Application of diagnostic procedures to epidemiological situations with special reference to arboviral infections.

The rationale behind the methodology employed to investigate a new disease that appears in an area hitherto unaffected is fundamentally different from that applied in an endemic disease situation. Special consideration must also be given to disease agents that appear and reappear at cyclical intervals. The authors present three separate approaches applicable in three different epidemiological situations. Each one may be used initially to determine the identity of the causal agent and the nature of the disease. These approaches, although described in a stepwise (flowchart) manner, are not meant to be applied rigidly, but rather should serve as a guideline for investigators. Special consideration is given to situations in which disease appears intermittently, using a sentinel model. Although this latter approach is expensive and time-consuming, it can yield excellent and reliable results when applied correctly.

The role of birds in the ecology of West Nile virus in Europe and Africa.

Surveys on wild birds conducted during the last two decades in Europe, notably Poland and the Czech Republic, to determine their infection rate with WN virus have revealed endemic foci of infection. Some species of seropositive birds were nonmigrators while others were hatchlings of migrating species. Persistently infected avian reservoirs are potential sources of viruses for mosquitoes that multiply in the temperate European zone in hot, wet summers. In the past, evidence for geographical circulation of WN viruses was based on antigenic analysis of strains from different countries while more recent epidemiological studies have relied on analysis of nucleotide sequences of the envelope gene. With the reappearance of epidemic WN fever in European countries, interest has been focused once again on the African origin of the causal agent carried by migrating wild birds. In some epidemics, isolates were made from human cases or mosquitoes and only serologic evidence for infection was available from domestic and wild bird populations. In this respect the unusual findings of anti-WN virus antibodies in a population of storks maintained in northern Germany could be interpreted as evidence for local infection. The unique susceptibility of young domestic geese in Israel in 1997-2000 to WN virus and the isolation of similar strains from migrating White storks in Israel and Egypt suggest that the recent isolates are more pathogenic for certain avain species and that migrating birds do play a crucial role in geographical spread of the virus. Knowledge of the routes taken by birds migrating between Africa and Europe will therefore help in selecting sites where attempts to isolate viruses will be most fruitful. The appearance of the disease in western European equine populations (Italy and France) requires that other birds and their migratory routes be investigated once more. It remains to be determined whether the European endemic foci of WN virus are in themselves sources of infection for other birds that migrate across Europe and do not necessarily reach sub-Saharan Africa. If this is the case it will be necessary to define the strategies for detection of virus overwintering in the European temperate climate.

Marek's disease in turkeys. I. A seven-year survey of commercial flocks and experimental infection using two field isolates.

Marek's disease virus (MDV) causes immunosuppression and tumors in chickens, but the turkey is an unusual host for the virus, and tumors caused by MDV in turkeys are unique. We describe the prevalence of turkey tumors in Israel between 1993 and 2000, their molecular diagnosis by polymerase chain reaction (PCR), and the natural distribution of herpesvirus of turkeys (HVT). Most clinical cases with tumors in commercial turkeys were diagnosed as MDV. The reproduction of Marek's disease (MD) in turkeys by two turkey MDV strains, Ar and La, was analyzed, and it was shown that these strains can induce tumors in experimental trials. The severity of experimental disease differed from those features of the original outbreak, since a less severe disease was recorded.

Marek's disease in turkeys. II. Characterization of the viral glycoprotein B gene and antigen of a turkey strain of Marek's disease virus.

Marek's disease virus (MDV) causes immunosuppression and tumors in chickens. As sporadic cases of Marek's disease (MD) were recorded in turkeys, the antigenic and genomic characteristics of the MDV glycoprotein B (gB) gene and antigen of turkeys were compared to the chicken MDV gB. The whole chicken and turkey gB genes were sequenced and found identical. By immunoblotting of infected-cell culture lysates using chicken convalescent and gB monoclonal antibodies, the antigenic epitopes of the chicken and turkey viruses were found to differ. The turkey MDV had a unique epitope, compared to the chicken MDV and compared with our previous findings. While the chicken MDV had two epitope types, heat-labile but dithiothreitol (DTT)-stable and heat-stable but DTT-labile, the turkey MDV gB epitope is both heat and DTT-labile.

West Nile fever in Israel 1999-2000: from geese to humans.

West Nile virus (WNV) caused disease outbreaks in Israel in the 1950s and the late 1970s. In 1998 an outbreak of WNV in goose farms and evidence of infection in dead migratory birds were reported. Consequently, human diagnostic services for WNV were resumed, including virus isolation, serology, and RT-PCR. Risk factors for infection were assessed by a serological survey in 1999, which revealed a seroprevalence of (a) 86% in people who had close contact with sick geese, (b) 28% in people in areas along bird migration routes, and (c) 27% in the general population. Following two fatal cases in Tel Aviv in September 1999 and one encephalitis case in the southern Eilot region, a regional serological survey was initiated there. The survey revealed two more WNV-associated acute encephalitis cases, an IgG seroprevalence of 51%, and an IgM seroprevalence of 22%. In the summer of 2000, acute cases of WN disease were identified in the central and northern parts of Israel, involving 439 people. The outbreak started in mid-August, peaked in September, and declined in October, with 29 fatal cases, primarily in the elderly. During the outbreak, diagnosis was based on IgM detection. Four virus isolates were subsequently obtained from preseroconverted frozen sera. Sequence and phylogenetic analysis of 1662 bases covering the PreM, M, and part of the E genes revealed two lineages. One lineage was closely related to a 1999 Israeli bird (gull) isolate and to a 1999 New York bird (flamingo) isolate, and the other lineage was closely related to a 1997 Romanian mosquito isolate and to a 1999 Russian human brain isolate.

Variations in biological features of West Nile viruses.

Pathological findings in humans, horses, and birds with West Nile (WN) encephalitis show neuronal degeneration and necrosis in the central nervous system (CNS), with diffuse inflammation. The mechanisms of WN viral penetration of the CNS and pathophysiology of the encephalitis remain largely unknown. Since 1996, several epizootics involving hundreds of humans, horses, and thousands of wild and domestic bird cases of encephalitis and mortality have been reported in Europe, North Africa, the Middle East, Russia, and the USA (see specific chapters in this issue). However, biological and molecular markers of virus virulence should be characterized to assess whether novel strains with increased virulence are responsible for this recent proliferation of outbreaks.

Use of live and inactivated vaccines in the control of West Nile fever in domestic geese.

The recent epizootic of West Nile fever in Israel affected predominantly young domestic geese between three and eight weeks old. Clinically, the birds presented paralytic signs while morbidity and mortality were severe in affected flocks. The condition was encountered from early September through late November on goose farms located throughout the country. Losses incurred by goose flocks were sufficiently great as to warrant investigation of ways to protect young geese against the neurological form of the disease. We have conducted a series of vaccination trials in which three-week old geese were immunized with an attenuated, commercial flavivirus vaccine derived from Israel turkey meningoencephalitis virus (TME). Birds were challenged two weeks later with a low Vero cell passage of West Nile virus by the intracerebral route. In a second group of experiments, inactivated and live TME vaccines were given in tandem at an interval of two weeks and challenged two weeks later. The third vaccination trial was based on West Nile virus (WNV) harvested from infant mouse brain, inactivated with formalin and oil adjuvanted. A single injection given either subcutaneously or intramuscularly resulted in 75% protection of the vaccinated groups, while two injections spaced two weeks apart resulted in 94% protection. Groups of geese, vaccinated at the farms and challenged under controlled conditions in the laboratory, showed levels of protection ranging from 39% to 72% for TME vaccine and 52% and 80% for WNV vaccine. The lower levels of protection are attributable to flocks being affected with intercurrent infections at the time of vaccination.

Diagnosis of turkey meningoencephalitis virus infection in field cases by RT-PCR compared to virus isolation in embryonated eggs and suckling mice.

Turkey meningoencephalitis (TME) is a paralytic epornitic disease of turkeys caused by turkey meningoencephalitis virus (TMEV), an arthopod-borne flavivirus belonging to the Ntaya serogroup VI. A TMEV specific RT-PCR was compared with classical techniques for TMEV diagnosis, which include virus isolation in 8-day-old chicken embryonated eggs and suckling mice, on 17 TME flocks with neurological signs that occurred during the fall of 1997. In 11/17 flocks both the RT-PCR and the virus isolation methods detected virus, in 4/17 flocks a negative diagnosis was obtained by both methods, and two flocks were positive by RT-PCR only. In four flocks RT-PCR only detected virus after inoculation into embryonated eggs or suckling mice. There was a dose response effect in the yield of the RT-PCR product. Direct examination of turkey brains yielded bands of low to medium intensity. Use of RT-PCR after embryo and/or mouse inoculation resulted in products of far greater intensity. Thus, RT-PCR can be successfully used to amplify TMEV RNA in the brains of diseased turkeys but a negative result would require egg and mouse inoculation for enrichment of virus prior to RT-PCR.

A live attenuated West Nile virus strain as a potential veterinary vaccine.

This article reviews the development of two attenuated West Nile virus (WNV) variants, WNI-25 and WNI-25A. These variants have lost the neuroinvasion trait of the parental virus. Attenuation was achieved through serial passages in mosquito cells and neutralization escape from WNV-specific monoclonal antibody. Genetic analysis reveals amino acid changes between the parental and each of the variants. The attenuated variants preserve the ability to replicate in mice and geese and to induce a protective immune response. WNI-25A was found to be a genetically stable virus. This variant was successfully used as a live vaccine to protect geese against a wild-type virulent WNV field isolate that closely resembles the WNV isolated during the 1999 New York epidemic.

Origin of the West Nile virus responsible for an outbreak of encephalitis in the northeastern United States.

In late summer 1999, an outbreak of human encephalitis occurred in the northeastern United States that was concurrent with extensive mortality in crows (Corvus species) as well as the deaths of several exotic birds at a zoological park in the same area. Complete genome sequencing of a flavivirus isolated from the brain of a dead Chilean flamingo (Phoenicopterus chilensis), together with partial sequence analysis of envelope glycoprotein (E-glycoprotein) genes amplified from several other species including mosquitoes and two fatal human cases, revealed that West Nile (WN) virus circulated in natural transmission cycles and was responsible for the human disease. Antigenic mapping with E-glycoprotein-specific monoclonal antibodies and E-glycoprotein phylogenetic analysis confirmed these viruses as WN. This North American WN virus was most closely related to a WN virus isolated from a dead goose in Israel in 1998.

Differential diagnosis of two paralytic conditions affecting young chickens with emphasis on PCR findings.

Two paralytic conditions affecting flocks of young commercial chickens that were indistinguishable histopathologically from lesions caused by Marek's disease virus (MDV) were investigated. The first, transient paralysis, was observed only in broilers while the second condition, a spastic paralysis, was seen in replacement White Leghorn pullets aged between 6 and 16 weeks. PCR examination of DNA extracted from the blood, brain and spleens of affected birds using primers specific for MDV serotype 1 confirmed the diagnosis of MD for transient paralysis only. None of the Leghorn DNA samples was positive by PCR. Some evidence for a genetic predisposition is offered for the neuritis affecting the pullets based on epidemiological considerations.

Booster immunization with a partially purified citrus tristeza virus (CTV) preparation after priming with recombinant CTV coat protein enhances the binding capacity of capture antibodies by ELISA.

Groups of rabbits and young lambs were immunized subcutaneously and intramuscularly with a recombinant citrus tristeza virus (CTV) coat protein (rCTV-CP) antigen. Three weeks after primary immunization the animals were divided into two groups that were boosted either with rCTV-CP or with a partially purified preparation of CTV particles (ppCTV). Twelve and 15 days after the last injection, the animals were bled and the binding capacity of the antisera for CTV detection was examined for capture antibodies by the indirect ELISA. Considerably higher ELISA titers were obtained from animals that were boosted with ppCTV than with rCP. Boosting with partially purified native antigens after priming with recombinant antigens is expected to extend the applicability of the antisera for detecting other structural and non-structural viral antigens by trapping ELISA.

Severe hepatopathy in geese and broilers associated with ochratoxin in their feed.

The feeding of a shipment of imported corn was associated with a severe reduction in growth and increased mortality in geese, and increased mortality in broilers. Pathological examinations revealed hepatopathy, visceral gout and mild nephropathy in geese, and in broilers an hepatopathy, which was often severe, and ascites. Samples of feed from affected geese farms were examined for up to 24 mycotoxins, and ochratoxin was found in 6 of 15 samples at levels up to 930 ng/g. The syndrome was experimentally reproduced by feeding geese and broilers suspect feeds with the natural ochratoxin contamination. It is believed that another, unidentified, mycotoxin was the major cause of the hepatotoxicity, and that ochratoxin served in this case as an indicator of a multi-mycotoxin involvement.

Replication of non-defective reticuloendotheliosis viruses in the avian embryo assayed by PCR and immunofluorescence.

Congenital transmission of reticuloendotheliosis virus (REV) is known to occur when one of the parents in either tolerantly infected or viraemic; however little attention has been paid to replication of the virus within the embryo itself. In this study five non-defective and one replication-defective strain of REV were inoculated into the yolk sac of 6-day-old embryonated eggs. With the exception of one non-defective strain, the others induced embryonic deformation or lesions that were apparent on the day of hatching. Replicative integration of the provirus was determined for several tissues of each embryo by the polymerase chain reaction employing a set of primers specific for the long terminal repeat. Indirect immunofluorescent detection of whole virus was performed on the same tissues in parallel. A variation in replication was found among the strains and between embryos inoculated with the same strain of virus, therefore, the results for each virus strain were expressed as cumulative values. The liver and spleen were the most commonly affected organs, while the bursa of Fabricius and thymus supported viral replication to a lesser degree. Detection of proviral DNA in blood was evidence that the chicks had been congenitally infected. Such an infection would likely lead to tolerance. Isolates REV-S and REV-D were found to replicate to the highest extent, and they also caused lesions in the commercial type of embryonated eggs, while REV-S and mainly the REV-T prototype were pathogenic for the SPF embryonated eggs.

A non-radioactive method for identifying enzyme-amplified products of the reticuloendotheliosis proviral env and LTR genes using psoralen-biotin labelled probes.

A novel polymerase chain reaction (PCR) system based on the env gene of reticuloendotheliosis virus (REV) strain REV-A for the detection of proviral DNA is described. The designed PCR product of 807 bp was identified using an internal probe of 278 bp produced by nested PCR from REV-infected DNA CEF. The env-gene PCR was then compared with the previously described PCR for proviral REV-long terminal repeat and the PCR product served also as the probe. The probes were labelled with the psoralen-biotin system by photoactivation and the southern blot hybridization signal was detected colorimetricaly. The advantages of using a non-radioactive means of probe labelling were demonstrated clearly in that study, as well as the effective labeling of probes with psoralen-biotin and the simple colorimetric method of detection. The env-gene PCR detected all eleven REV strains used in the study. These included three REV prototype strains and eight Israeli REV isolates. Both PCR systems had similar levels of sensitivity.

The immunodominant proteins of reticuloendotheliosis virus.

The antigenic profiles of three REV prototype strains, CSV, SNV and REV-T and eight Israeli isolates were analysed by SDS-PAGE and immunoblotting with convalescent chicken serum, three mAbs, 11A25, 11C237 and 11C100, a rabbit antiserum to REV-T whole virus (Cui et al., 1986) and a rabbit antiserum to REV-A p30 gag protein (Tsai et al., 1985). Under both reducing (+DTT) and non-reducing conditions of SDS-PAGE, a major immunodominant 75-100 kDa band was shared by all strains examined. In contrast to the chicken serum that recognized both continuous and discontinuous epitopes on the 75-100 kDa band of all the isolates, the mAbs and the two rabbit sera behaved otherwise. Only the DTT-resistant epitopes on the 75-100 kDa band of REV-T were recognized by the rabbit antisera and the mAb 11C237, and only the DTT-labile epitopes of REV-T 75-100 kDa antigen were detected by mAb 11C100. The two mAbs 11A25 and 11C237 detected discontinuous epitopes of all the strains except SNV, while the rabbit antisera recognized the discontinuous epitopes on the 75-100 kDa band of all the 11 strains. The rabbit antisera and mAb 11C237 detected additional lower molecular weight proteins and the mAb 11C237 also detected three proteins of high molecular weight under non-reducing conditions only. The p30 antiserum detected the low molecular weight proteins demonstrating their gag gene-encoded identity. From these results we conclude that the major immunogen of REV is the 75-100 kDa protein that contains both continuous and discontinuous epitopes. With this panel of antibodies the eight new isolates appeared to belong antigenically to REV subtype 3 (Chen et al., 1987).

Causes of losses including a Borna disease paralytic syndrome affecting young ostriches of one breeding organization over a five-year period (1989-1993).

Necropsy records and causes of mortality of ostriches up to 3 months old over a 5-year period (1989-1993) are presented. The data relate to one ostrich enterprise that comprises 10 breeding flocks, five rearing farms, and one hatchery. Causes of mortality are classified into nine major categories. The annual mortality percentages of all hatched ostriches over the 5-year period were 61%, 58%, 30%, 29%, and 16.6%, and the most significant cause of death was a paresis syndrome that accounted for 20%, 11%, 16%, 10.1%, and 2% mortality, respectively. Limb deformities and gastroenteritis were the other principal specific causes of mortality. The paresis syndrome was caused by an agent serologically related to Borna disease virus. Brain extracts from paralyzed ostriches, when given orally or intramuscularly to 5-week-old birds, reproduced the clinical signs and microscopic lesions. The mean time to death was less than 3 weeks for the intramuscularly infected group and was almost twice as long for the orally infected group.