Antimicrobial Resistance and Multilocus Sequence Types of Finnish Campylobacter jejuni Isolates from Multiple Sources.

Antimicrobial susceptibility was determined for 805 domestic Campylobacter jejuni isolates obtained from broilers (n = 459), bovines (n = 120), human patients (n = 95), natural waters (n = 80), wild birds (n = 35) and zoo animals/enclosures (n = 16) with known multilocus sequence types (MLST) for 450 isolates. The minimum inhibitory concentration (MIC) values for erythromycin, tetracycline, streptomycin, gentamicin and the quinolones ciprofloxacin and nalidixic acid were determined with the VetMIC method. MICs were compared with MLST types to find possible associations between sequence type and resistance. The proportions of resistant isolates were 5% (broilers), 6.3% (natural waters), 11.4% (wild birds), 11.6% (human patients), 16.7% (bovines) and 31.3% (zoo). The most common resistance among the human and bovine isolates was quinolone resistance alone while resistance to streptomycin alone was most often detected among the broiler isolates and tetracycline resistance was most commonly observed in the wild bird, water and zoo isolates. No or negligible resistance to erythromycin or gentamicin was detected. In all data, 12/26 of the tetracycline-resistant isolates were also resistant to streptomycin (P < 0.001) and the clonal complex (CC) ST-1034 CC showed a high proportion of 75% (9/12) of tetracycline-resistant isolates, most originating from the zoo and broilers with closely associated MLST types from these sources. No association between quinolone resistance and MLST type was seen. The low percentage of resistant isolates among the domestic Campylobacter infections is most probably due to the long-term controlled use of antimicrobials. However, the higher percentage of tetracycline resistance observed among the zoo isolates could present a risk for zoo visitors of acquisition of resistant C. jejuni. The resistance pattern of tetracycline and streptomycin most often found in ST-1034 CC could indicate a common resistance acquisition mechanism commonly present in this CC. Overall, MLST typing was found to be a useful method in recognition of potential genetic lineages associated with resistance.

Antimicrobial resistance in Campylobacter coli selected by tylosin treatment at a pig farm.

Limited knowledge is available regarding the dynamics of macrolide resistance under farm conditions with natural Campylobacter populations. We examined the dynamics of antimicrobial resistance in Campylobacter coli at a large pig farm. Faeces were sampled from untreated sows and piglets (n=57), weaned pigs treated with tylosin (n=68) and pigs of the same group 3-5 weeks after withdrawal of tylosin (n=15). Additionally, 48 weaned pigs were sampled after tylosin had not been administered for 7 months at the farm. MICs for seven antimicrobials were determined, isolates were genotyped by PFGE and mutations conferring macrolide resistance were identified. Resistance to at least one antimicrobial agent was higher (P<0.001) in the isolates from the treated pigs (30 of 56) than in those from the untreated animals (2 of 40). Resistance to ciprofloxacin, erythromycin, nalidixic acid and streptomycin was higher (P<0.05) in the isolates from the treated pigs than in those from the untreated animals. All 14 erythromycin-resistant isolates studied (MIC ≥ 512 µg/ml) contained mutation A2075G in 23S rRNA. Resistance against at least one antimicrobial was significantly lower (P<0.05) when tylosin had not been administered for 7 months. Resistance to erythromycin and streptomycin also decreased (P<0.05). PFGE analysis revealed a change of genotypes induced by tylosin treatment. In conclusion, tylosin treatment of pigs selected for a high-level of resistance to erythromycin and resistance to ciprofloxacin, nalidixic acid and streptomycin also increased in C. coli isolates within a few days.

Comparison of tests for detection of beta-lactamase-producing staphylococci.

Penicillin resistance identification tests are important in veterinary medicine. Six enzyme assays and a PCR test were compared for the detection of beta-lactamase production or the beta-lactamase gene in 175 staphylococcal isolates. We conclude that the PCR test and two nitrocefin-based assays can be recommended for routine clinical use.

A microbiological six-plate method for the identification of certain antibiotic groups in incurred kidney and muscle samples.

A microbiological method was developed for group level identification of antibiotics in incurred kidney and muscle samples from cattle and pigs. The method was composed of six test bacterium-plate growth medium combinations and the result was recorded as a profile of growth inhibition zones. The sample profiles were compared to two sets of references: one constructed with standard antibiotic solution profiles, and the other with these combined with profiles of microbiologically and chemically identified residues from incurred samples. The algorithm employed in profile comparison located the minimal sum of absolute pairwise differences over the tests, with the addition of a number of experimentally observed intra-test criteria. Chemical identification and quantitation of incurred residues was based on liquid chromatography. The method identified penicillin G as a penicillinase sensitive penicillin, enrofloxacin and ciprofloxacin belonging to fluoroquinolone group, and oxytetracycline belonging to tetracycline group. Each of these residues was microbiologically identified below the Maximum Residue Limit (MRL) for kidney tissue. Combining sample profiles with the standard reference data set did not enhance the resolution. Microbiological and chemical identification test results were in good agreement. The results of this study show that a microbiological identification method is a useful tool in preliminary characterisation of antibiotic residues in animal tissues.

Microbiological and chemical detection of incurred penicillin G, oxytetracycline, enrofloxacin and ciprofloxacin residues in bovine and porcine tissues.

Incurred penicillin G, oxytetracycline, enrofloxacin and ciprofloxacin residues in bovine and porcine muscle and kidney samples were analysed by microbiological and chemical methods, the former using Bacillus subtilis BGA as a test organism on agar media of pH 6, pH 7.2 and pH 8 and the latter using liquid chromatography. Least squares fits between the logarithms of the chemically obtained concentrations of the antimicrobials and the widths of the inhibition zones were used to estimate the inhibition zone widths corresponding to the maximum residue limit concentrations. In vitro sensitivities were determined with standard antimicrobial solutions. The results indicate that if B. subtilis BGA is used as a test organism, muscle tissue cannot be used as test material for screening oxytetracycline, enrofloxacin and ciprofloxacin residues on the plates used in this study, while penicillin G can be screened from muscle tissue. Because of the numerous factors causing or increasing variation in the analysis, the inhibition zone caused by a given antibiotic concentration cannot be predicted precisely. Therefore, a positive agar diffusion test needs to be confirmed chemically. If a kidney sample gives a positive agar diffusion test result, the antimicrobial concentration in a muscle sample from the same carcass should be checked chemically.

Microbiological and chemical identification of antimicrobial drugs in kidney and muscle samples of bovine cattle and pigs.

Microbiological and chemical identification of antimicrobial drug residues was attempted in 95 kidney and 76 muscle samples from 58 cattle, 36 pigs and one horse which had revealed kidneys positive to an inhibitor test. Information on pre-slaughter medication with one antimicrobial drug was available for 63% of the carcasses. Microbiological identification was performed by agar diffusion using 17 or 18 combinations of eight test bacteria, varying medium pH and three substances blocking the action of certain antimicrobials. Sample activity patterns compiled from inhibition zone diameters on test plates were compared with those obtained with standard antimicrobial solutions both visually and by locating the minimal sum of absolute pairwise differences over the tests. Chemical identification of residues was based on liquid chromatography. In kidney samples containing one microbiologically-identified antimicrobial the two methods gave fully consistent results with tetracyclines (15/15) and fluoroquinolenes (8/8). Preparation and storage of the kidney samples before chemical analyses appeared to influence the chemical identification of penicillin G. The results were consistent in 37 of the 41 samples stored without homogenization at -70 degrees C. The residue was identified by chemical means only in six and neither microbiologically nor chemically in four kidney samples with information on pre-slaughter medication. The same residue as in the kidney samples was identifiable microbiologically in 41% of the muscle samples of the same carcasses. The results show that the microbiological method is well suited for identification of antibiotic residues. They indicate further that an enhanced resolution with a reduced combination of plates is attainable.