Internet & Technology

Bovine Spongiform Encephalopathy: Investigation of Phenotypic Variation among Passive Surveillance Cases

Description
Bovine Spongiform Encephalopathy: Investigation of Phenotypic Variation among Passive Surveillance Cases
Published
of 12
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
Share
Transcript
  Bovine Spongiform Encephalopathy: Investigationof Phenotypic Variation among PassiveSurveillance Cases M. J. Stack * , S. J. Moore * , A. Davis † , P. R. Webb * , J. M. Bradshaw  ‡ ,Y.H.Lee x ,M.Chaplin * ,R.Focosi-Snyman * ,L.Thurston * ,Y.I.Spencer * ,S. A. C. Hawkins * , M. E. Arnold * , M. M. Simmons * and G. A. H. Wells * *Veterinary Laboratories Agency, Woodham Lane, Addlestone, Surrey KT15 3NB, UK,  † Dorevitch and Gippsland Vetnostics, 267 Moreland Road, Coburg, Victoria 3058, Australia,  ‡ Veterinary Laboratories Agency, Langford House,Langford, Bristol BS40 5DX, UK and   x Foreign Animal Disease Division, National Veterinary Research and Quarantine Service, Republic of Korea Summary Bovine spongiform encephalopathy (BSE) is a prion disease of domesticated cattle, first identified in GreatBritain (GB) in 1986. The disease has been characterized by histopathological, immunohistochemical, bio-chemical and biological properties, which have shown a consistent disease phenotype among cases obtainedby passive surveillance. With the advent of active surveillance in 2001, immunological tests for detection of the prion protein revealed some cases with different biochemical characteristics and, in certain instances, dif-ferences in pathology that have indicated variant phenotypes and the possibility of agent strain variation. Thisstudy examines a case set of 523 bovine brains derived from archived material identified through passive sur-veillance in GB. All cases conformed to the phenotype of classical BSE (BSE-C) by histopathological, immu-nohistochemical and biochemical approaches. The analyses consolidated an understanding of BSE-C and, bywestern blotting, confirmed differentiation from the known atypical BSE cases which exhibit higher or lowermolecular masses than BSE-C (BSE-H and BSE-L respectively).Crown Copyright  2010 Published by Elsevier Ltd. All rights reserved. Keywords:  bovine spongiform encephalopathy; immunohistochemistry; phenotype variation; western blotting Introduction Transmissible spongiform encephalopathies (TSEs),or prion diseases, are fatal neurodegenerative disor-ders affecting a wide range of mammalian hosts andinclude kuru and Creutzfeldt e  Jakob disease (CJD)in man, scrapie in sheep and goats and bovine spongi-form encephalopathy (BSE) in cattle (H € ornlimann et al. , 2006). BSE was first described in 1986 in GreatBritain (GB) (Wells  et al. , 1987), emerged as a feed-borne epidemic and, a decade later, was causallylinkedtotheoccurrenceofanewformofhumanpriondisease, new variant CJD (vCJD) (Collinge  et al. ,1996; Lasm  ezas  et al. , 1996; Will  et al. , 1996; Bruce et al. , 1997; Hill  et al. , 1997). Due to the novel anduncertain nature of the causal agent of TSEs, thecharacterization of these disorders has been basedon phenotypic features.In the early years of the BSE epidemic, the casedefinition of BSE was reliant on histopathologicalcharacterization and diagnosis was based principallyon detection of vacuolation in the brain stem of clini-cally suspect cattle (Wells  et al. , 1989). This was sup-plemented with the examination of brain tissue byelectron microscopy for disease-associated scrapie-as-sociated fibrils (SAFs) comprised of a disease-associ-ated protein, characterized as PrP or prion protein(Hope  et al. , 1988; Scott and Stack 1994; Stack  et al. ,1996). Biological characterization of a small numberof individual isolates by transmission from affectedbrain to mice indicated that the cases wereattributable to a single strain of agent that differed Correspondence to: M. J. Stack (e-mail: m.j.stack@vla.defra.gsi.gov.uk).0021-9975/$ - see front matter  Crown Copyright  2010 Published by Elsevier Ltd. All rights reserved .doi:10.1016/j.jcpa.2010.10.007 J. Comp. Path. 2011, Vol. 144, 277 e 288  Available online at www.sciencedirect.com www.elsevier.com/locate/jcpa  fromcharacterizedstrainsofthescrapieagentinsheep(Bruce  et al. , 2002; Green  et al. , 2005). Uniformity of the pathological changes in affected cattle in GBsupported evidence for a consistent diseasephenotype for BSE (Simmons  et al. , 1996). This hasalso been the case when the pathology of BSE casesin GB was compared with the majority of epidemicBSE cases in other European countries (Orge  et al. ,2000; De Beer  et al. , 2003; Casalone  et al. , 2006).With increasing recognition of the importance of a host-encoded cellular-membrane protein (PrP c ) inthe pathogenesis of TSE (Prusiner  et al. , 1984), bio-chemicaldemonstrationandanalysisofthepathogno-monic, conformationally altered form of the protein,srcinally designated PrP scrapie (PrP Sc ), in the dis-eased host (Prusiner, 1997) has become an essentialcomponentof the phenotypicdefinitionand diagnosisof these disorders ( Jeffrey  et al. , 2001; Gavier-Wid  en et al. , 2005). In diagnostic approaches where thedetection of PrP Sc involves protease digestion, suchas western blotting (WB), the partially protease-resistant and hydrolysis-resistant core of the proteinis termed PrP res (Stack  et al. , 2002), whereas whenprocessing of tissue samples does not involve proteasedigestion,asinimmunohistochemistry(IHC),thedis-ease-associated PrP detected is termed (PrP d )(Gonz  alez  et al. , 2002).Pathological phenotype characterization is carriedout principally in two ways: by WB of tissue extractswhere blots show characteristic patterns of PrP res with variable molecular masses (Stack  et al. , 2002; Jacobs  et al. , 2007), or by immunohistochemicaldetection of PrP d in formalin-fixed tissue (Miller et al. , 1993, van Keulen  et al. , 1995) and examinationof the configurations and distribution patterns of immunolabelling ( Jeffrey  et al. , 2001, Gonz  alez  et al. ,2003). The detection of PrP res and PrP d forms thebasis of current case definitions for diagnosis of TSEin many countries and WB (Arsac  et al. , 2007) andIHC (Gavier-Wid  en  et al. , 2005) techniques havebeen incorporated into statutory testing for BSE inboth passive and active surveillance.Since the introduction of active surveillance forBSE,therehavebeenreportsofBSEincattlewithvar-iant molecular and biological characteristics. Thesecases have been from several European countries(Biacabe  et al. , 2004; Casalone  et al. , 2004; DeBosschere  et al. , 2004; Polak  et al. , 2004; B  eringue et al. , 2006; Buschmann  et al. , 2006; Baron  et al. ,2007; Terry  et al. , 2007; Biacabe  et al. , 2008; Gavier-Wid  en  et al. , 2008; Stack  et al. , 2009a, b; Tester  et al. ,2009), Japan (Yamakawa  et al. , 2003; Masujin  et al. ,2008), Canada (Clawson  et al. , 2008) and the USA(Richt  et al. , 2007). Additionally, in some reports, his-topathological features (Casalone  et al. , 2004,Lombardi  et al. , 2008, Gavier-Wid  en  et al. , 2008)have also been described. These atypical cases fallinto two groups based on banding patterns of PrP res on WB analysis compared with that of classical BSE(BSE-C). The PrP res for BSE is, depending on thebinding of the antibody used, normally visualized asa triple banding pattern representing three regions of the protein: a diglycosylated region (upper band),a monoglycosylated region (middle band) and anunglycosylated region (lowest band). Relative tothe measured molecular mass (MM) of theunglycosylated PrP res band of BSE-C, one group isdefined as higher MM (BSE-H) and the other aslower MM (BSE-L) ( Jacobs  et al. , 2007). This varia-tion has raised the issues of possible agent variationin prion disease in cattle and occurrence of spontane-ous BSE cases (Baron  et al. , 2006; Buschmann  et al. ,2006; Capobianco  et al. , 2007; B  eringue  et al. , 2008).While no atypical BSE has been detected amongcases identified by passive surveillance, theproportion of such cases that have been fullycharacterized by WB and IHC is relatively small,mainly because these techniques were not routinelyapplied through much of the course of the epidemic.For example, in GB, at the peak of the epidemic,neither IHC nor WB for detection of PrP d andPrP res , respectively, were applied routinely resultingindiagnosesbeingbasedonarestrictedcasedefinition.The purpose of this study was to investigate an ar-chived set of BSE cases resulting from passive surveil-lance (clinical suspects) in GB using current WB andIHC techniques to extend the characterization of thecases and detect possible variations of phenotypic sig-nificance. Materials and Methods Tissue and Case Selection TheBSEbrainsstoredattheVeterinaryLaboratoriesAgency (VLA) were reviewed by a paper and data-base search to identify samples potentially suitableforinclusioninthestudy.Caseswereidentified,wherepossible, from which both fresh/frozen samples andformalin-fixed samples were available from the sameanimals(Table1).Forfixedsamples,blocksrepresen-tative of all major brain regions were retrieved asstoredwaxblocksorpreparedfrombraintissuestoredin 10% neutral buffered formalin for 1 e 13 years.Fixed tissue was processed routinely and embeddedin paraffin wax. Sections (5  m m) were stained withhaematoxylin and eosin (HE) or subject to IHC forthe detection of PrP d . Frozen material collected forbiochemical techniques was sampled  post mortem from the medulla oblongata region of the brain stem. 278  M.J. Stack  et al  .  While it was not anticipated from sample size orprevious diagnostic criteria that this case set wouldcontain the currently recognized phenotypic variantsof BSE ( Jacobs  et al. , 2007), cases were selected andgrouped by age as the presence of certain phenotypesamong cattle has been shown to vary with age, possi-blyreflectingdifferencesinincubationperiod.Groups(Table 1) comprised cases 4 e 6 years old (group 1),which are representative of the age of the majorityof BSE cases during the epidemic in GB, cases over6 years old (group 2) and cases under 4 years old(group 3). An additional group of older cattle, over6 years old, (group 4) comprised animals that showedBSE-like clinical signs, had statutory diagnosticresults, which were negative, and where histologicalexamination did not reveal evidence of other neuro-degenerative diseases. The available fresh/frozentissue for WB and formalin-fixed tissue for HE andIHC examinations, grouped according to age ranges,birth years and disease status for each group, areshown in Table 1. Western Immunoblotting and Analyses The BioRad TeSeE WB method (Arsac  et al. , 2007),acquired in kit form (BioRad Laboratories, Marnes-La-Coquette, France), was conducted on thawedsamples of frozen brain stem (medulla) as describedpreviously( Jacobs et al. , 2007) using avarietyofdiffer-entmonoclonalantibodies(mAbs).ThemAbsusedforlabelling PrP res and their reactivity with differentaminoacid(aa)sequencesofthePrPmolecule(accord-ingtothebovinesequencewithsixoctapeptiderepeats)were: mAb P4 (aa 97 e 112; R-BioPharm, Germany);Sha31 (aa 156 e 163; BioRad); SAF84 (aa 173 e 178;agiftfromDr.J.Grassi,CEA,Saclay,France).Thean-tibodies Sha31 and SAF84 have been shown to be use-ful in the Biorad TeSeE WB for further analysis of BSE-H cases ( Jacobs  et al. , 2007). All samples weretested using all three mAbs. Molecular mass markersused throughout the testing (SM) were obtained fromSigma, Poole, Dorset, UK (catalogue No: B2787).The methodologies for immunolabelling, quantifi-cation and analyses of results for WB have beenpublished previously (Stack  et al. , 2006). Briefly, theimmunolabelling was visualized by means of an en-hanced chemiluminescence system (CPD-StarTropix). Signals were quantified using Fluor SMultimager computer analysis (Quantity One,software, BioRad UK Ltd.).ControlmaterialexaminedbytheBioRadTeSeEWBwith the three mAbs (Sha31, SAF84 and P4) includedsheep scrapie and BSE-C samples from GB, an BSE-Htype isolate, the latest of the three detected so far in GB(Stack  et al. , 2009a) and a BSE-L-positive sample fromFrance (kindly supplied by Dr T. Baron, AFSSA,Lyon, France). Interpretation of results for known vari-antswasbasedontheestablishedmolecularprofilespub-lished for BSE-H and BSE-L ( Jacobs  et al. , 2007). Microscopical Findings Lesion profiling . Vacuolar lesion profiles were obtainedfrom HE sections of 310 BSE cases (sample groups1 e 3)usinga methodthat assesses semi-quantitativelythe severity of vacuolation in defined neuroanatomi-cal locations and presents the data as a mean scoreper location for each group of animals. The profileswerecomparedwiththeprofileobtainedfromaprevi-ous series of BSEcases ( n ¼ 100), obtained in the earlypart of the GB epidemic (Wells  et al. , 1992, Simmons et al. , 1996). IHC  . The IHC protocol for the statutory diagnosis of BSE in GB employing the rat anti-PrP R145 mAbwasapplied(Arnold et al. ,2007).MAbR145waspro-duced by VLA, Weybridge, with a synthetic peptidecorrespondingtothesequenceYQRESQAYYQRGA(221 e 233) of bovine PrP (Terry  et al. , 2003).For each case, histological sections at six coronalbrain levels were immunolabelled. At each level,8 e 17 neuroanatomical areas were examined (Table2), giving a total of 68 areas examined per case. Areaswere selected for examination on the basis that theywere readily identifiable and represented in sectionsin the majority of cases. Microscopy was undertakenby five observers. For each area the morphological Table 1Summary of cases and samples Group 1 * Group 2  * Group 3  * Group 4 † Age (months) 42 e 83 84 e 205 25 e 41 120 e 208Birth year 1987 e 1995 1984 e 1997 1988 e 1995 1988 e 1995Fixed and frozen 47 42 2 34Fixed only 170 18 48 20Frozen only 158 6 32 16Total cases 375 66 82 70 * Confirmed clinical cases of BSE. † Clinically suspected BSE: unconfirmed on statutory diagnosis andwith no histological evidence of other neurodegenerative diseases. Table 2NumbersofanatomicalareasexaminedbyIHCforeachlevel of brain Brain level Number of areas examined  Medulla oblongata (obex) 11Medulla oblongata (rostral medulla) 14Cerebellum 9Mesencephalon (rostral midbrain) 9Thalamus 17Frontal cortex and basal nuclei 8 Passive Surveillance of BSE  279  PrP d immunolabelling types present were recordedusing a separate recording sheet for each level of brain. Diagrammatic representations of each brainregion and standardized descriptive accounts of thepatterns of immunolabelling were also compiled, giv-ing the occurrence and relative prominence (withoutquantification) of labelling forms for each neuroana-tomical area, together with any obvious sublocaliza-tion of immunolabelling within a neuroanatomicalarea.Nocomparisonsweremadebetweenbrainlevels(Supplementary data Figs. S1 e S6).Eight labelling types consistent with previous de-scriptions were recognized: particulate (PT), granu-lar cytoplasmic (GC), perineuronal (PN),intramicroglial (IG), linear (L), stellate (ST), pla-que-like (PL) and aggregates (A) (Wells andWilesmith 1995; Orge  et al. , 2000; Debeer  et al. ,2003; Casalone  et al. , 2006). The presence orabsence of these labelling types was recorded andentered into the database in binary form (0, notpresent; 1, present). In addition, if an area was notrepresented in the section this was also recorded (1,not present) giving a nine-digit binary number foreach area. Thus, an area with PT, GC and STlabelling would be recorded as: 110001000 andwhere an area was missing the entry would be:000000001. Assessment of intensity of labelling wasnot considered a feasible observation in this studybecause of the wide variation in tissue storageconditions, the multiplicity of immunolabelling runsrequired to prepare the large numbers of sectionsand the difficulties of standardizing subjective semi-quantitative assessments.In those cases that were reactive with the mAb P4antibody in WB and for which appropriate sampleswere available, additional IHC was carried out. Sec-tions of rostral medulla were immunolabelled withmAbP4,whichisraisedtoanN-terminalaasequence(97 e 112) of the PrP protein (Harmeyer  et al. , 1998)and substituted for mAb R145 in the IHC protocolabove. For each of the 16 cases reactive with mAbP4 on WB, a medulla section of an age-matchedWB mAb P4-negative case was included. IHC withmAb P4 was not extended to brain levels other thanmedulla, since previous work has shown no variationin WB results among different brain regions of cattlewith BSE (Tester  et al. , 2009). Slides were examinedas described above. IHC data analysis . Data were exported to MicrosoftEx-cel for analysis. For each neuroanatomical area, thepresence/absence of each labelling type wasrecorded, the frequency with which it occurred ineach group was calculated and the frequencieswere graphed to allow comparisons betweengroups. Differences between groups were revealed bychi-squaredanalysisandfurtherinvestigatedusinglo-gisticalregression,wherethepresence/absenceofeachlabellingtypewas thedependent variableand the ageof the animal (plus a constant term) was the indepen-dent variable. This enabled comparisons of associa-tions between frequency of labelling type and age.ModelfitwasassessedbyaHosmer e Lemeshowgood-ness-of-fittest,whichtestedwhetheralogisticalregres-sion model was appropriate for the data (Hosmer andLemeshow, 1989).Slides immunolabelled with the mAb P4 weregrouped according to the PrP d immunolabellingtype(s) where present. Chi-squared analysis wasused to test for any correlation between mAb P4WB results and mAb P4 IHC results. Results Western Immunoblotting UsingtheBioRadTeSeEWBtest andthe threemAbs(Sha31, P4 and SAF84), molecular profiles of thethree highest MM protein bands were indistinguish-able from that of the BSE-C control sample andwere obtained for the majority of cases in each of the BSE-positive groups. However, it was clear thatwhensamples werestronglyreactive,twoextra,lowerMM protein bands could be detected with mAbsSha31 and SAF84 (Fig. 1).Although none of the retrospective samples testedshowed BSE-H or BSE-L molecular profiles, a sub-group of samples, which were identical to BSE-C intheir reaction to mAb Sha31, were also reactivewith the mAb P4 antibody. A total of 66 cases outof 287 (23%) were reactive to mAb P4 and were de-tected in all positive sample groups: 55 (27%) fromgroup 1 (4 e 6-year-old cases), eight (17%) fromgroup 2 (animals > 6 years old) and three (9%)from group 3 (animals < 4 years old). Examples areshown in Fig. 2. Lesion Profiling The vacuolation profiles for the three BSE confirmedgroups (1 e 3) were closely similar to that ofa previousseries of 100 BSE cases (Simmons  et al. , 1996), asshown in Fig. 3. Immunohistochemistry The descriptions of the patterns of immunolabellingforeachneuroanatomicalnucleusbyregionsuggestednoobviousdeviationinpatternofPrP d immunolabel-ling within each of the coronal brain regions (Figs.S1 e S6 in the supplementary information). 280  M.J. Stack  et al  .  Comparisons between the statistical analysis andthe descriptive findings were used to test mutuallythe credence of the findings from the different ap-proaches. All brains of the group 4 animals were ex-amined systematically as for other groups, butdisease-specific forms of immunohistochemical label-ling were not detected.Of the 32 medulla sections immunolabelled withthemAbP4,disease-specificreactivity,correspondingto the PT and A types seen with the R145 antibody,was observed in two cases. In both cases PT labellingwas present in the nuclei vestibularis, nucleus tractusspinalis nervi trigemini, formatio reticularis and nu-cleus raphe magnus, and A labelling was seen in theformatio reticularis and nucleus raphe magnus. Oneof these cases had a negative WB mAb P4 result andthe other case was a WB mAb P4 reactor. This is con-sistent with additional IHC studies of 24 randomlyselected confirmed cases of BSE in GB in whichmAb P4 detected PT immunolabelling in all diagnos-tic target areas of the medulla of all cases except one(P. R. Webb, unpublished data). The labelling wasnot obtained in sections from tissue treated with for-mic acid, suggesting that it is markedly protocol-de-pendent. Immunohistochemistry Data Analysis Usingthebinarynumbersystemtoassessfrequencyof labelling, the results were displayed graphically andfor most of the areas were similar for all groups. Ex-ample graphs for neuroanatomical areas used forthe statutory diagnosis of BSE are shown in Fig. 4.Graphs for the remaining 15 neuroanatomical areasalso examined for vacuolation lesion profiling areshown in the supplementary information (Fig. S7).Chi-square analysis indicated significant differencesbetween groups for some immunolabelling types insome areas. The Hosmer e Lemeshow goodness-of-fittest showed that the logistic regression analysis wasappropriateforthemajorityoflabellingtypes.Thelo-gistic regression revealed that older animals tended tohave less IG and L labelling types overall. Positive(more frequent in older animals) or negative (less Fig. 1. BSE-C PrP res molecular profiles from a selection of samples from group 1 (from cattle 42 e 83 months of age) showing a represen-tation of the majority of profiles found in the study. Bovine +ve and Ovine +ve represent the BSE- and scrapie-positive controls,respectively. Sigma molecular mass markers (SM). With mAbs Sha31 and SAF84 two lower bands are detected, which appear tovary in intensity depending on the concentration of the PrP res in individual samples. The BSE-C cases are not detected with theN-terminal mAb P4, whereas the ovine scrapie control is detected by this antibody. Passive Surveillance of BSE  281
Search
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks
SAVE OUR EARTH

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...

Sign Now!

We are very appreciated for your Prompt Action!

x