Abstract
Objectives
To investigate the pathogenicity of feline herpesvirus-1 (FHV-1) to the cornea, FHV-1 strains isolated from feline eyes with dendritic ulcers were subjected to genomic analysis to determine whether FHV-1 vaccine strains are involved in the formation of dendritic ulcers.
Methods
All open reading frame (ORF) sequences of the three F2 strains (Virbac, Intervet and Merial) and the FHV-1 clinical isolates from cats registered in GenBank were compared to detect nucleotide variants unique to the F2 strains, with those nucleotides then being used for simple genotyping of the F2 strains. In all isolates from feline eyes with dendritic ulcers, the regions including nucleotide variants of the F2 strain were amplified with PCR and sequenced. Isolates with nucleotide variants of the F2 strain were then subjected to next-generation sequencing to determine their full genome sequences, which were compared with all ORF sequences of the three F2 strains.
Results
Analysis of ORF sequences for simplified genotyping of F2 strains detected a single nucleotide variant in ORF28 and in ORF44. These were considered to be nucleotide variants unique to the F2 strain. Among the four FHV-1 isolates from eyes of four cats with dendritic ulcers, nucleotide variants of the F2 strain were detected in 1/4 strains (the NS strain). Next-generation sequencing of the NS strain was performed, and all ORF sequences of the NS strain were compared with the those of the three F2 strains. All ORF sequences of the NS strain were completely identical to those of two F2 strains (Virbac and Intervet) and some clones of the Merial vaccine strain.
Conclusions and relevance
The F2 strain was isolated from an eye with a dendritic ulcer, indicating that the strain has the potential to replicate in the corneal epithelium and form lesions.
Plain language summary
Since feline herpesvirus-1 (FHV-1) is a major cause of corneal ulcers, we investigated the pathogenicity of FHV-1 to the cornea via genomic analysis of FHV-1 isolated from feline eyes with dendritic ulcers. Previous genomic analyses of FHV-1 have not identified any nucleotide variants related to pathogenicity, but the FHV-1 modified live vaccine (MLV) F2 strain has unique nucleotide variants that may be distinguishable from field strains. FHV-1 MLVs are often administered, and so identification of feline clinical isolates as vaccine strains would be important information. Four FHV-1 isolates from the eyes of four cats with dendritic ulcers were subjected to genomic analysis. Genomic analysis of nucleotide variants unique to the F2 strain in these isolates resulted in one isolate (the NS strain) being tentatively classified as the F2 strain. The NS strain was further subjected to next-generation sequencing to determine its full genome sequence, which was compared with that of the F2 strain. All open reading frame sequences of the NS strain were completely identical to those of two F2 strains (Virbac and Intervet) and some of clones of the Merial vaccine strain, and the NS strain was identified as the F2 strain. The patient from which the NS strain was isolated had been vaccinated with the Merial vaccine 17 days before the virus was isolated. However, the NS strain was one of the strains included in the Merial vaccine, and so whether the strain originated from the Merial vaccine was unclear. Three of the four patients in this study had a history of corticosteroid treatment, and it was inferred that administration of corticosteroids may have been involved in the formation of dendritic ulcers. These results suggest that the F2 strain has the potential to replicate in the corneal epithelium and form lesions, and, in a cat with no previous history of allergy, corticosteroid administration is not recommended in the period immediately after MLV vaccination.
Introduction
Feline herpesvirus-1 (FHV-1) is widespread in cats around the world and is a virus that causes infectious ocular surface disease (IOSD). 1 Other pathogens that cause IOSD include feline calicivirus (FCV), Chlamydia felis and Mycoplasma species, but FHV-1 is the only pathogen that can infect and replicate in the corneal epithelium, and is the leading cause of corneal ulcers in cats. 2 Corneal ulcers are typically associated with ocular pain and cause visual impairment, and so they are a major ocular disease. Chronic corneal ulcers may also be associated with eosinophilic keratitis and corneal sequestration, which are ocular diseases characteristic of cats.3,4 The causes of these ocular diseases are not fully understood, and these diseases are intractable. Thus, research on the pathogenicity of FHV-1 to the cornea is important.
Genomic analysis is sometimes used to study viral pathogenicity. A study on human herpes simplex virus-1 (HSV-1) indicated that the HSV-1 genotypes identified in ocular samples from patients with herpetic corneal ulcers were associated with corneal ulcer morphology and recurrence. 5 The diagnosis of herpetic corneal ulcers in humans is largely based on the characteristics of the lesion according to a slit-lamp examination and the patient’s past medical history, 6 but obtaining a medical history can be particularly difficult with cats, hampering the diagnosis of herpetic corneal ulcers. That said, feline dendritic ulcers are a pathognomonic finding of herpetic corneal ulcers and can be diagnosed clinically based on observation of linear branching ulcers. 7 This finding indicates that FHV-1 has infected and replicated in the corneal epithelium, and so strains isolated from eyes with dendritic ulcers are likely to be the cause of corneal ulcers. Therefore, genomic analysis of isolates from eyes with dendritic ulcers is useful in studying the pathogenicity of FHV-1 to the cornea.
However, previous genomic analyses of FHV-1 clinical isolates have indicated that FHV-1 genome sequences are extremely uniform and that the severity of ocular disease may not be associated with genomic variation. 8 That said, attenuated FHV-1 strains include FHV-1 modified live vaccine (MLV) strains, and genome sequences of the representative F2 strain have indicated that it may have unique nucleotide variants. 9 In addition, strains with sequences very similar to the genome sequences of the F2 strain have been isolated from cheetahs vaccinated with vaccines containing the F2 strain, 10 and the F2 strain might be identified in feline clinical isolates. The FHV-1 vaccine is recommended as a core vaccine for routine vaccination of cats. Given the efficacy of live vaccines, vaccines containing the FHV-1 MLV strain are often administered, 11 and so identification of feline clinical isolates as vaccine strains would be important information.
Thus, the present study analysed the genome of FHV-1 strains isolated from eyes with dendritic ulcers. Genomic analysis was performed to determine whether the FHV-1 MLV strain is involved in the formation of dendritic ulcers. Owners were also interviewed regarding the medical history, medication history and vaccination status of cats to determine whether the administered FHV-1 MLV strain caused the formation of dendritic ulcers.
Materials and methods
Animals and specimen collection
Specimens were collected from cats undergoing an ophthalmic examination at Suga Animal Clinic between May 2019 and October 2021 that were suspected of having an ocular infection with FHV-1. Of the FHV-1 strains isolated from these specimens, four strains were isolated from eyes diagnosed with dendritic ulcers based on the finding of a linear branching ulcer on fluorescein staining after specimen collection, and were from four cats (patients 1, 2, 3 and 4). Dendritic ulcers had been diagnosed in the eyes of these cats when specimens were submitted. Their owners approved the use of those specimens for identification of a coinfection with FCV and for genomic analysis of FHV-1. Owners were interviewed regarding the cats’ vaccination history (including the name of the vaccine manufacturer), medical history and medication history. An antibody test for feline immunodeficiency virus (FIV) infection and an antigen test for feline leukaemia virus (FeLV) were performed using the SNAP FIV/FeLV Combo Test (IDEXX). A sterile Dacron swab (Alpha TX761; Texwipe) was used to collect a specimen by wiping tears from the conjunctiva. The swab was placed in a sterile plastic tube with 0.5 ml of Dulbecco’s modified Eagle’s medium – low glucose (Sigma-Aldrich) supplemented with gentamicin at a final concentration of 50 µg/ml. The tubes were refrigerated (2–6°C) and used for virus isolation. After specimen collection, fluorescein staining (FLUORES Ocular Examination Test Paper 0.7 mg; Ayumi) was performed, and a dendritic ulcer was identified based on the findings of a linear branching ulcer under a slit-lamp microscope (SLD-4; Topcon).
Cultured cells
Crandell-Rees feline kidney (CRFK) cells were used for virus isolation. 12 Eagle’s minimum essential medium (MEM) supplemented with 10% fetal bovine serum was used as growth medium, and the serum concentration in the growth medium was reduced to 4% for use as maintenance medium.
Virus isolation and DNA extraction
The swab was centrifuged at 700 g for 5 mins, and 100 µl of the supernatant after centrifugation was inoculated into a monolayer culture of CRFK cells in six-well plates. After adsorption at 37°C for 30 mins in 5% CO2, the cell surface was washed with MEM, and cells were cultured at 37°C in 5% CO2 with added maintenance medium. Viral DNA was extracted as total DNA from virus-infected cells 13 and used in a PCR. Viral DNA for next-generation sequencing was extracted from nucleocapsids in virus-infected cells. 14
Virus cloning
Virus cloning was performed by the plaque method according to conventional methods. A monolayer culture of CRFK cells in a six-well plate was inoculated with virus serially diluted 10-fold in maintenance medium and was allowed to adsorb for 1 h at 37°C. After adsorption, the cell surface was washed three times with MEM, layered with MEM supplemented with 0.9% Difco Agar (Becton, Dickinson and Company) and 4% fetal bovine serum, and incubated at 37°C for 3 days. Well-separated plaques were collected with a Pasteur pipette under an inverted microscope and suspended in maintenance medium to make viral clones.
PCR primers and PCR
Identification of FHV-1 in viral isolates was performed with PCR targeting the FHV-1 glycoprotein B (gB) gene as described by Vögtlin et al. 15 Primer sequences are shown in Table 1.
Primers used in a PCR to identify feline herpesvirus-1
Nucleotides were numbered based on GenBank accession number NC_013590 (C-27)
bp = base pair; gB = glycoprotein B; F = forward; R = reverse
To simply genotype the FHV-1 clinical isolates to the F2 strain, all open reading frame (ORF) sequences of the F2 strains from the three vaccine manufacturers (Feligen [GenBank accession number KR296657] from Virbac; Companion [GenBank accession number KR381803] from Intervet; and Purevax [GenBank accession number OL410296] from Merial), the C-27 strain (GenBank accession number NC_013590) as the reference FHV-1 strain, and the FHV-1 clinical isolates from cats that are registered in the GenBank gene database were compared using the software MEGA X. 16 Two ORFs with different nucleotides that appear to be unique to the F2 strain were identified, and primers specific to the regions containing the different nucleotides were designed using the software DNASIS PRO (Hitachi Software Engineering) (Table 2).
Primers used in a PCR to genotype the F2 strain
Nucleotides were numbered based on GenBank accession number KR296657 (Feligen strain from Virbac)
bp = base pair; F = forward; ORF = open reading frame; R = reverse
In the sequencing data for the F2 strain from Merial, there were three ORFs with undetermined nucleotides, and primers specific to the regions containing these undetermined nucleotides were designed using the software DNASIS PRO (Table 3).
Primers used in a PCR to sequence open reading frames with undetermined nucleotides of the F2 strain from Merial
Nucleotides were numbered based on GenBank accession number KR296657 (Feligen strain from Virbac)
bp = base pair; F = forward; ORF = open reading frame; R = reverse
PCR was performed using KOD-Plus-Neo (Toyobo) and PCR products were purified using the QIAquick PCR & Gel Cleanup kit (QIAGEN). Sequencing was performed by Hokkaido System Science and sequences were analysed with SnapGene (GSL Biotech). The obtained sequences were registered with DNA Data Bank of Japan (accession numbers LC820743–LC820750, in the order of ORF28 of YN strain from patient 1, TH strain from patient 2, NS strain from patient 3 and AS strain from patient 4 for LC820743–LC820746, and ORF44 of YN, TH, NS and AS strain for LC820747–LC820750).
Next-generation sequencing
Next-generation sequencing was performed by Macrogen Japan to determine the full genome sequences of clinical isolates in which markers of the F2 strain were detected with simplified genotyping. Sequencing was performed on the Novaseq 6000 platform (Illumina). All confirmed sequences were registered in GenBank (accession number OR514564) and detailed genomic analysis was performed with Geneious Prime version 2023.1.2. 10
Results
Animals
Clinical data on each patient with dendritic ulcers are shown in Table 4. Three patients were domestic shorthairs and one was an Exotic Shorthair. The two females were neutered and the two males were intact. Patients ranged in age from 2 to 120 months (mean: 53.3 months). Dendritic ulcers were diagnosed in all of the patients based on staining of linear branching ulcers (Figure 1). An ulcer was found in the right eye of patients 1 and 3, and in the left eye of patients 2 and 4 (Figure 1). The menace response was intact for all patients. The dazzle reflex, direct and consensual pupillary light reflexes, intraocular pressure, and a fundus examination were normal for all patients. The FIV antibody test and FeLV antigen test were negative for three of the patients; patient 1 tested positive for FIV. Patient 1 had no history of vaccination. Patient 2 was administered the MLV F2 strain (Purevax RCP; Merial) in a first vaccination 10 years previously and was administered the MLV FVR-PM strain (Fel-O-Guard Plus 3; Zoetis) 1 year later. Patient 3 was administered the MLV F2 strain (Purevax RCP; Boehringer) in a first vaccination 17 days before being diagnosed with a dendritic ulcer. Patient 4 was administered the inactivated 605 strain (Fel-O-Vax 3; Zoetis) in a first vaccination 9 months previously and was administered the MLV FVRm strain (Felocell CVR; Zoetis) 1 month later. The manufacturer and distributor of Purevax RCP in Japan changed when Merial Japan Inc. merged with Boehringer Ingelheim Japan in December 2017, but the F2 strain was the same; the F2 strain from Boehringer is referred to as the F2 strain from Merial in this study. Three of the four patients had previously received corticosteroid therapy. Patient 2 received systemic treatment for cholangiohepatitis at the same clinic, and patients 3 and 4 received topical treatment for allergic conjunctivitis at other clinics. Fourteen days after administration of a corticosteroid (oral prednisolone 1.0 mg/kg q24h), FHV-1 infection-like clinical signs appeared in patient 2. FHV-1 infection-like clinical signs were present in patients 3 and 4 when corticosteroids were administered and worsened 1–2 days after administration.
Clinical data on patients that had eyes with dendritic ulcers
Purevax RCP
Fel-O-Guard Plus 3
Fel-O-vax 3
Felocell CVR
FeLV = feline leukaemia virus; FHV-1 = feline herpesvirus-1; FIV = feline immunodeficiency virus; MLV = modified live vaccine
Photographs of dendritic ulcers stained with fluorescein in each patient. Linear branched ulcers are seen. The abbreviation above each photo indicates the patient’s designation and magnification is indicated below each photograph
Virus isolation and identification
Swabs from the four patients with dendritic ulcers were inoculated into CRFK cells for virus isolation, and a clear cytopathic effect with herpesvirus-like rounding of cells was noted by day 2 of culture. DNA extracted from the infected cells was used to perform PCR targeting the FHV-1 gB gene, and a DNA fragment approximately 740 bp in length was amplified. This was the same size as the expected amplified fragment of FHV-1 gB (data not shown). Thus, the four viral isolates were identified as FHV-1.
Simplified genotyping
The sequences of all ORFs of the F2 strains from the three vaccine manufacturers, the C-27 strain and the FHV-1 clinical isolates from cats were compared, and a difference in a single nucleotide was identified in ORF28 and in ORF44 (Table 5). These locatione were consistent with the locations of the unique single nucleotide polymorphisms reported by Vaz et al, 9 and so the nucleotide variants within these regions were used for simplified genotyping. Genomic analysis of these two regions in isolates from eyes with dendritic ulcers resulted in one of the four isolates (the NS strain) being tentatively classified as the F2 strain (Table 5). Nucleotides did not overlap in the chromatograms of the location of each nucleotide variant in these two regions (data not shown). This confirmed that the NS strain was not contaminated with field strains.
Nucleotide differences and locations in ORF28 and ORF44 of each feline herpesvirus-1 strain
Locations represent the number of nucleotides, with the first nucleotide in each ORF counted as 1
ORF = open reading frame
Next-generation sequencing
The NS strain was classified as the F2 strain according to simplified genotyping. The NS strain was subjected to next-generation sequencing to identify it as the F2 strain. The sequence was 134,720 bp in length (GenBank accession number OR514564). All ORF sequences in the NS strain were compared with those of the F2 strains from the three vaccine manufacturers, and the sequences were completely identical to the F2 strains from Virbac and Intervet. However, mixed nucleotides at R, Y and Y were found in ORF38, ORF41 and ORF48, respectively, in the F2 strain from Merial, suggesting the existence of multiple clones (Table 6). The nucleotides at locations R and Y in the F2 strain from Merial were A and T in the F2 strains from Virbac and Intervet and the NS strain, respectively (Table 6). When raw data from the next-generation sequencing of the NS strain were used to check the nucleotides at those locations, the nucleotide was A in 99.9% of the ORF38, T in 99.9% of the ORF41 and T in 99.8% of the ORF48. Thus, the NS strain was a clone with genome sequences similar to those of the F2 strain from Virbac and Intervet. The results suggested the existence of multiple clones of the vaccine strain from Merial, and the patient from which the NS strain was isolated had been vaccinated with the Merial vaccine. Given these facts, the vaccine strain from Merial was cloned using the plaque method, and an attempt was made to confirm the nucleotides in the 12 clones obtained (Table 6). The results indicated that 4/12 clones (clones 2, 9, 10 and 11) were identical to the NS strain.
Locations of the undetermined nucleotides in the open reading frames of the F2 strain from Merial and nucleotides at these locations in the NS strain, the two F2 strains (Virbac and Intervet) and the clones of the F2 strain from Merial
Locations represent the number of nucleotides, with the first nucleotide in each ORF counted as 1
ORF = open reading frame
Discussion
In the present study, a strain of FHV-1 that had all of the same ORFs as the F2 strain was isolated from a feline eye with a dendritic ulcer, indicating the potential for the F2 strain to infect the corneal epithelium and cause lesion formation. FHV-1 infection is a key aetiological factor in feline corneal ulcers, 2 and so this is clinically important information.
Feligen (Virbac) and Companion (Intervet) are both the F2 vaccine strain. A study has indicated that there are unique nucleotide variants of those two F2 strains in ORF28 and ORF44. 9 Purevax (Merial), which is another F2 strain, and the NS strain that was isolated in the present study had similar nucleotide variants at both locations. The three strains other than the NS strain that were isolated in this study included an isolate from patient 2, which had been vaccinated with the Merial vaccine 10 years previously. However, isolates with unique nucleotide variants of the F2 strain were not detected from patient 2. The NS strain was isolated from patient 3, which had been vaccinated with Merial vaccine 17 days before the patient was diagnosed with a dendritic ulcer, and was considered to be a possible Merial vaccine strain. However, the NS strain lacked the mixed sequences of the three ORFs of the Merial vaccine (ORF38, ORF41 and ORF48). Among the clones obtained from plaque cloning of the Merial vaccine, a clone with the same nucleotides as the NS strain was detected, indicating that the NS strain may be identical to some clones of the Merial vaccine strain.
Patient 3, from which the NS strain was isolated, was a cat that was picked up from outdoors and was in good physical condition, so patient 3 was vaccinated with Merial vaccine at another clinic. A few days later, patient 3 simultaneously developed sneezing, nasal discharge and ocular discharge from both eyes, which are typical clinical signs of FHV-1. This was consistent with the temporary development of FHV-1-like clinical signs in some cats as side effects of vaccination. 11 If the NS strain was a cloned strain of the administered Merial vaccine, possible causes are that patient 3 may have licked some of the vaccine that leaked from the injection site, 10 or the vaccine virus may have been squirted into the air when the vaccine was drawn out of the vaccine vial and become aerosolized during administration, resulting in the incidental exposure of patient 3 to the vaccine strain through its mouth, nose or eyes. However, the NS strain lacks the mixed nucleotides found in the Merial vaccine strain, which would mean that the contaminating clonal strains were out competed during replication in the cat, but the mechanism for this is not known. Another possibility is a scenario in which the F2 strain was already latently infecting patient 3 prior to vaccination with the Merial vaccine, and the strain was reactivated by vaccination.11,17 In this situation, patient 3 may have been vaccinated with the uncontaminated F2 strain or the F2 strain may have been circulating to an extent in the environment where patient 3 lived, but there is no evidence for these. Thus, the origin of the NS strain is unknown.
Corticosteroids had been administered to 3/4 patients in this study (patients 2, 3 and 4). Presumably, there is an association between the formation of dendritic ulcers induced by FHV-1 and corticosteroid administration. Patient 2 had no FHV-1-like clinical signs at the start of systemic corticosteroid administration and developed FHV-1-like clinical signs approximately 14 days after administration. Corticosteroids induce the reactivation of FHV-1 in 70% of cats with a latent infection, 18 and so the field strain that latently infected patient 2 may have reactivated and formed dendritic ulcers. By contrast, patient 4 was diagnosed with allergic conjunctivitis and had FHV-1-like clinical signs, but topical corticosteroids were started. The worsening of FHV-1-like clinical signs was noted 1–2 days later, suggesting that topically administered corticosteroids may have suppressed the natural resistance of patient 4 to a viral infection. 19 This facilitated the proliferation of a field strain and the formation of a dendritic ulcer. Patient 3 had a medical history similar to that of patient 4, so similar action of corticosteroids may have facilitated the proliferation of the F2 strain, leading to the formation of a dendritic ulcer. Therefore, when administering an MLV to cats with no previous history of allergy, the potential for a vaccine-related infection should be considered and corticosteroid administration is not recommended in the immediate post-vaccination period.
The present study indicated that there are several FHV-1 strains with different sequences in the Merial vaccine. Some of these strains have the same ORF sequences as the NS strain, suggesting that they are pathogenic to the cornea. The pathogenicity of the other FHV-1 strains in the Merial vaccine is unknown, and whether those strains were isolated alone or together is also unknown. Nucleotide variants of ORF28, ORF41 and ORF48 in these strains could be used to distinguish field strains from the Merial vaccine, and so the genomes of clinical isolates should be analysed and their pathogenicity should be examined in the future.
Conclusions
In the present study, a strain that had all of the same ORF sequences as the F2 strain was isolated from an eye with a dendritic ulcer, suggesting that the F2 strain may infect the corneal epithelium and cause lesion formation. In addition, the administration of corticosteroids may have been involved in the formation of dendritic ulcers due to the F2 strain, suggesting that in a cat with no previous history of allergy, corticosteroid administration is not recommended in the period immediately after MLV vaccination.
Footnotes
Author note
This paper was presented in part at the 2022 Japanese Society of Clinical Veterinary Medicine Symposium.
Conflict of interest
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The authors received no financial support for the research, authorship, and/or publication of this article.
Ethical approval
The work described in this manuscript involved the use of non-experimental (owned or unowned) animals. Established internationally recognised high standards (‘best practice’) of veterinary clinical care for the individual patient were always followed and/or this work involved the use of cadavers. Ethical approval from a committee was therefore not specifically required for publication in JFMS. Although not required, where ethical approval was still obtained, it is stated in the manuscript.
Informed consent
Informed consent (verbal or written) was obtained from the owner or legal custodian of all animal(s) described in this work (experimental or non-experimental animals, including cadavers, tissues and samples) for all procedure(s) undertaken (prospective or retrospective studies). No animals or people are identifiable within this publication, and therefore additional informed consent for publication was not required.
