Max Brugh

Preparation of Inactivated Oil-Emulsion Vaccines with Avian Viral or Mycoplasma Antigens

The influence of the composition of water-in-oil emulsions on their physical characteristics was determined by preparing experimental emulsions with various water-to-oil ratios and various emulsifiers. Emulsions containing Tween 80 in the aqueous phase and Arlacel A or Arlacel 80 in the oil phase were lower in viscosity than emulsions containing only an oil-phase emulsifier. Viscosity decreased as the concentration of oil increased. Oil-emulsion vaccines prepared with aqueous- and oil-phase emulsifiers had low viscosity, were stable for more than 12 weeks at 37 C, and induced a marked primary antibody response in chickens.

In ovo vaccination of chicken embryos with experimental Newcastle disease and avian influenza oil-emulsion vaccines.

Inactivated oil-emulsion (OE) Newcastle disease (ND) and avian influenza (AI) vaccines were injected into 18-day-old white rock (WR) and white leghorn (WL) chicken embryos to evaluate their immunologic efficacy and their effects on hatchability. Embryonating eggs were inoculated at 1.5 inches depth with various vaccine volumes and antigen concentrations. Serum hemagglutination-inhibition (HI) titers were first detected in chickens at 2 wk posthatch. Protection against morbidity and mortality was demonstrated in all of 10 chickens vaccinated as embryos and challenged with viscerotropic velogenic ND virus at 53 days of age and also in all of eight in ovo- vaccinated chickens challenged with highly pathogenic AI virus at 34 days of age. All of five unvaccinated control chickens for each respective ND- and AI-vaccinated group died. In pooled groups from successive hatches, the hatchability of WR or WL embryos injected with 100 microliters of vaccine was not significantly different (P > 0.05) from unvaccinated hatchmate controls when needle gauges of 22, 20, and 18 were used. Seroconversion rates of chickens vaccinated as embryos ranged from 27% to 100% with ND vaccination and 85% to 100% for AI vaccination. For ND, geometric mean HI titers of chickens per vaccine group ranged from 11 to 733, and in pooled groups, the range was 49 to 531. Titers for AI vaccine groups ranged from 156 to 1178. This study demonstrated that acceptable hatchability, seroconversion rates, and protective immunity can be attained with in ovo inoculation of ND or AI OE vaccines if the vaccines are prepared with sufficient antigen and administered properly.

An arg-lys insertion at the hemagglutinin cleavage site of an H5N2 avian influenza isolate

Recent isolations of H5N2 subtype avian influenza (AI) viruses in North America have raised questions concerning their origin, transmission to commercial poultry, and potential for virulence. One ratite-origin isolate of low pathogenicity, A/emu/TX/39924/93 (H5N2), was subjected to a procedure that rapidly selects and/or amplifies highly pathogenic (HP) strains. The resulting highly virulent derivative had an altered hemagglutinin (HA) gene containing an additional six nucleotides at position 970–975 in the HA1 coding region. This resulted in an arg-lys insertion near the proteolytic cleavage site of the HA protein. The remainder of the HA sequence differed by an additional seven amino acids from the parent. The HA precursor of the derivative, but not the parent, was readily cleaved during replication in cell culture without addition of trypsin. In experimentally infected chickens, the derivative produced lesions typical of highly pathogenic avian influenza. A reverse transcriptase-polymerase chain reaction (RT-PCR) primer set was designed to amplify exclusively from molecules with the inserted six nucleotides. The set yielded product only from the selected derivative samples and not the parent. Thus, the levels of the HP variants in the parent stock were undetectable, or the insertion occurred rapidly during the selection process.

Assessment of the Ability of Ratite-Origin Influenza Viruses to Infect and Produce Disease in Rheas and Chickens

Pathobiologic characteristics were determined for three mildly pathogenic (MP) ratite-origin avian influenza viruses (AIVs). Ratite-origin AIVs produced respiratory disease in rheas, and virus was reisolated from oropharyngeal and cloacal swabs on days 2-6 postinoculation. Inoculation of two ratite-origin AIVs in the upper respiratory tract of chickens resulted in viral infections, but the mean chicken infectious dose (CID50) for A/emu/Texas/39924/93 (H5N2) (Emu/Texas) virus was 500-fold lower than the CID50 for the A/rhea/North Carolina/39482/93 (H7N1) virus. In ovo and in vivo passage of the MP parent Emu/Texas isolate resulted in emergence of a highly pathogenic (HP) variant that had high plaquing efficiency in chicken embryo fibroblast cultures and was highly lethal in chicken pathotyping tests. This variant virus produced gross lesions in chickens similar to those reported for other HP AIVs. These findings demonstrated that ratite-origin AIVs can produce significant clinical disease in rheas and have a realistic potential for interspecies transmission to domestic poultry. Furthermore, HP variants can emerge from MP H5 ratite-origin AIVs if introduced and allowed to circulate in chicken populations.

Pathogenicity and Diagnosis of H5N2 Mexican Avian Influenza Viruses in Chickens

Chickens were inoculated with one of five H5N2 Mexican-origin avian influenza virus (AIV) isolates to determine their pathogenicity for chickens and to determine the ability of routine virologic and serologic tests to detect infections. In laboratory infections, three AIVs, H5/94, M5/94, and J12/94, produced sporadic illness and death and were categorized as mildly pathogenic. Q1/95 produced illness and death in all inoculated chickens and was categorized as highly lethal and highly pathogenic (HP). P11/94B commonly produced clinical illness, but deaths were infrequent. During the presence of clinical signs, oropharyngeal swabs were superior for isolation of AIV, but cloacal swabs were more successful after disappearance of clinical signs. Agar gel precipitin (AGP) serologic test was superior for detecting AIV infection during the clinical phase, but AGP and hemagglutinin inhibition tests were equally effective in detecting infections after recovery from clinical illness. Passage of P11/94B parent stock and selected 14-day-embryo-passed AIVs in adult hens resulted in emergence of some HP AIV derivatives. The hemagglutinin of Q1/95 and P11/ 94B parent stock and derivative AIVs had an identical proteolytic cleavage site of.... Pro-Gln-Arg-Lys-Arg-Lys-Thr-Arg-Gly, consistent with AIVs of high pathogenicity. However, no consistent differences were identified in the sequence of the hemagglutinin gene to explain the discrepancy in lethality patterns of the P11/94B AIVs. This suggests that genes other than the hemagglutinin impact the full expression of high lethality of Mexican-origin AIV infections in chickens.

Emergence of Highly Pathogenic Virus during Selective Chicken Passage of the Prototype Mildly Pathogenic Chicken/Pennsylvania/83 (H5N2) Influenza Virus

The prototype mildly pathogenic A/chicken/Pennsylvania/21525/83 (H5N2) avian influenza virus, which was isolated more than 5 months before the emergence of highly pathogenic virus in the major 1983 Pennsylvania outbreak, was examined for the presence of minority subpopulations of highly pathogenic virus. Selective serial passage of the parental mildly pathogenic virus in leghorn hens did not lead to recovery of highly pathogenic virus. However, several highly pathogenic reisolates were recovered from hens inoculated with either of two mildly pathogenic virus clones selected for their ability to efficiently produce plaques in trypsin-free chicken embryo fibroblasts. Unlike the parental virus, these reisolates caused high mortality in chickens and produced postmortem lesions typical of highly pathogenic avian influenza. Electrophoretic mobilities of the hemagglutinin glycoproteins of the highly pathogenic derivatives resembled those of the prototype highly pathogenic A/chicken/Pennsylvania/1370/83 (H5N2) virus isolated in October 1983. These results suggest that unrecognized subpopulations of highly pathogenic virus may have infected Pennsylvania chickens for several months before emerging as the clinically manifest component of the virus population.

Effects of chicken embryo age on time to death following infection by avian influenza viruses : implications for distinguishing highly pathogenic isolates

When white leghorn (WL) chick embryos ranging in age from 8 to 13 days were inoculated with a variety of avian influenza virus (AIV) isolates, strain-specific differences in embryo mean death times (MDT) were observed. Non-highly pathogenic (nHP) strains killed 8 or 9 day-old embryos much more rapidly than 12 or 13 day-old embryos. Highly pathogenic (HP) strains, however, were less sensitive to embryo age resulting in similar MDTs in both older and younger embryos. These observations were consistent over a broad range of virus doses for both HP and nHP strains. When a HP derivative of H5N2 AIV was compared to its nHP parent, the derivative killed older embryos more rapidly than the parent virus, while MDTs in younger embryos were the same for both parent and derivative. The two strains further exhibited clear differences in the structure of their respective hemagglutinin, a previously described pathogenicity determinant for this virus. Thus it may be possible to readily demonstrate the HP phenotype in AIV strains based on MDT measurements in WL embryos.

Simulation of Maternal Immunity by Inoculation of Immune Yolk Preparations into the Yolk Sac of 1-Day-Old Chickens

Yolk harvested from eggs laid by hens hyperimmunized with killed Newcastle disease virus (NDV) was inoculated into the yolk sac of 1-day-old specific-pathogen-free (SPF) chickens. Serum hemagglutination-inhibition antibody titers reached maximum levels 1 to 4 days after yolk inoculation and declined at a rate similar to that reported for naturally acquired maternal antibody. Expected levels of immune interference were observed when yolk-inoculated chickens were vaccinated with a conventional oil-emulsion NDV vaccine. These results show that yolk-sac inoculation with yolk antibody is a suitable approach for producing maternally immune chickens for laboratory studies.

Amantadine resistance among hemagglutinin subtype 5 strains of avian influenza virus.

Several avian influenza virus strains of hemagglutinin subtype 5 were assayed for sensitivity to the antiviral drug amantadine. Most strains exhibited little sensitivity to the drug as measured by plaque reduction. The A/Chicken/Scotland/59 (CS59), however, was highly sensitive, making it easily distinguishable from the other H5 strains. Drug sensitivity of the viruses was also assayed in chicken embryos. The in ovo patterns of amantadine sensitivity differed from those detected in cell culture. The CS59 isolate could not be distinguished from all the other strains on the basis of its response to amantadine in ovo. Although amantadine protected chickens inoculated with CS59 from morbidity and mortality, drug-resistant viruses were readily isolated from the infected birds. As found with other amantadine-resistant variants, the structure of the matrix gene was altered in the resistant isolates. These results demonstrate that amantadine resistance is widespread among avian influenza viruses of the H5 subtype, that drug sensitivity in cell culture does not necessarily reflect responses to amantadine in ovo and in vivo, and, as previously found, amantadine-resistant derivatives of H5 strains may be isolated from birds protected by the drug.

Re-evaluation of the Pathogenicity of A/Chicken/Alabama/75 (H4N8) Influenza Virus

Avian influenza (AI) virus A/chicken/Alabama/7395/75 (H4N8), a putatively non-pathogenic virus associated with a self-limiting outbreak of severe disease in commercial layers, was selectively passed in chickens or in cell cultures and then in chickens to determine whether virus with increased pathogenicity would emerge. When 20 derivatives of the parental virus were each inoculated intranasally and intratracheally in leghorn hens, mortality rates ranged from zero (0/24) to 25% (6/24); mortality was 4% (1/24) for hens inoculated with the parental virus. Many virus reisolates (51/144) from hens that died exhibited high pathogenicity, killing at least six of eight intravenously inoculated 4-week-old chickens. Most derivatives examined produced plaques in trypsin-free cell cultures more efficiently than the parental virus, but the highest plaquing efficiencies observed (10%) were lower than would be expected (100%) for highly pathogenic subtype H5 or H7 AI viruses. These results confirm that the Alabama H4N8 virus can acquire increased pathogenicity upon passage in chickens and suggest that it may have acted alone in producing the severe disease observed in laying chickens in Alabama.