Application of Neferm-Test to Identification of Psychrotrophic Bacteria Isolated from Foods

Veterinarni Medicina 40, 1995, 151-155

Identification of gramnegative nonfermentative bacteria by traditional methods requires much labor and time consumption. Therefore in this study we tested the use of an available commercial diagnostic kit and processing of its results by computerized numerical identification systems. Identification involved 83 gramnegative, psychrotrophic, nonfermentative, oxidase-positive bacteria isolated from just slaughtered and deep-frozen chickens. The strains were isolated from primary cultures in Pseudomonas F and Endo agar, Violet Red Bile agar (IMUNA, OXOID) and King B agar (King et al., 1954), under two cultivation regimes (room temperature for three days and 7 degrees C for 10 days). A commercial kit for diagnostics of nonfermentative bacteria in clinical microbiology under the name NEFERM-test (Lachema a.s., Brno) containing 12 biochemical assays was used for strain identification: GLU, AGR, IND, ESL, SCI, PHS, URE, MAN, XYL, MLT, LAG, NIT. All these assays were conducted parallelly by traditional methods. Additional tests involved the use of OXI and ONPG commercial strips (Lachema a.s., Brno), gelatin and Tween 80 hydrolysis (Pacova and Kocur, 1984) and fluorescein production (King et al., 1954). These identification systems were used to process the results of all tests: index and differentiation table (Lachema a. s., Brno) and numerical computerized systems TNW (Czech Collection of Microorganisms, Bmo) and IDENTI (Z. Svoboda, Jihlava). Various shortened procedures, identification keys and systems are used to speed up identification of gramnegative nonfermentative bacteria. Available commercial identification kits (API NE20) in form of microassay have been developed particularly for diagnostics of bacteria from clinical materials, which applies to NEFERM-test of the Czech make. We tested the use of NEFERM-test for identification of psychrotrophic bacterial contaminants isolated from raw material of animal source. Processing of the results of the assays obtained from a NEFERM-test plate only and reactions of ONPG strips brought about absolutely unsatisfactory identification of the investigated bacteria in all evaluated types of identification systems. Tab. I shows that the IDENTI system, TNW system and differentiation table had 35%, 31% and 11% abilities, respectively, to identify the tested strains correctly. Nortje et al. (1990) reported on similar experience, that means on a large failure of the API NE20 system to identify psychrotrophic bacteria in meat. On the basis of literary references (Molin and Ternstrom, 1982, 1986) and of our own practical experience we supposed that the tested psychrotrophic nonfermentative strains isolated from meat would be mainly fluorescent pseudomonas difficult to discriminate by NEFERM-test. Three additional tests (gelatin, Tween 80, King B) were chosen to facilitate differentiation of our isolates. Applying the results of the latter tests, identification was very satisfactory. It involved the identification system TNW only since only this test can be used together with additional tests. Tab. II shows the resultant identification. It is apparent from this procedure that generic identification was successful in 100% out of 83 tested strains, and specific identification was successful in 84.3% of the strains. TNW program classified 13 isolated strains as intermediary, but further tests confirmed correct specific identification in eight (9.7%) out of these strains. It is to note that the additional tests we used highly expressly increased the percentage of successful specific identification (94%). Tab. III shows an overview of isolated and identified bacteria which documents that the species of Pseudomonas genus, P. fluorescens 54 strains (65.1%) and P. fragi 13 strains (15.7%), were present most frequently in the set of psychrotrophic bacteria. Species representation of the bacteria we isolated from chickens (89% pseudomonads) is in keeping with literary data on meat and poultry contamination. Molin and Ternstrom (1982, 1986) tested 200 psychrotrophic pseudomonas isolated from meat (mainly from pork and beef). Unlike our study, the strains P. fragi (58%), P. fluorescens (24%) and P. lundensis (18%) were mainly represented in their set of bacteria. Even though the identification ability of the NEFERM-test used alone was very low, the results obtained with the kit cannot be evaluated negatively. Tab. IV shows a comparison of the results of the particular assays from a NEFERM-test microplate with the results of traditional methods. It documents a very high goodness if fit of the investigated tests ranging from 82 to 100%. To a lesser extent there occurred some falsely positive and/or negative results in comparison with the traditional tests, particularly in sugar fermentation where largest discrepancies were observed in lactose having 18.1% of falsely negative results in NEFERM-test. As mentioned by Rojickova (1994) the results of sugar fermentation often contain discrepancies

FULL PAPER on request
« back