Fifty-five serotype 3,4 isolates of Pasteurella multocida, isolated from turkeys dead from fowl cholera, were characterized (fingerprinted) genotypically for comparison with the serotype 3,4 live fowl cholera vaccine principally used in turkeys in California. Twenty-three isolates were obtained from turkeys vaccinated with the M9 live vaccine, and 32 additional isolates were from turkeys not vaccinated for fowl cholera. Methods of characterization included restriction endonuclease analysis of chromosomal DNA and ribotyping, a technique for highlighting restriction site heterogeneity of highly conserved ribosomal RNA genes and associated sequences using a radiolabeled rRNA probe. Eight different genotypes or ribotypes were detected in these isolates by the above methods. Of 23 isolates from M9-vaccinated turkeys flocks, 19 were the same ribotype as M9. Thirty of 32 isolates recovered from unvaccinated turkeys were different ribotypes from M9. The remaining two isolates resembled M9 and were recovered from two different flocks placed in succession on a turkey farm where a flock placed previously had been vaccinated with M9, suggesting interflock transmission. Ribotyping and restriction endonuclease analysis appear to be useful tools to aid in the determination of the role that the live vaccine plays in fowl cholera epidemiology.

The microcidal efficiency of a new pelleting process was evaluated in four trials. Also, different methods of measuring temperature and moisture were compared and attempts were made to determine the influence on efficiency of pH changes occurring during processing. In the new process, the traditional boiler-conditioner was replaced by an Anaerobic Pasteurizing Conditioning (APC®) System. Microcidal efficiency of the APC® System, by itself or in conjunction with a pellet mill, appeared to be 100% against Escherichia coli and nonlactose-fermenters, 99% against aerobic mesophiles, and 90% against fungi. These levels of efficiency were attained when the temperature and moisture of feed conditioned in the APC® System for 4.6 ± .5 min were 82.9 ± 2.4 C and 14.9 ± .3%, respectively. On-line temperature probes were reliable and provided quick, accurate estimates of feed temperature. The near infrared scanner and microwave oven methods of measuring moisture were much quicker but less accurate than the in vacuo method. There were no differences among the pH of samples of raw, conditioned, and pelleted feed.

Five hundred twenty isolates of Pasteurella multocida, collected in California from September 1985 to November 1988, were characterized in the laboratory. Characteristics examined included serotype, capsular type, biotype (subspecies), and possession of plasmid DNA. Three hundred thirty-three isolates recovered from turkeys dying from fowl cholera, 88 isolates from liver turkeys in flocks with fowl cholera outbreaks in the recent past, and 99 isolates from wildlife captured on fowl cholera-outbreak and non-outbreak turkey premises were studied in this manner. Characteristics were fairly homogeneous among isolates, especially those obtained from turkeys. The majority of isolates were serotype 3,4, capsular type A, subspecies multocida, and lacked plasmid DNA. Common serotypes of isolates from turkeys and wildlife sampled on the same premises were noted in eight of 13 cases examined.

Characteristics of Pasteurella multocida isolated from tissues of dead waterfowl and associated avian species found at 23 sites located in northern and central California, from January 1986 through January 1988 are reported. Two hundred ninety five isolates of P. multocida were obtained from 23 avian species. Most of the isolates belonged to the subspecies P. multocida multocida (63%), followed by P. multocida gallicida (37%), and by P. multocida septica (less than 1%). There appeared to be a higher prevalence of P. multocida multocida in Ross' geese (Chen rossi) and Snow geese (Chen coeruleus). All of the isolates belonged to somatic serotype 1, possessed the A capsule type and were susceptible to the 8 antimicrobial agents tested. None contained plasmid DNA.

The microcidal efficiency of a new pelleting process was evaluated in four trials. Also, different methods of measuring temperature and moisture were compared and attempts were made to determine the influence on efficiency of pH changes occurring during processing. In the new process, the traditional boiler-conditioner was replaced by an Anaerobic Pasteurizing Conditioning (APC) System. Microcidal efficiency of the APC System, by itself or in conjunction with a pellet mill, appeared to be 100% against Escherichia coli and nonlactose-fermenters, 99% against aerobic mesophiles, and 90% against fungi. These levels of efficiency were attained when the temperature and moisture of feed conditioned in the APC System for 4.6 +/- .5 min were 82.9 +/- 2.4 C and 14.9 +/- .3%, respectively. On-line temperature probes were reliable and provided quick, accurate estimates of feed temperature. The near infrared scanner and microwave oven methods of measuring moisture were much quicker but less accurate than the in vacuo method. There were no differences among the pH of samples of raw, conditioned, and pelleted feed.

Twenty-five isolates of the bacterium Pasteurella multocida were characterized (fingerprinted) phenotypically and genotypically in order to compare the abilities of various techniques to differentiate strains for epidemiologic studies of fowl cholera. Isolates were obtained over a 16-month period from turkeys dying from fowl cholera (six outbreak flocks) and from wildlife captured on premises with a history of the disease. The characteristics compared included (i) serotype, (ii) subspecies, (iii) antibiogram, (iv) presence of plasmid DNA, (v) restriction endonuclease analysis patterns of whole-cell DNA, and (vi) ribotype. Ribotyping, a method of highlighting DNA restriction site heterogeneity by using an rRNA probe, worked well for differentiating the strains of P. multocida when the majority of the other techniques could not. Ribotyping results correlated directly with and confirmed results obtained from restriction endonuclease analysis. Ribotyping demonstrated the presence of up to three strains of P. multocida in one outbreak flock, recurrence of a single strain in five of the flocks over an 11-month period, and the presence of common strains in turkeys and wildlife on the premises.

One hundred fifty-three isolates of Pasteurella multocida, representing the causative agent of 95% of all known outbreaks of fowl cholera occurring in California meat and breeder turkeys from August 1985 through February 1987, were examined for susceptibility to antimicrobial agents. Of the 153 isolates, 6 were shown to be resistant to one or more antimicrobial agents. Of the six resistant isolates, five contained R plasmids. All but one of the R plasmids were small (6 to 7 megadaltons) and nonconjugative, encoding resistance to tetracycline or kanamycin, streptomycin, and sulfonamides; the other was large (70 megadaltons) and conjugative, transferring resistance to kanamycin, streptomycin, sulfonamides, and tetracycline to P. multocida and Escherichia coli. The three plasmids encoding resistance to tetracycline alone appeared identical.

Swabs of the oropharynges of 801 live turkeys (621 meat birds and 180 breeders), collected from 15 flocks that had experienced an outbreak of fowl cholera and from 12 non-outbreak flocks, were screened for the presence of Pasteurella multocida. Turkeys from outbreak flocks were sampled within 2 to 9 weeks of the outbreak. Forty-nine isolates of P. multocida were recovered from turkeys in 11 of the outbreak flocks, and none were recovered from turkeys in non-outbreak flocks. Isolation rates varied from 0 to 72% of turkeys sampled in a flock. Nineteen isolates were tested for virulence by injecting them intravenously into turkeys, and 14 were lethal. Results demonstrated that for purposes of disease control, meat birds in fowl-cholera-outbreak flocks should be considered carriers of potentially virulent P. multocida for the life of the flock.

The efficacy of a new pelleting process in eliminating naturally occurring Escherichia coli and salmonella from poultry mash was assessed by comparing the microbial load in raw and processed mash. Instead of using steam produced in a boiler, the new process conditioned mash in an Original Vertical Conditioner with steam and other hot gases generated by direct combustion in a Vaporator. E. coli was isolated from 72-100% of samples of raw mash in all trials, and the mean number of colony-forming units of E. coli/g of samples was 6.8 +/- 4.0 X 10(4). Salmonellae (S. senftenberg, S. bredeney, and S. mbandaka) were isolated from 5-10% of samples of raw mash utilized in three of the seven trials. Within limitations of the sampling and analytical tests utilized, the new pelleting process eliminated salmonella from all mash in which the organism was known to be present but eliminated E. coli in only three trials. The process appeared to be 100% effective against both organisms when mash entering the pellet mill from the conditioner had a temperature of 85.7 C and a moisture content of 14.5% and had been retained and treated with steam and hot gases for 4.1 minutes.

The inoculation of Pasteurella multocida (P-1059) intravenously into turkeys increased significantly the plasma prostaglandin (PG) F2 alpha levels to 157% of the control values and the plasma PGE levels to 171% of control values at 3 hr after treatment. At 12 hr, the cloacal temperature of the inoculated birds was significantly higher than that of the control. The leukocyte count of inoculated birds remained unchanged from that of the control. However, the differential leukocyte count shifted in favor of significant increases in heterophils and decreases in lymphocytes and monocytes at 6 and 12 hr after inoculation. This study provides evidence that increases in plasma levels of PGF2 alpha and PGE may be partly responsible for the clinicopathological manifestations of acute fowl cholera.

Samples collected from the oropharynx of wild mammals and birds trapped on 36 turkey farms in California were evaluated for the presence of Pasteurella multocida. A total of 966 animals were collected from 18 premises that had experienced an outbreak of fowl cholera within the past 2-8 months; samples were collected from 16 of these 18 premises within 2-8 weeks of outbreak notification and while the infected flock was still present. A total of 939 animals were trapped from an additional 18 premises that had not reported any outbreaks of fowl cholera within at least 4 months, if ever. Forty-eight isolates of P. multocida, of a variety of somatic serotypes, were recovered from 6 species of mammals and 3 species of birds. On only 2 of 7 premises was the somatic serotype of the isolates obtained from wildlife the same as the isolate obtained from tissues of turkeys that had died of fowl cholera on the same premises. Tests for virulence to turkeys were conducted with 31 of the isolates. Seventeen of these isolates caused mortality in turkeys. Wide ranges in mortality rates and median times to death were observed.

One hundred sixty meat-turkey premises in California were monitored for outbreaks of fowl cholera from August 1985 through July 1986. Nearly 27 million turkeys in 720 flocks were at risk during the year. Fifty-three flocks of approximately 3 million turkeys on 34 different premises experienced confirmed fowl cholera outbreaks. The epidemic curve for the year indicated that the majority of outbreaks occurred in the late summer and fall, particularly in October. The incidence of outbreaks during this time was not significantly associated with seasonal variation in the size of the turkey population. The mean flock age at outbreak was 10.8 weeks, with a range of 5-18 weeks.

An enzyme-linked immunosorbent assay (ELISA) was developed for detecting paramyxovirus type 3 (PMV-3) antibody. This test had a higher sensitivity than the hemagglutination-inhibition test, and no cross-reactivity with various paramyxoviruses or avian pathogens was detected when the sera were tested diluted 1:200. The incidence of PMV-3 infection in California was studied by testing 2037 turkey sera from 174 meat and breeder flocks for the presence of PMV-3 antibody using the ELISA. The age at which the infection occurs was around 5 to 8 weeks for meat flocks and 10 to 12 weeks for breeder flocks. Infection with the PMV-3, as determined serologically, was more frequent than manifested cases of disease, and 95.2% of the flocks aged over 11 weeks had PMV-3 antibody. No typical manifestations of PMV-3 disease (respiratory disease plus drop in egg production) were observed, probably because of the early infection which occurred before laying age.

The incidence of hemorrhagic enteritis (HE) infection in California turkeys was studied by testing 2220 turkey blood samples from 173 flocks for HE virus (HEV) antibody by the enzyme-linked immunosorbent assay (ELISA). Maternal antibody was detected at 1 day of age in all flocks tested, and it vanished after 3 weeks. Acquired HEV antibody appeared at 8 to 10 weeks, and 100% of the meat and breeder turkey flocks were positive after 11 weeks of age. HEV infection occurred earlier in the meat flocks than in the breeder flocks, and it also occurred earlier during summer than during the fall and winter months.