To express the antigen in YSD system, asparagine (Asn-N) was replaced with glutamine (Gln-Q) in six N-glycosylation sites in HA1 protein gene sequence of influenza virus A H1N1/PR8 (978 bp), (accession number EF467821.1) (N10Q, N11Q, N23Q, N268Q, N286Q, and N320Q) before synthesis. To amplify HA1 sequences, a specific primer pair conjugated with NheI and BamHI enzyme sites (forward, aaaGCTAGCGACACAATATGTATAGGCTAC, reverse, aaaGGATCCTCTGGATTGAATGGACGGC) was used in PCR. The HA1-PCR product was treated with enzymes NheI and BamHI (NEB, United States) and was introduced to empty pCTCON plasmid (ampicillin-resistant). Previously, Saccharomyces cerevisiae EBY100 was used as a model for YSD (Cho et al., 2020; Hoang et al., 2022). In brief, the plasmids were transformed to EYB100 using electroporation (BioRad, United States). EBY100 harbored the antigen plasmid, HA1::YSD, which was selected in SD media without tryptophan supplementation (Clontech, Japan). The antigen was expressed in SGCAA media containing 2% galactose. The HA1::YSD expression was analyzed in six different colonies by comparing with negative control (EBY100 only) using Western blotting and enzyme-linked immunosorbent assay (ELISA) with primary anti-cMyc antibodies (Invitrogen, Waltham, Massachusetts, United States). Overall, 20 μL yeast cells were used in Western blotting, while the yeast dilution at an OD of 0.4–0.6 (100 μL/well) was added onto a maxibinding immunoplate (SPL Life Sciences, Republic of Korea). The HA1::YSD expression was validated three times in every batch of the experiment.

We used a previously described bio-panning technique to isolate HA1-specific candidates from human scFv libraries (Hoang et al., 2022). In brief, HA1::YSD was used as a target antigen to perform bio-panning with phage display scFv libraries. The scFv libraries (Tomlinson I + J) were expressed on phage using XL1 E. coli cells (tetracycline-resistant) and M13K07 helper phage. To isolate candidate-specific HA1::YSD, bio-panning was used as previously described (Lee et al., 2007; Cho et al., 2020; Hoang et al., 2022). In brief, the phage-displayed scFv libraries in a blocking buffer (3% BSA in Tris-buffered saline containing 0.1% (v/v) Tween 20 (TBS-T)) were first added to a 96-well maxibinding immunoplate (SPL Life Sciences, Republic of Korea) coated with PBS as a first negative selection. After incubation for 2 h at room temperature (RT, 25°C), the supernatant (SPNT) with non-binding scFv phages was transferred to the negative plate (EBY100 coated plate). The plate was incubated overnight at 4°C. Continuously, the non-binding SPNT phages were transferred to positive plate (HA1::YSD-coated plates) for 2 h at RT. Following the washing steps, the binding scFv phages were eluted using 100 μL/well of 100 mM triethylamine solution. The phages were amplified and used for the next round of bio-panning. After three rounds of bio-panning, the screening of HA1-specific scFv (HA1::scFv) was completed.

Phage ELISA was used to measure the affinity of the candidates to the positive samples. XL-1 E. coli blue at an OD₆₀₀ of 0.6 was infected with the HA1::scFv-isolated phages, which spread across the LB agar plates. The phage colonies were randomly selected and cultured in 96-well plates (SPL Life Sciences, Republic of Korea). The HA1::scFv was expressed on phages with the addition of M13K07 helper phage in 2TY growth medium with 0.1% glucose at 30°C overnight. The SPNT was collected and transferred to HA1::YSD-coated plate for 2 h at RT. After washing five times with TBS-T, the plate was incubated for 1 h at RT with a 1:1,000 dilution of anti-M13 HRP-conjugated antibodies (Sino biological, China). The plate was rinsed with TBS-T five times following the addition of TMB substrate solutions (GenDEPOT, United States). Before measuring at an absorbance of 450 nm, we added 1 M sulfuric acid to the plate.

The selected HA1-specific candidates were identified by PCR with a specific primer pair, LMB3, 5′-CAGGAAACAGCTATGAC-3′, and pHEN, 5′-CTATGCGGCCCCATTCA-3′ using 2X premix (Takara, Japan). The clones with 900 or 400 bp bands were sequenced using Macrogen (Republic of Korea). IgBlast tool from NCBI was used to blast the sequence of the CDRs and FR of each candidate (Martin, 1996; Johnson and Wu, 2000).

All the proteins in this research were expressed as soluble in E. coli. The sequences were introduced into pIg20 vector with the XmaI and NcoI enzymes (NEB, United States) that contain protein A at their C terminal and PhoA leader signal peptide at their N terminal. The expressed plasmids were transformed to E. coli BL21 (DE3 pLysE). The cells were cultured in Luria–Bertani (LB) broth with 100 μg/mL ampicillin and 25 μg/mL chloramphenicol at 37°C till OD ₆₀₀ reached approximately 0.8. The protein expression was induced by adding isopropyl 1-thiol-b-D galactopyranoside (IPTG) at 1 mM final concentration for 18 h at 25°C. Cell culture SPNT was filtered before being loaded into a column filled with IgG sepharose 6 fast flow resin (GE Healthcare, Chicago, United States). Next, the resin was washed by five bed volumes of PBS and five bed volumes of 5 mM ammonium acetate (pH 5.5). Later, the proteins were eluted from the resin with 10 bed volumes of 0.1 M acetic acid (pH 3.4) and neutralized by 1 M Tris–HCl (pH 9). The proteins were concentrated using Amicon® Ultra-15 Centrifugal Filter Units, 10 kDa (Merck, New Jersey, United States) and exchanged with PBS buffer, pH 7.4. The protein concentration was measured at OD₂₈₀ with each protein’s extinction coefficient. The protein purity was verified by being loaded on the SDS page with Coomassie blue staining or Western blotting using a 1:3,000 dilution of anti-6X His tag (Abcam, United Kingdom).

To obtain vRNPs directly from influenza virions, the virus (1 × 10⁵ PFU) stationed in TBS buffer was incubated at 50°C for 30 min or treated with 0.1% Triton X100 at room temperature. The mixture was then added to 1 μg of 3D8 scFv protein in the presence of Mg²⁺ for 1 h at 37°C; DW and BSA (1 μg) were used as negative controls. These samples were used as template for a one-step reverse transcription (RT)-PCR using SuPrimeScript RT-PCR premix (Genet Bio, Daejeon, Republic of Korea) to detect eight gene segments (HA, NA, NP, M1, PB1, PB2, PA, and NS1) of the influenza virus with a band length of approximately 500 bp by their specific primers (Supplementary Table S1), following the instructions of the kit, 50°C/15 min, 95°C/5 min, 40 cycles (95°C/20 s, 60°C/30 s, and 72°C/30 s), and 72°C/10 min. The PCR products were loaded to 1% agarose gel containing ethidium bromide.

Influenza virions were diluted in PBS and coated onto an maxibinding immunoplate (SPL Life Sciences, Republic of Korea) at 4°C overnight (1 × 10⁵ plaque-forming unit PFU/well). The NVLH8 protein (100 μL/well) in a 2X serial dilution (starting from the 100 μg/well) was added to the plate after incubation with a blocking buffer (5% skim milk in TBS-T buffer) for 2 h at RT. Following an incubation at 37°C for 1 h, 100 μL anti-6X His tag antibodies at a 1:3,000 dilution (Abcam, United Kingdom) were added to the TBS-T-washed plate for 2 h at RT. The virions coated in the plate were validated using polyclonal rabbit anti-HA antibodies (1:3,000 dilution; Invitrogen, United States). Goat anti-mouse IgG H&L (HRP) (1:5,000 dilution, Abcam, United Kingdom) and goat anti-rabbit IgG-HRP-conjugated antibodies (1:5,000 dilution, Invitrogen, United States) were added and incubated for 2 h at RT. After adding TMB and 1 M sulfuric acid, the plate was ready at OD450. The experiment was reproduced three different times independently. Each batch was designed with five data points, and three data points were chosen to be analyzed.

To titer the virus and evaluate the antiviral activity, we performed a plaque assay. In brief, MDCK cells were grown in 1 × 10⁶ cells/well in six-well plates (SPL Life Sciences, Republic of Korea) to fully confluence in Eagle’s minimal essential medium (Hyclone, United States) supplemented with 10% fetal bovine serum (Gibco, United States) and 0.1% antibiotic-antimycotic (Thermo Fisher Scientific, United States). The cells were incubated with the influenza virus for 1 h at 37°C. For the neutralizing test, serial dilutions of NVLH8 protein (0, 0.1, 1, 10, and 100 μg/mL) were pre-mixed with H1N1/PR8 viruses for 24 h at 37°C before the incubation step. The plate was overlaid with 1% Seaplaque agarose containing 1 μg/mL TPCK-trypsin in DMEM after withdrawing the inoculum. The plaques were stained with 0.5% crystal violet and counted after 3 days of incubation at 37°C. The assay was performed three times with different batches of protein expression. The data were expressed as the percentage of PFUs.

To test neutralization, influenza virus A H1N1/PR8, NVLH8 (100 μg/mL) was incubated with the viruses for 24 h at 37°C. The mixture was then incubated in MDCK cells (MOI 0.1) for 1 h at 37°C. After the removal of protein/virus complexes, the cells were cultured in virus growth medium [MEM-free medium supplemented with 0.2% BSA and TPCK-treated trypsin (1 μg/mL)]. The cells were harvested after 2, 4, 6, 8, 12, and 24 h of virus challenge and stored at −20°C for further RNA extraction.

To investigate the antiviral activity of 3D8 scFv post-treatment, the MDCK cells infected with H1N1/PR8 (MOI 0.1) were cultured in virus growth medium containing 3 μM of 3D8 scFv.

The neutralization activity of NVLH8 was analyzed using ICC. The ICC assay was conducted as described in a previous report (Hoang et al., 2022). In brief, following a neutralization assay of 24 h post-virus/antibody challenge to MDCK cells in 8-well chamber slides (SPL Life Sciences, Republic of Korea), the slides were rinsed with PBS. Ice-cold methanol was added for 15 min before the cells were permeabilized using Intracellular Staining Perm Wash Buffer (Biolegend, United States). After the cells were washed, a blocking buffer (1% BSA and glycine in PBST buffer) was added for 1 h. Polyclonal rabbit anti-HA antibodies at a 1:1,000 dilution (Invitrogen, United States) were added to detect the influenza viral HA protein, followed by an incubation with goat anti-rabbit IgG Alexa fluor 647 (1:1,000 dilution; Abcam, Cambridge, United Kingdom). The signals in the cells were visualized using a Zeiss LSM 900 confocal microscope with a 40X objective. The nucleus was stained with DAPI (LSbio, United States). The relative intensity percentage was calculated by dividing HA protein intensity (red signal) by DAPI signal (blue signal) in the same image and referenced to untreated samples (100%) with three different images.

Immunocytochemistry was also used to visualize the localization of neutralizing Abs. in cells. NVLH8 protein was premixed with H1N1/PR8 (MOI 10) at 300 μg/mL for 24 h at 37°C, and the protein at 300 μg/mL was used as a control (without the virus incubation). The virus/protein mixture was incubated with A549 cells for 1 h at 37°C and replaced with RPMI medium supplemented with BSA 0.2% and TPCK-trypsin 1 μg/mL. After 4 h of incubation, the cells were processed using the same method but with different antibodies. The cells were incubated with mouse monoclonal anti protein A antibodies (1:1,000; Sigma, United States) and goat anti-mouse IgG Alexa fluor 488 (1:1,000 dilution; Abcam, Cambridge, United Kingdom).

To evaluate the penetration of 3D8 scFv protein into cytoplasm using ICC, we treated MDCK cells with 3D8 scFv (5 μM) for 0.5, 1, 3, 6, 12, 24, and 48 h. The protein was detected using 3D8-specific antibodies (Abclon, #3B3, Incheon, Republic of Korea) at 1:1,000 dilution and goat anti-mouse IgG Alexa fluor 488 (1:1,000 dilution; Abcam, Cambridge, United Kingdom).

Hemagglutination (HA) assay was carried out to titer the influenza A viruses using HA units on chicken red blood cells (cRBCs) (Innovative Research Inc., Novi, MI, United States), as described previously (Kaufmann et al., 2017). Overall, 50 μL of cRBCs diluted to 1% PBS was distributed into 2-fold dilution of 25 μL viruses in 25 μL PBS followed by an incubation at room temperature for 1 h. The lowest virus titers that inhibit cRBC precipitation were determined at 1 HA units. Two further dilutions of the HA titration were defined as 4-fold HA units (4 HA) which were used for hemagglutination-inhibition (HI) assay.

For the HI assay, to determine inhibition activity of specific antibodies to HA antigen, 4 HA of 25 μL of influenza viruses were added to 25 μL of 2-fold serial dilution of five proteins, started at 1,000 μg/mL of NVLH8 in an immunology plate (SPL Life Sciences, Pocheon-si, Republic of Korea), followed by an incubation at RT for 1 h. After the addition of 50 μL of 1% cRBCs, the plate was incubated for 1 h/RT. The HI titer was determined at the lowest amount of protein (μg/mL) that inhibited hemagglutination by direct visualization.

Influenza virus A H1N1/PR8 at 2 × 10⁴ PFU/mL was neutralized using 100 μg of NVLH8 for 24 h at 37°C. The mixture was then incubated with MDCK cells (MOI 0.2) for 1 h at 37°C. Followed by the removal of the infection medium, the medium was changed to virus growth medium with 3D8 scFv (3 μM) and then incubated at 37°C with 5% CO₂ for 18 h. Single treatment either with NVLH8 or 3D8 scFv alone was conducted at the same time as non-treated sample. The cells or SPNT were collected and stored at −20°C for further analysis. While the cells were used for RNA extraction and protein extraction with RTqPCR and Western blotting, respectively, the SPNT was subjected to a plaque assay. The protein was harvested using RIPA buffer to lyse the cells, according to the manufacturer’s protocol (Biosolution, Republic of Korea). Western blotting was performed with 20 μg of protein from each sample using influenza A M2 polyclonal antibodies, polyclonal anti-HA antibodies (1:5,000 dilution, Invitrogen, United States), and monoclonal anti-GAPDH antibodies (1:100 dilution, Santa Cruz Biotechnology, United States) as the primary antibodies.

Total RNA was extracted from the harvested cells using the TRI reagent (MRC, United States) with chloroform and iso-propanol method. To measure the vRNA, cRNA, and mRNA of influenza, strand-specific qPCR was conducted. cDNAs specific to vRNA, cRNA, or mRNA of HA and NP genes were synthesized by specific primers using superscript IV first strand synthesis system (Thermo Fisher Scientific, USA), as shown in Supplementary Table S2, following the manufacturer’s protocol. The cDNAs were used as a template for qPCR using SYBR Premix Ex Taq and the Rotor-Gene Q system with the strand-specific primers to HA and NP genes (Supplementary Table S3). Data were analyzed using Rotor-Gene Q series software version 2.3.1. On the other hand, one-step RT-qPCR was performed on the extracted RNA using Accupower GreeenStar RT-qPCR Premix and Master Mix (Bioneer, Republic of Korea) and Rotor-Gene Q System (Qiagen) to measure the antiviral effect of the candidates. The relative expression level of each RNA type in 3D8 scFv-treated groups was normalized to that of the untreated groups (H1N1/PR8) (calibrated as 1). The primers are shown in Supplementary Table S4. GAPDH was amplified as an internal control and used for relative expression analysis.

The figures were presented in GraphPad Prism 8.0 software (GraphPad Software, United States). Statistical significance (asterisks) was determined using a one-way ANOVA, unpaired t-test, (ns: non-significant, ^*p < 0.05, ^**p < 0.001, ^***p < 0.0005, and ^****p < 0.0001). Data are shown as the mean with standard deviation (SD) of triplicate samples.

To optimize antigen expression in YSD, the removal of N-glycosylation sites in the sequences is required. Six N-glycosylation sites in HA1 protein gene sequence of influenza virus A H1N1/PR8 (978 bp) were changed from asparagine to glutamine (N to Q) (Supplementary Figure S1A). The amplified HA1-DNA fragment was cloned into an YSD expression vector (Supplementary Figures S1B,C). Six different yeast colonies expressed HA1 antigen successfully in contrast with the negative control (EBY100), which was confirmed using Western blotting and ELISA (Supplementary Figures S1D,E). Three rounds of bio-panning were performed, with the antigen binding affinities of scFv candidates to positive samples (HA1::YSD) increasing with each round, compared with negative controls (EBY100 yeast) (Figure 1A). Continuously, the nine clones that showed the highest affinity to HA1::YSD by phage ELISA were isolated (Figure 1B). Detection of the clones using PCR with specific primers indicated that these candidates appeared in different forms, either in scFv (900 bp) or in single-domain V_H or V_L (400 bp) (Figure 1C). The CDRs and FRs of the candidates were identified using the IgBlast Kabat antibody sequence database (Ye et al., 2013). Several stop codons existed in HA1-specific scFv (H1, H2, and H6) in CDRs and FRs (data not shown). Interestingly, H3, H4, and H5 scFv candidates shared the same amino acid sequences without any stop codons in CDRs and FRs, indicating the high frequency of the candidates and the diversity of the libraries. Three single-domain clones (H7, H8, and H9) with 400 bp sizes belonged to the VL sequences and shared differences in the first amino acid at CDR2 and a few amino acids at CDR3 (Figure 1D).

Because H3, H4, and H5 scFv candidates had identified sequences, one of them, H4 (renamed as NscFvH4), was picked to be produced at a protein level together with three different single-domain V_L, NVLH7, NVLH8, and NVLH9. These were cloned into pIg20 vector containing PhoA signal peptide and protein A (Figures 2A,B). The proteins were expressed in soluble form in E. coli and purified using IgG sepharose. The purified proteins were verified using Coomassie blue staining and Western blotting with anti-6X His tag antibodies sized 30.1 kDa for NscFvH4, and 22.5 kDa for NVLH7, NVLH8, and NVLH9 (Figure 2C). Each protein was obtained in a different yield and purity (Table 1). Although NscFvH4 and NVLH8 had similar purity (88 vs. 87%, respectively), NVLH8 was obtained at 3 mg/L, three times greater than NscFvH4, at 1 mg/L. In contrast, NVLH7 and NVLH9 were obtained at a low yield (0.3 and 0.5 mg/L, respectively) and at 70% purity despite similar constructs as NVLH8 (Figure 2D). The function of recombinant proteins was assessed using virion ELISA with a 2X serial dilution for direct antigen binding affinity to the influenza virus particle H1N1/PR8. Among the candidates, in a concentration-dependent manner, NVLH8 demonstrated the highest binding strength to virus H1N1/PR8 particles. NVLH9 demonstrated a weak interaction with virus particles, whereas NscFvH4 and NVLH7 did not demonstrate the same (Figure 2D).

To test whether NVLH8 neutralized virus infection, we performed a plaque inhibition assay. The protein caused a decrease in plaque number in a concentration-dependent manner. Specifically, NVLH8 at 0.1, 1, 10, and 100 μg/mL neutralized viruses with reductions of 96, 88, 57, and 38%, respectively, showing the EC50 value of 29.45 μg/mL (Figure 3A; Supplementary Figure S4). Moreover, the viral HA protein signal (red) in samples neutralized with 100 μg of single-domain NVLH8 was not visible when compared with positive samples using ICC (Figure 3B). The HA protein intensity decreased significantly in the presence of the protein (Figure 3C; Supplementary Figure S5).

The neutralizing NVLH8 exhibited binding affinity to virus particles (Figure 2D), resulting in virus inhibition (Figure 3). Therefore, we studied the neutralization mechanism of the single-domain V_L NVLH8. First, the protein exhibited HI activity at 31.25 μg/mL (Figure 4A). The findings indicated that NVLH8 had affinity for the HA1 subunit at a receptor binding site, which may prevent the viral binding to cell-surface receptors. Since the protein demonstrated a consistent relationship between the HI activity and the neutralization efficacy via plaque reduction, we hypothesized two possible neutralization mechanisms of the protein: (a) it inhibits the viral attachment to receptors or (b) it prevents the un-coating steps during viral entry by interfering with membrane fusion.

To figure out whether NVLH8 inhibits the virus entry or viral gene releasing steps, we quantified the vRNA (vHA) number of H1N1/PR8 levels in the infected cells in the early stages (2, 4, and 6 h post infection; hpi) (Figure 4B). At 2 and 4 hpi, there were no differences in HA vRNA levels between the NVLH8 treatment and the control (H1N1/PR8). The protein’s inhibitory activity was clearly demonstrated at 6 hpi by 45% reduction in vHA level. We then used NVLH8 to observe the time course of HA and NP expression in neutralized virus infection (Figure 4C). The HA and NP viral genomes decreased from 6 to 8–12 hpi. It is suggested that the viruses which were not inhibited by NVLH8 were able to replicate normally, which is why at 24 hpi, the HA and NP levels were slightly higher than at 12 hpi.

The results supported the hypothesis that the candidates inhibited the viral genome releasing steps, implying the blockage of membrane fusion steps rather than the entry steps. To further validate this, the virion binding proteins internalized with the virus were postulated. We tested the protein NVLH8 localization in the cells in the presence of viruses. To better visualize the protein, protein NVLH8 (300 μg/mL) was pre-incubated with H1N1/PR8 at MOI 10 for 24 h before its incubation with the cells. In addition, to prevent the degradation of the protein, the cells were fixed after 4 hpi. Using immunofluorescence staining, the bound NVLH8 (green) was found at 300 μg/mL dose in the presence of the viruses compared with no virus-treated samples (NVLH8) (Figure 4D). The finding that the Abs localized in the cells after being premixed with the viruses suggested that such a candidate prevented viral infection by inhibiting viral genome release.

We first evaluated the antiviral activity of 3D8 scFv after viral infection. The penetration of 3D8 scFv into MDCK cells was investigated. 3D8 scFv protein (green) was taken up at a very early time (0.5 h) and peaked at 6–12 h and remained in the cytoplasm for up to 48 h (Supplementary Figure S2A). Subsequently, the antiviral activity of 3D8 scFv post-treatment against H1N1/PR8 was evaluated in post-viral infection through the decrease of viral RNA levels of HA, M1, NP, and virus titer (Supplementary Figures S2B,C).

Our previous study reported the antiviral activity of 3D8 scFv against influenza H1N1/H275Y virus through hydrolyzing viral RNA in mRNA and vRNPs and cRNPs form (Lee et al., 2022). Here, we strongly agreed that 3D8 scFv targeted not only mRNA (naked RNA) but also vRNA and cRNA in RNP forms (vRNPs or cRNPs) of H1N1/PR8. Most of those bands were invisible or blurry in 3D8 scFv-treated samples when compared with negative controls (DW) or BSA (Figure 5B). However, those bands were detected in samples without non-0.1% Triton X100 even in the presence of 3D8 scFv (Figure 5A). The thickness of those 3D8 scFv-treated bands was less than that of the control sample bands. During the reverse transcription step at 50°C, vRNPs were assumed to be partially discharged, and 3D8 scFv, which was stable at high temperature, had bound and cleaved the free vRNPs. Consequently, the virus particles were incubated at 50°C for 30 min. The heating process could also release vRNPs that were digested by 3D8 scFv. However, the vRNPs were not released as much as in the 0.1% Triton X100-treated condition (Figure 5C). A post-viral infection at various times with 3D8 scFv post-treatment was conducted to analyze the changes in each type of RNA. At specific time points during virus replication, the expression of the three RNAs of the HA and NP segments was measured (Figure 5D). Particularly, reductions were observed clearly at 12 hpi (~30–40%) and 24 hpi (80%) for all three types of RNAs of HA and NP segments (Figure 6D) but not at earlier time points. The data suggested that 12 hpi was critical for 3D8 scFv antiviral activity when 3D8 scFv had fully penetrated and was released into the cytoplasm to bind and digest the three viral RNAs.

A synergistic efficacy of NVLH8 neutralizing Abs (i.e., targeting early stages) and 3D8 scFv (intermediate and late stage) was investigated. A combination treatment (Figure 6A) in which the neutralized H1N1/PR8 virus with NVLH8 was infected into MDCK cells and then treated with 3D8 scFv in virus growth medium was used. The antiviral activity of a combination treatment (3D8 scFv and single neutralizing Abs) was compared with that of a single treatment using intracellular viral gene expression (HA and NP) (Figure 6B). Individually, while 3D8 scFv caused relative viral gene reduction of approximately 56% (HA) and 65% (NP), neutralizing NVLH8 decreased HA and NP gene by 57 and 63%, respectively, and a combination treatment reduced HA and NP by 84 and 88%, respectively. These results suggested that combining a viral genome hydrolyzing 3D8 scFv with an influenza neutralizing Abs amplified the antiviral efficacy. The synergistic efficacy was consistently demonstrated in the reduction of virus titer using plaque assay (Figure 6C, Supplementary Figure S6) and viral protein using Western blotting (Figure 6D).

Therefore, a synergistic effect of a combination treatment of different therapeutic targets to different stages of influenza viral life cycles is proposed (Figure 7). NVLH8 neutralizing Abs bound to the influenza virus, internalized together into cells, and prevented the viral genome releasing steps, whereas 3D8 scFv localized in the cell cytoplasm by caveolae-mediated endocytosis and degraded viral genome in the intermediate and late stages of the life cycle of the virus. A synergistic effect can be achieved by using neutralized Abs-specific H1N1/PR8 as a vertical effect at viral entry steps, the action of 3D8 scFv hydrolyzing all types of IAV RNAs/RNP at viral protein biosynthesis, and exclusion of virus from the cytoplasm of infected cells as a horizontal effect.