Vet Dermatol 2015; 26: 57–e20
DOI: 10.1111/vde.12189
Antibody titres against canine papillomavirus 1 peak around clinical regression in naturally occurring oral papillomatosis € ller§ and Christian E. Lange†‡¶ Arda Sancak*, Claude Favrot†, Marco D. Geisseler†‡, Martin Mu *Division of Clinical Sciences, Faculty of Veterinary Medicine, Ankara University, Irfan Bastug Cadde, Ankara, 06110, Turkey †Dermatology Department, Clinic for Small Animal Internal Medicine, Vetsuisse Faculty, University of Zurich, Winterthurerstrasse 260, Zurich, ZH 8057, Switzerland ‡Institute of Virology, Vetsuisse Faculty, University of Zurich, Winterthurerstrasse 266a, Zurich, ZH 8057, Switzerland §Deutsches Krebsforschungszentrum, Im Neuenheimer Feld 242, Heidelberg, BW 69120, Germany ¶Microbiology and Immunobiology Department, Harvard Medical School, 77 Avenue Louis Pasteur, Boston, MA 20115, USA Correspondence: Christian E. Lange, Microbiology and Immunobiology Department, Harvard Medical School, 77 Avenue Louis Pasteur, Boston, MA 20115, USA. E-mail: [email protected]
Background – Most forms of canine papillomatosis are believed to be associated with papillomavirus infections. Canine papillomavirus type 1 (CPV1) is considered to be responsible for most oral cases and several forms of cutaneous papillomatosis. Hypothesis/Objectives – The aim of this study was to evaluate cases of naturally occurring oral papillomatosis with regard to the type of virus involved, antibody induction and remission time. Methods – Forty dogs showing different degrees of classical oral papillomatosis were included as a single study group. Tissue and serum samples were acquired upon initial presentation; serum samples were collected again upon remission (n = 13) and after 3 months of convalescence (n = 4). None of the dogs underwent antiviral therapy. Tissue samples were tested by PCR to detect CPV DNA, while serum samples were tested using a specific enzyme-linked immunosorbent assay for antibodies against the L1 capsid protein of CPV1. Results – All tissue samples were positive for CPV1 DNA, and 87.5% of all serum samples contained measurable levels of antibody against the virus (cut-off value 0.3). The average optical density measured in the enzyme-linked immunosorbent assay was 0.51 at initial presentation, 1.65 upon remission and 0.83 at 3 months postrecovery. Time to clinical regression varied between 1 month and 1 year. Conclusions and clinical importance – These data support existing evidence for a high prevalence of CPV1 in canine oral papillomatosis. The healing process seems to correlate with a strong antibody response, and antibody titres peaked around the time of clinical recovery. In contrast to previous data from laboratory settings, the variation in remission time was very high.
Introduction Most forms of canine papillomatosis are thought to be induced by papillomavirus infections. The best-known and probably most prevalent form is canine oral papillomatosis, which is clinically characterized by the presence of exophytic papillomas within the oral cavity or at mucocutaneous junctions. These papillomas are characteristic in their cauliflower-like appearance and easily diagnosed clinically. Clinical disease is typically transient and primarily seen in young dogs. The infectious agent is canine papillomavirus type 1 (CPV1), a Lambda papillomavirus, which can also be involved in cutaneous papillomatosis. Cutaneous manifestations include exophytic and endophytic papillomas, pigmented plaques and, in rare cases, squamous cell carcinomas.1
Accepted 13 October 2014 Sources of Funding: This study was self-funded. Conflict of Interest: No conflicts of interest have been declared. © 2014 ESVD and ACVD, Veterinary Dermatology, 26, 57–e20.
To date, 14 distinct CPV types have been reported.1–3 Of these, types CPV1 and CPV13 have been isolated from oral papillomas.1,2 Despite the apparent association with certain disorders, papillomavirus DNA can also be found in the absence of signs, which could indicate early or subclinical infections or simply carriage.4 Immunocompetent individuals are usually able to clear or control most papillomavirus infections, with cellular immunity being the primary factor.5 Antibodies protect from re-infection with the same virus type, and thus generic or recombinant vaccines can be used prophylactically.5–8 Serum antibody titres can also help detect past infections using specific enzyme-linked immunosorbant assays (ELISAs). Prevalence rates of 10.5 and 21.9% have been reported for CPV1 antibodies in asymptomatic dogs.9 Despite data showing antibody induction in controlled laboratory conditions, antibody responses to natural CPV infections, induced by supposedly lower viral loads, have not been studied in dogs.7 The study reported here was initiated to characterize the antibody responses that natural CPV1 infections induce and to monitor antibody titres prior to and beyond clinical regression using a 57
Sancak et al.
CPV1-specific ELISA. Such information may promote better understanding of the biology of this common disease, especially in the context of prognosis and potential vaccination regimens.
Materials and methods Forty dogs with clinical oral papillomatosis were recruited at the veterinary teaching hospital in Ankara, Turkey. The Ethics Committee of the University of Ankara approved all procedures (permission 2011107). Initially, a biopsy of the oral lesions and a serum sample were collected from all dogs and stored at 20°C. None of the dogs underwent antiviral or other therapies. Owners were asked to present the dogs for follow-up physical examinations and sampling upon and after clinical recovery. Clinicians verified the regression. Follow-up serum samples were collected at the time of remission from 13 dogs, at 3 months postremission in four dogs and at 6 months postremission from one dog. DNA was extracted from biopsy samples using a DNeasy extraction kit (Qiagen, Hombrechtikon, Switzerland). To verify successful extraction, a PCR amplifying canine glyceraldehyde 3-phosphate dehydrogenase (GAPDH) DNA was performed, while canPV/FAP64 and PPF1/CP5 primers were applied to screen for CPV DNA. Reported PCR conditions were used.4 Diluted CPV1 DNA served as a positive control, elution buffer from the DNeasy extraction kit as a negative control, and all canPV/FAP64 PCR products were determined (Microsynth, Balgach, Switzerland) by cycle sequencing using an ABI 377 sequencer (Applied Biosystems, Carlsbad, CA, USA) and compared with the NCBI database (BLAST X, http://blast.ncbi.nlm.nih.gov/Blast.cgi) to identify the virus type. To screen for serum antibodies, the L1 protein of CPV1 was expressed in Escherichia coli and coated onto ELISA plates as described previously.9 All dog sera were tested in triplicate diluted 1:500, avoiding more than one peripheral well per triplicate. Sera previously tested positive or negative served as controls.9 For the analysis, the mean of the two optical densities (ODs) closest to each other was taken, while the third OD of the triplicate was disregarded. To set a cut-off value, positive and negative control ODs were compared; the cut-off value was set as the OD resulting in an equal distribution of false positives and false negatives among the controls.
Results Of the 40 dogs sampled, 23 were male and 17 female, with an age range of 3 months to 13 years (average 2 years). Twelve golden retrievers, 14 individuals of other pure breeds and 14 mixed-breed dogs were included. Besides oral papillomatosis, two dogs also had periocular papillomas, while two more had other cutaneous sites affected. All dogs for which follow-up information was available achieved remission without specific treatment. The PCR for GAPDH amplified host DNA in 39 samples. All 40 samples were positive in the broad range PCR for CPV (canPV/FAP64) and 36 also in the narrow-range PCR for DNA of Lambda papillomaviruses (PPF1/CP5). The sequencing of the broad-range PCR products and comparison with the databases confirmed exact sequence identity with the CPV1 DNA reference in all cases. The cut-off value for the ELISA was set at an OD of 0.309 (data not shown). Of the 58 serum samples collected, 53 were classified as positive. Of the five negative samples, four were collected at initial presentation and one upon recovery. Two of the dogs that were seronegative at baseline had seroconverted at the time of resampling. No additional samples were available in the other two cases. Overall, the ODs ranged from 0.25 to 3.97 (mean 1.18). Among the 13 dogs where at least two 58
Figure 1. Individual antibody titres are represented as optical density (OD) at three time points, i.e. >1 month prior to resolution, around the time of resolution and >1 month postrecovery. Bars indicate the mean value and SD.
samples were available, antibody titres increased in 10 cases, whereas they decreased in three. The increase between initial presentation and recovery in this group overall was significant (Wilcoxon matched pairs signedranks test, P = 0.047) or highly significant (Mann–Whitney U-test; P = 0.003), respectively. The mean OD of serum samples collected more than 1 month prior to recovery (n = 9) was 0.51 (SD 0.26), whereas the mean OD for samples collected at the time of clinical remission (1 month; n = 18) was 1.65 (SD 1.2). The latter samples included seven dogs where both initial presentation and recovery fell within a 1 month time frame. At 3 months postrecovery (n = 4), the mean OD was 0.83 (SD 0.46; Figure 1). For 24 dogs, the initial presentation could not be linked to any recovery time due to lack of appropriate follow-up information. Eleven of the cases with follow-up (n = 16) went into complete regression within 3 months or less. In three dogs, the number of papillomas increased significantly before all regressed. Perhaps due to the limited number of dogs with complete follow-up, no statistical correlation between severity of clinical signs and antibody titres was found.
Discussion It is widely accepted that canine oral papillomatosis is induced by a papillomavirus infection, and the identification of CPV1 in all samples collected within this study further strengthens that association. There was no evidence of other CPV sequences in the electropherograms, which were obtained using primers with the potential to amplify most known CPVs. Although CPV DNA may also be found in samples of clinically healthy skin, the high prevalence (100%) and the large amount of amplified DNA in this study (data not shown) indicate an active viral infection.4 The induction of protective antibodies against CPV1 is likely to explain why oral papillomatosis is typically seen in young dogs with virus-na€ıve immune systems.7 The ODs found here at the time of initial presentation were © 2014 ESVD and ACVD, Veterinary Dermatology, 26, 57–e20.
Antibody titres to CPV1
variable. If results for only those dogs from which multiple samples were available are examined, a clear trend for an increase and peak in titres around the time of clinical regression is apparent (Figure 1). A similar peak occurred around the time of clinical regression (between 11 and 14 weeks) in dogs with experimentally induced CPV1 infection.7 The clinical nature of our study prohibited accurate determination of the time between exposure and appearance of clinical papillomatosis. This may explain the large variability of antibody titres among the serum samples at the time of initial presentation and why four sera taken at this point were deemed negative despite overt clinical disease. Alternatively, this variability may be a reflection of individual differences among the heterogeneous sample cohort. Lesion regression occurred within 3 months after initial presentation in 11 of the dogs, while the time between presentation and regression was >3 months in five dogs; therefore, the time to remission appears to be much more variable than in experimental conditions.7 As a consequence of this enormous variation, clinical studies on treatment regimens for oral papillomatosis should be interpreted with caution. Treatment effects may be either overshadowed or overestimated if test groups are small. We consider the observed remission of papillomas to be spontaneous healing because no specific therapy was initiated. It is possible that the biopsy procedure triggered an active immune response, because partial crushing or removal of oral papillomas has been anecdotally suggested to hasten resolution; however, the contrary effect has also been proposed.10 The follow-up information available is statistically not conclusive regarding this matter. However, the extended time required for spontaneous remission in five dogs and the increased numbers of papillomas in three dogs prior to recovery do not seem to support a beneficial effect. In conclusion, results of this study show that the healing process in naturally occurring infections is more variable than indicated by laboratory experiments. The data indicate, nevertheless, that clinical resolution generally correlates with a strong antibody response and
that antibody titres peak around the time of clinical remission.
Acknowledgements The authors thank Kurt Tobler and Mathias Ackermann of the Institute of Virology, Vetsuisse Faculty, University of Zurich, as well as Peter M. Howley of the Microbiology and Immunobiology Department at Harvard Medical School and the clinicians of the veterinary teaching hospital of the Ankara University for logistical and general support.
References 1. Lange CE, Favrot C. Canine Papillomaviruses. Vet Clin North Am Small Anim Pract 2011; 41: 1183–1195. 2. Lange CE, Ackermann M, Favrot C et al. Entire genomic sequence of novel canine papillomavirus type 13. J Virol 2012; 86: 10226–10227. 3. Lange CE, Tobler K, Schraner EM et al. Complete canine papillomavirus life cycle in pigmented lesions. Vet Microbiol 2013; 162: 388–395. 4. Lange CE, Zollinger S, Tobler K et al. The clinically healthy skin of dogs is a potential reservoir for canine papillomaviruses. J Clin Microbiol 2011; 49: 707–709. 5. Howley PM, Schiller JC, Lowy DR. Papillomaviruses. In: Knipe DM, Howley PM, eds. Fields’ Virology 6th edition. Philadelphia, PA: Lippincott, Williams & Wilkins, 2013; 1662–1699. 6. Bell JA, Sunberg JP, Ghim S et al. A formalin-inactivated vaccine protects against mucosal papillomavirus infection: a canine model. Pathobiology 1994; 62: 194–198. 7. Suzich JA, Ghim SJ, Palmer-Hill FJ et al. Systemic immunization with papillomavirus L1 protein completely prevents the development of viral mucosal papillomas. Proc Natl Acad Sci USA 1995; 92: 11553–11557. 8. Ghim SJ, Newsome J, Bell JP et al. Spontaneously regressing oral papillomas induce systemic antibodies that neutralize canine oral papillomavirus. Exp Mol Pathol 2000; 68: 147–151. 9. Lange CE, Tobler K, Favrot C et al. Detection of antibodies against epidermoplasia verruciformis-associated canine papillomavirus 3 in sera of dogs from Europe and Africa by enzyme-linked immunosorbent assay. Clin Vaccine Immunol 2009; 16: 66–72. 10. Collier LL, Collins BK. Excision and cryosurgical ablation of severe periocular papillomatosis in a dog. J Am Vet Med Assoc 1994; 204: 881–885.
sume Re es e ^tre associe es Contexte – La plupart des formes de papillomatoses canines sont cense a des infections a papillomavirus. Le papillomavirus canin de type 1 (CPV1) est conside re comme responsable de la plupart es. des cas oraux et de plusieurs formes de papillomatoses cutane ses/Objectifs – Le but de cette e tude e tait d’e valuer le type de virus implique , l’induction d’antiHypothe e de re mission les cas d’origine naturelle de papillomatose orale. corps et la dure thodes – Quarante chiens montrant diffe rents degre s de papillomatose orale classique ont e te inclus Me tude. Des e chantillons de tissus et de serum ont e te pre leve s te dans l’e a la visite initiale; les sera ont e leve s de nouveau au moment de la re mission (n = 13) et apre s 3 mois de convalescence (n = 4). Aucun pre rapie antivirale. Les e chantillons de tissu ont e te teste s par PCR pour de tecter des chiens n’a recßu de the chantillons de se rum ont e te teste s par ELISA pour les anticorps antil’ADN de CPV, alors que les e ique de capside L1 du CPV1. prote sultats – Tous les e chantillons tissulaires e taient positifs pour l’ADN de CPV1 et 87.5% de tous les sera Re s contre le virus (valeur seuil 0.3). La densite optique contenaient des taux mesurables d’anticorps dirige e par ELISA e tait de 0.51 re visite, 1.65 mission et 0.83 apre s 3 mois de moyenne mesure a la premie a la re mission. La dure e de re gression clinique variait de 1 mois re a 1 an.
© 2014 ESVD and ACVD, Veterinary Dermatology, 26, 57–e20.
59
Sancak et al.
es supportent les preuves existantes d’une haute Conclusions et importance clinique – Ces donne valence de CPV1 dans les papillomatoses orales canines. Le processus de gue rison semble corre ler pre ponse d’anticorps avec des titres d’anticorps les plus e leve s au moment de la gue rison cliavec une forte re rence des donne es ante rieures de laboratoire, la variation de la dure e de re mission e tait tre s nique. A la diffe elevee. Resumen n – la mayorıa de las formas de papilomatosis canina parecen estar asociadas con infeccio n Introduccio por papilomavirus. El papilomavirus canino de tipo I (CPV1) se considera responsable de la mayorıa de los casos orales y de varias formas de papilomatosis cut aneo. tesis/Objetivos – el propo sito de este estudio fue evaluar casos de papilomatosis oral espont Hipo anea en n al tipo de virus implicado, produccio n de anticuerpos, y tiempo de remisio n. relacio todos – 40 perros que mostraron diferentes grados de papilomatosis oral cl Me asica se incluyeron en un n inicial; y muestras de susolo grupo de estudio. Muestras de suero y de tejido se tomaron a la presentacio n (n = 13) y tras tres meses convalecencia (n = cuatro). Ninguno ero se recogieron de nuevo tras la remisio de los perros fue sometido a terapia antivırica. Las muestras de tejido fueron evaluadas mediante PCR para detectar el DNA de papilomavirus canino, mientras que las muestras de suero fueron probadas usando una n ligada a enzimas para anticuerpos frente a la proteına de la c prueba de inmunoabsorcio apsula L1 del papilomavirus canino de tipo I. Resultados – todas las muestras de tejido fueron positivas para DNA de papilomavirus canino de tipo I, y 87,5% de las muestras de suero contenıan niveles valorables de anticuerpos frente al virus (valor de corte ptica medida en el ensayo de inmunoabsorcio n fue de 0,51 a la presentacio n ini0,3). La media densidad o n, y 0,83 a los tres meses tras la recuperacio n. El tiempo para la regresio n clınica cial, 1,65 tras la remisio entre un mes y un an ~o. vario Conclusiones e importancia clınica – estos datos aportan nueva evidencia de una gran prevalencia del n parece correlacionpapilomavirus canino tipo I en casos de papilomatosis oral. El proceso de recuperacio arse con una fuerte respuesta de anticuerpos, y los tıtulos de anticuerpos fueron maximos alrededor del tin clınica. En contraste con datos previos en pruebas de laboratorio la variacio n en el empo de la recuperacio n fue bastante amplia. tiempo de remisio Zusammenfassung Hintergrund – Man geht davon aus, dass die meisten Formen der Papillomatose des Hundes durch Papil€r die lomavirus verursacht werden. Es wird angenommen, dass das canine Papillomavirus Typ 1 (CPV1) fu €r mehrere Formen der Papillomatose der Haut verantwortlich meisten F€alle oraler Papillomatose sowie fu ist. €rlich auftretender oraler PaHypothese/Ziele – Das Ziel dieser Studie war eine Evaluierung von F€ allen natu €rperinduktion und Remissionszeit. pillomatose in Bezug auf den involvierten Virustyp, die Antiko Methoden – Vierzig Hunde, die unterschiedliche Auspr€ agungen der klassischen oralen Papillomatose zeigten, wurden als eine Untersuchungsgruppe verwendet. Bei der Erstvorstellung wurden Gewebe- und Blutproben entnommen; zum Zeitpunkt der Remission (n=13) und 3 Monate nach der Heilung (n= 4) wurden €hrt. Die wieder Serumproben genommen. Bei keinem der Hunde wurde eine antivirale Therapie durchgefu Gewebeproben wurden mittels PCR auf CPV DNA getestet, w€ ahrend die Serumproben mittels spezifi€rper gegen L1 Kapsidprotein von CPV1 getestet wurscher Enzym-linked Immunosorbent Assay auf Antiko den. €r CPV1 DNA, und 87,5% aller Serumproben enthielten Ergebnisse – Alle Gewebeproben waren positiv fu €rpern gegen das Virus (Cut-off Wert 0,3). Die durchschnittliche optische Dichte, messbare Werte an Antiko die mittels Enzym-linked Immunosorbent Assay gemessen wurde, betrug 0,51 bei der Erstpr€ asentation, 1,65 bei Remission und 0,83 nach 3 Monaten der Heilung. Die Zeit bis zur klinischen Regression variierte zwischen 1 Monat und 1 Jahr. €tzen die Evidenz einer hohen Schlussfolgerungen und klinische Bedeutung – Diese Daten unterstu Pr€avalenz von CPV1 bei der oralen Papillomatose des Hundes. Der Heilungsprozess scheint mit einer star€rperantwort zu korrelieren, die Antiko €rpertiter waren zum Zeitpunkt der Heilung am ho €chsten. ken Antiko €heren Daten aus Laborsettings war die Variation der Dauer bis zur Remission sehr Im Gegensatz zu fru hoch.
要約 背景 – イヌの乳頭腫症のほとんどの型はパピローマウイルス感染と関連していると考えられている。イ ヌパピローマウイルス1型(CPV1)はほとんどの口腔の症例、および皮膚の乳頭腫症のいくつかの型の原因 となると考えられている。 e19
© 2014 ESVD and ACVD, Veterinary Dermatology, 26, 57–e20.
Antibody titres to CPV1
仮説/目的この研究の目的は関与するウイルス型、抗体誘導ならびに寛解時間に関して自然発生性口腔乳 頭腫症の症例を評価すること。 方法 – 異なった程度の古典的な口腔乳頭腫症を示す40頭のイヌを単一の研究グループに含めた。組織お よび血清サンプルは初診時に得た、血清サンプルは寛解時(n = 13)と回復期の3ヶ月後 (n = 4)に再度回収 した。抗ウイルス療法はどのイヌにも行わなかった。組織サンプルはCPV DNAを検出するPCRで検査し、血 清サンプルはCPV1のL1カプシドタンパクに対する抗体を、特異的酵素結合免疫吸着検査法を用いて検査 した。 結果 – すべての組織サンプルはCPV1 DNAに対して陽性で、すべての血清サンプルの87.5%がウイルスに対 する抗体を測定可能なレベルで含んでいた(カットオフ値0.3)。酵素結合免疫吸着検査法で測定した平均 可視化濃度は初診時に0.51、寛解時に1.65、および回復後3ヶ月で0.83であった。臨床的な退縮までの時 間は1ヶ月から1年まで様々であった。 結論および臨床的な重要性 – これらのデータはイヌの口腔乳頭腫症においてCPV1の高い保有率の既存の 証拠を支持する。治癒プロセスは強力な抗体反応と関連しているようであり、抗体価のピークは臨床的 な回復期付近のようである。研究室で行われた以前のデータと比較し、寛解時間のバリエーションは非 常に多かった。 摘要 背景——大部分犬乳头状瘤被认为与乳头状瘤病毒感染有关。犬乳头状瘤病毒1型(CPV1)是引起大部分口腔 和几种皮肤乳头状瘤的病因。 假设/目的——本次研究的目的是评估自然发生的口腔乳头状瘤,其感染病毒的类型、抗体感应和缓解时 间。 方法——四十只不同程度的典型口腔乳头状瘤患犬,分为一个单独研究组。发病初期取组织和血清样本;恢 复期(n=13)和3个月康复期后(n=4)再取血清样本。所有犬未经过抗病毒治疗。组织样本通过PCR实验检测 CPV DNA,血清样本则用特异性酶链免疫吸附实验,检测抗CPV1的L1衣壳蛋白抗体。 结果——所有组织样本为CPV1 NDA阳性,87.5%的血清样本包含达到可测量水平的病毒抗体(截止值0.3)。酶 链免疫吸附实验中测量的平均光度值,发病初期为0.51,恢复期为1.65,3个月康复期后为0.83。临床恢复时 间为1个月到1年不等。 总结与临床意义——这些数据显示,CPV1在犬口腔乳头状瘤中有很高的流行率。恢复过程似乎与强烈的抗体 反应有关,抗体滴度在临床症状恢复期到达峰值。对比之前的实验室数据,恢复时间差异很大。
© 2014 ESVD and ACVD, Veterinary Dermatology, 26, 57–e20.
e20
DOI: 10.1111/vde.12189
Antibody titres against canine papillomavirus 1 peak around clinical regression in naturally occurring oral papillomatosis € ller§ and Christian E. Lange†‡¶ Arda Sancak*, Claude Favrot†, Marco D. Geisseler†‡, Martin Mu *Division of Clinical Sciences, Faculty of Veterinary Medicine, Ankara University, Irfan Bastug Cadde, Ankara, 06110, Turkey †Dermatology Department, Clinic for Small Animal Internal Medicine, Vetsuisse Faculty, University of Zurich, Winterthurerstrasse 260, Zurich, ZH 8057, Switzerland ‡Institute of Virology, Vetsuisse Faculty, University of Zurich, Winterthurerstrasse 266a, Zurich, ZH 8057, Switzerland §Deutsches Krebsforschungszentrum, Im Neuenheimer Feld 242, Heidelberg, BW 69120, Germany ¶Microbiology and Immunobiology Department, Harvard Medical School, 77 Avenue Louis Pasteur, Boston, MA 20115, USA Correspondence: Christian E. Lange, Microbiology and Immunobiology Department, Harvard Medical School, 77 Avenue Louis Pasteur, Boston, MA 20115, USA. E-mail: [email protected]
Background – Most forms of canine papillomatosis are believed to be associated with papillomavirus infections. Canine papillomavirus type 1 (CPV1) is considered to be responsible for most oral cases and several forms of cutaneous papillomatosis. Hypothesis/Objectives – The aim of this study was to evaluate cases of naturally occurring oral papillomatosis with regard to the type of virus involved, antibody induction and remission time. Methods – Forty dogs showing different degrees of classical oral papillomatosis were included as a single study group. Tissue and serum samples were acquired upon initial presentation; serum samples were collected again upon remission (n = 13) and after 3 months of convalescence (n = 4). None of the dogs underwent antiviral therapy. Tissue samples were tested by PCR to detect CPV DNA, while serum samples were tested using a specific enzyme-linked immunosorbent assay for antibodies against the L1 capsid protein of CPV1. Results – All tissue samples were positive for CPV1 DNA, and 87.5% of all serum samples contained measurable levels of antibody against the virus (cut-off value 0.3). The average optical density measured in the enzyme-linked immunosorbent assay was 0.51 at initial presentation, 1.65 upon remission and 0.83 at 3 months postrecovery. Time to clinical regression varied between 1 month and 1 year. Conclusions and clinical importance – These data support existing evidence for a high prevalence of CPV1 in canine oral papillomatosis. The healing process seems to correlate with a strong antibody response, and antibody titres peaked around the time of clinical recovery. In contrast to previous data from laboratory settings, the variation in remission time was very high.
Introduction Most forms of canine papillomatosis are thought to be induced by papillomavirus infections. The best-known and probably most prevalent form is canine oral papillomatosis, which is clinically characterized by the presence of exophytic papillomas within the oral cavity or at mucocutaneous junctions. These papillomas are characteristic in their cauliflower-like appearance and easily diagnosed clinically. Clinical disease is typically transient and primarily seen in young dogs. The infectious agent is canine papillomavirus type 1 (CPV1), a Lambda papillomavirus, which can also be involved in cutaneous papillomatosis. Cutaneous manifestations include exophytic and endophytic papillomas, pigmented plaques and, in rare cases, squamous cell carcinomas.1
Accepted 13 October 2014 Sources of Funding: This study was self-funded. Conflict of Interest: No conflicts of interest have been declared. © 2014 ESVD and ACVD, Veterinary Dermatology, 26, 57–e20.
To date, 14 distinct CPV types have been reported.1–3 Of these, types CPV1 and CPV13 have been isolated from oral papillomas.1,2 Despite the apparent association with certain disorders, papillomavirus DNA can also be found in the absence of signs, which could indicate early or subclinical infections or simply carriage.4 Immunocompetent individuals are usually able to clear or control most papillomavirus infections, with cellular immunity being the primary factor.5 Antibodies protect from re-infection with the same virus type, and thus generic or recombinant vaccines can be used prophylactically.5–8 Serum antibody titres can also help detect past infections using specific enzyme-linked immunosorbant assays (ELISAs). Prevalence rates of 10.5 and 21.9% have been reported for CPV1 antibodies in asymptomatic dogs.9 Despite data showing antibody induction in controlled laboratory conditions, antibody responses to natural CPV infections, induced by supposedly lower viral loads, have not been studied in dogs.7 The study reported here was initiated to characterize the antibody responses that natural CPV1 infections induce and to monitor antibody titres prior to and beyond clinical regression using a 57
Sancak et al.
CPV1-specific ELISA. Such information may promote better understanding of the biology of this common disease, especially in the context of prognosis and potential vaccination regimens.
Materials and methods Forty dogs with clinical oral papillomatosis were recruited at the veterinary teaching hospital in Ankara, Turkey. The Ethics Committee of the University of Ankara approved all procedures (permission 2011107). Initially, a biopsy of the oral lesions and a serum sample were collected from all dogs and stored at 20°C. None of the dogs underwent antiviral or other therapies. Owners were asked to present the dogs for follow-up physical examinations and sampling upon and after clinical recovery. Clinicians verified the regression. Follow-up serum samples were collected at the time of remission from 13 dogs, at 3 months postremission in four dogs and at 6 months postremission from one dog. DNA was extracted from biopsy samples using a DNeasy extraction kit (Qiagen, Hombrechtikon, Switzerland). To verify successful extraction, a PCR amplifying canine glyceraldehyde 3-phosphate dehydrogenase (GAPDH) DNA was performed, while canPV/FAP64 and PPF1/CP5 primers were applied to screen for CPV DNA. Reported PCR conditions were used.4 Diluted CPV1 DNA served as a positive control, elution buffer from the DNeasy extraction kit as a negative control, and all canPV/FAP64 PCR products were determined (Microsynth, Balgach, Switzerland) by cycle sequencing using an ABI 377 sequencer (Applied Biosystems, Carlsbad, CA, USA) and compared with the NCBI database (BLAST X, http://blast.ncbi.nlm.nih.gov/Blast.cgi) to identify the virus type. To screen for serum antibodies, the L1 protein of CPV1 was expressed in Escherichia coli and coated onto ELISA plates as described previously.9 All dog sera were tested in triplicate diluted 1:500, avoiding more than one peripheral well per triplicate. Sera previously tested positive or negative served as controls.9 For the analysis, the mean of the two optical densities (ODs) closest to each other was taken, while the third OD of the triplicate was disregarded. To set a cut-off value, positive and negative control ODs were compared; the cut-off value was set as the OD resulting in an equal distribution of false positives and false negatives among the controls.
Results Of the 40 dogs sampled, 23 were male and 17 female, with an age range of 3 months to 13 years (average 2 years). Twelve golden retrievers, 14 individuals of other pure breeds and 14 mixed-breed dogs were included. Besides oral papillomatosis, two dogs also had periocular papillomas, while two more had other cutaneous sites affected. All dogs for which follow-up information was available achieved remission without specific treatment. The PCR for GAPDH amplified host DNA in 39 samples. All 40 samples were positive in the broad range PCR for CPV (canPV/FAP64) and 36 also in the narrow-range PCR for DNA of Lambda papillomaviruses (PPF1/CP5). The sequencing of the broad-range PCR products and comparison with the databases confirmed exact sequence identity with the CPV1 DNA reference in all cases. The cut-off value for the ELISA was set at an OD of 0.309 (data not shown). Of the 58 serum samples collected, 53 were classified as positive. Of the five negative samples, four were collected at initial presentation and one upon recovery. Two of the dogs that were seronegative at baseline had seroconverted at the time of resampling. No additional samples were available in the other two cases. Overall, the ODs ranged from 0.25 to 3.97 (mean 1.18). Among the 13 dogs where at least two 58
Figure 1. Individual antibody titres are represented as optical density (OD) at three time points, i.e. >1 month prior to resolution, around the time of resolution and >1 month postrecovery. Bars indicate the mean value and SD.
samples were available, antibody titres increased in 10 cases, whereas they decreased in three. The increase between initial presentation and recovery in this group overall was significant (Wilcoxon matched pairs signedranks test, P = 0.047) or highly significant (Mann–Whitney U-test; P = 0.003), respectively. The mean OD of serum samples collected more than 1 month prior to recovery (n = 9) was 0.51 (SD 0.26), whereas the mean OD for samples collected at the time of clinical remission (1 month; n = 18) was 1.65 (SD 1.2). The latter samples included seven dogs where both initial presentation and recovery fell within a 1 month time frame. At 3 months postrecovery (n = 4), the mean OD was 0.83 (SD 0.46; Figure 1). For 24 dogs, the initial presentation could not be linked to any recovery time due to lack of appropriate follow-up information. Eleven of the cases with follow-up (n = 16) went into complete regression within 3 months or less. In three dogs, the number of papillomas increased significantly before all regressed. Perhaps due to the limited number of dogs with complete follow-up, no statistical correlation between severity of clinical signs and antibody titres was found.
Discussion It is widely accepted that canine oral papillomatosis is induced by a papillomavirus infection, and the identification of CPV1 in all samples collected within this study further strengthens that association. There was no evidence of other CPV sequences in the electropherograms, which were obtained using primers with the potential to amplify most known CPVs. Although CPV DNA may also be found in samples of clinically healthy skin, the high prevalence (100%) and the large amount of amplified DNA in this study (data not shown) indicate an active viral infection.4 The induction of protective antibodies against CPV1 is likely to explain why oral papillomatosis is typically seen in young dogs with virus-na€ıve immune systems.7 The ODs found here at the time of initial presentation were © 2014 ESVD and ACVD, Veterinary Dermatology, 26, 57–e20.
Antibody titres to CPV1
variable. If results for only those dogs from which multiple samples were available are examined, a clear trend for an increase and peak in titres around the time of clinical regression is apparent (Figure 1). A similar peak occurred around the time of clinical regression (between 11 and 14 weeks) in dogs with experimentally induced CPV1 infection.7 The clinical nature of our study prohibited accurate determination of the time between exposure and appearance of clinical papillomatosis. This may explain the large variability of antibody titres among the serum samples at the time of initial presentation and why four sera taken at this point were deemed negative despite overt clinical disease. Alternatively, this variability may be a reflection of individual differences among the heterogeneous sample cohort. Lesion regression occurred within 3 months after initial presentation in 11 of the dogs, while the time between presentation and regression was >3 months in five dogs; therefore, the time to remission appears to be much more variable than in experimental conditions.7 As a consequence of this enormous variation, clinical studies on treatment regimens for oral papillomatosis should be interpreted with caution. Treatment effects may be either overshadowed or overestimated if test groups are small. We consider the observed remission of papillomas to be spontaneous healing because no specific therapy was initiated. It is possible that the biopsy procedure triggered an active immune response, because partial crushing or removal of oral papillomas has been anecdotally suggested to hasten resolution; however, the contrary effect has also been proposed.10 The follow-up information available is statistically not conclusive regarding this matter. However, the extended time required for spontaneous remission in five dogs and the increased numbers of papillomas in three dogs prior to recovery do not seem to support a beneficial effect. In conclusion, results of this study show that the healing process in naturally occurring infections is more variable than indicated by laboratory experiments. The data indicate, nevertheless, that clinical resolution generally correlates with a strong antibody response and
that antibody titres peak around the time of clinical remission.
Acknowledgements The authors thank Kurt Tobler and Mathias Ackermann of the Institute of Virology, Vetsuisse Faculty, University of Zurich, as well as Peter M. Howley of the Microbiology and Immunobiology Department at Harvard Medical School and the clinicians of the veterinary teaching hospital of the Ankara University for logistical and general support.
References 1. Lange CE, Favrot C. Canine Papillomaviruses. Vet Clin North Am Small Anim Pract 2011; 41: 1183–1195. 2. Lange CE, Ackermann M, Favrot C et al. Entire genomic sequence of novel canine papillomavirus type 13. J Virol 2012; 86: 10226–10227. 3. Lange CE, Tobler K, Schraner EM et al. Complete canine papillomavirus life cycle in pigmented lesions. Vet Microbiol 2013; 162: 388–395. 4. Lange CE, Zollinger S, Tobler K et al. The clinically healthy skin of dogs is a potential reservoir for canine papillomaviruses. J Clin Microbiol 2011; 49: 707–709. 5. Howley PM, Schiller JC, Lowy DR. Papillomaviruses. In: Knipe DM, Howley PM, eds. Fields’ Virology 6th edition. Philadelphia, PA: Lippincott, Williams & Wilkins, 2013; 1662–1699. 6. Bell JA, Sunberg JP, Ghim S et al. A formalin-inactivated vaccine protects against mucosal papillomavirus infection: a canine model. Pathobiology 1994; 62: 194–198. 7. Suzich JA, Ghim SJ, Palmer-Hill FJ et al. Systemic immunization with papillomavirus L1 protein completely prevents the development of viral mucosal papillomas. Proc Natl Acad Sci USA 1995; 92: 11553–11557. 8. Ghim SJ, Newsome J, Bell JP et al. Spontaneously regressing oral papillomas induce systemic antibodies that neutralize canine oral papillomavirus. Exp Mol Pathol 2000; 68: 147–151. 9. Lange CE, Tobler K, Favrot C et al. Detection of antibodies against epidermoplasia verruciformis-associated canine papillomavirus 3 in sera of dogs from Europe and Africa by enzyme-linked immunosorbent assay. Clin Vaccine Immunol 2009; 16: 66–72. 10. Collier LL, Collins BK. Excision and cryosurgical ablation of severe periocular papillomatosis in a dog. J Am Vet Med Assoc 1994; 204: 881–885.
sume Re es e ^tre associe es Contexte – La plupart des formes de papillomatoses canines sont cense a des infections a papillomavirus. Le papillomavirus canin de type 1 (CPV1) est conside re comme responsable de la plupart es. des cas oraux et de plusieurs formes de papillomatoses cutane ses/Objectifs – Le but de cette e tude e tait d’e valuer le type de virus implique , l’induction d’antiHypothe e de re mission les cas d’origine naturelle de papillomatose orale. corps et la dure thodes – Quarante chiens montrant diffe rents degre s de papillomatose orale classique ont e te inclus Me tude. Des e chantillons de tissus et de serum ont e te pre leve s te dans l’e a la visite initiale; les sera ont e leve s de nouveau au moment de la re mission (n = 13) et apre s 3 mois de convalescence (n = 4). Aucun pre rapie antivirale. Les e chantillons de tissu ont e te teste s par PCR pour de tecter des chiens n’a recßu de the chantillons de se rum ont e te teste s par ELISA pour les anticorps antil’ADN de CPV, alors que les e ique de capside L1 du CPV1. prote sultats – Tous les e chantillons tissulaires e taient positifs pour l’ADN de CPV1 et 87.5% de tous les sera Re s contre le virus (valeur seuil 0.3). La densite optique contenaient des taux mesurables d’anticorps dirige e par ELISA e tait de 0.51 re visite, 1.65 mission et 0.83 apre s 3 mois de moyenne mesure a la premie a la re mission. La dure e de re gression clinique variait de 1 mois re a 1 an.
© 2014 ESVD and ACVD, Veterinary Dermatology, 26, 57–e20.
59
Sancak et al.
es supportent les preuves existantes d’une haute Conclusions et importance clinique – Ces donne valence de CPV1 dans les papillomatoses orales canines. Le processus de gue rison semble corre ler pre ponse d’anticorps avec des titres d’anticorps les plus e leve s au moment de la gue rison cliavec une forte re rence des donne es ante rieures de laboratoire, la variation de la dure e de re mission e tait tre s nique. A la diffe elevee. Resumen n – la mayorıa de las formas de papilomatosis canina parecen estar asociadas con infeccio n Introduccio por papilomavirus. El papilomavirus canino de tipo I (CPV1) se considera responsable de la mayorıa de los casos orales y de varias formas de papilomatosis cut aneo. tesis/Objetivos – el propo sito de este estudio fue evaluar casos de papilomatosis oral espont Hipo anea en n al tipo de virus implicado, produccio n de anticuerpos, y tiempo de remisio n. relacio todos – 40 perros que mostraron diferentes grados de papilomatosis oral cl Me asica se incluyeron en un n inicial; y muestras de susolo grupo de estudio. Muestras de suero y de tejido se tomaron a la presentacio n (n = 13) y tras tres meses convalecencia (n = cuatro). Ninguno ero se recogieron de nuevo tras la remisio de los perros fue sometido a terapia antivırica. Las muestras de tejido fueron evaluadas mediante PCR para detectar el DNA de papilomavirus canino, mientras que las muestras de suero fueron probadas usando una n ligada a enzimas para anticuerpos frente a la proteına de la c prueba de inmunoabsorcio apsula L1 del papilomavirus canino de tipo I. Resultados – todas las muestras de tejido fueron positivas para DNA de papilomavirus canino de tipo I, y 87,5% de las muestras de suero contenıan niveles valorables de anticuerpos frente al virus (valor de corte ptica medida en el ensayo de inmunoabsorcio n fue de 0,51 a la presentacio n ini0,3). La media densidad o n, y 0,83 a los tres meses tras la recuperacio n. El tiempo para la regresio n clınica cial, 1,65 tras la remisio entre un mes y un an ~o. vario Conclusiones e importancia clınica – estos datos aportan nueva evidencia de una gran prevalencia del n parece correlacionpapilomavirus canino tipo I en casos de papilomatosis oral. El proceso de recuperacio arse con una fuerte respuesta de anticuerpos, y los tıtulos de anticuerpos fueron maximos alrededor del tin clınica. En contraste con datos previos en pruebas de laboratorio la variacio n en el empo de la recuperacio n fue bastante amplia. tiempo de remisio Zusammenfassung Hintergrund – Man geht davon aus, dass die meisten Formen der Papillomatose des Hundes durch Papil€r die lomavirus verursacht werden. Es wird angenommen, dass das canine Papillomavirus Typ 1 (CPV1) fu €r mehrere Formen der Papillomatose der Haut verantwortlich meisten F€alle oraler Papillomatose sowie fu ist. €rlich auftretender oraler PaHypothese/Ziele – Das Ziel dieser Studie war eine Evaluierung von F€ allen natu €rperinduktion und Remissionszeit. pillomatose in Bezug auf den involvierten Virustyp, die Antiko Methoden – Vierzig Hunde, die unterschiedliche Auspr€ agungen der klassischen oralen Papillomatose zeigten, wurden als eine Untersuchungsgruppe verwendet. Bei der Erstvorstellung wurden Gewebe- und Blutproben entnommen; zum Zeitpunkt der Remission (n=13) und 3 Monate nach der Heilung (n= 4) wurden €hrt. Die wieder Serumproben genommen. Bei keinem der Hunde wurde eine antivirale Therapie durchgefu Gewebeproben wurden mittels PCR auf CPV DNA getestet, w€ ahrend die Serumproben mittels spezifi€rper gegen L1 Kapsidprotein von CPV1 getestet wurscher Enzym-linked Immunosorbent Assay auf Antiko den. €r CPV1 DNA, und 87,5% aller Serumproben enthielten Ergebnisse – Alle Gewebeproben waren positiv fu €rpern gegen das Virus (Cut-off Wert 0,3). Die durchschnittliche optische Dichte, messbare Werte an Antiko die mittels Enzym-linked Immunosorbent Assay gemessen wurde, betrug 0,51 bei der Erstpr€ asentation, 1,65 bei Remission und 0,83 nach 3 Monaten der Heilung. Die Zeit bis zur klinischen Regression variierte zwischen 1 Monat und 1 Jahr. €tzen die Evidenz einer hohen Schlussfolgerungen und klinische Bedeutung – Diese Daten unterstu Pr€avalenz von CPV1 bei der oralen Papillomatose des Hundes. Der Heilungsprozess scheint mit einer star€rperantwort zu korrelieren, die Antiko €rpertiter waren zum Zeitpunkt der Heilung am ho €chsten. ken Antiko €heren Daten aus Laborsettings war die Variation der Dauer bis zur Remission sehr Im Gegensatz zu fru hoch.
要約 背景 – イヌの乳頭腫症のほとんどの型はパピローマウイルス感染と関連していると考えられている。イ ヌパピローマウイルス1型(CPV1)はほとんどの口腔の症例、および皮膚の乳頭腫症のいくつかの型の原因 となると考えられている。 e19
© 2014 ESVD and ACVD, Veterinary Dermatology, 26, 57–e20.
Antibody titres to CPV1
仮説/目的この研究の目的は関与するウイルス型、抗体誘導ならびに寛解時間に関して自然発生性口腔乳 頭腫症の症例を評価すること。 方法 – 異なった程度の古典的な口腔乳頭腫症を示す40頭のイヌを単一の研究グループに含めた。組織お よび血清サンプルは初診時に得た、血清サンプルは寛解時(n = 13)と回復期の3ヶ月後 (n = 4)に再度回収 した。抗ウイルス療法はどのイヌにも行わなかった。組織サンプルはCPV DNAを検出するPCRで検査し、血 清サンプルはCPV1のL1カプシドタンパクに対する抗体を、特異的酵素結合免疫吸着検査法を用いて検査 した。 結果 – すべての組織サンプルはCPV1 DNAに対して陽性で、すべての血清サンプルの87.5%がウイルスに対 する抗体を測定可能なレベルで含んでいた(カットオフ値0.3)。酵素結合免疫吸着検査法で測定した平均 可視化濃度は初診時に0.51、寛解時に1.65、および回復後3ヶ月で0.83であった。臨床的な退縮までの時 間は1ヶ月から1年まで様々であった。 結論および臨床的な重要性 – これらのデータはイヌの口腔乳頭腫症においてCPV1の高い保有率の既存の 証拠を支持する。治癒プロセスは強力な抗体反応と関連しているようであり、抗体価のピークは臨床的 な回復期付近のようである。研究室で行われた以前のデータと比較し、寛解時間のバリエーションは非 常に多かった。 摘要 背景——大部分犬乳头状瘤被认为与乳头状瘤病毒感染有关。犬乳头状瘤病毒1型(CPV1)是引起大部分口腔 和几种皮肤乳头状瘤的病因。 假设/目的——本次研究的目的是评估自然发生的口腔乳头状瘤,其感染病毒的类型、抗体感应和缓解时 间。 方法——四十只不同程度的典型口腔乳头状瘤患犬,分为一个单独研究组。发病初期取组织和血清样本;恢 复期(n=13)和3个月康复期后(n=4)再取血清样本。所有犬未经过抗病毒治疗。组织样本通过PCR实验检测 CPV DNA,血清样本则用特异性酶链免疫吸附实验,检测抗CPV1的L1衣壳蛋白抗体。 结果——所有组织样本为CPV1 NDA阳性,87.5%的血清样本包含达到可测量水平的病毒抗体(截止值0.3)。酶 链免疫吸附实验中测量的平均光度值,发病初期为0.51,恢复期为1.65,3个月康复期后为0.83。临床恢复时 间为1个月到1年不等。 总结与临床意义——这些数据显示,CPV1在犬口腔乳头状瘤中有很高的流行率。恢复过程似乎与强烈的抗体 反应有关,抗体滴度在临床症状恢复期到达峰值。对比之前的实验室数据,恢复时间差异很大。
© 2014 ESVD and ACVD, Veterinary Dermatology, 26, 57–e20.
e20
