Microcystin and related toxic peptides produced by cyanobacteria (blue-green algae) are potent and selective inhibitors of protein phosphatases 1 and 2A. We adapted existing enzymatic techniques to analyze the liver of rainbow trout after oral administration of hepatotoxic cyanobacteria. Liver tissue was removed 3 and 12 hours after treatment, and phosphatase activity was determined in liver extracts, using a specific phosphoprotein substrate. In all samples from fish exposed to toxic cyanobacteria, phosphatase activity was suppressed, whereas the control enzyme, lactate dehydrogenase, present in the same liver extract, was not affected by cyanobacteria. Thus, experimental poisoning by hepatotoxic cyanobacteria resulted in an abnormally low ratio of phosphatase to lactate dehydrogenase activity in the liver extracts. These results indicate that specific inhibition of phosphatases 1 and 2A may provide a useful diagnostic tool to determine the early effects of cyanobacteria toxic peptides directly in liver samples from poisoned animals. Although this test was developed with rainbow trout, it should be possible to extend the analysis of liver phosphatase activity to other species, including sheep and cattle, which are frequently affected by hepatotoxic cyanobacteria.

Growth of Pasteurella haemolytica A1 in RPMI 1640 medium containing 0.5% bovine serum albumin (BSA) for 2.5 hours enhanced culture supernatant leukotoxic activity [30,700 +/- 12,900 toxic units/ml, compared with leukotoxic activity of culture supernatants produced in RPMI 1640 medium alone (120 +/- 40 toxic units/ml)]. Gel filtration chromatography of the leukotoxic activity from RPMI 1640 medium supernatants in buffer containing 50 mM NaCl indicated a single leukotoxic activity peak (peak I) eluting near the gel resin molecular mass exclusion limit (estimated molecular mass of approx 8,000 kd). In contrast, culture supernatants produced in RPMI 1640 plus bovine serum albumin medium (RPMI + BSA) had peak I and 2 additional leukotoxic activity peaks (peaks II and III) with estimated molecular mass of approximately 80 and < 30 kd, respectively. All leukotoxic activity peaks were composed of approximately 100-kd molecular mass leukotoxin protomer, as determined by sodium dodecyl sulfate-polyacrylamide gel electrophoresis and immunoblotting with a monoclonal antibody against leukotoxin. Subjecting culture supernatant leukotoxic activity produced in RPMI + BSA to gel filtration chromatography in buffer containing 500 mM NaCl or 6M urea resulted in detection of only a single leukotoxic activity peak with estimated approximate molecular mass of 250 and 800 kd, respectively. These findings suggest that P haemolytica exists as a high molecular mass aggregate with low leukotoxic activity which, in the presence of BSA, partially disaggregates to multiple toxin forms with enhanced leukotoxic activity. Some of these leukotoxin forms interact with dextran-based gel resins at low ionic strength.

Naturally acquired turbinate atrophy in rabbits was associated with Pasteurella multocida infection. Several in vitro and in vivo studies were conducted to document toxin production from P. multocida isolates and to determine the relation of toxin to atrophic rhinitis in rabbits. Ten isolates of P. multocida serotype A:12 were obtained from adult New Zealand White rabbits with noninduced atrophic rhinitis. Specific-pathogen-free rabbits inoculated intranasally with isolates of P. multocida developed rhinitis and turbinate atrophy. However, inoculation with filtrates of the same bacteria failed to induce turbinate atrophy. Cytotoxicity was observed in assays, using bovine embryonic turbinate cell cultures with extracts of P. multocida, but not in agar overlay cytotoxicity assays, using bovine embryonic turbinate, bovine embryonic lung, or Vero cell cultures, or in a sandwich ELISA, using monoclonal antibodies to purified P. multocida toxin. Thus, turbinate atrophy was experimentally reproduced in rabbits with isolates of P. multocida, but toxin was only detected in vitro by cell culture assay of P. multocida extracts.

To assess the role of enterotoxigenic (ETEC), verotoxigenic (VTEC), and necrotoxigenic (NTEC) Escherichia coli in cattle with diarrhea, 1,524 colonies of E coli isolated from 197 calves with diarrhea and from 112 healthy controls were investigated for production of heat-labile and heat-stable enterotoxins, verotoxins, and cytotoxic necrotizing factors (CNF1 and CNF2). The ETEC were isolated from only 2 (1%) calves with diarrhea and from 5 (4%) healthy controls. In contrast, VTEC and NTEC that produced CNF2 were frequently identified. The VTEC were isolated from 18 (9%) calves with diarrhea and from 21 (19%) healthy cattle (P < 0.05), whereas NTEC that produced CNF2 were detected in 39 (20%) ill calves and in 38 (34%) controls (P < 0.01). Therefore, VTEC and NTEC that produced CNF2 were isolated significantly more frequently from healthy than diseased calves. Serogroups to which VTEC belonged differed considerably from the O groups involved with NTEC. Although, VTEC belonged to 18 serogroups, only 4 (O26, O103, O113, and O157) accounted for 56% (25 of 45) of verotoxigenic strains. The NTEC that produced CNF2 belonged to 26 serogroups; however, 64% (69 of 108) were from 6 serogroups (O1, O3, O15, O55, O88, and O123). Our results are compatible with cattle being a reservoir of VTEC that are pathogenic for human beings and with ETEC being an unusual cause of bovine colibacillosis in Galicia (northwestern Spain). Furthermore, results of this study indicate that VTEC and NTEC that produced CNF2 may be part of the normal intestinal flora of cattle.

Tonsils of 10 calves were inoculated with Pasteurella haemolytica (PH) and the degree of colonization was followed by collecting sequential tonsil wash specimens. Tonsils were colonized for at least 3 weeks after instillation of PH into the tonsillar sinus. Calves with colonized tonsils responded with serum and nasal secretion antibody responses to PH and to leukotoxin. Pasteurelia haemolytica was detected in nasal mucus specimens of 2 calves during the week after inoculation of the tonsils, but all other specimens were culture-negative. Infectious bovine rhinotracheitis virus-induced respiratory tract disease 25 days later did not elicit a population increase of PH in the tonsils, and did not elicit shedding of PH in nasal mucus.

An experimental model for subclinical edema disease was developed in weanling pigs. In multiple experiments, 3-week-old pigs were weaned, then inoculated intragastrically with 10(10) colony-forming units of an SLT-IIv-positive strain of Escherichia coli originally isolated from a pig with edema disease (principals). Control pigs were inoculated with a nonpathogenic E coli strain. Of 39 principals, 8 developed clinical edema disease within 14 days after inoculation. However, 20 of 21 principals that did not develop clinical signs of edema disease, but were submitted for necropsy examination at 14 days after inoculation, had characteristic vascular lesions of edema disease. Vascular lesions, found principally in ileum and brain, consisted of segmental necrosis of myocytes in the tunica media of small arteries and arterioles. None of the pigs inoculated with a nonpathogenic strain of E coli developed edema disease or vascular lesions. None of the principals necropsied at 2 days after inoculation had vascular lesions. Development of vascular lesions by 14 days after inoculation was used as the end point for detecting subclinical edema disease in the model.

A double-antibody sandwich ELISA for detection of antibodies directed against the exotoxin of Corynebacterium pseudotuberculosis, the cause of caseous lymphadenitis (CL) in small ruminants, was developed. A concentrated exotoxin was used. For interpretation of ELISA results, these sera were tested: sequentially obtained sera of C pseudotuberculosis-inoculated goats and sheep that were monitored for 68 weeks; sequentially obtained sera from 80 goats of 3 flocks with CL; sera from 652 goats of 7 flocks without CL; sera from 160 sheep of 4 flocks without CL; and 2,265 caprine and 208 ovine sera submitted for diagnostic testing. Data regarding the infection status and history of 10,454 of the 23,302 animals were collected after testing; most of these were goats that had been part of a CL control program. Specificity and sensitivity of the ELISA were nearly 100%. Subsequently, 31,978 animals from which no data on infection status of flocks had been collected were then tested. It was concluded that the ELISA is a useful diagnostic test for CL eradication programs. Sera with doubtful or inconclusive ELISA results were examined by use of immunoblot analysis. Proteins from C pseudotuberculosis culture supernatant were separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis and blotted onto nitrocellulose. Six proteins with molecular mass of 68, 65, 39, 38, 31, and 29 kDa reacted with sera from goats and sheep with experimentally induced or naturally acquired infection. Immunoblot analysis was valuable in classifying sera with doubtful or inconclusive results by ELISA.

The Escherichia coli heat-stable enterotoxin (STb) is the most prevalent toxin associated with diarrheagenic E coli isolates of porcine origin. Unequivocal biological activity of this toxin has been observed only in swine intestine. In this study, when endogenous protease activity was blocked with soybean trypsin inhibitor, intestinal secretion was stimulated by STb in jejunal loops of rats, mice, calves, and rabbits. Compared with pigs, rats, mice, and calves, rabbits were relatively insensitive to STb. These data demonstrate that the activity of STb is not a species-specific toxic activity; there is species variation in sensitivity to STb, and some common laboratory animals may have potential to be used to measure biological activity Of STb.

Compounds that prevent chloride transport in membrane vesicles have been tested for in vivo activity against the effects of intestinal secretory agents. Chloride channel blockers including diphenylamine-2-carboxylate, 4-acetamido-4'-isothiocyanostilbene-2,2'-disulfonate, 5-nitro-2-(2-phenylethylamino)benzoic acid, and alpha-phenylcinnamic acid were tested for effects on jejunal or ileal secretion in weanling pigs. Secretion was studied in ligated intestinal loops in a control state, during exposure to secretory concentrations of theophylline, and after prior treatment with cholera toxin. Increases in net fluid flux induced by either theophylline or cholera toxin were not prevented by adding chloride channel blockers into the intestinal lumen. Channel blocker concentrations that reduced chloride transport by > 50% in pig jejunal brush border vesicles did not cause significant changes in unidirectional blood to lumen chloride flux measured in situ. Several routes of administration of the specific chloride channel blocker alpha-phenylcinnamate failed to reduce fluid secretion induced by theophylline. Chloride channel blocker effectiveness appears to be significantly different between in vitro and in vivo experimental models. In contrast to the chloride channel blockers, loperamide significantly reduced net fluid and chloride flux in ileal loops secreting fluid in response to theophylline. Antagonism of the production or actions of second messenger by loperamide was more effective than the chloride channel blockers in reducing conductive chloride transport associated with intestinal secretion.