Classification of the spoilage flora of raw and pasteurized bovine milk, with special reference to Pseudomonas and Bacillus

Eighty-one bacterial strains isolated from refrigerated raw milk, 124 from pasteurized milk and cream stored at 5 degrees C and 7 degrees C, and 19 type and reference strains of Pseudomonas spp. and Bacillus spp. were characterized by numerical phenotypic analysis. Data were processed with simple matching (SSM) and Jaccard (SJ) coefficients, and UPGMA clustering. Fourteen clusters of Gram-negative bacteria were formed at SJ = 79% (SSM = 90%). Raw milk was exclusively spoilt by Gram-negative bacteria, the majority of which were Pseudomonas fluorescens biovar I, Ps. fragi, Ps. lundensis and Ps. fluorescens biovar III. Minor groups in raw milk included Enterobacteriaceae spp. and Acinetobacter spp. Pasteurized milk was spoilt by essentially the same Gram-negative organisms in 65% (5 degrees C) and 50% (7 degrees C) of the cases. The phenotypic characteristics of Gram-negative bacteria are given. Bacillus polymyxa (both temperatures) and B. cereus (only at 7 degrees C) were responsible for 77% of samples spoiled by the Gram-positive organisms. Minor milk spoilage groups included other Bacillus spp. and lactic acid bacteria. All Bacillus spp. grew fermentatively in milk, and most strains denitrified. It is suggested that: (i) industrial recontamination tests of pasteurized milk are directed against Pseudomonas; (ii) milk is stored at 5 degrees C or lower to avoid growth of B. cereus; and (iii) the significance of gas-producing and nitrate/nitrite-reducing Bacillus strains is recognized in cheese production.

Occurrence of different serotypes of Erysipelothrix rhusiopathiae in retail pork and fish

Retail pork (38 samples), cod (10 samples) and herring (10 samples) were obtained from 12 stores in the area of Lund in southern Sweden during September and October 1990. Erysipelothrix rhusiopathiae was isolated from 50% of the pork samples, 60% of the cod samples and from 30% of the samples from herring. Serotype 2 dominated on retail pork as well as on fish samples constituting 53% of the pork isolates (10 strains) and 33% of the cod isolates (2 strains). All E. rhusiopathiae isolates originating from herring were serotype 2 (3 strains). Serotypes 1b, 6, and 8 were isolated from retail pork only (6, 2 and 1 strains, respectively). Serotype 5 was isolated from cod only (3 strains) and so was serotype 9 (1 strain). The public health hazards with the occurrence of virulent strains of Erysipelothrix rhusiopathiae in retail pork and fish are discussed.

Occurrence of different serotypes of Erysipelothrix rhusiopathiae in retail pork and fish

Retail pork (38 samples), cod (10 samples) and herring (10 samples) were obtained from 12 stores in the area of Lund in southern Sweden during September and October 1990. Erysipelothrix rhusiopathiae was isolated from 50% of the pork samples, 60% of the cod samples and from 30% of the samples from herring. Serotype 2 dominated on retail pork as well as on fish samples constituting 53% of the pork isolates (10 strains) and 33% of the cod isolates (2 strains). All E. rhusiopathiae isolates originating from herring were serotype 2 (3 strains). Serotypes 1b, 6, and 8 were isolated from retail pork only (6, 2 and 1 strains, respectively). Serotype 5 was isolated from cod only (3 strains) and so was serotype 9 (1 strain). The public health hazards with the occurrence of virulent strains of Erysipelothrix rhusiopathiae in retail pork and fish are discussed.

Numerical taxonomy of fluorescent Pseudomonas associated with tomato roots

The phenetic taxonomy of 110 fluorescent bacterial strains, isolated from the roots of tomatoes and other plants was numerically studied through 97 features including 69 assimilation tests. Thirty-two reference strains of various Pseudomonas spp. were additionally included. The strains clustered into 16 clusters at the 74% similarity level when using Jaccard similarity coefficients. Almost all field strains belonged to the P. fluorescens/P. putida-complex while none clustered with P. syringae and allied bacteria. The biovar II branch, as well as the newly described biovar VI of P. fluorescens, made up 55% and 20% respectively, of the field strains; two % were allocated to P. fluorescens biovar I and three % to biovar IV. Eleven % of the root associated strains were designated P. putida; six strains were biovar A, three strains biovar B while four strains could not be referred to any known biovar. The continuum within the P. fluorescens/P. putida-complex as well as the taxonomic status of the six biovars of P. fluorescens and the three biovars of P. putida are discussed.

Occurrence of Erysipelothrix rhusiopathiae on pork and in pig slurry, and the distribution of specific antibodies in abattoir workers

Strains of Erysipelothrix rhusiopathiae isolated at 19 pig farms serving a certain abattoir, and on pork and in workers of this abattoir were studied. Mouse-pathogenic E. rhusiopathiae was found in pig slurry from two farms (11%). The strains belonged to serotypes 7 and 16 (both from the same farm) or were untypable. In pig slurry from the abattoir lairage only serotype 2 strains were found and all were pathogenic to mice. Mouse-pathogenic E. rhusiopathiae strains of serotype 2 were also recovered from 25 pork lions (25%). A mouse-pathogenic E. rhusiopathiae (serotype 2) strain was isolated from one of the 16 hand infections of slaughterhouse workers. The E. rhusiopathiae strains were phenotypically grouped by the API 50 CH system. Variations were demonstrated for the different serotypes. In 20 of 138 workers antibodies against E. rhusiopathiae were found; 14 had increased levels of IgG antibodies, seven had increased levels of IgM antibodies and one had an increased level of both.

The microbial flora of herring fillets after storage in carbon dioxide, nitrogen or air at 2 degrees C

Herring fillets from the Baltic Sea were stored in glass vessels in air, nitrogen or carbon dioxide (CO2) at 2 degrees C and the microbial development was studied. The microbiological shelf-life of the herring (the time to reach 10(7) organisms/g) was prolonged by a factor of 3.5 in CO2 as compared to air. The corresponding factor in nitrogen was 1.5. The microflora of fresh and spoiled herring was classified. The initial microflora was dominated by coryneforms, Flavobacterium spp., Moraxella-like organisms and Pseudomonas spp. The spoilage-flora in air (after 9 d) was dominated by Pseudomonas spp. and Moraxella-like organisms, and in nitrogen (14 d) Enterobacteriaceae, Vibrionaceae and Lactobacillus spp. were dominant. Homofermentative Lactobacillus spp. were the only organisms isolated from fillets stored in CO2 (28 d). It was concluded that storing fresh fish in pure CO2 at low refrigeration temperatures is a method with industrial potential. The method (1) improves the microbiological stability and (2) reduces the microbiological health hazards of the fish.

Numerical taxonomy of psychrotrophic pseudomonads

The taxonomy of 218 psychrotrophic pseudomonad strains (200 field strains from meat and 18 type and reference strains) was numerically studied by 174 biochemical and physiological tests. All strains were Gram-negative rods, oxidative positive and motile by means of one or more polar flagella. The strains clustered into 15 groups, of which 9 were regarded as major clusters. The major clusters were designated as Pseudomonas fragi (112 strains), P. fluorescens biotype III (7 strains), P. fluorescens biotype I (16 strains). P. aureofaciens/chlororaphis (3 strains), P. fluorescens biotype II (3 strains), P. putida biotype I (4 strains), Alteromonas putrefaciens (10 strains) and Aeromonas hydrophila biotype I (5 strains). One major cluster, containing 21 strains (cluster 2), was left unassigned. The phenotypic data indicate that this cluster might represent a new species. The P. fluorescens/P. putida complex matched closely the descriptions of Stanier et al. (1966), but the two largest clusters (1 and 2) were not in agreement with any species described in the eighth edition of Bergey's Manual of Determinative Bacteriology. Cluster 1 included the type strain (ATCC 4973) of the hitherto incompletely described P. fragi. A simplified scheme for the separation between P. fragi, P. fluorescens, P. putida and cluster 2 is presented.