Limited pathogenicity of Usutu virus for the domestic chicken (Gallus domesticus)

Abstract

In summer 2001, Usutu virus (USUV), a mosquito-borne flavivirus, was isolated for the first time in Europe during a mortality incident among Eurasian blackbirds (Turdus merula) in Austria. Chickens are frequently used as sentinel animals for arbovirus surveillance systems. In the present study, the pathogenicity of USUV for specific pathogen free chickens was investigated. Ten 2-week-old chickens were inoculated intravenously with 0.1 ml inoculum containing 10³ median (50%) tissue culture infectious dose of USUV strain Vienna 2001-blackbird (939/01). Clinical signs, viraemia, gross and microscopic lesions, contact transmission and immunological response were evaluated. No clinical signs were observed in the USUV-inoculated animals during the experimental period. Pathological examination showed moderate splenomegaly and follicular infiltrates in the liver of several inoculated animals. Mild non-suppurative encephalitis was observed in the brain tissue of one virus-inoculated chicken examined 7 days post inoculation (d.p.i.). USUV nucleic acid was detected by reverse-transcriptase polymerase chain reaction in the organs of six inoculated chickens, although immunohistochemistry for flavivirus antigen was negative in all tissues from all chickens. Virus shedding was shown in three inoculated birds by detecting USUV RNA in cloacal swabs of two chickens at 5 d.p.i., and in the cloacal and pharyngeal swabs of one chicken at 7 d.p.i. Based on detection of viral RNA in peripheral blood mononuclear cells, viraemia was detected only in two chickens (at 7 d.p.i.). Only one of the inoculated chickens developed an antibody response. There was no evidence of virus transmission to chickens kept in contact with inoculated birds. No USUV was isolated from in-contact birds and all in-contact and control animals lacked USUV-specific antibodies. The present data suggest that domestic chickens are not at risk of developing clinical disease following USUV infection and that chickens are unlikely to be useful for sentinel purposes in USUV surveillance programmes.

Pathogénicité limitée du virus Usutu pour le poulet domestique (Gallus domesticus)

Au cours de l'été 2001, un flavivirus, le virus Usutu (USUV), transmis par un moustique, a été isolé pour la première fois en Europe durant un épisode de mortalité observé chez des merles noirs eurasiens (Turdus merula) en Autriche. Les poulets sont souvent utilisés comme animaux sentinelles dans les systèmes de surveillance des arbovirus. Dans cette étude, la pathogénicité du l'USUV pour les poulets exempts de microorganismes pathogènes spécifiés, a été étudiée. Dix poulets âgés de 2 semaines ont été inoculés par voie intraveineuse avec 0,1 ml d'inoculum contenant 10³ TCID₅₀ de la souche Vienna 2001-blackbird (939/01) de l'USUV. Les symptômes, la virémie, les lésions macro et microscopiques, la transmission par contact et la réponse immunologique ont été évalués. Aucun symptôme n'a été observé chez les animaux inoculés avec l'USUV durant la période expérimentale. L'examen pathologique a montré une splénomégalie modérée et des infiltrats folliculaires dans le foie de plusieurs animaux inoculés. Une légère encéphalite non suppurative a été observée dans les tissus du cerveau d'un poulet inoculé avec le virus et examiné 7 jours après l'inoculation (d.p.i). L'acide nucléique de l'USUV a été détecté par RT-PCR dans les organes de 6 poulets inoculés, bien que l'immunohistochimie vis-à-vis de l'antigène flavivirus ait été négative pour tous les tissus de tous les poulets. La diffusion du virus a été mise en évidence chez 3 poulets inoculés, par détection d'ARN, au niveau des écouvillons cloacaux, 5 d.p.i., et au niveau des écouvillons cloacaux et trachéaux, 7 d.p.i.. La virémie a été recherchée par mise en évidence de l'ARN viral dans les cellules mononucléaires du sang périphérique, seuls deux poulets ont été positifs 7 d.p.i.. Un seul poulet inoculé a développé une réponse en anticorps. Il n'a pas été mis en évidence de transmission virale chez les poulets maintenus en contact de ceux inoculés. L'USUV n'a pas été isolé chez les animaux contact et aucun anticorps spécifique du USUV n'a été détecté chez tous les contacts et les témoins. Toutes ces données suggèrent que (1) les poulets ne risquent pas de développer une maladie clinique après l'infection par le USUV et (2) les poulets ne sont probablement pas utiles comme sentinelles dans les programmes de surveillance de l'USUV.

Beschränkte Pathogenität des Usutuvirus für das Haushuhn (Gallus domesticus)

Im Sommer 2001 wurde das Usutuvirus (USUV), ein durch Mücken übertragbares Flavivirus erstmals in Europa im Zusammenhang mit Todesfällen bei eurasischen Amseln (Turdus merula) in Österreich isoliert. Hühner werden oft als Indikatortiere in Arbovirus-Überwachungsprogrammen eingesetzt. In der vorliegenden Studie wurde die Pathogenität des USUV für spezifisch pathogen freie Hühner untersucht. 10 zweiwöchigen Hühnerküken wurde intravenös 0,1 ml Inokulum mit 10³ TCID₅₀ des USUV-Stamms Vienna 2001-Amsel (939/01) injiziert. Klinische Symptome, Virämie, pathologisch-anatomische und –histologische Befunde, Kontaktübertragung sowie Immunantworten wurden ermittelt. Während des Untersuchungszeitraums wurden keine klinischen Symptome bei den USUV inokulierten Tieren beobachtet. Bei der pathologischen Untersuchung wurden bei mehreren Tieren geringgradige Splenomegalie und follikuläre Infiltrate in der Leber beobachtet. Im Gehirn eines der mit Virus inokulierten Hühnerküken wurden 7 Tage post inoculationem (d.p.i.) eine geringgradige nichteitrige Enzephalitis festgestellt. USUV-Nukleinsäure wurde mittels Reverse Transkriptase Polymerasekettenreaktion (RT-PCR) in den Organen von 6 inokulierten Hühnerküken nachgewiesen, obwohl die immunhistochemische Untersuchung auf Flavivirus-Antigen in allen Organen aller Tiere negativ war. Die Virusausscheidung wurde bei drei inokulierten Tieren aufgezeigt, wobei USUV-RNS in Kloakenabstrichen von zwei Tieren am 5. d.p.i. und bei einem Tier im Kloaken- und Pharynxabstrich am 7. d.p.i. entdeckt wurde. Eine Virämie, nachgewiesen durch virale RNA in peripheren mononukleären Blutzellen, konnte nur bei zwei Küken (am 7. d.p.i.) festgestellt werden. Nur eines der inokulierten Tiere entwickelte eine serologische Immunantwort. Es gab keinen Nachweis einer Virusübertragung auf Hühnerküken, die in Kontakt mit den inokulierten Tieren gehalten wurden. USUV konnte nicht aus den Kontakttieren isoliert werden und weder die Kontakttiere noch die Kontrolltiere wiesen USUV spezifische Antikörper auf. Diese Ergebnisse lassen erkennen, dass 1. das Haushuhn kein Risiko hat, nach einer USUV-Infektion klinisch zu erkranken, und 2. es unwahrscheinlich ist, dass das Huhn sich als Indikatortier in USUV-Überwachungsprogrammen eignet.

Patogenicidad limitada del virus de Usutu en pollos domésticos (Gallus domesticus)

En verano del 2001, el virus de Usutu (USUV), un flavivirus transmitido por mosquitos, fue aislado por primera vez en Europa en el curso de un episodio de mortalidad en mirlo común (Turdus merula) en Austria. Los pollos se utilizan con frecuencia como animales centinelas en los programas de vigilancia de arbovirus. En el presente estudio, se investigó la patogenicidad del USUV en pollos libres de patógenos específicos. Diez pollos de 2 semanas de edad fueron inoculados vía intravenosa con 0.1 ml del inóculo que contenía 10³ TCID₅₀ del USUV cepa Vienna 2001-blackbird (939/01). Los síntomas clínicos, viremia, lesiones macroscópicas y microscópicas, transmisión por contacto así como las respuestas inmunológicas fueron evaluadas. No se observaron síntomas clínicos en las aves inoculadas con USUV durante el experimento. El examen patológico mostró moderada esplenomegalia e infiltrados foliculares en el hígado de varios animales inoculados. Se observó leve encefalitis no supurativa en un pollo inoculado, a los 7 días post inoculación (d.p.i.). El ácido nucleico del USUV fue detectado mediante transcriptasa reversa- reacción en cadena de la polimerasa (RT-PCR) en órganos de 6 pollos inoculados, aunque la inmunocitoquímica (IHC) para detectar antígeno de flavivirus fue negativa en los tejidos de todos los pollos. La excreción viral se demostró en 3 aves inoculadas mediante la detección de ARN de USUV en hisopos cloacales en dos pollos a los 5 d.p.i., y en hisopos cloacales y faríngeos de un pollo a los 7 d.p.i. En base a la detección de ARN viral en células mononucleares periféricas sanguíneas, se detectó viremia únicamente en dos pollos (a los 7 d.p.i.). Unicamente una de las aves infectadas desarrolló respuesta inmune. No hubo evidencia de transmisión vírica a los pollos que se encontraban en contacto con los inoculados. No se aisló USUV de las aves en contacto ni se detectaron anticuerpos específicos frente a USUV en ninguno de los animales en contacto ni en controles. Los datos presentados sugieren que (1) los pollos domésticos no están a riesgo de desarrollar enfermedad clínica tras la infección con USUV y (2) los pollos no parecen ser útiles para ser utilizados como centinelas en programas de vigilancia de USUV.

Introduction

Usutu virus (USUV) is a mosquito-borne flavivirus, which has been named after a river in Swaziland, South Africa. It is closely related to human pathogens such as Japanese encephalitis virus, Murray Valley encephalitis virus (MVEV), Saint Louis encephalitis virus (SLEV) and West Nile Virus (WNV) (Kuno et al., 1998; Bakonyi et al., 2004).

In late summer 2001 a die-off of Eurasian blackbirds (Turdus merula) and great grey owls (Strix nebulosa) was observed in and around Vienna, Austria. USUV was diagnosed as the causative agent (Weissenböck et al., 2002). This was the first time that USUV had emerged outside Africa, and it has caused fatalities in wild birds, predominantly blackbirds, in four consecutive summers since. The disease in birds is characterized by encephalitis, myocardial degeneration, and necrosis in the liver and spleen (Chvala et al., 2004).

Domestic chickens (Gallus domesticus) are frequently used to test infectivity and immune response to arboviruses. Chung (1970) showed that inoculation of chickens with MVEV and Sindbis virus resulted in viraemia and seroconversion, in the absence of clinical signs. MVEV activity in Australia is monitored by serosurveillance of chicken sentinel flocks (Broom et al., 2002b). The incidence and importance of WNV in wild birds and chickens has been described (van der Meulen et al., 2005). Also, WNV is reported to produce no clinical signs after experimental infection of chickens, but neutralizing antibodies are consistently found (Senne et al., 2000; Langevin et al., 2001). Thus chickens are used as sentinel species for WNV surveillance programmes in North America and in the Caribbean (Langevin et al., 2001; Quirin et al., 2004; Reisen et al., 2004). The enzootic activity of SLEV and Western equine encephalomyelitis virus was also monitored by seroconversions in chicken flocks in California (Reisen et al., 2000).

The potential pathogenicity of USUV in domestic chickens is unknown. As this virus seems to have become a resident pathogen in Austria, it is advisable to investigate whether it can pose a health threat to domestic chickens. In addition, by monitoring viraemia and seroconversion, the possibility that chickens may serve as amplifying hosts or may be used for serosurveillance purposes requires investigation. This study was performed to evaluate the clinical signs, viraemia, gross and microscopic lesions, contact transmission and immunological response in chickens experimentally inoculated with USUV.

Materials and Methods

Virus

USUV strain Vienna 2001-blackbird (939/01) was originally obtained from the brain homogenate of a blackbird found dead in Perchtoldsdorf (Mödling) near Vienna on 4 September 2001. The strain was isolated in Vero cell culture and was identified by immunohistochemistry (IHC), in situ hybridization, reverse transcriptase-polymerase chain reaction (RT-PCR), and sequencing. The strain was passaged two times in Vero cells to obtain a stock of virus for the experimental infection (USUV p3). The infectivity titre of the stock virus was 4.1×10⁵ median (50%) tissue culture infectious dose (TCID₅₀) per ml.

Experimental design

The experiment was performed in accordance with the Austrian legislation for animal experiments (permission number 68.205/100-BrGt/2002).

Eighteen 2-week-old specific pathogen free chickens (VALO; Lohmann, Cuxhaven, Germany) were used. Ten chickens were individually inoculated intravenously with 0.1 ml inoculum containing 10³ TCID₅₀ USUV. These chickens were placed together with three uninoculated chickens (in-contact birds) in an isolation cabinet. Five uninoculated chickens that served as a control group were housed separately. All chickens were observed daily for clinical signs. The experiment was terminated after 14 days.

From the inoculated chickens and the in-contact birds, cloacal and oropharyngeal swabs were taken 1, 2, 3, 4, 5, 7, 10 and 14 days post inoculation (d.p.i.), and blood samples were taken 1, 3, 5, 7, 10 and 14 d.p.i. At 3, 5, 7, 10 and 14 d.p.i., two inoculated chickens were killed. All in-contact chickens were killed 14 d.p.i.

From the control group one chicken was bled and killed at 3, 5, 7, 10 and 14 d.p.i.

All chickens were necropsied and examined for gross lesions. Between 1 and 4 h after death, tissue samples of the brain, heart, lung, liver, spleen, kidney, intestine, gizzard, proventriculus and pancreas were collected and fixed in neutral buffered 7% formalin for histopathological and immunohistochemical examination. In addition, the brain, heart, lung, liver, spleen, kidney and intestine were frozen and stored at −80°C for RT-PCR investigations.

Detection of viral RNA and USUV in chicken samples

For determination of viraemia, 0.5 ml blood was drawn from the ulnar vein into ethylenediaminetetraacetic acid-treated tubes (Sarstedt, Nürnbrecht, Germany). The blood was centrifuged at 6700×g for 5 min, and the plasma was saved for serological studies. Peripheral blood mononuclear cells (PBMCs) were purified using Erythrocyte Lysis buffer (Qiagen, Hilden, Germany) according to the manufacturer's recommendations.

All organ samples obtained from the same animal were pooled prior to RNA extraction. Approximately 200 mg each organ were mixed and homogenized in ceramic mortars using sterile sand. Thereafter the homogenates were suspended in 1500 µl diethyl-pyrocarbonate-treated distilled water, and the suspension was centrifuged at 8000×g for 10 min.

Cloacal and pharyngeal swabs were collected in microcentrifuge tubes containing 1 ml aliquots of 0.5% bovine serum albumin–minimum essential medium (Gibco-Invitrogen, Paisley, UK), pH 7.3, with 1% penicillin/streptomycin, 1% neomycin, and 0.2% econazole. Tubes were vortexed, swabs were squeezed and removed, and the tubes were centrifuged at 8000×g for 10 min.

RNA was isolated from 140 µl supernatant using the QiaAmp Viral RNA Mini Kit (Qiagen) according to the manufacturer's instructions. Reverse transcription and amplification were performed with a continuous RT-PCR method using the QIAGEN OneStep RT-PCR Kit (Qiagen) as previously described (Bakonyi et al., 2004).

Virus isolation attempts on Vero cell cultures were made from a subset of samples (pooled organs of chickens 1 to 6 and all samples per time point from other PCR-positive chickens) using standard cell culture and isolation techniques, including three subsequent blind passages (Weissenböck et al., 2002; Bakonyi et al., 2004).

Histology and IHC

Tissue samples were processed to haematoxylin and eosin-stained tissue sections and examined by light microscopy. Sections were stained for demonstration of viral antigen using the avidin–biotin complex technique with polyclonal antibodies to USUV, as described previously (Weissenböck et al., 2004).

Serological investigation

Plasma samples were tested by the haemagglutination inhibition (HAI) test in order to detect USUV-specific antibody response. The HAI test procedure was adapted from the OIE Manual of Standards: HA and HAI tests for Japanese encephalitis virus (http://www.oie.int/eng/normes/mmanual/A_00092.htm).

Prior to the HAI test, USUV antigen was titrated in a haemagglutination (HA) assay using goose erythrocytes. Serial two-fold dilutions of antigen were made in 0.4% bovine albumin/borate saline (pH 9.0) and mixed with 0.5% goose red blood cell (RBC). The optimal pH environment was determined using RBC diluent solutions with different pH values. Tests were performed in U-shaped microtitre plates. Mixtures were incubated at room temperature for 30 min and results were read thereafter. The virus suspension exhibited a titre of 16 HA units at pH 6.2. For HAI tests, 8 HA units of this virus (1:2 dilution of the virus suspension) were used.

Plasma samples were inactivated at 56°C for 30 min. Non-specific HA activity was inactivated by kaolin treatment and adsorption with packed goose RBCs. Serial two-fold dilutions were made from plasma samples in microtitre plates, and 8 HA units of antigen were added to each dilution. Thereafter the mixtures were incubated at room temperature for 1 h and 0.5% goose RBCs were added to the wells. Results were read after incubation at room temperature for a further 30 min. The antibody titre was defined as the reciprocal of the highest dilution of the test plasma sample, which showed complete inhibition of HA.

Results

Clinical features

No clinical signs were observed in the inoculated, in-contact or control animals during the entire experimental period.

Gross lesions and histopathology

At necropsy, all inoculated chickens showed mild to moderate splenomegaly. Histologically, the tunica muscularis of the gizzard and the lamina propria of the proventriculus of all USUV-inoculated chickens contained moderate to severe diffuse infiltrates of mononuclear cells. Between 7 and 14 d.p.i. the majority of the inoculated chickens had follicular hyperplasia of the spleen and moderate perivascular and follicular lymphocytic infiltrates in the liver. One chicken showed focal mild lymphocytic perivascular cuffs and endothelial proliferation in the brain at 7 d.p.i.

In the in-contact chickens and control animals only mild mononuclear infiltrates in the propria of the proventriculus and single follicular lymphocytic infiltrates in the liver were present. No abnormalities were detected in the samples of gizzard or brain examined.

Immunohistochemistry

No USUV antigen was detected in the examined organs of all chickens.

Detection of viral nucleic acid by RT-PCR

A summary of all RT-PCR results (internal organs, serum and swabs) of the inoculated chickens is presented in Table 1. In the pooled internal organs, USUV nucleic acid was detected in six inoculated chickens, killed on 3, 5 and 7 d.p.i. Organs of the remaining four inoculated chickens killed 10 and 14 d.p.i., and of in-contact and control animals, were negative.

Table 1.  Viral RNA recovery by RT-PCR in internal organs, swabs and PBMCs, and HAI serum titres to USUV in experimentally infected chickens

Chicken	RT-PCR organs, time euthanized	RT-PCR cloacal swabs (d.p.i.)	RT-PCR oropharyngeal swabs (d.p.i.)	RT-PCR PBMCs (d.p.i.)	Serum titres (d.p.i.)
1	Positive , 3 d.p.i.	–	–	3	–
2	Positive , 3 d.p.i.	–	–	3	–
3	Positive , 5 d.p.i.	5	–	5	–
4	Positive , 5 d.p.i.	5	–	5	–
5	Positive , 7 d.p.i.	1, 2, 3, 4, 5, 7	1, 2, 3, 4, 5, 7	3, 5, 7	–
6	Positive , 7 d.p.i.	1, 2, 3, 4, 5 , 7	1, 2, 3, 4, 5, 7	3, 5, 7	–
7	Negative, 10 d.p.i.	1, 2, 3, 4, 5, 7, 10	1, 2, 3, 4, 5, 7, 10	3, 5, 7, 10	Negative, 10
8	Negative, 10 d.p.i.	1, 2, 3, 4, 5 , 7, 10	1, 2, 3, 4, 5, 7, 10	3, 5, 7, 10	Negative, 10
9	Negative, 14 d.p.i.	1, 2, 3, 4, 5, 7, 10, 14	1, 2, 3, 4, 5, 7, 10, 14	3, 5, 7 , 10,	1:40, 10; 1:20, 14
10	Negative, 14 d.p.i.	1, 2, 3, 4, 5, 7 , 10, 14	1, 2, 3, 4, 5, 7 , 10, 14	3, 5, 7 , 10,	Negative, 10; Negative, 14
Time points with positive results are in bold italic.

Viral RNA was amplified from cloacal swabs of two virus-inoculated chickens 5 d.p.i. Also, both cloacal and pharyngeal swabs taken from a virus-inoculated chicken at 7 d.p.i. were positive. The swabs of all other chickens tested negative.

Evidence of viraemia was found only in two inoculated chickens and only at 7 d.p.i. following the amplification of viral RNA from PBMCs.

Isolation of USUV

No live virus could be isolated from any of the samples tested on VERO cells.

Serology

Only one inoculated chicken developed a serological response, with antibodies being detected at 10 and 14 d.p.i. (Table 1). Antibodies were not detected in any of the in-contact or control animals.

Discussion

The present study showed that USUV did not cause clinical signs in intravenously inoculated chickens and only infrequently lead to viraemia or viral excretion.

Pathologically, the inoculated chickens showed signs of immune stimulation such as splenomegaly and immune cell infiltrates in various tissues. Viral antigen, however, was not detected in the tissues of infected animals. These changes may be attributed to a general activation of the immune system, associated with low-level viral replication. In other bird species like Eurasian blackbirds (T. merula) and great grey owls (S. nebulosa), USUV causes necroses in the liver, spleen, heart and nervous system (Chvala et al., 2004), but such changes were absent in the virus-inoculated chickens.

The finding that USUV did not induce any clinical signs in chickens is in line with data from related viruses such as WNV, MVEV and SLEV, which are also not pathogenic for chickens (Smith et al., 1947; Chung, 1970; Senne et al., 2000; Langevin et al., 2001). In the case of these viruses, and also some alphaviruses, however, chickens develop reliable and robust viraemias and seroconversions (Maguire & Miles, 1965; Campbell & Hore, 1975; Calisher et al., 1986; Reisen et al., 1994; Komar et al., 2001), which allow the use of chickens in vector competence studies and monitoring programmes for virus activity (Morris et al., 1994; Komar, 2001; Broom et al., 2002a; Blackmore et al., 2003; Reisen et al., 2004; Quirin et al., 2004). Contrary to these findings, experimental USUV infection only inconsistently led to viraemia and seroconversion. These data clearly show that low-grade and short-lived USUV replication may occur in chickens and that chickens are rather inefficient hosts for USUV.

The serological assay chosen for this study, the HAI test, showed only one responder. It might be argued that other tests might have yielded different results. The other method currently available for detection of USUV antibodies, the plaque reduction neutralization test, showed comparable results in a previous experiment (unpublished observation). We are therefore certain that the HAI test is a reliable method for the detection of serological response to USUV.

The RT-PCR amplification of USUV RNA from organ pools until 7 d.p.i. is most probably due to the detection of viral replication rather than the detection of residual viral inoculum. Additionally, positive PCR results for PBMCs collected at 5 and 7 d.p.i. and the seroconversion of one animal are indicative of active viral replication. The use of IHC, however, a technique that easily detects the viral proteins present in lethal cases of USUV in different avian species, failed to identify viral antigen in the examined tissues. It is probable that the amount of viral protein produced during the transient, low-grade infection in chickens is below the sensitivity threshold of the IHC test used in this study.

Considering these experimental data, one has to keep in mind that the route and conditions of inoculation might not be absolutely representative of that used by mosquitoes. Due to practical reasons and reasons of comparison, we adopted methods used in previous experiments with WNV (Senne et al., 2000; Langevin et al., 2001). However, natural infection via mosquitoes might result in a slightly different outcome. Nonetheless, looking at the experimental resistance of chickens to WNV, which is also proven under natural conditions (Komar, 2001), it is highly likely that the result of this experimental study is valid for the natural situation as well.

In summary, the present data suggest that domestic chickens are not at risk of developing clinical disease following USUV infection and that chickens are unlikely to be useful for sentinel purposes in USUV surveillance programmes.

Translations of the abstract in French, Germany and Spanish are available on the Avian Pathology website.

The authors cordially thank Helga Lussy for her excellent technical assistance.