key: cord-345413-bsd32j8r
authors: Terada, Yutaka; Kuroda, Yudai; Morikawa, Shigeru; Matsuura, Yoshiharu; Maeda, Ken; Kamitani, Wataru
title: Establishment of a Virulent Full-Length cDNA Clone for Type I Feline Coronavirus Strain C3663
date: 2019-08-02
journal: Journal of Virology
DOI: 10.1128/jvi.01208-19
sha: 
doc_id: 345413
cord_uid: bsd32j8r

Feline infectious peritonitis (FIP) is one of the most important infectious diseases in cats and is caused by feline coronavirus (FCoV). Tissue culture-adapted type I FCoV shows reduced FIP induction in experimental infections, which complicates the understanding of FIP pathogenesis caused by type I FCoV. We previously found that the type I FCoV strain C3663 efficiently induces FIP in specific-pathogen-free cats through the naturally infectious route. In this study, we employed a bacterial artificial chromosome-based reverse genetics system to gain more insights into FIP caused by the C3633 strain. We successfully generated recombinant virus (rC3663) from Fcwf-4 cells transfected with infectious cDNA that showed growth kinetics similar to those shown by the parental virus. Next, we constructed a reporter C3663 virus carrying the nanoluciferase (Nluc) gene to measure viral replication with high sensitivity. The inhibitory effects of different compounds against rC3663-Nluc could be measured within 24 h postinfection. Furthermore, we found that A72 cells derived from canine fibroblasts permitted FCoV replication without apparent cytopathic effects. Thus, our reporter virus is useful for uncovering the infectivity of type I FCoV in different cell lines, including canine-derived cells. Surprisingly, we uncovered aberrant viral RNA transcription of rC3663 in A72 cells. Overall, we succeeded in obtaining infectious cDNA clones derived from type I FCoV that retained its virulence. Our recombinant FCoVs are powerful tools for increasing our understanding of the viral life cycle and pathogenesis of FIP-inducing type I FCoV. IMPORTANCE Feline coronavirus (FCoV) is one of the most significant coronaviruses, because this virus induces feline infectious peritonitis (FIP), which is a lethal disease in cats. Tissue culture-adapted type I FCoV often loses pathogenicity, which complicates research on type I FCoV-induced feline infectious peritonitis (FIP). Since we previously found that type I FCoV strain C3663 efficiently induces FIP in specific-pathogen-free cats, we established a reverse genetics system for the C3663 strain to obtain recombinant viruses in the present study. By using a reporter C3663 virus, we were able to examine the inhibitory effect of 68 compounds on C3663 replication in Fcwf-4 cells and infectivity in a canine-derived cell line. Interestingly, one canine cell line, A72, permitted FCoV replication but with low efficiency and aberrant viral gene expression.

sequences (Fig. 1A) . The full-length infectious cDNA clone was designated pBAC-FCoV-C3663, and sequence analysis showed that it possessed 25 nucleotide mutations compared with the C3663 reference sequence (Table 1) . Of the 25 mutations, 11 were synonymous and 14 were nonsynonymous (Table 1) . Two synonymous mutations at nt 9831 and nt 9834 were introduced as the genetic marker, which disrupts the EcoRI site (ΔEcoRI), confirming virus recovery from the cDNA clone ( Fig. 1B ; see also Table 1) .

Virus recovery by pBAC-FCoV-C3663 transfection. We produced rC3663 virus from Fcwf-4 cells, which are highly susceptible to FCoV infection, by transfecting the cells with pBAC-FCoV-C3663. Low levels of cytopathic effects (CPE) were observed at 2 days posttransfection (dpt), and the levels became higher by 3 dpt. To determine the rates of rC3663 virus recovery, we employed reverse transcription-PCR (RT-PCR) on isolated RNA from rC3663 and the parental strain. We further confirmed the ΔEcoRI genetic maker by analyzing EcoRI digestion and Sanger sequencing ( Fig. 1C and D) . Next, we analyzed the virological features of the rC3663 virus by comparing the growth The full-length C3663 sequence was divided into 11 fragments (Fr1 to Fr11), and each fragment was sequentially assembled into the plasmid backbone. The asterisk (*) indicates the site of the genetic marker. (B) Nucleotide sequence of the C3663 genome between nt 9826 and nt 9837. The EcoRI restriction site is underlined. rFCoV was mutated from GAATTC to GAGTTT for disruption of the EcoRI site (ΔEcoRI) to use as a genetic marker. Mutated nucleotides are shown in gray boxes with white letters. (C) Confirmation of the genetic marker in the rC3663 genome by EcoRI treatment. RT-PCR products of parental virus C3663 and rC3663 that amplified a region that included the genetic marker were treated with EcoRI. The treated samples were subjected to electrophoresis to confirm disruption of the EcoRI site in the rC3663 genome. (D) Sequence analysis of C3663 and rC3663 at the genetic marker site. The EcoRI restriction site and ΔEcoRI genetic marker are underlined. (E) Growth kinetics of parental virus C3663 and rC3663 in Fcwf-4 cells. Each virus was inoculated onto Fcwf-4 cells at an MOI of 0.01 and incubated for 24, 48, and 72 h. Viral titers of culture supernatants were measured by plaque assays using Fcwf-4 cells. LOD, limit of detection. The data represent means Ϯ standard deviations (SD) of results from three independent experiments. (F) Northern blot analysis for detecting viral RNA in C3663-infected or rC3663-infected Fcwf-4 cells. Total RNAs from Fcwf-4 cells infected with parental C3663 or rC3663 were extracted and subjected to electrophoresis. The transferred viral RNAs were hybridized with DIG-labeled RNA targeting ORF 7b and 3=-UTR. gRNA, genomic RNA; sgRNA, subgenomic RNA. kinetics of rC3663 and parental C3663 in Fcwf-4 cells after inoculating them with viruses at a multiplicity of infection (MOI) of 0.01. The results showed that the growth kinetics of rC3663 was similar to that of the parental C3663 (Fig. 1E) . Furthermore, we compared levels of viral RNA replication in parental C3663-infected and rC3663-infected Fcwf-4 cells by Northern blotting analysis (Fig. 1F ) and found that the amounts of genomic RNA (gRNA) and of sg mRNAs in rC3663-infected Fcwf-4 cells were similar to those seen with the parental C3663-infected cells (Fig. 1F ). Taking the results together, we were able to successfully generate infectious cDNA clones derived from type I FCoV strain C3663 using the BAC system. Our results indicate that the recovered rC3663 virus possesses virological features identical to those possessed by the parental C3663 virus ( Fig. 1E and F) .

Establishment of reporter rC3663 bearing the Nluc gene. In virology, recombinant viruses carrying reporter genes (green fluorescent protein [GFP] , red fluorescent protein [RFP], or luciferase) present many advantages for analyzing viral characteristics and screening for therapeutic compounds (24, 25) . Thus, we attempted to construct an infectious cDNA clone of type I FCoV strain C3663 carrying an Nluc gene. Following the protocols of Tekes et al. (26) , we inserted the Nluc gene into pBAC-FCoV-C3663 in place of the ORF 3abc gene to obtain pBAC-FCoV-C3663-Nluc ( Fig. 2A) . The Nluc gene replaced a region beginning at the start codon of ORF 3a and extending to 71 nt upstream of the ORF 3c stop codon to retain the transcription regulatory sequence (TRS) of the M gene ( Fig. 2A) .

To examine Nluc expression in Fcwf-4 cells infected with rC3663-Nluc virus, we inoculated Fcwf-4 cells with rC3663-Nluc at an MOI of 0.01. Infection with rC3663 was used as a control. After 24, 48, and 72 h postinfection (hpi), we found that Nluc activity in rC3663-Nluc-infected Fcwf-4 cells-but not in rC3663-infected or mock-infected cells-increased in a time-dependent manner (Fig. 2B) .

We further investigated the viral growth of rC3663-Nluc in Fcwf-4 cells by harvesting the supernatants of rC3663-Nluc-infected or rC3663-infected Fcwf-4 cells at 24, 48, and 72 hpi and then determining infectious titers of the supernatants by plaque assays. The level of production of infectious virus particles from rC3663-Nluc-infected cells was comparable to that from rC3663-infected cells (Fig. 2C ). As shown in Fig. 2B and C, the 

increase in Nluc activity was significantly correlated with viral replication in rC3663-Nluc-infected cells. Our data indicate that rC3663 carrying the Nluc reporter gene represents a powerful tool for investigating type I FCoV replication and production. Application of the rC3663 reporter virus in compound screening. Before application of the rC3663 reporter virus to compound screening, we determined the sensitivity of the rC3663-Nluc virus to treatment with known inhibitors of CoV replication, namely, cyclosporine (CsA) (27, 28) and lopinavir (29) . After adsorption of rC3663-Nluc onto Fcwf-4 cells at an MOI of 0.01, the infected cells were treated with various concentrations of CsA or lopinavir. As shown in Fig. 2D and E, both compounds inhibited luciferase activity in a dose-dependent manner. Furthermore, viral RNA levels in CsA-treated or lopinavir-treated cells were measured by real-time RT-PCR ( Fig. 2D and E). Dose-dependent reductions of intracellular viral RNA levels were found for both compounds and were correlated with luciferase activity, suggesting that the sensitivity of detection of luciferase expression levels in rC3663-Nluc-infected cells is comparable to that of viral RNA expression levels.

Next, to determine the usefulness of the rC3663 reporter virus for screening antiviral compounds, we utilized a commercially available library of 68 protease inhibitors. Fcwf-4 cells were inoculated with rC3663-Nluc (MOI ϭ 0.01), and 10 M concentrations of each protease inhibitor with CsA and with dimethyl sulfoxide (DMSO) were used as positive and negative controls, respectively. The side-effects of protease inhibitors were determined by 3-(4,5-dimethyl-2-thiazolyl)-2,5-diphenyl-2H-tetrazolium bromide (MTT) assays (Fig. 2F) Fig. 2E . Overall, our results support the idea of the suitability of rC3663-Nluc for compound screening. Identification of permissive cell lines for type I FCoV. In vitro propagation of type I FCoV is limited to a few cell lines, including Fcwf-4 cells, AKD cells, and CRFK cells, because type I FCoV shows a CPE only in such cell lines (23, 30, 31) . Thus, it is difficult to investigate the infectivity of type I FCoV in cell lines derived from other animals, such as dogs. Nevertheless, we explored novel cell lines for propagation of type I FCoV by inoculating three canine-derived cell lines, A72 (canine fibroblasts), MDCK (canine kidney epithelial cells), and DH82 (canine macrophages), with the rC3663-Nluc virus (MOI ϭ 0.1) and investigated infectivity by measuring Nluc activity. Although no CPE was observed for rC3663-Nluc-infected A72 cells, Nluc activity was significantly high at 24 hpi and increased in a time-dependent manner ( Fig. 3A and C) . On the other hand, rC3663-Nluc-infected MDCK and DH82 cells did not exhibit detectable Nluc activity (Fig. 3B ).

To determine viral RNA replication in A72 cells, A72, MDCK, DH82, and Fcwf-4 cells were infected with rC3663-Nluc virus at an MOI of 0.01, followed by real-time RT-PCR analysis of RNA extracted at 24, 48, and 72 hpi. Although the amount of viral RNA in rC3663-Nluc-infected A72 cells was lower than in Fcwf-4 cells, the amount of viral RNA in A72 cells (but not in DH82 and MDCK cells) was still significantly high at 48 and 72 hpi (Fig. 3D) . These results indicate that A72 cells permit replication of type I FCoV C3663 virus.

Next, we determined the levels of production of infectious virus particles from rC3663-Nluc-infected A72 cells by collecting the culture supernatants at 24, 48, and 72 hpi and measuring viral titers by plaque assays with Fcwf-4 cells (Fig. 3E ). The supernatant infectious titers of Fcwf-4 cells reached 1.67 ϫ 10 5 PFU/ml at 72 hpi, and the amount of viral RNA determined by real-time RT-PCR increased in a time-dependent manner in Fcwf-4 cells ( Fig. 3D and E). As shown in Fig. 3A and D, A72 cells supported rC3663 virus replication, but the production of infectious viruses was lower than that seen with Fcwf-4 cells. Meanwhile, infectious viruses were not produced by infected MDCK and DH82 cells (Fig. 3E ). These results indicate that A72 cells produce progeny viruses-albeit with low efficiency-while MDCK and DH82 cells are not permissive cell lines for type I FCoV.

To further determine the low level of production of progeny virus by rC3663-Nlucinfected A72 cells, we examined the propagation of rC3663 virus (MOI ϭ 0.1) in A72 and Fcwf-4 cells by indirect immunofluorescence assays (IFA) using confocal microscopy analysis. Using an anti-FCoV N monoclonal antibody (MAb), N protein expression in A72 cells was observed by IFA, and the A72 cells exhibited small foci at 48 hpi compared with those exhibited by Fcwf-4 cells (Fig. 3F) . Therefore, infectious particles were found to have been generated in A72 cells at a low efficiency of infection expansion by progeny particles to neighboring cells. We also examined N protein expression levels using immunoblotting. As expected, N protein expression levels in rC3663 virusinfected A72 cells were significantly low (Fig. 3G ). Although the levels of production of progeny virus and N protein were low, our results suggest that canine-derived A72 cells represent a permissive cell line for type I FCoV infections without cytotoxic effects. Fig. 3 , and unlike the results seen with A72 cells, neither viral RNA replication nor progeny virus production was observed in MDCK and DH82 cells infected with rC3663 virus. These results led us to speculate that the entry receptor for type I FCoV is perhaps not expressed in MDCK and DH82 cells or that viral RNA replication of type I FCoV is not permitted in these cell lines. Unfortunately, the type I FCoV receptor remains unknown. Thus, we examined viral replication levels in cells transfected with pBAC-FCoV-C3663-Nluc. As a negative control, we generated pBAC-FCoV-C3663-Nluc-PolDead by mutating the active site of viral RNA-dependent RNA polymerase (RdRp; nsp12) from SDD to SAA (Fig. 4A ) and confirmed that virus rescue did not occur in Fcwf-4 cells transfected with pBAC-FCoV-C3663-Nluc-PolDead because of disrupted RdRp activity (data not shown). After transfecting MDCK cells with pBAC-FCoV-C3663- Fig. 3G , expression levels of N protein in infected A72 cells were significantly low. Thus, to determine expression levels of sg N mRNA (sg mRNA6), total RNAs extracted from rC3663 virus-infected or mock-infected Fcwf-4 and A72 cells at 48 and 72 hpi were subjected to Northern blot analysis with a specific type I FCoV 3= untranslated region (3=UTR) probe (Fig. 5A) . We found that all viral RNAs, including gRNA and sg mRNA2 to mRNA7, were detected in Fcwf-4 cells infected with rC3663 virus (Fig. 5A) . However, the expression levels of gRNA and sg mRNA2 to mRNA5 were significantly lower in infected A72 cells than in infected Fcwf-4 cells (Fig. 5A) .

Next, to determine the specific RNA signal in infected A72 cells, we generated a set of specific digoxigenin (DIG)-labeled RNA probes against S, 3abc, M, and N genes (Fig.  5B ). As shown in Fig. 5C , we identified seven specific viral mRNAs (gRNA and sg mRNA2 to mRNA7) using a combination of S, 3abc, M, and N-specific RNA probes in infected Fcwf-4 cells. An unexpected RNA signal was observed between sg mRNA6 and sg mRNA7 (Fig. 5C) . These results indicate that the two RNAs detected in A72 cells infected with rC3663 were sg mRNA6 and sg mRNA7.

We further examined the expression levels of viral sg mRNAs in cells infected with the parental C3663 strain or type I FCoV strain Yayoi using Northern blot analysis with specific RNA probes against the 3=-UTR. Although gRNA and sg mRNA2 to mRNA7 were detected in Fcwf-4 cells infected with the C3663 strain or the Yayoi strain, two specific mRNAs-sg mRNA6 and sg mRNA7-were observed in A72 cells infected with C3663 or Yayoi (Fig. 5A) . These results suggest that the decreased synthesis of viral mRNAs was not specific to the infectious clone of C3663.

Although the expression levels of sg mRNA6 and sg mRNA7 in A72 cells infected with rC3663 were low (Fig. 5A) , the ratio of sg mRNA6 expression to sg mRNA7 expression in infected A72 cells was different from that in infected Fcwf-4 cells. The level of sg mRNA7 in infected A72 cells determined on the basis of the level of sg mRNA6 was lower than that in infected Fcwf-4 cells (Fig. 5A) Infectious cDNA Clone of Virulent Type I FCoV Journal of Virology we used a smaller amount of total RNA extracted from infected Fcwf-4 cells for Northern blot analysis to adjust the expression levels of sg mRNA6 between infected Fcwf-4 and A72 cells. As a result, the expression levels of sg mRNA6 in A72 cells were identical to those in Fcwf-4 cells, whereas the expression levels of sg mRNA7 in A72 cells were lower than those in Fcwf-4 cells (Fig. 5D) . These findings indicate that aberrant expression of viral RNA occurred in infected A72 cells.

Epidemiological research conducted on type I FCoV has led to the proposal of the presence of a virulent factor(s) in its viral genome (19, 20) . However, experimental confirmation of the presence of a virulent factor(s) is difficult in vivo due to the lack of a feasible FIP cat model of type I FCoV. Recently, we found that type I FCoV strain C3663 has the ability to induce FIP in SPF cats (9) . Thus, in the present study, we constructed an infectious cDNA clone derived from the virulence-retaining type I FCoV strain C3663 by utilizing the BAC system. As a result, we successfully rescued recombinant type I FCoV by transfection of the BAC cDNA clone into Fcwf-4 cells, where the recovered rC3663 showed growth kinetics similar to that of the parental virus (Fig. 1 ). In addition, we generated recombinant type I FCoV carrying Nluc as a reporter gene and applied our reporter rC3663 virus to drug screening (Fig. 2) .

Several host proteases, such as furin, TMPRSS2, and cathepsins, are required for the entry step of CoVs as mediated by cleaving spike proteins at the cell surface or in endosomes (33) 

Consistent with these findings, we identified nafamostat mesylate (compound 38) as an FCoV inhibitor that shows no cytotoxic effects, suggesting that FCoV utilizes the host protease TMPRSS2 for entry. Another compound, lopinavir, is an inhibitor of the human immunodeficiency virus type 1 (HIV-1) protease and is used for AIDS treatment (35) . In coronavirus infections, lopinavir inhibits MERS-CoV, severe acute respiratory syndrome coronavirus (SARS-CoV), and human CoV (HCoV) 229E infection by inhibiting the CoV 3C-like protease (29) . HCoV-229E belongs to the same genus as FCoV-Alphacoronavirus (3). Unsurprisingly, we identified lopinavir (compound 31) as an FCoV inhibitor that does not exhibit cytotoxicity. Our results suggest nafamostat and lopinavir as potential therapeutic agents against FCoV infection.

Specific chemical compounds that are active against type I FCoV can be applied in two different strategies against FIP, namely, cure and prevention. Several studies have attempted to develop curative therapeutic agents against FIP (36) (37) (38) , and although these compounds suppressed FCoV replication in vitro, the compounds failed to suppress FIP in vivo (39, 40) . While the differences between the in vitro and in vivo activities of these compounds remain controversial, the highly complicated pathological mechanism of FIP, including such components as antibody-dependent enhancement of infection (ADE) (41, 42) and type III hypersensitivity (43, 44) , makes developing efficient and curative therapeutics against FIP difficult.

On the other hand, the emergence mechanism of FIPV is best explained by an internal mutation theory (16) (17) (18) where type I FCoV has the potential to become type I FIPV by acquiring mutations in its viral genome during replication in kittens. Therefore, controlling the viral load of type I FCoV in kittens is important in preventing the onset of FIP disease. If chemical compounds can suppress the viral load of type I FCoV in kittens, then the probability of disease onset should be reduced. Thus, the use of therapeutics against type I FCoV would be an efficient method of controlling FIP. In this regard, our reporter virus represents a powerful tool for finding compounds for use in the cure and/or prevention of FIP, as the virus can be utilized in high-throughput screening.

Several reports have addressed the possibility that some viruses among the members of alphacoronavirus-1, such as type II FCoV (45, 46) , type I canine CoV (CCoV) (47) , and type IIb CCoV (48), emerged as chimeric viruses via recombination events. Feline aminopeptidase N (APN) works as a viral receptor for type II CCoV and porcine transmissible gastroenteritis virus (TGEV) (15, 49) , suggesting that cats, and especially kittens, play an important role in producing new chimeric viruses. However, the possibility of recombination in other animals, such as dogs, has not been well-studied. Furthermore, to the best of our knowledge, no reports have shown type I FCoV induction of apparent CPE in cell culture except for Fcwf-4 cells. Indeed, the C3663 strain of type I FCoV did not exhibit CPE in canine-derived cells (Fig. 3C) . However, the level of luciferase activity in A72 cells infected with the rC3663-Nluc virus was significantly increased without concomitant CPE ( Fig. 3A and C) , and although the progeny virus production levels were low, A72 cells permitted type I FCoV replication (Fig. 3D) . We reasoned that investigations of type I FCoV infectivity in other cell lines should not rely solely on CPE-based assays. Furthermore, our findings indicate that events of recombination between type I FCoV and CCoV may occur in dogs. Specifically, type I FCoV may be transmitted from cats to dogs and a new CoV may emerge in dogs through recombination events. In fact, CCoV replicates in A72 cells (50) .

Like other positive-strand viruses, CoVs require host factors for replication (51) (52) (53) in infected cells. Unlike A72 cells, MDCK and DH82 cells did not permit viral replication of type I FCoV from pBAC-FCoV-C3663-Nluc (Fig. 3) . Thus, we hypothesized that essential host factors exist in canine-derived A72 cells but not in MDCK cells. Future research should elucidate the host factors that support replication of type I FCoV in A72 cells, as these factors would be attractive targets for the development of therapeutics against type I FCoV.

Recently, many researchers have focused on bats as a major reservoir of novel viruses (54) . Several novel viruses have been identified and/or isolated from bats, and some are related to human-pathogenic viruses, such as SARS-CoV (55) and MERS-CoV (56) . However, other animal species are also potential novel virus reservoirs; for example, Lau et al. reported that the Deltacoronavirus porcine CoV HKU15 (PorCoV HKU15) can be transmitted from birds to pigs (57) . In addition, HCoV OC43 is known to originate from bovine CoV (58) . Therefore, not only bats but also other animal species should be investigated for their potential as reservoirs or sources of emerging viruses. Indeed, we found that the canine-derived A72 cell line permitted replication of type I FCoV (Fig. 3) . Although the host receptors of type I FCoV remain unclear, use of our reporter virus for FCoV infection may provide insights on viral host jumping and emergence of novel viruses.

Generally, expression levels of sg mRNA6 are the highest among all viral mRNAs in CoV-infected cells (59) . While sg mRNA6 expression levels in A72 cells were lower than in Fcwf-4 cells, the expression ratio of sg mRNA6 to sg mRNA7 in infected A72 cells was different from that in infected Fcwf-4 cells (Fig. 5A ). In fact, the expression levels of sg mRNA6 in 2 g of total RNA extracted from infected A72 cells were similar to its expression levels in 0.05 g of total RNA extracted from Fcwf-4 cells (Fig. 5D ). In contrast, the expression levels of sg mRNA7 in A72 cells were lower than in Fcwf-4 cells (Fig. 5D) , suggesting that synthesis of sg mRNAs is impaired in A72 cells.

Our real-time RT-PCR method amplifies the 3=-UTR of viral RNA, which is present in all viral RNAs. Therefore, the increase in viral RNA in A72 cells (Fig. 3D) was possibly caused by the synthesis of sg mRNA6 (Fig. 5A) . To our knowledge, there are no reports on aberrant RNA transcription of CoVs in infected cells, such as in A72 cells. Several host proteins, such as members of the heterogeneous nuclear ribonucleoprotein (hnRNP) family or the DEAD box RNA helicase family, interact with the TRS of CoV RNA and regulate transcription/replication (60) . Thus, our data indicate that a host factor(s) present in Fcwf-4 cells, but not in A72 cells, regulates viral RNA transcription. On the basis of the aberrant expression of viral mRNA seen in rC3663 virus-infected A72 cells, further studies are required for uncovering the CoV RNA transcription mechanisms in A72 cells.

In conclusion, we generated recombinant type I FCoV strain C3663 using a BACbased reverse genetics system. Our recombinant virus can potentially be used to expand research on type I FCoV-induced FIP. Furthermore, infection of the A72 cells with the rC3663 virus revealed unusual viral mRNA expression patterns that may provide novel insights into the mechanisms of CoV transcription. Although further study is required to test whether the recombinant virus can induce FIP in vivo, our established type I FCoV clone has the potential to be a powerful tool for understanding FIP pathogenesis. For example, because the S gene is a candidate for FCoV pathogenesis (19, 20) , a chimeric S gene virus corresponding to our recombinant virus may provide novel insights into type I FCoV pathogenesis. Additionally, other groups have previously established type I FCoV reverse genetics using the Black strain (26) and type I field virus (61); thus, exploiting multiple approaches can synergistically further our understanding of FIP.

Cat-derived Fcwf-4 cells (CRL-2787) (9, 62) and the canine-derived cell lines A72 (CRL-1542) (63, 64) , MDCK (CCL-34) (65), and DH82 (CRL-10389) (66) were grown in Dulbecco's modified Eagle's medium (DMEM; Nacalai Tesque, Kyoto, Japan) supplemented with 10% (for Fcwf-4, A72, and MDCK cells) or 15% (DH82 cells) heat-inactivated fetal bovine serum (FBS), 100 U/ml penicillin, and 100 g/ml streptomycin (Nacalai Tesque). Cells were maintained in a humidified 5% CO 2 incubator at 37°C. The type I FCoV strains C3663 and Yayoi were used in this study (9, 23) and propagated in Fcwf-4 cells as described previously (14) .

BAC construction. The BAC DNA of SARS-CoV-Rep (67) was kindly provided by Luis Enjuanes (Spanish National Center for Biotechnology [CNB-CSIC], Madrid, Spain) and was used as a backbone BAC sequence to generate infectious cDNA carrying the full-length type I FCoV strain C3663 sequence. The full-length genomic sequence of C3663 was divided into 11 fragments that were then assembled into pBeloBAC11 plasmids in sequential order (Fig. 1A) . Fragments Fr1, Fr2, Fr3, Fr4, Fr5, Fr6, Fr7, Fr8, Fr9, Fr10, and Fr11 correspond to nt 11218 to 13811, nt 13812 to 16360, nt 16361 to 18675, nt 18676 to 20998, nt 20999 to 23400, nt 23401 to 26227, nt 9210 to 11217, nt 26228 to 28545, nt 1 to 2069, nt 2070 to 5152, and nt 5153 to 9834, respectively. Red/ET recombination was employed for fragment integration with a Red/ET recombination system counterselection BAC modification kit (Gene Bridges, Heidelberg, Germany) according to the manufacturer's instructions (32) . The cDNA corresponding to the 3= end of the genomic RNA between nt 28046 and 28545 with 25 nt of adenine (pA) and a partial HDV ribozyme (Rz) sequence was generated by chemical synthesis (Eurofins, Brussels, Belgium); this cDNA was used as a PCR template for synthesizing Fr8 (Fig. 1A) . For synthesizing other fragments, RT-PCR was carried out using genomic RNA of the C3663 strain as described under "RNA extraction and RT-PCR" (see below). For Fr11 amplification, reverse primer YT648 (5=-TTTGCCTTATAACTTCCGTAGGTGTAAACTCATCACATAATGAGCCA TAAGACA-3=) was designed to disrupt the EcoRI site of C3663 between nt 9829 and nt 9834 (Fig. 1B) . The cDNA clone carrying the full-length C3663 sequence was designated pBAC-FCoV-C3663. Sequence analysis of the full-genome sequence of C3663 in pBAC-FCoV-C3663 was carried out by Eurofins. For construction of reporter rC3663, we replaced the ORF 3abc genes of pBAC-FCoV-C3663 with Nluc (Promega, Fitchburg, WI) using the recombination method described above ( Fig. 2A) and designated the infectious cDNA clone pBAC-FCoV-C3663-Nluc. The same recombination method was applied for the construction of pBAC-FCoV-C3663-Nluc-PolDead, which possesses amino acid substitutions (SDD to SAA) at the active site of RdRp (nsp12; Fig. 4A ).

RNA extraction and RT-PCR. Viral RNAs from the viral stock of type I FCoV strain C3663 and strain rC3663 were extracted using a PureLink RNA minikit (Thermo Fisher Scientific, Waltham, MA) according to the manufacturer's instructions. RNA was subjected to reverse transcription using a SuperScript III first-strand synthesis system for RT-PCR (Invitrogen, Waltham, MA) and random hexamers according to the manufacturer's instructions, and PCR was carried out using PrimeSTAR GXL DNA polymerase (TaKaRa, Shiga, Japan). C3663 cDNA was used as a PCR template for synthesizing the recombination template as described above. rC3663 cDNA was used for genetic marker confirmation.

Total RNA from rC3663 virus-infected or mock-infected Fcwf-4 cells and A72 cells was isolated and subjected to reverse transcription as described above.

Plasmid construction. To obtain DIG-labeled riboprobes for detecting FCoV RNA via Northern blot analysis, we constructed the pSPT18-FCoV plasmid. The nucleotide sequence between nt 27803 and nt 28329 of the C3663 genome was cloned into the pSPT18 plasmid through ligation using HindIII and EcoRI restriction sites (DNA ligation kit Mighty Mix; TaKaRa). Furthermore, we constructed Escherichia coli expression plasmid pGEX6P-1-3663N181-377/GST to produce a glutathione S-transferase (GST)-fused partial N protein. The nucleotide sequence between nt 26745 and nt 27335, encoding amino acid residues 181 to 377 of C3663, was inserted into pGEX6P-1 through ligation using BamHI and XhoI restriction sites (DNA ligation kit Mighty Mix).

Rescue of rC3663 from susceptible Fcwf-4 cells. Fcwf-4 cells were seeded onto 6-well plates (Violamo; Misumi, Schaumberg, IL) at 4.0 ϫ 10 5 cells/well. After incubation at 37°C overnight, the cells were transfected using 4 g cDNA clones and XtremeGene 9 DNA transfection reagent (Roche, Basel, Switzerland) according to the manufacturer's instructions. After 3 days of culturing at 37°C, CPE were observed and culture supernatants were harvested. The supernatants were stored as passage 0 (P0) viruses. P0 viruses were passaged twice in fresh Fcwf-4 cells in 10-cm-diameter dishes (Violamo), and then P2 viruses were used for conducting experiments.

rC3663 genetic marker confirmation. To determine the genetic marker of rC3663, rC3663 cDNA was used as a template for PCR using primer pair YT649 (5=-GCATGCAACTGGAGGGTACT-3=) and YT650 (5=-AGAGGATAGCCAAAGCGGTC-3=) and PrimeSTAR GXL DNA polymerase. PCR products were purified using a HighPure PCR purification kit (Roche), and DNA samples were treated with EcoRI at 37°C overnight. Treated samples were then subjected to electrophoresis and cleavage was verified. Purified DNA samples were also used for sequence analysis to confirm the genetic marker. Sequence analyses were carried out by Eurofins using primer YT649.

Fcwf-4, A72, MDCK, and DH82 cells were seeded onto 6-well plates at 4.0 ϫ 10 5 , 2.0 ϫ 10 5 , 2.0 ϫ 10 5 , and 2.0 ϫ 10 5 cells/well, respectively, and cultured at 37°C overnight. Viruses were used to inoculate each cell line at an MOI of 0.01. After adsorption at 37°C for 1 h, the cells were washed twice with DMEM and then fresh DMEM containing 10% FBS was added. Infected cells were cultured at 37°C for 24, 48, and 72 h, and culture supernatants were collected and stored at -80°C until further use in titration assays. Next, infected cells were collected and washed once using phosphate-buffered saline (PBS) and the cell pellets divided into two aliquots for real-time RT-PCR and Western blot analysis.

Measurement of viral titers by the plaque assay. Fcwf-4 cells were seeded onto 6-well plates at 1.0 ϫ 10 6 cells/well and cultured at 37°C overnight. Samples were diluted with a 10-fold serial dilution from a 10ϫ dilution using DMEM. Diluted viruses (400 l) were added to the Fcwf-4 cells, and the mixture was incubated at 37°C for 1 h for adsorption. After adsorption, supernatants were removed, and the cells were washed twice using DMEM. Next, a mixture of 0.8% agarose (SeaPlaque GTG agarose; Lonza, Basel, Switzerland) with DMEM containing 10% FBS was overlaid on to the infected cells, and the cells were incubated at 37°C until CPE were observed. Finally, infected cells were fixed with phosphate-buffered formalin (Nacalai Tesque) and stained with crystal violet. The number of plaques was then determined, and viral titers were calculated as numbers of PFU per milliliter.

Northern blot analysis. RNAs from C3663-infected or rC3663-infected Fcwf-4 cells were used for Northern blot analysis as described elsewhere (32) . RNA samples were diluted to 2 g in 5 l by the use of UltraPure distilled water (DW) (Invitrogen) and then mixed with 5 l 2ϫ loading dye (New England Biolabs, Ipswich, MA). After heating was performed at 65°C for 5 min, 10-l RNA samples were subjected to electrophoresis with 1.2% denaturing agarose gel and transferred onto a positively charged nylon membrane (Roche). Northern blot analysis was performed using a DIG wash and block buffer set and a DIG luminescence detection kit (Roche). DIG-labeled riboprobes for detecting viral RNAs were generated using pSPT18-FCoV and a DIG RNA labeling kit (SP6/T7; Roche) as described previously (68, 69) .

Measurement of Nluc activity. Fcwf-4 cells (1.0 ϫ 10 5 cells/well) and A72, MDCK, and DH82 cells (all 1.5 ϫ 10 5 cells/well) were seeded onto 24-well plates (Violamo) and cultured at 37°C overnight. After washing, rC3663-Nluc or rC3663 was inoculated at an MOI of 0.01 or 0.1. After adsorption, the cells were washed twice using DMEM and then fresh DMEM containing 10% FBS was added. The cells were then incubated at 37°C for 24, 48, and 72 h, after which culture supernatants were removed and cells lysed using passive buffer (Promega). Luciferase activity levels were measured using a Nano-Glo luciferase assay system (Promega) on a PowerScan HT luminometer (DS Pharma Biomedical, Osaka, Japan). The experiments were carried out in triplicate.

Measurement of inhibitory effects on type I FCoV replication. Fcwf-4 cells were seeded onto 24-well plates at 1.0 ϫ 10 5 cells/well and cultured at 37°C overnight. After washing of the cells was performed, 100 l rC3663-Nluc was used for inoculation at an MOI of 0.01. After adsorption at 37°C for 1 h, the infected cells were washed twice with DMEM and then fresh DMEM supplemented with 10% FBS and different concentrations of CsA (Sigma-Aldrich, St. Louis, MO) or lopinavir (Sigma-Aldrich) was added. Using a 2-fold serial dilution performed with DMSO, CsA was diluted from 50 M to 3.125 M and lopinavir from 30 M to 3.75 M. DMSO was also used as a control. After incubation at 37°C for 24 h, culture supernatants were removed, and the cells lysed with passive buffer. Nluc activity was then measured as described above.

Compound screening using a protease inhibitor library. We used a chemical library of 68 compounds (L1100) (Protease Inhibitor Library; TargetMol) (70) for compound screening ( Table 2) . Fcwf-4 cells were seeded onto 96-well plates (Violamo) at 2.0 ϫ 10 4 cells/well and cultured at 37°C overnight. Mixtures containing rC3663-Nluc at an MOI of 0.01 and a 10 M concentration of each compound from the library were prepared in DMEM and added to the cultured Fcwf-4 cells. CsA (10 M) and DMSO were used as the positive and negative controls, respectively. After incubation for 24 h, the culture supernatants were removed and the cells lysed in passive buffer. Nluc activity was measured as described above.

MTT assay. Fcwf-4 cells were seeded onto 96-well plates at 1.0 ϫ 10 4 cells/well and cultured at 37°C overnight. Each compound (10 M in DMEM) from the protease inhibitor library was added to the cultured Fcwf-4 cells. After incubation for 24 h, MTT assays were carried out using an MTT cell count kit (Nacalai Tesque) according to the manufacturer's instructions. Absorbance at 570 nm was then measured with the PowerScan HT luminometer.

Real-time RT-PCR. Cell pellets of infected Fcwf-4, A72, MDCK, and DH82 cells were lysed, and then total RNA was extracted using a PureLink RNA minikit. RNA was then used for real-time RT-PCR with Thunderbird Probe one-step quantitative PCR (qPCR) mix (Toyobo, Osaka, Japan) and run on a StepOne real-time PCR system (Applied Biosystems, Waltham, MA). For quantification of FCoV RNA, we used 3=-UTR targeting primers FcoV1128f (5=-GATTTGATTTGGCAATGCTAGATTT-3=; nt 28398 to 28422 of C3663) and FcoV1229r (5=-AACAATCACTAGATCCAGACGTTAGCT-3=; nt 28473 to 28499 of C3663) as well as a specific probe (5=-TCCATTGTTGGCTCGTCATAGCGGA-3=; nt 28446 to 28470 of C3663) labeled with 6-carboxyfluorescein (FAM) (71) . For quantification of GAPDH (glyceraldehyde-3-phosphate dehydrogenase) mRNA, wk1288 (5=-GAAGGTGAAGGTCGGAGT-3=) and wk1289 (5=-GAAGATGGTGATGGGATTTC-3=) were used as primers and FAM-labeled wk1290 (5=-CAAGCTTCCCGTTCTCAGCC-3=) was used as a probe (32) . Cycling conditions were 95°C for 1 min followed by 40 cycles of 95°C for 15 s and 58°C for 45 s. (Continued on next page) Construction of monoclonal antibody against N protein of type I FCoV. A GST-fused partial N protein, C3663N181-377/GST, was expressed in Escherichia coli carrying pGEX6P-1-C3663N181-377/GST and purified using glutathione Sepharose 4B beads (GE Healthcare, Chicago, IL). The purified protein was used as an antigen for immunization of BALB/c mice. The method of monoclonal antibody (MAb) establishment was described previously (72) . Screening of hybridomas was performed by enzyme-linked immunosorbent assays (ELISA) using purified C3663N181-377/GST or GST proteins. ELISA was carried out as described previously (73) . The experiments yielded a clone (designated 4D10) which produced MAbs against type I FCoV N protein.

Western blot analysis. Infected Fcwf-4, A72, MDCK, and DH82 cells were lysed in lysis buffer (100 mM Tris-HCl [pH 8.0], 150 mM NaCl, and 1% Triton X-100) and centrifuged at 16,000 ϫ g for 10 min at 4°C. Supernatants were then collected and mixed with 2ϫ sample buffer (0.1 M Tris-HCl [pH 6.8], 4% sodium dodecyl sulfate [SDS], 20% glycerol, 0.004% bromophenol blue, 10% 2-mercaptoethanol). Next, the boiled samples were subjected to electrophoresis by SDS-polyacrylamide gel electrophoresis (SDS-PAGE) and transferred onto a polyvinylidene difluoride (PVDF) membrane (Merck Millipore, Billerica, MA). The membranes were blocked in 3% skim milk-PBS containing 0.05% Tween 20 (PBS-T; Nacalai Tesque). The primary antibodies were mouse anti-FCoV-N monoclonal antibody (4D10) and mouse anti-␤-actin antibody (Sigma-Aldrich), while goat anti-mouse IgG horseradish peroxidase (HRP; Sigma-Aldrich) was used as a secondary antibody. ChemiLumi One Ultra (Nacalai Tesque) was used for visualization (32) .

IFA. Fcwf-4 and A72 cells were seeded onto 35-mm-diameter glass-bottom dishes (Matsunami Glass, Osaka, Japan) at 2.0 ϫ 10 5 and 1.0 ϫ 10 5 cells/well, respectively. After incubation at 37°C overnight, the cells were inoculated with rC3663 at an MOI of 0.1. Following adsorption at 37°C for 1 h, the supernatants were removed and fresh DMEM containing 10% FBS was added. The infected cells were cultured at 37°C for 48 h and then fixed with 4% paraformaldehyde-PBS. After washing was performed once with PBS, the cells were permeabilized for 15 min at room temperature with PBS containing 0.5% Triton X-100. The cells were then incubated with mouse anti-FCoV-N 4D10 monoclonal antibody at 4°C overnight, after which they were washed three times with PBS and incubated with CF488-conjugated anti-mouse IgG (Sigma-Aldrich) (1:500) for 1 h at room temperature. Finally, the cells were washed three times with PBS and then Fluoroshield with DAPI (4=,6-diamidino-2-phenylindole; ImmunoBioScience, Mukilteo, WA) was used as a mounting medium and for counterstaining of nuclei with DAPI. The cells were observed with a FluoView FV1000 laser scanning confocal microscope (Olympus, Tokyo, Japan).

Transfection of MDCK cells with pBAC-FCoV-Nluc. MDCK cells were seeded onto 24-well plates at 1.5 ϫ 10 5 cells/well and cultured at 37°C overnight. The cells were then transfected with 1 g pBAC-FCoV-Nluc and 0.1 g pcDNA3.1-fluc (32) using Lipofectamine 2000 (Thermo Fisher Scientific) according to the manufacturer's instructions. pBAC-FCoV-Nluc-PolDead was used as a control. Transfected cells were incubated at 37°C for 6 h, after which the supernatants were removed and fresh DMEM containing 10% FBS was added. After incubation was performed for an additional 18, 42, and 66 h (i.e., 24, 48, and 72 h after transfection), culture supernatants were removed and the cells lysed in passive buffer. Levels of Nluc and firefly luciferase activity were measured as described above and with a luciferase assay system (Promega), respectively. Nluc activity was normalized to firefly luciferase activity. For other virological tests (Fig. 4B to D) , the experiment was scaled up from 24-well plates to 6-well plates. 

Coronavirus pathogenesis and the emerging pathogen severe acute respiratory syndrome coronavirus

Coronavirus genomics and bioinformatics analysis

2012. Virus taxonomy: ninth report of the International Committee on Taxonomy of Viruses

Factors associated with pathogen seroprevalence and infection in Rocky Mountain cougars

Prevalence of antibodies to feline parvovirus, calicivirus, herpesvirus, coronavirus, and immunodeficiency virus and of feline leukemia virus antigen and the interrelationship of these viral infections in free-ranging lions in east Africa

Prevalence and implications of feline coronavirus infections of captive and free-ranging cheetahs (Acinonyx jubatus)

Pathogenicity studies of feline coronavirus isolates 79-1146 and 79-1683

Some important disorders of cats

Feline infectious peritonitis virus with a large deletion in the 5'-terminal region of the spike gene retains its virulence for cats

Antigenic comparison of feline coronavirus isolates: evidence for markedly different peplomer glycoproteins

Antigenic analysis of feline coronaviruses with monoclonal antibodies (MAbs): preparation of MAbs which discriminate between FIPV strain 79-1146 and FECV strain 79-1683

Prevalence of feline coronavirus types I and II in cats with histopathologically verified feline infectious peritonitis

Feline coronavirus serotypes 1 and 2: seroprevalence and association with disease in Switzerland

Differentiation of feline coronavirus type I and II infections by virus neutralization test

Differences in virus receptor for type I and type II feline infectious peritonitis virus

Significance of coronavirus mutants in feces and diseased tissues of cats suffering from feline infectious peritonitis

Two related strains of feline infectious peritonitis virus isolated from immunocompromised cats infected with a feline enteric coronavirus

Feline infectious peritonitis viruses arise by mutation from endemic feline enteric coronaviruses

Spike protein fusion peptide and feline coronavirus virulence

Mutation in spike protein cleavage site and pathogenesis of feline coronavirus

Disease outcome and cytokine responses in cats immunized with an avirulent feline infectious peritonitis virus (FIPV)-UCD1 and challenge-exposed with virulent FIPV-UCD8

Attempted immunization of cats against feline infectious peritonitis, using avirulent live virus or sublethal amounts of virulent virus

Antigenic and plaque variations of serotype II feline infectious peritonitis coronaviruses

Influenza A virus encoding secreted Gaussia luciferase as useful tool to analyze viral replication and its inhibition by antiviral compounds and cellular proteins

Highly sensitive real-time in vivo imaging of an influenza reporter virus reveals dynamics of replication and spread

Genome organization and reverse genetic analysis of a type I feline coronavirus

Cyclosporin A inhibits the replication of diverse coronaviruses

The SARS-coronavirus-host interactome: identification of cyclophilins as target for pan-coronavirus inhibitors

Screening of an FDA-approved compound library identifies four small-molecule inhibitors of Middle East respiratory syndrome coronavirus replication in cell culture

Plaque assay, polypeptide composition and immunochemistry of feline infectious peritonitis virus and feline enteric coronavirus isolates

Comparative properties of feline coronaviruses in vitro

MERS coronavirus nsp1 participates in an efficient propagation through a specific interaction with viral RNA

Host cell proteases: critical determinants of coronavirus tropism and pathogenesis

Identification of nafamostat as a potent inhibitor of Middle East respiratory syndrome coronavirus S protein-mediated membrane fusion using the split-protein-based cell-cell fusion assay

Lopinavirritonavir versus nelfinavir for the initial treatment of HIV infection

Synergistic antiviral effect of Galanthus nivalis agglutinin and nelfinavir against feline coronavirus

Potent inhibition of feline coronaviruses with peptidyl compounds targeting coronavirus 3C-like protease

Effect of chloroquine on feline infectious peritonitis virus infection in vitro and in vivo

Randomized, placebo controlled study of the effect of propentofylline on survival time and quality of life of cats with feline infectious peritonitis

An update on feline infectious peritonitis: diagnostics and therapeutics

Monoclonal antibodies to the spike protein of feline infectious peritonitis virus mediate antibody-dependent enhancement of infection of feline macrophages

Antibodydependent enhancement occurs upon re-infection with the identical serotype virus in feline infectious peritonitis virus infection

Isolation and characterization of feline C3 and evidence for the immune complex pathogenesis of feline infectious peritonitis

Antibody, immune complexes, and complement activity fluctuations in kittens with experimentally induced feline infectious peritonitis

Feline coronavirus type II strains 79-1683 and 79-1146 originate from a double recombination between feline coronavirus type I and canine coronavirus

Emergence of pathogenic coronaviruses in cats by homologous recombination between feline and canine coronaviruses

Genetic diversity of a canine coronavirus detected in pups with diarrhoea in Italy

Recombinant canine coronaviruses related to transmissible gastroenteritis virus of swine are circulating in dogs

Feline aminopeptidase N serves as a receptor for feline, canine, porcine, and human coronaviruses in serogroup I

Two genotypes of canine coronavirus simultaneously detected in the fecal samples of dogs with diarrhea

Efficient activation of the severe acute respiratory syndrome coronavirus spike protein by the transmembrane protease TMPRSS2

Mouse hepatitis coronavirus RNA replication depends on GBF1-mediated ARF1 activation

Nucleocapsid phosphorylation and RNA helicase DDX1 recruitment enables coronavirus transition from discontinuous to continuous transcription

Bats as reservoirs of severe emerging infectious diseases

Bats are natural reservoirs of SARS-like coronaviruses

Human betacoronavirus 2c EMC/2012-related viruses in bats, Ghana and Europe

Discovery and sequence analysis of four deltacoronaviruses from birds in the Middle East reveal interspecies jumping with recombination as a potential mechanism for avian-toavian and avian-to-mammalian transmission

Complete genomic sequence of human coronavirus OC43: molecular clock analysis suggests a relatively recent zoonotic coronavirus transmission event

SARS-coronavirus replication/transcription complexes are membrane-protected and need a host factor for activity in vitro

RNA-RNA and RNA-protein interactions in coronavirus replication and transcription

Reverse genetics for type I feline coronavirus field isolate to study the molecular pathogenesis of feline infectious peritonitis

Expression of feline infectious peritonitis coronavirus antigens on the surface of feline macrophage-like cells

Establishment of a canine cell line: derivation, characterization, and viral spectrum

Establishment of canine and feline cells expressing canine signaling lymphocyte activation molecule for canine distemper virus study

Characterization of an established line of canine kidney cells (MDCK)

A macrophagemonocyte cell line from a dog with malignant histiocytosis

Construction of a severe acute respiratory syndrome coronavirus infectious cDNA clone and a replicon to study coronavirus RNA synthesis

Severe acute respiratory syndrome coronavirus nsp1 protein suppresses host gene expression by promoting host mRNA degradation

Severe acute respiratory syndrome coronavirus nsp1 facilitates efficient propagation in cells through a specific translational shutoff of host mRNA

Characterization of recombinant Flaviviridae viruses possessing a small reporter tag

One-tube fluorogenic reverse transcription-polymerase chain reaction for the quantitation of feline coronaviruses

Production and characterization of monoclonal antibodies to Japanese encephalitis virus

Development and application of an indirect enzyme-linked immunosorbent assay for serological survey of Japanese encephalitis virus infection in dogs

We thank Luis Enjuanes (CNB-CSIC, Madrid, Spain) for providing the SARS-CoV-Rep BAC DNA. We also thank Kaede Yukawa for secretarial assistance and Kanako Yoshizawa and Nozomi Shimada for their technical assistance.This work was supported in part by grants-in-aid for the Research Program on Emerging and Re-Emerging Infectious Diseases from the Japan Agency for Medical Research and Development and Japanese Society for the Promotion of Science (JSPS) (KAKENHI grant JP16K08811, JP18fk0108058, and JP15J07066). Y.T. was supported by a JSPS Research Fellowship for young scientists. We have no potential conflicts of interest in relation to this study.