【病毒外文文献】2019 Differential recognition of peptides within feline coronavirus polyprotein 1 ab by sera from healthy cats and cats

Accepted Manuscript Differential recognition of peptides within feline coronavirus polyprotein 1 ab by sera from healthy cats and cats with feline infectious peritonitis Anastasia Chernyavtseva Nick J Cave John S Munday Magdalena Dunowska PII S0042 6822 19 30103 5 DOI https doi org 10 1016 j virol 2019 04 003 Reference YVIRO 9066 To appear in Virology Received Date 12 March 2019 Revised Date 8 April 2019 Accepted Date 10 April 2019 Please cite this article as Chernyavtseva A Cave N J Munday J S Dunowska M Differential recognition of peptides within feline coronavirus polyprotein 1 ab by sera from healthy cats and cats with feline infectious peritonitis Virology 2019 doi https doi org 10 1016 j virol 2019 04 003 This is a PDF file of an unedited manuscript that has been accepted for publication As a service to our customers we are providing this early version of the manuscript The manuscript will undergo copyediting typesetting and review of the resulting proof before it is published in its final form Please note that during the production process errors may be discovered which could affect the content and all legal disclaimers that apply to the journal pertain M A N U S C R I P T A C C E P T E D ACCEPTED MANUSCRIPT 1 Differential recognition of peptides within feline coronavirus polyprotein 1ab by sera from healthy cats and cats with feline infectious peritonitis Anastasia Chernyavtseva1 2 Nick J Cave1 John S Munday1 Magdalena Dunowska1 5 1 School of Veterinary Science Massey University Palmerston North New Zealand 2 Current address Animal Health Laboratory Ministry for Primary Industries Wallaceville New Zealand Author for correspondence M Dunowska massey ac nz 10 Keywords FIP feline coronavirus feline infectious peritonitis non structural proteins antibody response feline enteric coronavirus 15 Abbreviations Ab antibody FCoV feline coronavirus FECV feline enteric coronavirus 20 FIP feline infectious peritonitis neg negative nsp non structural protein pos positive Pp1ab polyprotein 1ab 25 M A N U S C R I P T A C C E P T E D ACCEPTED MANUSCRIPT 2 Abstract The aim of the study was to identify peptides within the polyprotein Pp 1ab that are differentially recognised by cats with either enteric or systemic disease following infection with feline coronavirus Overlapping 12 mer peptides n 28 426 across the entire Pp1ab were arrayed on peptide chips and 30 reacted with pooled sera from coronavirus seropositive cats and from one seronegative cat Eleven peptides were further tested in ELISA with individual serum samples and five were selected for further screening Two peptides 16433 and 4934 in the nsp3 region encoding the papain 1 and 2 proteases were identified for final testing Peptide 4934 reacted equally with positive sera from healthy cats and cats with feline infectious peritonitis FIP while peptide 16433 was recognized 35 predominantly by FIP affected cats The value of antibody tests based on these peptides in differentiating between the enteric and FIP forms of feline coronavirus infection remains to be determined 40 Highlights Cats develop antibodies to polyprotein 1ab Pp1ab of feline coronavirus This is most evident for cats with feline infectious peritonitis FIP Differences exist in responses to selected peptides between FIP and non FIP cats Such differences may be utilised for development of a serological test for FIP 45 M A N U S C R I P T A C C E P T E D ACCEPTED MANUSCRIPT 3 Introduction Feline enteric coronavirus FECoV infections are common among cats worldwide Pedersen 2009 2014 Infected cats typically remain healthy or develop mild self limited enteritis However in a 50 small proportion of FECoV infected cats viral variants that have lost tropism for enterocytes and gained ability to replicate in macrophages emerge This combined with ineffective immune response to the virus leads to development of a severe systemic disease termed feline infectious peritonitis FIP Addie et al 2009 Pedersen 2014 Throughout this manuscript the term feline coronavirus FCoV is used whenever distinction between the two pathotypes is not relevant The 55 terms FECoV and FIP virus FIPV are used when referring specifically to only one of the two pathotypes Clinically FIP is characterised by non specific signs such as fever loss of appetite and weight jaundice or diarrhoea that are accompanied by accumulation of protein rich exudate in abdominal or chest cavities wet form or development of neurological deficits or ocular lesions dry form 60 Pedersen 2009 There is no effective treatment and 95 of affected cats die within one year of the onset of disease Legendre et al 2017 While post mortem diagnosis of FIP is fairly straight forward due the presence of characteristic histopathological lesions of widespread vasculitis Kipar and Meli 2014 Pedersen 2009 the same is not true for ante mortem assessment A kitten from a multi cat environment that presents with compatible clinical signs is very likely to be affected by FIP 65 Pedersen 2009 However both attending veterinarians and owners of such cats often desire laboratory confirmation of the presumptive FIP diagnosis in order to facilitate an emotionally difficult decision to euthanize the cat The fact that FIP usually affects young animals combined with the variability in clinical and laboratory findings Riemer et al 2016 contributes to the problem As FIPV is highly macrophage associated detection of the virus ante mortem requires invasive 70 techniques and diagnostic sensitivity of the currently available tests is low Pedersen et al 2015 Tasker 2018 In one study the virus was detected in only approximately half of the effusion samples and none of the serum plasma samples from FIP cats using a commercially available qPCR test Felten et al 2017 Cats exposed to FECV raise antibodies against structural proteins of the virus and the titer of these antibodies often rise to high levels after macrophage tropic mutants arise 75 and FIP disease begins Pedersen et al 1977 However serology has been considered of limited diagnostic value due to inability to differentiate between immune responses to FECV and FIPV Feline coronaviruses are classified in the family Coronaviridae within the order Nidovirales King et al 2012 Other nidoviruses include members of Arteriviridae Roniviridae and Mesoniviridae families Typical for all nidoviruses coronavirus non structural genes are expressed soon after 80 M A N U S C R I P T A C C E P T E D ACCEPTED MANUSCRIPT 4 infection from two large open reading frames ORF 1a and 1b The two polyprotein Pp products Pp1a and Pp1ab are then auto cleaved into 16 non structural proteins nsps that are essential for viral replication Hagemeijer et al 2012 Perlman and Netland 2009 Thus nsps are one of the first viral proteins abundantly produced within the infected cells It is therefore logical to assume that cats infected with FCoV would raise an early immune response to at least some of FCoV nsps 85 However while a number of previous studies focused on immune responses to structural proteins of the virus Satoh et al 2011 Takano et al 2014 there are no data related to immune responses to nsps of FCoV Similarly studies with coronaviruses other than FCoV were designed to identify immunodominant epitopes within viral structural proteins but not those present within nsps Duan et al 2005 Yu et al 2007 90 Several nsps have been identified as targets for adaptive humoral immune responses in nidoviruses other than coronaviruses For example a total of 10 non linear B cell epitopes were identified in nsp1 nsp2 and nsp4 of porcine respiratory and reproductive syndrome virus PRRSV Oleksiewicz et al 2001b and sera from boars infected with PRRSV type I contained antibodies to both structural and non structural proteins of the virus Oleksiewicz et al 2001a In another study sera from pigs 95 infected with different PRRSV viruses reacted with nsp1 nsp2 and nsp7 Brown et al 2009 Johnson et al 2007 described the presence of cross reactive epitopes in nsp1 and nsp2 of various PRRSV strains as well as type specific epitopes within a hyper variable region of nsp2 The latter provided a basis for development of serological assays able to differentiate between antibody responses due to infection versus vaccination A number of nsps were also recognised by sera from 100 horses infected with equine arteritis virus EAV Go et al 2011 Interestingly there seemed to be a difference in the immune response to EAV nsps between horses that cleared the infection and those that became carriers Go et al 2011 There was also a difference between the antibody response to nsps of vaccinated horses and those experimentally infected with a virulent strain of EAV suggesting that serological responses to nsps may be useful as a diagnostic tool to differentiate between 105 infections with viruses of different virulence The aim of this study was to investigate humoral immune responses to FCoV nsps from Pp1ab in seropositive cats with different disease outcomes We hypothesised that identification of immunodominat epitopes that are recognised by sera from the majority of healthy FCoV seropositive cats but not by FIP affected cats would provide potential candidates for future 110 development of vaccines against FIP Immune responses raised by such vaccines may have the advantage of recognising FCoV infected cells early in infection without the disadvantage of antibody dependent enhancement of infection associated with humoral immune responses to structural proteins of the virus Balint et al 2014 Early clearance of FECoV infected cells would M A N U S C R I P T A C C E P T E D ACCEPTED MANUSCRIPT 5 minimize the chances of de novo emergence of FIP associated mutations and subsequent 115 development of FIP Identification of immunodominant epitopes that are recognised by sera from the majority of FIP affected cats but not by healthy FCoV seropositive cats would provide targets for development of FIPV specific serological test Materials and Methods Cat enrolment and sampling 120 Cats with presumptive diagnosis of FIP were enrolled into the study from cases presented at Massey University Veterinary Teaching Hospital as well as via local veterinary practices The samples collected included serum plasma formalin fixed tissue samples and abdominal thoracic effusion if present Fixed tissues were used for routine preparation of slides for histologic examination Haematoxylin eosin stained sections were assessed by a boarded pathologist JM for the presence 125 of lesions characteristic of FIP Kipar and Meli 2014 Pedersen 2009 On occasion immunohistochemistry IHC was performed to further support diagnosis of FIP Following de waxing and rehydration slides were subjected to heat induced antigen retrieval 98 C in citrate buffer pH 6 0 for 20 minutes loaded into Sequenza rack Thermo Fisher Scientific permeabilized 2 x 5 minutes in 0 25 TritonX in phosphate buffered saline PBS pH 7 0 washed 130 with PBS containing 0 2 Tween 20 PBST blocked Superblock with 0 1 Tween 20 Thermo Fisher Scientific for 30 minutes and incubated with primary antibody FIPV3 70 Santa Cruz Biotechnology diluted 1 200 in PBST at 4 C overnight The following morning slides were washed in PBST quenched with 3 H2O2 in methanol for 10 minutes and washed again The binding of primary antibody was detected using Mouse on Farma HRP polymer Biocare Medical and Betazoid DAB 135 chromogen kit Biocare Medical according to the manufacturer s instructions After the final wash in water the slides were counter stained with Gills haematoxylin Samples from non FIP cats included archival serum plasma samples from cats from Massey University Feline Nutrition Unit colony cats serum samples submitted to a diagnostic laboratory for unrelated reasons and serum samples from healthy cats from households with FIP cats Table 1 140 Samples collected for unrelated purposes either routine yearly health checks or nutrition trials were used whenever possible from colony cats to minimize the necessity for blood collection Samples collected from the same cat but at different dates were labelled with the same sample ID but different letter suffix e g 57 57a 57b The sampling protocol has been approved by the Massey University Animal Ethics Committee 145 M A N U S C R I P T A C C E P T E D ACCEPTED MANUSCRIPT 6 All serum plasma and effusion samples were tested for the presence of antibody to structural proteins of FCoV using a commercially available ImmunoComb Feline Coronavirus FIP test Biogal Laboratories according to the manufacturer s instructions The intensity of a blue colour of the sample dot was visually assessed on a scale 0 to 6 with the higher number indicating higher levels of FCoV antibody Addie et al 2015 150 Hybridisation of feline sera to peptide chips Custom peptide chips were commercially synthesised LLC Biosciences Each library included 28 426 12 mer sequences covering all available variants of the entire Pp1ab of FCoV with one amino acid walking distance Each chip was then hybridised with the following samples Chip1 control serum from a cat negative for FCoV antibody 66 155 Chip 2 Pooled sera n 5 from FCoV antibody positive healthy colony cats 45 57 67a 68 and 69 The selected cats represented surviving siblings of cats that had died due to FIP For each cat archival blood sample from the same year as the death of the FIP affected sibling was used in the study Chip 3 Pooled sera n 5 from FCoV antibody positive FIP cats 04 05 07 08 09 160 The level of fluorescence at each spot indicated the level of binding of the feline sera to a specific peptide In order to visualise the level of fluorescence across Pp1ab all peptide sequences as well as the reference sequence AAY16374 were back translated using a universal amino acid code The back translated nucleotide sequences were then mapped to the back translated Pp1ab sequence copied in the mapped order to an Excel spreadsheet and then linked to the fluorescence data 165 Peptide ELISAs Selected peptides n 11 Table 2 were used as antigens in ELISA based format and tested with each individual serum sample that contributed to sample pools used for hybridisation to peptide chips The selection of peptides was based on the presence of a comparatively stronger signal with a serum pool from FIP cats than with a serum pool from healthy FCoV seropositive cats peptides 22880 170 28424 16431 24480 24481 and 16433 or vice versa peptides 4929 4934 25438 4774 and 4775 Five peptides 25438 26242 28423 25438 16433 were further tested with an extended number of serum plasma effusion samples n 50 and two of those 25438 16433 were further evaluated with additional samples n 53 for a total of 103 samples each CovaLink NH plates Nunc Thermo Scientific were coated overnight at room temperature with 100 175 L of a relevant peptide Sulfo NHS solution 10 g mL of peptide 0 184 mg mL of Sulfo N M A N U S C R I P T A C C E P T E D ACCEPTED MANUSCRIPT 7 hydroxysuccinimide in carbonate buffer 15 mM Na2CO3 35 mM NaHCO3 pH 9 6 Following three washes with distilled water plates were blocked with 300 L of 10 Goat Serum Gibco 16210 072 in carbonate buffer at room temperature for two hours The wells were then emptied and 100 L well of each test sample diluted 1 10 in dilution buffer phosphate buffered saline PH 7 2 PBS 180 with 10 goat serum and 0 05 Tween 20 were added in duplicate to the plate The plate was incubated at room temperature for one hour washed five times with CovaBuffer 2 M NaCl 40 mM MgSO4 7H20 0 05 Tween 20 in PBS Horseradish HRP conjugated Goat Anti Cat IgG Fc Abcam ab112801 diluted 1 100000 in dilution buffer 100 L well at room temperature was then added to each well and the plate was incubated for one hour at room temperature After five washes with 185 CovaBuffer 100 L TMB ELISA Substrate Highest Sensitivity Abcam ab171522 was added to well and the plate was incubated at 37 C for 10 min The reaction was stopped by the addition of 100 L of 1M H2SO4 to each well The results were presented as corrected optical density values at 450 nm OD450 which were calculated by subtracting the OD450 of the no peptide well from the OD450 of the peptide coated well tested with the same serum sample 190 The diagnostic sensitivity and specificity of ELISA tests for detection of FCoV antibody positive cats 4934 ELISA or FIP affected cats 16433 were calculated using an on line calculator available at https www medcalc org calc diagnostic test php Table 3 As similar corrected OD450 values were obtained when testing different sample types from the same cat whenever available see Figure 5 values for only one sample type in the order of preference serum plasma or effusion were 195 included in the analysis if more than one sample type was available from the same cat on the same sampling date Results Cats A total of 92 samples from 72 cats were available for the study Table 1 The majority 28 42 66 7 200 of the non FIP cats were clinically normal at the time of sampling These comprised colony cats 39 samples from 19 cats two kittens 11 and 34 from the same households as FIP cases and seven healthy cats kittens from a breeding colony of Siberian Forest cats 35 to 41 Two of the FIP affected kittens 09 and 30 from private households were Siberian Forest cats originally obtained from the same breeder A small number 5 42 11 9 of non FIP cats presented with clinical signs 205 suggestive of FIP but did not have histological lesions typical for FIP 3 6 10 12 15 The remaining 9 42 21 4 non FIP samples were opportunistically sourced from submissions to a diagnostic laboratory 56 to 66 These were deemed to represent non FIP cats based on the M A N U S C R I P T A C C E P T E D ACCEPTED MANUSCRIPT 8 stated age of animals and the type of tests requested but detailed clinical histories were not provided 210 With the exception of one case 18 a diagnosis of FIP was confirmed histologically The 18 cat was considered FIP affected based on clinical history alone as tissues were not available for the post mortem examination Finally archival blood samples from three FIP affected colony cats 70 71 72 were collected while the cats appeared clinically normal two to four months before they were euthanised due to FIP 215 The range of ages for FIP 0 3 to 12 years seropositive non FIP 0 2 to 18 years and seronegative non FIP 0 2 to 17 years cats were similar However the median age of FIP cats 3 0 years was lower than an average age of seropositive non FIP cats 4 3 years p 0 047 Figure 1 The distribution of sexes within FIP and non FIP groups was also similar with 54 8 and 52 5 of males including both entire and castrated males in each group respectively 220 Hybridisation to peptide chips The results of the testing are visually presented in Figures 2 and 3 There was minimal binding of the negative control serum to Chip1 average fluorescence per peptide 8 units range 0 616 with clear binding detected to Chip 2 average fluorescence per peptide 487 units range 0 5 424 and Chip 3 average fluorescence per peptide 2 562 units range 0 21 745 Sera from FIP cats Chip 3 225 appeared to recognise more antigens with stronger binding to selected peptides than sera from healthy cats Chip 2 One region spanning about 34 amino acids within nsp3 aa 1017 to 1051 in reference sequence AAY16374 showed a comparatively low level of binding to sera from seropositive cats The average fluorescence per peptide n 228 in that region was 415 45 and 11 for chips 1 2 and 3 230 respectively Peptide ELISAs All 11 peptides tested showed some level of binding to FCoV antibody positive sera and no binding to the control FCoV antibody negative serum in ELISA However there was considerable variability between cats Figure 4 None of the peptides tested reacted exclusively with sera from FIP or 235 serologically positive non FIP cats Out of five peptides selected for further testing three 25438 26242 28423 produced inconsistent binding patterns and were discarded The ELISA