Bacteriological evaluation of dog and cat diets that claim to contain probiotics

Abstract

Nineteen commercial pet foods claiming to contain probiotics were evaluated. Selective bacterial culture was performed to identify organisms that were claimed to be present. Twelve diets claimed only to contain specific bacterial fermentation products, which does not necessarily indicate that live growth would be expected, but these products also included the term “probiotic” somewhere on the package, suggesting that live, beneficial organisms were present. No products contained all of the listed organisms, while 1 or more of the listed contents were isolated from 10 out of 19 products (53%). Eleven products contained additional, related organisms including Pediococcus spp, which was isolated from 4 products. No relevant growth was present in 5 (26%) products. Average bacterial growth ranged from 0 to 1.8 × 10⁵CFU/g. Overall, the actual contents of the diets were not accurately represented by the label descriptions.

Introduction

Probiotics are microorganisms that, when ingested, exert beneficial effects beyond that of their nutritional value (1). Probiotic therapy is becoming increasingly popular in veterinary medicine; however, few results from objective research are available, particularly for dogs and cats. Probiotic therapy has been recommended for the treatment or prevention of a variety of conditions in different species. A number of probiotic products are available commercially for use in dogs and cats; they are available in tablet, capsule, paste, and liquid forms. Some commercial dog and cat foods also claim to contain probiotics. Incorporation of probiotics into diets may have the advantage of easy, daily administration of beneficial organisms.

Because probiotics are considered food supplements, not drugs, there are no regulations regarding their use as supplements or food additives. Various studies have reported that quality control among probiotic supplements intended for human or animal use is poor, with a significant percentage of products either not containing the organisms stated on the label, not containing the numbers of organisms stated on the label, or containing additional species (2,3,4,5) Studies evaluating the quality control of pet foods claiming to contain probiotics have not been reported. The goal of this study was to isolate, enumerate, and identify probiotic bacteria in pet foods claiming to contain probiotics.

Materials and methods

Commercially available pet (dog and cat) foods claiming to contain probiotics were purchased from local retailers. Products listing bacterial species in the ingredient list or claiming to contain ‘probiotics’ on the packaging were included. Lot numbers and expiry dates were recorded and all products were tested prior to their expiry date. Quantitative culture was performed on all products. Ten grams of food were added to 90 mL of phosphate buffered saline (PBS, pH 7.2) and homogenized in a blender. Serial 10-fold dilutions were performed in PBS and 100 μL of each dilution was inoculated onto deMan, Rogosa, Sharpe (MRS) agar and incubated anaerobically for the isolation of lactic acid bacteria, and onto blood agar and incubated aerobically and anaerobically for the isolation of enterococci and bacilli. All plates were incubated at 37°C for 48 h.

Enrichment culture was used to aid in the isolation of low levels of organisms that might be overlooked in quantitative culture. For enrichment, 500 mg of each food was added to 10 mL each of MRS broth for growth of lactic acid bacteria, tryptone soya broth for growth of aerobes, and brain heart infusion broth for growth of anaerobes. After 24 h of incubation, 100 μL of broth was inoculated onto MRS agar and incubated under anaerobic conditions for the isolation of lactic acid bacteria, onto blood agar under aerobic conditions for the isolation of aerobes, and onto blood agar under anaerobic conditions for the isolation of anaerobes. Plates were incubated at 37°C for 48 h. Organisms were identified via colonial morphology, Gram stain reaction, and biochemical characteristics. Lactobacilli were identified by using the API 50 CHL biochemical assay (Biomerieux, St. Laurent, Quebec) and bacilli were identified via standard biochemical tests (6). Enterococci were identified by using the API 20 Strep biochemical assay (Biomerieux). All testing was performed in duplicate. No attempt was made to identify or enumerate organisms unrelated to those listed on the labels.

Results

Nineteen diets were tested. Thirteen were for dogs and 6 were for cats. All claimed to contain specific organisms or probiotics. Two products listed 1 or more bacterial species as ingredients, while 12 listed fermentation products of probiotic species, and 5 listed both organisms and fermentation products. One product claimed to contain Streptococcus faecium, which was reclassified as Enterococcus faecium in 1984 (7). Five products misspelled 1 of the listed species.

Bacterial growth was present in all products; however, as the purpose of this study was to evaluate the contents of the diets compared with those claimed on the label, no attempt was made to identify organisms that were not included on the list of ingredients. No products contained all of the claimed organisms, while 1 or more of the listed contents were isolated from 10 out of 19 (53%) products (Table 1). Eleven products contained additional, related organisms, including Pediococcus spp, which were isolated from 4 products. Five (26%) products did not contain any relevant growth.

Table 1.

Discussion

Interpretation of these results is confounded somewhat by the questionable labelling of some products. Twelve diets listed only specific bacterial fermentation products (L. acidophilus fermentation product) as ingredients, while 5 diets claimed to contain both specific organisms and fermentation products. Fermentation products of lactic acid bacteria or bacilli are typically included as a source of enzymes. This does not necessarily indicate that live organisms are present and, based on the definition provided above, these would not be considered to be probiotics. However, the diets in this study that listed only fermentation products in the ingredients also stated directly that they contained probiotics. Therefore, the variable levels of live growth may better reflect misclassification or incorrect advertising of the diets, rather than indicate poor survival of probiotic organisms.

The diets that were tested contained between 0 and 1.8 × 10⁵ CFU/g. It is unclear whether the level of supplementation present in these diets would be adequate, even if the organisms that were present possessed probiotic properties. Doses of 1 × 10¹⁰ to 1 × 10¹¹ CFU/d are often used in human probiotic efficacy studies (8,9,10). Kailasapathy et al (11) have suggested a minimum therapeutic dose of 1 × 10⁸ to 1 × 10¹⁰ CFU/d in humans, while others have claimed that at least 1 × 10⁹ to 1 × 10¹⁰ viable organisms must reach the intestinal tract in humans to be effective (3,12,13). Dose requirements for probiotic organisms have not been evaluated adequately in dogs and cats, and they likely vary between probiotic strains. However, Lactobacillus rhamnosus strain GG was only variably isolated from the feces of dogs fed doses of 1 × 10⁹ and 1 × 10¹⁰ CFU/d (Weese and Anderson, unpublished data). Approximately 5.5 kg/d of the diet containing the highest level of growth would be required to provide 1 × 10⁹ CFU/d. It was interesting to note that bacterial names were misspelled on the labels of 5 products. While this does not necessarily indicate a poor quality product, such errors raise concerns.

Enterococcus faecium was the most commonly isolated organism and has been reported to have probiotic properties in vitro and in vivo (14,15). However, concern has been expressed over the use of enterococci, because they can be opportunistic pathogens and probiotic strains of enterococci are able to transfer the vanA gene; the gene responsible for vancomycin resistance (16,17). Lactococcus lactis was identified in 7 products, despite not being included on the label of any product, while L. plantarum was present in 4 diets. Certain strains of L. plantarum and Lc. lactis have been demonstrated to possess probiotic properties; however, it is unclear whether the strains isolated from these diets possessed any such properties (18,19). Lactobacillus acidophilus was purported to be present in 13 diets, yet it was not identified in any. This was surprising and an explanation is not readily apparent. Lactobacillus acidophilus has been isolated routinely from probiotic supplements and fecal samples by the author's laboratory, so it is unlikely that laboratory error accounted for this finding. Inaccurate identification by the biochemical assay could be considered; however, in 9 out of 14 (64%) of the diets claiming to contain L. acidophilus, no organisms isolated with a similar Gram stain appearance, let alone similar biochemical characteristics, were isolated. Despite being listed on the label of 3 products, Bacillus licheniformis was not isolated from any diet, and B. subtilis was identified in only 1 out of 12 (8.3%) diets. Once it was determined that the stated bacilli were not present, no further attempts were made to identify contents, so it is possible that other bacilli were present. Mislabeling of probiotics claiming to contain specific bacilli has been reported elsewhere (20,21). The isolation of Pediococcus spp. from 4 diets was surprising. Pediococci have not been reported to possess beneficial properties and are considered to be opportunistic pathogens (22). The failure to detect Bifidobacterium spp. from any product likely relates to the strict anaerobic nature of this genus (23) and is not surprising, as similar results have been reported in human and veterinary probiotic supplements (2,4,5).

Two products claimed to contain yogurt. This does not adequately indicate which bacterial species would be expected to be present. Most yogurt products contain L. bulgaricus and Streptococcus thermophilus, which are not regarded as being probiotics because they are poorly able to survive bile environments and are typically unable to colonize the intestinal tract (24,25). Yogurt containing probiotic strains is available, but there is no indication that probiotic-containing yogurt was added to these diets.

The culture techniques that were employed here would be expected to be adequate for isolation of all of the probiotic organisms listed on the labels. The use of both direct and enrichment culture techniques should have decreased the possibility that some organisms were overlooked. It is possible that some strains of similarly appearing lactobacilli were not individually identified. Representative colonies from plates containing differently appearing colonies were identified, but it cannot be ruled out that species with similar colonial and Gram stain appearances were not distinguished. This could have affected the results in diets with multiple organisms; however, it would not affect the overall bacterial counts.

The yeast component of the diets was not evaluated. Some species of yeast, notably Saccharomyces boulardii, have been shown to possess probiotic properties (26,27); however, most yeast additives act as nutritional supplements, not probiotics. There is no evidence that any of the yeast species listed on the labels of the tested diets possess probiotic properties.

Overall, commercial pet foods that claim to contain probiotics appear to contain very low numbers of viable organisms, and often do not contain the species listed on the label. Whether this relates to improper addition of organisms during processing, failure to survive processing, or poor viability during storage is unclear. Regardless of the contents of any diets containing lactic acid bacteria or bacilli, it is debatable as to whether they should be considered to contain probiotics without demonstration of species-specific efficacy. Based on extrapolation from other species, it seems likely that certain probiotics will be shown to be effective for the treatment or prevention of certain conditions in dogs and cats. Currently, there is very little information available regarding research on probiotic therapy in dogs. Results of this study indicate that these commercial diets are not good sources of probiotics. While production of a probiotic-containing diet should be possible, research must be performed to select bacterial species with beneficial in vitro and in vivo properties, and the ability to survive commercial processing and storage. CVJ