Detection of anti-fesavirus 4 antibodies in plasma samples from cats in Japan

Abstract

Fesaviruses, members of the group of picorna-like viruses, were discovered in 2014 in stool samples from cats in an animal shelter in the United States. In this study, we investigated the serological evidence of fesavirus 4 infection in domestic cats in Japan using a purified recombinant antigen corresponding to a portion of the large open reading frame of fesavirus 4. An enzyme-linked immunosorbent assay was developed and used to test 100 cat plasma samples to detect specific antibodies against the fesavirus 4 antigen and Western blotting confirmed presence of specific antibodies against the fesavirus 4 antigen in 12 samples. These findings highlight the importance of investigating newly discovered viruses circulating in companion animals in Japanese animal shelters.

Animal shelters often house stray animals that have not received enough veterinary care or preventive treatments before being admitted. As a result, these animals may have an increased likelihood of exposure to infectious agents, making regular surveillance within shelters essential for identifying pathogens present in the resident animals [4]. Cats are popular companion animals and commonly share living environments with humans, domestic animals, and wildlife. Their close interactions with these groups raise the possibility of transmitting certain viruses that are also capable of infecting humans, such as rabies, SARS, and influenza A viruses [1, 2]. Therefore, research on newly identified viruses in cats is crucial for safeguarding public health. Feline stool-associated RNA virus (fesavirus) 1, 2, 3 and 4 were identified in 2014 from fecal samples collected from cats in animal shelters in the United States [8]. The sequences of fesavirus 1–4 are markedly diverse, despite sharing the same virus designation [8]. Fesaviruses are classified as picorna-like viruses based on their limited sequence similarity in the open reading frames (ORFs) corresponding to the RNA-dependent RNA polymerase (RdRp) domains of members of the order Picornavirales [3]. Picorna-like viruses possess positive-sense, single-stranded RNA genomes ranging from 1.4 to 9.5 kilobases in length [3]. More recently, the fesavirus 4 sequence was also detected in fecal samples from cats in an animal shelter in Japan [5]. So far, all fesaviruses have been identified exclusively through metagenomic analyses of fecal samples [5, 8]. Such approaches may detect nucleic acids derived not only from viruses actively infecting cats but also from undigested intestinal contents, including insects or other ingested animals. Because experimental infections are not feasible for picorna-like viruses, whose isolation is often challenging, serological detection of specific antibodies may represent the only practical method to demonstrate fesavirus infections in their feline hosts. For the reasons described above, we established an ELISA-based screening approach using a purified recombinant protein corresponding to a partial open reading frame (ORF) sequence of fesavirus 4 expressed in Escherichia coli, and applied it to detect specific antibodies against this antigen in plasma samples collected from cats in Japan. Our aim was to obtain serological data on fesavirus 4 to better understand the prevalence and geographic distribution of this viral infection in cats in Japan and potentially worldwide.

Use of animal specimens in this study was reviewed and certified by Research Animal Testing and Research Ethics Review Committee of Tokyo University of Agriculture and Technology. The certificate document is submitted as a supplemental document. We analyzed a total of 100 plasma samples of domestic cats in Japan whose information was listed in Supplementary table. Plasma samples from cats in which fesavirus 4 was detected in the reference [5] were not available and not included in this study. Fesavirus 4 sequence in this study was obtained from fesavirus 4-C12 in our previous study [5], whose accession number of the DNA Data Bank of Japan was LC822348. All the chemicals were obtained from Fuji film (Tokyo, Japan) unless otherwise noted. The antigenic epitopes and hydrophobicity predictions were performed with Bepipred Linear Epitope Prediction 2.0 method in the DTU Health Tech Department of Health Technology (IEDB) site (https://services.healthtech.dtu.dk/services/BepiPred-2.0/) and Kyte-Doolittle scale plot of the ProtScale tool (https://web.expasy.org/protscale/), both tools were accessed on December, 2025, respectively. The region of the fesavirus 4 gene with high antigenic potential was amplified using the KOD-Plus-Neo kit (Code No. KOD-401, Toyobo, Osaka, Japan) with a pair of specific primers (Forward, 5′-TTCCCGGGTACGAATCCCAACAATGCCA-3′; and Reverse, 5′-CTCGAGTCAAGCTCTCGTGTCCTCTT-3′) containing a SmaI and XhoI restriction sites, respectively, and inserted into an expression vector pGEX-4T-1 (Code No. GE28-9545-49, GE Healthcare, Tokyo, Japan) for expression of a foreign gene as a fusion protein with glutathione S-transferase (GST), and the resulting plasmid was named as pGEX-4T-fesa4. Escherichia coli BL21 (DE3) was transformed with the pGEX-4T-fesa4 plasmid and cultured in the LB medium (Code No. 63-6528-52, Becton, Dickinson and Co., Franklin Lakes, NJ, USA) with ampicillin (Code No. 016-23301, 100 μg/mL final) for 5 hr at 37°C after addition of Isopropyl β-D-1-thiogalactopyranoside (IPTG) (Code No. 096-05143, 260 μM final) and for 9 hr at reduced temperature of 24°C to increase soluble recombinant protein. The bacterial cells were pelleted and sonicated in 1% Triton X-100 (Code No. A16046) in phosphate buffered saline (PBS) with a protease inhibitor cocktail (Code No. 03969-21, Nacalai Tesque, Kyoto, Japan) followed by centrifugation at 15,000 rpm for 20 min at 4°C for the supernatant for purification. For purification of the GST-fused fesavirus 4 protein (Fesa4-GST), bacterial supernatant from 150 mL E. coli culture was adjusted to 0.8% Triton X-100, 125 mM Tris-HCl, 150 mM NaCl, 1 mM fresh DTT, and 1 mM EDTA and mixed with glutathione Sepharose 4B (Code No. 17075601, GE Healthcare) and Fesa4-GST was eluted from the attached beads with elution buffer of 100 mM Tris-HCl, 500 mM NaCl, 1 mM DTT, 15 mM reduced glutathione. Protein samples were heated at 95°C for 5 min in SDS protein sample buffer (Sample Buffer Solution (2ME-, Code No. 193-11032) (×4)), separated in 10% SDS-PAGE gel for Coomassie brilliant blue (CBB) staining or transfer on a polyvinylidene difluoride (PVDF) membrane (Clear blot membrane-P Plus, Code No. WSE-4050, ATTO, Tokyo, Japan). The sample buffer without 2-mercaptoethanol (2-ME) was used after confirmation of some positive bands in WB with some anti-fesa4 antibody-positive plasma samples. The membrane was blocked with Blocking One (Code No. 03953-95, Nacalai Tesque), incubated in the primary samples and the secondary antibodies of horseradish peroxidase (HRP)-conjugated anti IgG, at room temperature for 1 hr for all the incubations, with washes three times with 0.1% Tween 20 in PBS (PBST) between incubations. The antibodies were visualized with the TMB substrate (TMB Solution for Western Blotting, Code No. 05298-80, Nacalai Tesque) for 2 min. For ELISA, purified Fesa4-GST or GST were used to coat 96-well plates (ELISA plate, Iwaki, Tokyo, Japan) using carbonate-bicarbonate buffer (44 mM NaHCO₃, 6 mM Na₂CO₃, pH 9.3) at 4°C for overnight, washed three times with PBST (PBS with 0.1% Tween-20). Optimum antigen concentrations for coating (625 and 1,250 ng/mL for GST and Fesa4-GST, respectively) were determined with serial dilutions of the antigens (Supplementary Fig. 1–3). Plates were incubated with 5% skimmed milk in PBST for blocking, with the primary antibodies and secondary antibodies, for 1 hr at room temperature, with 4-time washes with PBST between the incubations. After antibody incubations and washes, antibodies were detected with ELISA POD Substrate TMB Solution (Code No.05299-54, Nacalai Tesque) for 10 min at room temperature, added with an equal volume of 1 M H₂SO₄ to stop the reaction, and measurement of the absorbance at 450 nm (Multiskan FC, Thermo Scientific, Tokyo, Japan). The combinations of primary and secondary antibodies and dilutions are listed in Complementary document. All reactions were performed in duplicate, and averages were calculated.

Potential antigenic epitopes within the deduced amino acid sequence of the large ORF of fesavirus 4 were analyzed using antigenic epitope prediction tools. In order to increase the chance of success in expressing a soluble recombinant antigen protein, we decided to express a part of the ORF with higher antigenicity than the other regions. As shown in Fig. 1, we selected a 493 amino acid (aa) region (positions 937–1429) for recombinant expression based on BepiPred Linear Epitope Prediction and Kyte-Doolittle hydropathy scale, for its antigenicity and hydrophilicity scores (Fig. 1, indicated by arrows), and expressed it as a recombinant protein to use it as an antigen to detect specific antibodies against fesavirus 4.

Fig. 1.

Analysis of epitope prediction and hydropathy plot of the deduced amino acid sequence of fesavirus 4 and the corresponding fesavirus 4-C12 genome. (Upper Panel) Bepipred Linear Epitope Prediction 2.0. Regions with scores above 0.5 (bold center line) surpass the threshold and are predicted to contain B-cell epitopes. (Middle Panel) Hydropathy plot based on the Kyte and Doolittle Scale. Regions below zero (bold central line) indicate hydrophilic regions. (Lower Panel) Schematic representation of the fesavirus 4-C12 genome. The region corresponding to the RNA-dependent RNA polymerase (RdRp) and the protein segment expressed in this study are indicated. The N- and C- termini are also shown.

A portion of the fesavirus 4 large ORF selected above was inserted into pGEX-4T-1 plasmid vector (pGEX-4T-fesa4) for expression of the recombinant Fesa4 protein. E. coli cells transformed with pGEX-4T-fesa4 and empty pGEX-4T-1 were cultured to express Fesa4-GST and GST, respectively. Fesa4-GST and GST with the predicted molecular sizes of 81 and 26 kDa, respectively, were expressed in the supernatants of transformed E. coli (Fig. 2, lanes 1 and 2) and were affinity-purified with glutathione-conjugated Sepharose 4B (Fig. 2, lanes 3 and 4 for CBB-stained gel, and lanes 5 and 6 for WB probed with an anti-GST antibody). The GST was used to detect anti-GST specific antibodies in the cat plasma samples and served as a negative control since it might contain the same minute amount of E. coli contaminants as the fesa4-GST that could not be eliminated by affinity purification.

Fig. 2.

Expression and purification of Fesa4-glutathione S-transferase (GST) and GST. Coomassie brilliant blue (CBB)-stained sodium dodecyl sulfate (SDS)-polyacrylamide gel electrophoresis (PAGE) gel (lanes 1–4) and Western blotting with an anti-GST mouse monoclonal antibody (lanes 5 and 6). Numbers on the left indicate molecular mass markers in kilo Dalton (kD). Lanes 1 and 2, cell supernatants of Escherichia coli transformed with pGEX-4T-fesa4 and pGEX-4T. Lanes 3 and 4, purified Fesa4-GST and GST obtained from E. coli supernatants using glutathione Sepharose 4B. Lanes 5 and 6, Western blotting of the purified Fesa4-GST and GST probed with an anti-GST mouse monoclonal antibody. Loading amounts were the amount corresponding to E. coli culture of 0.06 mL (lanes 1 and 2), and 1.7 μg (lane 3), 3.4 μg (lane 4) and 50 ng (lanes 5 and 6) of the purified proteins. The arrowhead and arrow indicate the positions of Fesa4-GST and GST, respectively.

Although we attempted to remove GST from the Fesa4-GST fusion protein by thrombin digestion at the specific cleavage site between Fesa4 protein and GST, the yield of recovered Fesa4 protein from the digestion reaction was not good enough to perform subsequent experiments. Therefore, we decided to use Fesa4-GST and GST directly to develop an ELISA system for detecting specific antibodies against the Fesa4 antigen. For this purpose, absorbance at 450 nm (A450) was measured for both Fesa4-GST and GST, and a normalized A450 (NA450) was calculated by subtracting the GST A450 value from that of Fesa4-GST, thereby eliminating the effect of antibodies directed against GST.

To identify cat plasma samples containing Fesa4-specific antibodies, we performed ELISA using 100 cat plasma samples collected from veterinary clinics in Japan and obtained NA450 values for each sample (Fig. 3). Sixteen samples with the highest NA450 ELISA values (Fig. 3, within the arrow) were tested with Western blotting to find specificity of the antibodies and 12 out of 16 plasma samples appeared to have antibodies against Fesa4-GST but not against GST (Fig. 4), while 3 samples reacted to both Fesa4-GST and GST (Fig. 4, lane pair 7, 8 and 13), and a sample did not react to either of the proteins (Fig. 4, lane pair 9). These findings demonstrated presence of specific antibodies against the fesavirus 4 antigen in plasma samples from cats in Japan.

Fig. 3.

Normalized absorbance at 450 nm (NA450) in the ELISA assay using 100 cat plasma samples to detect antibodies specific to Fesa4. Normalized A450 (NA450) was calculated by subtracting the glutathione S-transferase (GST) A450 value from that of Fesa4-GST. Sixteen samples with the highest NA450 (arrow) were also tested with Western blotting using Fesa4-GST and GST.

Fig. 4.

Western blotting with the 16 cat plasma samples from the top in Normalized A (NA450). Molecular mass markers in kilo Dalton (kD) are indicated on the left. Pairs of purified Fesa4-glutathione S-transferase (GST) and GST were used for each blot, and NA450 of the cat plasma samples used were indicated below the panels. Lanes 1–16, 16 plasma samples from the top in NA450. Lane 17, a plasma sample which did not react to either Fesa4-GST or GST. Arrowhead and arrow denote the positions of Fesa4-GST and GST, with predicted sizes of 81 and 26 kDa, respectively.

To our knowledge, this study represents the first serological investigation of antibodies against the fesavirus 4 antigen and demonstrates presence of specific antibodies against fesavirus 4 antigen in plasma samples from cats in Japan. Previous studies reported fesavirus 4 detection rates of 25% (5/20) by reverse transcription quantitative PCR (RT-qPCR) and 10% (2/20) by metagenomic sequencing in stool samples collected from cats in an animal shelter in Japan [8]. Although accurate positive rate of specific antibody against the fesavirus 4 antigen cannot be calculated in this study, at least 12 of the 100 tested samples showed Fesa4-specific reactivity by Western blotting among samples selected based on high ELISA reactivity (Fig. 4). Further investigations are required to establish a more accurate epidemiological understanding of fesavirus 4 in cats, including refinement of the antibody detection system, evaluation of antibody specificity using absorption of specific antibodies with soluble antigens, and assessments of active infection rates and seroprevalence across various feline populations in Japan and other countries. Also, in this study, the experiments were conducted using only proteins expressed in E. coli, and use of an indirect immunofluorescence assay with infected cells or Western blotting on lysates from infected cells as positive controls should be more suited if the virus is isolated. In this regard, it should be noted that the plasma samples from the cats in which fesavirus 4 RNA was detected in reference [5] were not available for the sero-detection of anti-fesavirus 4 antibodies in this study, and the antibody test of plasma samples collected 2–3 weeks after viral nucleic acid detection will more accurately prove active infections of fesavirus 4 in the infected cats in the future.

The possibility of inter-species transmission of fesaviruses, including fesavirus 4, is not yet known, although viruses with sequences highly similar to those of fesaviruses have been detected in species other than domestic cats. In a metagenomic study of fecal samples from fruit bats in Cameroon, a virus termed bat felisavirus, whose sequence was the closest to fesavirus 1, was detected [7]. Another metagenomic study of fecal samples from humans in isolated Amazonian villages detected a virus sequence most closely related to bat felisavirus and fesavirus 1, which was designated a fesa-like virus [6]. Although the former study suggested a possible insect origin of bat felisavirus, active infections in either bats or humans cannot be ruled out. Because our findings provide evidence of active fesavirus 4 infection in cat populations in Japan, similar sero-epidemiological studies targeting other fesaviruses in cats are needed to determine whether these viruses actively infect feline hosts and to assess their potential for cross-species transmission.

CONFLICT OF INTEREST

The authors declare that there is no conflict of interest involved in this study.