Fig 1.
Effects of serine, cysteine, metalloproteinase, and aspartic protease inhibitors on FIEC viability and FCoV replication in FIECs.
The cytotoxicity of each protease inhibitor on feline intestinal epithelial cells (FIECs) was evaluated using an MTT assay (green bars). FIECs were first pre-treated with neuraminidase for 1 hour, followed by a 2-hour incubation with the respective inhibitors. The cells were then inoculated with type I FCoV strains (UCD, UG-FH8, and ABA; MOI = 0.05) for 1 hour in the continued presence of the inhibitors. At 12 hours post-infection (hpi), cells were fixed and stained with an anti-nucleocapsid antibody to reveal infected cells. The relative infection rate was calculated by normalizing the infection rate (infected cells/total cells) to that of the untreated control. Error bars represent the standard error of the mean (SEM), n = 3. Statistical significance was determined by two-way ANOVA, except for camostat, which failed the normality test and was analyzed using the Kruskal–Wallis test. *, **, and ***, indicate p-values < 0.05, 0.01, and 0.001 respectively. Panels (A–F) show the effects of specific protease inhibitors on the replication of type I FCoV strains: (A–C) Serine protease inhibitors AEBSF (F(2, 18) = 119.773, p < 0.001), CMK (F(2, 18) = 201.437; p < 0.001) and camostat (F(2, 18) = 310.521; p < 0.001); (D) Cysteine protease inhibitor E-64d (F(2, 18) = 2.558; p = 0.074); (E) Metalloproteinase inhibitor marimastat (F(2, 18) = 2.625; p = 0.1); (F) Aspartic protease inhibitor pepstatin A (F(2, 18) = 0.7; p = 0.510).
Fig 2.
Effect of pH on the replication of FCoV.
The cytotoxicity of each inhibitor on feline intestinal epithelial cells (FIECs) was assessed using an MTT assay (green bars). FIECs were first pre-treated with neuraminidase for 1 hour, followed by a 2-hour incubation with acidification inhibitors. The cells were then inoculated with type I FCoV strains (UCD, UG-FH8, and ABA; MOI = 0.05) for 1 hour in the presence of the inhibitors and maintained in inhibitor-containing medium thereafter. At 12 hours post-infection (hpi), cells were fixed and stained with an anti-viral nucleocapsid antibody to reveal the infected cells. The relative infection rate was calculated by normalizing the infection rate (infected cells/total cells) to that of the untreated control. Error bars represent the standard error of the mean (SEM), n = 3. Statistical significance was determined by two-way ANOVA, except for bafilomycin A1(Baf) and combined treatment of Baf with camostat, which failed the normality test and was analyzed using the Kruskal–Wallis test. *, **, and ***, indicate p-values < 0.05, 0.01, and 0.001 respectively. (A) Effect of NH₄Cl (F(2, 18) = 96.783; p < 0.001) on the replication of type I FCoV strains. (B) Effect of Baf (F(2, 18) = 67.061; p < 0.001) on the replication of type I FCoV strains. (C) Effect of combined treatment with Baf and camostat (F(4, 30) = 239.767; p < 0.001) on the replication of type I FCoV strains. C: camostat.
Fig 3.
Impact of pancreatic serine proteases on FCoV replication.
Feline intestinal epithelial cells (FIECs) were pre-treated with neuraminidase for 1 hour prior to inoculation with type I FCoV strains (UCD, UG-FH8, and ABA). Virus adsorption was performed in the presence of pancreatic serine proteases for 1 hour, after which cells were maintained under the same treatment conditions. At 12 hours post-infection (hpi), cells were fixed and stained with anti-FCoV nucleocapsid antibody (10A12). (A) Representative immunofluorescence images showing enhanced viral infection and syncytia formation following treatment with serine proteases. (B) Quantification of serine protease-mediated enhancement of viral infection. The relative infection rate was calculated as the number of infected cells divided by total cell number, normalized to the untreated control. Statistical significance was determined by one-way ANOVA (F(6, 42) = 27.296; p < 0.001) followed by Dunnett’s post hoc test to compare treated groups with the untreated control. *, **, and *** indicate p-values < 0.05, 0.01, and 0.001, respectively. (C) Quantification of syncytia formation following protease treatment. Syncytia were counted from three replicate experiments and normalized to the untreated control. C: chymotrypsin; T: trypsin; E: elastase. Error bars represent the standard error of the mean (SEM), n = 3. Due to interaction effects between virus strains and inhibitor concentrations, data were analyzed separately for each strain. Statistical significance was determined by one-way ANOVA (F(6, 42) = 52.687; p < 0.001) followed by Dunnett’s post hoc test to compare treated groups with the untreated control. *, **, and *** indicate p-values < 0.05, 0.01, and 0.001, respectively.
Fig 4.
Cleavage of the FCoV spike protein by serine proteases.
(A) Schematic of the feline coronavirus spike protein and its putative cleavage sites. Multiple sequence alignment of coronavirus spike proteins was performed using MEGA12 with the MUSCLE algorithm. Predicted protease cleavage motifs are highlighted in bold red, with scissor icons marking the predicted cleavage positions. CS: cleavage site; SP: signal peptide; HR1: heptad repeat 1; HR2: heptad repeat 2; TM: transmembrane domain; CT: cytoplasmic tail. (B) Schematic of the wild-type and mutant spike constructs derived from the type I FCoV strain ABA. The native transmembrane and cytoplasmic regions were replaced with a foldon trimerization domain and a C-terminal V5/His tag. Cleavage at the S1/S2 site generates the S2 protein, and cleavage at the S2′ site yields the S2′ protein. WT: wild-type spike; MS1/S2: mutant at the S1/S2 site (RRNRRS → AANAAS); MS2′: mutant at the S2′ site (RRS → AAS); 2M: double mutant at both sites. tPA SP: tissue plasminogen activator signal peptide. (C) Western blot analysis of spike cleavage in supernatants following serine protease treatment. HEK293T cells were transfected with spike constructs and cultured in serum-free medium for 48 h. Supernatants were collected and incubated with indicated concentrations of serine proteases at 37 °C for 30 min. Cleavage products were detected using anti-V5 western blotting. Solid line: spike multimers; solid arrow: monomeric spike; open arrow: S2 protein (from S1/S2 cleavage); dashed lines: S2′ proteins (from S2′ or alternative cleavages).
Fig 5.
Entry of pseudotyped viruses carrying wild-type or mutant FCoV Spike.
(A) Optical microscopical images showing cytopathic effects induced by four Spike-pseudotyped lentiviruses (wild-type, MS1/S2, MS2’, and 2M). Neuraminidase (NA)-pretreated FIEC monolayers were inoculated with each pseudotyped virus, and viral entry–associated CPE was documented at 5 hpi and 24 hpi. (B) Representative fluorescence microscopy images of FIEC monolayers at 48 hpi following inoculation with MS2’ or 2M Spike-pseudotyped lentiviruses. Cells were pretreated with NA and infected under the following conditions: untreated, supplemented with trypsin (10 µg/mL), treated with chlorpromazine (CPZ, 6 µM), or treated with both CPZ and trypsin. (C) Quantitative analysis of viral entry for MS2’ and 2M pseudoviruses under the treatment conditions described in panel (B). Data represent the mean EGFP fluorescence intensity measured from 10 random fields per condition. Statistical significance was determined by two-way ANOVA followed by Tukey HSD post hoc test. F(3, 16) = 6.796, p = 0.004.** indicates p < 0.01. T0: without trypsin; T10: 10 µg/mL trypsin; C6: 6 µM CPZ.
Fig 6.
Natural variation at the S2′ cleavage site alters trypsin-mediated processing of the FCoV spike protein.
(A) Sequence logo of the S2′ cleavage site generated from multiple sequence alignment of 167 type I FCoV spike protein sequences. The canonical basic residue at the P2 position (K/R) is replaced by non-basic amino acids (M, V, T, E) in 10.2% of isolates, with methionine (M) being the most frequent substitution (7.2%). (B) Trypsin-mediated cleavage profiles of three spike variants: wild-type (RRS), a non-cleavable mutant (AAS), and a naturally occurring mutant (MRS). HEK293T cells were transfected with spike constructs and cultured in serum-free medium for 48 h. Supernatants were collected and incubated with increasing concentrations of trypsin (2, 10, and 50 μg/mL) at 37 °C for 30 min. Cleavage products were detected using anti-V5 western blotting. Solid line: spike multimers; solid arrow: monomeric spike; open arrow: S2 protein (from S1/S2 cleavage); dashed lines: S2′ proteins (from S2′ or alternative cleavages).
Fig 7.
The identification and cleavage of feline fecal serine proteases.
(A) Fecal serine protease zymography. Colonic fecal samples were collected and prepared as 1% and 10% suspensions before zymographic analysis. For comparison, three additional fecal samples from healthy shelter cats were collected and prepared as 10% suspensions. Transparent bands indicate the presence of serine proteases. (B) Cleavage of feline fecal serine proteases. Supernatants containing secreted spike protein from spike-transfected HEK293T cells were collected and incubated with 10% colonic fecal suspensions at 37°C for 30 minutes. Cleaved products were analyzed by western blotting using an anti-V5 antibody. Solid line: spike multimers; solid arrow: monomeric spike; open arrow: S2 protein (from S1/S2 cleavage); dashed lines: S2′ proteins (from S2′ or alternative cleavages).
Table 1.
Serine proteases identified in feline colonic fecal samples by LC-MS/MS.
Fig 8.
Impact of TMPRSS2 and TMPRSS11D on FCoV replication and spike cleavage.
(A) TMPRSS2 and TMPRSS11D enhance viral infection. Feline intestinal epithelial cells (FIECs) stably expressing TMPRSS2 or TMPRSS11D were pre-treated with or without neuraminidase (NA), followed by infection with type I FCoV strains (UCD, UG-FH8, or ABA) at an MOI of 0.005. At 12, 24, and 48 hours post-infection (hpi), cells were fixed and stained with an anti-nucleocapsid antibody. Representative immunofluorescence images and quantification show increased infection rates in TMPRSS2- and TMPRSS11D-expressing cells compared to controls. Infection was quantified as the percentage of nucleocapsid-positive cells among total cells. Data represent mean ± SEM (n = 3). Statistical analysis was performed separately for each virus strain, neuraminidase treatment, and time point due to interaction effects, using one-way ANOVA F(2, 108) = 576.919 with Dunnett’s post hoc test to assess differences between cell lines. *p < 0.05, **p < 0.01, ***p < 0.001. (B, C) Spike cleavage in FIECs expressing TMPRSS2 or TMPRSS11D. Cells were transfected with V5-tagged spike constructs, and at 48 hours post-transfection, cell lysates (B) and supernatants (C) were collected. Cleavage products were analyzed by western blot using an anti-V5 antibody. Expression of full-length and cleaved spike fragments was evaluated to determine proteolytic activity. Solid line: spike multimers; solid arrow: monomeric spike; open arrow: S2 protein (from S1/S2 cleavage); dashed lines: S2′ proteins (from S2′ or alternative cleavages).
Fig 9.
Proposed model of FCoV entry and spike protein cleavage.
The three pancreatic serine proteases (trypsin, chymotrypsin, and elastase) are secreted by the pancreas and activated in the intestine. Within this intestinal environment, feline coronavirus (FCoV) employs two distinct entry pathways into host cells: the endosomal route and the cell surface route. During viral biosynthesis, the spike protein is pre-cleaved at the S1/S2 cleavage site by furin in the Golgi. A second cleavage event is required at the S2′ region to activate membrane fusion. This cleavage can be mediated either by soluble serine proteases (trypsin, chymotrypsin, elastase) or by membrane-bound proteases (TMPRSS2 and TMPRSS11D). In the endosomal pathway, cleavage occurs in the endosome after viral internalization, whereas in the cell surface pathway, it occurs at the plasma membrane. Trypsin and TMPRSS2/11D primarily target the canonical S2′ site, while chymotrypsin and elastase cleave at two distinct sites flanking S2′, referred to as S2′a CS and S2′b CS. CS: cleavage site.