which is predicted to bind only to a bisecting GlcNAc branch or tri- or tetra-antennary complex-type N-glycan cores (Figure 2c, Figure S2a), indicates the presence of biantennary N-glycan cores on the Spike glycoprotein. The high binding to PHA-E (Figure 4f), which is also predicted to bind to a bisecting GlcNAc branch (Figure 2c, Figure S2a) might be explained by the presence of extension sequences on the Spike glycoprotein that are bound by PHA-E.

Lectins binding to O-core glycans: We analyzed the binding of the Spike glycoprotein by the lectins predicted to bind to O-glycan core sequences (Figures 3b, 4g). We found that ACG, ACL, MOA, SBA, and WGA bound with high binding intensities to Spike glycoprotein whereas AAA, BPA, CNL, CSA, DBA, HAA, Jacalin, PA-IL, PNA, UEA-II, VVA, and WFA bound with relatively low intensities (Figure 4g). Considering that all these lectins also bind to extension sequences (Figure S2a), this does not necessarily indicate the presence of O-glycans on the Spike glycoprotein of SARS-CoV-2.

To further explore the identity of lectins capable of binding to the O-glycans, if any, on the recombinant Spike glycoprotein expressed from HEK293 cells, we subjected the Spike glycoprotein to beta-elimination for the removal of O-glycans, and then performed lectin array assays (Figures S3a, b). No change in mass was observed upon beta-elimination (Figure S3a), indicative of a low degree of modification, if any, with O-glycans. No significant inhibition was found (considering a threshold of >5-fold change and p-value <0.1 in at least two replicates in a paired t-test) for any of the possible top-binding O-glycan-binding lectins (ABL, ACL, LSL-N, RCA60, and SBA (Figure S3c). Further, similar levels of reduction in binding (as observed for these O-glycan-binding lectins) were also observed for some N-glycan-binding lectins (Figure S3c), perhaps due to the peeling of N-glycans during the β-elimination reaction.

Monosaccharide competition assays: We performed lectin array assays of the Spike glycoprotein in the presence of the monosaccharides, 500 mM mannose, 100 mM L-fucose, and 100 mM Neu5Ac, to assess competitive inhibition, if any, of lectin binding by these monosaccharides (Figure 6a, Figures S4a, b). We found that the binding of the lectins, PA-IIL, RS-Fuc and AAL, was inhibited by 100 mM L-fucose (considering a threshold of >5-fold change and p-value <0.1 in at least two replicates in a paired t-test) (Figure 6b), thus validating the L-fucose specific binding of these lectins to the glycans on the Spike glycoprotein. No significant inhibition was observed in the binding of any of the mannose and sialic acid binding lectins upon incubation with 500 mM mannose and 100 mM sialic acid, respectively (considering a threshold of >5-fold change and p-value <0.1 in at least two replicates in a paired t-test), although the N-glycan binding lectins, GS-II and PSA, did show a modest degree of inhibition upon D-mannose competition (Figures S4c, d). It is possible that these and other mannose and sialic acid binding lectins have low binding affinities for the monosaccharide, which consequently is unable to competitively inhibit binding to the more complex glycan motifs on the Spike glycoprotein.

Lectins binding to the cultured SARS-CoV-2:

We performed lectin array analysis of SARS-CoV-2 virions propagated in Vero E6 cells (Figures S1a, b), enriched using centrifugal membrane filters and either PEG or a Dynabeads™ Intact Virus Enrichment kit (Figures S1c, d), inactivated by heat and UV [45], and biotinylated with biotin-NHS. Enrichment was confirmed by RT-PCR for E-gene (HEX channel) and Orf1ab gene (FAM channel). (Figure S1c, d) by significant differences in the Ct values for the RT-PCR performed with the un-enriched virus and the enriched virus (Figure S1c, d and Supplementary Data 2). For controls for the lectin array, we subjected culture supernatants of uninfected Vero E6 cells to the same enrichment procedures mentioned above. We detected no fluorescence for any of the lectin spots in these controls (Figures 7a, b). High-intensity fluorescence signals were observed for many of the lectin spots in the arrays incubated with SARS-CoV-2 virus enriched either by PEG or by the Dynabeads™ Intact Virus Enrichment kit, indicative of the presence of glycosylation on the SARS-CoV-2 virions (Figures 1a, b).

Figure 7c shows an individual dot plot containing the average values of three replicates each of SARS-CoV-2 enriched either by PEG or by the Dynabeads™ Intact Virus Enrichment kit and assayed on the lectin array. We found many lectins with binding intensities exceeding that of the positive control in the lectin array and these included RCA120, BANLEC, NPA, PTL-2, PA-IIL, AMA, PHA-P, GNA, TL, STL, ASA, CPA, AAL, SBA, MAA, and ConA in the case of SARS-CoV-2 virions enriched by PEG, and BANLEC, GNA, RCA120, AMA, and NPA in the case of SARS-CoV-2 virions enriched by the Dynabeads™ Intact Virus Enrichment kit (Figure 7c). The top glycan motif (among the different N- and O-glycans expected in mammalian cells) predicted to be bound by each of these lectins is represented in Figure 5, and includes high/oligo-mannose or hybrid N-glycans (predicted to be bound by ConA, AMA, BANLEC, GNA, and NPA) or complex-type N-glycans (predicted to be bound by ASA, AMA, RCA-120, PTL-2, PA-IIL, PHA-P, TL, STL, AAL, and MAA), as well as O-glycans and extension sequences (SBA). The glycan motif bound by CPA is unknown; it is reported to haemagglutinate rabbit erythrocytes with inhibition by fetuin, and GalNAc [48,49].

Overall, we found better lectin binding for SARS-CoV-2 enriched by PEG, with 16 lectins displaying fluorescence intensities exceeding that of the positive control spots in SARS-CoV-2 enriched by PEG compared to only five lectins for SARS-CoV-2 enriched by the Dynabeads™ Intact Virus Enrichment kit (Figure 7c, d). Interestingly, the glycan binding profiles of these lectins are also quite distinct. The five lectins binding to SARS-CoV-2 enriched by the Dynabead kit mainly bind to high/oligo-mannose or hybrid N-glycan cores, with the exception of RCA-120, whose top glycan motif is a galactose terminating biantennary N-glycan (Figure 5).

Lectins binding to N-core glycans:We analyzed the binding of the SARS-CoV-2 by the lectins predicted to bind to N-glycan core sequences (Figure 7e). We found that the lectins, AAL, AMA, ASA, BANLEC, ConA, DSA, GNA, HHA, LcHA, NPA, ORYSATA, PA-IIL, PHA-E, PHA-L, TL, and UDA showed significant binding to SARS-CoV-2 virions enriched by PEG (Figure 7e). The lectins ABA, BC2L-A, BC2LCN, CA, CAA, GS-II, LBA, LEA, Lotus, MNA-G, MNA-M, PSA, RPA, SNA-II, VFA, and WGA showed low or negligible binding to SARS-CoV-2 virions enriched by PEG or the Dynabeads™ Intact Virus Enrichment kit (Figure 7e). The absence of binding to PSA is suggestive of the absence of core-fucosylation on the SARS-CoV-2 virions propagated in Vero E6 cells (Figure 7e).

Lectins binding to O-core glycans: We analyzed the binding of the SARS-CoV-2 virions by the lectins predicted to bind to O-glycan core sequences (Figure 7f). We found that only CSA, DBA, MOA, SBA, and WFA bound with high (>50% normalized %RFU) binding intensities to SARS-CoV-2 virions enriched by PEG (Figure 7f). Considering that all these lectins also bind to extension sequences (Figure S2a), this does not necessarily indicate the presence of O-glycans on SARS-CoV-2.

Lectins binding to recombinant Spike glycoprotein vs. SARS-CoV-2 virions: We found four additional lectins, viz., PA-IIL, STL, MAA, and CPA, that bound SARS-CoV-2 enriched by PEG with high intensity but did not bind to the Spike glycoprotein with average intensities exceeding those of the positive control spots (Figures 4e, 7d). We noticed that PA-IIL and MAA showed >100% normalized percent fluorescence intensity in the Spike glycoprotein assay in one replicate and two replicates, respectively, although the average normalized percent fluorescence intensity was <100% (Supplementary Data 2). Lectins that bound with high intensity to the Spike glycoprotein but did not bind to SARS-CoV-2 enriched by PEG with intensities exceeding those of the positive control spots include AAL, ABL, ACL, BC2L-A, CA, CALSEPA, GAL3, GRFT, GS-II, HHA, LcHA, Lentil, LSL-N, MNA-M, ORYSATA, PALa, PHA-E, PHA-L, PSA, RCA60, and RS-Fuc. The differential binding by these lectins might be explained by different levels of variously linked fucose (perhaps, a higher abundance of core-fucosylation (α1,6-fucose), and α1,2-fucose in the Spike glycoprotein expressed from HEK293F cells, and a higher abundance of α1,3/4-fucose in the SARS-CoV-2 virions propagated from Vero E6 cells) and α2,3-Neu5Ac (in SARS-CoV-2 virions propagated from Vero E6 cells).

The SARS-CoV-2 viral envelope is covered with highly glycosylated spike protein, essential in viral binding and entry to the host cells [50,51]. In the therapeutic field, targeting glycosylation offers possibilities for inhibiting viral infections such as COVID-19. Lectins are, therefore, lead molecules for developing new antiviral agents based on the ability to inhibit viral entry in the host cell. Here, we have screened 95 different lectins from different sources (animals (12), plants (65), fungi (10), microbes (5), slime moulds (2), and red alga (1)) on a commercial lectin array for their potential to bind to the SARS-CoV-2 Spike glycoprotein expressed from HEK293F cells and the SARS-CoV-2 virions propagated in Vero E6 cells. Although a few studies have recently used lectin arrays – Senicar et al to distinguish the lectin profile of SARS-CoV-2 negative and SARS-CoV-2 positive nasopharyngeal swabs [52], Wuo et al to understand the glycome of SARS-CoV-2 Spike glycoprotein in relation to antibody neutralization [53], Hoffman et al to identify mammalian lectins that bind to the RBD of Spike [54], and Zheng et al to understand the changes in the glycome associated with different variants of concern of Spike [55], our study represents a notable expansion compared to the previous studies, with a more detailed examination of lectin-binding specificities associated with SARS-CoV-2. Further, our study benefited from a comprehensive examination of the specificities of these lectins, as determined from the available glycan array data on Carbogrove [37] (68 lectins out of 95 lectins) using the MotifFinder tool [35] against N-core, O-core, N-extension, and O-extension glycans.

There are several studies that have examined the site-specific glycosylation profile of SARS-CoV-2 Spike glycoprotein using mass spectrometry [11,14,18,56-60].The glycan shield of SARS-CoV-2 is accepted to be sparse and mainly dominated by complex-type glycans as compared to other viral glycan shields, such as that of HIV, which is densely packed and contains around 60% of oligomannose-type N-glycans [61-63]. Similar to HIV, the S1 subunit harboring the RBD domain of the Spike glycoprotein of MERS, is predominantly covered with hybrid-type N-glycans and similar levels of oligomannose-type and complex-type N-glycans, but the Spike glycoprotein of SARS-CoV-1, predominantly contains hybrid and complex-type N-glycans, with only a few glycosites occupied by oligomannose N-glycans. The S1 subunit of SARS-CoV-2 is decorated with complex-type N-glycans with fewer hybrid-type and oligomannose-type N-glycans [59]. Out of 22 N-linked glycosylation, 8 sites contain a considerable population of oligomannose-type glycans, while 14 sites are dominantly covered with complex-type glycan [9,56].

Here, we have studied the glycomes of SARS-CoV-2 Spike glycoprotein expressed and purified from HEK293F cells, and SARS-CoV-2 virions propagated in Vero E6 cells. We present a panel of 32 lectins (including lectins that bind to oligomannose/hybrid and complex-type N-glycans, extension sequences, and O-glycans) that robustly bind to the Spike glycoprotein, and five lectins that bind to SARS-CoV-2 virions, irrespective of the method (PEG or Dynabead kit) used for enrichment. Banlec and Lentil, among the top binding lectins in our study, have previously been shown to block the entry of SARS-CoV-2 and its variants [29,34]. GRFT from Griffithsia sp., is a broadly specific lectin that has previously been shown to exhibit anti-SARS-CoV-2 activity[28], and the broad spectrum lectin from stinging nettle, UDA potentially blocks the pseudotyped SARS-CoV-2 in A549 cells [33]. HHA, NPA, GNA, and MNA-M lectins, also listed as top binders in our study, have been shown to have antiviral activities against SARS-CoV and FIPV [64]. Besides these previously reported lectins, our panel includes the following lectins, AAL, ABL, ACL, AMA, ASA, BC2L-A, CA, ConA, GAL3, GS-II, LcHA, LSL-N, ORYSATA, PALa, PHA-E, PHA-L, PHA-P, PSA, RCA60, RCA120, SBA and TL, which can be further studied for their potential to inhibit viral entry in future studies. Published literature also suggests the anti-viral activity of Triticum vulgaris (Wheat Germ Agglutinin) lectin against SARS-CoV-2 and its Variants of Concern (VoC), Alpha and Beta in Vero B4 cells and a pulldown assay indicated the direct binding of this lectin with the virus [30]. Another investigation found that Gal-9 and Gal-3 facilitate SARS-CoV-2 transmission by promoting binding with ACE2 in a glycan-dependent manner [65,66]. Furthermore, it has been demonstrated that Gal-3 antagonist Prolectin-M (PL-M) tablets can reduce SARS-CoV-2 activity in patients [67]. In our experiments, Gal-9 displayed moderate binding intensity using the PEG enrichment method, whereas Gal-3 showed low binding intensity regardless of the method used, while GAL-3 is along the top binder for SARS-CoV-2 spike glycoprotein. Previously, HHA, BPA, WFA, MPA, AlA, GSl-II, PHA-E, PHA-L, WGA, and SNA were shown to interact with the nasopharyngeal samples of SARS-CoV-2 positive patients [52]. However, in our lectin array experiment, WGA does not bind with as high a binding intensity as the lectins mentioned above. We cannot rule out the possibility that WGA (or VFA, which did not display binding in any lectin array assay, or any of the other lectins that did not bind with high intensity) was not fully functionally active on the lectin array or optimally accessible without any steric hindrances. Whereas our initial characterization of the lectin array with biotinylated polyacrylamide probes validated the function of only some of the spotted lectins, several lectins showed high binding intensities to Spike glycoprotein, suggesting that the absence of binding by some mannose-binding lectins was due to a low affinity for the monosaccharide mannose rather than inactive lectin in the lectin array.

Our study indicates differences in the glycan profiles of recombinant Spike glycoprotein and SARS-CoV-2 virions enriched by two different methods. These differences may stem from the different cell types (HEK293F vs Vero E6) [68], different processing routes (recombinant glycoprotein ectodomain secretion vs virus budding), different glycoconjugate moieties (Spike glycoprotein vs Spike and other glycoproteins on the intact virions; for instance, Griffithsin inhibited hACE2-mediated direct infection better when the coronaviral M protein was present than when it was absent, which may explain why it inhibits authentic SARS-CoV-2 better than pseudotyped viruses, which do not contain M [69]), different degrees of enrichment and resulting contamination with other glycoproteins secreted by the cells or present in the serum-free culture medium (due to the different principles of enrichment, i.e., molecular exclusion by the non-ionic polymer PEG [70] and charge-based enrichment by the Dynabead kit), and the effect of enrichment conditions on the Spike glycans (effect of exposure to acidic pH during enrichment using the Dynabead kit [71], and longer processing times with the enrichment method by PEG).

Previous studies in the literature have also indicated that Spike glycosylation can vary according to the lab (and researchers) performing the study, expression cell lines, and the sample being studied (glycoprotein or pseudovirus or clinical sample) [56]. The structural mapping of recombinant spike glycoprotein of SARS-CoV-2 revealed that CHO-expressed spike glycoprotein exhibits more complex glycans and higher sialylation as compared to HEK-expressed spike glycoprotein, which exhibits more high-mannose glycans and lower sialylation [14]. Site-specific glycosylation also differs in these cell lines, N1158 site was found unoccupied in CHO-expressed spike, and N122, N282, and N1158 positions were found unoccupied in HEK-secreted spike glycoprotein [14]. Although the conservation of glycosylation sites and glycan composition differs at each site in different variants of concern, Wuhan-Hu-1 Spike has a high amount of complex glycan with higher fucosylation as compared to other variants (Alpha, Beta, Gamma, Delta, and Omicron); Delta and Omicron strains have significantly high oligo-mannose content and low fucose-containing complex glycans as compared to the Wuhan-Hu-1 Spike [18]. The site-specific glycan analysis of virally derived Spike glycoprotein and recombinant Spike glycoprotein purified from labs in Amsterdam, Harvard, Southampton/Texas, Switzerland, and Oxford showed that all the recombinant samples contained at least 30% oligomannose content and varied composition at different glycan sites, whereas N234 showed homogeneity in all the samples [56]. The recombinant Spike glycoprotein possessed a low level of oligomannose content at several sites compared to the virally derived Spike glycoprotein. In contrast, virally derived Spike glycoproteins have higher occupancy at each site [56]. Our comparison of lectins binding to the Spike glycoprotein and SARS-CoV-2 virions enriched by PEG, importantly, provides additional information as to the presence of α2,3-sialic acid and the different glycosidic linkages of fucose that might be present in the recombinant glycoprotein and the intact virions.

The native-like trimeric soluble Spike glycoprotein exhibits only T323 O-glycan sites with low occupancy. At the same time, the RBD monomer displays several O-glycan sites and almost wholly occupied T323 sites when N-glycosylation is inhibited by kifunensine [60]. According to reports, the T323 site possesses Tn antigen, core-1, sialylated core-1, core-2, and sialylated core-2 glycans [57,72]. The top lectins that bound to recombinant Spike glycoprotein in our study included ACL, which we predicted to bind to core-1 and core-2 O-glycans, and SBA, which we predicted to bind to Tn, core-5, and core-7 O-glycans. Besides binding to O-cores, ACL and SBA were predicted to bind to GalNAc in N- glycan extension sequences. Our lectin array assays with Spike glycoprotein subjected to beta-elimination, however, did not indicate a significant change in the binding of any O-glycan-specific lectin. Considered together with the report that about 99% of O-glycan sites are unmodified in the native soluble spike glycoprotein [11], it is unclear whether ACL and SBA bound to O-glycans on the Spike glycoprotein.

Viral proteins are generally covered with under-processed high mannose and hybrid glycans) [61,73], resulting in the use of mannose-binding lectins as antiviral agents [64]. In contrast, the SARS-CoV-2 spike protein is mainly decorated with complex-type glycans [56] thus highlighting a requirement for lectins that bind preferentially to complex glycans present on the virus and not on the host cells. Apart from mannose-binding lectins, this study also identifies several complex N-glycan-binding lectins. The identification of lectins with specific binding profiles to SARS-CoV-2 holds promising implications for therapeutic development and diagnostic assays. These lectins could serve as potent inhibitors or markers for virus detection. However, further experimental validation is warranted to elucidate their precise roles and mechanisms of action.

Data availability

All lectin models used in the study are available in Supplementary Data 1, and all source data behind the graphs in the manuscript are being provided in Supplementary Data 2. Other data associated with the study are available from the corresponding authors upon reasonable request.

Funding

This work was supported by the Council of Scientific and Industrial Research, Government of India (MLP0037 to TNCR, AR, and RPR). N acknowledges the University Grants Commission, Government of India, for her fellowship. SA acknowledges CSIR for JRF and SRF.

Acknowledgment

The authors acknowledge Dr. Ipsita Roy, Associate Professor, NIPER Mohali, and her group for facilitating the use of the Amersham Typhoon Molecular Imager at NIPER Mohali, Mr. Zachary Klamer, Van Andel Institute, Michigan, for guidance in using MotifFinder for glycan array data analysis, and CSIR-IMTECH (manuscript communication number: 047/2024) for the research facilities and infrastructure. The following reagent was obtained through BEI Resources, NIAID, NIH: Modified pαH Vector Containing the SARS-Related Coronavirus 2, Wuhan-Hu-1 HexaPro Spike Glycoprotein Ectodomain, NR-53587. We acknowledge the BSL3 facility of CSIR-IMTECH for facilitating the virus work.

Author Contributions

N: Methodology, Investigation, Data Curation, Formal analysis, Writing- Original draft preparation, Visualization

SA: Methodology, Investigation, Formal analysis, Writing - Original draft preparation.

RD: Investigation, Writing- Review and editing

RPR: Funding acquisition, Project administration, Methodology, Writing- Review and editing, Supervision, Conceptualization

AR: Funding acquisition, Project administration, Methodology, Writing- Review and editing, Supervision, Conceptualization

RTNC: Funding acquisition, Project administration, Formal analysis, Methodology, Writing- Review and editing, Visualization, Supervision, Conceptualization.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Letko M, Marzi A, Munster V. Functional assessment of cell entry and receptor usage for SARS-CoV-2 and other lineage B betacoronaviruses. Nat Microbiol. 2020 Apr;5(4):562-569.
Hoffmann M, Kleine-Weber H, Schroeder S, et al. SARS-CoV-2 Cell Entry Depends on ACE2 and TMPRSS2 and Is Blocked by a Clinically Proven Protease Inhibitor. Cell. 2020 2020/04/16/;181(2):271-280.e8.
Shang J, Ye G, Shi K, et al. Structural basis of receptor recognition by SARS-CoV-2. Nature. 2020 May;581(7807):221-224.
Zhou P, Yang XL, Wang XG, et al. A pneumonia outbreak associated with a new coronavirus of probable bat origin. Nature. 2020 Mar;579(7798):270-273.
Wrapp D, Wang N, Corbett KS, et al. Cryo-EM structure of the 2019-nCoV spike in the prefusion conformation. Science. 2020 Mar 13;367(6483):1260-1263.
Xia S, Liu M, Wang C, et al. Inhibition of SARS-CoV-2 (previously 2019-nCoV) infection by a highly potent pan-coronavirus fusion inhibitor targeting its spike protein that harbors a high capacity to mediate membrane fusion. Cell Research. 2020 2020/04/01;30(4):343-355.
Walls AC, Park YJ, Tortorici MA, et al. Structure, Function, and Antigenicity of the SARS-CoV-2 Spike Glycoprotein. Cell. 2020 Dec 10;183(6):1735.
Allen JD, Ivory DP, Song SG, et al. The diversity of the glycan shield of sarbecoviruses related to SARS-CoV-2. Cell Rep. 2023 Apr 25;42(4):112307.
Watanabe Y, Allen JD, Wrapp D, et al. Site-specific glycan analysis of the SARS-CoV-2 spike. Science. 2020 May 4.
Watanabe S, Amagai Y, Sannino S, et al. Zinc regulates ERp44-dependent protein quality control in the early secretory pathway. Nat Commun. 2019 Feb 5;10(1):603.
Watanabe Y, Allen JD, Wrapp D, et al. Site-specific glycan analysis of the SARS-CoV-2 spike. Science. 2020 Jul 17;369(6501):330-333.
Watanabe Y, Berndsen ZT, Raghwani J, et al. Vulnerabilities in coronavirus glycan shields despite extensive glycosylation. Nature communications. 2020;11(1):2688.
Grant OC, Montgomery D, Ito K, et al. Analysis of the SARS-CoV-2 spike protein glycan shield reveals implications for immune recognition. Sci Rep. 2020 Sep 14;10(1):14991.
Wang Q, Wang Y, Yang S, et al. A Linkage-specific Sialic Acid Labeling Strategy Reveals Different Site-specific Glycosylation Patterns in SARS-CoV-2 Spike Protein Produced in CHO and HEK Cell Substrates. Front Chem. 2021;9:735558.
Struwe WB, Chertova E, Allen JD, et al. Site-Specific Glycosylation of Virion-Derived HIV-1 Env Is Mimicked by a Soluble Trimeric Immunogen. Cell Reports. 2018 2018/08/21/;24(8):1958-1966.e5.
Shajahan A, Supekar NT, Gleinich AS, et al. Deducing the N- and O-glycosylation profile of the spike protein of novel coronavirus SARS-CoV-2. Glycobiology. 2020 Dec 9;30(12):981-988.
Brun J, Vasiljevic S, Gangadharan B, et al. Assessing Antigen Structural Integrity through Glycosylation Analysis of the SARS-CoV-2 Viral Spike. ACS Central Science. 2021 2021/04/28;7(4):586-593.
Shajahan A, Pepi LE, Kumar B, et al. Site specific N- and O-glycosylation mapping of the spike proteins of SARS-CoV-2 variants of concern. Sci Rep. 2023 Jun 21;13(1):10053.
Casalino L, Gaieb Z, Goldsmith JA, et al. Beyond Shielding: The Roles of Glycans in the SARS-CoV-2 Spike Protein. ACS Central Science. 2020 2020/10/28;6(10):1722-1734.
Yang J, Petitjean SJL, Koehler M, et al. Molecular interaction and inhibition of SARS-CoV-2 binding to the ACE2 receptor. Nature Communications. 2020 2020/09/11;11(1):4541.
Reily C, Stewart TJ, Renfrow MB, et al. Glycosylation in health and disease. Nat Rev Nephrol. 2019 Jun;15(6):346-366.
Shental-Bechor D, Levy Y. Effect of glycosylation on protein folding: a close look at thermodynamic stabilization. Proceedings of the National Academy of Sciences. 2008;105(24):8256-8261.
Zhao H, Wang T-C, Li X-F, et al. Long-term stability and protection efficacy of the RBD-targeting COVID-19 mRNA vaccine in nonhuman primates. Signal Transduction and Targeted Therapy. 2021 2021/12/24;6(1):438.
Martina CE, Crowe JE, Jr., Meiler J. Glycan masking in vaccine design: Targets, immunogens and applications. Front Immunol. 2023;14:1126034.
Baghaie L, Leroy F, Sheikhi M, et al. Contemporaneous SARS-CoV-2-Neutralizing Antibodies Mediated by N-glycan Shields. Viruses. 2023 Oct 12;15(10).
Mazalovska M, Kouokam JC. Lectins as Promising Therapeutics for the Prevention and Treatment of HIV and Other Potential Coinfections. Biomed Res Int. 2018;2018:3750646.
Mason CP, Tarr AW. Human lectins and their roles in viral infections. Molecules. 2015 Jan 29;20(2):2229-71.
Ahan RE, Hanifehnezhad A, Kehribar ES, et al. A Highly Potent SARS-CoV-2 Blocking Lectin Protein. ACS Infect Dis. 2022 Jul 8;8(7):1253-1264.
Chan JF, Oh YJ, Yuan S, et al. A molecularly engineered, broad-spectrum anti-coronavirus lectin inhibits SARS-CoV-2 and MERS-CoV infection in vivo. Cell Rep Med. 2022 Oct 18;3(10):100774.
Auth J, Froba M, Grosse M, et al. Lectin from Triticum vulgaris (WGA) Inhibits Infection with SARS-CoV-2 and Its Variants of Concern Alpha and Beta. Int J Mol Sci. 2021 Sep 22;22(19).
Yamasaki K, Adachi N, Tun MMN, et al. Core fucose-specific lectin (PhoSL) as a potent broad-spectrum inhibitor of SARS-CoV-2 infection. Febs J. 2023 Jan;290(2):412-427.
Liu YM, Shahed-Al-Mahmud M, Chen XR, et al. A Carbohydrate-Binding Protein from the Edible Beans Effectively Blocks the Infections of Influenza Viruses and SARS-CoV-2. Cell Reports. 2020 Aug 11;32(6).
Vanhulle E, D'huys T, Provinciael B, et al. Carbohydrate-binding protein from stinging nettle as fusion inhibitor for SARS-CoV-2 variants of concern. Front Cell Infect Mi. 2022 Aug 30;12.
Wang WB, Li QQ, Wu JJ, et al. Lentil lectin derived from exhibit broad antiviral activities against SARS-CoV-2 variants. Emerg Microbes Infec. 2021 Jan 1;10(1):1519-1529.
Klamer Z, Haab B. Combined Analysis of Multiple Glycan-Array Datasets: New Explorations of Protein-Glycan Interactions. Anal Chem. 2021 Aug 10;93(31):10925-10933.
Klamer Z, Staal B, Prudden AR, et al. Mining High-Complexity Motifs in Glycans: A New Language To Uncover the Fine Specificities of Lectins and Glycosidases. Anal Chem. 2017 Nov 21;89(22):12342-12350.
Klamer ZL, Harris CM, Beirne JM, et al. CarboGrove: a resource of glycan-binding specificities through analyzed glycan-array datasets from all platforms. Glycobiology. 2022 Jul 13;32(8):679-690.
Stanley P, Moremen KW, Lewis NE, et al. N-Glycans. In: Varki A, Cummings RD, Esko JD, et al., editors. Essentials of Glycobiology. 4th ed. Cold Spring Harbor (NY)2022. p. 103-16.
de Haan N, Pučić-Baković M, Novokmet M, et al. Developments and perspectives in high-throughput protein glycomics: enabling the analysis of thousands of samples. Glycobiology. 2022;32(8):651-663.
Magalhaes A, Duarte HO, Reis CA. The role of O-glycosylation in human disease. Mol Aspects Med. 2021 Jun;79:100964.
Garg H, Ringe RP, Das S, et al. UVC-Based Air Disinfection Systems for Rapid Inactivation of SARS-CoV-2 Present in the Air. Pathogens. 2023;12(3):419.
Lucansky V, Samec M, Burjanivova T, et al. Comparison of the methods for isolation and detection of SARS-CoV-2 RNA in municipal wastewater. Frontiers in Public Health. 2023 03/07;11.
Zheng X, Wang M, Deng Y, et al. A rapid, high-throughput, and sensitive PEG-precipitation method for SARS-CoV-2 wastewater surveillance. Water Res. 2023 Feb 15;230:119560.
Pedersen K, Kullmann A, Reed BM, et al. Rapid bead-based isolation of SARS-CoV-2 for multiomic viral research. 2022.
Devitt G, Johnson PB, Hanrahan N, et al. Mechanisms of SARS-CoV-2 Inactivation Using UVC Laser Radiation. ACS Photonics. 2024 Jan 17;11(1):42-52.
Carpentier G. Dot Blot Analyser, code & documentation, ImageJ News. September 2008.
Schneider CA, Rasband WS, Eliceiri KW. NIH Image to ImageJ: 25 years of image analysis. Nat Methods. 2012 Jul;9(7):671-675.
Wakankar MS, Krishnasastry MV, Jaokar TM, et al. Solution and in silico studies on the recombinant lectin from Cicer arietinum seeds. Int J Biol Macromol. 2013 May;56:149-55.
Gautam AK, Srivastava N, Nagar DP, et al. Biochemical and functional properties of a lectin purified from the seeds of Cicer arietinum L. 3 Biotech. 2018 Jun;8(6):272.
Zhang J, Xiao T, Cai Y, et al. Structure of SARS-CoV-2 spike protein. Curr Opin Virol. 2021 Oct;50:173-182.
Huang C, Tan Z, Zhao K, et al. The effect of N-glycosylation of SARS-CoV-2 spike protein on the virus interaction with the host cell ACE2 receptor. iScience. 2021 Nov 19;24(11):103272.
Seničar M, Roubinet B, Daniellou R, et al. Lectin Analysis of SARS-CoV-2-Positive Nasopharyngeal Samples Using GLYcoPROFILE(®) Technology Platform. Diagnostics (Basel). 2022 Nov 18;12(11).
Wuo MG, Dugan AE, Halim M, et al. Lectin Fingerprinting Distinguishes Antibody Neutralization in SARS-CoV-2. ACS Central Science. 2023 2023/05/24;9(5):947-956.
Hoffmann D, Mereiter S, Jin Oh Y, et al. Identification of lectin receptors for conserved SARS-CoV-2 glycosylation sites. Embo j. 2021 Oct 1;40(19):e108375.
Zheng L, Wang K, Chen M, et al. Characterization and function of glycans on the spike proteins of SARS-CoV-2 variants of concern. Microbiology Spectrum. 2022;10(6):e03120-22.
Allen JD, Chawla H, Samsudin F, et al. Site-Specific Steric Control of SARS-CoV-2 Spike Glycosylation. Biochemistry. 2021 Jul 13;60(27):2153-2169.
Shajahan A, Pepi LE, Rouhani DS, et al. Glycosylation of SARS-CoV-2: structural and functional insights. Anal Bioanal Chem. 2021 Dec;413(29):7179-7193.
Chalk R, Greenland WEP, Moreira T, et al. Identification, mapping and relative quantitation of SARS-CoV-2 Spike glycopeptides by Mass-Retention Time Fingerprinting. Commun Biol. 2021 Aug 3;4(1):934.
Cho BG, Gautam S, Peng W, et al. Direct Comparison of N-Glycans and Their Isomers Derived from Spike Glycoprotein 1 of MERS-CoV, SARS-CoV-1, and SARS-CoV-2. J Proteome Res. 2021 Sep 3;20(9):4357-4365.
Eldrid CFS, Allen JD, Newby ML, et al. Suppression of O-Linked Glycosylation of the SARS-CoV-2 Spike by Quaternary Structural Restraints. Anal Chem. 2021 Nov 2;93(43):14392-14400.
Bonomelli C, Doores KJ, Dunlop DC, et al. The glycan shield of HIV is predominantly oligomannose independently of production system or viral clade. PloS one. 2011;6(8):e23521.
Waheed AA, Gitzen A, Swiderski M, et al. High-Mannose But Not Complex-Type Glycosylation of Tetherin Is Required for Restriction of HIV-1 Release. Viruses. 2018;10(1):26.
Crispin M, Ward AB, Wilson IA. Structure and immune recognition of the HIV glycan shield. Annual review of biophysics. 2018;47(1):499-523.
Keyaerts E, Vijgen L, Pannecouque C, et al. Plant lectins are potent inhibitors of coronaviruses by interfering with two targets in the viral replication cycle. Antiviral Res. 2007 Sep;75(3):179-87.
Du L, Bouzidi MS, Gala A, et al. Human galectin-9 potently enhances SARS-CoV-2 replication and inflammation in airway epithelial cells. Journal of Molecular Cell Biology. 2023;15(4).
Sigamani A, Mayo K, Chen-Walden H, et al. Galectin approach to lower covid transmission - Drug Development for clinical use. medRxiv; 2022.
Sigamani A, Mayo KH, Miller MC, et al. An Oral Galectin Inhibitor in COVID-19-A Phase II Randomized Controlled Trial. Vaccines (Basel). 2023 Mar 25;11(4).
Tian Y, Parsons LM, Jankowska E, et al. Site-Specific Glycosylation Patterns of the SARS-CoV-2 Spike Protein Derived From Recombinant Protein and Viral WA1 and D614G Strains. Front Chem. 2021;9:767448.
Bains A, Guan W, LiWang PJ. The Effect of Select SARS-CoV-2 N-Linked Glycan and Variant of Concern Spike Protein Mutations on C-Type Lectin-Receptor-Mediated Infection. Viruses. 2023;15(9):1901.
Parray ZA, Hassan MI, Ahmad F, et al. Amphiphilic nature of polyethylene glycols and their role in medical research. Polymer Testing. 2020 2020/02/01/;82:106316.
Niu Z, Zhang P, Rafailovich M, et al. Molecular Dynamics Modeling of the SARS-CoV-2 Spike Protein at PH2 Through PH11.5. Research Square; 2021.
Gstottner C, Zhang T, Resemann A, et al. Structural and Functional Characterization of SARS-CoV-2 RBD Domains Produced in Mammalian Cells. Anal Chem. 2021 May 4;93(17):6839-6847.
Watanabe Y, Bowden TA, Wilson IA, et al. Exploitation of glycosylation in enveloped virus pathobiology. Biochim Biophys Acta Gen Subj. 2019 Oct;1863(10):1480-1497.
Cheng K, Zhou Y, Neelamegham S. DrawGlycan-SNFG: a robust tool to render glycans and glycopeptides with fragmentation information. Glycobiology. 2016;27(3):200-205.

There is NO Competing Interest.

SupplementaryData213Aug2024.zip
Supplementary Data 1
SupplementaryData123July2024.7z
Supplementary Data 2
SupplementaryMaterial13August2024.pdf
Supplementary Figures
SupplementaryMaterial13August2024.pdf

Download PDF

Version 1

posted

You are reading this latest preprint version

SENSING SWEET SPOTS: A comprehensive study of SARS-CoV-2-binding lectins

Status:

Version 1

Abstract

Figures

Introduction

Material and methods

Results

Discussion

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1