University of Zurich, Vetsuisse Faculty

Home , Ovine gammaherpesvirus 2, Persian cat

Favrot, C. Canine and feline viral dermatoses with a particular emphasis to papillomavirus infections. 2007, University of Zurich, Vetsuisse Faculty. Postprint available at: http://www.zora.uzh.ch University of Zurich Posted at the Zurich Open Repository and Archive, University of Zurich. Zurich Open Repository and Archive http://www.zora.uzh.ch Originally published at: University of Zurich, Vetsuisse Faculty, 2007.

Winterthurerstr. 190 CH-8057 Zurich http://www.zora.uzh.ch

Year: 2007

Canine and feline viral dermatoses with a particular emphasis to papillomavirus infections

Favrot, C

Favrot, C. Canine and feline viral dermatoses with a particular emphasis to papillomavirus infections. 2007, University of Zurich, Vetsuisse Faculty. Postprint available at: http://www.zora.uzh.ch

Posted at the Zurich Open Repository and Archive, University of Zurich. http://www.zora.uzh.ch

Originally published at: University of Zurich, Vetsuisse Faculty, 2007. Vetsuisse-Fakultät- Universität Zürich Klinik für Kleintiermedizin (Direktorin: Prof. Dr. Claudia Reusch)

Canine and feline viral dermatoses with a particular emphasis to papillomavirus infections (Virale Dermatosen bei Hunden und Katzen unter spezieller Berücksichtigung der Papillomavirus- Infektionen)

HABILITATIONSSCHRIFT

Zur Erlangerung der Venia Legendi an der Vetsuisse-Fakultät der Universität Zürich

vorgelegt von

Dr. med.vet. Claude Favrot Diplomate, European College of Veterinary Dermatology Masters in Sciences Zürich, 2007

Chapter 1 General introduction - Aims and scope of the thesis 3

Chapter 2 Virale Dermatosen bei Hunden und Katzen 11

Chapter 3 Parvovirus infection of keratinocytes as a cause of 24 canine erythema multiforme

Chapter 4 Two cases of FeLV-associated dermatoses 28

Chapter 5 Evaluation of papillomaviruses associated with cyclosporine-induced hyperplastic verrucous lesions in dogs 35

Chapter 6 Detection of novel papillomaviruses in canine mucosal, cutaneous and in situ squamous cell carcinomas 42

Chapter 7 Detection of novel papillomavirus-like DNA sequences in paraffine embedded samples of feline invasive and in situ squamous cell carcinomas 52

Chapter 8 Clinical, histological and immunohistochemical study of feline viral plaques and bowenoid in situ carcinomas 59

Chapter 9 Summarizing discussion and further studies 68

Chapter 10 Zusammenfassung und weitere Studien 76

Acknowledgments

The studies in this thesis were conducted at the University of Zurich, Switzerland and at the Clinique Vétérinaire de Ferrette, France. They were financially supported by these institutions as well as the Waltham Foundation and the

European College of Veterinary Dermatology.

Chapter 1

General introduction Aims and scope of the thesis

Skin lesions are a predominant feature of many viral diseases in humans and large domestic mammals [1, 2]. In comparison, viral dermatoses are rarely described in dogs and cats [3]. However, they may be underdiagnosed due to the difficulty in detecting viruses. The development and increased availability of diagnostic techniques such as electron microscopy, immunohistochemistry, viral amplification (PCR) is making detection more routine [4]. To complement advances in diagnosis, several effective antiviral agents for treating at least some viral dermatoses have been developed in the last few years [1]. Consequently, making the correct diagnosis may actually have important consequences for therapy as well as for prognosis.

Definition and classification:

Viruses form a diverse group of non-cellular infectious agents that share a distinct composition and a unique mode of replication. These agents lack much of the enzymatic machinery necessary for their multiplication. They are consequently obligate intracellular parasites that multiply inside cells and use the synthetic apparatus of the host cell to produce their own components [1]. The animal viruses are divided in several families according to their shape, structure of virion and the type of nucleic acid within it [5-7]. In fact, in the virion, the genome of the virus consists of only one type of nucleic acid (DNA or RNA). The viral particle (virion) is made of the nucleic acid surrounded by a protective protein core called the capsid. Some viruses do additionally possess an envelope which play a major in the infection of host cells [5-7].

Viral Replication and host response:

Viruses make use of the host-cell machinery to synthesize and assemble viral particles (replication) because they do not possess the required enzymes. Viruses encode for structural (capsid) proteins and non structural proteins that usually regulate the viral replication. Some

of these non-structural proteins (transforming proteins) are responsible of the oncogenic potential of tumour viruses, like high-risk papillomavirus [8, 9]. Some viruses, like papillomaviruses depend on epithelial cell differentiation for completion of their replication [10]. After an initial inoculation in the basal layer of the epithelium, non-structural (early) proteins are expressed in the suprabasal layers and in the stratum spinosum. Capsid (late) proteins are subsequently produced in the stratum spinosum and the viral particles assembled and released in the upper stratum granulousm and stratum corneum, respectively [10]. Epithelial host-cells are infected by papillomaviruses through inoculation but for most other epithelium-infecting viruses, the replication cycle begins with an attachment phase called adsorption [1, 11]. This attachment requires specific interaction between host-cell receptors and virus. Cells lacking virus-specific receptors are not susceptible to infection. Following adsorption and penetration, viral envelopes and capsids are destroyed (uncoating) and viral genome can therefore instruct the host-cell machinery to produce its own proteins. Viral particles of most viruses infecting epithelial cells (herpes, papillomavirus, poxvirus) are produced inside the cells and released after cell death or cytolysis [1]. Each virus has its own site of replication: Herpesviruses and papillomaviruses replicate in the nuclei whereas poxviruses multiply in the cytoplasm. Viral replication usually causes gross cytopathic changes and host-cells sometimes die. These cytopathic effects may be pathognomonic of one specific viral infection (viral inclusions and pseudo-inclusions, syncytium formation (herpesvirus, retrovirus, paramyxovirus), modified keratinisation process (papillomaviruses). Some viruses, like papillomaviruses or distemper virus, can however, at least in some instances, replicate without causing irreversible damage to the host keratinocytes (true commensality, chronic infections)[12-15]. Another form of non cytocydal infection is the latency (herpesviruses, papillomaviruses). In this instance, very few or no virion are produced in the infected cells but reactivation of the infection can occur at any moment. Last but not least, some viruses (papillomaviruses, retroviruses) are able to induce host-cell immortalization and neoplastic transformation. Most of the time, transformed host-cells loose their ability to sustain productive infection [8, 9].

Viral infection and skin lesions:

Virus-induced skin lesions are usually the direct consequence of the virus replication. Examples of these direct effects are wart formations associated with papillomaviruses infections, pock formation in poxviruses infections or vesiculation associated with herpesvirus infections. Some skin lesions are however due to the host response or to the interaction of replication and host response. Erythema multiforme, for example, are exanthematous skin lesions associated with herpesvirus infection in humans and cats and may be, with parvovirus infections in dogs [16-18]. This reaction is considered to be due to the destruction of infected keratinocytes by cytotoxic T-cells. Finally, viruses may also modify skin biology and cause indirect changes, like in the so-called “hard pad disease” [15, 19-21].

Diagnosis of viral infections:

Several approaches are now available to diagnose viral infection: virus isolation and culture, microscopy, serology or detection of viral antigens or nucleic acids. As these techniques demonstrate increasing sensitivity, results should always be interpreted in the context of the clinical and histological setting. In fact, one must always keep in mind that virus infection can be fortuitous and unrelated to the disease. Cultivation of the virus and/or direct identification (pathognomonic cytopathic effects) from the clinical material represent the “gold standard” for viral diagnosis because they establish at the same time that the virus is present and actively replicates in the lesional sample. These techniques are however limited by the low sensitivity and the difficulty to cultivate some viruses like papillomaviruses. The recent development of techniques that allow the amplification and multiplication of viral nucleic acids (PCR) has dramatically increased the sensitivity of virus detection. The main pitfall of PCR is however its great sensitivity itself, as false-positive assays may result from the amplication of minute amount of nucleic acid of viruses unrelated to the disease. Electron microscopy and immunohistological identification of viral antigens in lesional samples are also available. As they detect productive infections, these techniques may be regarded as more specific. They are however less sensitive than PCR techniques.

Serologic studies to detect antiviral antibodies are important for epidemiological studies, to determine the prevalence of one specific virus in a population, and to detect individuals that have been previously affected by the condition or in contact with the virus.

Aims and scope of the thesis:

As mentioned above, viral dermatoses are rarely reported in domestic carnivores [3, 22]. The aim of this thesis was to report some unusual aspects of viral infections of the skin of domestic carnivores and to show that at least some of these viral cutaneous infections remained under-diagnosed. Furthermore we aimed to broaden our knowledge of the genetic diversity of carnivores papillomaviruses and to demonstrate that these latter viruses may contribute to the development of skin cancer in these species. We have first shown that canine parvovirus 2 is able to induce, in some instances, clinical and histological changes that mimic human erythema multiforme (EM) [17]. Similar changes have already been described in dogs but were, most of time, attributed to drug reactions [23]. On the contrary, true EM is virtually always associated in Man with herpesvirus infections and almost never with drug reaction [16]. Feline EM has also been shown to be due to herpesvirus infections [24]. The disclosure of canine EM associated with virus infection should encourage veterinary dermatologists to look for virus antigens or nucleic acids in skin samples of dogs affected by this condition. We have described and studied two cases of FeLV-induced skin conditions [25]. We have first shown that FeLV, like other retroviruses, is able to induce syncytium formation in the skin of infected cats. Similar cases have already been described but with a very different clinical phenotype [26]. More interestingly, FeLV antigens and nucleic acids were uncovered in cutaneous lymphoma samples in a serologically negative cat. These findings suggest that focal skin FeLV replication may occur in some instances. Dogs treated with cyclosporine A sometimes develop lichenoid plaques and warts of unknown origin. We have evaluated such lesions in nine affected dogs and demonstrated that the majority of these plaques are not papillomavirus-induced. Some however harbour papillomavirus DNA and antigens and are probably due to the reactivation of a latent PV- infection of the skin [27]. Anecdotal reports have suggested that PV could play a role in the development of skin squanous cell carcinomas in dogs and cats [28-33]. We have consequently tried to amplify

PV DNA from skin samples of canine and feline squamous cell carcinomas and demonstrated that nucleic acids of these viruses are present in a significant amount of such samples [34, 35]. Furthermore, amplified PV sequences revealed that these samples are infected by PV of great genetic diversity. These findings suggest that PV could play an active role in the development of such cancers in dogs and cats and that domestic carnivore, like humans, may be infected my numerous different PVs. A last study carried out on a subset of feline squamous cell carcinomas (Bowenoid in situ carcinomas: BISC) in situ has shown that BISC are often infected by PVs and that these viruses actively replicate in such lesion [36]. These results further suggest an active role of these viruses in the development of such lesions.

1. Lowy, D.R., Viral diseases: General considerations, in Fitzpatrick's Dermatology in General Medicine, I.M. Freedberg, et al., Editors. 2003, Mc Graw-Hill: New-York. p. 2035-2041. 2. Scott, D.W., Viral diseases, in Large Animal Dermatology, D.W. Scott, Editor. 1988, W.B. Saunders: Philadelphia. p. 96-119. 3. Scott, D.W., W.H. Miller, and C.E. Griffin, Viral, rickettsial and protozoal diseases, in Muller & Kirk's Small animal dermatology, D.W.M. Scott, W.H. Griffin, C.E., Editor. 2001, W.B. Saunders: Philadelphia. p. 517-542. 4. Sellon, R.K., Update on molecular techniques for diagnostic testing of infectious disease. Vet Clin North Am Small Anim Pract, 2003. 33(4): p. 677-93. 5. Condit, R.C., Principles of Virology, in Fields Virology, D.M. Knipe and P.M. Howley, Editors. 2001, Lippincoot, Williams & Wilkins: Philadelphia. p. 19-51. 6. Harrison, S.C., Principles of virus structure, in Fields Virology, D.M. Knipe and P.M. Howley, Editors. 2001, Lippincott, Williams & Wilkins: Philadelphia. p. 53-85. 7. Murphy, F.A., et al., The Nature of Viruses as Etiologic Agents of Veterinary and Zoonotic Diseases, in Veterinary Virology, F.A. Murphy, et al., Editors. 1999, Academic Press: San Diego. p. 3-22. 8. Harwood, C.A. and C.M. Proby, Human papillomaviruses and non-melanoma skin cancer. Curr Opin Infect Dis, 2002. 15(2): p. 101-114. 9. zur Hausen, H., Papillomaviruses causing cancer: evasion from host-cell control in early events in carcinogenesis. J Natl Cancer Inst, 2000. 92(9): p. 690-698. 10. Doorbar, J., The papillomavirus life cycle. J Clin Virol, 2005. 32(Supplement 1): S7- S15. 11. Lowy, D.R. and P.M. Howley, Papillomaviruses, in Fields Virology, D.M.H. Knipe, P.M., Editor. 2001, Lippincott, Williams & Wilkins: Philadelphia. p. 2231-2264. 12. Antonsson, A., et al., Prevalence and type spectrum of human papillomaviruses in healthy skin samples collected in three continents. J Gen Virol, 2003. 84(Pt 7): p. 1881-1886. 13. Antonsson, A., et al., The ubiquity and impressive genomic diversity of human papillomavirus suggest a commensalic nature of the viruses. J Virol, 2000. 74(24): p. 11636-11641

14. Antonsson, A. and B.G. Hansson, Healthy skin of many animal species harbours papillomaviruses which are closely related to their human counterparts. J Virol, 2002. 76(24): p. 12537-12542. 15. Haines, D.M., et al., Immunohistochemical detection of canine distemper virus in haired skin, nasal mucosa, and footpad epithelium: a method for antemortem diagnosis of infection. J Vet Diagn Invest, 1999. 11(5): p. 396-399. 16. Assier, H., et al., Erythema multiforme with mucous membrane involvement and Stevens-Johnson syndrome are clinically different disorders with distinct causes. Arch Dermatol, 1995. 131(5): p. 539-543. 17. Favrot, C., et al., Parvovirus infection of keratinocytes as a cause of canine erythema multiforme. Vet Pathol, 2000. 37(6): p. 647-649. 18. Huff, J.C., Erythema multiforme. Dermatol Clin, 1985. 3(1): p. 141-152. 19. Grone, A., M.G. Doherr, and A. Zurbriggen, Up-regulation of cytokeratin expression in canine distemper virus-infected canine footpad epidermis. Vet Dermatol, 2004. 15(3): p. 168-174. 20. Grone, A., P. Engelhardt, and A. Zurbriggen, Canine distemper virus infection: Proliferation of canine footpad keratinocytes. Vet Pathol, 2003. 40(5): p. 574-578. 21. Koutinas, A.F., et al., Histopathology and Immunohistochemistry of Canine Distemper Virus-induced Footpad Hyperkeratosis (Hard Pad Disease) in Dogs with Natural Canine Distemper. Vet Pathol, 2004. 41(1): p. 2-9. 22. Favrot, C. and S. Wilhelm, Virale Dermatosen bei Hunden und Katzen. Tierärtzliche Praxis, 2006. 24 (K): p. 307-318 23. Scott, D.W. and W.H. Miller, Erythema multiforme in dogs and cats: literature review and case material from the Cornell University College of Veterinary Medicine (1988-1996). Vet Dermatol, 1999. 10: p. 297-309. 24. Prost, C., A case of exfoliative erythema multiforme associated with herpesvirus 1 infection in a cat (abstract). Vet Dermatol, 2004. 15 (Suppl. 1): p. 51. 25. Favrot, C., et al., Two cases of FeLV-associated dermatoses. Vet Dermatol, 2005. 16 (6): p. 407-412. 26. Gross, T.L., et al., Giant cell dermatosis in FeLV-positive cats. Vet. Dermatol., 1993. 4(3): p. 117-122. 27. Favrot, C., et al., Evaluation of papillomaviruses associated with cyclosporine- induced hyperplastic verrucous lesions in dogs. Am J Vet Res, 2005. 66(10): p. 1764- 1769. 28. LeClerc, S.M., E.G. Clark., and D.M. Haines,. papillomavirus infection in association with feline cutaneous squamous cell carcinoma in situ (Abstract). in AAVD/ACVD Meeting. 1997: p. 125-126. 29. Schwegler, K., J.H. Walter, and R. Rudolph, Epithelial neoplasms of the skin, the cutaneous mucosa and the transitional epithelium in dogs: an immunolocalization study for papillomavirus antigen. Zentralbl Veterinarmed A, 1997. 44(2): p. 115-123. 30. Sundberg, J.P., R.E. Junge, and W.D. Lancester, Immunoperoxidase localization of papillomaviruses in hyperplastic and neoplastic epithelial lesions in animals. Am J Vet Res, 1984. 45 (7): p. 1441-1446. 31. Teifke, J.P., et al., Detection of papillomavirus-DNA in mesenchymal tumour cells and not in the hyperplastic epithelium of feline sarcoids. Vet Dermatol, 2003. 14(1): p. 47-56. 32. Teifke, J.P., C.V. Lohr, and H. Shirasawa, Detection of canine oral papillomavirus- DNA in canine oral squamous cell carcinomas and p53 overexpressing skin

papillomas of the dog using the polymerase chain reaction and non-radioactive in situ hybridization. Vet Microbiol, 1998. 60(2-4): p. 119-130. 33. Watrach, A.M., E. Small, and M.T. Case, Canine papilloma: progression of oral papilloma to carcinoma. J Nat Cancerol Inst, 1970. 45: p. 915-920. 34. Nespeca, G., et al. Detection of novel papillomavirus-like sequences in paraffin- embeddedspecimens oe invasive and in situ squamous cell carcinomafrom cats. Am J Vet Res, 2006. 26(12): 2036-2041 35. Zaugg, N., et al. Detection of novel papillomaviruses in canine mucosal, cutaneous and in situ squamous cell carcinomas. Vet Dermatol, 2005. 16(5): p. 290-298. 36. Wilhelm, S. et al. Clinical, histological and immunohistochemical study of feline viral plaques and bowenoid in situ carcinomas. Vet Dermatol, 2006. 17: 424-431

Chapter 2

Virale Dermatosen bei Hunden und Katzen

C. Favrot1, S. Wilhelm1

Tierärtzlich Praxis, 2006, 34 (5): 307-318

1Klinik für Kleintiermedizin, Dermatologie, Vetsuisse-Fakultät der Universität Zürich, Schweiz

11 Einl Vi Key dog Einge Vi C. Vi Au Vi Hun Sch die he me be dings der fe we dir we so ma er rus rus ne höhe lc itsbilde st ra ek is be Favro sd tho lz ral lü Nac d–K ehen he Er wo –c nL le ew ga el te –d –D ss ka eitung ab er de nF re le ngen: n. Der rz at rds el hw äsi ird nn eD er erm ge nm ng erm at Kl de t, ytopathisc wö In rv äl : ze on ma rs inik eis au nk da die le S. 07. el at er at odif rt er n, ch durc ch ing tos osi linis ist er: mutlic vi ose 04. ermatos Wilhelm st Di fü we in ra er ind en izie s– te 2005; rK ke agno hn le – em ch de rz sh in ire ko he rA leintiermedizin, re hl en oc ne ui akze se rE no äu Fa mme kt au an nti hk nd Bi de fi ll sr mitte schäd ff ge de ge ptier ld ei ge ie ei ek nti we nb ne rd ne übe in ch r, Vi te t: fi en rde ls Ve ei Ka Vi en ige ie ode 05. zi re rse –g Hi Vi mu as and schaf cat ski will se ma th vi dann have di pose zwi no Vi Sum Diagn fik tr in Hunden Zi Vi Zusam re nd er rbindun de Kl nd Wa us ovi 01. ra n. rectl di el ral ral at ral at n, rN ch he st nl nn en est nucl nol ls bei io schen rB rekt ei al le onl Para- di 2006 ol io ie rus, Di are ma ie eD .I ch De die derm esi n. ns consequent derm of nti ei itä nE te keinen el os ukleins eser nm ogi ach og tw si yb me Ant yb ew es st ngegangen, ei Durc Ke no ry: lt. erm th ons ma rm von nd tb es er di und rkra indir Lent ie umsra gz ca schen ill yv em achen Poxvi nf ei is zu gesc at Hu Ve rectl en ra Vi at der ,w tell ew ige at yb Übers fo –d s( ass at does ub rev os Vi we ir ci tinoz os hb rände nk Katzen re Bewei ologi iv ent ek cus äu sel ie un uses osen ea niz urde da ren es Ant 18, em hi ir yc es nk ringe unge rus, ung: ie is te sen nd td esser re Reakt us, g. io te den eht ssoc en ic wi itä are not ly on mp yte aused of ni ig de Vo der ned urch önne All nv e( runge 45). ht th kom Herpes we been sf Mo we tv eni dies st nn condi und th darz n( lific rH nd vo sarbei Nachw nd prove eN cons rha er at Le und io or ia ür Gattu ed op wh rde on nn tä rbi 89). ding ur ra me nd nen, Vi te Üb mo er iter: eK at au nde ach by n, urc das ust (86). td mi fa og res en ren dw lliv n. tion Ep besagt id ass io lle nb be vi tz di as li ci di th ti er vi sd Aller ng td ered el hd ne Auch we ra ns ei ch rus, th wie and ir el fy ent lita mi id ed Dr Vo st, at tr ruses ith ozii ei sd ve le Ge- us Ka Haut sk en nk- arf ey ei le em Da of- er er is nabl th .C th n. as rl te Kl die rurs ir canceri ir n n n th Retrovi bei le rare eA - - - cat und ert - ie ep are un ev ekt ni ect d. ei Berück .F ec in zell gen tz cht and et kl nt spi dF 12 acht ssoz How ir si rol ster av fr ly auf Parvovi in in lin us ie lok st Apo Gef de Vi mögl sen Pa Pa Ät ne nd, en he e equent el am vergess ife or is ei rot) eh ren dom condi rus, ic zati rk na re pil is grund sw ia ever chen vi ner iologie al aus. si we det rat in zu al ahr In ilie be tb pt na pil tio th ra linis cht eI lom ichk ,V di nur ei on us Lent est and osei io de iH ec rden, unt ect le nP rus nen vi ,o ei rectl se tion uc ly mmun et Sol ge np ein ig lom und rD or ral rf en ei ic av se ause ch ne report ei unde erschei io hd ung api su de ei iv hi nd prä verursacht ner olgende st che in ani immu ropert er NA- en te ire no lte we yc en ir nem st zei mu rs is hi llo uk as us, nd hat der und gten ol nv n( ma find av kar Derm aused sy se st of nu Pr rden, fm Vi chne ma Mo st ch ed tion ogi lok und ol Mo er nol rus Eryt st th or äsen PV) are den. ls ie Li kee Fa zin nd in ne en ogi iren vi in em di cal ed te nÜ so al Pl hä but und ogi ut in rbi at ns ru RNA-M Pat das ku ll (Er fika hem th om rat by som eI Ka schen in pi at ta fe ea is uf ose we Auf ge rt s, fe lliv al el be cr wa are ltät, sbe ic te ease. tion kti nm mmun vi ur yth sd we tze tion igste atö at he Ortho sf höre ite mo hd rden. am rsic hog ne et ir ruses eact ch on Let darst ures häuf er son us probabl rden n. ime in em re rat se ant ,d pith enge Un Ve en unt urch sen ul htsa dt Nachw nn nz en ztere md It and nE io er -u ure. de am sy tifor rände Neu ig sa ene ig iv können is so of el of hat tst ns el de ve ud ach Hi eni nd nm st ersit re lt. rbe man erw er verändert -T Parvovi ntar th vi ssoc an fv nE ult yo Som em me, rm such wir es st det ral ka De Ve ci eg Ti en Parapoxvi ropis ge ögl ei runge se it de ir olo ähnt if erärz ut fte te ty dat ät tung. nnt di al rände sv schäd mz li pider ect wie ia enera derm ei or mit ed DN lic ns ic Te eg nu of as gie ©2 te mv nhergehen. Zür DN ory ir me). uf rus. hu mus ha io tl erm we chni al nv senen und td dw in Kan sk A- nderdi eryt ol eP Prax no 006 at malz und Aa orl er runge d, Nic ich, nd ige to uch as in Papi ge rden. ir Vi New oses at rus, ith fv zero DN ken Im und al ie (52). di rol hem es cel Sch verei nd n. oses re 2006; ht is de er genden st Sch ir häuf el Schweiz vir vi agnosed. llo ife Ai te n, nu ko al zw Herpes in ls RNA. te wie ru nB lpo ge Ei sol in Derm zu we am ma al ri .T nH gui at nf su di nucl chni ni mmt Das ses. are 34 ei nd ne sa ig dom le en ta chen, pu rg ed eh achen hese ge vi DN ul te sh aut nerl (K): tz uer übersehen. tion se ew Di ru at Art caused zei an la tifor ei ques nF Der ir Th tk vi Der Vi est di A- zu agnoses osen lä s, ca tion od ekt The 307-18 GmbH dlungs- rus, ic äss se ik ge seas el mo ru di al si Amp önne Ortho me. ic ma so ht mato- er el ci ei ige on em such ne lb at si oder lic pur di ani di Re- au ne nur die im- bei di to- te in en es n- fi- e- li- h, st i- n n it e f r - r - - -

H UND /KATZE 307 308 Virale Dermatosenbei Hund und Katze C. Favrot, S. Wilhelm

normalerweise speziesspezifisch, ansteckend und wird oft durch gebaut. DieFreisetzung desVirions erfolgt gleichzeitig mit der

ZE Mikroläsionenübertragen(74). BeiHundenwird die Mehrheit Desquamation (52). derInfektionendurchdas kanine orale PV (COPV) verursacht. Papillomavireninduzierenauchdie Syntheseder so genann- AT Allerdingsließensichauchandere Stämme auskaninenHautlä- ten transformierendenProteine E6 und E7, die für die proliferati-

/K sionenisolieren(60, 103). BeiKatzen istdiese Krankheit selten venund karzinogenenEigenschaftendes Virusverantwortlich

D und wird durchfeline oder bovine Papillomavirenhervorgerufen sind (45). (85, 98). Bisherkonnte erst beieinemkaninen(COPV) und einem EineInfektion durchein unproduktivesPapillomavirus ist

UN felinen(FdPV-1) Papillomavirus dasGenom aufgeschlüsseltwer- ebenfalls möglichund wird mitder Entwicklung einesfelinen

H den(99, 102). Sarkoids oder einesFibropapilloms assoziiert (85). Es wird an- Die Replikation desVirus läuftparallelzur Differenzierung genommen,dass es sich hierbeiumdas Pendant desequinen desPlattenepithelsabund kann in eine „Early Phase“ (E) und ei- Sarkoids handelt. In derMehrzahlder Fälle gelang es, in solchen ne „LatePhase“ (L) unterteilt werden. Während derE-Phase im LäsionenPV- DNAnachzuweisen,wobeidie Blast-Analyse der Stratum spinosumund dersuprabasalen Schicht werdendie vira- L1-Sequenz starkeHomologienmit einembovinenPVaufwies lenProteine synthetisiert.ImStratum granulosum findetwährend (42, 85, 105). derL-Phase die Syntheseder Kapsidproteine statt. Erstimoberen Bereich des Stratum granulosum wird dasKapsidzusammen- Klinik Hund Es existierenverschiedene klinischePräsentationen,die unter- schiedlichenVirenzugeschriebenwerden, derenkomplettesGe- nom jedochnochnicht entschlüsseltist.Dennochist es möglich, dass jede einzelne klinische Präsentation (z.B.orale Infektion oderinvertiertes Papillom)mit einemspezifischen PV assoziiert ist(9, 30). Kanineorale Papillomatose(COPV): Hierbei handelt es sich um dieklassischsteund auch häufigste transiente virale HauterkrankungbeimHund. Betroffensind meistjunge Hunde. EineRassen-oderGeschlechtsprädisposition gibtesnicht. Die Läsionenbefinden sich aufder Schleimhaut derMaulhöhle,der Lippen, desPlanumnasaleund/oder derKonjunktiven (87). Aus- nahmsweise könnenauchdie peribukkale Haut,die Augenlider oderder obere Gastrointestinaltrakt betroffen sein(Abb.1). In der Regel sind es mehrereLäsionen, die durchihre wuchernde,hyper- Abb. 1 Typische viraleWarzen (kanine oralePapillomatose) plastische Struktur dasTierstören können. DerDurchmesser va- riiertvon 2–3 mm bis zu über3cm. Ihrexophytisches Aussehen, ihreGröße,die blumenkohlartige Oberflächeund die Lokalisa- tion derLäsionenlassen oft bereitseine klinischeDiagnosezu(6). Kanine orale Papillomebildensichinden meistenFälleninner- halb voneinembis drei Monaten spontan zurück(14). BeiVorlie- geneinerImmundefizienzwurdenpersistierendePapillomavirus- Infektionenbeschrieben(63, 73). Papillomatosen des adultenHundes: Beidiesen Läsionen kommt es zu keinerspontanenRückbildung. Hunde jedenAlters können betroffensein. Die Veränderungentretenoftmals an mu- kokutanenÜbergängen, voralleminder Maulhöhle (hieramhäu- figsten und gravierendsten),dem Gesicht undzwischen den Ze- henauf (Abb.2). Es handelt sich um multiple typische Warzen, die miteinerdickenKeratinschicht bedecktseinkönnen(Abb.3). In vielenFällensind die Läsionenpigmentiert,manchmalstehen siedicht nebeneinander stehendund könnenbis in denPharynx reichen(117), wo siezuSchluckstörungenführen. Die Ätiologie Abb. 2 Persistierende Warzen am Schwanz einer erwachsenen istnochungeklärt, dochwerdenvielfach Immundefizienzen als Deutschen Dogge (Papillomatose desadultenHundes) Ursache angenommen(7, 63, 73, 75, 100, 117).

13 Ab Abb ru Hu eben ne Po schei si fi sen Ep ne sam pla Hun ma Th es nd ge nge nD ln re b. zu id er nde sm nc In Mul .3 me rv ei den sel fa ap er 5 (9). dies de he ei nt nz at ve urc lls mod or nh ie au te ne n( nH is re ig tip rt Pap Wa ko eP Das hmes n, ch zu f( an und rm mög te ar 9). ie nnte le ys ko rze en unde Abb rk nv gg rt tige illo la al pla pi ur eP mme qu wurde lic ser ompl or Gr ma ign eb am gme sä ei .4 si his nf es h. ap al ra an ne chl vi vo av en ). Pf nj le ch Di tolo ru kli ühr il et ula nt Ei nm n1 sp ot er ic ma Tr lom s-indu ed te t( ea ie nisc enba he ne ontane ru te gis an nA –2 au oc rt 60, it mA ci ufge e: PV St sf da ep f( ch hu hb immun for cm bhe lle or zi Di re 72, sA 60). eS bdomen ap er ei sch wölbten nd uung mation nm mis Abhe ese au il te il ei tr 96, bs ung, ei histopa fu lom it Be uk pi nigen su (72, unge et nt Ho gm übe il nd 111). zen tur ppre in ung si atöse wie (54, und rnbi ent 111). Ve wöhnliche ch en be rd ic Ra thol de ssi ht de rä be si ie ld en am vo In mit 72, ssen ri nde tz rt Pla ve ung ru immun oba And og mC ge eP manc en Ki mb mmun nT 96, eo ru ques is sam ch (z la nn im ere ch si OPV nge he .B ei nH ques 111). te nophil he su junger te Ze ra tw de dage an su (A .M nw nF nK au ppri -1 pien ntr rh ppre de er bb tv ops zu äl ei gen de ör en re um umane mi .5 er le ad sen in nk ssi pe n. unter )h än ): nä zy er ult we ei ri ve Zu Be äu de Si te am to- ei ne h- er st i- n n n e - - - - - i 14 Pa fi sei de li he bu serk schu re Be Ka Abb. Abb. vo (10 Ka Ätio ch nden nG rL ute nC pil nge nG id rzi 1, tze er Be Pl at lo ppigen, 6 4 lamo en no ite esi we OP wur 10 zen att nc gi en iK sich mi ra ei me ch 4, Pap In is om enepi V- ha at de vir tur em st und ve ts 11 in zen is nur ra ns illo zw hä pigm erst ko rt us te duzi kterisie 2). nu ult ie thelkarzi lfte itu nK im -i ma ar mple we ko rt rd nduz ein if Bei es er an munsupprim mmt en ni vi oka (B at un er te ru zu ge tie tt Papi ein zen rt ow nV da eine ier s-indu len, Fa ne en en rt Beri es zig en’ nom uf ten en llbe eränder hm äu llo ts nur Ve mH ch es fe sD m( ße de cht Pl en zi Wa rä ric st lüs kat sel rt er rZ aq Vi (Abb Abb ,w und nde eü si ier is ht rale rz te si sel ue ungen tz te ease) ze unge urde ber te ch en sP ei ru en .: nz s( nsp t( wir .6 Der D. ne Ti nge de dies la an Abb uo Fd ei bi dü (98). ): er rK ez Carl tt ma bericht ne nu in de enepi PV sd nf ra es if eK In .7 to Pl ber ma at le nA ot is nd at -1) ind se ind Ans at )( ze nL ch ti) de ra oa lig nb th te C. ei et uge durc 12, en nkhe (102). es rV mit ber el äsi nepi pr nm ne ei Fav ons (35) karzi PV si nli äd 20, Hund one et hl Tr ni rot, ch ex ten it ul th erinär is de ansf cht .E en ei in 62, tiz el oph nom po n. in S. rn und ex ch tde karzinom ent Fo in bestät Die nie Wi de or 98). yti (13). is te me ev ck in rm Katze lh ma risches ru tie rt se schen Erhe el si tu di .B ir rt Pe nte- ig m tio vo tu üb- zi al nd in t. is r- n n n e e -

H UND /KATZE 309 310 Virale Dermatosenbei Hund und Katze C. Favrot, S. Wilhelm

Regelmäßig wird überKatzen mit multizentrischen Karzino- Hundenunterscheidetdie histologischeUntersuchung zwischen

ZE men in situ berichtet(Abb.8)(2, 67). In einerdiesbezüglichen einerechtenPapillomatoseund einernichtviralenwarzenähn- Arbeitwurde PV beibis zu 40% deruntersuchten Probennachge- lichen Läsion. Histopathologischerkennt man eine papillomatöse AT wiesen (59). Solche Veränderungenwerdenauchmit Demodex- epidermaleHyperplasiemit ballonierender Degenereration von

/K InfektioneninZusammenhanggebracht(38). ZellenimStratum spinosum(Koilozytose). Zusätzlichkommtes

D Die letzte Form der felinen PV-Hautinfektionen ist das Fibropapil- zu einerHypergranulosemit prominentverklumpten Kerato- lomoder feline Sarkoid(42, 85, 105). Diese Läsionen im hyalingranula (118). Eosinophile intranukleäre viraleEinschluss-

UN Gesichtoder an den Extremitätenbestehen aus festenKnoten in der körperchenund basophiles intrazytoplasmatisches fibrilläres

H Dermis, wahrscheinlichaufgrund einer nichtproduktivenPV-Infektion. Materialbestätigendie ätiologischeDiagnose(118). Fehlen virale Einschlusskörperchen, kann eine virusinduziertePapilloma- Diagnosestellung tosenicht ausgeschlossen werden. Zum Nachweis der Ätiologie sind auch immunhistochemischeTechniken und DNA-Amplifi- Oftgenügt beijungenHundenmit oraler Papillomatosebereits kation nützlich. die klinischeUntersuchung. In fraglichenFällenund beiadulten Epidermale Hypermelanose, unregelmäßige Akanthose und Hypergranulose mitverklumpten Keratohyalingranula werden normalerweise beiVorliegenvon kaninen multiplen pigmentier- tenpapillomatösen Plaquesgesehen (72, 96, 111). In einigen Fäl- lenkonnten viraleEinschlusskörperchennachgewiesen werden (60, 96). BeiKatzen liegenähnlichehistopathologische Verände- rungenvor,wobeivirale Einschlusskörperchenhäufigerbeob- achtetwerden(12, 20, 62, 98). HistologischeUntersuchungenvon felinenSarkoidenerga- beneine pseudokarzinomatöseAkanthose und dicht gepackte mesenchymale Zellen, die Kollagenbündelumgaben (85, 105). Therapie

Es existierenverschiedene Therapievorschläge,deren Erfolg je- docheherauf spontanerRegression alsauf echterHeilungberuht (74). Beistarkem Verdachtauf eine spontan abheilende COPV- Entzündung ist es durchaus vernünftig, alsErstesabzuwartenund Abb.7 Feline Papillomavirus-induzierte pigmentiertePlaques die Läsionenzubeobachten(87).Am effektivstenerweist sich ein (Abb.:C.Mege) chirurgisches Vorgehen,besonders in Form vonKryochirurgie und Lasertherapie,dochwurde die chirurgischeVorgehensweise auch mit Resistenz und Rezidivenassoziiert (61). Zusätzlich gibt es Protokollefür Behandlungen mitRetinoidenund Interferon, überderen Wirkung bisherabernochkeine ausführlichenStudien publiziertwurden(114, 115). BeimMenschen, inklusive immunsupprimierten Patienten,werden vireninduzierte Neoplasien derzeitig mit Imiquimodund anderen Imidazoquinolonen therapiert(92).Diese Medikamente regen dieToll-like-Rezeptoren an und führen so zur Freisetzung vonZyto- kinen vomTyp Th1 und vonInterferon-α .Über dieerfolgreiche An- wendung vonImiquimodbeim Hund liegen nur wenige anekdoti- scheBerichte vor. Aufgrund seiner spezifischen antiviralenWirkung wird beimMenschen auch Cidofovir eingesetzt (79).Über einen möglichen Einsatz bei Hunden ist noch nichtsbekannt. Aktuelle Fortschritte in der Präventionund der Immuntherapie der humanenPapillomatose, speziellbeim Papillomaviren- Abb. 8 Felines induzierten zervikalenKarzinomder Frau, könnten als Basis für im- Papillomavirus-in- muntherapeutischeBehandlungen vonPV- Infektionen dienen(61). duziertesPlatten- Bis dato wurde nicht übereine erfolgreiche Behandlung kani- epithelkarzinom in nerpigmentierter Plaquesberichtet, doch kann eine spontane Re- situ (Ohrgrund) gression auftreten. Zusätzlich empfiehlt es sich,die Ursache der

15 Di Hund Klinik Wi Pat Po Po Immuns diene Ät uns zi Ko se Tr ko be Vi 109, Infe kutane „P noti ph fa und La sel Wi Po die An tion (65, ropä Über we rüc Ku Deut su er sst m) opismus iz oc ru mme xvir em eu ck ge te ge wu ldle ed ntakt hpoc ti-O iologie kz pe en Die Im Übe sches en ktione Ex Ti ck isc 69). sm nb ke ho ah sc en ,e ve nw 110). er nu nd uführe be Zo zi rde ,d er it us no be is be hl Ge nläs rt vir lre he ine kä rs nk fi et ke re im it .E uppre env ge zu te de onos führ ae hopo nd übri ar Pa and Es nd ch im sche rof nL ue ic ge en Po mit de identif de nv ns nz an inig en ine ga ion“ ra Hun au Die Nag sb eN he ie si nit r, au nw rP ns fe rl ro Hund te ire ge tz sc n( oll nz po si en hilfe de än nd ei xvir bis fw Fl ssi nS iren Lä n. nz ch ei oka at ev Pa nd ag hei nF n( um ne ra ar ät eB er df xvir ,d 96). Ans ne ten de ei (Abb ockenv ei zz on nur Bis ia izie de si zu pe ra on et le en no die en sa sch sen len 4, nt gr rn ie ind lz one en be vo an uH or ult de re ie nW ch zi po us tec be ,w he -A 10% dies rt oße zw de di 10, nK au ub gv re ka men fre dbre nT .9 nh .B es meis di et Re ipl ir .M in xvir n( rw unde eI urde kung ntikör rK inf rW fK ei ur eg nnt ve schei ). sser er si be esi en is en he pli is 27, Ul ic ar nz DN de gis urde it Fa Anschl ch rs iz wa eh ten top en-Infe ie schr ir ht ontakt (27). da uhpocken-Inf ka ra ni tz ea (z ka ze idenz Ve (88, ng sp eF ie rg llbe de und ch A-V en- re er 37, ne nur en pe vo ndte pie to .B niner pr rt ra at ti nS ora rä de ibt et re folgt rk au ie on en nd .D rn al re holo äd tione be wur ric .d nde 109). est ri Nac be 50, zu äu Na ei ir ie pe disc lbe ch re na Inf es pa mi St is ha ktion und ie as as Hu en hte ach ne ße nf n, rA et er lati zi po ru den und ge gisc udien be noc sem tho ric üb nde te 76, n, hd da Ku ,d en izie hb Gen ekt nd n( es ei nde inz nge nie we üb ti bhe iK vh produz lich hte Kr ine pti ge nige er ie hk 4, hpoc zu he Kat lt (O er er Po ve 106, er Er rt ige fe is rt at us oc n( ion e) ust ne er il we ic wie ri 10, nU en ei er rf, en ck en inf rt ei ko li Infe zen üb Or re ung zen te hz hte rF (66). 27). ne ne Or da ei nf we Er en) nh ke rde ,w en ge nnten Ti ier nP er 108, iz Ec sen Ra tho 27, ekt lt nters al vo ns rP us izie t( in ktion re nv vir is th ra ie er pos unde de n. ts es auf th lbe esh ss Infe 17, ge eü po ns ox re ei de Hun ir eh 37, op ta ls seku en rH ap yma rt ion oe en 110). si uc us it n, n( rv .D ric xv rk vir nm al en ch be Re ox be 21, -Infe illoma iv sp ch al ktione be 50, )s au ine hunge und zu en bm de Mensch on ire ht ei ie us te iF ndä rd viru ezif se (78, Kat ei co ginnt ind via tlä 47, ne nv ide Di n- Auc (1 si 76, Mensch kutane rv pr st ktione schl en in ntagio- üc an re si nd nz ze is In 16 vir en nm s oir agno imä in Ly 94). orhe ng 108). ntif one Nor hs ch 106, um- pe fe hi und ). oo- nu da os zu en eu de au en, m- zu en es k- e- re r- i- it n n n n s s r - - - - - r r f 16 Hunde ch de st si mi en Ze Knötc Abb. Üb Diagno Di Kuh rie Ko zei bio tre we dropi schr Haa ge zy kör ode sen Abk moni Be Kat kult ters fe st el ch eh kti ir te tw rW tolo eP it re te ti stäti nigen ch pf, lic pta zi pe lt rF ze uc ure ur rfollik t. t, ns ie la nH ni on nf tr ic pocken- 9 en en sche he rc rimärlä be hunge inf gis (1 he ts Be oc gis te IV et im übe ntr ke ic izie gung n, nd zei he ist ch rw be na 16 en us n( n( wi Feli he hd -inf or ch ib ln ch Fäl azy sero Na n, rd prä se el ka gten ei rt er ten ). De ge na mie us eA 27). 4, Ul eU ie ei ne si en en ), ni se iz ei le nnt Mi ck as top ri si Re Zw pa de ch ge 10, (76). Lä zer lo Pr b. ie tfe ne nw von und ng. nore rt nB Kuh on re Kat en knoti nters Infek Ri Ge kro ra nw gis gel ne rt la äsen si (4, ei sk auf at we ic rn Be 27, en ts sm sik one be et Tr ode ra urde wöhnlich po zen ion ht ch ve Die Ko tw xie linis an 5). urde ii rde ras ra ka ti uc Poc Kat st otzde ge zg de at o, ck 37, eT si ei on ns en ra njunkt ch ti mmun er eht nn ch ode is Die hung n( ke ei und typi ne mg en das on ch ei und roße de zen est pontan t( ch nd 50, ne ei .G au de lfo ne 47, -D ms ev ken histol en aus rN n( 76). ne er Wa schen su rO an f. für rK bei Fi en erm si rm el en ri ulze iv su 76, he Ve In ei 91, Vi 94). ag oll Inne zen eb nd Infektion eg hrsc nd Ein itis ei ntr tue ppri ne vo il at rale rt er Gl immun vi rda zu og at er er 106, Ka ten ne en en ri hopo Fäl ion 109). at sF Ke azy lvie el re itis ie schl ll nv rüc er Kö rha is al he me ch tlic mi ino Der be ek die dma tzen ch le die sB en pa nnz ei n- inl top un er (A ts ru rp k, 109, oba lb er xvir tro us hk ma zy eU töd su nna de tho inz mute Lä bb. ng ko de ichk er te be eg Inf Be ei ße au la sk vo te ne ppri to ch önne n, Ve lic ch gn en el si nters mme sm le im de :O n. sek 110). fg ör n( se ve si ns de nmik te ekt one ne ei ite en hv t. -Infe ha omo rä tz nb ru las pe mi .F at ntuell C. Be an tw rS ink te undä ie nK Be er nde ns rs up nd is uc umf is rc nw ei er Fav ni pir er de be reit ine ionen Auc ch ro ch ch cher) er lus ei nis ie he la ktion ts Hund hung ys te nne re ru nb ei noten rot, at sk ei de ei ne re er äch le uf rh assen sn ine n) temisc Ne iv ch nm ne Pe so nge hd te nung opisc imhä n. ri en is Ein S. Pa rha ed umane rh und ach sK lic rs mT krose vo (76). sei nf de de ie us er Te Wi ni 12 pe one in au lb hF schl er ei se nH izie kra st ontakts ute rP re n( he Katze il lh he au ie ln mG kurz Vi ne er Ta fd we we en el nb .D ulze rb nI eL ne f. nkte Pr ine us ru rt An- au Un- m und ge Zu hy be de em is ni- si V- en il- u- n- In er e- e- ä- ie s- s- t- n e e s - - r - r

H UND /KATZE 311 312 Virale Dermatosenbei Hund und Katze C. Favrot, S. Wilhelm

Therapie überlebt, bleibt aber chronisch infiziert.Bei ungeschütztenWel- ZE penwerdendurchdas Virusnach und nach verschiedene Organe Einespezifische Behandlunggibt es zwar nicht, dochheilendie (z.B.Lymphknoten,Milz,Niere,Leber und Lungen) befallen. AT Läsionenheilenmeist innerhalb weniger Wochen ab.Immunsup- Diemeisten Hunde überleben eine solche Infektionnicht (11).

/K primierte Tierekönnenallerdingssystemische Anzeichen ent- Infektionenmit demfelinenHerpesvirus (felinesRhinotra-

D wickeln.Hin undwiederist eine Behandlung sekundärerbakte- cheitisvirus)geschehen durchden direktenKontakt mitakut infi- rieller Infektionennotwendig. ziertenTieren, mit latent infizierten Katzen während einerReak-

UN tivierungsphase oderüberdie kontaminierte Umgebung. DieIn-

H fektionswege sind üblicherweiseoral, nasal oderokulär und die Replikation erfolgt in derMukosader Naseund derTonsillen. Das Herpesviren-Infektionen AuftreteneinerVirämie istselten. Beilatentinfizierten Tieren ziehtsichdas Virusindie Gangliendes Nervus trigeminus und Ätiologie auch in epitheliale Gewebe zurück(28).

DieseErreger gehörenzuden Doppelstrang-DNA-Viren, die bei zahlreichenSpezies die verschiedenstenErkrankungenhervorru- Kanine Herpesvirus-Infektion aufHaut und fenkönnen. HerpesvirenzeigeneinenTropismus für Nervenge- Schleimhaut webe,Haut, lymphatischeZellenund respiratorisches Epithel so- wiefür Epitheliender Genitalregionen(81). DasCharakteristi- Kanine Herpesvirenzeigeneinenausgeprägten genitalen Tropis- kum allerbekannten Herpesvirenist ihreEigenschaft, latent in ih- mus. NebstAbort, Unfruchtbarkeitund letalen neonatalenSepti- renWirtenzuverharren.InZellen, die VirenimLatenzstadium kämienkönnenHerpesvirenauchvesikulopapilläreLäsionenan beherbergen, werdennur wenige viraleGeneexprimiert. Solche denGenitalien, am Abdomen und aufder Maul- undGenital- Virenbehaltenihre Replikationsfähigkeitbei,wodurchesbei ei- schleimhaut hervorrufen(11, 46, 49). nerReaktivierung zumRezidivkommen kann (81). Beikaninen und felinenHerpesvirenhandelt es sich um α -Herpesviren. Beide Aujeszky'scheKrankheit/Pseudotollwut weisen große genetische Homologienauf (11). InfizierteHündin- nensind normalerweiseasymptomatisch,dochkanneswährend DieAujeszky’sche Krankheit wird durchein um ein α -Herpesvi- derTrächtigkeitoderdurchImmunsuppression zu einerReakti- rusverursacht, daseinenTropismus für Nervengewebe aufweist. vierung derlatentenInfektion kommen. In derFolge wird dasVi- Das Hauptreservoir dieses Virussind Schweine.Hunde infizieren rusübernasale, genitale, okuläre und orale Sekretion freigesetzt sich durchdenVerzehrvon kontaminiertemSchweinefleisch (86). (11). Dieprimäre Replikation beiinfizierten Welpengeschieht in Der Großteil dererkranktenHunde weisteinenintensiven Juck- denEpithelzellen deroronasalen Mukosa. Anschließend erfolgt reiz im Gesicht auf, wobeieszuschwerenselbstinduzierten Lä- die virämische Phase (in Makrophagen). Die Mehrheit derbetrof- sionenkommt(51, 68). Zusätzliche Symptome sind Speicheln fenenWelpen, die durchmaternale Antikörpergeschützt werden, und neurologischeAusfälle. DieKrankheit führtunweigerlich zumTod. Feline Herpesviren-Infektionen

Das feline Herpesvirus1(FHV-1) verursachtbei Katzen Rhino- tracheitis undKonjunktivitis (28).Nach einerInkubationsperiode vonüblicherweise weniger alseinerWoche entwickeln die betrof- fenenKatzen eine schwere Apathie, Fieber,Augen- und Nasen- ausfluss. In derFolge treten Konjunktivitis undRhinotracheitis auf(28). Im GegensatzzuCalicivirus-Infektionenkommtessel- tenzuoralenUlzerationen (24). Es wurden auch fazialeerosive Dermatitidenbeschrieben(Abb.10) (25, 43). BeidurchFHV-1 bedingtenHauterkrankungentretenVesikel, Erosionen,Ulzera- tionen,Krusten und Stomatitis auf. Histopathologisch istneben denulzerösen und verkrustetenLäsioneneine ausgeprägte eosi- nophiledermale Infiltration zu erkennen. EinigeKeratinozyten weisen intranukleäre Einschlüsse auf (43, 44). DieDiagnosewird üblicherweise mittels Immunhistochemie und/oderPCR-Ampli- fikation derviralenDNA gestellt (97).Ebensoist eine kulturelle Abb.10 Feline Herpesvirus-Dermatitis(Abb.: L. Beco) Anzüchtungdes Virusmöglich (28).

17 Pr Di Infe Di Re Er He Fel si übe Ät zw schen ko de Fäl Gamm nus ke fi notr er si ne ten le nis ge zy Ha zi we ei ma an Immunit fe Antibi Pe In-vitro- hil fe 82, mple nd sch, nz eh ne wäh kti ron, äsen it rF re n. tä de nz ythema ntamini eÜ es freic ei nn utt ch te iologie le rd 83, Impfunge Le rpe ach ine re en ei tr ei Tr eh on un Fi re yklo xfol rn Az IV en Vi kti umor pu si ne nt ne in we en oti be nti x- L-L zv nk eb Exo otzde ov aret al si dd h. äu -a 95). ei ee re yklo bl ode Lite sE rt ät Vi wird. svi de en ch on iati rt ka er vir sh vir Sp ode tis sso typi on fi nv am Anti iz er er ntw ys ra ro mul iren-I zy ra en re rR ge al ge und ie au ei ve us rd ts An mk zu en ode vir gung ra vir zi in Sch ry re ns ri Fr na tos bd er sche es nh bore rt ch nE be ed fd eg ei ). vir nD ie tu ti rK Im tw pla -T em ursa tikö nl und us er the n (A an Ly for rC n. el e, uf zu rt ri FI er ie rh we el ab as rkra al ic he )u ch en .A ontr er imitie nw zen Ge se da ne Ap Be bb Vi mph de vo Re (54, ido me eM we ke ly rp ei re en äu ch nfektionen Fe bo ra ma ma te nd nti Hau nde .1 Kr sf mph sz ne ge nkun iK ne er lte optose nduz ak ta fi de t: pie nige er t. LV oll fi vine zu ist ad tos da ge ed 1) gü rS ns re 113). uI ank vir de ns rt ti el Di da rf ine at me re -V ed tv vie en nI mult oz en sf ni ika at ei ine zu (77, nfe zen e. ei no ge pontanhe el ier sf em sL Ta be ic er so Kör he rK ir zd hr ne opa ytä ne er el ine nfe nd Hi ru hk nb und me ex än us nL rF ge wi te kti el Be ei de we it ine ak rd ei at ng azu 80). sek ha st re pe ine pe . im de nH ne thi kti ine er ei if ed im utane nte rv one eL st na iK ze tof lpe olo eu ra Sat be Immu orme rfl ri ru rk en munol Kat undä e. In on Ep V- ei ie ch wu or er st ohne vo mente Immu Le er Ein kä at nb n( il nb gisc nge üs wi te ne el Sch is The asso de linis pe ung id rin zen (16). ei rlie zen mi ihr Ve uk rde rf si litos ch eA est ei rl 28). re er rf ne sv pa nd ns ace- gk he af ev og äm rä He er ra en zei Kat ge nb zi mis nd ll ch eh wu (77, at ir olgende we rI ar be in ef ft zy ind ei nde pie e. ir en is Unte ie Die us nde er en rp ie en gten ef tz Ab In nfe Der iz te Ta es us ak de zen rde ch rde klo rt .E Im vir st din -In esv te iz St ie nk ude mit ru 80). je eD ge te da rv na eA nA en Ve kti Fäl er nI ge ie nv st vir ni nz rsuc ma do we udien us nge fe ei rie ihr ir önne sI hme st nz na nA et md rbe gt α on schi er Impfung ritt vir ne ktio ns ,G ch le ntw rt us or lpe titis en nfe lle (F er -u mmun eW ni .D ch hunge ma ik na ge nF -s sser us mi us al ie inä eL nicht du nI be an nk nd ne el Sy eht or nb kti (3, de nF pe ,l ie ei sam le tos sagen Mö tH V, (F irk kuter al im zy ta rm nur nfe ne mpto rP omm ung, md ω on be zi rch ympho- me ei lf or 38, IV sy en nz pri un Genus klo -Inte er fi sp tz uf rR ge gli ed nf Men- ke me ühr kti rä ,G st is sche urc kurz pe si gi ei mär ezi- und we tr al iz vir be ch va 64, em me te ine die en In- hi- im nd on g- a- e- s- te ls r- i- i- n h n - - - i , , 18 (16). nicht mark be in nicht, knoten, 43, ko 107). (Ab Abb Ku Retro Über Klinik Ka Auc su Neopl de de mind mute ne ne si kli tös plas B-Z au one ppre in ck nA nk nL ch nnte nisc eP de tan b. tzen 58). .1 ien el hk tw tw :C sy est nk Über ymphom rF ic önne Fe Int vir l-L rt 1 as la dies be ssi he es vire utane htv st ik .P qu LV Fe urde we er Ha ie önne be er (FeL äm fa ymp olge em on el Ly Feli Sy ros nb de eI LV nj es, essan irä lle -induz ie di ite ut is n- we is mp ,d n( li mpto nfe ef ne ed t) -A nu T- ine ei ns mis ho ch ri ch ode Kn eg übe en V) ass rde ar Ze 24, sH oka oc ser ,a ntige me ho kti oli te ne mT eS en nd ch oten fe ie er rD ma rw hn ll- rl nd bhä m. erpesvi olo on tär rt ozi le ympt kra en 48). au pit un ine ei Ly dor ei es än no tik ünnda ic ie Re zu ld gisc sf ode Be ngig te und he se nk ie mph ht Kat se ge Ly ührl lok tp rt La ie pli is ome de li be ib gibt en rt re rm au ei rus-i Erk hF mph otr ser ts zen Ulz te er r- ka al sei eH rm is vo ome lt. iche sg nor nten eh Ze ult ek ope si eL au es Tu ti ra zu Vi nd tige DN esch nduziert er om, st on gibt rv in rale it V- male nkunge autv ize f. au mor utane si at ie 75% Be er An und Hi n, ne ode ar be tr nd ion re ch Mi ntr los Le Der ri an es Re ia ga en ka n. st rw iv ch hä er Be isc en rn lz uk sen be ach ol de si es Re ma nichte ti nn di ak irä ei nn vir Das ,K te uf en ve änder og ri lu ha äm rL ic ve (A se to Eryt tion pli ge mis da ch we ig al vo t, ic ht nK is se rs nd noc uft ymphom bb ie äl es wie ka ht sV ch te bet Fe r( pe rde pit eP nb hem pr te C. ch .1 un de re at ti Kap umf üb nä he LV we 16). rm od re ir rof he sen ei Fav ungen en on zen n( ten ub sI 2). dK us he er nmark, am -pos Katzen. Hund uk rde li an fe rot, li asst si ode 107). mmun de otr rb (87, da Fe Im Hä (107). ka we noc ti ev ne en da ul nd sV ve iti S. sK LV ope esp ti ra tifor und to nti uf rde vo on er Ti he one v( ie 107). Wi bei An Infe -pos ir be ig ythema noc ge sy Ly roc rlie er Di de se nmark- n( Ka Katze me us lh 56, T- Da ra n( st tr ne el mph rg kuta- eL Neo- ktio- he wi he ge iti tze et em ve 56). m und uc Die zu- 24, 87, nten ve n. n- rd n- ä- ar r- h n s - -

H UND /KATZE 313 314 Virale Dermatosenbei Hund und Katze C. Favrot, S. Wilhelm

Hautveränderungen assoziiert mithoher Rate an ZE FeLV/FIV-positiven serologischen Befunden

AT Plasmazell-Pododermatitis. Bei dieser sehrschmerzhaftenEr-

/K krankung sind einodermehrere metakarpaleodermetatarsale

D Pfotenballen betroffen. NebeneinerSchwellungtretenmit der Zeit auch Ulzeraauf. Die Diagnosewird mittels zytologischer

UN oderhistologischer Untersuchung gestellt. Häufig sind FIV- posi-

H tive Katzen betroffen(39, 88). Die Ursacheist bis heute unbe- kannt. Glukokortikoide,Doxycyclin(10 mg/kg KM/Tag)und auch chirurgischeEingriffe werdenals Therapiemöglichkeiten vorgeschlagen(19, 39, 40, 84). Rezidivierende Plasmazell-Polychondritis. Hierbei handelt es sich um einesehrseltene Krankheit,die durcheine asymmetri- sche Schwellungder Pinna,gefolgt voneinerpermanentenEin- drehung oderDeformation desOhrknorpels charakterisiert wird Abb. 12 FeLV-induziertes kutanes Lymphom (8, 29, 88). Auch diesePlasmazell-Polychondritis tritt häufig bei FeLV-oderFIV-positivenKatzen auf. Hautveränderungen assoziiert mitImmundefizienz FIV- positiveKatzen leidenhäufigunter verschiedeneninfektiö- sen Krankheiten wiezum BeispielAbszesse, Pyodermien, Der- matophytosen,Kryptokokkoseund Demodikose (86). Diagnose retroviralerInfektionen

DieDiagnoseeinerFeLV- Infektion beruht aufdem Nachweis des FeLV-Kapsidproteins p27, normalerweise mittels ELISA.Virämi- sche Katzen weisen einpositives Testergebnis auf. Beitransient virämischenoderlatentinfizierten Tieren kann dasResultat falsch negativsein(16). Zusätzlichbestehtdie Möglichkeit, mit Gewebe vonKatzen,einschließlichdes vorherformalinfixierten und paraffinisierten,eine PCRdurchzuführen (16). Abb. 13 FeLV-induzierteRiesenzell-Dermatitis Prävention und Therapieeiner FeLV-Infektion

ZurPrävention einerFeLV- Infektion wurden verschiedene wirk- Riesenzell-Dermatitis. EineRiesenzell-Dermatitis wurde sameImpfstoffe entwickelt. Beiallenhandelt es sich um Tot- bei verschiedenenanFeLVerkranktenKatzen beschrieben(22, oder Vektorvakzinen (basiertauf genetisch entwickeltemProtein 36). Klinisch lagenErosionen, Ulzerationen, Krusten, Schuppen gp70). undAlopezievor (Abb.13). Die Tierezeigten Juckreiz, systemi- Zur Behandlung vonFeLV- induzierten Lymphomenund sche Symptome(Fieber,Anorexie, Apathie) und auch hämatolo- FeLV-induzierter Knochenmarksuppression existierenverschie- gischeVeränderungen(Anämie).Alle betroffenen Katzen wurden dene Arbeiten,auf die im vorliegendenArtikel nicht weitereinge- euthanasiertoderstarben innerhalb weniger Wochen.Das Auffäl- gangenwird. ligste derhistologischen Untersuchung warenein Keratinozyten- Synzytium und eine Dyskeratose. BeiallenbeschriebenenFällen konnte FeLV-Antigenoder-DNAnachgewiesen werden. Zurzeit istnochungeklärt, ob es sich beidieser Synzytium-Formation um Staupe und “Hard PadDisease” eine neoplastische Anordnung handelt. KutaneHörner. Im Zusammenhang mit FeLV-Infektionen Ätiologie und Pathogenese wurden auch kutane Hörner beschrieben. In derdarunter lie- gendenEpidermiskonnte virales Antigennachgewiesen werden Staupe wird durchein RNA-Virus derFamilie Paramyxoviridae (86). verursacht(70). Derbreite Einsatzvon wirksamen Impfprogram- menhat zu einerdrastischen Reduktion desAuftretens dieser

19 sche tra si kutane top Ohn Diagno Di Klinik Kr va oder im An chen chiti ke zell au we die tos Sym tane neurol und obe änd zi ge es oder ten ten au kurzem über ge wi de Vo da Ve Vi si tino vo ve si ve ch nd rka wi ch ria ne nnt ff sam rS ss ru ea nK si ei kt ankh rda munkr rde rha ed bal ore la Infe Hu er en re Die zy ch sS s, ep pt ss ,D na oll es ev so Sym ei bl mV Der sm mit ohne na kut rfolgt bei ei un ynthes nA ei es nP uung er len nde man xi og te om te nden nen en be nM er entuell um ch mit nv Bron assen ikul yndrom kt äm so eit ne urchf bestä at gen e, at Ans nd np eu Pf is ma ei er Hu oba te apeln io an (Hard we ea ptom ns Gr ino digi is we De ch se tlic Hy Vo ir Wo otenba ei ne un gef an st da mw ei ni nd är sp ,K ch rimäre khe auc es ei titis chopneu nden ale en tec is öße uftre e( ne ra rde de ch tig al hy ne zy vo re er rg ife uungs perk nF ät nk tale he en en ch üh sub rämpfe) l) kt, auch ea eg rz Ep „O hi te kung es un drat it te t, de Ana er Pa eu lls rs n. ), Eb st Sp fe ,o ,d e, or rt und ten tw und an ep dass eu nv chichte (P ,j ll te uft auf, nd at dP aku ith utl erat tän yt ld dD Har tra hldi Hype nu .N ko kul Mit me en ät en ass wo at ed nv nd ithe tr ion em mnes opat er er nd mo du re Do el folge er iche lis (71, io di nge ak e. er ustel su te is Fo ach ose nur nd äre oc das nw be ag di tels in an de sf ten ira ei phigus, en n, rch er gen ge Ha ea folgt ie Te lia de gE tk ni rk Fo eb nd es rm hi hn ie ne nos tw n( nit eg re seku respirat le ek ke vo or se ür 86). Sym des e), Vi Hu sel er il wi nT ne dies ut le sc (71). omm rm de im Im nce nd nR sz or Im ne ei de we ic e, ,A al 119). tr tho- nnze nS at (Le ru ti her an vo ev ei te Ge in nd rS nt ons an mu ht Pla eE tlic ga pe urol zi er sP einer os Au ndä der pfs ue pt sn Be er is bb phal ne ni ep nd rk Lu dec tau er ntz t( si or Fe de st we imme tig ek ta om e) ode die nol ich Ef num An .1 la hi orische il ahezu Infe id ge om chut lika ine rkra en ntr ro tw re og Da Sta ie up 70). pus ,P vo hlen rR len pe, nu om kt ist be itis fek 5) önne st e( nu net in Infe en Bod og nl tikö te rp is „Har mt an Infektion rV e( ;e rübe ap er upe tion ktion mn (33, zv nkun vir nasal tes ), (86). eg ch r. “) Ko es tk (70). is be können anifes In ymphatisc ar De Ha ukleä nd de bei 70). in dies (71). illoma (41). er kt Neur Mal ch rp ie el na eS tin erf al und/ Sym fe fa ak je nj asal onnt is dP rg n. te io in ut rw s) di 35). er Ha er eE ktio lle all e( g( do übe unkt „Hard td ale zu er üg uc eA ilun eh nm ve Die ym rekt nutr de His ti re in ad Bei Na Te ol oder ei em pt ab at Ab en ut. n: ch urch eb hs en. eren ins rän en fa tos er nE bei Der rd nL nd ko nti te og pt oti om Sympt iv chn de it gd Di hä ch topatholo er an miliä it Inkuba ke Re b. de wi ynz Ers it ch re We om is eo de he itis ntr Pa is en ion kör dem sche er ymphknot nK ngig. .I der ins sease“ e( ma 14) Zus Fi einiger nne her azi St Bef es ik spir ch lüs ch Ve ru nO dD nv Pr tk lp e( oll de ,I yti ba Hu eber Re en aup ch pe ma re nche (z tig, ngen Vi eS dophil ome er en rla se oli am und ri n. all rh Sta Hy ürz sal ni iere iele at al Ve ti is lüs inkr rg r, spir st at doz wu ru Be to idio ee ympt on uf cht ion ,A eR kann fe ea nicht al en, me gisc ep ka vo ino an en au rha upe pe Zeit li nT lo si sen nF der n, ri sA id hande rde rkra se (E ykl at ch ist esp pat at nn da en gi st nN ch en en ie rk nhang pa er Br Ke nd nac zy lten “t ion vi ier ohne rbr he äll oku- ra sc ,m er ome Pfo- sen- uto- ue er si itis ung lie vo vo hi thi- im da on- nk- on vo de zy so die ru te ri ra er kt, en en he en en ch h- a- e- s- s- e, r- rt lt tt it n n n ß s s - ) r r - - - - - 20 me zu Pa Ab Be Pa Da St Abb. ex pie pe de ne ei vier we im zw ne is rD ri mK rv muni b. ie rde ei au ve rv be mente im ten tie rv ov ma Im 14 15 rw ine ia orha st ni rt pe olostr iro Vi gn muni ov Fa eht li si en rz nd und er Staupe St re se mA os ll ll nde impf de ,i we iros eS ung en au nd est vo ei um t. st si nur ns bs pe imon ne er tud zu en Mat mA ei el tand ung. un zwi vo ra ind ne lun rz bei be die Hu e ie er ei mW uft at im nz kuten ,w proph mH ge gh sch nd tv na ige Mö vo Im re Hu ei er le er er nicht en n nv und: el ten nD gten, Allgeme de gli nd: we an Antikör ylaktisc pe Infektion ie na de ch ei ge nde na og nasal Hard rW au ne Vi rs ke da zo uc ge rale ufge sr ten sE it oc ge ss ha ech ei inen he pe eH Pad wurde ei he n( ei ry Der ch Pr nde nomm r, ne st Im ke ne ng the ype 26, en odukten die en Di ma wird rs ine re pfung Va ds ma seas ei rke to ympt im 57). Te tr und mpf kz en se ch ef an ei ra multif ch nb inationen Zu e ütze fe C. nW we sp tose unumgänglich. t. omatisc nur nik 16. kti ei Fav sam la rde nk el Hund en ve zen or rot, die pe Le n, me an me wi Be ta S. Hüllpr zu be he ve und n. mi nh ro eR handlu Wi (EM rG ns rhinde nT De ti an de Katze lh T- wo na ru gm el sh he rm PC )b otei- m Ex ch kti nd- al ra ng rn e- R it it b e - - -

H UND /KATZE 315 316 Virale Dermatosenbei Hund und Katze C. Favrot, S. Wilhelm

8. Bunge MM, Foil CS, Taylor HW,GlazeMB. Relapsing polychondritis in a cat. JAmAnim Hosp Assoc 1992; 28: 203–6. ZE 9. Campbell KL, SundbergJP, Goldschmidt MH, Knupp C, Reichmann ME. Cutaneousinverted papillomas in dogs. VetPathol1988; 25 (1): 67–71. AT 10. Canese MG, Lavazza A, Massone A, GaleanoF,Boldini M. Feline poxvirus infection. Acase report. Schweiz ArchTierheilk 1997; 139 (10):454–7. /K 11. Carmichael LE, Greene CE. Canine herpesvirus infection. In: Infectious D Diseases of the Dogand Cat. GreenCE, ed. Philadelphia: Saunders 1998; 28–32.

UN 12. CarneyHC, England JJ,Hodgin EC, WhiteleyHE, Adkison DL, Sundberg Abb. 16 JP.Papillomavrius infection of aged Persian cats. JVet DiagnInvest 1990; 2: H Kutane Parvovirose: 294–9. Ulkus am Pfoten- 13. Carpenter JL, Kreider JW,AlroyJ,Schmidt GM. Cutaneousxanthogranulo- ballen ma and viralpapillom on aeyelid of acat. VetDermatol 1992; 3(4/5): 187–90. 14. Chambers VC, Evans CA. Canine oral papillomatosis. I. Virusassayand observations on the various stages of the experimental infection. Cancer Res richtet(23). BeiEMwerdendie Antigene derKeratinozytenmo- 1959; 19: 1188–95. difiziert,worauf durcheine anschließendeLymphozyteninfiltra- 15. Collier L, Collins BK. Excision and cryosurgical ablation of severe periocu- tion eine selektiveApoptose induziert wird (1, 55). Neben den lar papillomatosis in adog.JAm VetMed Assoc 1994; 204 (6): 881–3. 16. Cotter SM.Feline viral neoplasia. In: InfectiousDiseases of the Dogand Cat. klassischen Symptomen einer Parvovirose zeigte derWelpe vesi- GreenCE, ed. Philadelphia: Saunders 1998; 71–84. kulopapuläre kutane und mukokutaneVeränderungenimGesicht, 17. Czerny CP,Eis-Hübingeer AM, Mayr A, Schneweiss KE. PfeiffB:Animal Maul,anderVulva, an derventralenKörperhälfte, denPfoten(mit poxviruses transmitted from cat to man: current event withlethal end. JVet Med B1991, 38: 421–31. fokalenUlzerationen an denPfotenballen)und überden Kno- 18. DenmanAM, Rager-Zisman B. Viruses and autoimmune diseases–adapting chenvorsprüngen (Abb.16). Erythematöse Plaqueswaren am Koch's postulates. AutoimmunityRev 2004; 3(5): 355–61. Kinn und Abdomensichtbar (23). Da der Hundnicht überlebte, 19. Dias Pereira P, Faustino AMR. Feline plasma cell pododermatitis: astudy of konnte die Entwicklung derHautveränderungennicht weiterbe- 8cases. VetDermatol 2003; 14 (6): 333–7. 20. Egberink HF,Berrocal A, Bax HA, vanden Ingh TS, Walter JH, Horzinek obachtetwerden. Histopathologischwurdeneine regenerative ne- MC. Papillomavirus associated skin lesions in acat seropositive for feline krotisierende Enteritis,vereinbarmit Parvovirose, und kutane Lä- immunodeficiencyvirus. VetMicrobiol 1992; 31 (2–3):117–25. sionenähnlich einemErythema multiforme major beschrieben. 21. Eis-Hübinger AM, Gerritzen G, Schneweis KE, PfeiffB,Pullmann H, Mayr DieHautveränderungensetzten sich auseinerlymphozytärenIn- A, Czerny CP.Fatal cowpox-like virus infection transmitted by cat. The Lan- cet 1990; 336:880. terface-Dermatitis,Apoptose und Satellitose, verteilt über die ge- 22. Favrot C, Grest P, Meli M, KiparA, Wilhelm S, ZauggN.P-32 Giant cell der- samte Epidermis,sowie ausapoptotischenKeratinozytenmit in- matosis in aFeLV-infectedcat: detection of the virus in infected skinbyim- tranukleären basophilen viralen Einschlüssen zusammen. Dieim- munohistology and PCR. VetDermatol 2004; 15 (s1): 50. 23. Favrot C, OlivryT,Dunston SM, Degorce-Rubiales F, Guy JS. Parvovirus in- munhistochemischeUntersuchung bestätigte dasVorhandensein fection of keratinocytes as acause of canineerythema multiforme.VetPathol vonkaninemParvovirus CPV- 2inder Hautund in denintestina- 2000; 37 (6): 647–9. lenDrüsen. 24. Favrot C, Wilhelm S, Mezler K, Grest P, Kipar A, Meli M. Acase of FeLV positive cutaneous T- cell lymphoma in aserologically FeLV negative cat. Vet Danksagung Derm2005; (16): 196. 25. Flecknell PA,Orr CM Wright AI, Gaskell RM, KellyDF. Skin ulceration as- Wirdanken IngridWilhelm-Dotterherzlich für die Durchsicht des Manuskripts. sociated with svirus infection in cats. VetRec 1979; 104 (14): 313–5. 26. Frisk AL, Konig M, Moritz A, Baumgartner W. Detection of canine distemper virus nucleoprotein RNAbyreverse transcription-PCRusing serum, whole blood,and cerebrospinal fluid from dogs with distemper.JClin Micro- Literatur biol 1999;37(11): 3634–43. 1. Assier H, Bastuji-Garin S, Revuz J, Roujeau JC. Erythema multiforme with 27. Gaskell RM, Gaskell CJ,Evans RJ,DennisPE, Bennett AM, Udall ND, mucous membrane involvement and Stevens-Johnson syndrome are clinical- Voyle C, Hill TJ.Natural and experimental poxvirus infection in the domes- ly different disorders with distinct causes. Arch Dermatol 1995; 131 (5): tic cat. VetRec 1983; 112 (8): 164–70. 539–43. 28. Gaskell RM, Dawson S. Feline respiratorydisease. In: Infectious Diseases of 2. Baer KE, Helton K. Multicentric squamous cell carcinoma in situ resembling the Dogand Cat. GreenCE, ed. Philadelphia: Saunders 1998; 97–105. Bowen'sdisease in cats. VetPathol1993; 30 (6): 535–43. 29. Gerber B, CrottazM,vonTscharner C, Scharer V. Feline relapsing polychon- 3. Beaumont SL, Maggs DJ,ClarkeHE. Effects of bovine lactoferrin on in vi- dritis: twocases and areviewofthe literature. JFeline Med Surg 2002; 4(4): tro replication of feline herpesvirus. VetOphthalmol 2003; 6(3): 245–50. 189–94. 4. Bennett M, Gaskell CJ,BaxbyD,Gaskell RM,KellyDF, Nadoo J. Feline 30. Ghim S, Newsome J, Bell J, Sundberg JP,Schlegel R, Jenson AB.Spontane- cowpoxinfection. JSmall Anim Pract1990; 31: 167–73. ouslyregressing oral papillomas induce systemic antibodies that neutralize 5. Bennet L, Gaskell RM, BaxbyD.Feline poxvirus infection. In: Infectious canine oral papillomavirus. Exp Mol Pathol 2000; 68 (3):147–51. Diseases of the Dogand Cat. GreenCE, ed. Philadelphia: Saunders 1998; 31. GodfreyDR, Blundell CJ,EssbauerS,Pfeffer M, Shearer DH, Rest JR, Ba- 110–1. kerJFM. Unusal presentations of cowpoxinfection in cats. JSmall Anim 6. Bernays ME, Flemming D, Peiffer RL Jr.Primarycorneal papilloma and Pract2004; 45: 202–5. squamous cell carcinoma associated with pigmentarykeratitisinfour dogs. 32. Greene CE, Appel MJ.Caninedistemper. In: Infectious Diseases of the Dog JAmVet Med Assoc 1999; 214 (2): 215–17. and Cat.Green CE, ed. Philadelphia: Saunders 1998;9–22. 7. Brooks-Mill A, Campbell KL. Concurrent hypothyroidism, IgM deficiency, 33. Gröne A, Engelhardt P, Zurbriggen A. Canine distemper virus infection: impaired Tcell mitogen response and multifocal cutaneous squamouspapil- Proliferation of canine footpad keratinocytes. VetPathol2003; 40 (5): lomas in adog.Can Pract 1992; 17: 15–21. 574–8.

21 317 Virale Dermatosenbei Hund und Katze C. Favrot, S. Wilhelm

34. Gröne A, GroetersS,Koutinas A, Saridomichelakis M, Baumgartner W. 60. Le Net JL, Orth G, SundbergJP, CassonnetP,Poisson L, Masson MT,George Non-cytocidal infection of keratinocytes by caninedistempervirus in the so- C, LongeartL.Multiple pigmented cutaneous papules associated with ano- ZE called hardpad disease of caninedistemper. VetMicrobiol 2003; 96 (2): velcaninepapillomavirus in an immunosuppressed dog. VetPathol1997; 34 157–63. (1): 8–14. 35. Gross TL, Fau-Brimacomb BH. Multifocal intraepidermal carcinoma in a 61. Li J, Sun Y, Garen A. Immunization and immunotherapyfor cancers invol- AT doghistologicallyresembling Bowen'sdisease in Man. Am JDermatopathol ving infection by ahuman papillomavirus in amouse model. Proc Natl Acad 1986; 8(6): 509–15. Sci USA2002; 99 (25):16232–6. /K 36. Gross L, Clark EG, Hargis AM, Head LL, Haines DM. Giant cell dermatosis 62. Lozano-Alarcon F, Lewis TP 2nd. Clark EG, BradleyGA, ShupeMR, Har- D in FeLV-positive cats. VetDermatol 1993; 4(3): 117–22. gis AM. Persistent papillomavirus infection in acat. JAmAnim Hosp Assoc

37. Groux D, Degorce-Rubiales F, Capelli JL. Feline poxvirosis, areportconcer- 1996; 32 (5): 392–6. UN ning twocases. Prat Med ChirAnim Cie 1999; 34 (3): 215–29. 63. LucroyMD, Hill FI, Moore PF,Madewell BR. Cutaneous papillomatosis in 38. Guaguère E, OlivryT,Delverdier-PoujadeA, Denerolle P, Pages JP,Magnol JP. adog with malignant lymphoma following long-term chemotherapy. JVet H Demodexcati infestation in associationwith felinecutaneous squamous cell DiagnInvest 1998; 10 (4): 369–71. carcinoma in situ:areport of five cases.Vet Dermatol 1999;10(1): 61–7. 64. Maggs DJ,Nasisse MP,Kass PH. Efficacyoforal supplementationwith 39. GuaguereE,HubertB,Delabre C. Feline pododermatoses. VetDermatol L-Lysine in cats latentlyinfectedwith felinesvirus. Am JVet Res 2003; 64 1992; 3(1): 1–12. (1): 37–42. 40. GuaguereE,Prelaud P, Degorce-Rubiales F, MullerA, HubertT,Lebon S. 65. Mayr A, Lauer J, CzernyCP. Neue Fakten über die Verbreitung vonOrtho- FC-23 Feline plasma cell pododermatitis: aretrospective study of 26 cases. poxvirusinfektionen. Prakt Tierarzt 1995; 7: 961–7. VetDermatol 2004; 15 (s1): 27 66. MeyerH,SchayC,Mahnel H, Pfeffer M. Characterization of orthopoxviru- 41. Haines DM, MartinKM, ChelackBJ, Sargent RA, Outerbridge CA, Clark ses isolated from man and animals in Germany. Arch Virol 1999;144 (3): EG. Immunohistochemical detection of canine distemper virus in haired 491–501. skin, nasal mucosa, and footpad epithelium:amethodfor antemortem diag- 67. Miller WH, Jr., Affolter VK, Scott DW,Suter MM. Multicentric squamous nosis of infection. JVet DiagnInvest 1999; 11 (5): 396–9. cell carcinomas in situ resembling Bowen'sdisease in five cats. VetDerma- 42. Hanna PE, Dunn D. Cutaneousfibropapilloma in acat (feline sarcoid). Can tol 1992; 3(4/5): 177–82. VetJ2003; 44 (7): 601–2. 68. Monroe WE. Clinical signs associated with pseudorabiesindogs. JAmVet 43. Hargis AM, Ginn PE, Mansell JEKL, GarberRL. Ulcerative facial and nasal Med Assoc 1989; 195 (5): 599–602. dermatitis and stomatitis in cats associated withfeline svirus1.Vet Derma- 69. Muller T, Henning K, Kramer M, Czerny CP,Meyer H, Ziedler K. Seropre- tol 1999; 10 (4): 267–74. valenceoforthopoxvirus specific antibodies in red foxes(Vulpes vulpes) in 44. Hargis AM, Ginn PE. Feline svirus 1-associated facial and nasal dermatitis the Federal State Brandenburg, Germany. JWildl Dis 1996; 32 (2):348–53. and stomatitis in domesticcats. VetClin North Am Small Anim Pract1999; 70. MurphyFA, Gibbs EPJ,Horzinek MC, StuddertMJ. Pathogenesis of viralin- 29 (6): 1281–5. fection. In: VeterinaryVirology.MurphyFA, Horzinek MC, StuddertMJ, 45. Harwood CA, ProbyCM. Human papillomaviruses and non-melanoma skin eds. San Diego: Academic Press1999; 167–9. cancer.CurrOpin InfectDis 2002; 15 (2):101–14. 71. MurphyFA, Gibbs EPJ,Horzinek MC, StuddertMJ. Paramyxoviridae. In: 46. Hashimoto A, Hirai K, Fukushi H, FujimotoY.The vaginal lesions of abitch VeterinaryVirology.MurphyFA, Horzinek MC, StuddertMJ, eds. San Die- with ahistoryofcaninesvirus infection. Japan JVet Sci 1983; 45: 123–6. go: Academic Press1999; 423–5. 47. Hawranek T, TritscherM,Muss WH, Jecel J, NowotnyN,Kolodziejek J, Em- 72. Nagata M, NankoH,Moriyama A, Washizu T, Ishida T. Pigmentedplaques berger,M,SchaeppiH,Hintner H. Feline orthopoxvirus infection transmit- associated withpapillomavirus infection in dogs: is this epidermodysplasia ted fromcat to human. JAmAcad Dermatol 2003; 49 (3): 513–8. verruciformis? VetDermatol 1995; 6: 179–86. 48. HayesKA, RojkoJL, Mathes LE. Incidence of localized feline leukemia vi- 73. Nicholls PK, KlaunbergBA, Moore RA, SantosEB, ParryNR, GoughGW, rusinfection in cats. Am JVet Res 1992; 53 (4): 604–7. StanleyMA. Naturallyoccurring,nonregressing canineoral papillomavirus 49. HillH,Mare CJ.Genital disease in dogs caused by caninesvirus. Am JVet infection: Host immunity,virus characterization, and experimental infection. Res 1983; 35: 669–73. Virol 1999;265 (2): 365–74. 50. Hoare CM, Gruffydd-Jones TJ,BennettM,Gaskell RM,BaxbyD.Cowpox 74. Nicholls PK, StanleyMA. The immunology of animal papillomaviruses. Vet in cats. VetRec 1984; 114 (1): 22. Immunol Immunopathol 2000;73(2): 101–27. 51. Howard DR. Pseudorabies in dogs and cats. In: Current VeterinaryDermato- 75. Nicholls PK, Moore PF,Anderson DM, Moore RA, ParryNR, GoughGW, logy.Kirk, RW,ed. Philadelphia: Saunders 1986; 116–9. StanleyMA. Regression of canineoral papillomas is associated with infiltra- 52. HowleyPL, Lowy DR. Papillomaviruses and their replication. In: Field's Vi- tionofCD4+ and CD8+ lymphocytes. Virol 2001;283 (1): 31–9. rology.Knipe DM, ed. Philadelphia: Lippincott, Williams &Wilkins 2003; 76. NowotnyN,FischerOW, SchilcherF,Schwendenwein I, LoupalG,Schwarz- 2197–2264. mann T, MeyerJ,HermannsW.Pockenvirusinfektionen bei Hauskatzen: kli- 53. Hubert B, Magnol JP.Hepatozoon canis,afortuitous or apathogenicagent in nische, pathologische, virologische und epidemiologische Untersuchungen. canine dermatology.Vet Dermatol 2002; 13: 224–7. Wien Tierärztl Mschr1994; 81: 362–9. 54. HuffJC. Erythema multiforme. Dermatol Clin 1985; 3(1): 141–52. 77. OlivryT.Newly recognized felinedermatoses:Selected topics. In: DVMFall 55. HuffJC, Weston WL, Tonnesen MG. Erythema multiforme: acritical review Seminar,1997, KeyWest. of characteristics, diagnostic criteria,and causes. JAmAcad Dermatol 1983; 78. Pfeffer M, Pfleghaar S, vonBomhard D, Kaaden OR, MeyerH.Retrospec- 8(6): 763–75. tive investigation of feline cowpoxinGermany. VetRec 2002; 150 (2): 50–1. 56. Jackson ML, Wood SL, Misra V, Haines DM. Immunohistochemical iden- 79. PranskySM,Albright JT,MagitAE. Long-termfollow-up of pediatric recur- tification of Band Tlymphocytes in formalin-fixed,paraffin-embedded rent respiratorypapillomatosis managed withintralesional cidofovir.Laryn- felinelymphosarcomas: relation to feline leukemia virus status, tumor site, goscope 2003; 113 (9): 1583–7. and patient age. Can JVet Res 1996; 60 (3): 199–204. 80. Prost C. Acase of exfoliative erythema multiforme associated withsvirus 1 57. Koutinas AF,Baumgartner W, Tontis D, Polizopoulou Z, Saridomichelakis infection in acat (abstract). VetDermatol 2004; 15 (Suppl.1). MN,Lekkas S. Histopathology and immunohistochemistryofcaninedis- 81. Roizmann B, Pellett PE. The Familysviridae: abrief introduction: In: Field's tempervirus-inducedfootpad hyperkeratosis (HardPad Disease) in dogs Virology,Vol. 2, 4th ed. Knipe DM, HowleyPM, eds. Philadelphia:Lippin- withnatural caninedistemper. VetPathol2004; 41 (1): 2–9. cott, Williams &Wilkins 2001; 2381–97. 58. Kovacevic S, Kipar A, Kremendahl J, Teebken-Schuler D, Grant CK. Immu- 82. SandmeyerLS, Keller CB,Bienzle D. Effects of interferon-alpha on cytopa- nohistochemical diagnosis of feline leukemia virus infection in formalin- thic changes and titers for felineherpesvirus-1 in primarycultures of feline fixedtissue. Eur JVet Pathol 1997; 3(1): 13–9. corneal epithelial cells. Am JVet Res 2005; 66 (2): 210–6. 59. LeClerc SM, Clark EG, Haines DM. papillomavirus infection in association 83. SandmeyerLS, Keller CB,Bienzle D. Effects of cidofovir on cell death and with feline cutaneous squamous cell carcinoma in situ. In: AAVD/ACVD replication of feline herpesvirus-1 in cultured feline corneal epithelial cells. Meeting, 1997. Am JVet Res 2005; 66 (2): 217–22.

22 318 Virale Dermatosenbei Hund und Katze C. Favrot, S. Wilhelm

84. Scarampella F, Ordeix L. FC-22Doxycycline therapyin10cases of feline 104.TeifkeJP, Lohr CV,Shirasawa H. Detection of canine oral papillomavirus- plasma cell pododermatitis: clinical,haematological and serologicalevalua- DNAincanineoral squamous cell carcinomas and p53 overexpressingskin ZE tions. VetDermatol 2004; 15 (s1): 27 papillomas of the dogusing the polymerase chain reaction and non-radio- 85. Schulman FY,Krafft AE, Janczewski T. Feline cutaneousfibropapillomas: active in situ hybridization. VetMicrobiol 1998; 60 (2–4):119–30. AT Clinicopathologic findings and association with papillomavirus infection. 105. TeifkeJP, KidneyBA, Lohr CV,Yager JA.Detection of papillomavirus- VetPathol2001; 38 (3): 291–296. DNAinmesenchymal tumour cellsand not in the hyperplastic epithelium /K 86. Scott DW,Miller WH, GriffinCE. Viral, rickettsial and protozoal diseases. of D In: Muller &Kirk'sSmall AnimalDermatology.Scott DW,Miller WH, Grif- feline sarcoids. VetDermatol 2003; 14 (1): 47–56. finCE, eds. Philadelphia: Saunders 2001; 517–42. 106.Thomsett LR, BaxbyD,Denham EM. Cowpoxinthe domestic cat. VetRec

UN 87. Scott DW,Miller WH, GriffinCE. Neoplastic and non-neoplastic tumors. In: 1978; 103 (25): 567. Muller &Kirk'sSmall AnimalDermatology.Scott DW,Miller WH, Griffin 107. TobeyJC, Houston DM, Breur GJ,JacksonML, Stubbington DA.Cutane- H CE, eds.Philadelphia: Saunders 2001; 1236–413. ous T- cell lymphoma in acat. JAmVet Med Assoc 1994; 204 (4): 606–9. 88. Scott DW.Feline dermatology 1983–1985:„the secret sits“. JAmAnim Hosp 108.Tryland M. Sandvik T, Holtet L, Nilsen H, OlsvikO,Traavik T. Antibodies Assoc 1986; 23: 255–74. to orthopoxvirus in domestic cats in Norway.Vet Rec 1998; 143 (4): 105–9. 89. Sellon RK. Update on molecular techniques for diagnostic testing of in- 109.von Bomhard D, Pflegehaar S, MahnelH,Schneekloth-Ducker I. Fall- fectious disease.VetClin NorthAm SmallAnim Pract2003; 33 (4): 677–93. bericht: Katzenpockeninfektion als Zoonosefür Hund und Mensch. Klein- 90. Simon M, Horvath C, PauleyD,King N, Hunt R, Ringler D. Plasma cell po- tierprax 1991; 32: 511–4. dodermatitis in feline immunodeficiencyvirus-infected cats. VetPathol 110. vonBomhard D, Pfleghaar S, MahnelH.Zur Epidemiologie, Klinik, Patho- 1993; 30 (5): 477. logie und Virologie der Katzen-Pocken-Infektion. Kleintierprax 1992;37 91. Smith KC, Bennett M, Garrett DC. Skin lesions caused by orthopoxvirus in- (4): 219–30. fection in adog.JSmallAnim Pract 1999; 40 (10): 495–7. 111. Walder EJ.Malignanttransformation of pigmented epidermal nevus in a 92. StanleyMA. Imiquimod and the imidazoquinolones: mechanism of action dog(Abstract). VetPathol1997; 34: 505. and therapeutic potential. ClinExp Dermatol 2002; 27 (7): 571–7. 112. WatrachAM, Small E, Case MT.Caninepapilloma: progression of oral pa- 93. SteffanJ,Alexander,D,Brovedani F, Fisch RD.Comparison of cyclosporine pillomatocarcinoma. JNat Cancerol Instit 1970; 45: 915–20. Awith methylprednisolone for treatment of canine atopic dermatitis: aparal- 113. Weston WL, Morelli JG. Herpes simplexvirus-associated erythema multi- lel blinded randomizedcontrolled trial. VetDermatol 2003; 14: 11–22. forme in prepubertal children. Arch Pediatr Adolesc Med 1997; 151 (10): 94. SteinbornA,EssbauerS.Marsch W. Kuh-/Katzenpocken Infektion beim 1014–6. Mensch: Ein potenziell verkanntes Krankheitsbild.Dtsch Med Wschr 2003; 114. White SD.Newly introduceddrugs in veterinarydermatology.In: Proc 128 (12):607–10. World Congress VetDermatol, Edinborough1996. 95. Stiles J, Townsend WM, Rogers QR, Krohne SG. Effect of oral administrati- 115. White SD,Rosychuk RA, Scott KV,Trettien AL, Jonas L, DenerolleP.Use on of L-lysine on conjunctivitis caused by felineherpesvirus in cats.Am JVet of isotretinoin and etretinate for the treatment of benign cutaneous neopla- Res 2002; 63 (1): 99–103. sia and cutaneous lymphoma in dogs. JAmVet Med Assoc 1993; 202 (3): 96. Stokking LB,EhrhartEJ, Lichtensteiger CA, Campbell KL. Pigmented epi- 387–91. dermal plaques in three dogs. JAmAnim HospAssoc 2004; 40 (5): 411–7. 116. Wilkinson GT,Prydie J, Scarnell J. Possible „orf“(contagious pustular der- 97. SuchyA,Bauder B, Gelbmann W, Lohr CV,TeifkeJP, Weissenbock H. Diag- matitis, contagious ecthyma of sheep) infection in the dog.VetRec 1970; 87 nosis of feline herpesvirus infection by immunohistochemistry, polymerase (25): 766–7. chain reaction, and in situ hybridization. JVet DiagnInvest 2000; 12 (2): 117.WithrowSJ, Mac Ewen EG. Viral papillomatosis. In: Clinical Veterinary 186–91. Oncology.WithrowSJ, Mac Ewen EG, eds. Philadelphia:Lippincott 1989. 98. SundbergJP, VanRanst M, Montali R, Homer BL, Miller WH, Rowland PH, 118. Ya ger JA,WilcockBP. Squamous papilloma. In: Color Atlasand Text of Scott DW,England JJ,Dunstan RW,Mikaelian I, JensonAB. Feline papillo- Surgical Pathology of the Dogand Cat. Dermatopathology and Skin Tu- mas and papillomaviruses. VetPathol2000; 37 (1): 1–10. mors. 99. SundbergJP, O'Banion MK, Schmidt-Didier E, Reichmann ME, Cloning and Ya ger JA,WilcockBP, eds. London: Wolfe 1994; 251–2. characterization of acanineoral papillomavirus. Am JVet Res 1986; 47: 119. Ya ger JA,WilcockBP. Perivascular dermatitis with diffuse orthokeratotic 1142–4. hyperkeratosis. In: Color Atlas and Text of Surgical Pathology of the Dog 100. SundbergJP, Smith EK, Herron AJ,JensonAB, Burk RD,Van RanstM.In- and Cat.Dermatopathologyand Skin Tumors. Ya ger JA,WilcockBP, eds. volvement of canine oralpapillomavirus in generalized oraland cutaneous London: Wolfe 1994; 66–70. verrucosis in aChinese Shar Peidog.Vet Pathol 1994; 31 (2):183–7. 101. SundbergJP, Junge RE, Lancester WD.Immunoperoxidase localization of papillomaviruses in hyperplastic and neoplastic epithelial lesions in animals. Am JVet Res 1984; 45 (7): 1441–6. Claude Favrot, DVM,MsSc, Dip. ECVD 102. Tachezy R, DusonG,RectorA, JensonAB,SundbergJP, VanRanst M. Clo- Klinik fürKleintiermedizin, Dermatologie ningand genomic characterization of Felis domesticus papillomavirus type 1. Virol 2002;301 (2): 313–21. Vetsuisse Fakultät, UniversitätZürich 103. Tanabe C, Kano R, Nagata M, NakamuraY, Watanabe S, Hasegawa A. Mo- Winterthurerstrasse260 lecular characteristics of cutaneous papillomavirus from the canine pig- CH-8057 Zürich mented epidermal nevus. JVet Med Sci 2000; 62 (11): 1189–92. E-Mail:[email protected]

Chapter 3

Parvovirus Infection of Keratinocytes as a Cause of Canine Erythema Multiforme

C. Favrot1, T. Olivry2, S. M. Dunston2, F. Degorce- Rubiales3 and J. S. Guy2

Veterinary Pathology, 2000: 37 (6): 647-649

1Clinique Vétérinaire de Ferrette, Ferrette, Haut-Rhin, France

2 Department of Clinical Sciences, College of Veterinary Medicine, North Carolina State University, Raleigh, NC, USA

3 LAPVSO, Toulouse, France

24 Vet Pathol 37:647±649 (2000)

BRIEF COMMUNICATIONS AND CASE REPORTS

Parvovirus Infection of Keratinocytes as a Cause of Canine Erythema Multiforme

C. Favrot, T. Olivry, S. M. Dunston, F. Degorce-Rubiales AND J. S. Guy

Abstract. Erythema multiforme major was diagnosed in a dog with necrotizing parvoviral enteritis. Skin lesions consisted of ulceration of the footpads, pressure points, mouth, and vaginal mucosa; vesicles in the oral cavity; and erythematous patches on the abdomen and perivulvar skin. Microscopic examination of mucosal and haired skin specimens revealed lymphocyte-associated keratinocyte apoptosis at various levels of the epidermis. Basophilic cytoplasmic inclusions were seen in basal and suprabasal keratinocytes. Immunohistochem- ical staining, performed with canine parvovirus-2±speci®c monoclonal antibodies, con®rmed the parvovirus nature of the inclusions in the nucleus and cytoplasm of oral and skin epithelial cells. This is the ®rst case of canine erythema multiforme reported to be caused by a viral infection of keratinocytes. This case study indicates that the search for epitheliotropic viruses should be attempted in cases of erythema multiforme in which a drug cause cannot be identi®ed.

Key words: Canine parvovirus-2 (CPV-2); dog; immunology; infection; skin; virus.

In humans, the classi®cation of erythema multiforme were present on the abdomen and chin. In spite of ¯uid ther- (EM) variants recently has been revised with an emphasis apy and intravenous cephalexin and metoclopramide, the on clinical manifestations.2 The relevance of this modi®ed dog died 2 days after presentation. clinical nosology has been supported by subsequent epide- A necropsy was performed and necrotic lesions were seen miologic and pathologic case studies. Dermatoses described throughout all intestinal sections. Histopathologic analysis of clinically as EM minor and major most commonly seem to small intestine specimens consisted of severe segmental nec- be caused by viral infections leading to lymphocyte-medi- rotizing enteritis suggestive of an acute infection due to ca- ated keratinocyte apoptosis.1,3,6,7 Human EM generally is nine parvovirus-2 (CPV-2). Viral inclusions were not iden- caused by herpes simplex virus,3,6,7 but it also can be trig- ti®ed in the intestinal specimens, presumably because of the gered by other infectious agents such as parvovirus B19.8 severe necrosis of the digestive epithelium. Skin biopsy In 1998, the consensus clinical classi®cation used for human EM was adapted to the canine species.4 That case study established that, in contrast to previous reports,10 canine cases of EM (minor or major) rarely were associated with previous drug exposure.4 In non±drug-related cases, offending causes could not be determined but a viral etiology was considered plausible. The purpose of the present paper is to describe a canine case of EM major in which parvovirus infection of epidermal and mucosal keratinocytes led to lymphocyte-associated apoptosis and clinical signs of EM major. A 2-month-old female Great Dane puppy was presented, 3 days after adoption, with acute-onset diarrhea, vomiting, dehydration, and skin lesions. Because parvovirus enteritis had been diagnosed recently at the facility of the dog's breeder, parvovirus was suspected as the cause of diarrhea. However, 6 days before the initial presentation, the dog had received a tetravalent vaccine (distemper, parvovirus, par- ain¯uenza, and hepatitis). Dermatologic examination revealed well-demarcated ulceration of the footpads (Fig. 1) and pressure points, as well as mouth and vaginal mucosae. Fig. 1 Skin, footpad; dog. A sharp-edged ulcer is pres- Vesicles were seen in the oral cavity. Erythematous patches ent.

647 25 648 Brief Communications and Case Reports Vet Pathol 37:6, 2000

Fig. 2 Haired skin; dog. Clusters of lymphocytes (white arrowheads) are located in the immediate vicinity of apoptotic keratinocytes (black arrows). Viral inclusions are visible in an intracellular vacuole (black arrowhead). H.E. Scale bar ϭ 18 ␮m. Fig. 3 Haired skin; dog. A lymphocyte (white arrowhead) is situated near an apoptotic keratinocyte (black arrow) that contained viral inclusions (black arrowheads). H.E. Scale bar ϭ 4 ␮m. Fig. 4 Haired skin, abdomen; dog. Dark-staining viral inclusions ®ll the cytoplasm of basal and juxtabasal keratinocytes. Viral aggregates are occasionally present in the super®cial dermis. Small inclusions are visible within keratinocyte nuclei (black arrow). Avidin±biotin±peroxidase immunohistochemistry, aminoethylcarbazole chromogen, hematoxylin counterstain, CPV2c2A parvovirus-speci®c monoclonal antibodies. Scale bar ϭ 11␮m. specimens were obtained from lesional skin and oral mu- cyte apoptosis with satellitosis were restricted to sites of epi- cosa. Focal mononuclear interface gingivitis was identi®ed dermal hyperplasia (Figs. 2, 3). Numerous basophilic cyto- in biopsy samples collected from the gum. Additionally, con- plasmic inclusions were seen in the lower third of the hy- ¯uent basal keratinocyte vacuolation progressing to vesicu- perplastic epidermis (Figs. 2, 3). lation with epithelial ulceration and neutrophil accumulation To verify the viral origin of cytoplasmic inclusions, a was observed. Prominent lymphocyte exocytosis was present three-step avidin±biotin±peroxidase immunohistochemical in preblistered mucosal epithelium. Keratinocyte apoptosis, technique was performed as previously described.9 Immu- often in close contact with lymphocytes (e.g., satellitosis), nostaining of paraf®n-embedded sections was done with two was observed at all levels of the epithelium. In some spec- monoclonal antibodies speci®c for CPV-2 (CPV2c2A and imens, basophilic inclusions were observed in the cytoplasm CPV103B10A, 1 2,000 dilution, MeÂrial, Lyon, France). Ex- of basal and suprabasal keratinocytes. Examination of haired amination of negative controls, consisting of sections im- skin specimens revealed varying degrees of the same path- munostained with irrelevant monoclonal antibodies, was un- ologic process. The epidermis exhibited focal hyperplasia, remarkable. In mucosal specimens, multiple intracellular crusting, and erosion. Lymphocyte exocytosis and keratino- parvovirus inclusions were seen throughout the epithelium.

26 Vet Pathol 37:6, 2000 Brief Communications and Case Reports 649

In haired skin samples, parvovirus inclusions were seen most Lyon, France) for providing distemper- and parvovirus-spe- commonly coalescing in basal and juxtabasal keratinocytes ci®c monoclonal antibodies. of hyperplastic epidermis (Fig. 4). The smallest viral inclusions were identi®ed in the keratinocyte's nucleus (Fig. 4). References Inclusions further aggregated and ®lled-up the cytoplasm of 1 Assier H, Bastuji-Garin S, Revuz J, Roujeau JC: Erythe- epithelial cells leading to displacement of the nucleus to the ma multiforme with mucous membrane involvement and cell's periphery and subsequent cell degeneration. Immuno- Stevens±Johnson syndrome are clinically different dis- staining of digestive specimens similarly demonstrated CPV- orders with distinct causes. Arch Dermatol 131:539±543, 2 particles in the epithelial crypts of the small intestine. Fur- 1995 thermore, viral inclusions of skin and mucosal sections were 2 Bastuji-Garin S, Rzany B, Stern RS, Shear NH, Naldi L, negative when immunohistochemistry was performed using Roujeau J-C: Clinical classi®cation of cases of toxic epi- monoclonal antibodies speci®c for either distemper virus (1 : dermal necrolysis, Stevens±Johnson syndrome and ery- 50, MeÂrial, Lyon, France) or papillomavirus-group±speci®c thema multiforme. Arch Dermatol 129:92±96, 1993 antigens (AR087±5R, undiluted, Biogenex, San Ramon, 3 Brice SL, Leahy MA, Ong L, Krecji S, Stockert SS, Huff CA). JC, Weston WL: Examination of non-involved skin, pre- According to the recently proposed classi®cation of canine viously involved skin, and peripheral blood for herpes EM, the skin lesions exhibited by this patient ®t the criteria simplex virus DNA in patients with recurrent herpes- for a clinical diagnosis of EM major (erythematous patchy associated erythema multiforme. J Cutan Pathol 21:408± lesions with ulcerations on less than 10% of the body surface 412, 1994 and with more than one mucosa affected).4 Our histologic 4 Hinn AC, Olivry T, Luther PB, Cannon AG, Yager JA: and immunohistochemical investigations suggested CPV-2 Erythema multiforme, Stevens±Johnson syndrome and infection of mucosal and epidermal keratinocytes as the pri- toxic epidermal necrolysis in the dog: clinical classi®- mary cause of EM in this dog. Remarkably, most parvoviral cation, drug exposure and histopathological correlations. inclusions were identi®ed in the cytoplasm of keratinocytes, J Vet Allergy Clin Immunol 6:13±20, 1998 whereas only rare viral particles were seen in cell nuclei. 5 Hullinger GA, Hines ME, Styer EL, Frazier KS, Baldwin However, these observations are identical to those described CA: Pseudocytoplasmic inclusions in tongue epithelium in glossal specimens of dogs naturally infected with CPV- of dogs with canine parvovirus-2 infections. J Vet Diagn Invest 10:108±111, 1998 2.5 Indeed, viral replication initially occurs in the nucleus 6 Imafuku S, Kokuba H, Aurelian L, Burnett J: Expression but large virion clusters appear as cytoplasmic aggregates. of herpes simplex virus DNA fragments located in epi- However, these inclusions still are surrounded by the nuclear dermal keratinocytes and germinative cells is associated membrane and should be referred to as pseudocytoplasmic.5 with the development of erythema multiforme lesions. J A viral infection of keratinocytes suggests a logical patho- Invest Dermatol 109:550±556, 1997 genesis of EM lesions in this dog. We hypothesize that an 7 Kokuba H, Imafuku S, Aurelian L, Burnett JW: Erythe- infection of stem cells and transient amplifying keratinocytes ma multiforme lesions are associated with expression of most likely occurred following hematogenic dissemination a herpes simplex virus (HSV) gene and qualitative alter- of the parvovirus. Viral peptides could be presented by class ations in the HSV-speci®c T-cell response. Br J Dermatol I major histocompatibility complex molecules at the surface 138:952±964, 1998 of epithelial cells. Recognition of viral antigens by T-lym- 8 Lobkowicz F, Ring J, Schwarz TF, Roggendorf M: Ery- phocytes, possibly sensitized by the previous parvovirus thema multiforme in a patient with acute human parvo- vaccination, would trigger these cytotoxic cells to induce the virus B19 infection. J Am Acad Dermatol 20:849±850, apoptosis of virus-infected keratinocytes. 1989 The present case study supports the concept that a viral 9 Olivry T, Moore PF, Naydan DK, Puget BJ, Affolter VK, etiology is possible in some forms of canine EM. We pro- Kline AE: Antifollicular cell-mediated and humoral im- pose that a search for epitheliotropic viruses (e.g., distemper, munity in canine alopecia areata. Vet Dermatol 7:67±79, papilloma-viruses, parvoviruses, and herpesviruses) should 1996 be attempted in cases of canine EM in which a causative 10 Scott DW, Miller WH: Erythema multiforme in dogs and drug cannot be clearly established. cats: literature review and case material from the Cornell University College of Veterinary Medicine (1988±1996). ACKNOWLEDGEMENTS Vet Dermatol 10:297±309, 1999 We thank Dr. GuagueÁre for his comments in the initial Request for Reprints from Dr. C. Favrot, Clinique VeÂteÂri- phase of the study and Drs. Latour and Soulier (MeÂrial, naire, 32 rue de Mulhouse, F-68300 St. Louis (France).

Chapter 4

Two cases of FeLV-associated dermatoses

C. Favrot1, S. Wilhelm1, P. Grest2, M. L. Meli3, R. Hofmann-Lehmann3 and A. Kipar4

Veterinary Dermatology, 2005, 16:407-412

1Clinic for Small Animal Internal Medicine, Dermatology Unit, Vetsuisse Faculty, University of Zurich, Zurich, Switzerland

2Institute of Veterinary Pathology, Vetsuisse Faculty, University of Zurich, Zurich, Switzerland

3Clinical Laboratory, Vetsuisse Faculty, University of Zurich, Zurich, Switzerland

4Department of Veterinary Pathology, Faculty of Veterinary Science, University of Liverpool, Liverpool, UK

Veterinary Dermatology 2005, 16, 407–412

Blackwell Publishing Ltd Case report Two cases of FeLV-associated dermatoses

C. FAVROT*, S. WILHELM*, P. GREST†, M. L. MELI§, R. HOFMANN-LEHMANN§ and A. KIPAR‡

*Clinic for Small Animal Internal Medicine, Dermatology Unit, †Institute of Veterinary Pathology, and §Clinical Laboratory, Vetsuisse Faculty, University of Zürich, Zürich, Switzerland, ‡Department of Veterinary Pathology, Faculty of Veterinary Science, University of Liverpool, Liverpool, UK

(Received 25 July 2005; accepted 10 July 2005)

Abstract Two cases of feline leukaemia virus (FeLV)-associated dermatosis are described. The first cat was affected by an ulcerative dermatitis identified as a giant-cell dermatosis. The second case was a cutaneous lymphoma. In both cases, FeLV antigens and FeLV genome were demonstrated in the affected skin immunologically and with polymerase chain reaction, respectively. The first case suggests that, like other retroviruses, at least some strains of FeLV can induce syncytium formation. As FeLV antigens and genome were demonstrated in a serologically negative cat, the second case suggests that focal skin FeLV replication may occur. FeLV-associated dermatoses are rare skin conditions that may be under-diagnosed.

INTRODUCTION MATERIAL AND METHODS

Feline leukaemia virus (FeLV), a member of the on- Animals cornavirus subfamily of retroviruses, occurs worldwide Two castrated domestic indoor–outdoor male cats and replicates in many tissues including respiratory (the first aged 3 and the second aged 15 years), in epithelium, salivary gland and bone marrow.1 It causes reduced general condition and with skin plaques and approximately one-third of feline lethal cancers and ulcerations, were presented at the Clinic for Small numerous cats die of anaemia or infectious diseases Animal Internal Medicine, Dermatology Unit, Vetsuisse as a consequence of the immunosuppressive effects of Faculty, Zürich. Both cats were given clinical and the virus.1 FeLV infection has also been associated with dermatological examinations. numerous infectious dermatoses of fungal, parasitic and/or bacterial aetiology.2 A direct cytopathic effect Serological examination of the virus in the skin, however, has been rarely The presence of plasma FeLV p27 antigen as a measure demonstrated, but is associated with two different for viraemia was determined using double-antibody syndromes: giant-cell dermatosis and epidermal horns.3,4 sandwich enzyme-linked immunosorbent assay (ELISA) Lymphoma accounts for about 90% of the haemat- as previously described.6 opoietic tumours in cats and is often a consequence of Plasma samples were tested for feline immunodefi- FeLV infection.5 Cutaneous lymphomas, however, are ciency virus (FIV) by ELISA, measuring antibodies rare and usually occur in older FeLV-negative cats.5 against the FIV transmembrane protein.7 The purpose of this article is to present two new cases of FeLV-induced dermatoses with evidence of Histological and immunohistological examination viral antigens and proviral sequences in the skin: one Skin samples for histopathological examination of T-cell lymphoma in a serologically negative cat and were taken (during the first examination of both cats) by one case of giant-cell dermatosis in a serologically biopsy from the lesions with a 6-mm skin punch, fixed positive cat. in 4% neutral buffered formalin and embedded in paraffin wax. Sections were cut and either stained with haematoxylin and eosin or used for immunohistological examination. Correspondence: Claude Favrot, Clinic for Small Animal Internal Skin lesions were examined immunohistologically Medicine, Vetsuisse Faculty, Winterthurerstrasse 260, 8057 Zürich, for FeLV antigens using a cocktail of mouse monoclonal Switzerland. E-mail: [email protected] antibodies against the FeLV envelope protein gp70 The clinical cases have been reported at the 5th World Congress of Veterinary Dermatology, Vienna, 25–28 August 2004 (case 1) and and the group-specific protein p27 (clones C11D82i 20th North America Veterinary Dermatology Forum, Sarasota, 6–10 and PF12J-10A; Custom Monoclonals, Sacramento, April 2005 (Case 2). CA, USA). The peroxidase–antiperoxidase method

408 C Favrot et al. was applied as previously described.8 In case 2, the skin lesions were also stained for the pan-T-cell marker CD3 and the pan-B-cell marker CD45R as previously described.9

Real-time polymerase chain reaction (real-time PCR) for exogenous FeLV and feline herpesvirus-1 (FHV-1) Polymerase chain reaction (PCR) analysis was performed on skin with lesions after deparafﬁnization of two 20-µm thick sections of the same lesional tissues as mentioned previously and DNA extraction using the DNeasy tissue kita. FeLV provirus was detected by real-time PCR with primers that recognize the unique region (U3) of the long-terminal repeat (LTR) of exogenous FeLV-A,-B,-C as described previously.10,11 Examination for feline herpesvirus (FHV-1) sequences was undertaken as described elsewhere.12

RESULTS

Case 1 The cat was presented with a pruritic dermatosis of 3 months’ duration and a previous history including vaccination for FeLV during the first year of life and Figure 2. Case 1. A skin lesion. Folliculitis (red arrow) and booster injection in the second year. Physical exam- syncytium formation (black arrow) of epithelial cells of a hair ination revealed well-demarcated ulcerative lesions follicle. Haematoxylin and eosin stain. of the head, limbs and paws (Fig. 1). The cat was also depressed and febrile (39.7 °C). A staphylococcal infec- Histological examination revealed an ulcerative tion was identified by cytological and bacteriological dermatitis with folliculitis, dyskeratotic keratinocytes examination (Staphylococcus intermedius), but multiple and syncytia formation within the epidermis and the skin scrapings were negative for parasites and fungal sebaceous glands (Figs 2 and 3). The epidermis was culture was negative. A nonregenerative normochrome acanthotic and hyperkeratotic. Multiple giant kerati- normocytic anaemia (hematocrit: 18% (reference range: nocytes and scattered apoptotic cells were present in 33–45%) reticulocytes: 0.6%) was diagnosed. The ELISA the superficial epidermis, sebaceous glands and hair for FeLV antigen was positive, whereas that for FIV follicles. In the dermis, a severe perifollicular to diffuse antibodies was negative. The cat was started on cefalexin inflammatory infiltration with numerous lymphocytes, (25 mg kg−1 twice daily) therapy. plasma cells and neutrophils was present. Immunohistological analyses revealed numerous epithelial cells that expressed viral proteins with variable intensity (Fig. 4) in the epidermis of the skin surface, hair follicles and sebaceous glands. Lymphocytes in the dermal infiltrates were often positive as well. A 131-bp long proviral FeLV DNA was amplified from skin samples of this cat. Skin samples evaluated for FHV-1 DNA were deemed negative. A diagnosis of FeLV-induced giant-cell dermatosis was made. Despite the treatment and a marked but temporary improvement of the skin lesions, the general condition deteriorated and the cat was euthanized. Necropsy was not permitted.

Case 2 The cat was presented with a dermatosis of 2 months’ duration and weight loss. It had previously been treated Figure 1. Case 1. Well-demarcated ulcerated lesion on the face. with megestrol acetate and prednisolone (variable dos- ages) on the basis of a tentative diagnosis of eosinophilic ° a plaques. The cat was depressed and febrile (39.6 C). Phys- QIAGEN, Hombrechtikon, Switzerland ical examination of the skin revealed multiple nodules

FeLV-induced dermatoses 409

Figure 5. Case 2. Ulcerated nodule on the carpus. Figure 3. Case 1. Sebaceous glands. Syncytium formation (black arrow) in epithelial cells. Haematoxylin and eosin stain.

Figure 6. Case 2. Histology of the cutaneous nodule in Fig. 5. Figure 4. Case 1. Epithelial cells in a hair follicle express FeLV Superficial ulceration (black arrow) and diffuse infiltration of antigen with variable intensity (arrows). Immunohistological the dermis by neoplastic round cells (red arrow). Haematoxylin and demonstration of FeLV gp70, peroxidase antiperoxidase method, eosin. Papanicolaou’s haematoxylin counterstain. and ulcerated lesions that affected the face, feet and results a diagnosis of cutaneous non-epitheliotropic T-cell abdomen (Fig. 5). Multiple skin scrapings were negative lymphoma was made. Immunohistology for FeLV antigen for parasites. Additionally, the cat was anaemic (Ht. 23%) revealed variably intense viral protein expression by numer- and lymphopenic (170 lymphocytes per microlitre). Serum ous neoplastic cells and weak expression by epidermal ELISA tests for FeLV antigens and FIV antibodies were cells in all layers (Fig. 9). A 131-bp long proviral FeLV both negative. Radiographic examination of the thorax DNA fragment was amplified from skin samples of this cat. revealed a nodular opacity in the left lung. Sonographic Despite Lomustine therapy (10 mg once daily) examination of the abdomen showed the presence of a started after histological diagnosis, the cat’s general small nodule in the liver. Fine needle aspirates of pul- condition deteriorated and it was euthanized. Necropsy monary and liver nodules were, however, unremarkable. was not permitted so the nodular liver and pulmonary Histological examination of the skin lesions revealed lesions could not be further evaluated. extensive superficial ulceration and focally extensive dense dermal infiltration by pleomorphic round cells, resembling lymphoblasts with round to indented nuclei DISCUSSION containing fine chromatin and one single medium-sized nucleolus (Fig. 6). Cellular atypia such as anisocytosis This report describes two FeLV-associated skin condi- and anisocaryosis were observed, multiple large nucleoli tions in cats, presenting clinically as dermatoses with and abnormal mitoses were also seen (Fig. 7). A large poor response to treatment: giant-cell dermatosis and proportion of neoplastic cells exhibited peripheral and/or cutaneous lymphoma. The presence of proviral FeLV cytoplasmic CD3 expression (Fig. 8). Based on these sequences as well as FeLV antigens in the skin with

410 C Favrot et al.

Figure 7. Case 2. The neoplastic inﬁltrate is composed of pleomorphic round cells, resembling lymphoblasts. They exhibit variably chromatin-dense nuclei (red arrows) and distinct nucleoli (black arrow). Haematoxylin and eosin.

Figure 9. Case 2. Neoplastic cells (arrowheads) and epidermal keratinocytes (arrow) express FeLV antigens. Immunohistological demonstration of FeLV gp70 and p27, peroxidase antiperoxidase method, Papanicolaou’s haematoxylin counterstain.

carcinoma), infectious (alpha-herpesvirus infections) and immunologic disorders such as lupus erythema- tosus, Hailey–Hailey disease and psoriasis.13 Retroviruses including human lentiviruses such as human immuno- Figure 8. Case 2. Neoplastic cells exhibit a variably intense deficiency virus and feline gammaretroviruses (e.g. FeLV), peripheral and/or nuclear reaction for the pan T-cell marker CD3 possess fusion proteins and are sometimes seen to induce (arrowheads). Blood vessels and hair follicles are negative. syncytium formation in lymphoid tissues.14–16 How- Peroxidase–antiperoxidase method, Papanicolaou’s haematoxylin ever syncytial keratinocytes are mainly observed in counterstain. AIDS patients that are concomitantly infected with herpesvirus or papillomavirus.17,18 Gross and coworkers lesions of both cats was demonstrated by PCR and suggested that syncytium formation observed in feline immunohistological analysis, respectively. cases is not a consequence of a direct cytopathic effect So far, six cases of FeLV-induced giant-cell dermatosis of the virus but of carcinomatous transformation of have been described in the literature.4 Most presented the epidermis.4 HIV-induced squamous cell carcinoma, as scaling and crusting dermatoses affecting mainly however, has rarely been observed in humans.18 The the face and the neck. Vesicular and ulcerative lesions oncogenic potential of FeLV is much greater than that were also reported with involvement of the footpads of HIV but Rohn and coworkers have also demonstrated and mucous membranes.4 The clinical and histological that FeLV variants do exhibit various pathogenic and presentation of this case was similar to those previously cytopathic effects, including syncytium formation in described, although more ulcerative and less hyper- one strain.16 It thus appears possible that FeLV-induced keratotic. Mucous membranes were unremarkable. giant-cell dermatosis is the result of a specific and Affected cats usually decline quickly with death or probably rare viral variant. Confirmation of this euthanasia days to weeks after initial presentation. hypothesis needs further investigation. The histological hallmark of giant-cell dermatosis Lymphomas are frequent neoplasms in cats and is the presence of syncytial keratinocytes and dys- often a consequence of FeLV infection.5 Cutaneous keratotic cells. The former have also been observed in lymphomas, however, are rare and usually occur in FeLV-infected cats in association with cutaneous older serologically FeLV-negative cats.5 Attempts to horns.3 Multinucleated keratinocytes are observed in identify FeLV genomic sequences and antigens in humans in association with neoplastic (squamous cell epitheliotropic and nonepitheliotropic cutaneous

FeLV-induced dermatoses 411 lymphomas are occasionally successful and confirm 2. Scott DW, Miller WH, Griffin CE. Viral, rickettsial FeLV involvement in the development of at least some and protozoal diseases. In: Muller and Kirk’s Small cutaneous lymphomas in cats.19,20 Tobey and co- Animal Dermatology. Philadelphia: W.B. Saunders, 2001: workers suggested defective or latent infection as they 517–42. did not detect FeLV antigens in the neoplastic cells of a 3. Center SA, Scott DW, Scott FW. Multiple cutaneous horns on the foot pads of a cat. Feline Practice 1982; 12: cutaneous lymphoma.21 In the case reported here, both 26–30. proviral genome and antigens were demonstrated. The 4. Gross TL, Clark EG, Hargis AM et al. Giant cell der- presence of viral antigens in neoplastic cells and kerat- matosis in FeLV-positive cats. Veterinary Dermatology inocytes but not in the peripheral blood suggests focal 1993; 4: 117–22. productive infection of both cell types. As the greater 5. Vonderhaar MA, Morisson WB. Lymphosarcoma. In: sensitivity of RT-PCR detects proviral DNA in the Morisson WB ed. Cancer in Dogs and Cats. Jackson, serum of cats with undetectable antigenaemia, negative Wyoming: Teton New Media, 2002: 641–70. ELISA does not rule out generalized infection. Local- 6. Lutz H, Pedersen NC, Durbin R et al. Monoclonal anti- ized FeLV replication has been reported after experi- bodies to three epitopic regions of feline leukemia virus mental infection with viral antigens in the spleen, bone p27 and their use in enzyme-linked immunosorbent assay marrow, lymph nodes or small intestine.22 Restricted, of p27. Journal of Immunological Methods 1983; 56: 209–20. localized FeLV replication was also shown in another 7. Calzolari M, Young E, Cox D et al. Serological diagnosis study in naturally infected cats. The same organs 19 of feline immunodeficiency virus infection using were examined but no viral antigen was found. This recombinant transmembrane glycoprotein. Veteri- case report supports the hypothesis that infections nary Immunology and Immunopathology 1995; 46: 83– restricted to the skin may sometimes occur in cats and 92. subsequently induce cutaneous lymphomas. 8. Kipar A, Kremendahl J, Grant CK et al. Expression of As PCR assays and immunohistology for FeLV are viral proteins in feline leukemia virus-associated enteritis. not routinely carried out on cutaneous lymphomas, Veterinary Pathology 2000; 27: 129–36. the frequency of this association is unknown and may 9. Kipar A, Bellmann S, Kremendahl J et al. Cellular com- be underestimated. It is possible, however, that FeLV position, coronavirus antigen expression and production tumorigenesis of dermal T cells is rare and caused of of specific antibodies in lesions in feline infectious peri- tonitis. Veterinary Immunology and Immunopathology particular FeLV variants. A previous study failed to 1998; 65: 243–57. detect FeLV nucleic acid in a portion (20%) of feline 10. Tandon R, Cattori V, Gomes-Keller MA et al. Quantita- lymphoma samples and suggested that FeLV lympho- tion of feline leukaemia virus viral and proviral loads by magenesis can be associated with clearance of viral nucleic TaqMan((R)) real-time polymerase chain reaction. 23 acids from cancer cells. This ‘hit and run’ mechanism Journal of Virological Methods 2005; 130(1–2): 124–32. has already been associated with other conditions induced 11. Hofmann-Lehmann R, Huder JB, Gruber S et al. Feline by retroviruses.24 This study was, however, carried out leukaemia provirus load during the course of experimen- with a single-round PCR system23 that has a lower dia- tal infection and in naturally infected cats. Journal of gnostic sensitivity than the more recently developed FeLV- General Virology 2001; 82: 1589–96. specific nested and TaqMan PCR systems.10,11 As real-time 12. Vogtlin A, Fraefel C, Albini S et al. Quantification of TaqMan PCR can detect fewer nucleic acid copies feline herpesvirus 1 DNA in ocular fluid samples of clinically diseased cats by real-time TaqMan PCR. Journal compared to conventional PCR we may be able to detect of Clinical Microbiology 2002; 40: 519–23. FeLV in a larger portion of lymphosarcoma cases. 13. Kimura SK, Hatano H. Multinucleated epidermal cells In conclusion, this report suggests that FeLV-induced in non-neoplastic dermatoses. British Journal of Derma- dermatoses are probably due to particular viral variants tology 1978; 99: 485–9. and their frequency might be underestimated. 14. White JM. Membrane fusion. Science 1992; 258: 917–24. 15. Fenyo EM, Morfeldt-Manson L, Chiodi F et al. Distinct replicative and cytopathic characteristics of human immu- ACKNOWLEDGEMENTS nodeficiency virus isolates. Journal of Virology 1988; 62: 4414–9. Thanks to the Centre for Clinical Studies, the Vetsuisse 16. Rohn JL, Moser MS, Gwynn SR et al. In vivo evolution Faculty of the University of Zürich for use of logistics of a novel, syncytium-inducing feline leukemia virus variant. Journal of Virology 1998; 72: 2686–96. and financial support from the Swiss National Science 17. Fagan WA, Collins PC, Pulitzer DR. Verrucous Herpes Foundation (31-65231). R.H.-L. is the recipient of a virus infection in human immunodeficiency virus patients. professorship from the Swiss National Science Foun- Archives of Pathology and Laboratory Medicine 1996; dation (PP00B-102866). 120: 956–8. 18. Orem J, Otieno MW, Remick SC. AIDS-associated cancer in developing nations. Current Opinion in Oncology REFERENCES 2004; 16: 468–76. 19. Kovacevic S, Kipar A, Kremendahl J et al. Immunohis- 1. Cotter SM. Feline viral neoplasia. In: Green CG ed. tochemical diagnosis of feline leukemia virus infection in Infectious Diseases of the Dog and Cat. Philadelphia: formalin-fixed tissue. European Journal of Veterinary W. B. Saunders Co, 1998: 71–84. Pathology 1997; 3: 13–9.

412 C Favrot et al.

20. Jackson ML, Wood SL, Misra V et al. Immunohis- feline leukemia virus infection in cats. American Journal tochemical identification of B and T lymphocytes in of Veterinary Research 1992; 53: 604–7. formalin-fixed, paraffin-embedded feline lymphosarco- 23. Jackson ML, Haines DM, Meric SM et al. Feline leuke- mas: relation to feline leukemia virus status, tumor site, mia virus detection by immunohistochemistry and and patient age. Canadian Journal of Veterinary polymerase chain reaction in formalin-fixed, paraffine- Research 1996; 60: 199–204. embedded tissue of cats with lymphosarcoma. Canadian 21. Tobey JC, Houston DM, Breur GJ et al. Cutaneous Journal of Veterinary Research 1993; 57: 269–76. T-cell lymphoma in a cat. Journal of the American 24. Nath A. Pathobiology of human immunodeficiency virus Veterinary Medical Association 1994; 204: 606–9. dementia. Journal of Biology and Chemotherapy 1999; 22. Hayes KA, Rojko JL, Mathes LE. Incidence of localized 19: 113–27.

Résumé Deux cas de dermatoses associées au FeLV sont rapportées. Le premier chat présentait une dermatite ulcérative identiﬁée comme une dermatose à cellules géantes. Le second cas était un lymphome cutané. Dans les deux cas, des antigènes du FeLV et du génome ont été mis en évidence immunologiquement et par PCR respec- tivement. Le premier cas suggère que comme d’autres rétrovirus, au moins certaines souches de FeLV peuvent provoquer la formation de syncytium. Comme les antigènes et l’ADN du FeLV ont été retrouvés chez un chat séronégatif, le second cas suggère qu’une réplication focale du FelV peut se dérouler dans la peau. Les dermatoses associées au FeLV sont des dermatoses rares peut être sous diagnostiquées.

Resumen Describimos dos casos de dermatosis en gatos asociados con la presencia del virus de la Leucemia Felina (FeLV). El primer gato presentó una dermatitis ulcerativa con presencia de células gigantes. El segundo caso fue un linfoma cutáneo. En ambos casos se demostró la presencia de antígeno de FeLV y del genoma de FeLV en la piel afectada, mediante una prueba inmunológica y con una reacción de polimerasa en cadena, respectivamente. Las características del primer caso sugieren que, al igual que otros retrovirus, al menos algunas variantes de FeLV pueden inducir la formación de células sincitiales. En el segundo caso, ya que tanto el antígeno como el genoma de FeLV fueron detectados en un gato serológicamente negativo, se sugiere que puede ocurrir una replicación local del virus de la Leucemia Felina. Las dermatosis asociadas con la presencia de FeLV son procesos raros, y tal vez no suﬁcientemente reconocidos.

Zusammenfassung Zwei Fälle einer FeLV-assoziierten Dermatose sind beschrieben. Die erste Katze hatte eine ulzerierende Dermatitis, die identiﬁziert wurde als eine Riesenzell-Dermatose. Der zweite Fall zeigte kutanes Lymphom. In beiden Fällen wurden FeLV Antigene immunologisch, sowie FeLV Genom mittels Polymerase Chain Reaction nachgewiesen. Der erste Fall lässt darauf schließen, dass zumindest einige FeLV Stämme, sowie andere Retroviren, eine Synzytium Formation induzieren können. Da FeLV Antigene und Genom auch in einer serologisch negativen Katze demonstriert wurden, weist der zweite Fall darauf hin, dass eine fokale FeLV Replikation in der Haut vorkommen kann. FeLV-assoziierte Dermatosen sind seltene Erkrankungen der Haut, die möglicherweise zu selten diagnostiziert werden.

Chapter 5

Evaluation of papillomaviruses associated with cyclosporine-induced hyperplastic verrucous lesions in dogs C. Favrot 1, T. Olivry2, A.H. Werner3, G. Nespecca1, A. Utiger1, P. Grest4, M. Ackermann5

American Journal of Veterinary Research, 2005, 66; 10:

1764-1769

1Clinic for Small Animal Internal Medicine, Dermatology Unit, Vetsuisse Faculty, University of Zurich, Zurich, Switzerland 2Department of Clinical Sciences, College of Veterinary Medicine, North Carolina State University, Raleigh, NC, USA 3Valley Veterinary Speciality Services, Studio City, CA, USA 4 Pathology Institute, Vetsuisse Faculty, University of Zurich, Zurich, Switzerland 5 Virology Institute, Vetsuisse Faculty, University of Zurich, Zurich, Switzerland

35 04-12-0466r.qxp 9/14/2005 10:43 AM Page 1764

Evaluation of papillomaviruses associated with cyclosporine-induced hyperplastic verrucous lesions in dogs

Claude Favrot, DrVet, MSc; Thierry Olivry, DrVet, PhD; Alexander H. Werner, DVM; Gilles Nespecca; Anna Utiger, DVM; Paula Grest, DVM; Mathias Ackermann, DVM, PhD

helper T lymphocytes. In humans, cyclosporine A has Objective—To determine whether cyclosporine A- been used for more than a decade to prevent transplant induced hyperplastic skin lesions of dogs were asso- rejection and for treatment for dermatologic conditions ciated with papillomavirus infections. that include severe psoriasis and atopic dermatitis.1 Animals—9 dogs that were treated with cyclosporine The usefulness of cyclosporine A in treatment for A and developed hyperplastic skin lesions. atopic dermatitis in dogs has been reported,2-6 and the Procedure—History and clinical and histopathologic drug has also been approved for treatment for immune- data were collected. Paraffin-embedded skin biopsy mediated conditions such as perianal fistulae and seba- specimens from hyperplastic skin lesions were ceous adenitis.7,8 immunostained for common papillomavirus genus- specific structural antigens by use of a polyclonal rab- Hyperplastic skin lesions are known adverse bit anti-bovine papillomavirus type 1 antiserum. effects of long-term treatment with cyclosporine A in Sections from each tissue block underwent DNA humans. Most lesions appear to be papillomavirus- extraction, and polymerase chain reaction (PCR) induced verruca vulgaris, but malignant carcinoma- assays were performed with several sets of primers tous transformations are also possible.9 Similar lesions to amplify a wide range of papillomavirus DNA from have also been described in dogs,10,11 but evidence for humans and other animals. causative involvement by papillomavirus is lacking. Results—In 7 of 9 dogs, there were more than 10 Most lesions in dogs resemble those reported as pso- hyperplastic skin lesions that microscopically resem- riasiform lichenoid dermatosis, and skin nodules usu- bled those of psoriasiform lichenoid dermatosis. In ally regress spontaneously or in response to antimicro- those dogs, results of testing for papillomavirus via bial treatment.10,11 Because papillomavirus has been immunohistochemical analyses and PCR assays were detected in most cyclosporine A-induced hyperplastic negative. In the other 2 dogs, there were only 1 and skin lesions in humans, a similar role for papillo- 3 verrucous lesions, and in those dogs, histologic evaluation revealed koilocytes and nuclear viral inclu- mavirus in the development of similar lesions in dogs sions that were immunoreactive for papillomavirus warrants investigation. We observed that these drug- antigens. Papillomavirus DNA was amplified from induced nodules appear to be heterogeneous in nature. both dogs. One of the sequences was characteristic Most lesions are numerous and resemble those of pso- for the canine oral papillomavirus, whereas the other riasiform lichenoid dermatosis, with staphylococci in had similarities with the recently described canine the stratum corneum and absence of detectable papil- papillomavirus 2. lomavirus DNA and antigens. In 2 dogs, however, Conclusions and Clinical Relevance—In dogs, cyclosporine A administration was associated with the hyperplastic skin lesions occasionally develop during eruption of few skin nodules diagnosed as viral papil- treatment with cyclosporine A. Most of the lesions lomas. The objective of this study was to determine resemble those of psoriasiform lichenoid dermatosis, whether cyclosporine A-induced hyperplastic skin although papillomavirus can be detected in some lesions in dogs contained papillomavirus DNA and instances. (Am J Vet Res 2005;66:1764–1769) genus-specific structural antigens. Materials and Methods yclosporine A is a potent immunosuppressive History and clinical information regarding 9 dogs that Cagent that acts primarily by selectively inhibiting were treated with cyclosporine A and developed verrucous skin lesions were recorded retrospectively. Limited clinical and pathologic information regarding 1 of the dogs has Received December 16, 2004. 2 Accepted February 2, 2005. already been published. From the Clinic for Small Animal Internal Medicine, Dermatology Histologic and immunohistochemical evaluations— Unit (Favrot, Nespecca, Utiger), the Pathology Institute (Grest), Punch biopsy specimens of the skin were obtained from derm- and the Virology Institute (Ackermann), Vetsuisse Faculty, University of Zürich, Winterthurerstrasse 260, CH 8057 Zurich, atologic lesions of the 9 dogs. Biopsy specimens were fixed in Switzerland; the Department of Clinical Sciences, College of formalin, embedded in paraffin, and processed routinely for Veterinary Medicine, North Carolina State University, Raleigh, NC histologic assessment. Five-micrometer-thick skin sections 27606 (Olivry); and Valley Veterinary Specialty Services, 13125 were stained with H&E and Gram stains by use of standard Ventura Blvd, Studio City, CA 91604 (Werner). methods. Presented in part at the Annual Meeting of the American Academy For the detection of papillomavirus antigens, a 3-step of Veterinary Dermatology and American College of Veterinary immunohistochemical method was used. The primary Dermatology, Monterey, Calif, April 2003. immunoreagent was a polyclonal rabbit antiserum directed Address correspondence to Dr. Favrot. against chemically disrupted bovine papillomavirus type 1.a

1764 AJVR, Vol 66, No. 10, October 2005 36 04-12-0466r.qxp 9/14/2005 10:43 AM Page 1765

This reagent detects papillomavirus genus-specific common both genomes. The forward primer was 5’-ATGGCGGM- structural antigens regardless of the host species. The prim- TARAAAAGGTA-3' and the reverse primer was 5’-AACAGCT- ary antiserum was used at a 1:200 dilution, and other reagents GYTTTTTARCYTTTTT-3'. Internal control was made by use (ie, biotinylated goat anti-canine rabbit IgG and streptavidin of the same forward primer with the reverse primer PapE1 5’- peroxidaseb) were diluted at 1:40. Diaminobenzidine was used ACAGTTGCAGGGAAAGTC-3' to amplify an internal 184-bp as a chromogen. The positive control consisted of paraffin- fragment. The PCR reactions were performed in 30-µL vol- embedded sections from a dog with canine oral papillo- umes containing 1 µL of genomic DNA, 50mM KCl, 3mM µ µ mavirus (COPV)-induced oral papillomas, whereas the nega- KCl2, 200 M of each dNTP, 0.3 M each of consensus sense tive control antiserum consisted of normal rabbit serum. and antisense primers, and 2.5 units of DNA polymerase.e The Skin sections stained with H&E, Gram stain, and amplification involved an initial denaturation at 95oC for 4 immunostain for papillomavirus were coded and evaluated by minutes and 30 cycles at 95oC for 1 minute, 50oC for 1 minute, an author (TO) who was unaware of the origins of the speci- and 74oC for 1 minute, with a final elongation step at 74oC for mens. Sections stained with H&E were examined for changes 5 minutes. Reaction mixture with no DNA served as a negative typically associated with papillomavirus infection in dogs, control, and COPV-positive papilloma DNA samples and feline and those changes were recorded as present or absent, includ- papilloma-positive DNA samples were used as positive con- ing epidermal dysplasia, hypergranulosis, coalescing kerato- trols. The PCR products were resolved via electrophoresis in hyalin granules, koilocytes, intranuclear viral inclusions in 2% agarose gel stained with ethidium bromide. Amplified the stratum spinosum and stratum granulosum, intrafollicu- DNA was sequenced by use of fluorescent sequencing and flu- lar or intraepidermal pustules, and bacteria in the stratum orescent dye terminator.f corneum. Sections stained with Gram stain were evaluated for epidermal bacteria. Sections immunostained for papillo- Detection of papillomavirus with CP4, C5P, and PPF1 mavirus group-specific antigens were evaluated for staining in primers—To amplify the DNA of as many different papillo- keratinocytes of the stratum spinosum, stratum granulosum, maviruses as possible, the CP4, CP5, PPF1 primer set was selected because of its ability to detect the nucleic acids of up and stratum corneum. Transmission electron microscopy was 12 performed on the stratum corneum of 1 specimen. to 64 human papillomaviruses. The PCR reactions were performed in 30-µL volumes containing 1 µL of genomic DNA, µ µ Polymerase chain reaction assays—Amplification of 50mM KCl, 3mM KCl2, 200 M of each dNTP, 0.45 M of the papillomavirus DNA via polymerase chain reaction (PCR) CP4 and CP5 primers, 0.3µM of the PPF1 primer, and assays was performed on formalin-fixed paraffin-embedded 2.5 units of DNA polymerase.d Amplification involved an ini- specimens. Thirty-micrometer-thick sections were cut from tial denaturation at 95oC for 10 minutes and 40 cycles at 95oC tissue blocks by use of a new disposable microtome blade for each block to avoid cross-contamination between samples. Each section was deparaffinized twice with 1.2 mL of xylene at 20oC for 10 minutes, washed with 100% ethanol, and air-dried. Desiccated samples were suspended in a lysis buffer (50mM Tris-HCl [pH, 8.5], 1mM EDTA, 2.8% sodium dodecylsulfate, and 20 mg of proteinase K/mL) and incubated for 10 hours at 56oC on a rocking platform. After lysis, samples were transferred to a spin columnc and centrifuged to reduce viscosity. Viral DNA was precipitated with absolute ethanol and extracted by use of a commercially available kit.d Phylogenetic studies have revealed that COPV and feline papillomavirus are closely related and that this group of viruses is closer Figure 1—Photomicrographs a section of an exophytic papilloma from a dog. Notice to some genera of human papillomavirus than hypergranulosis, koilocytes, and intranuclear viral inclusions (arrowheads) in ker- to papillomaviruses in other animals, includ- atinocytes of the stratum spinosum, stratum granulosum, and lower stratum corneum. H&E stain; bar = 1 mm (A) and 25 µm (B). ing bovine papillomavirus. Therefore, 2 sets of primers were designed. The first set of primers (PapE1-forward and PapE1-reverse) amplified DNA from COPV and feline papillomavirus, whereas the second set of primers (CP4, CP5, and PPF1) amplified human papillomaviruses, including the oncogenic strains. To amplify genomic sequences of canine, feline, or closely related papillomaviruses in clinical samples, nucleotide sequences conserved among known canine and feline papillomaviruses were reviewed and sequences encoding the E1 early gene were found to be the most highly conserved. Therefore, E1 sequences of feline (LOCUS AF480454) and canine (LOCUS NC001619) Figure 2—Photomicrographs of a section of a papilloma in a dog. Notice papilloma- papillomaviruses were aligned with the aim of tous epidermal hyperplasia with prominent koilocytosis and intranuclear viral inclu- designing consensus primer pairs able to sions (arrowheads) predominantly in the stratum spinosum. H&E stain; bar = amplify an approximate 341-bp fragment of 0.1 mm (A) and 25 µm (B).

AJVR, Vol 66, No. 10, October 2005 1765 37 04-12-0466r.qxp 9/14/2005 10:43 AM Page 1766

for 1 minute, 47oC for 1 minute, and 74oC for 1 minute, with Eight breeds were represented, and there were 2 West a final elongation step at 75oC for 5 minutes. Reaction mix- Highland White Terriers; 8 dogs were male, and 1 dog ture with no DNA served as a negative control, and COPV- was female. Age at the time hyperplastic skin lesions positive papilloma DNA samples and feline papilloma-posi- developed ranged from 6 months to 9 years (median, 3 tive DNA samples were used as positive controls. The PCR years). Two dogs (dogs 1 and 2) had 1 to 3 lesions with products were resolved via electrophoresis in 2% agarose gel the typical appearance of a papilloma. The other dogs stained with ethidium bromide. Amplified DNA was sequenced on an automated sequencer with fluorescent dye (dogs 3 to 9) had numerous variably pigmented, slight- terminator,e and sequences were compared with those includ- ly raised verrucous papules on the trunk and limbs. In ed in the GenBank database by use of alignment software.g dog 1, the skin nodules were removed surgically. In 5 Samples were considered to have positive results for dogs, the lesions regressed after the dose of cyclosporine detection of papillomavirus DNA if they had a band of the A was reduced and administration of antibimicrobials expected size after gel electrophoresis and if amplified DNA was instituted. In the remaining 3 dogs, lesions was sequenced and the protein encoded by the sequence had regressed spontaneously after cyclosporine A adminis- homology with the E1 protein of a previously established tration was discontinued or the dose was tapered. papillomavirus sequence. Comparisons were made with alignment software.f Histopathology—In dogs 1 and 2, examination of H&E-stained sections of biopsy specimens revealed Results severe focal epidermal hyperplasia and dysplasia, Clinical information—In 8 of the 9 dogs, adminis- koilocytosis, and intranuclear viral inclusions with tration of cyclosporine A at the median dosage of margination of chromatin (Figures 1 and 2). Focal 5 mg/kg every 24 hours was associated with the eruption hypergranulosis and coalescing keratohyalin granules of multiple hyperplastic and verrucous skin lesions. A also were observed in 1 of those dogs. Such changes single lesion developed in the other dog. The duration of were absent in sections from the other dog, suggesting treatment with cyclosporine A before development of that the differing cytopathic effects seen in the 2 dogs lesions varied from 1 to 24 months (median, 4 months). resulted from infection with different viruses.13 Among the remaining 7 dogs, microscopic findings in H&E-stained sections were similar. Findings included epidermal acanthosis without dysplasia, hypergranulosis, variable lymphocyte exocytosis, intraepidermal or intrafollicular pustules, and bacteria in the stratum corneum (Figure 3). The upper portion of the dermis contained bands of lymphocytes and plasma cells, although a true interface dermatitis was not detected. These features were considered similar to changes referred to as psoriasiform lichenoid dermatitis.14-16 Examination of stained sections revealed gram- negative rods in epidermal crypts in 1 dog and clusters of gram-positive cocci in the stratum corneum of hair follicle infundibula in dogs 3 to 9. Transmission electron microscopy revealed bacteria with features similar to those of staphylococci.

Figure 3—Photomicrographs of a section of a hyperplastic verrucous skin Figure 4—Photomicrograph of a portion of an exophytic papillo- lesion in a dog. Notice irregular epidermal hyperplasia with luminal (open ma from a dog. Notice that intranuclear inclusions in ker- arrowheads) and mural (solid arrowhead) folliculitis and a band of lym- atinocytes in the stratum granulosum are immunohistochemi- phocytes and plasma cells in the superficial portion of the dermis. cally stained for papillomavirus group-specific antigens. Lymphocytes are also in the lower epidermal layers. H&E stain; bar = 0.1 Diaminobenzidine chromogen with hematoxylin counterstain; mm (A) and 25 µm (B). bar = 10 µm.

1766 AJVR, Vol 66, No. 10, October 2005 38 04-12-0466r.qxp 9/14/2005 10:43 AM Page 1767

Figure 5—Photomicrographs of a section of a hyperplastic verrucous skin lesion in a dog. Notice intracorneal clusters of bacteria immunohistochemically stained (inset [arrowheads]) with an antiserum specific for papillomavirus antigens (A) but not with irrelevant control rabbit serum (B). Diaminobenzidine chromogen with hematoxylin counterstain; bar = 0.1 mm.

canine COPV E1 gene. In dog 1, papillomavirus DNA was amplified with the CP4, CP5, PPF1, and PapE1 primers (Figure 6). The amplified sequence was 98% homologous with that of COPV. In dog 2, papillomavirus DNA was amplified with the CP4, CP5, and PPF1 primers. The amplified sequence was 97% homologous with that of a recently described canine papillomavirus (GenBank No. AY725239). Moreover, this sequence was homologous at the predicted amino-acid level with E1 protein of human papillomavirus 63 (76% homology), bovine papillomavirus 3 (74% homology), and feline papillomavirus (72% homology). Papillomavirus DNA was not amplified from any other specimens from dogs 3 to 9.

Discussion Figure 6—Gel electrophoretogram of a polymerase chain reac- In humans, administration of cyclosporine A is tion assay performed with CP4, CP5, and PPF1 primers for detection of papillomaviruses in verrucous skin lesions in 7 often associated with numerous cutaneous adverse dogs. Notice a 450-bp amplicon that indicates papillomavirus effects.9,17,18 Most of those changes are associated with DNA in extracts of skin biopsy specimens from 2 dogs (lane 3 the development of viral, bacterial, or fungal [dog 1] and lane 4 [dog 2]) and a positive control specimen (pos [canine oral papillomavirus]). A similar amplicon is not evident infections. Papillomaviruses are the most frequently in a negative control specimen (neg) or lanes 5 to 9 (dogs 3 to reported virus detected in the associated infections, 7). Lanes 1 and 11 represent the ladder of molecular weight but herpes simplex and molluscum contagiosum virus markers. infections also have been recorded.9,17,19 Noninfectious Immunohistochemical analyses—In dogs 1 and changes such as hyperpigmentation, skin tags, lichen simplex, acne, cysts, and sebaceous hyperplasia are 2, the results of immunohistochemical staining con- 17,18,20 firmed the intranuclear keratinocyte viral inclusions reported less frequently. Follicular dystrophy, increased hair growth (hypertrichosis),21,22 and gingi- to be derived from papillomavirus. In dog 1, papillo- 23 mavirus inclusions were in nuclei in the stratum gran- val hyperplasia are frequently recorded. ulosum and lower stratum corneum, whereas in dog 2, Furthermore, compared with the general population, the inclusions were in cells from the stratum spin- the incidence of squamous cell carcinoma is higher in human patients treated with cyclosporine A for longer osum to the lower stratum corneum (Figure 4). In 24 dogs 3 to 9, intrakeratinocyte staining with the papil- than 2 years. lomavirus-specific antiserum was not detected. In dogs, lesions resembling psoriasiform lichenoid dermatitis have been reported in association with However, there was bacterial uptake of stain in the 2,10,25 stratum corneum (Figure 5). administration of cyclosporine A. Gingival hyperplasia and hypertrichosis have been reported to be rare PCR assay—The PCR assay detected papillo- adverse drug events.7,11,25,26 Results of our study indicat- mavirus DNA from sections of the 3 positive con- ed that psoriasiform lichenoid dermatitis was the most trols with both sets of primers, and the sequence of common diagnosis for hyperplastic verrucous skin amplicons was 99% homologous with that of the lesions in 9 dogs treated with cyclosporine A. It has

AJVR, Vol 66, No. 10, October 2005 1767 39 04-12-0466r.qxp 9/14/2005 10:43 AM Page 1768

been hypothesized10 that this reaction is induced by cyclosporine A-induced skin lesions develop most fre- staphylococcal infection, a premise that was supported quently in humans and dogs treated with high doses, by the observation of cocci in 6 of 9 specimens in our for extended periods of time, or with a combination of study. However, in 3 of 8 dogs, all lesions regressed immunosuppressive drugs.1,25 Most of those changes without the administration of antimicrobials after dis- regress after treatment is discontinued. continuation or decreasing the administration of Our results suggested that hyperplastic and verru- cyclosporine A. Immunohistochemical staining for cous skin lesions observed in dogs after cyclosporine A papillomavirus with the polyclonal antiserum stained administration may have multiple causes. In most bacteria in the stratum corneum. In 1 dog with such dogs, the lesions are typical of those characterized as bacteria, transmission electron microscopy revealed lichenoid psoriasiform dermatosis, but infection by bacteria with features consistent with staphylococci in papillomaviruses may develop in some dogs. In all the stratum corneum but no papillomavirus. Thus, instances, decreasing or discontinuing administration care must be taken in the interpretation of immuno- of cyclosporine A, with or without concurrent admin- staining for papillomavirus with this reagent. istration of antimicrobials, resulted in regression of the Several pharmacologic properties of cyclosporine lesions. A can explain the development of hyperplastic lesions in the skin of dogs and humans. Cyclosporine A mod- a. B0580, Dako Corp, Carpinteria, Calif. ifies cytokine secretions in several cell types.27 In b. Biogenex, San Ramon, Calif. humans, development of gingival hyperplasia is caused c. QIA shredder, Qiagen, Basel, Switzerland. by increased production of extracellular matrix in asso- d. QIAamp DNA mini kit, Qiagen, Basel, Switzerland. e. Pfu Turbo DNA polymerase, Stratagene, La Jolla, Calif. ciation with secretion of transforming growth 28 f. Applied Biosystems, Foster City, Calif. factor-β. Moreover, inhibition by cyclosporine A of g. Blast, version 2.2.11. Available at: www.ncbi.nlm.gov/BLAST. the calcineurin-nuclear factor of activated T-cell 1 Accessed Jan 15,0205. pathway in follicular keratinocytes stimulates hair 21 growth and induces hypertrichosis. References Results of our study also suggested that 1. Gupta AK, Brown MD, Ellis CN, et al. Cyclosporine in der- cyclosporine A-induced verrucous skin lesions can matology. J Am Acad Dermatol 1989;21:1245–1256. sometimes be associated with infections with various 2. Olivry T, Rivierre C, Jackson HA, et al. Cyclosporine papillomaviruses, for which emergence might be pro- decreases skin lesions and pruritus in dogs with atopic dermatitis: a blinded randomized prednisolone-controlled trial. Vet Dermatol moted by suppression of cell-mediated immunity. 2002;13:77–87. However, it is not known whether treatment with 3. Olivry T, Steffan J, Fisch RD, et al. Randomized controlled cyclosporine A favors recurrence of a latent infection trial of the efficacy of cyclosporine in the treatment of atopic derm- or promotes de novo infection. In dog 1, clinical find- atitis in dogs. J Am Vet Med Assoc 2002;221:370–377. ings, histologic findings, and results of immunohisto- 4. Fontaine J, Olivry T. Treatment of canine atopic dermatitis chemical and PCR testing were consistent with those with cyclosporine: a pilot clinical study. Vet Rec 2001;148:662–663. 29 5. Olivry T, Mueller RS. Evidence-based veterinary dermatol- typically associated with infection with COPV ogy: a systematic review of the pharmacotherapy of canine atopic (LOCUS NC001619). In dog 2, results of histologic dermatitis. Vet Dermatol 2003;14:121–146. and immunohistochemical staining were different from 6. Steffan J, Alexander D, Brovedani F, et al. Comparison of those in dog 1, mainly with regard to koilocytes in the cyclosporine A with methylprednisolone for treatment of canine stratum spinosum and the absence of hypergranulosis atopic dermatitis: a parallel, blinded, randomized controlled trial. Vet or coalescing keratohyalin granules.13 These differences Dermatol 2003;14:11–22. 7. Mathews KA, Sukhiani HR. Randomized controlled trial of may have indicated infection with a virus other than cyclosporine for treatment of perianal fistulas in dogs. J Am Vet Med COPV. Moreover, DNA from papillomavirus was only Assoc 1997;211:1249–1253. amplified by use of the CP4, CP5, PPF1 set of primers; 8. Linek M, Boss C, Haemmerling R. Effects of cyclosporine A the subsequent amplification indicated infection with a on clinical and histological abnormalities in dogs with sebaceous recently described papillomavirus (GenBank No. adenitis. J Am Vet Med Assoc 2005;226:59–64. AY725239). In humans, the use of degenerated primers 9. Iraji F, Kiani A, Shahidi S, et al. Histopathology of skin lesions with warty appearance in renal allograft recipients. Am J has broadened the spectrum of capability of PCR Dermatopathol 2002;24:324–325. assays and enabled detection of unknown papillo- 10. Werner AH. Psoriasiform-lichenoid-like dermatosis in three 30 maviruses. It is possible that the epidermis of affected dogs treated with microemulsified cyclosporine A. J Am Vet Med dogs may be infected by papillomaviruses that are dif- Assoc 2003;223:1013–1016. ficult to detect by use of traditional techniques that 11. Seibel W, Sundberg JP, Lesko LJ, et al. Cutaneous papillo- amplify COPV DNA. Likewise, it is possible that the matous hyperplasia in cyclosporine-A treated beagles. J Invest Dermatol 1989;93:224–230. cyclosporine A-induced hyperplastic skin lesions in 12. Iftner A, Klug SJ, Garbe C, et al. The prevalence of human dogs 3 to 9 might have been caused by papillo- papillomavirus genotypes in nonmelanoma skin cancers of nonim- maviruses that were undetectable with available tech- munosuppressed individuals identifies high-risk genital types as pos- niques. In humans, DNA of papillomavirus is often sible risk factors. Cancer Res 2003;63:7515–7519. amplified from lesions of psoriasis, a dermatologic con- 13. Croissant O, Breitburd F, Orth G. Specificity of cytopathic effect dition with similarities to canine psoriasiform of cutaneous human papillomaviruses. Clin Dermatol 1985;3:43–55. 31 14. Gross TL, Halliwell RE, McDougal BJ, et al. Psoriasiform lichenoid dermatosis. It has been hypothesized that lichenoid dermatitis in the springer spaniel. Vet Pathol 1986;23:76–78. papillomavirus induces autoimmune reactions in the 15. Mason KV, Halliwell RE, McDougal BJ. Characterization of epidermis and could play a role in the development of lichenoid-psoriasiform dermatosis of springer spaniels. J Am Vet Med such lesions. Finally, it must be kept in mind that Assoc 1986;189:897–901.

1768 AJVR, Vol 66, No. 10, October 2005 40 04-12-0466r.qxp 9/14/2005 10:43 AM Page 1769

16. Yager JA, Wilcock BP. Interface dermatitis. In: Yager JA, 24. Paul CF, Ho VC, McGeown C, et al. Risk of malignancies in Wilcock BP, eds. Color atlas and text of surgical pathology of the dog psoriasis patients treated with cyclosporine: a 5 y cohort study. and cat dermatopathology and skin tumors. London: Mosby Year Book J Invest Dermatol 2003;120:211–216. Inc, 1994;85–106. 25. Rosenkrantz WS, Griffin CE, Barr RJ. Clinical evaluation of 17. Euvrard S, Kanitakis J, Cochat P, et al. Skin diseases in chil- cyclosporine in animal models with cutaneous immune-mediated disease dren with organ transplants. J Am Acad Dermatol 2001;44:932–939. and epitheliotropic lymphoma. J Am Anim Hosp Assoc 1989;25:377–384. 18. Lugo-Janer G, Sanchez JL, Santiago-Delpin E. Prevalence 26. Seibel W, Yahia NA, McCleary LB, et al. Cyclosporine- and clinical spectrum of skin diseases in kidney transplant recipi- induced gingival overgrowth in beagle dogs. J Oral Pathol Med 1989; ents. J Am Acad Dermatol 1991;24:410–414. 18:240–245. 19. Van der Leest RJ, Zachow KR, Ostrow RS, et al. Human 27. Esposito C, Fornoni A, Cornacchia F, et al. Cyclosporine papillomavirus heterogeneity in 36 renal transplant recipients. Arch induces different responses in human epithelial, endothelial and Dermatol 1987;123:354–357. fibroblast cell cultures. Kidney Int 2000;58:123–130. 20. Bencini PL, Montagnino G, Sala F, et al. Cutaneous lesions 28. Stabellini G, Carinci F, Luigi Bedani P, et al. Cyclosporin A in 67 cyclosporin-treated renal transplant recipients. Dermatologica and transforming growth factor beta modify the pattern of extracel- 1986;172:24–30. lular glycosaminoglycans without causing cytoskeletal changes in 21. Gafter-Gvili A, Sredni B, Gal R, et al. Cyclosporin A- human gingival fibroblasts. Transplantation 2002;73:1676–1679. induced hair growth in mice is associated with inhibition of cal- 29. Yager JA, Wilcock BP. Squamous papilloma. In: Yager cineurin-dependent activation of NFAT in follicular keratinocytes. J,Wilcock BP, eds. Color atlas and text of surgical pathology of the dog Am J Physiol Cell Physiol 2003;284:C1593–C1603. and cat: dermatopathology and skin tumors. London: Mosby Year Book 22. Heaphy MR Jr, Shamma HN, Hickmann M, et al. Inc, 1994;251–252. Cyclosporine-induced folliculodystrophy. J Am Acad Dermatol 30. Meyer T, Arndt R, Christophers E, et al. Importance of 2004;50:310–315. human papillomaviruses for the development of skin cancer. Cancer 23. Wysocki GP, Gretzinger HA, Laupacis A, et al. Fibrous Detect Prev 2001;25:533–547. hyperplasia of the gingiva: a side effect of cyclosporin A therapy. Oral 31. Majewski S, Jablonska S, Favre M, et al. Papillomavirus and Surg Oral Med Oral Pathol 1983;55:274–278. autoimmunity in psoriasis. Immunol Today 1999;20:475–476.

AJVR, Vol 66, No. 10, October 2005 1769 41

Chapter 6

Detection of Novel Papillomaviruses in Canine Mucosal, Cutaneous and in situ Squamous Cell Carcinomas

N. Zaugg1, G. Nespeca1, B. Hauser2, M. Ackermann3, C. 1 Favrot

Veterinary Dermatology, 2005; 16: 290-298

1 Clinic for Small Animal Internal Medicine, Dermatology Unit, Vetsuisse Faculty, University of Zurich, Zurich, Switzerland. 2 Institute for Veterinary Pathology, Vetsuisse Faculty, University of Zurich, Zurich, Switzerland 3 Virology Institute, Vetsuisse Faculty, University of Zurich, Zurich, Switzerland

The study has been funded by grants of the Waltham Foundation and The European Society of Veterinary Dermatology (ESVD)

Veterinary Dermatology 2005, 16, 290–298

BlackwellDetection Publishing, Ltd. of novel papillomaviruses in canine mucosal, cutaneous and in situ squamous cell carcinomas

N. ZAUGG*, G. NESPECA*, B. HAUSER†, M. ACKERMANN‡ and C. FAVROT*

*Clinic for Small Animal Internal Medicine, Dermatology Unit, †Institute for Veterinary Pathology and ‡Institute for Virology, Vetsuisse Faculty, University of Zürich, Switzerland

(Received 31 January 2005; accepted 20 May 2005)

Abstract Papillomavirus (PV) DNA is frequently uncovered in samples of human skin squamous cell carcinomas (SCC). However, the role of these viruses in the development of such cancers in canine species remains controversial. While approximately 100 human PVs are known, only one single canine oral PV (COPV) has been identified and studied extensively. Therefore, we applied a narrow-range polymerase chain reaction (PCR) suitable for the detection of classical canine and feline PVs, as well as a broad-range PCR, which has been used for the detection of various novel PVs in humans, in order to analyse 42 paraffin-embedded samples, representing three different forms of canine SCCs. Ten samples of skin tissues with various non-neoplastic conditions served as controls. While none of the negative controls reacted positively, PV DNA was discovered in 21% of the tested SCC samples. Interestingly, the classical COPV was amplified from only one sample, while the other positive cases were associated with a variety of thus far unknown PVs. This study suggests that a fraction of canine SCC is infected with PVs and that a genetic variety of canine PVs exists. Therefore, these results will facilitate the future study of the role of PVs in the development of canine skin cancers.

INTRODUCTION The aetiology of canine skin and mucous membrane SCC remains largely unknown, although environmental Squamous cell carcinomas (SCC) are malignant factors such as sunlight exposure or burns have been tumours that arise from the squamous epithelium of implicated.1,2 The role of papillomaviruses (PV) also the skin and the mucous membranes. Squamous cell remains unclear and very few reports have confirmed carcinomas account for up to 5% of the skin tumours the association between PV and SCC in dogs.10–13 in dogs and are the most frequent malignant canine Additionally, the role of these viruses in the develop- tumours of the digits, tongue and gingiva.1,2 Cutane- ment of cancer has not been established and the ous SCCs may be either exophytic or ulcerative, genomes of potentially causative viruses have not yet whereas mucous membrane SCCs are usually exo- been cloned and analysed. phytic.1,2 Aside from these two classical forms of inva- Similarly, the role of human PVs (HPV) in the deve- sive SCC, a case of multifocal in situ SCC arising from lopment of human skin SCC remains controversial. pigmented papules and plaques has been described in a HPV DNA is, however, frequently uncovered in at least dog.3 In this case, tumoral cells were confined to the three types of human skin SCCs: Epidermodysplasia epidermis, while the basal membrane remained verruciformis, Bowen’s disease and Bowenoid papulo- unaffected. Furthermore, cases of canine multiple sis.7,9,14,15 Additionally, links between HPVs and cervi- pigmented plaques that evolved into SCC have been cal and anal human SCCs have been well demonstrated reported.4,5 and the causality established.16–18 In humans, solar keratosis, Bowen’s disease and Papillomaviruses are host-specific epitheliotropic Bowenoid papulosis are regarded as forms of in situ DNA viruses that infect skin and mucous mem- SCC.6–9 Some of them subsequently develop into inva- branes. The complete genome and the biological prop- sive SCCs. Mucosal forms of SCCs also affect the erties of only one canine papillomavirus are well cervical, anal and oral mucous membranes. Further- known.19–21 However, the existence of up to six dif- more, Epidermodysplasia verruciformis is a rare genetic ferent types has been suggested.4,21–23 predisposition to develop viral warts with high risk of In contrast, classification of HPVs has recently carcinomatous transformation. been reviewed, and nearly 100 HPV types have been described based on isolation and sequencing of complete genomes.15 de Villiers has additionally proposed crite- The study has been funded by grants of the Waltham Foundation and ria to define genera, species, types, subtypes and vari- The European Society of Veterinary Dermatology (ESVD). ants within the papillomaviridae family.15 It has also Correspondence: C. Favrot, Clinic for Small Animal Internal Medicine, Dermatology Unit, Vetsuisse Faculty Zürich, recently been shown that phylogenetic classification Winterthurerstrasse 260, CH-8057 Zürich, Switzerland. E-mail: based on the L1 gene of PVs correlates, at least par- [email protected] tially, with the biological and pathological properties.15

Detection of novel papillomaviruses 291

Papillomavirus infection may either be subclinical, to a QIA shredder column (Quiagen, Basel, Switzer- or induce microlesions or benign neoplasias.14,15,24,25 land) and centrifuged in order to reduce viscosity. Additionally, a subset of HPVs and animal PVs is Desoxyribonucleic acid was precipitated with absolute clearly implicated in the development of cancer in ethanol and extracted with the QIAamp® DNA Mini humans and animals.14,18,26–28 In humans, these PVs Kit (Quiagen). cause mucosal carcinomas and are referred to as high- risk PVs.14 Papillomavirus detection and sequencing The nature and the biological properties of canine Phylogenetic studies have shown that COPV and feline PVs, as well as the aetiology of canine SCC, remain PV are closely related and that this group is closer to largely unknown.2 As HPVs that induce benign warts some human PVs, including oncogenic PVs, than to in humans are usually different from those that induce PV in other animals, such as bovine PV.15 The investi- cancer, it can be speculated that the well known canine gators consequently selected two sets of primers: the oral PV (COPV) does not usually induce SCC in dogs first one (PapE1-Forward, PapE1-Reverse) is designed and that some other unknown canine PVs, the canine to amplify specifically COPV and FdPV DNA and the counterparts of the high-risk HPVs, may be respon- second one (CP4, CP5, PPF1) is designed to amplify sible or contribute to such development.21 various HPV DNA, especially the oncogenic ones. The purpose of our study was therefore to detect a broad spectrum of PV DNAs in samples from three Narrow-range PCR with PapE1 primers forms of canine SCC (cutaneous invasive, cutaneous All samples were coded before being assayed. The in situ and mucosal invasive) and to sequence the sequences encoding E1 are most highly conserved amplified DNA. amongst canine, feline or closely related PVs.29 Therefore, the E1 sequences of feline [LOCUS AF480454] and canine [LOCUS NC001619] papillomaviruses were MATERIALS AND METHODS aligned with the aim of designing degenerated consensus primer pairs able to amplify an estimated 341-bp Materials fragment from both phylogenetically related genomes. Fifty-seven samples of paraffin-embedded skin were The forward primer used was 5′-ATGGCGGM- included in the study: TARAAAAGGTA-3′ and the reverse primer used was 5′-AACAGCTGYTTTTTARCYTTTTT-3′. To • seventeen samples of canine invasive cutaneous amplify an internal 184-bp fragment using the same SCC; forward primer, a second reverse primer 5′-GAAACA- • twenty-three samples of canine mucosal SCC; GTTGCAGGGAAAGTC-3′ was designed. • two samples of canine in situ SCC; The PCR reactions were performed in 30-µL volumes, µ   • ten samples of canine skin with various nontumoral containing 1 L of genomic DNA, 50 m KCl,, 3 m µ µ conditions; KCl2, 200 of each dNTP, 0.3 each of consensus • three samples of virus-induced canine wart; sense and antisense primers and 2.5 U of PfuTur- • one sample of virus-induced feline in situ SCC; and boDNA polymerase (Stratagene, CA, USA). Polymer- • one sample of virus-induced bovine fibropapilloma. ase chain reaction amplification involved an initial denaturation step at 95 °C for 4 min, followed by 30 Except for one case (#2 provided by Dr T. L. Gross), cycles at 95 °C for 1 min, 50 °C for 1 min and 74 °C for all samples were selected by one board-certified patho- 1 min, with a final elongation step at 74 °C for 5 min. logist (BH) at the Institute of Veterinary Pathology of Reaction mixture with no DNA served as negative con- the University of Zürich, Switzerland. trol, and COPV-positive papilloma DNA samples and feline papilloma positive DNA samples were used as Methods positive controls. The PCR products were resolved by The study was carried out using a PCR technique on electrophoresis in 2% agarose gel stained with ethid- formalin-fixed, paraffin-embedded samples. Thirty-µm- ium bromide. Amplified DNA was sequenced using thick sections of each sample were cut from each tissue AB-3100-based fluorescent sequencing and BigDye block, using a new disposable microtome blade for terminator chemistry. each block. Rigorous precautions were taken in order to avoid cross-contamination between samples. Broad-range PCR with CP4, CP5 & PPF1 primers In order to amplify as many different PVs as possible, DNA extraction. Each section was deparaffinized twice the CP4, CP5, PPF1 was selected, because of its ability with 1.2 mL xylene at room temperature for 10 min, to uncover up to 64 different HPVs.29 washed with ethanol 100% and then dried. The desiccated The PCR reactions were performed in 30-µL vol- samples were suspended in an ATL lysis buffer (50 m umes, containing 1 µL of genomic DNA, 50 m KCl,   µ µ Tris-HCl, pH 8.5 1 m ethilenediaminetetraacetic 3 m KCl2, 200 of each dNTP, 0.45 of the CP4 acid, 2.8% sodium dodecylsulphate and 20 mg mL−1 and CP5 primers and 0.3 µ of the PPF1 primer, Proteinase K) and incubated at 56 °C on the rocking and 2.5 U of PfuTurboDNA polymerase (Stratagene). platform overnight. After lysis, samples were transferred Polymerase chain reaction amplification involved an

292 N. Zaugg et al. initial denaturation step at 95 °C for 10 min, followed by 40 cycles at 95 °C for 1 min, 47 °C for 1 min and 74 °C for 1 min, with a final elongation step at 75 °C for 5 min. Reaction mixture with no DNA served as negative control, and COPV-positive papilloma DNA samples and feline papilloma positive DNA samples were used as positive controls. The PCR products were resolved by electrophoresis in 1% agarose gels stained with ethidium bromide. Amplified DNA was sequenced using AB-3100-based fluorescent sequencing and BigDye terminator chemistry, and obtained sequences were compared with entries in the GenBank database.

Interpretation of the results and sequence analyses Samples were deemed positive if the two following criteria were fulﬁlled:

• samples exhibited a band of the expected size after gel electrophoresis; • the ampliﬁed DNA exhibited a signiﬁcant homology with DNA coding for the E1 protein of a previously established PV. Comparisons were made Figure 2. Histological section from lesion in Fig. 1. H&E. Canine with the BLAST software (GenBank – National SCC in situ. Irregular acanthosis (white arrow), hyperpigmentation Center of Biotechnology Information: NCBI). and pigmentary incontinence (blue arrow), follicular involvement (green arrow). The basement membrane is intact (yellow arrow). µ Sequences were subsequently compared to each Bar: 200 m. other on the amino acid sequence level in order to establish their homology on the protein level.

RESULTS

Clinical and histological criteria Available SCC cases were assigned to clinical and histological groups according to the following criteria. Two dogs (cases 1 and 2) exhibited numerous hyperpigmented, scaly maculae, plaques and nodules. One of the plaques of case 1 ulcerated and was subsequently biopsied (Fig. 1). Case 2 exhibited several ulcerated nodules that were biopsied. Histopathological examination of the two cases revealed marked acanthosis,

Figure 3. Histological section from lesion in Fig. 1. H&E. Canine in situ SCC. Proliferation of basaloid cells (red arrow). Clumped keratohyalin granules (white arrow), hyperpigmentation (blue arrow), anisocaryosis (double arrow). Bar: 50 µm.

orthokeratotic hyperkeratosis and hypergranulosis (Fig. 2) with keratohyalin granule clumping (Fig. 3). The epidermis was disorganized, with numerous atypical cells and premature keratinization. Proliferation of basaloid cells was observed in some areas but most of the atypical cells were of the squamous type (Fig. 3). Atypia consisted of macrokaryosis, anisokaryosis (Fig. 3), hyperchromasia, prominent nucleoli, multinucleated cells and abnormal mitoses (Fig. 4). However, Figure 1. In situ squamous cell carcinoma. Hyperpigmented plaque. the basement membrane was intact and the dermis Case 1. was not affected (Figs 2, 3 and 4). Therefore, these dogs

Detection of novel papillomaviruses 293

Figure 4. Histological section from lesion in Fig. 1. H&E. Canine in Figure 6. Histology from lesion in Fig. 5. H&E. Canine invasive situ SCC. Intact basement membrane (green arrow). Numerous SCC: cords (red arrow) of keratinocytes invade the dermis. The atypias: giant tumour cells/abnormal mitosis (red arrow), epidermis is ulcerated (left, white arrow) and hyperkeratotic (right, prominent nucleoli (yellow arrow), multinucelated cells (blue arrow). blue arrow). Horn pearls (green arrow). Bar: 200 µm. Bar: 50 µm.

Figure 7. Histology from lesion in Fig. 5. H&E. Canine invasive Figure 5. Invasive squamous cell carcinoma. Clawbed. SCC. Cords of atypical keratinocytes. Several malignancy criteria are present: numerous mitoses (white arrows), macrokaryosis (green arrow), prominent nucleoli (blue arrow). Bar: 50 µm. were considered to represent cases of in situ SCC (Table 1). and the canine papillomas exhibited changes typical of Seventeen additional cases represented the group of papillomavirus infections, such as koilocytosis, clump- skin-derived invasive SCC. Some cases exhibited a ing of the keratohyalin granules and viral inclusion proliferative, exophytic and invasive pattern (Fig. 5), bodies (data not shown). The feline in situ SCC (data whereas others were also invasive but more ulcerative not shown) had previously been shown by immuno- (Table 1). Histologically, they all consisted of cords chemistry to react positively with papillomavirus- or islands of atypical cells that invaded the dermis specific antibodies. (Fig. 6). Large nuclei with prominent nucleoli were present in all cases, as well as numerous mitoses PCR studies (Fig. 7). Premature keratinization and intercellular Papillomavirus DNA was detected in the positive con- bridges were also present in most of the samples. trol tissues but not in the negative control tissues. The Finally, 23 cases of mucous membrane-derived invasive narrow-range PCR, optimized to detect known feline SCC were available, which histologically all exhibited and canine papillomaviruses, reacted positively with similar changes as those described above for skin- extracts from the canine papillomas as well as with the derived SCCs. Seventeen of the latter lesions arose immunohistologically positive case of feline in situ from the gingiva, four from the tongue, one from the SCC. In contrast, this set of primers was unable to nasal mucosa and one from the lips. discover bovine papillomavirus in extracts from the Three canine papillomas, one feline squamous cell bovine fibropapilloma tissue. However, the broad-range carcinoma in situ and one bovine fibropapilloma were PCR detected papillomavirus DNA in both carnivorous included as positive controls. The bovine fibropapilloma and bovine positive control tissues.

Table 1. Identiﬁcation and properties of clinical SCC cases

Case # Group Breed Sex Age (years) Localization 1 In situ Rhodesian ridgeback M 4 Diffuse 2 In situ Retriever mixed F 5 Abdomen, limbs 3 Skin German shepherd dog F 13 Limb 4 Skin Flatcoat retriever F 12 Clawbed 5 Skin Golden retriever M 13 Nasal planum 6 Skin Giant schnauzer F 10 Clawbed 7 Skin Schnauzer M Unknown Clawbed 8 Skin Giant schnauzer F 8 Clawbed 9 Skin Bernese mountain dog M 7 Dorsum 10 Skin Weimaraner M 8 Flank 11 Skin Golden retriever M 12 Limb 12 Skin Flatcoat retriever M 8 Limb 13 Skin Giant schnauzer M 7 Carpus 14 Skin Siberian husky F 11 Carpus 15 Skin Mongrel M 10 Limb 16 Skin Giant schnauzer F 9 Clawbed 17 Skin Mongrel M 10 Clawbed 18 Skin Weimaraner M 8 Face 19 Skin Labrador retriever M 4 Clawbed 20 Mucous membrane English cocker M 7 Nose 21 Mucous membrane Yorkshire terrier F 8 Gingiva 22 Mucous membrane Golden retriever M 10 Gingiva 23 Mucous membrane Airedale terrier M 9 Gingiva 24 Mucous membrane Standard poodle F 10 Tongue 25 Mucous membrane Standard poodle M 7 Gingiva 26 Mucous membrane Yorkshire terrier M 12 Tongue 27 Mucous membrane Golden retriever M 5 Gingiva 28 Mucous membrane Cocker spaniel M 14 Gingiva 29 Mucous membrane Old English sheepdog F 11 Gingiva 30 Mucous membrane Maltese F 10 Gingiva 31 Mucous membrane Irish setter M 8 Gingiva 32 Mucous membrane Irish wolfhound F 7 Lips 33 Mucous membrane West Highland white terrier F 12 Gingiva 34 Mucous membrane German Wachtelhund F 11 Tongue 35 Mucous membrane Pekingese F 11 Gingiva 36 Mucous membrane Mongrel M 13 Gingiva 37 Mucous membrane Golden retriever M 11 Gingiva 38 Mucous membrane Pekingese F 9 Gingiva 39 Mucous membrane Schnauzer M 13 Gingiva 40 Mucous membrane Briard M 6 Gingiva 41 Mucous membrane Appenzeller F 12 Tongue 42 Mucous membrane Golden retriever M 4 Gingiva

Table 2. Papillomavirus-positive cases

Case # Group* Breed Sex Age Localization Related to† 1 In situ Rhodesian ridgeback M 4 Diffuse HPV65‡ 2 In situ Retriever mix F 5 Abdomen, limbs BPV1 11 Invasive skin Golden retriever M 12 Limb HPV85 18 Invasive skin Weimaraner M 8 Face HPV59 20 Mucous membrane Cocker M 7 Nose COPV§ 33 Mucous membrane West Highland white terrier F 12 Gingiva HPV65 36 Mucous membrane Mongrel M 13 Gingiva HPV59 40 Mucous membrane Briard M 6 Gingiva HPV65 42 Mucous membrane Golden retriever M 4 Gingiva HPV4 *Clinical SCC type; †closest relative detected by NCBI-BLAST analysis; ‡unless otherwise stated, detected exclusively by broad-range PCR; §detected by both narrow-range and broad-range PCR.

The results of the test samples included in this situ SCC, 2/17 invasive skin SCCs, and 5/23 cases of study are presented in Table 2. Interestingly, the narrow- mucous membrane-derived invasive SCC. range PCR detected only one case of SCC associated with a papillomavirus infection, namely a mucous Sequencing studies membrane-derived SCC (case 20). However, this case The above PCR results strongly suggested that thus far and eight additional cases were detected with the unknown papillomaviruses had been detected with the broad-range technique, including 2/2 samples from in broad-range PCR. To address this issue, the nucleotide

© 2005 European Society of Veterinary Dermatology, Veterinary Dermatology, 16, 290–298 47 Detection of novel papillomaviruses 295 sequences of the amplification products of the indi- the use of formalin-fixed, paraffin-embedded tissue vidual PCRs were determined and compared by BLAST decreases the amplification rate of viral DNA.31 Our analysis to known papillomavirus sequences. If avail- study demonstrates that COPV DNA is rarely present able (7/42 dogs), several independent samples from each in SCC samples. dog were used for PCR and sequencing, and sequenc- The absence of PV DNA in the nontumoral skin in ing results were identical for different locations of the present study suggests that dogs, unlike humans, the lesions from each individual dog. However, the are rarely affected by occult infections. This conclusion sequences differed largely between different dogs. As is corroborated by Antonnsson et al. who have un- expected, the sequence obtained from case 20 turned covered PV DNA in the skin of various healthy animals out to be closely related to the published sequence of but not in dogs and cats.24 canine oral papillomavirus.21 In contrast, the remain- Invasive skin SCCs appear to be infrequently ing eight positive cases were more closely related to infected by PVs (2/17: 12%). These results are in other papillomaviruses, such as human and bovine contrast with those of mucous membrane SCCs that papillomaviruses (Table 2). Although the classification appeared to be more frequently affected (5/23: 22%). of papillomavirus is based on the L1 sequences, these Last but not least, the two cases of skin in situ SCC results strongly support that novel canine papillomavi- were deemed positive. The viral aetiology of this latter ruses were detected in canine SCCs throughout this condition has already been suggested by two different study. However, 78% of the tested samples still remained case reports.4,5 However, the presence of PV DNA in negative, which indicated that either a large proportion the two tested samples of canine in situ SCC are not of canine SCC is not associated to papillomavirus proof that these viruses are directly responsible for the infection or that the corresponding virus strains are development of these tumours. In fact, establishing still more different, which would mean that the range causality between papillomaviruses and skin cancers of PCR detection would still need to be further remains problematic. Criteria for causality were first enlarged. proposed by zur Hausen and have been modified by Harwood.16,17 According to these authors, epidemio- Breed, sex and age distribution logical evidence that the viral infection represents a These data are included in Tables 1 and 2. Although risk factor, regular presence of the viral nucleic acid, schnauzers appeared to be over-represented in the total stimulation of proliferation upon cross-infection of the sample, retrievers (flatcoated and golden) (19% of the viral genome in tissue culture cells and demonstration sample but 33% of papillomavirus-positive individuals) that induction of proliferation depends on functions of emerged as a possible breed group with a tendency to the viral DNA are essential criteria to demonstrate SCC caused by papillomavirus. The total sex distribu- causality. At the present time, not all these criteria are tion resulted in 60% male and 40% female dogs. How- fulfilled by the canine in situ SCCs. ever, 70% of the papillomavirus-positive dogs were The clinical relevance of the presence of PVs in inva- male. The average age of dogs with SCC was 9.25 years, sive mucosal and skin SCC lesions is another import- with in situ SCC, 4.5 years, with invasive skin SCC, ant question. Assuming that PVs play a role in the 9.38 years, and of dogs with mucous SCC, 9.57 years. development of canine SCC, the absence of PV DNA The average age of papillomavirus-positive dogs with in numerous canine SCC can be explained by the ‘hit SCC was 7.89 years. and run’ model, which postulates an initial transformation of the infected cell and a subsequent loss of PV-DNA.32 Interestingly, in a previous study in canines, DISCUSSION a PV antigen-negative SCC arose from a PV antigen- positive Epidermodysplasia verruciformis-like lesion.4 In this study, it was possible to uncover PV DNA in This model, however, implies a deletion of viral genes 9/42 samples of canine SCC (21.4%) with a broad-range during integration of the viral DNA in the host chro- PCR-assay and in 1/42 (2.3%) with a narrow-range mosomes. Such a deletion has only been demonstrated PCR-assay designed to amplify DNA from COPV, in humans with the E2 gene.33 As human high-risk FdPV and closely related PVs. Interestingly, the last papillomaviruses are genetically different from low- figure is in line with that of the only other study which risk HPVs one can postulate that canine high-risk PV used a PCR technique:12 Teifke et al. used a narrow- (provided they do exist) should be genetically different range PCR and amplified COPV DNA in 3/53 SCC from COPV. As we have only used sets of primers that samples (6%). These numbers are, however, lower than were designed to uncover COPV DNA and high-risk those obtained in previously published immunohisto- HPV DNA, we cannot rule out that the DNA of some chemistry studies. In one of these studies, PV antigens unknown canine PV remained undetectable. were demonstrated in 10/100 canine SCC samples, with The amplified DNA sequences suggest the presence 17 additional questionable positive results (10–27%).11 of several different PVs in the canine SCCs. Further- In the second one, PV antigens were uncovered with an more, eight out of nine positive samples were infected immunoperoxidase technique in 6/20 (30%) samples.30 by these unknown PVs. Moreover, our findings con- It is, however, important to consider that the PCR firm that dogs, as well as humans, can be infected with technique cannot amplify the DNA of all PVs and that PVs of great genetic diversity.

The detection of PV DNA in SCC tissues and not in 3. Gross TL, Fau-Brimacomb BH. Multifocal intraepider- normal skin or skin affected by other conditions might mal carcinoma in a dog histologically resembling Bowen’s result from the increased replication of latent virus in disease. American Journal of Dermatopathology 1986; response to tumoral cell cytokine secretions.34 8: 509–15. Mitsuishi et al. have uncovered HPV DNA in 74 and 4. Nagata M, Nanko H, Moriyama A et al. Pigmented plaques associated with papillomavirus infection in dogs: 67% of the samples of human actinic keratosis and 9 is this epidermodysplasia verruciformis? Veterinary Bowen’s diseases, respectively. In the same study, no Dermatology 1995; 6: 179–85. differences in p53, p21, Ki 67 and PCNA were found 5. Stokking LB, Ehrhart EJ, Lichtensteiger CA et al. between HPV-positive and HPV-negative samples. This Pigmented epidermal plaques in three dogs. Journal of set of data suggests that HPVs probably play a role in the American Animal Hospital Association 2004; 40: the pathogenesis of both conditions but that the viruses 411–17. alone are not able to induce cancer transformation. 6. Anwar J, Wrone DA, Kimyai-Asadi A et al. The deve- Human SCCs also occur many years after the initial lopment of actinic keratosis into invasive squamous cell infection, which usually has a benign course, even with carcinoma: evidence and evolving classification schemes. oncogenic HPV types.14 Malignant transformation Clinical Dermatology 2004; 422: 189–96. thus implies the presence of continuing infection and 7. Arlette JP, Trotter MJ. Squamous cell carcinoma in situ of the skin: history, presentation, biology and treatment. prolonged E6/E7 oncogene expression.35 Such chronic Australasian Journal of Dermatology 2004; 45: 1– infections occur in human anal or cervical PV infec- 11. tions and Epidermodysplasia verruciformis but have 8. Cockerell CJ. Pathology and pathobiology of the actinic 20 not been demonstrated with canine PV infections. (solar) keratosis. British Journal of Dermatology 2003; Although the present results cannot be regarded as 149: 34–6. sufficient proof for the carcinogenic potential of canine 9. Mitsuishi T, Kawana S, Kato T et al. Human papilloma- PVs, the detection of novel members of this large virus infection in actinic keratosis and Bowen’s disease: family of viruses is important for further research on comparative study with expression of cell-cycle regula- this issue. tory proteins p21waf1/cip1, 53, pcna, ki-67, and bcl-2 in Importantly, the novel papillomaviruses were un- positive and negative lesions*1. Human Pathology 2003; covered in the two cases of SCC in situ. These two dogs 34: 886–92. 10. Sundberg JP, Junge RE, Lancester WD. Immunoperox- were younger than the average age of the available dogs idase localization of papillomaviruses in hyperplastic and with SCC, which also favours a viral aetiology for this neoplastic epithelial lesions in animals. American Journal type of lesions. Indeed, the presence of PVs in such of Veterinary Research 1984; 45: 1441–6. 4 lesions has been established previously. Additionally, 11. Schwegler K, Walter JH, Rudolph R. Epithelial neo- one of the lesions described by this author subse- plasms of the skin, the cutaneous mucosa and the transi- quently developed into invasive SCC and the similarities tional epithelium in dogs: an immunolocalization study between these cases and the human Epidermodysplasia for papillomavirus antigen. Journal of Veterinary verruciformis has already been emphasized.4 The pres- Medicine A 1997; 44: 115–23. ence of specific oncogenic PVs in a significant number 12. Teifke JP, Lohr CV, Shirasawa H. Detection of canine of such canine lesions consequently warrants further oral papillomavirus-DNA in canine oral squamous cell investigation. carcinomas and p53 overexpressing skin papillomas of the dog using the polymerase chain reaction and non- The study has also demonstrated the great diversity radioactive in situ hybridization. Veterinary Microbiology of the canine PVs. Cloning these new viruses will allow 1998; 60: 119–30. phylogenetic comparison with human commensal and 13. Watrach AMSe, Case MT. Canine papilloma: progres- high-risk PVs. These comparisons can eventually sion of oral papilloma to carcinoma. Journal of National provide clues for the interpretation of past and future Cancerology Institute 1970; 45: 915–20. studies. 14. Lowy DRH, Howley PM. Papillomaviruses. In: Knipe DM, Howley PM eds. Fields Virology, 4th edn. Philadelphia: Lippincott Williams & Wilkins, 2001: 2231–64. ACKNOWLEDGEMENTS 15. de Villiers E-M, Fauquet C, Broker TR et al. Classifica- tion of papillomaviruses. Virology 2004; 324: 17–27. The authors would like to acknowledge Dr Thelma Lee 16. Harwood CA, Proby CM. Human papillomaviruses and non-melanoma skin cancer. Current Opinion in Infectious Gross who has provided one of the samples (case #2). Diseases 2002; 15: 101–14. 17. zur Hausen H. Papillomavirus infections – a major cause of human cancers. Biochimica et Biophysica Acta 1996; REFERENCES 1288: F55–78. 18. zur Hausen H. Papillomaviruses and cancer: from basic 1. Morrison WB. Cancers of the head and neck. In: studies to clinical application. National Review of Morrison WB ed. Cancer in Dogs and Cats. Jackson: Cancer 2002; 2: 342–50. Teton New Media, 2002: 489–93. 19. Chambers VC, Evans CA. Canine oral papillomatosis. I. 2. Thomas RCF, Fox LE. Tumors of the skin and the Virus assay and observations on the various stages of the subcutis. In: Morisson WB ed. Cancer in Dogs and Cats, experimental infection. Cancer Research 1959; 19: 1188– 2nd edn. Jackson: Teton New Media, 2002: 473–88. 95.

20. Nicholls PK, Stanley MA. Canine papillomavirus – A human papillomavirus genotypes in nonmelanoma skin centenary review. Journal of Comparative Pathology cancers of nonimmunosuppressed individuals identifies 1999; 120: 219–33. high-risk genital types as possible risk factors. Cancer 21. Sundberg JP, O’Banion MK, Schmidt-Didier E et al. Research 2003; 63: 7515–19. Cloning and characterization of a canine oral papilloma- 30. Sundberg JP, Smith EK, Herron AJ et al. Involvement of virus. American Journal of Veterinary Research 1986; canine oral papillomavirus in generalized oral and cuta- 47: 1142–4. neous verrucosis in a Chinese Shar Pei dog. Veterinary 22. Le Net JL, Orth G, Sundberg JP et al. Multiple pig- Pathology 1994; 31: 183–7. mented cutaneous papules associated with a novel canine 31. Albini S, Zimmermann W, Neff F et al. Identification papillomavirus in an immunosuppressed dog. Veterinary and quantification of ovine gammaherpesvirus 2 DNA in Pathology 1997; 34: 8–14. fresh and stored tissues of pigs with symptoms of porcine 23. Campbell KL, Sundberg JP, Goldschmidt MH et al. malignant catarrhal fever. Journal of Clinical Microbiology Cutaneous inverted papillomas in dogs. Veterinary 2003; 41: 900–4. Pathology 1988; 25: 67–71. 32. Smith KT, Campo MS. ‘Hit and run’ transformation of 24. Antonsson A, Hansson BG. Healthy skin of many animal mouse C127 cells by bovine papillomavirus type 4: the species harbours papillomaviruses which are closely viral DNA is required for the initiation but not for main- related to their human counterparts. Journal of Virology tenance of the transformed phenotype. Virology 1988; 2002; 76: 12537–42. 164: 39–47. 25. Antonsson A, Erfurt C, Hazard K et al. Prevalence and 33. Ordonez RM, Espinosa AM, Sanchez-Gonzalez DJ type spectrum of human papillomaviruses in healthy skin et al. Enhanced oncogenicity of Asian-American human samples collected in three continents. Journal of General papillomavirus 16 is associated with impaired E2 repres- Virology 2003; 84: 1881–6. sion of E6/E7 oncogene transcription. Journal of 26. Pfister H. Human papillomavirus and skin cancer. Journal General Virology 2004; 85: 1433–44. of National Cancer Institute Monographs 2003: 52–6. 34. de Villiers EM, Ruhland A. Do specific human papillo- 27. Saveria Campo M. Animals models of papillomavirus mavirus types cause psoriasis? Archives of Dermatology pathogenesis. Virus Research 2002; 89: 249–61. 2001; 137: 384. 28. Smith KT, Campo MS. Papillomaviruses and their 35. Duensing S, Münger K. Mechanisms of genomic insta- involvement in oncogenesis. Biomedical Pharmaco- bility in human cancer: insights from studies with human therapy 1985; 39: 405–14. papillomavirus oncoproteins. International Journal of 29. Iftner A, Klug SJ, Garbe C et al. The prevalence of Cancerology 2004; 109: 157–62.

Résumé L’ADN de papillomavirus (PV) est fréquemment retrouvé dans des achantillons de carcinome épider- moïde (SCC) chez l’homme. Cependant le rôle de ces virus dans le développement de ces cancers est controversé. Alors qu’environ une centaine de PV sont recensés chez l’homme, un seul PV canin oral (COPV) a été identifié et étudié. Nous avons utilisé une technique de PCR spécifique des PV canin et félin, ainsi qu’une technique de PCR utilisée pour la détection des nouveaux PV humains, afin d’analyser 42 biopsies paraffinées, représentant trois formes différentes de SCC canins. 10 échantillons de tissus affectés par des maladies non néoplasiques ont servi de contrôle. Aucun des témoins négatifs n’a réagi positivement, et de l’ADN de PV a été retrouvé cajs 21% des prélèvements testés de SCC. Le COPV classique n’a été amplifié qu’une seule fois, alors que les autres cas positifs étaient associés à la présence d’une variété inconnue de PV humain. Cette étude suggère qu’une fraction des SCC canins est infectée par le PV, et qu’il existe une variété génétique des PV canins. Ces résultats vont faciliter les études futures qui s’intéresseront au rôle des PV dans le développement des cancers cutanés du chien.

Resumen El ADN del virus papiloma (PV) se descubre de forma frecuente en muestras de carcinoma de células escamosas (SCC) de la piel humana. Sin embargo, el papel de estos virus en el desarrollo de estas neoplasias en el perro es aún controvertido. Mientras que se conocen aproximadamente 100 virus papiloma en humanos, tan sólo un virus papiloma canino oral (COPV) ha sido identificado y estudiado de forma exhaustiva. Por ello, para analizar 42 muestras incluidas en parafina que representaban tres formas diferentes de carcinomas de células escamosas en perros, aplicamos una reacción de polimerasa en cadena (PCR) de estrecho rango, válida para detectar virus papiloma clásicos caninos y felinos; y también una PCR de amplio rango, que ha sido utilizada para la detec- ción de nuevos virus papiloma en humanos. Diez muestras de piel con lesiones no neoplásicas se utilizaron como controles. Mientras que ninguno de los controles negativos dio resultado positivo, ADN de virus papiloma se encontró en un 21% de las muestras de carcinoma de células escamosas. Curiosamente, el clásico virus papiloma oral canino solo se amplificó de una muestra, mientras que los otros casos positivos se asociaron con variedades hasta ahora desconocidas de virus papiloma. Este estudio sugiere que una fracción de carcinomas de células escamosas caninos está infectada con el virus papiloma, y que existe una diversidad genética de virus papiloma caninos. Por lo tanto, estos resultados facilitarán futuros estudios sobre el papel del virus papiloma en el desarrollo de cáncer de piel en perros.

Zusammenfassung Papillomavirus (PV) DNA wird häuﬁg in Hautproben von Plattenepithelkarzinomen des Menschen gefunden. Beim Hund bleibt die Rolle dieser Viren bei der Entstehung derartiger Tumoren allerdings

umstritten. Während etwa 100 humane Papillomaviren bekannt sind, wurde erst ein einziges canines orales Pap- illomavirus (COPV) identifiziert und umfangreich untersucht. Daher haben wir eine ‘narrow-range’ Polymerase Chain Reaction (PCR) angewendet, die passend ist für den Nachweis von klassischen caninen und felinen Pap- illomaviren, sowie eine ‘broad-range’ PCR, die verwendet worden war für den Nachweis von verschiedenen neuen Papillomaviren beim Menschen, um 42 in Paraffin eingebettete Proben zu analysieren, die drei unterschiedliche Formen von caninem Plattenepithelkarzinom repräsentierten. Zehn Hautproben von verschiedenen nicht- neoplastischen Zuständen dienten als Kontrollen. Während keine der Negativkontrollen positiv war, wurde PV DNA in 21% der untersuchten Proben der Plattenepithelkarzinome gefunden. Interessanterweise wurde das klassische COPV nur aus einer Probe isoliert, während die anderen positiven Fälle im Zusammenhang mit einer Variation von bisher unbekannten PVs gefunden wurden. Diese Studie weist darauf hin, dass ein Teil der caninen Plattenepithelkarzinome mit PV infiziert ist und dass eine genetische Variation der caninen Papillomaviren besteht. Daher werden diese Ergebnisse zukünftige Studien über die Rolle von Papillomaviren bei der Entstehung von caninen Hauttumoren erleichtern.

Chapter 7

Detection of novel papillomavirus-like DNA sequences in paraffine-embedded samples of feline invasive and in situ squamous cell carcinomas

G. Nespeca1, P. Grest2, W. S. Rosenkrantz3, M. 4 1 Ackermann , C. Favrot

American Journal of Veterinary Research, 2006; 67 (12): 2036-2041

1 Clinic for Small Animal Internal Medicine, Dermatology Unit, Vetsuisse Faculty, University of Zurich, Zurich, Switzerland 2 Institute for Veterinary Pathology, Vetsuisse Faculty, University of Zurich, Zurich, Switzerland 3 Animal Dermatology Clinic, San Diego, CA, USA 4 Virology Institute, Vetsuisse Faculty, University of Zurich, Zurich, Switzerland

Funded by grants from the Waltham Foundation and the European Society of

Veterinary Dermatology

52 Detection of novel papillomaviruslike sequences in paraffin-embedded specimens of invasive and in situ squamous cell carcinomas from cats

Gilles Nespeca, Med Vet; Paula Grest, DVM; Wayne S. Rosenkrantz, DVM; Mathias Ackermann, DVM, PhD; Claude Favrot, DrVet, MS

Objective—To detect and partially characterize papil- ABBREVIATIONS lomavirus (PV) DNA in squamous cell carcinoma AK Actinic keratosis (SCC) tumor specimens from cats. SCC Squamous cell carcinoma Sample Population—54 formalin-fixed paraffin- PV Papillomavirus embedded skin biopsy specimens were examined. HuPV Human PV Specimens originated from Bowenoid in situ SCC FePV Feline PV (BISC; n = 21), invasive SCC (22), and skin affected by BISC Bowenoid in situ SCC miscellaneous nonneoplastic conditions (11). CaPV Canine PV Procedures—Samples from each tissue block under- BoPV Bovine PV went DNA extraction after deparaffinization, and PCR assays were performed. Two sets of primers derived from PV E1 were used. The first set of primers was AK, a precancerous skin growth associated with sun designed for the narrow-range PCR assay and was 3 able to generate amplification products of feline PV exposure. However, AK is often regarded as a form of (FePV), canine oral PV, or closely related PVs. The sec- SCC, which is confined to the epidermis; thus, AK is ond set of primers was selected for the broad-range also referred to as in situ SCC.3-6 A second form of in PCR assay because of its ability to amplify DNA from situ SCC, precancerous dermatitis (termed Bowen’s dis- 64 human PVs. Sequence analysis of each amplified ease), presents as 1 or more flat red scaly patches up to DNA was performed. several centimeters wide, often found in large num- Results—1 of the 21 specimens of BISC was positive bers.1,4,6 In situ SCC can persist as such; regress; or for PV DNA on the basis of narrow-range PCR assay develop into a third, even more malignant form, inva- results, whereas all the other specimens (BISC, invasive SCC. Similar skin cancers are also observed in vet- sive SCC, and controls) had negative results for PV erinary medicine, specifically in cats.7 Squamous cell DNA. In contrast, 5 of 21 BISC specimens and 4 of 22 carcinoma has been linked to a variety of causative invasive SCC specimens were positive for PV DNA on the basis of broad-range PCR assay results. associations, which include exposure to UV or ionizing Sequence analysis revealed that only 1 specimen was radiation; arsenic ingestion; toxic exposure to tars and infected by a virus closely related to classic FePV. In oils; immunosuppression from drugs such as cortico- the 8 other specimens positive for PV DNA, DNA of steroids, azathioprine, and cyclosporine; and last but unknown PVs was uncovered. not least, to PV infection.1,8-10 Conclusions and Clinical Relevance—Bowenoid in Papillomaviruses are host-specific epitheliotropic situ SCC and invasive SCC of cats may be associated DNA viruses that infect skin and mucous membranes. with PVs of genetic diversity. (Am J Vet Res In general, PV infections are benign, result in a latent 2006;67:2036–2041) infection, or induce microlesions or benign neoplasias.11-14 However, a subset of HuPVs and other animal PVs is clearly implicated in the development of quamous cell carcinoma is, after basal cell carcino- cancer.10, 14-17 Human PVs that cause mucosal and skin Sma, the second most common cancer of the skin in carcinomas in humans are referred to as high-risk PVs humans.1,2 Squamous cell carcinoma involves cancerous or epidermodysplasia verruciformis–associated HuPV changes to the cells of the middle portion of the epider- types, respectively.14 mal skin layer. This cancer may begin in normal skin; Close to 100 HuPV types have been described on in skin at the site of a burn, injury, or scar; or at a site the basis of isolation of complete genomes.13 of chronic inflammation.1 Most often, it originates from Knowledge on the combination of biological properties and sequence similarities led to the definition of new Received June 29, 2006. criteria to define genera, species, types, subtypes, and Accepted July 31, 2006. 13 From the Clinic for Small Animal Internal Medicine, Dermatology variants within the Papillomaviridae family. Unit (Nespecca, Favrot), the Institute for Veterinary Pathology In contrast to the numerous HuPVs, only a single (Grest), and the Virology Institute (Ackermann), Vetsuisse FePV has been identified.18 However, the existence of a Faculty, University of Zürich, Winterthurerstrasse 260, CH 8057 few other FePVs has been suggested on the basis of Zurich, Switzerland; and the Animal Dermatology Clinic, 5610 findings from several clinical and immunohistochemi- Kearny Mesa Rd, Ste B, San Diego, CA 92111 (Rosenkrantz). 19-22 Supported by grants from the Waltham Foundation and the cal studies. European Society of Veterinary Dermatology. Until now, little has been known about the pres- Address correspondence to Dr. Favrot. ence of PVs in SCCs of cats. Investigators in 1 study23

2036 AJVR, Vol 67, No. 12, December 2006 53 failed to uncover PV antigen in SCCs of cats, whereas AACAGCTGYTTTTTARCYTTTTT-3′) for narrow-range PCR findings in another study24 revealed the presence of PV assay, which is able to generate amplification products of antigens in 44% of tumor specimens of BISC from cats. approximately 341 bp of FePV, CaPV, or closely related PVs. Furthermore, PV DNA has been uncovered in tumor The second set of primers (ie, CP4, CP5, and PPF1 specimens from fibropapillomas, another type of cuta- primers), also derived from E1, was selected for broad-range 25 PCR assay with the objective of amplifying as many PVs as neous proliferative disease, of cats. possible. With this set of primers, up to 64 HuPVs are iden- Similar to the human disease types, 3 varieties of tifiable.30 The expected size of the PCR product was approxi- SCC have been described for cats, which are AK, BISC, mately 450 bp. and invasive SCC.26 Actinic keratosis usually occurs as a solitary lesion on sun-exposed, lightly haired areas, PCR assay and agarose gel electrophoresis— such as ear tips, external nares, or eyelids. White cats Polymerase chain reaction conditions for PapF and PapR were performed. Volumes of 30 mL were used. Each reaction are predisposed for the development of such lesions. contained 1 µL of genomic DNA, 200µM of each deoxynu- On the other hand, BISC is characterized usually by cleoside triphosphate, 0.3µM of each of the sense and anti- multiple well-circumscribed, hyperpigmented lesions sense primers, and 2.5 units of a DNA polymerase.c After an that occur frequently on the face, neck, and limbs.27,28 initial denaturation step at 95oC for 4 minutes, PCR assay To our knowledge, comparative studies of these 2 early was performed for 30 cycles at 95oC for 1 minute, 50oC for 1 forms of cancer have not been performed in cats. minute, and 74oC for 1 minute, with a final elongation step The purpose of the study reported here was to at 74oC for 5 minutes. Deoxyribonucleic acid extracted from detect PV DNA in specimens representing the various canine warts served as positive control, whereas DNA- and types of SCC in cats and in specimens from feline skin RNA-free water was used as negative control. The PCR mix with CP4, CP5, and PPF1 primers was with various nontumor conditions. We wanted to test identical to the mix for narrow-range PCR assay, except that whether tumor specimens from cats with SCC were 0.45µM of the CP4 and CP5 primers and 0.3µM of the PPF1 more often infected by PVs than nontumor skin speci- primer were used. The PCR assay consisted of a denaturation mens. Two types of PCR assays, narrow and broad step at 95oC for 10 minutes, followed by 40 cycles at 95oC for range, were applied to extend the range of targeted PVs 1 minute, 47oC for 1 minute, and 74oC for 1 minute, with a as far as possible. final elongation step at 75oC for 5 minutes. An extract from 1 bovine fibropapilloma served as an additional positive control. Materials and Methods Polymerase chain reaction products were segregated by Tissue specimens—Tissues were obtained from the col- agarose gel electrophoresis, and bands were viewed under UV lections of the Prairie Diagnostics Services, Western College light after ethidium bromide staining. Bands on the gel were d of Veterinary Medicine, University of Saskatchewan, excised, and DNA was extracted with a gel extraction kit. Amplified DNA was sequenced by use of fluorescent Saskatoon, SK, Canada; the Institut de Pathologie et de géné- e tique, Loverval, Belgium; Rest associates, London; and, the sequencing and terminator chemistry. Pathology Institute, Vetsuisse Faculty, University of Berne, Berne, Switzerland. Fifty-four formalin-fixed paraffin- Sequence analysis—Samples were considered positive embedded skin biopsy specimens were examined. Specimens for PV DNA if they met the following requirements: they originated from BISC (n = 21), invasive SCC (22), and mis- had a band of the expected size after gel electrophoresis and cellaneous skin conditions other than skin cancers (eg, aller- the sequenced DNA had homology with E1 of previously gic dermatitis; 11) that were used as negative controls. sequenced PVs. Homologous DNA sequences were searched for by use of the National Center for Biotechnology Specimens from white cats or from locations typical for AK f such as ear tips and eyelids were excluded from this study. Information GenBank via a BLAST search. Sequence Thirty-micrometer-thick sections were cut from each tissue alignments and phylogenetic trees were made from the clustal algorithm obtained by use of a software block, with a new disposable microtome blade for each g block, before DNA extraction. Two canine warts, which had program. histopathologic characteristics of typical PV-induced inclusion bodies, and 1 bovine fibropapilloma served as positive Results controls. Macro- and microscopic analysis—A careful macroscopic selection and microscopic confirmation DNA extraction—The protocol of Albini et al29 was used of affected specimens was a major prerequisite prior for DNA extraction. Briefly, each section was deparaffinized to the virologic analysis. Specimens representing twice with 1.2 mL of xylene at room temperature (approx invasive SCC had been resected from sun-exposed or 20oC) for 10 minutes, washed with 100% ethanol, and then dried at 37oC for 30 minutes. Desiccated samples were sus- white areas of 21 domestic shorthaired cats and 1 pended in a tissue lysis buffer (50mM Tris-HCl [pH 8.5], Persian cat (12 males and 10 females). Twenty-two 1mM ethilenediaminetetraacetic acid, and 2.8% sodium specimens from ear tips (n = 11), nose (3), eyelids dodecylsulfate) and proteinase K (20 mg/mL) and incubated (3), digits (3), and lips (2) met criteria required for at 56oC on a rocking platform overnight. After lysis, samples invasive SCC. These criteria included the macroscop- were transferred to a columna and centrifuged to reduce vis- ic presence of scaly-to-crusty and erosive-to-plaque- cosity. The DNA was precipitated with absolute ethanol and like or ulcerative lesions (Figure 1). The growth extracted with a commercial DNA kit.b process was always endophytic. Histologically, cords Primers—Two sets of primers were used for the PCR or islets of infiltrative cells were detected in all speci- assay. Because the sequences encoding E1 are highly con- mens. Furthermore, anisocytosis; anisokaryosis; served, the E1 regions of FePV (GenBank accession No. large, hyperchromatic nuclei; prominent nucleoli; AF480454) and CaPV (GenBank accession No. NC001619) increased mitotic index; and abnormal mitoses were were aligned to design a set of consensus primers (ie, PapF, encountered in all specimens with variable intensity 5′-ATGGCGGGMTARAAAAGGTA-3′ and PapR, 5′- and in variable proportion. In addition, keratin pearls

AJVR, Vol 67, No. 12, December 2006 2037 54 and intercellular bridges were present in 16 and 12 BISC were available for virologic analysis by PCR assay specimens, respectively. and sequencing. A second group of 21 tumor specimens met the criteria for BISC. These specimens were obtained PCR assays—The narrow-range PCR assay ampli- from the face (n = 16), neck (12), and limbs (3) or fied PV DNA extracted from canine warts but not DNA were scattered (2). Thirteen domestic shorthair cats, extracted from bovine fibropapilloma (Figure 2). In 3 domestic longhair cats, 2 Siamese, 1 Persian, 1 contrast, the broad-range PCR assay amplified PV DNA Himalayan, and 1 Cornish Rex were affected. from canine warts and bovine fibropapilloma. It was Macroscopically, the lesions were squamous crustosus concluded that both PCR assays were able to specifi- and grossly circular. Two lesions had a single center, cally amplify selected PV DNAs. but 19 were multicentric (Figure 1). Microscopically, The narrow-range PCR assay was applied to sam- the following criteria were met for BISC: moderate-to- ples from invasive SCC and BISC specimens; 1 BISC severe parakeratotic hyperkeratosis, acanthosis with sample (BISC sample No. 15; Appendix; Figure 3) had papillomatous hyperplasia (n = 1) or irregular hyper- positive results for PV DNA, whereas the others had plasias (20), loss of polarity, and scattered dyskeratot- negative results. Next, the broad-range PCR assay was ic keratinocytes atypia in all layers of the epidermis applied to the same samples. Interestingly, 5 of 21 BISC and usually also in the infundibulum and reaching samples (BISC sample Nos. 2, 5, 6, 10, and 15) as well the isthmus. Furthermore, the following types of as 4 of 22 SCC samples (SCC sample Nos. 15, 24, 28, atypia were recorded: enlarged nuclei, anisokaryosis, and 29) had positive results for PV DNA (Figure 2). monster cells (bizarre multinucleated giant cells), One of the samples that had positive results for PV and abnormal mitotic figures. Hyperpigmentation DNA on the broad-range PCR assay (BISC sample No. was found in all but 3 specimens. Fifteen of the 21 15) also had positive results for PV DNA on the nar- specimens had clumped keratohyalin granules, which row-range PCR assay. These results suggested that the were considered as suggestive for PV infection. broad-range PCR assay was indeed able to uncover PVs However, other signs such as koilocytosis and nuclear that were different from the known FePV and CaPVs. inclusion bodies were not detected. With the exception of the ulcerated lesions, the dermis of all samples was considered normal and not heavily inflamed (Figure 1). Thus, a total of 22 samples representing invasive SCC and 21 samples representing

Figure 1—Macroscopic and microscopic lesions of SCCs in cats. A—Photograph of invasive SCC in a cat with ulceration of the eyelid. B—Photomicrograph of a section of the invasive SCC from panel A at low magnification. Notice invasive proliferation of atypical keratinocytes with pearl formation (white arrow) and the cornified layer (red arrow). The basal membrane is not dis- cernible. The dermis (long arrow) is invaded by cords of atypical Figure 2—Establishment of broad-range and narrow-range PCR keratinocytes (short black arrow, pointing towards such an inva- assays for detection of carnivore PVs. Polymerase chain reaction sive site). H&E stain; bar = 200 µm. C—Photomicrograph of a products were loaded on agarose gels and stained with ethidi- section of the invasive SCC from panel A at high magnification. um bromide. A—Amplification of cloned DNA by either broad- Notice islets of keratinocytes with features of malignancy, such range (450 bp) or narrow-range PCR assay (341 bp). Lane 1 = as anisokaryosis, anisocytosis, multinucleated cells (short Water in place of DNA added to the reaction. Lane 2 = DNA arrow), and abnormal mitosis (long arrow). H&E stain; bar = 50 from a commercially available phagemid.h Lane 3 = DNA from µm. D—Photograph of BISC in a cat with circular, crusted, ero- oral CaPV cloned into the phagemid.h Lane 4 = DNA from CPV3 sive, and hyperpigmented plaques (arrow). E—Photomicrograph (GenBank accession No. DQ295066) cloned into the phagemid.h of a section of the BISC from panel D at low magnification. The M1 = Molecular weight marker (100-bp ladder). B—The DNA basal membrane (white arrow) is intact, and the dermis is not was extracted from tissues before being amplified by either the invaded. Irregular acanthosis (long black arrow) is obvious. narrow range or the broad-range PCR assays. M2 = 1-kilobase Notice that hair follicles are affected (short arrow). H&E stain; ladder. M1 = 100-bp ladder. Lane 1 = Negative control with no bar = 200 µm. F—Photomicrograph of a section of the BISC DNA added to the reaction. Lane 2 = Extract from canine wart from panel D at high magnification. Notice acanthosis, hyper- tissue, which had typical PV-induced inclusion bodies on histo- pigmentation (brown cells), clumped keratohyalin granules logic examination. Lane 3 = Extract from a tumor specimen of a (white arrow), loss of polarity, and anisokaryosis (branched cat with SCC (GenBank accession No. DQ085784). Lane 4 = arrow). H&E stain; bar = 50 µm. Extract from SSC sample No. 39.

2038 AJVR, Vol 67, No. 12, December 2006 55 Discussion The purpose of our study was to detect and partially characterize PV DNA in samples representing in situ and invasive types of SCC in cats to learn more about PV variants in cats and about possible associations of these viruses with individual forms of SCC in cats. Two types of PCR-assays, a narrow range and a broad-range PCR, were applied to extend the range of targeted PVs as far as possible. Careful macroscopic selection and microscopic confirmation resulted in the identification of 22 samples representing invasive SCC and 21 samples representing BISC, which were available for virologic analysis by PCR assay. Papillomavirus DNA was detected in 4 of 22 samples representing invasive SCC and in 5 of 21 samples representing BISC, whereas all nontumor control samples had negative results for PV DNA. Only 1 (BISC sample No. 15) of the 9 viral DNAs had been Figure 3—Phylogenetic relationships of the newly detected PV sequences with known PV E1 sequences (ie, CaPV, FePV, HuPV, revealed by the narrow-range PCR assay. However, the and BoPV are represented by GenBank accession Nos. D55633, same narrow-range PCR assay amplified PV DNA AF377865, AY330623, and AJ620206, respectively). IS = BISC. extracted from canine warts, which was expected SCC= Invasive SCC. Units indicate the number of substitution events (percentage of nucleotides), because the primers had been chosen for their homology with conserved sequences within E1 of FePV and Sequence analysis—The nucleotide sequence of CaPV. Yet, the restricted range of these primers was the amplified DNA was determined and the resulting confirmed, as they proved unable to amplify DNA from sequences were compared to assess whether novel PV- the more distantly related bovine fibropapilloma virus. like sequences had been detected. Indeed, use of the These results indicate that the remaining samples with basic local alignment search toole revealed relatedness to positive results for PV DNA did not harbor conven- PV E1 sequences for all 9 samples that were positive for tional FePV or CaPV. PV DNA. Relation to HuPV, FePV, CaPV, rat PV, and Eight samples, which had negative results for PV BoPV was evident. Clustal alignmentsg revealed the var- DNA on narrow-range PCR assay, had positive results ious degrees of relationship of the newly determined for PV DNA on broad-range PCR assay. The broad- sequences among each other as well as in comparison range PCR assay made use of a second set of primers with known PVs. Overall, CaPVs and FePVs were most that were also derived from E1 but known to uncover closely related to the newly detected viral sequences, a large variety of HuPVs.30 In our study, this second set with a relative amino acid identity of 56% to 71%. of primers also amplified DNA from bovine fibropapil- Among the HuPVs, types 4, 55, 63, 65, 71, and 74 were loma virus as well as viral DNA from canine warts. aligned most frequently but type 71 most often had the Sequencing of the amplification products obtained closest relationship to the new sequences with 58% to from the 8 samples revealed novel PV-related DNAs, 61% amino acid identity. Among the BoPVs, type 5 was although relations to HuPV, FePV, CaPV, rat PV, and the closest relative, having 55% to 62% amino acid iden- BoPV were evident. A phylogenic tree drawn from the tity. A phylogenic tree drawn from the aligned sequences aligned sequences divided the new sequences into 4 divided the new sequences into 4 clusters (Figure 3). In clusters. Three of those clusters had close relationship cluster 1, BISC sample No. 15 was situated most closely to CaPV, FePV, and HuPV. Interestingly, the fourth clus- with CaPV and FePV. Cluster 2 was occupied by BISC ter, represented by 6 amplification products, was clear- sample No. 10 and was between FePV and HuPV type ly distinct from BoPV type 5 and from HuPV type 71, 71. Cluster 3 was represented by BISC sample No. 2 and FePV, and CaPV. Judging from the limited sequence found close to HuPV type 71. The remaining sequences information available, it appeared as if this fourth clus- (SCC sample Nos. 15, 24, 28, and 29 and BISC sample ter represented a novel group of FePVs that had not Nos. 5 and 6) represented a fourth cluster, which was been detected previously and that may be associated clearly distinct from BoPV type 5 on the most distant with SCC in cats. Notably, all PVs detected in associa- side and HuPV type 71, FePV, and CaPV on the less dis- tion with invasive SCC were found to belong to this tant side. These results suggested the presence of thus novel cluster. far unidentified PVs in tissues representing invasive SCC This represents, to our knowledge, the first evi- and BISC. Interestingly, sequences obtained from speci- dence of thus far unknown PV-like sequences associat- mens of 3 cats with invasive SCC (SCC sample Nos. 15, ed with SCC in cats. Interestingly, some of the novel 24, and 28) had identical sequences. Furthermore, it sequences were found in association with invasive was observed that not a single sample from invasive SCC SCC. Notably, previous attempts to detect convention- specimens had been associated with the more classic al PV antigens in such lesions had failed,23 which led to FePV and CaPV. However, because the classification of the hypothesis that invasive carcinomas of cats are PVs is based on the sequence of L1, the exact taxonom- probably not virally induced, whereas instances of ic position of these novel PV-like sequences could not be Bowen’s disease in cats are probably PV-induced.7 Our assigned. findings clearly challenge the former opinion, although

AJVR, Vol 67, No. 12, December 2006 2039 56 a causative correlation between the disease and the 7. Scott DW, Miller WH, Griffin CE. Neoplastic and non-neo- novel PV strains has not yet been shown. Results of plastic tumors. In: Scott DW, Miller WH, Griffin CE, eds. Muller and another study24 did reveal PV antigens in 44% of BISCs. Kirk’s small animal dermatology. 6th ed. Philadelphia: WB Saunders Co, 2001;1236–1413. Although the proportion of BISC samples with positive 8. Paul CF, Ho VC, McGeown CE, et al. Risk of malignancies results for PV DNA in our study is lower (5 of 21 BISC in psoriasis patients treated with cyclosporine: a 5 y cohort study. samples), it should be kept in mind that the broad- J Invest Dermatol 2003;120:211–216. range PCR assay may not be able to reveal all variants 9. zur Hausen H. Papillomavirus infections—a major cause of of FePVs. Furthermore, it is well-known that PCR human cancers. Biochim Biophys Acta 1996;1288:F55–F78. detection of viral nucleic acids in formalin-fixed and 10. zur Hausen H. Papillomaviruses and cancer: from basic studies to clinical application. 2002;2:342–350. paraffin-embedded tissues may be decreased in com- Nat Rev Cancer 29 11. Antonsson A, Erfurt C, Hazard K, et al. Prevalence and type parison to fresh tissue. Finally, the absence of PV spectrum of human papillomaviruses in healthy skin samples col- DNA in SCC samples can also be explained by the so- lected in three continents. J Gen Virol 2003;84:1881–1886. called hit-and-run model, which postulates an initial 12. Antonsson A, Hansson BG. Healthy skin of many animal transformation of the infected cell and a subsequent species harbors papillomaviruses which are closely related to their loss of PV DNA.31 human counterparts. J Virol 2002;76:12537–12542. 13. de Villiers EM, Fauquet C, Broker TR, et al. Classification of Full proof of the existence of the novel PVs that papillomaviruses. Virology 2004;324:17–27. are predicted through the results of our study still 14. Lowy DR, Howley PM. Papillomaviruses. In: Knipe DMH, needs to be provided. However, we suggest that a Howley PM, Griffin DE, et al, eds. Fields’ virology. Vol 2. 4th ed. great diversity of FePVs may exist that is in need of Philadelphia: Lippincott Williams & Wilkins, 2001;2231–2264. detection and characterization. The future use of 15. Pfister H. Chapter 8: human papillomavirus and skin cancer. the technique applied here will help in identifying J Natl Cancer Inst Monogr 2003;(31):52–56. more affected cats with papilloma-associated dis- 16. Saveria-Campo M. Animal models of papillomavirus pathogenesis. Virus Res 2002;89:249–261. eases. Virologic studies can be initiated with the 17. Smith KT, Campo MS. Papillomaviruses and their involve- aim to better characterize these novel viruses. ment in oncogenesis. Biomed Pharmacother 1985;39:405–414. Cloning and sequencing of the entire genomes of 18. Tachezy R, Duson G, Rector A, et al. Cloning and genomic these viruses will allow phylogenetic comparisons characterization of Felis domesticus papillomavirus type 1. Virology with HuPVs as well as discrimination between 2002;301:313–321. benign and high-risk variants. Such studies can 19. Carpenter JL, Kreider JW, Alroy J, et al. Cutaneous xan- thogranuloma on an eyelid of a cat. Vet Dermatol 1992;3:187–190. eventually provide insights into the molecular path- 20. Egberink HF, Berrocal A, Bax HA, et al. Papillomavirus asso- ways underlying the pathogenesis of these viruses ciated skin lesions in a cat seropositive for feline immunodeficiency in cats. virus. Vet Microbiol 1992;31:117–125. 21. Lozano-Alarcon F, Lewis TP II, Clark EG, et al. Persistent a. QIAshredderTM column b, Qiagen, Basel, Switzerland. papillomavirus infection in a cat. J Am Anim Hosp Assoc b. QIAamp DNA Mini Kit, Qiagen, Basel, Switzerland. 1996;32:392–396. c. Pfu Turbo DNA polymerase, Stratagene, La Jolla, Calif. 22. Sundberg JP, VanRanst M, Montali R, et al. Feline papillomas d. QIAquick gel extraction kit, Qiagen, Basel, Switzerland. and papillomaviruses. Vet Pathol 2000;37:1–10. e. AB-3100-based fluorescent sequencing and BigDye terminator 23. Sundberg JP, Junge RE, Lancester WD. chemistry, Synergene Biotech GmbH, Biotech Center Zurich, Immunoperoxidase localization of papillomaviruses in hyperplas- Schlieren, Switzerland. tic and neoplastic epithelial lesions in animals. Am J Vet Res f. NCBI BLAST, National Center for Biotechnology Information, 1984;45:1441–1446. Bethesda, Md. Available at: www.ncbi.nlm.nih.gov/BLAST/. 24. LeClerc SM, Clark EG, Haines DM. Papillomavirus infection Accessed Oct 10, 2005. in association with feline cutaneous squamous cell carcinoma in situ, g. Lasergene Biocomputing Software for the Macintosh, version in Proceedings. Am Assoc Vet Derm Am Coll Vet Derm x.x, DNAStar Inc, Madison, Wis. 1997;13:125–126. h. pBluescript, Stratagene, La Jolla, Calif. 25. Schulman FY, Krafft AE, Janczewski T. Feline cutaneous fibropapillomas: clinicopathologic findings and association with References papillomavirus infection. Vet Pathol 2001;38:291–296. 1. Grossmann D, Leffell DJ. Squamous cell carcinoma. In: 26. Gross TL, Ihrke PJ, Walder EJ, et al. Epidermal tumors. In: Freedberg IM, Eisen AZ, Wolff K, et al, eds. Fitzpatrick’s dermatology Gross TL, Ihrke PJ, Walder EJ, et al, eds. Skin diseases of the dog and in general medicine. Vol 1. 6th ed. New York: McGraw-Hill Book Co, cat: clinical and histopathological diagnosis. Oxford, England: 2003;xxx–xxx. Blackwell Publishing Ltd, 2005;562–577. 2. Kane CL, Kheen CA, Smithberger E, et al. Histopathology of 27. Baer KE, Helton K. Multicentric squamous cell carcinomas cutaneous squamous cell carcinoma and its variants. Semin Cutan in situ resembling Bowen’s disease in cats. Vet Pathol Med Surg 2004;23:54–61. 1993;30:535–543. 3. Anwar J, Wrone DA, Kimyai-Asadi A, et al. The develop- 28. Miller WH, Affolter VK, Scott DW, et al. In situ resembling ment of actinic keratosis into invasive squamous cell carcinoma: evi- Bowen’s disease in five cats. Vet Dermatol 1992;3:177–182. dence and evolving classification schemes. Clin Dermatol 29. Albini S, Zimmermann W, Neff F, et al. Identification and 2004;22:189–196. quantification of ovine gammaherpesvirus 2 DNA in fresh and stored 4. Arlette JP, Trotter MJ. Squamous cell carcinoma in situ of the tissues of pigs with symptoms of porcine malignant catarrhal fever. skin: history, presentation, biology and treatment. Australas J J Clin Microbiol 2003;41:900–904. Dermatol 2004;45:1–11. 30. Iftner A, Klug SJ, Garbe C, et al. The prevalence of human 5. Cockerell CJ. Pathology and pathobiology of the actinic papillomavirus genotypes in nonmelanoma skin cancers of nonim- (solar) keratosis. Br J Dermatol 2003;149:34–36. munosuppressed individuals identifies high-risk genital types as pos- 6. Mitsuishi T, Kawana S, Kato T, et al. Human papillomavirus sible risk factors. Cancer Res 2003;63:7515–7519. infection in actinic keratosis and bowen’s disease: comparative 31. Smith KT, Campo MS. “Hit and run” transformation of study with expression of cell-cycle regulatory proteins mouse C127 cells by bovine papillomavirus type 4: the viral DNA is p21(Waf1/Cip1), p53, PCNA, Ki-67, and Bcl-2 in positive and neg- required for the initiation but not for maintenance of the trans- ative lesions. Hum Pathol 2003;34:886–892. formed phenotype. Virology 1988;164:39–47.

2040 AJVR, Vol 67, No. 12, December 2006 57 Appendix New papillomaviruslike sequences.

GenBank Sample type accession No. and No. Sequences DQ085782 BISC 15 1 ATGGTACAAT GGGCATTTGA CAATAAGTAC ACAGATGAAG CAGAGATAGC TTTTCATTAT 61 GCACGTTTGG CAGAGGAGGA TGCAAATGCA GAGGCTTGGT TAAAAAGCAA CTCCCAAGCT 121 AAATATGTCC GAGATTGTGC GCAAATGGTG AAGCTGTATC TTAGACAAGA AATGAGGCAG 181 ACTACTATTT CTGAATGGAT TGACAAGTGC TGCCAGTCAG TGACAGAGGA CGGTGACTGG 241 GGGGATATTA TGCGCTTCTT AAAATATCAG CAAGTTAATT TCACTCAGTT TTTAACTGCC 301 ATGAGAAATG CTTTAGAGGG TAAACCTAAA AAAAACTGCT TAGTATTTTA TGGGCCTCCA 361 GATACTGGCA AGTCATATTT CTGCTTTAGT TTGGTTAGTT TTATGCAGGG GAAAGTGGTG 421 AATTTTATGA ATAGCAA DQ085783 BISC 2 1 ANGAGGAACG ATATAGCCTA CCACTATGCA TTGCTAGCCG ACGAGGACAC AAATGCAGCG 61 GCATGGCTAG GTACAAACTC ACAGGCCAAG CATGTCAGGG ACTGCGCAGT GATGGTCAAG 121 CATTACAGGC GTGCCATAAT GTCTGCCATG AGTATGTCCG AATGGATAAA CAGACGAATG 181 GGCCTGATAG AGGAGGAAGG AGACTGGAAA AACATAGGCA ATTTCCTCAG ATACCAGGGT 241 ATAGAGGTTA TTACATTTAT AGGGGCGCTG AGGGACATGT TAAAGGGCAT TCCAAAAAGG 301 ACATGTATGT GTATAGTGGG ACCACCAGAC ACAGGGAAAT CAGCGTTTTG CCTTAGCCTG 361 CTAGACTTCT TCGGGGGTAG GGTACTGTCA TTCACCAATT ACAAAAGCCA TTTTTGNTGN 421 CCNACCCTCA A DQ085784 SCC 15, 24, and 28 1 TTATGGTACA NGTGGGCATT TGACAATAAG TACACAGATG AAGCAGAGAT AGCTTTTCAT 61 TATGCACGTT TGGCAGAGGA GGATGCAAAT GCAGAGGCTT GGTTAAAAAG CAACTCCCAA 121 GCTAAATATG TCCGAGATTG TGCGCAAATG GTGAAGCTGT ATCTTAGACA AGAAATGAGG 181 CAGACTACTA TTTCTGAATG GATTGACAAG TGCTGCCAGT CAGTGACAGA GGACGGTGAC 241 TGGGGGGATA TCATGCGCTT CTTAAAATAT CAGCAAGTTA ATTTCACTCA GTTTTTAACT 301 GCCATGAGAA ATGCTTTAGA GGGTAAACCT AAAAAAAACT GCTTAGTATT TTATGGGCCT 361 CCAGATACTG GCAAGTCATA TTTCTGCTTT AGTTTGGTTA GTTTATGCAT GGAAAGTGGA 421 TTTNA DQ085785 SCC 29 1 TTATGCACGT TTGGCAGAGG AGGATGCAAA TGCAGAGGCT TGGTTAAAAA GCAACTCCCA 61 AGCTAAATAT GTCCGAGATT GTGCGCAAAT GGTGAAGCTG TATCTTAGAC AAGAAATGAG 121 GCAGACTACT ATTTCTGAAT GGATTGACAA GTGCTGCCAG TCAGTGACAG AGGACGGTGA 181 CTGGGGGGAT ATTATGCGCT TCTTAAAATA TCAGCAAGTT AATTTCACTC AGTTTTTAAC 241 TGCCATGAGA AATGCTTTAG AGGGTAAACC TAAAAAAAAC TGCTTAGTAT TTTATGGGCC 301 TCCAGATACT GGCAAGTCAT ATTTCTGCTT TAGTTTGGTT AGTTTATGCT TGAAAGTGGA DQ085786 BISC 6 1 TTTTATGGTA CAGTGGGCAT TTGACAATGA ATACTTTGAG GAAAGTGAGA TAGCATATCA 61 GTATGCATGC CTTGCAGAAA CAGAAGAAAA TGCTGCAGCC TTCCTAAATT CTAACAGCCA 121 AGCTAAGCAT GTCAGGGACT GTGCAACTAT GTGCAGATAT TATAAGAGAG CAGAAATGCA 181 GAGAATGTCA ATGTCCGCCT GGATTCACAA GAGATGTAAG GAGACCAGCC TGCAGGGAGA 241 TTGGAAAGAA ATAGTCAAGT TTCTTAGACA TCAAAGTGTA GAGTTTATTA CCTTTCTCTG 301 CAGCTTCAAG AAATTTCTCA GGGGTGTGCC TAAAAAAAAT TGCATGCTTT TCTGGGGTCC 361 TCCTAACACA GGCAAATCTA TGTTTTGCAT GAGCTTACTT TCTTTCCTAA AGGCANAGAT 421 TCTTTANC DQ085788 BISC 5 1 TTCTTATGGT ACAGTGGGCA TTTGACAATA AGTACACAGA TGAAGCAGAG ATAGCTTTTC 61 ATTATGCACG TTTGGCAGAG GAGGATGCAA ATGCAGAGGC TTGGTTAAAA AGCAACTCCC 121 AAGCTAAATA TGTCCGAGAT TGTGCGCAAA TGGTGAAGCT GTATCTTAGA CAAGAAATGA 181 GGCAGACTAC TATTTCTGAA TGGATTGACA AGTGCTGCCA GTCAGTGACA GAGGACGGTG 241 ACTGGGGGGA TATTATGCGC TTCTTAAAAT ATCAGCAAGT TAATTTCACT CAGTTTTTAA 301 CTGCCATGAG AAATGCTTTA GAGGGTAAAC CTAAAAAAAA CTGCTTAGTA TTTTATGGGC 361 CTCCAGATAC TGGCAAGTCA TATTTCTGCT TTAGTTTGGT TAGTTTATGC AGGGAAAGTG 421 TATTTAAAA DQ085789 BISC 10 1 TTTTTNTGGT NNCCAGTGGC NTACGATAAC GACTTCCGTG ACGAGTGCCA AATTGCCTAC 61 GAATATGCAC GGCTTGCCAC GGAGGACAGC AATGCATTGG CATGGTTGGA ATGCAATAAT 121 CAGGCCAAAT TTGTCAAAGA CTGTGCACGT ATGGTCGGGT ACTATAAGCG CGCTGAAATG 181 CAAAATATGT CTATCTCTGC TTGGATACNT AAGCAAATTA AAGATAGGCA GTGCACTACC 241 GATTGGAAAG TAATTNTGAA TTTTCNTAAG TTTCANCATG TGGAGGTTAT AATTTTTTTA 301 AATGCAATGA TGCATTTGCT CCGTGGCACG CCAAAGAAAA ATTGTCTGGT TCTGTACGGT 361 CCCCCAAATA CAGGGAAATC CATGTTCGCA ATGAGCTTAA TTCAGTGTCT GAAAGGACGT 421 GTATTGTNGT ATGTGAATTC ACGTAGTCAG TTNTGGTTGC ANCCCTTGGC AGATGCAAAA 481 ATAGCACTGC TGGACGATGC AACCAGACCA TGCTGGGAAC TATATAGATA TTTATTGAGA 541 AATGCATTGG ATGGTAATCC TATATGCCTG ACTAANCNAG C

AJVR, Vol 67, No. 12, December 2006 2041 58

Chapter 8

Clinical, histological and immunohistochemical study of feline viral plaques and bowenoid in situ carcinomas

1 2 3 1 S. Wilhelm , F. Degorce-Rubiales , D. Godson , C. Favrot

Veterinary Dermatology, 2006; 17: 424-431

1 Clinic for Small Animal Internal Medicine, Dermatology Unit, Vetsuisse Faculty, University of Zurich, Zurich, Switzerland

2 LAPVSO, Toulouse, France

1 Prairie Diagnostic Service, Saskatoon, Saskatchewan, Canada

59 Clinical,Blackwell Publishing Ltd histological and immunohistochemical study of feline viral plaques and bowenoid in situ carcinomas

Sylvia Wilhelm*, Frederique Degorce-Rubiales†, cats in the FVP + BISC group. On the other hand, only Dale Godson‡ and Claude Favrot* one of the nine BISC cats was positive. The presence of both FVP and BISC lesions in some cats and the high *Clinic for Small Animal Internal Medicine, Dermatology Unit, detection rate of PV antigens in the FVP and FVP + Vetsuisse-Faculty University of Zurich, Zurich, Switzerland BISC groups suggest that both conditions might have †LAPVSO, Toulouse, France the same viral cause and that some BISC may evolve ‡Department of Veterinary Microbiology, Western College of from FVP. The low rate of viral antigen detection in the Veterinary Medicine, University of Saskatchewan, Saskatoon, BISC group indicates another cause or a loss of viral Saskatchewan, Canada Correspondence: C. Favrot, Clinic for Small Animal Internal Medicine, replication during the cancerogenesis. Dermatology Unit, Vetsuisse-Facility, University of Zurich, Accepted 31 August 2006 Winterthurerstrasse 260, CH-8057 Zurich, Switzerland. Tel. +41 44 635 81 12; Fax: +41 44 635 89 20; Email: [email protected]; Email: swilhelm@vetclinics,unizh.ch

Introduction What is known about the topic of this paper Papillomaviruses (PV) are highly diverse viruses that usually • Reports of papillomavirus-induced dermatitis in cats are induce benign skin or mucous membrane proliferation rare. • Lesions of feline viral plaques have been described as in mammals and birds but can also cause squamous cell 1 feline hyperpigmented plaques and are clinically indistin- carcinomas. In humans, the PVs that induce benign hyper- guishable from lesions of bowenoid in situ carcinomas. plasia and those that induce cancers are phylogenetically • Feline bowenoid in situ carcinoma could be, like feline different.1 Benign hyperplasias (warts) usually regress viral plaques, papillomavirus-induced. after a few months, a regression associated with the development of cell-mediated immunity.2 What this paper adds to the field of veterinary In contrast with dogs, where PV infections are frequently dermatology • Clinically, feline viral plaques and feline bowenoid in situ observed, reports of PV-induced dermatoses are rare in 3–7 carcinomas are indistinguishable. cats. Lesions are usually flat and hyperpigmented, rather • Feline viral plaques and feline bowenoid in situ carcinomas than exophytic and flesh colour warts, and spontaneous might have the same viral cause. regression is rare.3–7 These lesions are usually, but not • Feline viral plaques could be a precursory lesion of feline always, multiple and have been described as feline viral bowenoid in situ carcinoma. plaques (FVP).8 Feline multicentric in situ squamous cell carcinomas also usually occur as multiple hyperpigmented plaques that resemble those of human Bowen’s disease.9,10 Gross Abstract and coworkers, however, recently remarked that there are Feline viral plaques (FVP) induced by papillomavirus major differences between the human and the feline (PV) are often hyperpigmented and flat warts. The fact diseases, and have coined the term ‘bowenoid in situ that up to 47% of bowenoid in situ carcinomas (BISC), carcinoma’ (BISC) to describe the feline condition.8 As which also usually occur in the form of hyperpigmented FVP clinically resembles BISC, it was suggested that both plaques, are positive for PV antigen in immunochemistry conditions may have the same cause, and one report suggests that BISC could evolve from FVP. mentions the association of both FVP and BISC on the The relationship between the presence of PV anti- same cat.11,12 Furthermore, it has been shown immuno- gens and the clinical and histological features of 26 histologically that up to 47% of feline BISC samples are cases of feline dermatoses (clinically described as positive for PV antigen, suggesting that BISC is virally pigmented plaques and with histological diagnosis of induced and that FVP could be, at least in some instances, FVP and/or BISC) was therefore determined. The precursory lesions of feline BISC.11 cases were classified into one of the three following Using records of the clinical, histological and immuno- groups: FVP, FVP + BISC or BISC. Immunohistological histological features of 26 cases of feline dermatoses detection of papillomavirus group-specific antigen clinically described as pigmented plaques and with an initial was performed using a polyclonal rabbit antibovine histological diagnosis of FVP and/or BISC, the hypotheses papillomavirus antiserum. that both lesions are often associated in the same samples, Of the seven cases in the FVP group, six were and that PV antigens are present in the majority of these deemed positive by immunohistology as were all 10 lesions, were tested.

60 FVP and BISC in cats with hyperpigmented plaques

for 20 min at 42 °C and treated with a 1 : 2000 dilution of rabbit Materials and methods antibovine papillomavirus type-1 antibody (Dako Diagnostics Canada Animals Inc., Missisauga, ON, Canada). A goat-biotinylated antirabbit IgG (Vector Laboratories Inc., Burlington, ON, Canada) was used at a History and clinical information was obtained from 26 cats with hyper- 1 : 400 dilution as the secondary antibody. Replicate sections were pigmented plaques. Cats were included, provided that a histological stained as above without protease digestion, and additional sections diagnosis of FVP and/or BISC had been made previously, and clinical were stained with a normal rabbit antiserum as the primary antibody to data (including concurrent diseases, immunosuppressive therapy and provide negative control. A positive control tissue, canine cutaneous evolution of the lesions, when available) were subsequently analysed papilloma, was included in each assay run. for each of the three histological groups: FVP, FVP + BISC and BISC. Both diaminobenzidine (DAB) (Electron Microscopy Sciences, Fort Washington, PA, USA) and Nova Red (Vector Laboratories Inc., Statistical analysis Burlington, ON, Canada) were used as chromogens on two different Data were analysed using nonparametric statistical methods sections for each sample. (GraphPad PRISM® for Windows, version 4.0; GraphPad Software, Inc., San Diego, CA, USA). Kruskal–Wallis one-way ANOVA by ranks and the Dunn’s post-test for multiple comparisons were used to compare ages among the three histological groups. Results Clinical information Histological evaluation The clinical data are summarized in Table 2. Differences Archival specimens of all 26 cats were compiled. These samples have between ages of cats in FVP, FVP + BISC and BISC groups been previously collected by biopsy from all 26 cats, fixed in formalin, (median 11.5, 12 and 13, respectively) were not statistically and processed routinely to paraffin wax for histological assessment. Sections (5 µm) were cut, routinely processed and stained with hae- significant. The sizes of the groups did not allow a proper matoxylin and eosin. The following criteria were systematically evaluation of potential breed or sex predispositions. assessed: severity and nature of the acanthosis, hypergranulosis and On clinical examination, FVP and BISC lesions were size of the keratohyalin granules, premature keratinization, involve- often indistinguishable and usually presented as solitary or ment of the hair follicle in the pathological process, disorderly or multiple grey, tan to black papules or small flat plaques abnormal maturation of the epidermis, atypia (pleomorphic or abnor- (Figs 1 and 2). Some, more frequently the BISC, appeared mally large nuclei, multinucleate cells), mitoses more than three cell ulcerated (Fig. 2). Solitary lesions were observed in only layers above the basal cell layer, koilocytosis, clear cells and presence of intracytoplasmic pseudo-inclusions and intranuclear inclusions. three of the 26 cats. The face, neck and limbs were mostly Koilocytes were defined as keratinocytes with swollen cytoplasms affected by BISC. FVP occurred mostly on the trunk, even and shrunken nuclei.8 Clear cells were defined as keratinocytes if other areas, including face and neck, were also affected. with swollen cytoplasm but rather enlarged, vesicular nuclei. These Cats with both conditions usually presented lesions on modified keratinocytes (clear cells and koilocytes) have been reported more than one body area and all body regions could be 13 to be also regularly associated with human PV infection. When affected. Very little follow-up information was available but observed, the margins of the lesions were checked for changes cases of transformation of FVP into BISC after the initial suggestive of viral infection such as koilocytes and clear cells, pseudo- inclusions, and clumped keratohyalin granules. histological diagnosis were not recorded. None of the Samples were subsequently classified into one of three groups: affected cats had a known history of immunosuppressive FVP, FVP + BISC (when both lesions were present on the same cat or on drug administration or concurrent disease. the same section) or BISC in accordance with standard criteria (Table 1) 8 for the diagnosis of FVP and BISC. When changes overlapping typical Histological examination FVP and BISC lesions were observed, lesions were designated as The results are summarized in Table 3. FVP, provided that the acanthosis remained moderate and atypia was absent. Lesions were classified as BISC if the acanthosis was marked and loss of polarity as well as atypia was evident. FVP The diagnosis of FVP was made in seven cases (Table 3). Immunohistochemical analysis Lesions consisted of well-demarcated epidermal hyper- Papillomavirus antigen was detected (at the Immunology Laboratory plasia with acanthosis, hyperpigmentation, hypergranulosis of Prairie Diagnostic Services, Saskatoon, Saskatchewan, Canada) with clumped keratohyalin granules and numerous koilocytes using an avidin–biotin complex technique adapted for an automated (Fig. 3). Some of these keratinocytes contained blue- slide stainer (Codon Histomatic Stainer, Fisher Scientific, Edmonton, grey fibrillar pseudo-inclusions (one of seven). Larger and AB, Canada) as previously described.14 This method has already been compact amphophilic intracytoplasmic pseudo-inclusions validated for the detection of feline PV antigens.7 Briefly, sections from each tissue block were mounted on slides (Codon Slides, Fisher were present in four cases (Fig. 3). In one case, both Scientific, Edmonton, AB, Canada) coated with 0.1% poly-D-lysine, pseudo-inclusion types were present in the same sample digested with protease XIV (Sigma Chemical Co., St. Louis, MO, USA) and compact ones (present in the stratum granulosum)

Table 1. Histological features of feline viral Feline viral plaque Bowenoid in situ carcinoma plaque and bowenoid in situ carcinoma Acanthosis Mild to moderate Moderate to severe Follicular involvement Sometimes Yes Differentiation Normal Dysplastic epidermis, loss of polarity Clumped keratohyalin granules Yes Yes Koilocytes Yes Yes Intracytoplasmic pseudo-inclusions Yes Yes Atypia No Yes Mitotic activity No Moderate

61 Wilhelm et al.

Table 2. Clinical ﬁndings

Case Breed Sex Age (years) Lesions Multiple/Solitary Localization 1 DSH M 14 Crusty plaques Multiple Thorax, face, shoulder 2 Cornish Rex M 12 Papillary plaques Multiple Shoulder, paw, trunk, abdomen 3 DSH F 11 Plaque Solitary Neck 4 DSH F 15 Plaques, papules Multiple Face, feet, abdomen 5 DSH 18 Papules Multiple Flank 6 DSH M 12 Crusts, plaques Multiple Face and digits 7 DSH F 13 Crusts, plaques Multiple Eyelids, face 8 American curl M 14 Crusts, plaques Multiple Face, neck 9 DSH F 13 Papules Multiple Unknown 10 DSH F 12 Papules Multiple Flank 11 Sphynx F 6 Macules Multiple Neck 12 DSH M 3 Ventral papules Multiple Abdomen 13 DSH M 16 Plaque Solitary Abdomen 14 DSH F 8 Papules Solitary Dorsum 15 DSH F 13 Crusts, erythema, erosion Multiple Axilla, feet 16 DSH M 13 Papules, crusts Multiple Face, neck 17 DLH F 9 Scaly papules Multiple Face 18 DSH F 12 Erythematous papules Multiple Neck, face 19 DLH M 11 Erythematous plaques Multiple Neck, face 20 Himalayan F 8 Crusty plaque Multiple Dorsum, neck 21 DSH F 15 Crusty plaque Multiple Leg, toe 22 DSH F 11 Crusty plaques Multiple Face, lip 23 DLH M 17 Plaques Multiple Dorsum 24 DSH F 11 Crusty plaques Multiple Face, lip 25 DSH M 16 Crusty plaques Multiple Face, neck, shoulders, foot pads 26 DSH M 12 Crust papules Multiple Face, neck DSH, domestic shorthair cat; DLH, domestic longhair cat.

epidermis with irregular acanthosis and broad rete ridges. Irregular acanthosis frequently descended around hair follicles. The epidermis was disorganized with a marked loss of cellular polarity and loss of normal stratification of the stratum basale and spinosum in all cases (wind-blown appearance). Keratinocytes with a hyperchromatic nucleus were present throughout the whole epidermis. Atypia was variable in nature and intensity (anisocytosis, anisocryosis and rare binucleated keratinocytes). Rare mitotic figures were present in all samples. Scattered apoptotic keratinocytes were present in four BISC samples. Koilocytes were present in all of them (Fig. 6). Other clear cells with rather enlarged vesicular nuclei were also observed. The cells (koilocytes and clear cells) contained sometimes intracytoplasmic additional blue-grey fibrillar pseudo-inclusions (three of nine cases). Clumped kerato- Figure 1. Cat no. 6. Pigmented plaque on the head diagnosed as hyalin granules were seen in one of nine BISC cases. Ero- feline viral plaque: Note the slightly raised and hyperpigmented lesion sions or ulcerations were present in five of nine cases. with a small central ulceration. Courtesy of Catherine Mège. Immunohistochemical examination seemed to result from the condensation of fibrillar ones Results are summarized in Table 3. Of the seven cases of (more prevalent in the stratum spinosum) (Fig. 4). Intranu- the FVP group, six were positive for PV antigen. Interest- clear inclusions were not observed. ingly, all of the 10 samples with BISC and FVP lesion types were positive (Fig. 7). Only one of the nine BISC cases FVP + BISC was deemed positive (11%). PV antigens were always vis- Interestingly, both BISC and FVP changes were present in ualized in the nucleus of the koilocytes; intracytoplasmic 10 cats, sometimes in the same, sometimes in different, pseudo-inclusions remained unstained (Fig. 4). skin samples (Fig. 5a,b). Transition lesions exhibiting both FVP and BISC features were also sometimes observed. Discussion BISC The clinical resemblance between BISC and FVP and the The diagnosis of BISC was made on nine cases. These presence of both lesions in some cats suggest that some lesions consisted of sharply demarcated expansion of the BISC evolve from FVP. Furthermore, despite the absence

62 FVP and BISC in cats with hyperpigmented plaques

Figure 3. Cat no. 10. Histology of a feline viral plaque. Note the presence of clear cells (ballooned cytoplasm, rather swollen nucleus (black arrow) with intracytoplasmic pseudo-inclusions (red arrow). Haematoxylin and eosin. Magniﬁcation ×10. Bar = 200 µm. Figure 2. Cat no. 26. Pigmented plaques at the base of the ear and the pinna diagnosed as feline bowenoid in situ carcinoma. Note the slightly raised and ulcerated lesions partially covered by crusts.

Table 3. Histopathological ﬁndings

Case Margins? Hyperpig. Koilocytes/clear cells Dyskerat. Comp. ps. incl. Fibr. ps. incl. KH Gran. Diagnosis PV-Ag 1Yes Yes Yes No No Yes No BISC Neg 2Yes Yes Yes No No No Yes BISC + FVP Pos 3YesYesYes Yes No No Yes FVP Pos 4Yes Yes Yes Yes Yes No Yes BISC + FVP Pos 5NoYes Yes No No No No BISC Neg 6Yes Yes Yes Yes Yes No Yes BISC + FVP Pos 7Yes Yes Yes No No No No BISC Neg 8Yes Yes Yes Yes No Yes No BISC Neg 9YesYesYes No Yes No Yes FVP Pos 10 Yes Yes Yes Yes Yes No Yes FVP Neg 11 Yes Yes Yes No Yes No Yes FVP Pos 12 Yes Yes Yes No Yes No Yes FVP Pos 13 Yes Yes Yes No No No Yes FVP Pos 14 Yes Yes Yes No No No Yes FVP Pos 15 No Yes Yes No No No No BISC Neg 16 Yes Yes Yes Yes No Yes Yes BISC + FVP Pos 17 Yes Yes Yes No No No No BISC Neg 18 No Yes Yes Yes No No Yes BISC Pos 19 Yes Yes Yes Yes No No Yes BISC + FVP Pos 20 Yes Yes Yes Yes Yes No Yes BISC + FVP Pos 21 Yes Yes Yes Yes No No No BISC Neg 22 Yes Yes Yes Yes No Yes No BISC Neg 23 Yes Yes Yes Yes No No Yes BISC + FVP Pos 24 Yes Yes Yes No Yes No Yes BISC + FVP Pos 25 Yes Yes Yes No No No Yes BISC + FVP Pos 26 Yes Yes Yes Yes No No Yes BISC + FVP Pos Margins?, presence of lesional margins; Hyperpig., hyperpigmentation; Dyskerat., dyskeratosis; Comp. ps. incl., compact pseudo-inclusions; Fibr. ps. inclus., ﬁbrillar pseudo-inclusions; KH Gran., clumped keratohyalin granules; PV-Ag, papillomavirus antigen. Pos, positive; Neg, negative.

63 Wilhelm et al.

Figure 4. Cat no. 9. Immunohistochemical analysis of a feline viral Figure 6. Cat no. 21. Histology of a feline bowenoid in situ plaque. Note the presence of positive nuclei (black arrow). The fibrillar carcinoma. Note the marked acanthosis (black stars: acanthotic (red arrow) and the solid (green arrow) intracytoplasmic inclusions epidermis), the follicular involvement (black points), the loss of remained unstained. Diaminobenzidine. Magnification ×40. polarity and the presence of numerous koilocytes (arrow). Bar = 50 µm. Haematoxylin and eosin. Magnification ×10. Bar = 200 µm.

of statistically signiﬁcant difference, cats affected by FVP affected. This ﬁnding does not seem to support the tended to be younger than those affected by BISC: this hypothesis that BISC evolve from FVP but the discrepancy could imply that FVP are precursor lesions of BISC. How- could be explained by a higher cancerization rate of lesions ever, while BISC affected the face, neck or the limbs in located on the face and neck, for example as a result of most cases, FVP lesions were more often present on the increased ultraviolet radiations exposure, compared to trunk even if other areas, including neck and face, were those in other regions of the body.

Figure 5. Cat no. 16. Histology of two lesions present on the same biopsy sample. Haematoxylin and eosin. Magniﬁcation ×40. Bar = 50 µm. (a) Feline viral plaque. Note the moderate acanthosis. The stratiﬁcation and the differentiation of the epidermis are conserved. Koilocytes and clumped keratohyalin granules are the most obvious papillomaviruses’ cytopathic characteristics on this lesion. (b). Early bowenoid in situ carcinoma. Note the acanthosis, the obvious disorganization of the epidermis and the abnormal differentiation of most keratinocytes. Clumped keratohyalin granules and one single koilocyte are the only papillomavirus cytopathic effects noticed on this lesion.

64 FVP and BISC in cats with hyperpigmented plaques

acids are often uncovered in normal mammalian skin. However, genome copy number is usually very low and productive infection rarely occurs in such cases.15,16 Establishing causality between the presence of viruses in skin lesions and oncogenesis remains problematic, and the presence of replicating viruses cannot be regarded as a sufficient proof. In vitro studies are mandatory to establish such causality.17 Almost all cats affected by BISC were deemed negative by IHC. These findings might suggest that BISC has two distinct causes and that only a subgroup of BISC is virally induced. A loss of viral replication during the cancerization process could also explain these findings. In fact latent PV infection or infection with minimal replication may remain undetected by IHC, because of the relatively low sensitivity of such techniques. The ‘hit and run’ model, which postu- Figure 7. Cat no 19. Immunohistochemical analysis of a feline bowe- lates an initial cellular transformation by the virus and a noid in situ carcinoma. Note the presence of numerous koilocytes and subsequent loss of viral genome, could account for the clear cells (red arrow) with positive nuclei. Novared. negative IHC in some BISC lesions.18 Furthermore, it was Magnification ×10. Bar = 200 µm. recently demonstrated that PVs maintained productive infections in precursory lesions of cervical cancer but that FVP usually conserved the general organization of the capsid antigens were no longer produced in late cervical epidermis and atypia was absent, whereas BISC lesions cancers.19 In conclusion, a loss of viral protein expression were disorganized and abnormal keratinocytes were present in advanced cases of BISC seems likely. throughout the epidermis. However, both conditions share Feline BISC has long been considered the counterpart numerous histological features: irregular acanthosis with of human Bowen’s disease (BD) – an in situ squamous cell rete ridges formations, presence of clumped keratohyalin carcinoma that presents as solitary, well-circumscribed, granules, koilocytes and clear cells. The presence of erythematous plaques and occurs on the face, extremities koilocytes or clear cells in all BISC lesions (including IHC- and genitalia.20,21 Koilocytes are usually not present in such negative ones) might be regarded as a proof of presence lesions.21 Human bowenoid papulosis is characterized of the virus. These cells with vacuolated cytoplasm and by genital pigmented verrucous papules or plaques.21 This shrunken, pycnotic nuclei are usually considered highly condition is also histologically characterized by in situ SCC suggestive of PV infections.8,13 All the authors who have lesions but, in contrast to BD, bowenoid papulosis lacks studied feline BISC have recognized these cells, but two full-thickness epidermal atypia. PV DNA is uncovered of three have not used the term ‘koilocyte’ to describe in virtually all samples of bowenoid papulosis but data them.8–10 In situ hybridization studies could be helpful to concerning the presence of PV in human BD remain determine if these cells actually harbour PV nucleic acids contradictory.22–25 Furthermore, PVs that infect human and if the term ‘koilocyte’ is appropriate. bowenoid papulosis and BD are usually to mucosal and In both FVP and BISC samples, fibrillar and compact not to cutaneous strains.23,24 These data show that feline pseudo-inclusions were seen. In one case both were BISC lesions display substantial differences from both present in the same sample, and compact ones (more human conditions and justify the use of a specific denom- present in the stratum granulosum) seemed to result from ination, as emphasized by Gross and coworkers.8 the condensation of fibrillar ones (more prevalent in the The results of the present study support the hypothesis stratum spinosum) (Fig. 4). This condensation has already that some BISC evolve from FVP lesions and the causative been described by Carney and coworkers.3 role of PV. However, evidence that these PVs are able to Our study demonstrates that the association between induce cancerization in mammalian skin is lacking and FVP and BISC is frequent and occurs sometimes on the further studies are warranted. Nucleic acids amplification same skin lesion. Additionally, cases of overlapping BISC techniques could establish which PVs are present in FVP and FVP lesions have been detected. This association was and BISC lesions and whether BISC samples without FVP already described before.11,12 These similarities support are really sterile or infected by dormant PV. As well, in vitro the hypothesis that FVP could be precursory lesions of studies addressing the transforming potential of feline PV BISC. are required to better understand the role that these All except one FVP and FVP + BISC cases were positive viruses play in this condition. for PV antigen by immunohistochemistry (IHC). As pseudo-inclusions were present in the negative case, it can be considered that all these samples were infected by References PV. Furthermore, as IHC detects capsid antigens, it can be concluded that productive infection occurred in all positive 1. de Villiers E-M, Fauquet C, Broker TR et al. Classification of papillomaviruses. Virology 2004; 324: 17–27. samples (all FVP lesions and positive BISC). These findings 2. Nicholls PK, Stanley MA. The immunology of animal papilloma- support the hypothesis that PVs play an active role in the viruses. Veterinary Immunology and Immunopathology 2000; 73: development of such lesions. It must, however, be borne 101–27. in mind that PVs are sometimes commensal, and nucleic 3. Carney HC, England JJ, Hodgin EC et al. Papillomavirus infection

65 Wilhelm et al.

of aged Persian cats. Journal of Veterinary Diagnostic Investigation fixed paraffin-embedded tissues for diagnostic pathology. Journal 1990; 2: 294–9. of Veterinary Diagnostic Investigation 1991; 3: 101–12. 4. Carpenter JL, Kreider JW, Alroy J, Schmidt GM. Cutaneous xan- 15. Antonsson A, Hansson BG. Healthy skin of many animal species thogranuloma and viral papilloma on a eyelid of a cat. Veterinary harbours papillomaviruses which are closely related to their Dermatology 1992; 3: 187–90. human counterparts. Journal of Virology 2002; 76: 12537–42. 5. Egberink HF, Berrocal A, Bax HAD et al. Papillomavirus associated 16. Majewski S, Jablonska S. Human papillomavirus and oncogenesis: skin lesions in a cat seropositive for feline immunodeficiency critical evaluation of recent findings. International Journal of virus. Veterinary Microbiology 1992; 31: 117–25. Dermatology 2002; 41: 319–20. 6. Lozano-Alarcon F, Lewis II TP, Clark EG et al. Persistent papillo- 17. Harwood CA, Proby CM. Human papillomaviruses and non- mavirus infection in a cat. Journal of the American Animal Hospital melanoma skin cancer. Current Opinion in Infectious Diseases Association 1996; 32: 392–6. 2002; 15: 101–14. 7. Sundberg JP, van Ranst M, Montali R et al. Feline papillomas and 18. Smith KT, Campo MS. ‘Hit and run’ transformation of mouse papillomaviruses. Veterinary Pathology 2000; 37: 1–10. C127 cells by bovine papillomavirus type 4: the viral DNA is 8. Gross TL, Ihrke PJ, Walder EJ, Affolter VK. Epidermal tumors. In: required for the initiation but not for maintenance of the trans- Gross TL et al., eds. Skin Diseases of the Dog and Cat: Clinical and formed phenotype. Virology 1988; 164: 39–47. Histopathological Diagnosis. Oxford: Blackwell Science, 2005: 19. Doobar J. Molecular biology of human papillomavirus infection 562–577. and cervical cancer. Clinical Science 2006; 110: 525–41. 9. Baer KE, Helton K. Multicentric squamous cell carcinoma in situ 20. Arlette JP, TrotterMJ. Squamous cell carcinoma in situ of the skin: resembling Bowen’s disease in cats. Veterinary Pathology 1993; history, presentation, biology and treatment. Australasian Journal 30: 535–43. of Dermatology 2004; 45: 1–11. 10. Miller WH Jr, Affolter V, Scott DW, Suter MM. Multicentric 21. Duncan KO, Lefell DJ. Epithelial precancerous lesions. In: Freed- squamous cell carcinomas in situ resembling Bowen’s disease in berg IM et al., eds. Fitzpatrick’s Dermatology in General Medicine. five cats. Veterinary Dermatology 1992; 3: 177–82. New-York: Mc Graw-Hill, 2003: 719–36. 11. LeClerc SMC, Haines EG. Papillomavirus infection in association 22. Mitsuishi T, Kawana S, Kato T, Kawashima M. Human papilloma- with feline cutaneous squamous cell carcinoma in situ. In: virus infection in actinic keratosis and Bowen’s disease: comparative Proceedings of the AAVD/ACVD Meeting 1997: 125–126. study with expression of cell-cycle regulatory proteins p21waf1/ 12. Gross TL, Affolter VK. Advances in skin oncology. In: Kwochka cip1, 53, pcna, ki-67, and bcl-2 in positive and negative lesions*1. KW, Willemse T, von Tschaner C, eds. Advances in Veterinary Human Pathology 2003; 34: 886–92. Dermatology III. Boston: Butterworth-Heinemann, 1998: 382– 23. Mitsuishi T, Sata T, Matsukura T, Iwasaki T, Kawashima M. The 385. presence of mucosal human papillomavirus in Bowen’s disease 13. McLeod K. Prediction of human papillomavirus antigen in cervical of the hands. Cancer 1997; 79: 1911–7. squamous epithelium by koilocytes nuclear morphology and ‘wart 24. Quereux G, N’Guyen JM, Dreno B. Human papillomavirus and scores’: confirmation by immunoperoxydase. Journal of Clinical extragenital in situ carcinoma. Dermatology 2004; 209: 40–5. Pathology 1987; 40: 323–8. 25. Lu S, Syrjanen K, Havu VK. Failure to demonstrate human 14. Haines DM, Chelack BJ. Technical considerations for developing papillomavirus (HPV) involvement in Bowen’s disease of the enzyme immunohistochemical staining procedures on formalin- skin. Archives of Dermatology Research 1996; 289: 40–5.

Résumé Les plaques virales du chat (FVP) induites par les papillomavirus (PV) se présentent souvent comme des plaques hyperpigmentées. Le fait que jusqu’à 47% des carincomes in situ bowenoides (BISC), qui se présentent aussi sous la forme de plaques hyperpigmentées, sont positifs pour l’antigène de PV par immunohistochimie suggère que les BISC pourraient provenir de FVP. La relation entre la présence d’antigènes de PV et les données cliniques et histologiques de 26 cas de dermatoses félines cliniquement répertoriées comme des plaques hyperpigmentées avec un diagnostic histologique de FVP et/ou de BISC a été recherchée. Les cas ont été classés en trois groupes : FVP, FVP + BISC ou BISC. La recherche immunohistochimique de papillomavirus a été réalisée en utilisant un antisérum polyclonal de lapin anti-bovin. Sur les sept cas du groupe FVP, six étaient positifs à l’immunohistochimie, un seul des neuf BISC était positif. La présence de lésions de FVP et de BISC chez certains chats, et la fréquence importante de découverte d’antigènes de PV dans les groupes FVP et FVP + BISC suggère que ces deux maladies ont une même cause virale, et que certains BISC peuvent provenir de FVP. Le faible taux de détection d’antigène viral dans le groupe BISC indique une autre cause, ou la perte de la réplication virale pendant la cancérogénèse.

Resumen Las placas virales felinas (FVP) inducidas por el virus papiloma son a menudo verrugas hiperpigmentadas y planas. El hecho de que hasta un 47% de los carcinomas Bowenoides in situ (BISC), que también ocurren como placas hiperpigmentadas, son positivos al antígeno del virus papiloma mediante inmunohistoquímica sugiere que los BISC pueden evolucionar a partir de placas virales felinas. Se determinó la relación entre la presencia de antígenos del virus del papiloma y las características clínicas e histológicas de 26 casos de dermatosis (clínicamente descritas como placas pigmentadas y con diagnostico histológico de FVP y/o BISC). Los casos se clasiﬁcaron en uno de los tres grupos siguientes: FVP, FVP + BISC o BISC. La detección inmunohistológica de antígeno especiﬁco del grupo del virus papiloma se realizó utilizando un antisuero policlonal de conejo frente al papiloma bovino. De los siete caso en el grupo FVP, seis fueron considerados positivos mediante inmunohistoquímica así como los diez gatos del grupo FVP + BISC. Por otro lado, solo uno de los nueve gatos con BISC fue positivo. La presencia de ambas lesiones FVP y BISC en algunos gatos y el elevado nivel de detección de antígenos del virus papiloma en los grupos FVP y FVP + BISC sugiere que ambas condiciones podrían tener la misma causa vírica y que algunos BISC podrían

66 FVP and BISC in cats with hyperpigmented plaques

progresar desde FVP. El bajo porcentaje de detección de antígeno vírico en el grupo BISC sugiere otra causa o una pérdida de replicación viral durante el proceso de carcinogénesis.

Zusammenfassung Feline virale Plaques (FVP), die von Papillomavirus (PV) verursacht werden, sind oft hyperpigmentierte und flache Warzen. Die Tatsache, dass bis zu 47% der ‘Bowen’-ähnlichen in situ Karzinome (BISC), die normalerweise auch in Form von hyperpigmentierten Plaques erscheinen, mittels Immunchemie positiv sind für PV-Antigen, weist darauf hin, dass BISC sich aus FVP entwickeln könnte. Der Zusammenhang zwischen dem Auftreten von PV Antigenen und den klinischen und histologischen Erscheinungsbildern von 26 Fällen von felinen Dermatosen (die klinisch als pigmentierte Plaques beschrieben und histologisch als FVP und/oder BISC diagnostiziert wurden) wurde daher bestimmt. Die Fälle wurden in eine der drei folgenden Gruppen eingeteilt: FVP, FVP + BISC oder BISC. Die immunhistologische Bestimmung des gruppenspezifischen Papillomavirus Antigens wurde mit einem polyklonalen Kaninchen Antiserum gegen bovines Papillomavirus durchgeführt. Von den sieben Fällen in der FVP Gruppe wurden sechs mittels Immunhistologie als positiv angesehen, genauso wie alle 10 Katzen in der FVP + BISC Gruppe. Andererseits war nur eine der neun BISC Katzen positiv. Das Vorhandensein von beiden, FVP und BISC Läsionen bei manchen Katzen und das häufige Auftreten von PV-Antigenen in den FVP und FVP + BISC Gruppen ist ein Hinweis darauf, dass beide Formen dieselbe virale Ursache haben und einige BISC sich aus den FVP entwickeln könnten. Das seltene Auftreten von viralem Antigen in der BISC Gruppe bedeutet, dass eine andere Ursache vorliegt oder der Verlust von viraler Replikation während der Kanzerogenese besteht.

67 Chapter 9

Summarizing discussion and further studies

Viruses replicate inside cells by synthesizing their own proteins and assembling them into virions. This replication is associated with various cytopathic effects, which are, usually, typical or pathognomonic of one specific virus. Poxviruses infections are associated with large intracytoplasmic inclusions and herpesvirus infections with intranuclear inclusions, for example. Aside from these cytopathic effects, viruses induce macroscopic changes which are sometimes, easily recognizable. Papillomaviruses induce cauliflower-like lesions, the so- called warts, which are virtually pathognomonic. Poxviruses and herpesviruses induce pock lesions and vesicles, respectively, which are very typical of these infections. These virus- associated changes have long been described in dogs and cats [1]. Viruses may also induce some less obvious changes, which are described in humans but remained often undescribed in canine and feline. These changes may be due to various pathogenic states like minimal viral replication, latency or non-productive infections. This thesis aims to describe some of these undescribed virus-induced skin changes in dogs and cats and, especially, papillomavirus-induced ones.

We first described a case of canine erythema multiforme presumably associated with parvovirus infection [2]. We hypothesized that an infection of stem cells and primary amplifying keratinocytes occurred following hematogenic dissemination of the parvovirus. Viral antigens could have been presented by class I major histocompatibility complex molecules at the surface of the keratinocytes. Recognition of the viral antigens by T- lymphocytes, possibly sensitised by a previous parvovirus vaccination would have triggered these cytotoxic T-cells to induce apoptosis of infected keratinocytes. In this case, clinical and pathological lesions are not due to the cytopathic effect of the virus itself but to the T lymphocyte-induced cytolysis. Interestingly, virus infections (especially herpes simplex infection but also B19 parvovirus infections) are the most frequent causes of erythema multiforme in humans [3, 4]. This case was the first report of virus-associated erythema multiforme in dogs. This report leaves however some moot questions that warrant some further studies. The most important question is to know whether canine parvovirus usually replicates in the skin of affected dogs without causing any cytopathic effects or if the skin contamination reported in this study was incidental or due to a specific parvovirus strains. Second, as parvovirus antigens have been uncovered in the affected skin, one cannot exclude a direct effect of the virus infection associated to a secondary lymphocytic reaction.

The second article of this thesis aimed to describe some previously unknown cutaneous consequences of FeLV infection. FeLV is a member of the oncornavirus subfamily of retroviruses, which replicates in many tissues like bone marrow, salivary glands and respiratory epithelium. Its replication in the feline skin was already described by Gross and coworkers and associated with the so-called giant cell (multinucleated keratinocytes) dermatosis and horn formation [1, 5]. Multinucleated keratinocytes are sometimes observed in humans in association with neoplastic conditions, infectious diseases like herpesvirus infection and immunologic disorders [6]. Retroviruses also possess fusion proteins, which are able to induce syncytium formation in infected tissues [7, 8]. However, although FeLV infection is a frequent disease, syncytium are rarely observed in the affected skin of infected cats [5]. We described another case of FeLV-associated giant cell dermatosis with an ulcerative phenotype and demonstrated the presence of both FeLV antigens and proviral sequences in the lesional skin. Gross and coworkers suggested that these cytopathic effects were not the direct consequence of FeLV infection but the early stage of carcinomatous transformation [5]. The presence of FeLV antigens in the affected skin of the cat we observed, suggested an active replication of the virus and supported the hypothesis of a direct cytopathic effect. Furthermore, Rohn and coworkers demonstrated that FeLV variants do possess various pathogenic and cytopathic effects[9]. All in all, we considered more likely that these changes are the direct consequence of infection with a specific and rare variant of FeLV. This hypothesis however warrants further investigation. Feline internal lymphomas are often the consequence of FeLV infection. Cutaneous lymphomas, however, usually occur on FeLV-negative cats [10]. FeLV genomic sequences have however already been sometimes amplified from cutaneous lymphomas [11]. The originality of the case we reported lies on the fact that FeLV antigens have been demonstrated in the affected skin of a serologically negative cat. These findings suggest that productive FeLV infection may in some instances occur and may be restricted to the skin. Further studies are needed to demonstrate the existence of multiple FeLV strains with various physiologic and pathologic properties.

Papillomaviruses (PV) are host-specific epitheliotropic viruses that infect the skin and mucous membranes. As these viruses do not possess the enzymatic machinery required for replication, they depend upon host-cell machinery to achieve this process and upon host-cell

differentiation for completion of their life cycle [12]. As more than 150 different PV have been isolated from the human lesional or healthy skin, only a few PV have been identified in carnivores [13, 14]. In this thesis, we have demonstrated the existence of new papillomavirus-like sequences in various canine and feline lesions, including cyclosporine A- associated exophytic lesions, in situ and invasive carcinomas [15, 16]. We have used two sets of primers designed for the amplification of a sequence of the E1 gene of PV. The narrow- range set of primers was supposed to amplify canine and feline PV and their close relatives [16]. The broad range PCR system was designed to amplify up to 64 human PVs and several animal PV such as canine and feline PV[16, 17]. These studies have shown the existence of at least six feline and five canine unknown papillomavirus-like sequences. As the classification of papillomaviruses is based upon L1 gene, the amplification of sequences of the E1 gene does not allow proper evaluation of these sequences and classification of the newly uncovered PVs but these results suggested however that canine and feline lesional skin can be infected by PV of great genetic diversity. It would be of great interest to amplify and clone these novel canine and feline PVs. Fortunately, a new technique, the rolling-circle amplification (RCA) technique, was recently introduced to amplify and isolate circular DNA and, especially human and animal PVs [18, 19]. RCA is a multiple random primed, sequence- independent amplification of circular DNA. Furthermore, the amplification is as effective as PCR. Therefore, only minute amount of crudely isolated DNA from tissue can be used for amplification of papillomavirus genomic DNA. RCA analysis of canine and feline skin samples will permit to determine whether healthy skin harbors PVs and to sequence PVs that are present in lesional skin. This descriptive study is the mandatory initial step for a better understanding of the role that play PV in the development of skin lesion in dogs and cats, and, especially, in the development of skin cancers. We have already applied this new technique to the isolation and cloning of a new canine PV (CPV3)[20].

In mammals and birds, PV induce a wide range of cutaneous and mucous changes such as exophytic and flat warts, precancerous and cancerous lesions. They are considered important carcinogens in humans and some high-risk PVs are directly responsible for the development of cervical cancers in women[21-23]. Even though the link between cervical carcinoma and human PV is clear, the role of PVs in the development of cutaneous squamous cell carcinoma (SCC) is not as definite [24]. There is however emerging epidemiological evidence to suggest

that PV might play an important role in skin cancerogenesis, especially in epidermodysplasia (EV) associated-one. Establishing causality between the presence of PV in a skin lesion and the development of the lesion is nevertheless problematic [25]. Criteria have been proposed to establish this relationship but difficulties in culturing PVs have made their fulfillment often impossible [22, 25]. Additionally the use of extremely sensitive nested polymerase chain reaction (PCR) makes possible the detection of minute amounts of viral DNA (even 0.05 viral genome per cell). The presence of such an amount of PV nucleic acid does not indicate a productive infection and can also be found in healthy skin [26, 27]. Evidence also suggests that ultraviolet (UV) radiation contributes to the cancerization of some PV- associated skin cancers [25, 28]. All in all, the role of PV in the development of skin cancer in humans remains questionable. Some animals models support the causative role of PV in the induction of skin SCC: A few decades ago, it was demonstrated that cottontail-rabbit PV (CRPV) are able to induce skin cancers in rabbit [29-30]. Other studies have also established that attenuated life canine oral PV (COPV) vaccine induce SCC in Beagles [32]. Additionally, canine and feline can be affected by skin conditions that share some similarities with human EV and cancerization has been reported in some patients [33]. Epidemiologic studies have demonstrated association between carnivores SCC and PV but causality has never been established [16, 33-40]. This thesis has confirmed the epidemiologic association between some groups of feline and canine skin cancers and PV infections and the genetic diversity of carnivore PVs [16, 20, 34, 40]. Aside from the identification and cloning of these new PVs, the most important studies to carry out would be to determine the relative prevalence of each new carnivore PV and to demonstrate in vitro that, at least some of them, are able to induce keratinocyte transformation and immortalization. We have, for example, detected, cloned, and sequenced a novel PV (CPV3), which was associated with a case of canine epidermodysplasia verruciformis [20]. The affected dog developed multiple plaques and one single interdigital lesion of in situ SCC [20]. DNA of CPV3 was uncovered in each lesion tested (including SCC) but was not present in intact skin of the same dog. Sequence-independent, multiply primed rolling-circle amplification was used to amplify, clone, and sequence the entire genome of CPV3. Indeed, analysis of the cloned and sequenced canine papilloma genome allowed its classification as a member of a new papillomavirus genus (GenBank accession DQ295066). Additionally, mRNA for the putative transforming protein E6 was discovered in each tested lesion: These findings

demonstrated that CPV3 was transcriptionally active in the mentioned skin lesions and supported the hypothesis of a causative role of CPV3 in the pathogenesis of canine EV [41]. Complementary DNA of the p53 transcript of the affected dog was cloned from blood, intact skin as well as skin lesions and its nucleotide sequence was found not to differ from the wild type canine p53 sequence. This finding suggested that the development of malignancy could not be attributed to UV-dependent mutagenesis. Therefore, the hypothesis was supported that transforming proteins of CPV3 induced cancer development by different mechanism than human EV-associated PVs [41]. As well, we are currently studying the relative prevalence of CPV3 in canine sera. We have cloned CP3-L1 gene (codes for major CPV3 capsid protein) and generated antisera against this protein. Our goal would be to establish an ELISA test and to evaluate 500 already collected canine sera. All in all, the findings in this thesis, the subsequent and future studies open new avenues to study PV-induced skin cancerogenesis. As at least some carnivore conditions bear major resemblances with human ones, these breakthroughs will reveal helpful for both veterinary and human oncology.

1. Scott, D.W., W.H. Miller, and C.E. Griffin, Viral, rickettsial and protozoal diseases, in Muller & Kirk's Small Animal Dermatology, D.W.M. Scott, W.H. Griffin, C.E., Editor. 2001, W.B. Saunders: Philadelphia. p. 517-542. 2. Favrot, C., et al., Parvovirus infection of keratinocytes as a cause of canine erythema multiforme. Vet Pathol, 2000. 37(6): p. 647-649. 3. Huff, J.C., Erythema multiforme. Dermatol Clin, 1985. 3(1): p. 141-152. 4. Huff, J.C., W.L. Weston, and M.G. Tonnesen, Erythema multiforme: a critical review of characteristics, diagnostic criteria, and causes. J Am Acad Dermatol, 1983. 8(6): p. 763- 775. 5. Gross, T.L., et al., Giant cell dermatosis in FeLV-positive cats. Vet. Dermatol., 1993. 4(3): p. 117-122. 6. Kimura, S.K. and H. Hatano, Multinucleate epidermal cells in non-neoplastic dermatoses. Brit J Dermatol, 1978. 99: p. 485-489. 7. Fenyo, E.M., et al., Distinct replicative and cytopathic characteristics of human immunodeficiency virus isolates. J Virol, 1988. 62(11): p. 4414-4419. 8. White, J.M., Membrane fusion. Science, 1992. 258: p. 917-924. 9. Rohn, J.L., et al., In vivo evolution of a novel, syncytium-inducing feline leukemia virus variant. Journal of Virology, 1998. 72(4): p. 2686-2696. 10. Vonderhaar, M.A. and W.B. Morisson, Chapter 45. Lymphosarcoma, in Cancer in dogs and cats, W.B. Morisson, Editor. 2002, Teton New Media: Jackson, Wyoming. p. 641- 670.

11. Tobey, J.C., et al., Cutaneous T-cell lymphoma in a cat. J Am Vet Med Assoc, 1994. 204(4): p. 606-609. 12. Doorbar, J., The papillomavirus life cycle. J Clin Virol, 2005. 32(Supplement 1): p. 7-12. 13. de Villiers, E.-M., et al., Classification of papillomaviruses. Virolo, 2004. 324: p. 17-27. 14. Scott, D.W., W.H. Miller, and C.E. Griffin, Neoplastic and non-neoplastic tumors, in Kirk and Muller's Small Animal Dermatology, D.W. Scott, W.H. Miller, and C.E. Griffin, Editors. 2001, W.B. Saunders Co.: Philadelphia. p. 1236-1413. 15. Favrot, C., et al., Evaluation of papillomaviruses associated with cyclosporine-induced hyperplastic verrucous lesions in dogs. Am J Vet Res, 2005. 66(10): p. 1764-1769. 16. Zaugg, N., et al., Detection of novel papillomaviruses in canine mucosal, cutaneous and in situ squamous cell carcinomas. Vet Dermatol, 2005. 16(5): p. 290-298. 17. Iftner, A., et al., The prevalence of human papillomavirus genotypes in nonmelanoma skin cancers of nonimmunosuppressed individuals identifies high-risk genital types as possible risk factors. Cancer Res, 2003. 63(21): p. 7515-7519. 18. Rector, A., et al., Isolation and cloning of a papillomavirus from a North American porcupine by using multiply primed rolling-circle amplification: the Erethizon dorsatum papillomavirus type 1. Virol, 2005. 331(2): p. 449-456. 19. Rector, A., R. Tachezy, and M.A. Van Ranst, Sequence independant strategy for detection and cloning of circular DNA virus genome by using multiply primed rolling circle amplification. J. Virol., 2004. 78(10): p. 1993-1998. 20. Tobler, K.et al., Detection of the prototype of a potential novel genus among the papillomaviridae in association with canine epidermodysplasia verruciformis. J Gen Virol, 2006. 67(12): p. 2036-2041 21. Saladi, R.N. and A.N. Persaud, The causes of skin cancer: a comprehensive review. Drugs Today (Barc), 2005. 41(1): p. 37-53. 22. zur Hausen, H., Papillomavirus infections--a major cause of human cancers. Biochim Biophys Acta, 1996. 1288(2): p. F55-78. 23. zur Hausen, H., Oncogenic DNA viruses. Oncogene, 2001. 20(54): S. 7820-7823. 24. Sterling, J.C., Human papillomaviruses and skin cancer. J ClinVirol, 2005. 32(Supplement 1): p. 67-72. 25. Harwood, C.A. and C.M. Proby, Human papillomaviruses and non-melanoma skin cancer. Curr Opin Infect Dis, 2002. 15(2): p. 101-114. 26. Antonsson, A. and B.G. Hansson, Healthy skin of many animal species harbours papillomaviruses which are closely related to their human counterparts. J Virol, 2002. 76(24): p. 12537-12542. 27. Majewski, S. and S. Jablonska, Human papillomavirus and oncogenesis: critical evaluation of recent findings. Int J Dermatol, 2002. 41(6): p. 319-320. 28. Termorshuizen, F., et al., Sunlight Exposure and (Sero) Prevalence of Epidermodysplasia Verruciformis-Associated Human Papillomavirus. J Invest Dermatol, 2004. 122(6): p. 1456-1462. 29. Breitburd, F., J. Salmon, and G. Orth, The rabbit viral skin papillomas and carcinomas: a model for the immunogenetics of HPV-associated carcinogenesis. Clin Dermatol, 1997. 15(2): p. 237-47. 30. Rous, P. and J.W. Beard, The progression to carcinoma of virus-induced rabbit papillomas (Shope). J Experim Med, 1935(62): p. 523-548. 31. Brandsma, J.L., The cottontail rabbit papillomavirus model of high-risk HPV-induced disease. Methods Mol Med, 2005. 119: p. 217-235. 32. Bregman, C.L., et al., Cutaneous neoplasms in dogs associated with canine oral papillomavirus vaccine. Vet Pathol, 1987. 24(6): p. 477-87.

33. Gross, T.L., et al., Epidermal tumors, in Skin diseases of the dog and cat: Clinical and histopathological diagnosis, T.L. Gross, et al., Editors. 2005, Blackwell Science: Oxford. p. 562-577. 34. Nespeca, G., et al. Detection of novel papillomavirus-like sequences in paraffin- embeddedspecimens oe invasive and in situ squamous cell carcinomafrom cats. Am J Vet Res, 2006. 26(12): p. 2036-2041 35. LeClerc, S.M., E.G. Clark and D.M Haines, Papillomavirus infection in association with feline cutaneous squamous cell carcinoma in situ (Abstract). in AAVD/ACVD Meeting. 1997.p. 125-126. 36. Schwegler, K., J.H. Walter, and R. Rudolph, Epithelial neoplasms of the skin, the cutaneous mucosa and the transitional epithelium in dogs: an immunolocalization study for papillomavirus antigen. Zentralbl Veterinarmed A, 1997. 44(2): p. 115-123. 37. Sundberg, J.P., R.E. Junge, and W.D. Lancester, Immunoperoxidase localization of papillomaviruses in hyperplastic and neoplastic epithelial lesions in animals. American JtVet Rest, 1984. 45(7): p. 1441-1446. 38. Teifke, J.P., et al., Detection of papillomavirus-DNA in mesenchymal tumour cells and not in the hyperplastic epithelium of feline sarcoids. Vet Dermatol, 2003. 14(1): p. 47-56. 39. Teifke, J.P., C.V. Lohr, and H. Shirasawa, Detection of canine oral papillomavirus-DNA in canine oral squamous cell carcinomas and p53 overexpressing skin papillomas of the dog using the polymerase chain reaction and non-radioactive in situ hybridization. Vet Microbiol, 1998. 60(2-4): p. 119-130. 40. Wilhelm, S. et al.Clinical, histological and immunohistochemical study of feline viral plaques and bowenoid in situ carcinomas. Vet Dermatol, 2006. 17: p. 424-431 41. Erne, M.L. Further Characterization of a novel canine papillomavirus (CPV3) and its potential role in the context of Epidermodysplasia Verruciformis. Dissertation. Zürich 2006.

75 Chapter 10

Zusammenfassung und weitere Studien

Viren replizieren sich innerhalb von Zellen indem sie ihre eigenen Proteine synthetisieren und diese zu Virionen assemblieren. Diese Replikation wird von verschiedenen zytopathischen Effekten begleitet, die in der Regel typisch oder pathognomonisch für ein bestimmtes Virus sind. Zum Beispiel werden Pockenvirusinfektionen von grossen intrazytoplasmischen Inklusionen begleitet und Herpesvirusinfektionen von intranuklearen Inklusionen. Abgesehen von diesen zytopathischen Effekten rufen Viren auch makroskopische Veränderungen hervor, die manchmal einfach zu erkennen sind. Papillomaviren rufen blumenkohlähnliche Läsionen hervor, sogenannte Warzen, die nahezu pathognomonisch sind. Pocken- und Herpesviren rufen Pockenpusteln bzw. Blasen hervor, die sehr typisch sind für diese Infektionen. Diese Viren begleitenden Veränderungen wurden schon lange für Hunde und Katzen beschrieben [1]. Viren können auch weniger offensichtliche Veränderungen hervorrufen, die für Menschen beschrieben sind, für Hunde und Katzen aber grösstenteils nicht. Diese Veränderungen können auf unterschiedlichen pathogenischen Zuständen beruhen, wie minimale virale Replikation, Latenz oder nicht-produktive Infektion. Ziel dieser Arbeit ist es, einige der von Viren hervorgerufenen Hautveränderungen zu beschreiben, die bei Katzen und Hunden bisher nicht bekannt waren, besonders die von Papillomaviren verursachten.

Wir waren die ersten, die einen Fall von Erythema multiforme (EM) bei Hunden beschrieben haben, das mit einer Parvovirusinfektion assoziiert war [2]. Wir haben angenommen, dass die Infektion von Stammzellen und primären amplifizierenden Keratinozyten nach einer hämatogenen Ausbreitung des Parvovirus auftrat. Virale Antigene wurden demnach von Haupthistokompatibilitätskomplex-Molekülen der Klasse I auf der Oberfläche von Keratinozyten präsentiert. Wir gingen davon aus, dass die Erkennung des viralen Antigens von T-Lymphozyten, die möglicherweise durch eine vorgängige Parvovirus-Impfung sensibilisiert wurden, diese zytotoxischen T-Zellen dazu gebracht haben, die Apoptose von infizierten Keratinozyten herbeizuführen. In diesem Fall sind die klinischen und pathologischen Läsionen nicht auf den zytopathischen Effekt des Virus selber zurückzuführen, sondern auf die von T-Lymphozyten hervorgerufene Zytolyse. Interessanterweise sind virale Infektionen (besonders die Herpes simplex Infektion, aber auch die B19 Parvovirus Infektion) die häufigste Ursache von EM bei

Menschen [3,4]. Unser Bericht war der erste veröffentlichte Fall von Virus assoziiertem EM bei Hunden. Er liess aber einige Fragen offen, die weitere Untersuchungen rechtfertigen. Die wichtigste Frage war, ob Parvoviren von Hunden sich in der Haut von betroffenen Hunden ohne zytopathische Effekte replizieren oder ob die Kontaminierung der Haut, von der wir in unserer Studie berichteten, zufällig oder auf einen spezifischen Stamm von Parvoviren zurückzuführen war. Zweitens konnten wir eine direkte Wirkung des mit einer sekundären lymphotischen Reaktion assoziierten Virus nicht ausschliessen, weil Parvovirus-Antigen in der befallenen Haut gefunden wurde.

Das Ziel des zweiten Teils dieser Arbeit war es, einige vorgängig unbekannte kutane Folgen der FeLV-Infektion zu beschreiben.

FeLV ist ein Mitglied der Oncornavirus-Unterfamilie von Retroviren und repliziert sich in vielen Geweben wie zum Beispiel im Knochenmark, in Speicheldrüsen und im Atemwegsepithel. Seine Replikation in feliner Haut wurde schon von Gross und Mitarbeitern beschrieben und ist von Riesenzellen (multinukleare Keratinozyten) Dermatose assoziiert. Mulitnukleare Keratinozyten werden manchmal bei Menschen im Zusammenhang mit neoplastischen Konditionen, infektiösen Krankheiten wie Herpesvirus Infektion und immunologischen Störungen beobachtet [6]. Retroviren besitzen auch Fusionsproteine, die in infizierten Geweben die Bildung von Synzytium herbeiführen können [7,8]. Dennoch werden Synzytia selten in der Haut von infizierten Katzen beobachtet, obwohl die FeLV Infektion eine häufige Krankheit ist. Wir haben einen Fall von FeLV assoziierter Riesenzellen-Dermatosis mit einem ulzerativen Phänotyp beschrieben und haben sowohl die Präsenz von FeLV-Antigenen als auch von proviralen Sequenzen nachgewiesen. Gross und Mitarbeiter haben vorgeschlagen, dass diese zytopathischen Effekte nicht die direkte Folge einer FeLV Infektion sind, sondern ein frühes Stadium von karzinomatöser Transformation [5]. Das Vorhandensein von FeLV-Antigenen in den Hautläsionen der Katze in unserer Studie legte die aktive Replikation des Virus nahe und stützte die Hypothese eines zytopathischen Effektes. Ferner haben Rohn und Mitarbeiter gezeigt, dass FeLV-Varianten verschiedene pathogenische und zytopathische Effekte haben [9]. Angesichts dieser Ergebnisse haben wir vermutet, dass es wahrscheinlicher ist, dass die Veränderungen die direkte Folge

einer Infektion mit spezifischen und seltenen Varianten von FeLV sind. Diese Auffassung bedarf jedoch weiterer Untersuchungen. Feline Lymphome sind oft das Resultat einer FeLV-Infektion. Ein kutanes Lymphom tritt jedoch normalerweise bei FeLV-negativen Katzen auf [10]. Feline Genomsequenzen von Leukämieviren wurden manchmal durch kutane Lymphome amplifizert [11]. Die Einzigartigkeit unseres Falles liegt in der Tatsache, dass FeLV-Antigene in den Hautläsionen einer serologisch negativen Katze nachgewiesen wurden. Diese Erkenntnisse legen nahe, dass eine produktive FeLV-Infektion in gewissen Fällen auftreten und auf die Haut beschränkt sein kann. Weitere Studien sind nötig um zu untersuchen, ob multiple FeLV-Stämme mit verschiedenen physiologischen und pathologischen Eigenschaften existieren.

Papillomaviren sind wirtespezifische epitheliotrope Viren, die Haut und Schleimhäute infizieren. Weil diese Viren keine enzymatische Ausrüstung besitzen, die für eine Replikation benötigt wird, sind sie von Ausrüstungen der Wirtszellen abhängig, um diesen Prozess zu vollbringen, von einer Differenzierung der Wirtszelle, um ihren Lebenszyklus zu vollenden [12]. Obwohl mehr als 150 verschiedene Papillomaviren von gesunder und kranker menschlicher Haut isoliert wurden, wurden nur wenige in Karnivoren identifiziert [13,14]. In der vorliegenden Studie wurden neue Papilloma- ähnliche Sequenzen in verschiedenen caninen und felinen Läsionen nachgewiesen, einschliesslich Cyclosporin A-assoziierte exophytische Läsionen und in-situ und invasive Karzinome [15,16]. Es wurden zwei Sets von Primern verwendet, die für die Amplifikation einer Sequenz des E1-Gens des Papillomavirus entwickelt wurden. Das Set von Primern mit begrenzter Wirkung wurde entwickelt, um canine und feline Papillomaviren und deren nahe Verwandten zu amplifizieren [16]. Das umfassende PCR System wurde entwickelt, um bis zu 64 humane Papillomaviren und mehrere Papillomaviren von Tieren einschliesslich die caninen und felinen zu amplifizieren [16, 17]. Diese Untersuchungen haben die Existenz von mindestens sechs felinen und fünf caninen zuvor unbekannten Papillomavirus-ähnlichen Sequenzen gezeigt. Weil die Klassifikation der Papillomaviren auf dem L1-Gen basiert, erlaubt die Amplifikation von Sequenzen des E1-Gens keine einwandfreie Evaluation dieser Sequenzen oder

Klassifikation der neu entdeckten Papillomaviren. Diese Resultate haben jedoch nahegelegt, dass canine und feline Hautläsionen von Papillomaviren von beträchtlicher genetischer Diversität infiziert werden können. Es wäre von enormem Interesse, diese canine und feline Papillomaviren zu amplifizieren und zu klonen. Glücklicherweise wurde kürzlich eine neue Technik, das Rolling-circle Amplifikationsverfahren (rolling- circle amplification, RCA) eingeführt, um zirkuläre DNA und insbesondere Papillomaviren von Menschen und Tieren zu amplifizieren und isolieren [18,19]. Das Rolling-circle Amplifikationsverfahren ist eine multiple „random primed“, Sequenz unabhängige Amplifikation von zirkulärer DNA. Zudem ist diese Amplifikation so wirkungsvoll wie PCR. Daher werden für die Amplifikation von Papilloma genomischer DNA nur kleinste Mengen von grob isolierter DNA benötigt. Das Rolling-circle Amplifikationsverfahren kann angewendet werden um zu bestimmen, ob gesunde Haut Papillomaviren in sich trägt und um Papillomaviren zu sequenzieren, die in Hautläsionen von Hunden und Katzen vorhanden sind. Diese beschreibende Studie ist der unumgängliche erste Schritt für ein besseres Verständnis der Rolle, die Papillomaviren in der Entwicklung von Hautläsionen bei Hunden und Katzen spielen, besonders in der Entwicklung von Hautkrebsen. Wir haben diese neue Technik bereits bei der Isolation und beim Klonen eines neuen caninen Papillomavirus angewandt (CPV3) [20].

Bei Säugetieren und Vögeln führen Papillomaviren ein grosses Spektrum an kutanen Veränderungen und Veränderungen der Schleimhaut herbei, wie zum Beispiel exophytische und Flachwarzen und präkanzeröse und kanzeröse Läsionen. Sie werden für wichtige Karzinogene bei Menschen gehalten und einige risikoreiche Papillomaviren sind direkt für die Entwicklung von Gebärmutterhalskrebs bei Frauen verantwortlich [21-23]. Obwohl die Verbindung zwischen zervischen Karzinomen und humanen Papillomaviren klar ist, ist die Rolle von Papillomaviren in der Entwicklung von kutanem Plattenepithel- Karzinom (SCC) nicht so klar [24]. Es gibt jedoch vermehrte epidemiologische Evidenzen, die nahelegen, das Papillomaviren eine wichtige Rolle bei der Hautkanzerogenese spielen, besonders bei der Epidermodysplasia verruciformis (EV). Es ist dennoch problematisch, eine Kausalität zwischen dem Vorhandensein von Papillomaviren in Hautläsionen und der Entwicklung der Läsionen herzustellen [25]. Es

wurden Kriterien vorgeschlagen, um diese Beziehung zu bestätigen, Schwierigkeiten bei der Kultivierung von Papillomaviren haben deren Erfüllung aber grösstenteils unmöglich gemacht [22, 25]. Zusätzlich erlaubt der Gebrauch von extrem sensitiver Nested-PCR den Nachweis einer winzigen Menge von viraler DNA (0.05 virale Genome per Zelle). Die Anwesenheit von kleinen Mengen von Nukleinsäure des Papillomavirus induziert keine produktive Infektion und kann auch in gesunder Haut vorkommen [26, 27]. Vieles deutet auch darauf hin, dass ultraviolette Strahlung (UV) zur Kanzerisation von einigen Papillomavirus assoziierten Hautkrebsen beiträgt [25, 28]. Alles in allen bleibt angesichts all dieser Ergebnisse die Rolle von Papillomaviren bei der Entwicklung von Hautkrebs beim Menschen unklar. Einige Tiermodelle stützen die kausative Rolle von Papillomaviren bei der Induktion des Squamosazell-Karzinoms (SCC): Vor einigen Jahrzehnten wurde gezeigt, dass Cottontail rabbit Papillomaviren (CRPV) in der Lage sind, Hautkrebs bei Hasen zu induzieren [29- 31].

Andere Studien haben auch festgestellt, dass attenuierter Lebend-Impfstoff für Canines orales Papillomavirus (COPV) bei Beagle SCC induziert [32]. Hinzu kommt, dass Hunde und Katzen von Hautkonditionen betroffen sein können, die einiges gemein haben mit EV von Menschen und bei einigen Patienten wurde von einer Kanzerisation berichtet [33]. Einige Studien haben eine Verbindung zwischen SCC und Papillomaviren bei Karnivoren gezeigt, es wurde aber keine Kausalität hergestellt [16, 34-40]. Diese Arbeit hat die epidemiologische Verbindung zwischen einigen Gruppen von felinem und caninem Hautkrebs und Papillomavirus-Infektion sowie die genetische Diversität von karnivoren Papillomaviren bestätigt [16, 20, 34, 40]. Nebst der Identifikation und dem Klonen dieser neuen Papillomaviren sind die wichtigsten Untersuchungen diejeinigen, die die relative Prävalenz von jedem neuen karnivoren Papillomavirus feststellen und in vitro zeigen, dass zumindest einige davon in der Lage sind, die Transformation und Immortalisation von Keratinozyten zu induzieren. Wir haben zum Beispiel ein neuartiges Papillomavirus (CPV3) entdeckt, geklont und sequenziert, das mit einem Fall von Caniner Epidermodysplasia verruciformis assoziiert war [20]. Der befallene Hund entwickelte multiple Plaques und eine einzige interdigitale Läsion von in-situ SCC [20]. In jeder untersuchten Läsion (einschliesslich SCC) wurde

DNA von CPV3 isoliert, diese war aber nicht vorhanden in der intakten Haut desselben Hundes. Es wurde Sequenz-unabhängige, „multiple primed“ RCA verwendet, um das ganze CPV3 Genom zu amplifizieren, klonen und sequenzieren. Tatsächlich liess die Analyse des geklonten und sequenzierten caninen Papilloma-Genoms seine Klassifikation als Mitglied einer neuen Gattung von Papillomaviren (GenBank-Eintrag DQ295066) zu. Zusätzlich wurde in jeder untersuchten Läsion mRNA für das mutmassliche transformierende Protein E6 gefunden. Diese Resultate zeigten, dass CPV3 in den erwähnten Hautläsionen transkriptional aktiv war und stützten die Hypothese einer kausativen Rolle von CPV3 in der Pathogenese von Caniner EV [41]. Komplementäre DNA des p53 Transkriptes des betroffen Hundes wurde sowohl von intakter Haut als auch von Hautläsionen geklont, und seine Nukleotid-Sequenz wich nicht von der Wildtyp-caninen p53-Sequenz ab. Dieses Ergebnis deutete darauf hin, dass die Entwicklung eines bösartigen Tumors nicht auf UV-abhängige Mutagenese zurückzuführen war. Folglich wurde die Hypothese gestützt, dass transformierende Proteine von CPV3 die Entwicklung von Krebs mit Mechanismen induzieren, die anders sind als die Mechanismen, die bei humanen EV-assoziierten Papillomaviren angestrengt werden [41]. Wir untersuchen gegenwärtig auch die relative Prävalenz von CPV3 in caninen Seren. Wir haben die CP3-L1 Gene geklont (Kodes für CPV3 Haupt-Kapsidprotein) und Antiseren gegen dieses Protein generiert. Unser Ziel ist es, einen ELISA Test zu etablieren und 500 vorgängig gesammelte canine Seren zu evaluieren. Schliesslich öffnen die Resultate dieser Arbeit and hoffentlich zukünftiger Studien neue Zugänge zur Erforschung der Papillomavirus-induzierten Hautkanzerogenese. Das sollte sich sowohl für die veterinär- als auch die human-medizinische Onkologie als hilfreich erweisen, wenn man bedenkt, dass es mehrere ähnliche Hautkonditionen bei Karnivoren und Menschen gibt.

4. Huff, J.C., W.L. Weston, and M.G. Tonnesen, Erythema multiforme: a critical review of characteristics, diagnostic criteria, and causes. J Am Acad Dermatol, 1983. 8(6): p. 763- 775. 5. Gross, T.L., et al., Giant cell dermatosis in FeLV-positive cats. Vet. Dermatol., 1993. 4(3): p. 117-122. 6. Kimura, S.K. and H. Hatano, Multinucleate epidermal cells in non-neoplastic dermatoses. Brit J Dermatol, 1978. 99: p. 485-489. 7. Fenyo, E.M., et al., Distinct replicative and cytopathic characteristics of human immunodeficiency virus isolates. J Virol, 1988. 62(11): p. 4414-4419. 8. White, J.M., Membrane fusion. Science, 1992. 258: p. 917-924. 9. Rohn, J.L., et al., In vivo evolution of a novel, syncytium-inducing feline leukemia virus variant. Journal of Virology, 1998. 72(4): p. 2686-2696. 10. Vonderhaar, M.A. and W.B. Morisson, Chapter 45. Lymphosarcoma, in Cancer in dogs and cats, W.B. Morisson, Editor. 2002, Teton New Media: Jackson, Wyoming. p. 641- 670. 11. Tobey, J.C., et al., Cutaneous T-cell lymphoma in a cat. J Am Vet Med Assoc, 1994. 204(4): p. 606-609. 12. Doorbar, J., The papillomavirus life cycle. J Clin Virol, 2005. 32(Supplement 1): p. 7-12. 13. de Villiers, E.-M., et al., Classification of papillomaviruses. Virolo, 2004. 324: p. 17-27. 14. Scott, D.W., W.H. Miller, and C.E. Griffin, Neoplastic and non-neoplastic tumors, in Kirk and Muller's Small Animal Dermatology, D.W. Scott, W.H. Miller, and C.E. Griffin, Editors. 2001, W.B. Saunders Co.: Philadelphia. p. 1236-1413. 15. Favrot, C., et al., Evaluation of papillomaviruses associated with cyclosporine-induced hyperplastic verrucous lesions in dogs. Am J Vet Res, 2005. 66(10): p. 1764-1769. 16. Zaugg, N., et al., Detection of novel papillomaviruses in canine mucosal, cutaneous and in situ squamous cell carcinomas. Vet Dermatol, 2005. 16(5): p. 290-298. 17. Iftner, A., et al., The prevalence of human papillomavirus genotypes in nonmelanoma skin cancers of nonimmunosuppressed individuals identifies high-risk genital types as possible risk factors. Cancer Res, 2003. 63(21): p. 7515-7519. 18. Rector, A., et al., Isolation and cloning of a papillomavirus from a North American porcupine by using multiply primed rolling-circle amplification: the Erethizon dorsatum papillomavirus type 1. Virol, 2005. 331(2): p. 449-456. 19. Rector, A., R. Tachezy, and M.A. Van Ranst, Sequence independant strategy for detection and cloning of circular DNA virus genome by using multiply primed rolling circle amplification. J. Virol., 2004. 78(10): p. 1993-1998. 20. Tobler, K.et al., Detection of the prototype of a potential novel genus among the papillomaviridae in association with canine epidermodysplasia verruciformis. J Gen Virol, 2006. 67(12): p. 2036-2041 21. Saladi, R.N. and A.N. Persaud, The causes of skin cancer: a comprehensive review. Drugs Today (Barc), 2005. 41(1): p. 37-53. 22. zur Hausen, H., Papillomavirus infections--a major cause of human cancers. Biochim Biophys Acta, 1996. 1288(2): p. F55-78. 23. zur Hausen, H., Oncogenic DNA viruses. Oncogene, 2001. 20(54): S. 7820-7823. 24. Sterling, J.C., Human papillomaviruses and skin cancer. J ClinVirol, 2005. 32(Supplement 1): p. 67-72. 25. Harwood, C.A. and C.M. Proby, Human papillomaviruses and non-melanoma skin cancer. Curr Opin Infect Dis, 2002. 15(2): p. 101-114. 26. Antonsson, A. and B.G. Hansson, Healthy skin of many animal species harbours papillomaviruses which are closely related to their human counterparts. J Virol, 2002. 76(24): p. 12537-12542.

27. Majewski, S. and S. Jablonska, Human papillomavirus and oncogenesis: critical evaluation of recent findings. Int J Dermatol, 2002. 41(6): p. 319-320. 28. Termorshuizen, F., et al., Sunlight Exposure and (Sero) Prevalence of Epidermodysplasia Verruciformis-Associated Human Papillomavirus. J Invest Dermatol, 2004. 122(6): p. 1456-1462. 29. Breitburd, F., J. Salmon, and G. Orth, The rabbit viral skin papillomas and carcinomas: a model for the immunogenetics of HPV-associated carcinogenesis. Clin Dermatol, 1997. 15(2): p. 237-247. 30. Rous, P. and J.W. Beard, The progression to carcinoma of virus-induced rabbit papillomas (Shope). J Experim Med, 1935(62): p. 523-548. 31. Brandsma, J.L., The cottontail rabbit papillomavirus model of high-risk HPV-induced disease. Methods Mol Med, 2005. 119: p. 217-235. 32. Bregman, C.L., et al., Cutaneous neoplasms in dogs associated with canine oral papillomavirus vaccine. Vet Pathol, 1987. 24(6): p. 477-87. 33. Gross, T.L., et al., Epidermal tumors, in Skin diseases of the dog and cat: Clinical and histopathological diagnosis, T.L. Gross, et al., Editors. 2005, Blackwell Science: Oxford. p. 562-577. 34. Nespeca, G., et al. Detection of novel papillomavirus-like sequences in paraffin- embeddedspecimens oe invasive and in situ squamous cell carcinomafrom cats. Am J Vet Res, 2006. 26(12): p. 2036-2041 35. LeClerc, S.M., E.G. Clark and D.M Haines, Papillomavirus infection in association with feline cutaneous squamous cell carcinoma in situ (Abstract). in AAVD/ACVD Meeting. 1997.p. 125-126. 36. Schwegler, K., J.H. Walter, and R. Rudolph, Epithelial neoplasms of the skin, the cutaneous mucosa and the transitional epithelium in dogs: an immunolocalization study for papillomavirus antigen. Zentralbl Veterinarmed A, 1997. 44(2): p. 115-123. 37. Sundberg, J.P., R.E. Junge, and W.D. Lancester, Immunoperoxidase localization of papillomaviruses in hyperplastic and neoplastic epithelial lesions in animals. American JtVet Rest, 1984. 45(7): p. 1441-1446. 38. Teifke, J.P., et al., Detection of papillomavirus-DNA in mesenchymal tumour cells and not in the hyperplastic epithelium of feline sarcoids. Vet Dermatol, 2003. 14(1): p. 47-56. 39. Teifke, J.P., C.V. Lohr, and H. Shirasawa, Detection of canine oral papillomavirus-DNA in canine oral squamous cell carcinomas and p53 overexpressing skin papillomas of the dog using the polymerase chain reaction and non-radioactive in situ hybridization. Vet Microbiol, 1998. 60(2-4): p. 119-130. 40. Wilhelm, S. et al.Clinical, histological and immunohistochemical study of feline viral plaques and bowenoid in situ carcinomas. Vet Dermatol, 2006. 17: p. 424-431 41. Erne, M.L. Further Characterization of a novel canine papillomavirus (CPV3) and its potential role in the context of Epidermodysplasia Verruciformis. Dissertation. Zürich 2006.