Immunohistochemical Study of Rabies Virus Within the Central Nervous

Diagnostic Pathology (Domestic Mammals)

Veterinary Pathology 47(4) 630-636 ª The American College of Immunohistochemical Study of Rabies Veterinary Pathologists 2010 Reprints and permission: Virus Within the Central Nervous sagepub.com/journalsPermissions.nav DOI: 10.1177/0300985810370013 System of Domestic and Wildlife Species http://vet.sagepub.com

L. T. Stein1, R. R. Rech1, L. Harrison1, and C. C. Brown1

Abstract Immunohistochemistry using a commercial polyclonal antibody for lyssavirus was applied to 39 archival cases of rabies. Paraffin blocks from 13 different species were available, including 3dogs,4cats,1pig,6cattle,4horses,1llama,7skunks(Mephitis mephitis), 7 raccoons (Procyon lotor), 1 bat (Myotis species), 1 white-tailed deer (Odocoileus virginianus), 1 bobcat (Lynx rufus), 2grayfoxes(Urocyon cinereoargenteus),and1redfox(Vulpes vulpes). All cases had previously been diagnosed as rabies using histopathology and/or fluorescent antibody testing. The immunohistochemistry technique successfully detected lyssavirus antigen in all cases. In species for which 3 or more samples were available, distributional trends were seen in 4 main brain regions: brainstem, cerebellum, hippocampus, and cerebrum. The best site for rabies virus detection in dogs and cats was the hippocampus. For cattle, viral antigen was most prominent in the brainstem, followed by the cerebellum. In horses, the cervical spinal cord and adjacent brainstem were the optimal sites for detecting rabies virus antigen. In raccoons and skunks, positive labeling was widely dispersed, so selection might be less important for these wildlife reservoir species. Immunohistochemistry should prove useful in enhancing the accuracy of rabies diagnosis through informed selection of brain sampling sites when composite sampling is not feasible. This immunohistochemical technique could provide reliable virus detection in formalin-fixed tissues in any potentially infected species.

Keywords central nervous system, immunohistochemistry, lyssavirus, rabies, viral encephalitis

Rabies is a neurologic disease characterized by acute, presence of Negri bodies has been considered pathognomonic progressive, fatal encephalitis.1,3,19,20,26 It is probably the for rabies, they are absent in 20% to 60% of rabies oldest known zoonosis, recognized for more than 5,000 cases.1,10,15,16,20,24,26 Because the encephalitis of rabies can years.9,19,20 The main viral reservoirs include animals of the be challenging to distinguish from other viral encephalitides, orders Carnivora and Chiroptera,8 but all mammalian species the diagnosis of rabies should include ancillary tests such as are susceptible at any time, and the disease is found on every direct fluorescent antibody test (FAT) and intracerebral inocu- continent except Antarctica.2,13,19,25 Rabies virus, a lyssavirus lation in suckling mice, which are the standard tests for rabies of the family Rhabdoviridae, is a neurotropic virus. Inocula- today.10,16,18,20,24,27 Despite its efficiency in quickly detecting tion is generally through a carnivore or bat bite; the virions rabies, FAT has many drawbacks that can limit its usefulness, travel retrograde along axons and across synapses to the cen- including equipment cost, potential exposure of laboratory staff tral nervous system (CNS). Viral replication and dissemina- to live virus, fading nature of fluorochrome dyes, and, in the tion occur widely throughout the CNS before the virus developing world, necessity of getting fresh specimens to spreads centrifugally to the salivary glands.3,6,13,22,25,26 It is potentially distant laboratories in adequate condition for test- not possible to detect rabies infection during the incubation ing.1,7,10 An estimated 55,000 humans die annually of rabies, period, which is generally 1 to 8 weeks.6,13,20,22,26 Therefore, with almost all of these in the developing world (http:// the diagnosis of rabies can only be achieved with 100% cer- tainty through postmortem examination of the central nervous 6,19,20 system or trigeminal ganglia. It is essential that rabies 1 Department of Pathology, College of Veterinary Medicine, University of diagnosis in animals be rapid and accurate to facilitate postex- Georgia, Athens, GA, USA posure treatment decisions for exposed humans.27 Corresponding Author: Histologically, rabies is characterized by a viral encephalitis Corrie C. Brown, Department of Pathology, College of Veterinary Medicine, with intracytoplasmic Negri bodies in neurons, but these University of Georgia, 501 D.W. Brooks Drive, Athens, GA 30602-7388, USA 13,18,20 changes can be mild or even absent. Although the Email: [email protected]

630 Stein et al 631

Table 1. Intensity of Rabies Antigen Immunoreactivity in Major Brain Areas of 13 Mammalian Speciesa

Brain Region

Case No. Species Brainstem Cerebellum Hippocampus Cerebrum

1 Raccoon 1 þþ þ þþþ þþþ 2 Raccoon 2 þþ NAb þþþ þþþ 3 Raccoon 3 þþ þ þþþ þþþ 4 Raccoon 4 þþ þþ þþþ þþþ 5 Raccoon 5 NA NA þþþ þþþ 6 Raccoon 6 þþ þþ þþþ þþþ 7 Raccoon 7 þþ þ þþþ þþ 8 Skunk 1 þþ NA þþþ þþ 9 Skunk 2 þ þ þþ þþ 10 Skunk 3 þþ þþ þ þþ 11 Skunk 4 þþþ þþ þþþ þþ 12 Skunk 5 þþ þþ þþ þþ 13 Skunk 6 NA þ NA þþ 14 Skunk 7 þþþ þþþ þþþ þþ 15 Bovine 1 þþþ NA þ þ 16 Bovine 2 þþþ NA – – 17 Bovine 3 þþþ þþ – þþ 18 Bovine 4 þþ þþþ – þþ 19 Bovine 5 þþ þ – þ 20 Bovine 6 þþþ þþþ þþþ þþþ 21 Cat 1 þþ þ þþþ þþþ 22 Cat 2 þþ þþ þþþ þþþ 23 Cat 3 þþ þþ þþþ þþþ 24 Cat 4 þ þ þþ þ 25 Dog 1 þþ þþ þþþ þþ 26 Dog 2 þ – þ – 27 Dog 3 þþ þ þþ þþ 28 Horsec 1 þþþ þ þþ þ 29 Horse 2 þþ+þ þþ NA þþ 30 Horse 3 þþ þ – þ 31 Horse 4 þþ – – þ 32 Swine þþ þ þ þ 33 Llama þþþ þþþ þþþ þþ 34 Red fox þ –NAþþ 35 Gray fox 1 þþ – þ þþ 36 Gray fox 2 þ þ þ þ 37 White-tailed deer þþ NA NA þþ 38 Bobcat þþþ þþþ þþþ þþþ 39 Bat þþþ þþ þ þ a Immunoreactivity was scored on the following scale: absent (–), weak (þ), moderate (þþ), strong (þþþ). NA, not available. b Segment of brain was not available. c The cervical spinal cord was also analyzed for each horse; signal intensities for case Nos. 28–31 were þþþ, þþþ, þþ, and þþþ, respectively. www.who.int/rabies/en/). Improved diagnostic testing for this Because rabies does not always infect all regions of the at-risk population is essential. Over the last 2 decades, immu- brain equally and viral dissemination varies among species, the nohistochemistry (IHC) has become increasingly popular for erratic distribution of viral antigen can compromise the relia- detecting numerous antigens in fixed tissues.7,10,20,26 Studies bility of test results if all parts of the brain are not sampled have shown that the sensitivity of IHC for rabies is equal to that or if a particular sample has low levels of viral antigen. The of FAT.16,27 IHC may be even more sensitive in early diagnosis collection of optimal samples is therefore essential for optimiz- of suspected cases when traditional histological and FAT tech- ing diagnosis.2 The main objectives of this study were to niques could not detect viral antigens or lesions;1,11,26 however, develop an IHC protocol with a polyclonal antibody suitable the disadvantage of IHC, so far, is that most publications were for rabies diagnosis in various mammalian species, examine based on the use of in-house antibodies, mostly monoclonal, the reliability of that technique, and develop a preliminary and therefore with specificity for the regional strains of the assessment concerning the regions of the brain most useful for virus.1,11,16,26,27 diagnosis of rabies in various species.

631 632 Veterinary Pathology 47(4)

Methods to diffuse granularity throughout the perikaryon of neurons and Cases their axonal and dendritic processes. In some cases, inclusions appeared as brown oval homogeneous structures within the All formalin-fixed, paraffin-embedded (FFPE) samples were perikaryon of neurons (Fig. 1). The presence of viral antigen obtained from the archives of the Department of Pathology, in the neuronal processes gave a granular appearance to the Athens Diagnostic Laboratory, and Southeastern Cooperative neuropil, with variable extension into axons (Fig. 2). Wildlife Disease Study at the College of Veterinary Medicine, Some distributional differences of rabies antigen for each University of Georgia, Athens. These cases had been previ- species were highlighted (Table 1). For the domestic carnivores ously diagnosed as rabies by FAT and/or histopathological (cats and dogs, case Nos. 21–27), results were very similar, examination. Immunohistochemistry was performed on a total with most intense signal in the hippocampus (Fig. 3), followed of 39 cases, encompassing 13 different mammalian species, by the cerebrum, brainstem, and cerebellum in descending and included 7 skunks (Mephitis mephitis), 7 raccoons (Pro- order of immunoreactivity. cyon lotor), 6 cattle, 4 cats, 3 dogs, 4 horses, 1 pig, 1 llama, In cattle (case Nos. 15-20), the most intense signal was in 2 gray foxes (Urocyon cinereoargenteus), 1 red fox (Vulpes the perikaryon of large neurons of brainstem nuclei. In most vulpes), 1 bat (Myotis species), 1 white-tailed deer (Odocoileus cases, cytoplasmic reactivity was intense and diffuse. The cer- virginianus), and 1 bobcat (Lynx rufus). Pertinent data are in ebellum had the next most intense signal; Negri bodies in the Table 1. Purkinje cells appeared as strongly labeled, oval, homogeneous structures (Fig. 1). Throughout the molecular layer of the cer- Immunohistochemistry ebellum, dendritic processes had strong signal. There was pat- chy cytoplasmic labeling of neurons of the granular cell layer, All tissues were examined by immunohistochemistry using the including Golgi neurons. following protocol to detect lyssaviral ribonucleoprotein. After For horses (case Nos. 28-31), the cervical spinal cord had deparaffinization in CitriSolv, tissue sections were quenched the strongest signal, followed by the brainstem. In all 4 horses, with 3% hydrogen peroxide and rinsed in water. The antigenic the distribution of the viral antigen was diffuse throughout the sites were exposed by microwaving in a citrate buffer solution neuraxis (Fig. 4), with sparing or even absence in sections of (1X Vector Ag-Retrieval Unmasking Solution) for 10 minutes the hippocampus, cerebellum and cerebrum. In general, antigen 0 0 0 (4 -4 -3 intervals), followed by blocking in 1:10 diluted Power- granularity in the brain of horses, especially in the neuronal block solution (Biogenex Universal Blocking Reagent) for perikaryon, was less prominent than in the other species. 7 minutes. Primary antibody (rabies polyclonal DFA reagent, IHC was only applied to 1 pig (case No. 32), which had made in goat, Chemicon/Millipore, Billerica, MA) at a moderate positivity in the brainstem, with milder reactions in 1:2,000 dilution was incubated for 1 hour at 37 C followed the other segments. The 1 llama studied (case No. 33) had by two 20-minute room temperature incubations: biotinylated intense signal in all segments. link antibody (LSAB2 kit) and horseradish peroxidase–conju- In raccoons (P. lotor) (case Nos. 1-7) and skunks (M. mephi- gated streptavidin (LSAB2 kit). The reaction was revealed in tis) (case Nos. 8-14), IHC reactivity varied between moderate diaminobenzidine peroxidase substrate (Dako, Carpenteria, and strong in all examined segments (Fig. 5); however, in all CA) for no longer than 5 minutes. The slides were counter- cases, the hippocampus was the site where the signal was con- stained with Mayer’s hematoxylin and coverslipped with Per- sistently very strong. In these 2 wildlife species, IHC of any mount. Positive controls were a rabies FAT-positive brain. brain segment yielded positive results. Negative controls included substituting primary antibody with Both red fox (V. vulpes) (case No. 34) and gray foxes PBS. (U. cinereoargenteus) (case Nos. 35 and 36) had the strongest immunoreactivity in the cerebral cortex, followed by the brain- Viral Antigen Distribution stem (Fig. 2). In the white-tailed deer (O. virginianus, case No. 37), bobcat (L. rufus, case No. 38), and bat (Myotis species, For each species, 4 anatomic locations in the brain were exam- case No. 39), positive signal was detected in the available ined: brainstem, cerebellum, cerebrum, and hippocampus. All regions of the brain (Fig. 6). 4 equine cases also included cranial cervical spinal cord. The IHC reactivity was scored as follows: absent (–); weak (þ), mean of 1 focus of positive signal/high-power field (hpf); mod- Discussion erate (þþ), mean, 2 to 5 foci of positive signal/hpf; and strong Immunohistochemistry demonstrated rabies viral antigen in all (þþþ), mean, more than 5 foci of positive signal/hpf. 39 cases examined. Because clinical manifestations of rabies in many species are unpredictable, the optimal rabies diagnostic test must detect viral antigen in any mammalian species. Some Results IHC studies for rabies only use monoclonal antibody prepara- Immunohistochemistry results are in Table 1. Rabies virus anti- tions, which may not be suitable, especially if their activity gen was detected in all 39 cases. The signal was observed depends on conformational epitopes.19 The polyclonal anti- almost exclusively in gray matter and characterized by sparse body used in this study accurately detected rabies virus in all

632 Stein et al 633

Figure 1. Cerebellum; bovine, case No. 18. Several brown, circular, strongly positive inclusion bodies for rabies viral antigen are within the cytoplasm of Purkinje cells. The dendrites of many Purkinje cells are also strongly positive. The granular cell layer and the basket cells of the molecular layer have widespread fine staining. Immunohistochemistry (IHC), labeled streptavidin biotin (LSAB) method with diaminobenzidine (DAB) substrate, Mayer’s hematoxylin counterstain. Figure 2. Cerebral cortex; red fox (V. vulpes), case No. 34. Labeling of rabies virus antigen within the cytoplasm and processes (axons and dendrites) of neurons. IHC, LSAB method with DAB substrate, Mayer’s hematoxylin counterstain. Figure 3. Hippocampus; dog, case No. 27. Distinct labeling for rabies virus antigen within the perikaryon of several neurons.

633 634 Veterinary Pathology 47(4)

13 tested species. Immunohistochemistry for rabies viral of Negri bodies, however, which are found only in 40% to antigen detection in FFPE sections has been successfully 80% of rabies cases.7 Along with such variable and inconsistent applied in several countries, although most of the antibodies are Negri body formation, some hematoxylin and eosin–stained not commercially available,1,5,7,10,11,14,16,21 which makes intro- inclusions in the cytoplasm of neurons resemble Negri bodies duction of the technique to other laboratories challenging. but might not be rabies-specific. Such pseudo-Negri bodies are Although we did not evaluate the polyclonal antibody in cases protein-related inclusions and could lead to an initial false- from other parts of the world in this study, we have used it on positive diagnosis of rabies. These cytoplasmic inclusion bod- cases of rabies in cattle in Brazil with positive results. Because ies have been described in the neurons of nonrabid cats, cattle, the antibody is polyclonal, it should detect a wide range of moose, woodchucks, and skunks.13,14,22,23,24 For these reasons, rabies viral strains. rabies diagnosis should not be based solely on the presence of Although FAT is the standard technique for quick rabies Negri bodies in the hippocampus, especially in cases without diagnosis, it requires the use of expensive ultraviolet light inflammation, and IHC or FAT should be applied to rabies- microscopes11 and fresh samples, which contain live virus and suspect cases to confirm the presence of viral antigen. entail public health risk.26 Transporting fresh samples is a prob- In this study, the distribution of rabies virus antigen in the lem in countries where diagnostic laboratories are not well brain varied among species, demonstrating that sampling site established or where lack of refrigeration and high ambient in some species might be important. Numerous studies have temperatures can interfere with the FAT.1,2,11 Field collection analyzed the spread of rabies virus in the brain using sensitive of brain samples often occurs far from diagnostic laboratories. diagnostic tools such as FAT.2,3,17,24 A comprehensive study of Delays in sample collection and/or shipping in some countries multiple species using FAT concluded that when composite may add 5 or more days from the death of the animal to the sampling of the brain is not possible, the thalamus should be the point of laboratory testing; the ensuing autolysis further hinders area of choice for testing.2 However, considering results of our diagnostic accuracy.6 In the formalin-fixed specimens used in study, although supporting the collection of the whole brain, we IHC, the rabies virus is rapidly inactivated by formaldehyde, also advocate choosing specific parts for the preliminary IHC. making the transport and laboratory processing of specimens Knowledge of the optimal area of the brain for detection of much safer.10 This is even more important when the diagnosis rabies viral antigen will enhance the quality and efficiency of must be made overseas, requiring international transportation the diagnosis. of samples with zoonotic or Office International Epizooties Dogs and cats, the 2 domestic carnivores studied, had very (OIE)listed diseases. Recently, a new technique, known as similar viral antigen distribution with strongest immunoreac- direct immunohistochemical test (dRIT), has been evaluated tivity in the hippocampus. Although rabies in dogs and cats has under field and laboratory conditions to detect rabies virus anti- been greatly reduced in the United States, unvaccinated dogs gen in glycerol-preserved, field-collected brain samples.11 This and cats may contract rabies from wildlife and be the source technique uses touch-imprint brain samples that are further fixed of infection for humans.6,9 Globally, most cases of rabies in in formalin with application of a monoclonal antibody. Diagno- humans are due to exposure to infected dogs.7,19,21,23 sis using the dRIT technique is rapid, with high sensitivity and In the 5 cattle examined, the brainstem proved to be the area specificity.11 However, if the rabies test is negative with dRIT, of the brain with strongest IHC signal. Consequently, diagnos- the touch-imprint specimen is not suitable for additional diag- tic testing should target this region along with the cerebellum, nostic testing. In contrast, a major advantage of IHC is that the which also had moderate to strong immunoreactivity. The pathologic changes in the brain are clearly observable. Formalin greatest economic and public health impact of rabies in cattle fixation of brain preserves the tissue architecture and allows his- exists in Latin America.6,17 In Brazil, it is estimated that more tologic evaluation to formulate a differential diagnosis.10,19 than 842,000 cattle deaths are caused by rabies each year, cre- Certain parts of the brain have historically been considered ating abundant potential for human exposure.3,12,22 reliable for detecting Negri bodies or were routinely sent for In this study, horses had the strongest immunoreactivity in FAT to confirm the diagnosis of rabies. These parts included the cervical spinal cord, followed by the brainstem, with a con- the hippocampus in carnivores and the cerebellum in herbi- sistent positive signal present throughout the neuraxis but weak vores.2,7,10,14,18,26,27 The initial diagnostic focus on hippocam- or even absent in sections of the hippocampus, cerebellum, and pus can be attributed to its high frequency of large antigen cerebrum. In horses, the variation in distribution of rabies viral aggregates or inclusion bodies, which are easily identified by antigen, reported from several studies using Negri body map- traditional histological stains.10 The accuracy of laboratory ping or FAT,3,12,17 emphasizes the difficulty in selecting the diagnosis was inconsistent when based solely on the presence optimal sampling site for FAT diagnosis. Most studies used

Figure 3. continued. IHC, LSAB method with DAB substrate, Mayer’s hematoxylin counterstain. Figure 4. Spinal cord; horse, case No. 31. Small granules of rabies viral antigen within the perikaryon and neuronal processes within neuropil. IHC, LSAB method with DAB substrate, Mayer’s hematoxylin counterstain. Figure 5. Cerebral cortex; raccoon (P. lotor), case No. 6. Rabies virus antigen is labeled within the cytoplasm of neurons. Viral antigen also extends through neuronal processes. IHC, LSAB method with DAB substrate, Mayer’s hematoxylin counterstain. Figure 6. Hippocampus; bobcat (L. rufus), case No. 38. The cytoplasm of neurons throughout the hippocampal formation have multiple fine stippling for rabies viral antigen immunoreactivity. IHC, LSAB method with DAB substrate, Mayer’s hematoxylin counterstain.

634 Stein et al 635

FAT, which has been shown to be less sensitive in horses than 3. Carrieri ML, Peixoto ZM, Paciencia ML, Kotait I, Germano PM: in other animal species.3 In an FAT study of equine rabies, only Laboratory diagnosis of equine rabies and its implications for 29% of samples were positive in all 4 CNS regions (cortex, hip- human postexposure prophylaxis. J Virol Methods 138:1–9, 2006. pocampus, brainstem, and medulla),3 confirming this highly 4. Green SL, Smith LL, Vernau W, Beacock SM: Rabies in horses: selective and limited viral distribution. Horses usually develop 21 cases (1970–1990). J Am Vet Med Assoc 200:1133–1137, far fewer Negri bodies than any other species, sometimes in 1992. only 30% of cases,4,12,17,25 perhaps paralleling the inconsistent 5. Hicks DJ, Nunez A, Healy DM, Brookers SM, Johnson N, Fooks and limited viral distribution. Many diagnosticians proclaim AR: Comparative pathological study of the murine brain after the cerebellum as the best tissue for FAT and detection of Negri experimental infection with classical rabies virus and European bodies simply based on bovine rabies; this could lead to mis- bat lyssaviruses. J Comp Pathol 140:113–126, 2009. diagnosis in horses.3 Given variable viral distribution and spar- 6. Jackson AC, Wunner WH: Rabies, 2nd ed. Academic Press, Lon- seness of Negri bodies in equine rabies, performing don, UK, 2007. immunohistochemistry in the spinal cord and brainstem could 7. Jogai S, Radotra BD, Banerjee AK: Immunohistochemical study increase diagnostic accuracy. of human rabies. Neuropathology 20:197–203, 2000. The application of IHC in the raccoon (P. lotor) and skunk 8. Krebs JW, Williams SM, Smith JS, Rupprecht CE, Childs JE: (M. mephitis) indicated that the hippocampus is the most pro- Rabies among infrequently reported mammalian carnivores in the ductive source of viral antigen in rabies. However, in these spe- United States. 1996–2000. J Wildl Dis 39:253–261, 2003. cies, rabies antigen was abundant throughout the brain. In 2 9. Lackay SN, Kuang Y, Fu ZF. Rabies in small animals. Vet Clin domestic species (swine and llama) and some wildlife species Am Small Anim Pract 38:851–861, 2008. [gray foxes (U. cinereoargenteus), red fox (V. vulpes), bat 10. Last RD, Jardine JE, Smit MME, Van Der Lugt JJ: Application of (Myotis species), white-tailed deer (O. virginianus), and bobcat immunoperoxidase techniques to formalin-fixed brain tissue for (L. rufus)], only 1 or 2 cases of rabies from each species were the diagnosis of rabies in southern Africa. Onderstepoort J Vet available. Although the rabies antigen distribution in these Res 61:183–187, 1994. cases cannot be assessed given the limited number of cases, this 11. Lembo T, Niezgoda M, Velasco-Villa A, Cleaveland S, Ernest E, protocol proved accurate to detect rabies virus. Further IHC Rupprecht CE. Evaluation of a direct, rapid immunohistochem- studies are necessary to evaluate the best sampling site for these ical test for rabies diagnosis. Emerg Infect Dis 12:310–313, 2006. species, but the information presented here could be used as a 12. Lima EF, Riet-Correa F, Castro RS, Gomes AAB, Lima FS: Clin- starting point for further assessment. ical signs, distribution of the lesions in the central nervous system In summary, IHC for rabies detection using targeted sec- and epidemiology of rabies in northeastern Brazil. Pesq Vet Bras tions of brain could enhance accurate diagnosis in various spe- 25:250–264, 2005. cies. The public health implications of this disease warrant 13. Maxie MG, Youssef S: Nervous system. In: Jubb, Kennedy, and continued efforts to develop more accurate sampling and test- Palmer’s Pathology of Domestic Animals, ed. Maxie MG, 5th ing modalities. This IHC protocol provides an alternative to ed., vol. 1., pp. 283–457. Elsevier, Philadelphia, PA, 2007. FAT and can be used safely, even in tropical and remote areas. 14. Nietfeld JC, Rakich PM, Tyler DE, Bauer RW: Rabies-like inclusions in dogs. J Vet Diagn Invest 4:333–338, 1989. Acknowledgments 15. Nuovo GR, DeFaria DL, Chanona-Vilchi JG, Zhang Y: Molecular We thank Dr. M. K. Keel for providing the wildlife cases in this study. detection of rabies encephalitis and correlation with cytokine We also thank Jian Zhang, Kelley Gibson, and Trey Harden for valu- expression. Modern Pathol 18:62–67, 2005. able technical support. 16. Palmer DG, Ossent P, Suter MM, Ferrari E. Demonstration of rabies viral antigen in paraffin tissue sections: comparison of the Declaration of Conflicting Interests immunofluorescence technique with the unlabeled antibody The author(s) declared no conflicts of interest with respect to the enzyme method. Am J Vet Res 46:283–286, 1985. authorship and/or publication of this article. 17. Peixoto ZMP, Cunha EMS, Sacramento DRV, Souza MCAM, Silva LHQ, Germano PL, Kroeff SS, Kotait I: Rabies laboratory Funding diagnosis: peculiar features of samples from equine origin. Braz The author(s) received no financial support for the research and/or J Microbiol 31:72–75, 2000. authorship of this article. 18. Perl DP, Good PF: The pathology of rabies in the central nervous system. In: The Natural History of Rabies, ed. Baer GM, 2nd ed., References pp. 163–190. CRC Press, Boca Raton, FL, 1991. 1. Arslan A, Saglan YS, Temur A: Detection of rabies viral antigens 19. Rupprecht CE, Hanlon CA, Hemachudha T: Rabies re-examined. in non-autolysed and autolysed tissues by using an immunoperox- Lancet Infect Dis 2:327–343, 2002. idase technique. Vet Rec 155:550–552, 2004. 20. Rupprecht CE, Stohr K, Meredith C: Rabies. In: Infectious Dis- 2. Bingham J, Van Der Merwe M: Distribution of rabies antigen in eases of Wild Mammals, ed. Williams ES and Barker IK, 3rd infected brain material: determining the reliability of different ed., pp. 3–36. Blackwell, Ames, IA, 2001. regions of the brain for the rabies fluorescent antibody test. J Virol 21. Suja MS, Mahadevan A, Madhusudhana SN, Vijavasarathi SK, Methods 101:85–94, 2002. Shankar SK: Neuroanatomical mapping of rabies nucleocapsid

635 636 Veterinary Pathology 47(4)

viral antigen distribution and apoptosis in pathogenesis in street Rupprecht CE, Shoemake H, Wright C, Yager P: A comparative dog rabies—an immunohistochemical study. Clin Neuropathol study of the fluorescent antibody test for rabies diagnosis in fresh 28:113–124, 2009. and formalin-fixed brain tissue specimens. J Virol Methods 22. Summers B, Cummings JF, De Lanhunta A: Inflammatory 95:145–151, 2001. diseases of the central nervous system. In: Veterinary Neuro- 25. Wilkins PA, Del Piero F: Rabies. In: Equine Infectious Diseases, pathology, ed. Summers B, Cummings JF, and De Lanhunta A, ed. Sellon DC and Long MT, pp. 185–191. Saunders Elsevier, St pp. 95–188. Mosby, St. Louis, MO, 1995. Louis, MO, 2007. 23. Swanepoel R: Rabies. In: Infectious Diseases of Livestock, ed. 26. Woldehiwet Z: Clinical laboratory advances in the detection of Coetzer JAW and Tustin RC, 2nd ed., vol. 2., pp.1123–1182. rabies virus. Clin Chim Acta 351:49–63, 2005. Oxford University Press, Cape Town, Africa, 2004. 27. Zimmer K, Wiegand D, Manz D, Frost, JW, Reinacher M, Frese 24. Whitfield SG, Fekadu M, Shaddock JH, Niezgoda M, Warner CK, K: Evaluation of five different methods for routine diagnosis of Messenger SL, Hanlon C, Morrill P, Murray K, Orciari L, rabies. J Vet Med B 37:392–400, 1990.

636