TMPRSS2 Activates Hemagglutinin-Esterase Glycoprotein of Influenza C Virus

ABSTRACT

Influenza C virus (ICV) has only one kind of spike protein, the hemagglutinin-esterase (HE) glycoprotein. HE functions similarly to hemagglutinin (HA) and neuraminidase of the influenza A and B viruses (IAV and IBV, respectively). It has a monobasic site, which is cleaved by some host enzymes. The cleavage is essential to activating the virus, but the enzyme or enzymes in the respiratory tract have not been identified. This study investigated whether the host serine proteases, transmembrane protease serine S1 member 2 (TMPRSS2) and human airway trypsin-like protease (HAT), which reportedly cleave HA of IAV/IBV, are involved in HE cleavage. We established TMPRSS2- and HAT-expressing MDCK cells (MDCK-TMPRSS2 and MDCK-HAT). ICV showed multicycle replication with HE cleavage without trypsin in MDCK-TMPRSS2 cells as well as IAV did. The HE cleavage and multicycle replication did not appear in MDCK-HAT cells infected with ICV without trypsin, while HA cleavage and multistep growth of IAV appeared in the cells. Amino acid sequences of the HE cleavage site in 352 ICV strains were completely preserved. Camostat and nafamostat suppressed the growth of ICV and IAV in human nasal surface epithelial (HNE) cells. Therefore, this study revealed that, at least, TMPRSS2 is involved in HE cleavage and suggested that nafamostat could be a candidate for therapeutic drugs for ICV infection.

IMPORTANCE Influenza C virus (ICV) is a pathogen that causes acute respiratory illness, mostly in children, but there are no anti-ICV drugs. ICV has only one kind of spike protein, the hemagglutinin-esterase (HE) glycoprotein on the virion surface, which possesses receptor-binding, receptor-destroying, and membrane fusion activities. The HE cleavage is essential for the virus to be activated, but the enzyme or enzymes in the respiratory tract have not been identified. This study revealed that transmembrane protease serine S1 member 2 (TMPRSS2), and not human airway trypsin-like protease (HAT), is involved in HE cleavage. This is a novel study on the host enzymes involved in HE cleavage, and the result suggests that the host enzymes, such as TMPRSS2, may be a target for therapeutic drugs of ICV infection.

INTRODUCTION

Influenza C virus (ICV) was first isolated from a human with upper respiratory symptoms in 1947 (1). ICV usually causes mild upper respiratory illness, but it can also cause lower respiratory infections, such as bronchitis and pneumonia, particularly in children <2 years old (2). Seroepidemiological studies have revealed that the influenza C virus is widely distributed globally, and recurrent infection with this virus frequently occurs in children and adults (3–8).

ICV is a member of the Orthomyxoviridae family, including the influenza A virus (IAV), influenza B virus (IBV), and influenza D virus. The ICV genome consists of seven RNA segments encoding three polymerase proteins (PB2, PB1, and P3), a hemagglutinin (HA)-esterase (HE) glycoprotein, a nucleoprotein (NP), a matrix (M1) protein, a CM2 protein, and two nonstructural proteins (NS1 and NS2). HE possesses receptor-binding, membrane fusion, and receptor-destroying activities, thus combining the functions of hemagglutinin (HA) and neuraminidase (NA) of IAV and IBV (9). However, HE has been uniquely evolved compared with HA and has a low capability to induce mutations, weak and infrequent selective bottlenecks, and a low evolutionary rate (10–13).

HA is synthesized as a precursor protein, HA0, and is cleaved by host cell proteases into the subunits HA1 and HA2 to gain its fusogenic capacity (14–16). Similarly, the HE present in infectious virus particles is composed of two subunits. The N-terminal 432 amino acids are the HE1 polypeptide: the remaining 209 amino acids, including the hydrophobic fusion peptide, the transmembrane domain, and the cytoplasmic tail, are called HE2 (17, 18). HA and HE proteins are cleaved into HA1 and HA2 and HE1 and HE2 within a membrane-proximal surface loop at Arg/Lys↓Gly and Arg↓Ile residues, respectively (17, 19). (The downward arrow represents the cleavage site.) For highly pathogenic IAV strains, the HA cleavage site consists of a polybasic amino acid forming the characteristic protease recognition motif Arg₄Xaa₃Arg/Lys₂Arg₁↓ for ubiquitous intracellular furin-like serine proteases, such as furin and PC5/6 (16, 20–22). Alternatively, a monobasic Arg (or rarely Lys) residue found at the HA cleavage site of low-pathogenicity IAV and seasonal human influenza virus and the HE cleavage site of ICV is not cleaved by such furin-like serine proteases (23, 24).

Homma and Ohuchi first reported that the Sendai virus’s surface fusion glycoprotein was cleaved by trypsin in vitro (25, 26). Since then, trypsin and other trypsin-like proteases, such as plasmin and club cell tryptase, have been identified as influenza virus HA-activating proteases in vitro (14, 15, 27). In a recent in vitro study, several type II transmembrane serine proteases (TTSPs), including transmembrane protease serine S1 member 2 (TMPRSS2) and human airway trypsin-like protease (HAT; also known as TMPRSS11D) in the human respiratory tract, are reportedly involved in HA cleavage of IAV with a monobasic cleavage site. However, the cleavage rate by TMPRSS2 and HAT varies, depending on the difference in HA subtypes (19, 28–32). Similarly, the HA of IBV is activated by TMPRSS2 and HAT (33). Furthermore, in in vivo analysis using a mouse model, TMPRSS2 was found to be a host cell factor essential for viral spread and pathogenesis of H1N1, including the swine flu pandemic 2009 and H3N2 influenza A viruses. However, IBV is still lethal in TMPRSS2 knockout mice and wild-type mice (34, 35). However, the HAT knockout mouse model remains uninvestigated.

As described above, TTSPs that activate HA of IAV and IBV have been analyzed in detail. However, the enzyme expressed in the human respiratory tract catalyzing HE cleavage of ICV has not been identified in vitro or in vivo. Thus, this study investigated whether TMPRSS2 and HAT are involved in the cleavage of HE glycoprotein in vitro and ex vivo.

RESULTS

Establishment of MDCK-TMPRSS2 and MDCK-HAT cells.

IAV and ICV do not undergo multicycle replication in MDCK cells without trypsin because the cell line lacks an endogenous protease that activates HA/HE with monobasic cleavage sites. However, these viruses significantly propagate in the MDCK cell line in the presence of trypsin. For that reason, we attempted to generate a cell line with permanent expression of either TMPRSS2 or HAT. MDCK cells were infected with a lentivirus vector incorporating cloned tmprss2 and tmprss11d cDNA, and then blasticidin-resistant cells were selected. While the original cells did not express tmprss2 and tmprss11d mRNA, each selected cell revealed proper expression of the mRNAs (Fig. 1A). Furthermore, the levels of expression of each serine protease in these cells by indirect immunofluorescence assay (Fig. 1B) were similarly observed. These data confirmed that MDCK-TMPRSS2 and MDCK-HAT cells could be established.

FIG 1.

mRNA and protein expression of each serine protease in MDCK-TMPRSS2 and MDCK-HAT cells. (A) Real-time RT-PCR-quantified expression levels of tmprss2 and tmprss11d mRNA in MDCK, MDCK-TMPRSS2, and MDCK-HAT cells. Each column represents the mean ± SD from triplicate cultures. ND, not detected. (B) Levels of protein expression of TMPRSS2 and HAT (also known as TMPRSS11D) in MDCK, MDCK-TMPRSS2, and MDCK-HAT cells were analyzed by indirect immunofluorescence using specific antibodies and Alexa Fluor 546 (orange)-conjugated secondary antibodies. Nuclei were stained with Hoechst 33342 (blue). Size bars, 20 μm.

Multicycle replication and HE cleavage of ICV by TMPRSS2 as opposed to HAT.

Each MDCK cell was infected with ICV (C/Ann Arbor/1/50) and IAV (A/Sendai-H/N633/09) as a control and cultured with or without trypsin. Viral infectivity titers in these supernatants were measured by 50% tissue culture infective dose (TCID₅₀) endpoint assay instead of plaque assay because ICV does not induce obvious plaques (Fig. 2A). IAV revealed multicycle viral replication without trypsin in both MDCK-TMPRSS2 and MDCK-HAT cells. On the other hand, ICV revealed multicycle viral replication without trypsin in MDCK-TMPRSS2 cells, but not in MDCK-HAT cells.

FIG 2.

Analysis of viral replication and spike protein cleavage in MDCK-TMPRSS2 and MDCK-HAT cells. (A) Each MDCK cell was infected with ICV (C/Ann Arbor/1/50) and IAV (A/Sendai-H/N633/09) at an MOI of 0.01 and then cultured with or without 5 μg/ml trypsin. Viral titers in these supernatants were measured by TCID₅₀ endpoint assay. Each symbol represents the mean ± SD from three independent experiments. HE and HA glycoproteins in the supernatants (B) and whole-cell lysates (C) of each MDCK cells at 48 and 24 h postinfection with ICV (C/Ann Arbor/1/50) and IAV (A/Sendai-H/N633/09) at MOI of 10, respectively, in the presence or absence of 5 μg/ml trypsin were detected by Western blotting. The cleavage rates of HE and HA were quantified from the band density using Image Lab software: % HE cleavage = [HE1/(HE1 + HE0)] × 100, and % HA cleavage = [HA1/(HA1 + HA0)] × 100. Each column represents the mean ± SD from three independent experiments. Significant difference in MDCK cell culture without trypsin: *, P < 0.05; **, P < 0.01; ***, P < 0.005. NS, not significant. hpi, hour postinoculation.

To investigate whether HE was cleaved by TMPRSS2 or HAT, each MDCK cell was infected with ICV and IAV, cultured with or without trypsin. These supernatants and whole-cell lysates were analyzed by Western blotting using anti-HE and anti-HA monoclonal antibodies, respectively (Fig. 2B and C). HA in these supernatants was cleaved by both TMPRSS2 and HAT without trypsin. However, in MDCK-HAT cells, cleavage of HA was very slight compared with that in MDCK-TMPRSS2 cells and was not detected in the whole-cell lysate. In contrast, although cleavage of HE in these supernatants without trypsin was observed in MDCK-TMPRSS2 cells, the results were comparable and not significantly different between MDCK-HAT and MDCK cells. Also, HE in the whole-cell lysate was slightly cleaved by TMPRSS2. These data suggest that TMPRSS2 activates the ICV HE.

Amino acid sequence comparison of the cleavage site.

Since the sites of cleavage by HAT were different between the HE and HA proteins, we analyzed the amino acid sequence of the ICV HE protein cleavage site while comparing it with those of autocatalytic domains of TMPRSS2 and HAT (36, 37) and the cleavage sites of IAV/IBV HA proteins (Table 1). HE proteins possess a monobasic Arg residue at the P1 position of the cleavage site and HA proteins of IAV and IBV. In contrast, Pro and Lys residues at the P4 and P5 positions were characteristic of the HE protein. C/Ann Arbor/1/50, used in this study, is a laboratory strain isolated in 1950 (38). All 352 ICV strains detected worldwide between 1947 and 2018 possessed the completely preserved 10 amino acid residues from the P7 position to the P′3 position in the HE cleavage site (12, 39, 40), suggesting that the same results could be obtained in the laboratory and clinical strains. The clinically isolated ICV strain of C/Yamagata/8/2002 was found to exhibit multicycle viral replication without trypsin in MDCK-TMPRSS2 cells, but not in MDCK-HAT cells (Fig. 3).

TABLE 1.

Residues of the cleavage site in serine proteases and virus membrane fusion proteins

Protein	Residue at positiona:
	P7	P6	P5	P4	P3	P2	P1	↓	P′1	P′2	P′3
Proteases
TMPRSS2 (autocatalytic site)	N	S	S	R	Q	S	R	↓	I	V	G
HAT (autocatalytic site)	I	T	L	S	E	Q	R	↓	I	L	G
Membrane fusion proteins
IAV H1	I/V	P	S	I	Q	S	R	↓	G	L	F
IAV H3	I/V	P	E	K	Q	T	R	↓	G	I/L	F
IBV	A	K	L	L	K	E	R	↓	G	F	F
ICV	V	T	K	P	K	S	R	↓	I	F	G

^aThe arrow indicates the cleavage site.

FIG 3.

Analysis of viral replication in MDCK-TMPRSS2 and MDCK-HAT cells. Each MDCK cell was infected with ICV (C/Yamagata/8/2002) at an MOI of 0.01 and then cultured with or without 5 μg/ml trypsin. Viral titers in these supernatants were measured by TCID₅₀ endpoint assay. Each symbol represents the mean ± SD from three independent experiments. Significant difference in MDCK cell culture without trypsin: **, P < 0.01; ***, P < 0.005.

Suppression of ICV replication in the presence of serine protease inhibitors.

Camostat mesylate (camostat) and nafamostat mesylate (nafamostat) have serine protease inhibitor activity. They inhibited coronavirus envelope protein (S protein) cleavage by TMPRSS2 and HAT (41–45). To confirm that these serine protease inhibitors suppressed multicycle viral replications by TMPRSS2 and HAT without trypsin, we treated MDCK-TMPRSS2 and MDCK-HAT cells with camostat and nafamostat after inoculation with ICV and IAV (Fig. 4). Cytotoxic effects were not observed in these serine protease inhibitor-treated cells (data not shown). Camostat and nafamostat suppressed replications of IAV in MDCK-TMPRSS2 and MDCK-HAT cells. Similarly, camostat and nafamostat also suppressed ICV replication by TMPRSS2. Since multicycle replication and HE cleavage of ICV without trypsin were not detected in MDCK-HAT cells, also the effects of these serine protease inhibitors were not observed (Fig. 4).

FIG 4.

Analysis of serine protease inhibitors for viral replication in MDCK cells. MDCK-TMPRSS2 and MDCK-HAT cells were infected with ICV (C/Ann Arbor/1/50) and IAV (A/Sendai-H/N633/09) at an MOI of 0.01 in the presence of camostat or nafamostat at 10 μg/ml (20 and 19 μM, respectively). Viral titers in these supernatants at 7 days postinoculation were measured by TCID₅₀ endpoint assay. Each column represents the mean ± SD from three independent experiments. NS, not significant. *, P < 0.05; **, P < 0.01; ***, P < 0.005.

Serine protease inhibitors inhibit ICV replication in primary cultures of human nasal surface epithelial cells.

In a previous study, camostat and nafamostat also inhibited replication and HA cleavage of IAV with a monobasic cleavage site in human tracheal epithelial cells and human nasal epithelial (HNE) cells (46, 47). Because ICV causes upper respiratory illness, serine proteases were also examined in the HNE cells. The tmprss2 and tmprss11d mRNAs were expressed in HNE cells (Fig. 5A). The mRNA expression levels of TMPRSS2 in MDCK-TMPRSS2 and HNE cells were comparable. However, expression of tmprss11d mRNA in MDCK-HAT cells was lower than that in HNE cells (Fig. 1A and 5A). HNE cells were treated with camostat and nafamostat after inoculation with ICV and IAV. In a previous study, these protease inhibitors at 10 μg/ml did not show cytotoxicity to HNE cells (46). Camostat and nafamostat suppressed replications of ICV and IAV at 10 μg/ml (Fig. 5B). The suppression effects of these serine protease inhibitors for viral replications were dose-dependently observed (Fig. 5B). Similar to the previous study results for IAV (46), the inhibitory effect of nafamostat on the growth of ICV was higher than that of camostat.

FIG 5.

Analysis of serine protease inhibitors for viral replication in HNE cells. (A) Expression levels of tmprss2 and tmprss11d mRNA in HNE cells were quantified by real-time RT-PCR. Each column represents the mean ± SD from triplicate cultures. (B) HNE cells were infected with ICV (C/Ann Arbor/1/50) and IAV (A/Sendai-H/N633/09) at an MOI of 0.1 in the presence of camostat or nafamostat at the indicated dose, and viral titers in these supernatants at 7 days postinoculation were measured. Each symbol represents the mean ± SD from three independent experiments. Significant difference for inhibitors at 0 μg/ml: *, P < 0.05; **, P < 0.01; ***, P < 0.005. Significant difference between the inhibitors: ‡, P < 0.05; ‡‡, P < 0.01.

DISCUSSION

This study identified that TMPRSS2 was one of the enzymes involved in HE protein cleavage of ICV. Alternatively, HE cleavage by HAT was not observed, while HA cleavage of IAV was done. These differences suggest that the amino acid sequences cleaved by TMPRSS2 and HAT may be different. The substrate for TMPRSS2 has not been characterized, suggesting that broadly utilized trypsin substrates are tolerated. In the study on the difference in the cleavage of the HA subtypes, the conserved cleavage sites that were most efficiently cleaved by TMPRSS2 were a small hydrophobic amino acid in the P4 position and a Ser/Thr in the P2 position. However, these features are not absolute determinants for cleavage (19). Here, we revealed that the HE cleavage site of ICV had the amino acid sequence cleaved by TMPRSS2. Alternatively, the substrate that has been shown to be optimally hydrolyzed by HAT is Arg₄Gln₃Asp₂Arg₁↓ (48). However, proline in the P3 and P4 positions strongly suppresses the proteolysis of peptides by HAT (48). The HE cleavage site sequence of all ICV strains was Pro₄Lys₃Ser₂Arg₁↓, suggesting that the sequence of the HE cleavage site was unrecognized by HAT; therefore, ICV may not be able to use HAT, which is expressed in the respiratory tract for viral replication.

The amount of HA protein cleaved by TMPRSS2 was higher than that by HAT, suggesting that the catalytic activity of TMPRSS2 was higher than that of HAT. TMPRSS2 is expressed on the cell surface and in the trans-Golgi network, while HAT is only expressed on the cell surface (49–51). Since the cleavage of HE and HA by TMPRSS2 was observed in the cell lysate, TMPRSS2 may have high cleavage activity by intracellular expression in our established MDCK-TMPRSS2 cells. ICV and IAV showed efficient proliferation in MDCK-TMPRSS2 cells without trypsin, although the amounts of HA and HE protein cleaved by TMPRSS2 were significantly lower than those by trypsin. These data suggest that these viruses have sufficient infectivity by a slight cleavage of this spike protein.

TMPRSS2 is also involved in the infections of parainfluenza virus, coronavirus, and human metapneumovirus other than the influenza virus (52). Therefore, it can serve as a target for therapeutic drugs for these viruses. Other host enzymes, such as TMPRSS4, TMPRSS13, matriptase, DESC1, KLK5, KLK12, and plasmin, were reported to be involved in the HA cleavage (52). Since the HE cleavage site has no amino acid sequence activated by matriptase and TMPRSS13 (49), it may be cleaved by TMPRSS4, DESC1, KLK5, KLK12, and plasmin. The serine protease inhibitors camostat and nafamostat completely suppressed ICV and IAV growth in MDCK-TMPRSS2 cells and IAV growth in MDCK-HAT cells. Alternatively, nafamostat strongly suppressed the viral growth in HNE cells, whereas camostat showed only weak suppression. Similar to a previous study (46), in this study using HNE cells, the suppression effects of nafamostat on ICV and IAV growth were higher than those of camostat. This result suggests that nafamostat suppressed a wide range of host enzymes and that ICV also utilizes host enzymes other than TMPRSS2, which was inhibited by nafamostat, as well as IAV. However, the spectrum of the serine proteases inhibited by camostat and nafamostat is not well known. Thus, we would like to analyze them in the future.

In this study, we demonstrated that camostat and nafamostat suppressed ICV growth in MDCK-TMPRSS2 and HNE cells. Because these serine protease inhibitors were not specific for TMPRSS2, the extent of involvement of TMPRSS2 in ICV infection was not determined. In vivo studies using TMPRSS2 knockout mice have been conducted for IAV and IBV (34, 35). For ICV proliferation, knockdown experiments, such as by utilizing small interfering RNA (siRNA), or knockout experiments, such as by utilizing CRISPR/Cas9, in vitro or in vivo are warranted.

In conclusion, TMPRSS2 is involved in the HE cleavage and ICV multicycle replication. Therefore, nafamostat could be a candidate as a therapeutic drug for ICV infection.

MATERIALS AND METHODS

Plasmids and transfection.

We cloned a cDNA containing tmprss2 and tmprss11d from Huh-7 cells, a human hepatocarcinoma cell line, and the sequence was confirmed to be the same as those of NM_005656.3 and NM_004262.2, respectively. The cDNA was cloned into a lentivirus vector (pLVSIN, purchased from Clontech) under the control of a cytomegalovirus (CMV) promoter with a blasticidin-resistant gene to make the vesicular stomatitis virus G (VSV-G)-pseudotyped recombinant virus for infection to Madin-Darby canine kidney (MDCK) cells. The infected MDCK cells were incubated with 20 mg/ml blasticidin (Sigma-Aldrich, St. Louis, MO, USA) for 2 weeks. The selected blasticidin-resistant cells that stably expressed TMPRSS2 and HAT (MDCK-TMPRSS2 and MDCK-HAT) were used in further experiments. Those cells were cultured in Eagle’s minimal essential medium (MEM; Sigma-Aldrich) with 10% fetal bovine serum (FBS; Thermo Fisher Scientific, Waltham, MA, USA), 100 U/ml penicillin G, and 100-μg/ml streptomycin at 37°C in a 5% CO₂ atmosphere.

Human nasal epithelial cell culture.

Human nasal surface epithelial (HNE) cells were isolated and cultured in Dulbecco’s modified Eagle’s medium plus nutrient mixture F-12 (DMEM–F-12; Gibco Life Technologies, Grand Island, NY, USA) containing 2% Ultroser G (USG; Pall, Port Washington, NY, USA) serum substitute as described previously (46, 53). Nasal samples for cell culture were obtained from the patients undergoing endoscopic surgery (mean ± standard deviation [SD] age, 55.2 ± 11.6 years; n = 13 [4 females and 9 males]). This study was approved by the Tohoku University Ethics Committee (IRB no. 2016-1-802).

Extraction of RNA and quantitative real-time RT-PCR.

Total RNA was extracted from 350 μl cell lysate cultured in a 96-well plate (Greiner Bio-One, Baden-Württemberg, Germany) using the RNeasy minikit (Qiagen, Venlo, Netherlands). The cDNA was synthesized from the extract using high-capacity cDNA reverse transcription (RT) kits (Applied Biosystems, Foster, CA, USA), following the manufacturer’s instructions. Quantitative real-time RT-PCR was performed in 20-μl reaction mixtures, using gene-specific primers and the IQ Supermix (Bio-Rad Laboratories, Hercules, CA, USA) in the MiniOpticon real-time PCR detection system with CFX Manager software (Bio-Rad Laboratories). Table 2 shows the primer sequences for amplification. The reaction efficiency with each primer set was determined using standard amplification. Target gene expression levels and the expression level of the gene coding for hypoxanthine-guanine phosphoribosyltransferase (hprt) as a reference gene were calculated for each sample using the reaction efficiency. The results were analyzed using a relative quantification procedure and are illustrated as relative expression compared with hprt mRNA expression.

TABLE 2.

Primers for real-time RT-PCR

Gene	Primer (5′→3′)
	Forward	Reverse
Human tmprss2	GCGTGGAAAAACCTCTTAACAATC	ATCCGGCTCCATAGAACATGAAA
Human tmprss11d	CGGCTCAATAATGCCCACCAC	AGTCACGAGGATTAGAGTTGCTTC
Human/dog hprt	CGAGATGTGATGAAGGAGATGGG	ATGTAATCCAGCAGGTCAGCAA

Indirect immunofluorescence assay.

Cells were cultured in a multiwell glass-bottom dish (Matsunami Glass Ind., Ltd., Osaka, Japan), fixed with 4% paraformaldehyde in phosphate-buffered saline (PBS) for 15 min at room temperature, and permeabilized with 0.5% Triton X-100 in PBS for 15 min at room temperature. After that, the mixture was incubated with polyclonal anti-TMPRSS2 (GTX100743; GeneTex, Irvine, CA, USA) or anti-TMPRSS11D (GTX117370; GeneTex) antibody as a primary antibody for 1 h at room temperature and then incubated with Alexa Fluor 546-conjugated goat anti-rabbit IgG (Molecular Probes, Eugene, OR, USA) as a secondary antibody. Nuclei were stained with Hoechst 33342 (Molecular Probes). Cells were observed using an LSM700 laser scanning confocal microscope (Carl Zeiss, Oberkochen, Germany). Image capture, analysis, and processing were performed using Zen v.2011 software (Carl Zeiss) and Photoshop CS5 (Adobe Systems, San Jose, CA, USA).

Viruses.

ICV (C/Ann Arbor/1/50) (54), which had been grown in the embryonated chicken eggs, and IAV H1N1pdm09 (A/Sendai-H/N633/09) (53), which had been passaged four times in MDCK cells, were used in this study. The cell culture supernatants and amniotic fluid were centrifuged and stored at −80°C.

Multicycle replication assays.

Confluent MDCK, MDCK-TMPRSS2, and MDCK-HAT cell monolayers cultured in a 6-well plate were inoculated with ICV and IAV at a multiplicity of infection (MOI) of 0.01 for 1 h. Cells were washed three times and then cultured with or without 5 μg/ml trypsin (type I; Sigma-Aldrich) at 33°C in a 5% CO₂ atmosphere. The culture medium supernatants were harvested every 24 h from 1 to 7 days postinfection. For multicycle replication assays in the presence of serine protease inhibitors, confluent MDCK-TMPRSS2, MDCK-HAT, and HNE cell monolayers cultured in a 96-well plate were inoculated with IAV and ICV at an MOI of 0.01 for 1 h and then washed three times. The assay mixtures were then cultured with an indicative dose of camostat mesylate (Ono Yakuhin Co., Ltd., Osaka, Japan) or nafamostat mesylate (Fujifilm Wako Pure Chemical Corporation, Osaka, Japan) at 33°C in a 5% CO₂ atmosphere. Each supernatant was harvested 7 days later.

Endpoint assays were used to determine viral titers. Confluent monolayers of MDCK cells (Dainippon Pharma, Osaka, Japan) cultured in a 96-well plate were inoculated with 50 μl of 10-fold serially diluted supernatants and centrifuged at 2,000 rpm for 30 min at room temperature. The cells were washed three times and cultured in MEM containing 5 μg/ml trypsin at 33°C in a 5% CO₂ atmosphere for 7 days. To identify positive wells, the cells were monitored by light microscopy as they underwent cell death due to apparent cytopathic effect. We used four wells per dilution. The infectivity titer was calculated using Reed and Muench’s method and expressed as the 50% tissue culture infective dose (TCID₅₀)/ml (55).

Western blotting.

Each confluent MDCK cell monolayer cultured in the 96-well plate was inoculated with ICV and IAV at an MOI of 10 for 1 h and then cultured with or without 5 μg/ml trypsin (Sigma-Aldrich) at 33°C in a 5% CO₂ atmosphere, and the supernatant was harvested 48 and 24 h postinoculation with ICV and IAV, respectively. Each harvest was ultracentrifuged at 25,000 rpm for 2 h at 4°C, and after being washed twice with PBS, the pellets were lysed using EzApply (Atto, Tokyo, Japan) and heated at 100°C for 5 min. After collection of the supernatant, the cells were washed twice and lysed using EzApply. Proteins were separated by SDS-PAGE on NuPAGE 4 to 12% Bis-Tris protein gels with MES (morpholineethanesulfonic acid)-SDS running buffer (Invitrogen, Carlsbad, CA, USA) and transferred to polyvinylidene difluoride (PVDF) membranes using the iBlot system (Invitrogen). The membranes were soaked in PBS with Tween 20 (PBS-T) containing 4% skim milk for 1 h at room temperature for blocking. The blots were incubated with monoclonal anti-HA (WS26) (47) or anti-HE (S16) (56) antibody as a primary antibody for 1 h at room temperature and then incubated with peroxidase-conjugated polyclonal goat anti-mouse IgG (Bio-Rad Laboratories) as a secondary antibody for 1 h at room temperature. The proteins were detected by ECL Prime Western blotting detection reagent (GE Healthcare, Little Chalfont, United Kingdom). Image capture, analysis, and processing were performed using the ChemiDoc XRS Plus system and Image Lab software (Bio-Rad Laboratories).

Sequence data.

The HE gene’s full-length nucleotide sequence data from 352 strains between 1947 and 2018 were obtained as described previously (12, 13, 39, 40). The amino acid sequences of the HE protein cleavage site were deduced from the nucleotide sequence, and the diversity was calculated using MEGA v.7.0.26 (57).

Statistical analyses.

Data were analyzed using JMP Pro v.12.2.0 software (SAS Institute Japan, Tokyo, Japan). Group differences were assessed by analysis of variance with post hoc Tukey-Kramer test. A P value of <0.05 was considered significant for all tests.