Water quality changes during establishment and maintenance of nitrification in four 70-liter recirculating systems each containing 20 channel catfish (Ictalurus punctatus) (total weight 150 g) are described. Establishment of nitrification took 33 days. During establishment of nitrification, all catfish in the systems died. Death appeared to be related to nitrite concentrations greater than 2.5 mg/liter NO 2 -N. The pH declined with progressive nitrification. When the pH was less than 6.0, nitrification ceased. Within 2 wk of introducing the catfish, total bacterial counts stabilized at 10 6 organisms per milliliter. Counts of lactose fermenters and aeromonads fluctuated around 10 4 organisms per milliliter when fish were in the system and decreased markedly after the fish were removed. After establishment of nitrification, 40 channel catfish (total weight 100 g) were maintained successfully for 37 days.

Streptococcus iniae is an important bacterial pathogen of fish, causing up to 50% mortality in stocks, which has recently been associated with human infections. To determine whether S. iniae isolates from humans and fish are similar, the present authors examined the biochemical profiles and genetic relatedness of these isolates by random amplified polymorphic DNA (RAPD) analysis and repetitive primer polymerase chain reaction(REP PCR). The biochemical profiles differentiated between the human and fish isolates of S. iniae using pyrrolidonyl arylamidase, arginine dehydrogenase, ribose, beta-glucoronidase and glycogen as markers. These biochemical results suggest that the fish and human S. iniae isolates are genetically different. However, RAPD and REP PCR do not have the discriminatory power to differentiate between these streptococcus isolates using five different RAPD primers and BoxA primer.

An avirulent, wild-type avian Escherichia coli (E. coli Av) was electrotransformed with a plasmid coding for the production of microcin 24 (pGOB18) and was designated E. coli AvGOB18. The transformant inhibited the growth of seven serotypes of Salmonella commonly associated with colonization and contamination of poultry products and seven strains of E. coli O157:H7 in the in vitro colicin/microcin assay. The transformant did not inhibit the replication of multiple isolates of Listeria monocytogenes or Campylobacter jejuni in similar assays. The transformant is nonconjugative, indicating that the plasmid would not be transmitted to other intestinal microflora in the environment. The transformant also survived in sterile tap and deionized water incubated at 25 C and 37 C in the laboratory for 30 days and was recovered from drinkers and birds in in vivo floor pen studies. In in vivo studies, E. coli AvGOB18 did not colonize the intestinal tract of broiler chicks when given as a single or multiple dose and did not reduce the Salmonella load in the broilers. But Salmonella typhimurium was reduced significantly in the intestinal tracts of broiler chickens when E. coli AvGOB18 was administered continually in the water supply.

Species-specific polymerase chain reaction (PCR) primers have been developed for the identification of the causative agents of warmwater and marine finrot in fish: Flavobacterium columnare (Flexibacter columnaris) and Flexibacter maritimus. Differences in gene sequence in the bacterial small-subunit (16S) ribosomal RNA (rRNA) were used to design the species-specific PCR primers. The previously reported species-specific PCR primers Psy1 and Psy2 for the identification of Flavobacterium psychrophilum (Flexibacter psychrophila),the causative agent of coldwater finrot, were also used to develop a speciation scheme for all three bacterial finrots. These three primer sets were successful in discriminating among yellow-pigmented bacteria as well as in speciating the three major pathogenic flexibacteria to fish. The primer sets were designed to produce uniquely sized subproducts of 16S rRNA for each species: Flavobacterium psychrophilum (1,100 base pairs, bp), F. columnare (800 bp), and Flexibacter maritimus (400 bp). These primers were shown to correctly speciate field isolates in double-blind experiments (P = 0.01).

The purpose of this study was to determine the rate at which resistance developed in avian coliform bacteria when exposed to nalidixic acid, sarafloxacin, or enrofloxacin. In in vitro studies, the rates of mutation of avian isolates of Escherichia coli and Salmonella were determined following nalidixic acid, sarafloxacin, or enrofloxacin pressure. The rates of mutation were similar for nalidixic acid and sarafloxacin, whereas a lower rate of mutation was seen after enrofloxacin pressure. In in vivo studies, the quinolones were administered in the drinking water to broiler chickens at a concentration of 40 ppm for five consecutive days. Samples of feces were inoculated onto appropriate media and the frequency of resistance was determined. The frequency rates of resistance to nalidixic acid and sarafloxacin were similar. Enrofloxacin-medicated birds did not develop enrofloxacin-resistant coliform bacteria. The in vitro and in vivo data appear to correlate.

The nearly complete small subunit 16S ribosomal RNA (rRNA) gene sequence amplified by polymerase chain reaction was determined for Flavobacterium psychrophilum (formerly Flexibacter psychrophilus) by using automated nucleotide sequencing. The sequence was found to be 1,465 base pairs (bp) in length, a size consistent with previously determined sequences for 14 other bacterial species from various taxa, including the yellow-pigmented bacteria. Sequence signatures confirmed that this organism was a member of the bacterial division Bacteroides-Flavobacterium. Parsimonious and additive phylogenetic trees were constructed with homology, and pairwise evolutionary distances, were used to estimate phylogenetic relationships. Data show that F. psychrophilum, F. columnare, and Flexibacter maritimus are closely related, have a common descent, and represent a distinct group within the division Bacteroides-Flavobacterium. This group also included other organisms from the genera Flavobacterium and Cytophaga. Further, Flavobacterium aquatile, the type species For the genus Flavobacterium, was also determined to be a member of this "Flavobacterium-Cytophaga-Flexibacter subcomplex." This supports previous assertions that the type strain of Flavobacterium should be changed to a more representative species such as Flavobacterium breve.

Three methods of infecting channel catfish Ictalurus punctatus with Edwardsiella ictaluri were compared. Groups of fish were exposed to the bacterium by intraperitoneal (IP) injection, bath immersion with stress, or bath immersion without stress. We determined infection rate by culturing tissues for the presence of E. ictaluri over a 3-week period. Fish in the IP-injected and bath-stressed groups had higher rates of infection than the bath-unstressed fish. Three weeks after exposure, the cumulative percentage of fish that cultured positive for E. ictaluri was 46.6% in the IP-exposed group, 52.7% in the bath-stressed group, and only 15.7% in the bath-unstressed group. Tissues from the IP-injected and the bath-stressed fish also cultured positive at an earlier time postexposure. Edwardsiella ictaluri were isolated from liver, head kidney, brain, spleen, and trunk kidney at some time during the 3-week sampling period. Anti-E. ictaluri antibody levels determined by agglutination and enzyme-linked immunosorbent assay indicated significantly different antibody titers depending on method of exposure to E. ictaluri. The bath-stressed group showed measurable antibody levels earlier and maintained higher titers than the bath-unstressed group throughout the experiment. This study confirms that stressed f.sh are more susceptible to infection with E. ictaluri than unstressed fish.

Two groups of newly hatched chicks received standard broiler rations supplemented with 0.6% acid (either butyric or lactic) continuously starting at one day of age, while a control group received unsupplemented feed. At two days of age, chicks were inoculated by oral gavage with 1 mL of 106 CFU Salmonella typhimurium. Ceca were obtained from six chicks of each group at day 7, 14, and 21. After seven days there were no differences in level of Salmonella colonization among the supplemented and control groups. After fourteen days, lactic acid decreased colonization by 1.6 logs (P < .05), while no significant decrease was observed with butyric acid. After twenty-one days, both acids had significantly reduced intestinal colonization; butyric and lactic acid produced 1.67 and 1.95 log reductions, respectively. There were no significant differences in body weights among the groups at day 21. It is important to demonstrate that the level of Salmonella in the digestive tract and ceca of broiler chickens can be significantly reduced by the addition of selected carbohydrates to the water or feed. Reduced levels in the intestines means less salmonellae to be spread around in the grow house and during processing, and therefore less contaminated final product.

Previous studies have revealed a reduction of cecal Salmonella carriage from feeding either carbohydrate or short-chain fatty acids (SCFAs). This in vitro study presents a profile of the relative SCFA content of the ceca when chicks are fed an unmedicated diet with 2.5% carbohydrate. Subsequent incorporation of these acids into culture medium was used to demonstrate their antagonistic activity toward in vitro growth of Salmonella typhimurium. Commonly found concentrations of SCFAs based upon the above findings reduced in vitro Salmonella growth by at least 50%, and 10 x concentrations inhibited growth more than 80%. An explanation of the mechanism(s) involved in growth reduction is offered.

An outbreak of salmonellosis in a gerbil colony was investigated. The clinical, bacteriologic, and pathologic findings are reported. Clinical signs included an occasional sudden death, depression, emaciation, dehydration, rough hair coat, and testicular enlargement. Not every sign was observed in every infected gerbil. At necropsy, 11 animals had lesions consistent with salmonellosis. Histopathologic lesions consisted of interstitial pneumonia, hepatic and splenic necrosis, meningitis, and suppurative orchitis. Splenic and intestinal amyloidosis were also noted. Salmonella, group D, was recovered from gerbil feces, a container in which adult mosquitos were reared, filarial inoculum, and a cockroach. An epizootiologic investigation led to salmonella-infected cockroaches as the possible source of animal contamination via mosquitos and the subsequent filarial inoculum.

Indwelling urinary catheters with a closed urine collection system were maintained in 30 male cats for 3 days after induction of irritant cystitis. All cats received subcutaneous fluids during the 3 days the catheters were in place. The effects of four different treatment regimens on urinary tract infection rates, incidence of urethral obstruction, and development of urinary tract lesions over a 10-day period were compared with results in a nontreated group. Treatments were 1) amoxicillin for 5 days PO; 2) prednisolone for 5 days PO; 3) both amoxicillin and prednisolone for 5 days PO; and 4) dimethylsulfoxide (DMSO) for 3 days intravesicularly. Euthanasia was done before the end of the 10-day experimental period if the cats had two bouts of urethral obstruction or if the cats became uremic for causes unrelated to urethral obstruction. Seven cats were euthanatized before the conclusion of the experiment. These cats had been treated with prednisolone, prednisolone and amoxicillin, or DMSO. All cats that received amoxicillin alone or no therapy survived the 10-day period. Mortality was due to repeated urethral obstruction or to uremia associated with pyelonephritis or papillitis. Urinary tract infection rate was similar in all groups. The group treated with prednisolone alone had the highest incidence of renal infection. Inflammatory lesions in the lower urinary tract were similar in all groups. In conclusion, persistent urinary tract infection often develops in cats with cystitis after indwelling urethral catheterization even when closed systems of urine drainage are used.(ABSTRACT TRUNCATED AT 250 WORDS)

Biochemical and electrophoretic studies were done on bacterial isolates collected from rainbow trout Oncorhynchus mykiss from 2 private hatcheries in the Pacific Northwest, USA. Five isolates originated from post-spawning mortalities and morbidities; 2 other isolates were obtained from reproductively active 17 mo old rainbow trout. After S-35-methionine-labeled protein lysates were prepared from each isolate, correlation coefficients of electrophoretic profiles were calculated for each pair of isolates using an automated microbiological identification system (AMBIS). The Gram-positive isolates appeared to represent 1 or possibly 2 genera: Carnobacterium and Lactobacillus. Five isolates were identified as C. piscicola and two (106891,106892) did not correspond to any previously described species. Comparisons based upon correlation coefficients calculated for the 2 unidentified isolates and for biochemically similar strains, including C. piscicola (ATCC 15434), L. alimentarius (ATCC 29643) and L. homohiochi (ATCC 15434), resulted in values no greater than 0.792. This dissimilarity suggests that the 2 unidentified isolates represent a different biotype of C. piscicola or a previously undescribed species of Lactobacillus.

A quantitative microtiter method for determining the degree of complement resistance or sensitivity of microorganisms is described. The microtiter method is compared with a quantitative automated system and the standard plate-count technique. Data were accumulated from 30 avian Escherichia coli isolates incubated at 35 C with either chicken plasma or heat-inactivated chicken plasma. Analysis of data generated by the automated system and plate-count techniques resulted in a classification of the microorganisms into three groups: those sensitive to the action of complement; those of intermediate sensitivity to the action of complement; and those resistant to the action of complement. Although the three methods studied did not agree absolutely, there were statistically significant correlations among them.

Two-day-old chicks were orally inoculated with 1 ml of Salmonella typhimurium (10(5) colony-forming units/ml) and divided into four groups. Three groups were fed 2.5% carbohydrates starting on day 1 (arabinose, galactose, and lactose), while the fourth group served as the control. Ceca were obtained from each group at 7, 14, and 21 days. At the end of 14 days, all three carbohydrates statistically reduced Salmonella recovery. However, lactose failed to reduce recovery between day 14 and day 21. Arabinose and galactose continued to show significant reductions of recovery through 21 days. No statistical difference was found between Salmonella recovery from whole ceca (with cecal material) and inverted ceca (washed free of cecal material).

A study of the microbiological flora isolated from cultures of normal and lesional skin tissue samples collected from 19 bowhead whales (Balaena mysticetus) over a 4 yr period is presented. These cultures were obtained from 30 tissue samples (17 normal, 13 lesion) and 248 swab samples (157 normal, 91 lesion). Seven hundred-thirty bacterial and yeast isolations were made (285 normal, 445 lesion). Distribution revealed that 56% of the gram positive bacterial isolates, 75% of the gram negative bacterial isolates and 64% of the yeast isolates recovered were associated with lesional skin. It was found that 80% of one group of Corynebacterium sp. isolates, 90% of the Acinetobacter sp. isolates and 94% of the Moraxella sp. isolates were associated with lesional skin. Although the primary yeasts recovered were Candida spp., they were found on both normal and lesional skin. Enzymatic assays of isolates from normal and lesional skin demonstrated production of enzymes capable of causing necrosis. The majority of the microorganisms recovered were facultative anaerobes and many of them could be considered potential pathogens of mammalian hosts.

Yersinia enterocolitica (Serotypes 0:3 or 0:8), Yersinia frederiksenii, Yersinia kristensenii, or Yersinia intermedia along with 10(8) cells of each of three extraneous organisms (Escherichia coli, Enterobacter aerogenes, Pseudomonas fragi), all commonly found on market poultry, were inoculated into five enrichment media followed by streaking onto 11 plating media to determine the most-efficacious combinations for future surveys or assessment studies. For Yersinia enterocolitica (0:8), infrequent recoveries were made using yeast extract-rosebengal-bile oxalate sorbose broth and phosphate-buffered saline (4 C) followed by plating onto pectin, DNase-Tween 80 (polyoxyethylene sorbitan monooleate)-sorbitol, MacConkey-Tween 80, or cefsulodin-irgasan-novobiocin (CIN) agars. With Y. enterocolitica (0:3), recoveries were most frequently made using phosphate-buffered saline, sorbitol bile (incubated for 17 days) and yeast extract-rosebengal-bile oxalate sorbose broth followed by plating onto pectin, CIN, bismuth sulfite (Difco Laboratories, Detroit, MI), or modified Rimler-Shotts agar. For Y. frederiksenii, Y. kristensenii, and Y. intermedia, incubation in sorbitol bile for 17 days or in yeast extract-rosebengal-bile oxalate sorbose broth, followed by plating onto CIN, pectin, DNase-Tween/80-sorbitol, cellobiose-arginine-lysine agar, or MacConkey-Tween 80 agar yielded the most-frequent recoveries. Overall, the CIN and pectin agars performed best for the recovery of the Yersinia bacterium; the modified selenite broth and the bismuth-sulfite plating agars were unsatisfactory in the present study.

Abstract. Motile Aeromonas spp. virulent for fish were studied with regard to their adhesion profile. Electron microscopy demonstrated the presence of fimbriae (pili) on Aeromonas cells regardless of virulence potential. The results show no significant correlations between ability to haemagglutinate, yeast cell co‐agglutination and virulence. All strains expressed mannose‐sensitive haemagglutinin activity against guinea pig erythrocytes. Using four types of red blood cells no characteristic haemagglutinin pattern related to virulence could be discerned. Expression of surface haemagglutin(s) on Aeromonas hydrophila appears to be medium dependent; strains grown in liquid media demonstrated enhanced haemagglutination activity. Both virulent and avirulent strains had in vitro epithelial cell adhesive capabilities. Cell surface characteristics measured by agglutination in acriflavine and stability after boiling indicated that most virulent strains agglutinated in the presence of acriflavine, but not all sedimented after boiling. The ability of 10 selected strains of A hydrophila to grow in normal pooled catfish serum was determined. Only 17% of the virulence variations can be explained by their sensitivity to serum.

Sixty ready-to-cook broiler carcasses obtained from several local supermarkets were tested for the presence of Yersinia enterocolitica and other Yersinia species. In the present study, the authors used two enrichment broths, yeast-extract/rosebengal-bile oxalate sorbose (YER-BOS) and phosphate-buffered saline with a postenrichment KOH treatment (PBS-KOH), and two plating media, cefsulodin-irgasan-novobiocin (CIN) and pectin agar. Yersinia organisms were found on 34 of 60 carcasses (56.7%) and Y. enterocolitica, on 16 of 60 carcasses (26.7%). There was no significant difference between CIN and pectin agar; however, PBS-KOH yielded a significantly higher (P less than or equal to .05) detection rate than YER-BOS, regardless of the plating media used. In addition to Y. enterocolitica, Y. frederiksenii and Y. intermedia were also isolated from the market broilers. None of the Y. enterocolitica isolates were found to be presumptively virulent.