Abstract: This study analyzed the viability of the main micro-organisms involved in outbreaks of bacterial food borne diseases together with two fungal strains experimentally inoculated into six different commercial dry extruded pet foods during six months. Growth of all micro-organisms analyzed decreased along the experimental period, indicating that dry extruded pet food is not an adequate substrate for microbial development and it is safe as pet food since most pathogenic micro-organisms did not adapt to this substrate. However, Salmonella enterica serovar Typhimurium (S. typhimurium) remained viable for six months even in coexistence with other micro-organisms. Viability of the different micro-organisms showed significant differences depending on the levels of fat and acid in the kibble coating. These results indicate that the kibbles characteristics could help to predict the dynamics of microbial contamination.
INTRODUCTION
Recently, some outbreaks of severe gastroenteritis have been attributed to pet food contaminated by different micro-organisms (Finley et al., 2006), being Salmonella sp. the most remarkable (Cave et al., 2002; CDC, 2005).
Salmonella transmission from pets or their food to their owners, mainly children, has been described in raw pet food (Finley et al., 2006), treats (Clarck et al., 2001) and pet chews (Wong et al., 2007). This fact represents a big hazard for public health and further knowledge in this field is required in order to assure the wellbeing of both pets and their owners.
Some other microorganisms, such as Escherichia coli, Clostridium perfringens and Staphylococcus aureus can be transmitted the same way and cause health problems (Stogdale and Diehl, 2005). Although, some studies have been performed to characterize the microbiota in commercial food (D`Aoust, 1978; Wojdat et al., 2004, 2005; Bontempo, 2005; Medina et al., 2007), remains unclear whether dry pet food is a good or poor substrate for microbial growth and survival (Bueno et al., 2001; Wesse et al., 2005; Strohmeyer et al., 2006). In addition, it is important to point out that another hazard in pet food is the presence of mycotoxins, (Magan, 2006) and the dynamics of contaminated ready-processed products with those micro-organisms has not yet been studied.
The objective of the present study was to analyze the viability of different microorganisms related to outbreaks of bacterial food borne diseases and 2 mycotoxin-producer fungal strains experimentally inoculated into six commercial dry extruded pet foods for dogs and cats during 6 months. This way, the microbiological behavior can be assessed in this type of pet food so that safety for both pets and owners can be evaluated. These products had different composition, but apart from the ingredients, other technological parameters like the levels of added acid, coated fat, colouring and size and shape, were taken into account in the analysis to observe if they have any influence in the viability and growth dynamics of the micro-organisms.
MATERIALS AND METHODS
This study was conducted from June to December 2007, in the Laboratory of the
Research Group of Applied and Environmental Microbiology, in the Faculty of
Veterinary, at the Universitat Autònoma de Barcelona (Spain).
| Table 1: | Composition and other characteristics of the assayed commercial products |
| * Adjusted to 100% in each formulation | |
Selection of samples and strains: Six commercial dry extruded pet foods belonging to different commercial categories (named A to F) were selected for this study taking into account their composition and characteristics such as size and shape of the kibble, added acid, level of coated fat and colour, so that a wider range of products could be compared in the study (Table 1). Most dry extruded pet foods have a best before date which varies between 12 and 18 months and most market products have a rotation of less than 6 months, therefore 6 months was chosen as the experimental period.
The bacterial strains used to prepare the inoculuas were those mentioned that according to current EU legislation on Food Hygiene EC 1774/2002 (2002) play a main role in the outbreaks of bacterial food borne diseases: Salmonella typhimurium ATCC 13311, Escherichia coli ATCC 9637, Clostridium perfringens ATCC 9856 and Staphylococcus aureus ATCC 9144. In addition, 2 mycotoxin-producer fungal strains (Aspergillus flavus ATCC 76688 and Fusarium moniliforme ATCC 200567) (Magan, 2006) were also inoculated.
The samples were inoculated according to official methods for qualitative and quantitative studies of microbial viability in foodstuffs outlined by the Association of Official Analytical Chemists.
Preparation and distribution of the samples: Two bags of 5 kg of each commercial product were received in the laboratory in their original packaging. For each commercial product two different batches were inoculated in separate studies to check that results were repeatable. Each commercial pet food was divided into 78 containers (50 g per container), 26 containers per group. Every 2 weeks, two containers of each sample were taken at random from each group to analyse for growth and viability of the analyzed micro-organisms.
Group 1 was the negative control group and only 1 mL of sterile ¼ Ringer solution was added to these samples to maintain the same level of humidity as in the rest of the groups. Viability and growth of 8 microbiological parameters were investigated: E. coli, C. perfringens, S. aureus and S. typhimurium for being involved in most of the food borne diseases and A. flavus and F. moniliforme, for their ability of producing mycotoxins. However, total aerobic mesophile bacteria (TAMB) and Enterobacteriaceae were also analysed since are considered two important hygiene indicator bacterial groups.
Group 2 was inoculated with 1 mL of a suspension with a concentration of 1.0x107 Colony Forming Units per mL (cfu mL-1) of each one of the selected micro-organisms, so the effect of coexistence among them could be analyzed. This suspension was used based on the results previously obtained by plate-counting. The micro-organisms investigated in this group were the same than those in group 1. Group 3 was inoculated with 1 mL of a suspension of a titrated solution of 1.0x107 cfu mL-1 of S. typhimurium only.
All samples were kept at room temperature, in a cool, dry place throughout the duration of the study to mimic their normal commercial storage.
Culture conditions for the micro-organisms evaluated: In group 1, half of the container (25 g) was placed into a sterile flask containing 225 mL of sterile Ringer ¼, as the starting dilution. Serial 10-fold dilutions were carried out up to five dilutions in sterile Ringer ¼. For each micro-organism 0.1 mL of each dilution was surface-sown in triplicates in their corresponding plates. For TAMB, each dilution was surface-sown onto TSA plates (Liofilchem, Teramo, Italy), for Enterobacteriaceae McConkey agar (Liofilchem) plates were used and for E. coli Coli ID plates (BioMérieux, Marcy l`Etoile, France) were used and their β-D-glucuronidase checked. These 3 micro-organisms were kept in cultures at 37°C for 24-48 h. Detection of C. perfringens was carried out in SPS medium (Liofilchem) and cultures were kept at 37°C for 48 h under strict anaerobic conditions using Anaerocult system (Merck, Darmstadt, Germany). For S. aureus, Baird Parker plates (Liofilchem) were placed at 37°C for 24-48 h and colonies were confirmed by the coagulase test (BioMérieux). For the fungi Sabouraud Cloramphenicol plates (Liofilchem) were used and samples were placed at 28°C for 5-7 days.
For the detection of S. typhimurium, 25 g of the sample were placed into a flask containing 225 mL of buffered peptone water to pre-enrich the sample for 18 h at 37°C. The enrichment step was carried out in Rappaport-Vassiliadis broth (Liofilchem) at 42°C for 24 h and the colonies were recovered in selective media SS and XLD (Liofilchem). One to three suspected colonies were picked out and biochemically identified using the API20E system (BioMérieux).
Statistical analysis: The statistical analysis of the results was carried out by SPSS v14.0 software (SPSS Inc., Chicago, IL, USA) and SAS v9.1 (SAS Institute Inc., Cary, NC, USA) using the Mann-Whitney-Wilcoxon Test for bivariate, non-parametric tests. p<0.05 was considered significant.
RESULTS
Significant differences were observed among the control group and groups 2 and 3. In the control group no growth was observed for any of the analyzed micro-organisms, with exception of TAMB. These bacteria showed an increase in their counts toward the end of the experimental period although always lower than 2 log cfu g-1 (data not shown).
The dynamics of the micro-organism growth in each commercial pet food product analyzed are shown in Table 2-4.
In group 2, counts for all micro-organisms decreased over time, with the exception
of TAMB, that in most of the samples, as occurred in the control group, increased
towards the end of the study but never reached the inoculation counts.
| Table 2: | Results obtained for group 1 and 2 2 in products A1, A2, B1 and B2, expressed as log cfu g-1 |
| TAMBC: Total aerobic mesophile bacteria count A: Absence P: Presence <1: Under the detection limit (10 cfu g-1) | |
| Table 3: | Results obtained for group 2 in products C1, C2, D1 and D2, expressed as log cfu g-1 |
| TAMBC: Total aerobic mesophile bacteria count A: Absence P: Presence <1: Under the detection limit (10 cfu g-1) | |
| Table 4: | Results obtained for group 2 in products E1, E2, F1 and F2, expressed as log cfu g-1 |
TAMBC: Total aerobic mesophile bacteria count A: Absence P: Presence <1: Under the detection limit (10 cfu g-1) |
|
Total Enterobacteriaceae counts had results of a high variability depending on the product, while E. coli remained viable for a maximum of 8 weeks. S. aureus was generally recovered up to the fourth week, although in group B counts were obtained even the tenth week. C. perfringens showed low and irregular counts during the study.
Salmonella typhimurium was detected during the whole experimental period when coexisting with the other micro-organisms (group 2) or inoculated alone (group 3). No statistical differences were found when comparing both groups.
With respect to the fungi, F. moniliforme did not grow in any of the analyzed pet foods and it was only detected during the first month in concentrations lower than those at inoculation. On the other hand, A. flavus showed maximum counts of 2x104 cfu g-1, but it was detected irregularly throughout the experiment and also increased towards the end of the experimental period.
In order to compare the results more accurately and to know where lay the differences, products were compared all together and by pairs. These pairs (A-B, C-D and E-F) were established according to the similarities in composition and category of pet food to which they belong.
Among the commercial categories of pet extruded food analyzed, product B showed significantly higher counts of TAMB (p<0.05) compared to products C, D, E and F. When compared by pairs of similar products product B had higher counts of TAMB than product A (p = 0.001) and product C had higher counts of TAMB than product D (p = 0.013). Similarly, product B showed significant differences of Enterobacteriaceae counts compared to the rest of the products (p<0.05) and higher counts of Enterobacteriaceae were observed in product B compared to A (p = 0.002) and in product C compared to product D (p = 0.059). Product C showed significant differences compared with product A (p = 0.059) and E (p = 0.008). Escherichia coli counts showed differences between products B and D (p = 0.043) and F and D (p = 0.043). Clostridium perfringens counts were also higher in product B compared to products A (p = 0.006), C (p = 0.056) and D (p = 0.011) and product F compared to D (p = 0.028). In the case of the fungi, product B also showed statistically significant higher counts of A. flavus compared to E (p = 0.014) and F (p = 0.002) and significantly higher counts were observed among product C compared to E (p = 0.008) and F (p = 0.008). Fusarium moniliforme could not be compared among products due to its irregular presence along the study. No differences were observed among products regarding S. aureus counts.
DISCUSSION
The present study showed that the inoculated strains of micro-organisms such as TAMB, Enterobacteriaceae, E. coli, C. perfringens, S. aureus, S. typhimurium, A. flavus and F. moniliforme, did not adapt properly to extruded dry pet food and that these pet foods are not a good substrate for microbial growth. In fact, all counts decreased along the experimental period, although some micro-organisms remained after inoculation in dry pet food. Dry pet food has a low water activity which preserves the quality of pet food throughout its scheduled shelf life (Gómez et al., 1997) and in the case of dry extruded pet food, in the extrusion process, kibbles and the ingredients used to coat kibbles generally undergo temperatures above 130°C. These treatments seem enough to avoid growth and survival of most of the experimentally inoculated micro-organisms analyzed in the present study. However, S. typhimurium was detected in the inoculated samples during the 6 months of the study, even in interaction with other micro-organisms. Salmonella typhimurium is very resistant to adverse conditions such as the acidity or temperature variations (Medina et al., 2007). On the contrary, S. aureus, F. moniliforme and E. coli showed a low persistence in the kibbles.
According to the results obtained, products B and C seemed to be the products where the microbial viability remained for a longer time. A possible explanation for this fact is that the rest of the products had an acidic coating on the kibbles` surface and its influence was checked by the significant differences in viability found in the statistical analysis.
If acid coating was not present higher viability of all the bacteria analyzed was observed. Products B and C showed higher counts and viability of micro-organism, especially TAMB and Enterobacteriaceae. The acidic coating causes a drop in pH and limits the growth of the inoculated micro-organisms in different degrees (Prescott et al., 1999), especially in the case of S. aureus (Del Río et al., 2007). Taking this fact into account it can be predicted that uncoated particles are more susceptible to providing a better substrate for different micro-organisms and they can therefore develop more easily. On the contrary fungal counts of A. flavus were higher in kibbles with acidic coating, since fungi can develop at low pH values (Prescott et al., 1999).
Some micro-organisms showed very low viability in dry pet food such as F. moniliforme that showed a lack of adaptation to the substrate. This genus is very sensitive to adverse conditions and probably because of the characteristics of its conidia lost viability faster. On the other hand, the other fungus analyzed, A. flavus that is a well known producer of mycotoxins (Bhatnagar et al., 2002), showed higher viability. Other micro-organisms such as C. perfringens were very irregular along the experimental period. This genus can form spores when finds itself in a hostile environment until the right conditions are re-established (Velugoti et al., 2007).
Levels of Enterobacteriaceae and TAMB counts were always lower than 3x104 cfu g-1 in control cultures and these levels could be used as a hygiene indicator for the extruded pet food.
Even if dry extruded pet food is a poor substrate for microbial development, the need for good hygienic practices in processing and packaging cannot be ignored (Fischer et al., 2007). Knowing the kibbles composition and performing a microbiological analysis prior to commercialization may help to predict the microbiological dynamics and viability of these products so that higher safety for both owners and pets can be achieved.
CONCLUSION
To our knowledge this is the first study to analyze viability and dynamics of the main micro-organisms involved in outbreaks of bacterial food borne diseases inoculated in dry extruded pet food. The results indicated that dry extruded pet food is not suitable for microbiological growth, especially if kibbles are coated with acid. However, contamination with Salmonella Thyphimurium can remain in this food up to 6 months. Further studies will need to analyze other pet food formulas such as dry extruded kibbles in order to predict the microbiological behaviour of these important micro-organisms.