Assessment of interplay between UV wavelengths, material surfaces and food residues in open surface hygiene validation.

The use of UV-visible radiation for detecting invisible residue on different surfaces as a means of validating cleanliness was investigated. Wavelengths at 365, 395, 435, 445, 470 and 490 nm from a monochromator were used to detect residues of beef, chicken, apple, mango and skim milk. These were on three surfaces: aluminium, fibre re-enforced plastic (FRP; Q-Liner®) and stainless steel, pre- and post a cleaning step using commercial detergent. The area covered by residues as detected by specific wavelengths was compared statistically. The sensitivity of the wavelengths for detection differed significantly (p < 0.05) for various residues depending on the material surfaces. Generally, wavelengths 365-445 nm were consistently able to illuminate all residue before cleaning, though sensitivity varied, while 490 nm showed more of the surface structural features instead of residue. The 365-395 nm wavelengths were significantly more sensitive (p < 0.05) for detecting beef and chicken residues on aluminium and stainless steel both before and after cleaning. The 435-445 nm wavelengths were significantly more sensitive for detecting apple and mango residues on the FRP both before and after cleaning. It is important when UV-systems are used as real-time tools for assessing cleanliness of surfaces that the surface materials being illuminated are taken into account in the choice of lamp wavelength, in addition to expected residue. This will ensure higher confidence in results during the use of UV-light for real-time hygiene validation of surfaces.

Screening for Bacillus subtilis group isolates that degrade cyanogens at pH 4.5-5.0.

Cyanogenic food crops abound in nature with important crops like cassava forming the staple food for over half a billion people. Detoxification by hydrolysis of cassava cyanogenic glycosides often involves acid fermentation, and in some of these processes Bacillus species are encountered. Forty Bacillus spp. (20 Bacillus subtilis, 11 Bacillus licheniformis, 7 Bacillus sonorensis, 2 Bacillus cereus) isolated from acid fermented primary starters to produce Gergoush, a Sudanese fermented snack, were screened for their ability to grow and to hydrolyze linamarin, the major cyanogen found in cassava at pH levels below 5.0; also the cyanogen amygdalin was assessed. The B. subtilis isolates grew in both HCl and lactic acid environments from pH 4.5-6.0 while being able to break down the cyanogenic glycosides. The B. licheniformis and B. sonorensis isolates grew and degraded cyanogens at pH 5.0 in a HCl environment, while two B. cereus isolates used in the study showed no breakdown reaction under all conditions tested. One B. subtilis isolate was observed to have substrate specificity between the breakdown of linamarin and amygdalin. We conclude that some Bacillus spp. isolates are important in the microbiological breakdown of cyanogens in cassava fermentations even at pH 4.5-5.0 though further investigations are required.

Attachment behaviour of Escherichia coli K12 and Salmonella Typhimurium P6 on food contact surfaces for food transportation.

The role of cargo container lining materials aluminium, a fibre reinforced plastic (FRP) and stainless steel in bacterial cross contamination during transport was assessed. For this, attachment and detachment of Escherichia coli K12 and Salmonella Typhimurium P6 on the three surfaces in the absence or presence of residues were evaluated. Observations were correlated with water contact angles of the materials (hydrophobicity) and roughness profile (R(a)). Attachment of the organisms was negatively correlated to the hydrophobicity of the three materials with r = -0.869 and -0.861 for E. coli K12 and S. Typhimurium P6 respectively. Correlation with roughness average was poor; r = -0.425 and -0.413 respectively for E. coli K12 and S. Typhimurium P6. Presence of residue caused significant reduction (p < 0.05) in the levels of bacteria attached to all materials, but made attached bacteria significantly more difficult to detach by either of two rinsing systems from all three surfaces. Explanation for these observations could be made in part from scanning electron micrographs which showed significantly more bacteria sitting on patches of residue when it was introduced to the surfaces, compared to the bare material sections of the same surfaces. We report these observations for the first time for aluminium and the FRP material and in part for stainless steel. The S. Typhimurium P6 strain also had significantly higher level of attachment than the E. coli K12 strain. Our findings show that food residue and soils affect the extent and amount of bacteria attaching to abiotic surfaces by altering the surface contact properties for the bacteria. Physicochemical properties like hydrophobicity appear to be a better basis for material selection for hygienic design of containers, than the traditional use of R(a).

Identification and safety evaluation of Bacillus species occurring in high numbers during spontaneous fermentations to produce Gergoush, a traditional Sudanese bread snack.

Gergoush is a naturally fermented Sudanese Bread snack produced in three fermentation steps (primary starter, adapted starter and final dough), followed by three baking steps for a half to one hour at above 200 °C. This study examines the microbiota of two sets of fermentations performed at a traditional production site in Khartoum, Sudan in 2006 and 2009, respectively. In 2006 four different milk/legume based primary starters (faba bean, chick pea, lentil and white bean) were sampled in order to enumerate and identify the Bacillus at species or group level. In 2009 specific focus was on the enumeration and safety evaluation of the dominant Bacillus cereus group species occurring during various Gergoush productions (including the three fermentations steps and after baking). In 2006, the primary starters contained Bacillus spp. in the order of between 7.7 and 8.1 log(10) CFU/g. Species identifications were performed by M13-PCR typing using the Escherichia coli phage M13 derived primer PM13 combined with internally transcribed 16-23S rRNA PCR, 16S rRNA gene and gyrA or gyrB gene sequencing, and selected phenotypic tests. Depending on the legume used, 40-68% of the isolates were identified as B. cereus sensu lato, 16-27% as Bacillus licheniformis, 8-32% as Bacillus subtilis and 4-20% as Bacillus sonorensis. During the second set of fermentation trials performed in 2009, the Bacillus spp. and B. cereus occurred in numbers of between 7.7-9.9 and 6.1-7.8 log(10) CFU/g, respectively, while no bacteria were detected after baking. A total of 180 B. cereus sensu lato isolates from four different primary starters, adapted starters and final doughs were further identified as B. cereus sensu stricto (118 isolates) and Bacillus thuringiensis (62 isolates). The safety of Gergoush was evaluated based on the counts and toxin gene profiles of the dominant B. cereus species. Considering that no bacteria survived the baking process, and that the cereulide synthetase gene cesB involved in the production of the heat stable emetic toxin cereulide was not detected in any of the investigated B. cereus isolates, the results indicate, that Gergoush produced at the traditional production site is safe for human consumption. This study is the first to identify the Bacillus of Gergoush to species level, and it is the first to perform a safety evaluation of the product, based on the dominant B. cereus species.