The manufacturing processes have been reported to influence the properties of probiotics with potential impact on health properties. The aim was to investigate the effect of different growth media and inactivation methods on the properties of canine-originated probiotic bacteria alone and in combination mixture.
Methods and Results
Three established dog probiotics, Lactobacillus fermentum VET9A, Lactobacillus plantarum VET14A and Lactobacillus rhamnosus VET16A, and their combination mixture were evaluated for their adhesion to dog mucus. The effect of different growth media, one reflecting laboratory and the other manufacturing conditions, and inactivation methods (95°C, 80°C and UV irradiation) on the mucus adhesion of the probiotic strains was characterized. Evaluation of dog probiotics was supported by cell visualization using transmission electron microscopy (TEM). Higher adhesion percentage was reported for probiotic strains growing in laboratory rather than in manufacturing conditions (P < 0·05). Inactivation by heat (95°C, 80°C) decreased the adhesion properties when strains were cultivated in soy-based growth media compared with those grown in MRS broth (P < 0·05). TEM observations uncovered differences in cell-surface components in nonviable forms of probiotic strains as compared with their viable forms.
Manufacturing process conditions such as growth media and pretreatment methods may significantly affect the adhesive ability of the tested strains.
Significance and Impact of the Study
Growth conditions, growth media, pretreatment methods and different probiotic combinations should be carefully considered for quality control of existing probiotics and for identification of new probiotics for dogs. These may also have an impact on health benefits for the host.
Adhesion of probiotic bacteria to the intestinal mucus has been reported to be associated with probiotic health benefits. Adhesion is also regarded a prerequisite for temporary colonization and extended faecal recovery of probiotics (Finlay and Falkow 1997; Juntunen et al. 2001). Probiotic is defined as ‘live microorganism which when administered in adequate amounts confers a health benefit on the host’ of humans or animals (FAO/WHO 2002). Some health effects have also been reported for inactivated forms of probiotic bacteria (Lopez et al. 2008; Lahtinen and Endo 2012). Nonviable preparations may have global economic advantages in terms of extended shelf life in nonrefrigerated conditions and for storage in extreme temperature or humidity. Improved stability could facilitate the use and maintenance of probiotic products for areas where their use has been expensive or impossible due to lack of cold storage. The method of inactivation appears to influence the properties and potentially the efficacy of probiotics (Ouwehand and Salminen 1998; Ananta and Knorr 2009; Lahtinen and Endo 2012). Physical inactivation treatments may alter the adhesive abilities of the bacterial cell wall, which in turn may affect the properties of probiotics (Tuomola et al. 2000). Similar changes have been reported for different production processes of probiotic products (Grześkowiak et al. 2011).
Our hypothesis was that different growth media, one reflecting laboratory and the other manufacturing conditions, modify the properties of canine probiotic bacteria alone and in combination. These and different inactivation methods were assessed for their adhesion impact to choose optimal strains or strain combinations to improve the health of dogs (Manninen et al. 2006). Three established dog probiotics, Lactobacillus fermentum VET9A, Lact. plantarum VET14A and Lact. rhamnosus VET16A (Beasley et al. 2006), and their combination mixture were assessed in this study.
Materials and methods
Bacterial strains and culture conditions
The three probiotic strains (called here: VET probiotics) used in the study were Lact. fermentum VET9A (NCIMB 41636), Lact. plantarum VET14A (NCIMB 41638) and Lact. rhamnosus VET16A (NCIMB 41640), provided by the Vetcare Ltd. (Salo, Finland).
For assays, VET probiotics were grown in de Man Rogosa Sharpe broth (MRS; Oxoid Ltd., Basingstoke, Hampshire, UK) in 37°C to reflect standard laboratory conditions. For a comparison of the effect of growth media on the adhesion properties, the parallel assays were performed where the strains were grown in patented soy-based growth media provided by Galilaeus Ltd. (Kaarina, Finland). Soy-based growth media contain mainly of soy meal as a source of nitrogen for the lactic acid bacteria strains. Strains have a carbon and mineral sources as well (Patent application numbers: FI 20095836 (A) and PCT/FI2010/050538).
The process was completed in 30°C to reflect manufacturing conditions. For all VET probiotics, the culture was inoculated into broth in a 1·5-ml Eppendorf tube to a final concentration of one per cent and incubated for 18 h without agitation and under aerobic conditions.
Dog mucus samples
Mucus was prepared from the canine jejunal chyme essentially as described earlier (Kirjavainen et al. 1998; Ouwehand et al. 2001). In brief, jejunal chyme was centrifuged at 12 000 g to remove particulate matter. Mucus was precipitated from the clear supernatants by dual ethanol precipitation and freeze-dried. Equal amounts of mucus from each dog were pooled, and a stock suspension of 5 mg ml−1 in HEPES (N-2-hydroxyethylpiperazine-N0-2-ethanesulfonic acid)-Hanks buffer (HH; 10 mmol l−1 HEPES; pH 7·4) was prepared and stored at −20°C until use. Before use, the protein concentration was determined by a Bradford method as described by Kruger (1994), using bovine serum albumin (BSA; Sigma, St. Louis, MO, USA) as a standard. Dog mucus was dissolved to concentration of 0·5 mg ml−1 in HEPES (N-2-hydroxyethylpiperazine-N-2-ethanesulfonic acid)-Hanks buffer (HH; 10 mmol l−1 HEPES, pH 7·4).
Treatments of the microbial suspensions
The microbial suspensions of VET probiotics were divided into 1-ml aliquots and incubated for 10 min at 95°C and 30 min at 80°C. For UV irradiation, 1 ml of the microbial suspension was transferred to a Petri dish and irradiated for 10 min at a distance of 10 cm from a 30-W UV-C lamp (Philips, Roosendaal, the Netherlands).
In vitro assay of adhesion to dog intestinal mucus
In brief, 100 μl (0·5 mg ml−1) of dog jejunal mucus was immobilized onto 96 wells of polystyrene microtiter plates (Maxisorp, Nunc, Denmark) by incubation overnight at 4°C, as previously described (Grześkowiak et al. 2011). The wells were washed twice with 200 μl of HH to remove unbound mucus.
For adhesion assays, VET probiotic strains Lact.fermentum VET9A, Lact. plantarum VET14A and Lact. rhamnosus VET16A were metabolically labelled by the addition of 10 μl ml−1 tritiated thymidine (5-3H-thymidine 1·0 mCi ml−1; Amersham Biosciences, Little Chalfont, UK) to the culture media and were grown overnight. After incubation, bacteria were harvested and washed twice with phosphate-buffered saline (PBS) buffer. The treatments of the microbial suspensions were performed. Thereafter, the optical density of radiolabelled bacteria at 600 nm was adjusted to 0·25 ± 0·05 to standardize the bacterial concentration (108 cells ml−1). A suspension of 100 μl radioactively labelled bacteria single or in combination mixture was added to each well. The combination mixture was prepared by mixing bacterial solutions in equal proportions. After incubation at 20°C for 1 h, the wells were washed twice with 200 μl of HH to remove unbound bacteria. Bound bacteria were released and lysed by incubation at 60°C for 1 h with 250 μl of 1% sodium dodecyl sulfate (SDS) in 0·1 mol l−1 NaOH. Adhesion was assessed by quantifying the amount of radioactivity by liquid scintillation and was expressed as the percentage of radioactivity recovered after adhesion relative to the radioactivity of the bacterial suspension added to the immobilized mucus. Adhesion was determined in three independent experiments, and each assay was performed in triplicate to calculate intra-assay variation.
Transmission electron microscopy of VET probiotics and their nonviable forms
The VET probiotics were cultured individually in the patented soy-based growth broth for 18 h. The cultures were separated into two: one was observed as viable cells and another was heat-treated by the following method and observed as heat-inactivated cells. Heat inactivation was performed by a cyclic heating process three times 10 min at 80°C. The VET probiotic strains were allowed to cool to 40°C between heating cycles. Viable and heat-inactivated cells were fixed with 5% glutaraldehyde (Electron Microscopy Sciences, Fort Washington, PA, USA) in 0·16 mol l−1 s-collidine–HCl buffer, pH 7·4 for 3 h, postfixed with 1% OsO4 (Merck KGaA, Darmstadt, Germany) containing 1·5% potassium ferrocyanide for 2 h, dehydrated with a series of increasing ethanol concentrations (70%, 96% and twice at 100%) and embedded in 45359 Fluka Epoxy Embedding Medium kit (Sigma-Aldrich, Buchs, Switzerland). Thin sections were cut with an ultramicrotome (REICHERT ULTRACUT E ultramicrotome, Leica Corp., Wien, Austria) to a thickness of 70 nm, stained with 5% uranyl acetate and 5% lead citrate in Ultrostainer (Leica Corp., Wien, Austria). The sections were examined with a JEOL JEM-1200EX transmission electron microscope (JEOL, Tokyo, Japan) operating at 60 kV acceleration voltage.
Statistical analysis was performed using the SAS for Windows 9.3 (SAS Institute, Inc., Cary, NC). Data were subjected to three-way anova. All results are shown as the average of at least three independent experiments; variation is expressed as standard deviation. Significance was considered at P < 0·05.
Adhesion properties of the VET probiotics cultivated in MRS and patented soy-based growth media
Adhesion properties of the study probiotics are shown in Fig. 1. All the tested VET probiotics and their nonviable forms had a good ability to adhere to dog jejunal mucus, although this varied between strains grown in different media. Two strains, that is, VET9A (viable, nonviable UV) and VET16A (viable), did not show any statistical differences (P > 0·05) in the adhesion percentage between micro-organisms cultivated in environment reflecting manufacture and laboratory conditions. All the other VET probiotics and their nonviable forms alone and in combination mixture showed differences (P < 0·05) amongst the adhesion properties depending on the growth conditions of micro-organisms – VET probiotics and their nonviable forms showed higher adhesion ability when cultivated in MRS media compared with the soy-based growth media.
Adhesion properties between viable vs nonviable forms of VET probiotics cultivated in MRS media
The adhesion of probiotic strains grown in MRS media alone and in combination mixture varied between 2·88% (SD 0·34) and 10·11% (SD 1·78) (Fig. 1). The heat inactivation influenced the adhesion properties of the strain VET14A causing adhesion loss when heat-inactivated at 95 and 80°C (P = 0·044, P = 0·022, respectively). On the other hand, cells inactivated by UV showed similar adhesion properties to viable cells and showed better properties than cells inactivated by heating at 95°C and at 80°C (P = 0·006, P = 0·003, respectively) of the VET14A strain.
In addition, adhesion ability of the probiotics in their combination mixture (viable and inactivated forms) differed from that of VET14A and VET16A alone in their viable and inactivated forms (P < 0·05).
Adhesion properties between viable vs nonviable forms of VET probiotics cultivated in patented soy-based growth media
The adhesion of probiotic strains grown in patented soy-based growth media alone and in combination mixture varied between 1·44% (SD 0·05) and 4·07% (SD 0·61) (Fig. 1). The treatments had a different effect on the adhesion properties of a strain VET9A; the adhesion of this strain in viable form was higher than that of heat-inactivated at 95 and 80°C (P = 0·001, P < 0·001, respectively). On the other hand, cells inactivated by UV had similar adhesion properties to viable cells and showed better properties than cells inactivated by heating at 95°C and at 80°C (P = 0·001, P < 0·001, respectively) of a VET9A strain. VET14A strain behaved similarly (P < 0·001, P < 0·001, P < 0·001, P < 0·001, respectively) as well as the combination mixture of the three strains VET9A, VET14A, VET16A (P = 0·007, P < 0·001, P < 0·001, P < 0·001, respectively). The adhesion of a strain VET16A in viable form was higher than that of the heat-inactivated strain at 95 and 80°C (P = 0·013, P = 0·006, respectively); inactivation of this strain by UV increased the adhesion percentage compared with inactivation by heating at 80°C (P = 0·043).
The adhesion properties of the VET probiotics in their combination mixture (viable and inactivated forms) differed from that of VET9A alone in its viable and inactivated forms (P < 0·05), VET14A alone in its viable and UV-inactivated form (P < 0·05) and VET16A alone in its UV-inactivated form (P < 0·05).
TEM observation of viable and nonviable forms of VET probiotics
TEM analysis indicated that the viable VET9A cells have a cell-surface coat of fine filamentous material, but components of this are rarely seen on the surface of the heat-inactivated cells grown in the patented soy-based growth broth (Fig. 2a,b). Moreover, interestingly, several intracellular components were seen in viable VET9A cells but not in inactivated VET9A cells (Fig. 2a). Similar results were obtained from VET14A and VET16A, and cell components were only seen in viable cells (Fig. 2a). No major differences were found on cell surface of the two strains after inactivation (Fig. 2a,b).
We have earlier reported that manufacturing conditions have an impact on the properties of human probiotic bacteria even for the same specific strain (Grześkowiak et al. 2011). Here, we demonstrate that probiotic bacteria of canine origin present significantly different properties when cultivated in different media and inactivated by different treatments.
The probiotic definition by FAO/WHO (2002) requires that probiotics should be viable to exert beneficial health effects. However, also the inactivated forms of probiotic bacteria have been reported to provide health benefits (reviewed by Taverniti and Guglielmetti 2011).
There are more studies on the use of viable than that of nonviable forms of probiotics (Mercenier et al. 2003; Collado et al. 2007; Marsella et al. 2012). Generally, nonviable forms of probiotics may be less able to bind to the intestinal mucosa, but also translocate less and thereby have lesser ability to modulate the immune system but also less likely to be harmful (Haller et al. 2000; Cross et al. 2004). As the method of inactivation can disrupt the bacterial cells, the potential interaction of intracellular bioactive bacterial compounds with host cells can be decreased but at times also increased. To test the hypothesis of impact of viability, we assessed the effect of three different methods of inactivation on the in vitro adhesion of dog probiotic strains. The aim was to facilitate selection of an appropriate method for the culturing and processing and maintaining the benefits of both forms of bacteria.
The probiotics adhesion capacities differed significantly depending on the growth conditions of micro-organisms. Previously, Ouwehand et al. (2001) reported that the growth media and the food matrix significantly affect the adhesive ability of certain human probiotic strains. In addition, we also confirmed that the inactivation methods had an impact on the adhesion of VET probiotics in the present study. When VET probiotics were cultivated in MRS broth, heat-treatments reduced the adhesion property of Lact. plantarum VET14A, but the UV irradiation treatment had no significant effects on that of the strain. However, in VET9A, VET16A and mixture of the VET strains, inactivation did not have a significant impact when they were cultured in MRS broth. This might be a merit to use the inactivated cells as probiotics to expect longer shelf life and comparable probiotic properties to viable cells. On the other hand, all VET probiotics cultivated in soy-based growth medium exerted significant reduction in the adhesion after bacteria were heat-treated but not after UV irradiation. Our TEM analysis indicated that heat inactivation at 80°C had a visual impact on cell-surface components of VET9A cultured in the soy-based growth medium. Because of the expected polysaccharide components on the cell wall, we used potassium ferrocyanide additive (Kanovsky 1971) in the osmium fixative for TEM, because it enhances contrasting of polysaccharides by osmium. The observed components might be exopolysaccharides (EPS), which are known to have both positive and negative impacts on adhesion properties of probiotics (Lebeer et al. 2009). On the other hand, major differences were not seen on cell surface of viable and heat-inactivated cells in VET14A and VET16A by TEM, but these strains cultured in the soy-based growth medium also reduced adhesion ratio after heating. These might mean nonvisible components by TEM, for example proteins, were denatured by heating. Several cell-surface proteins have been reported as adhesins of mucosa (Kleerebezem et al. 2010). Thus, future study should focus on analysing the changes and differences in cell wall composition and structure. The adhesion of Lact. fermentum VET9A and Lact. rhamnosus VET16A was not affected by any treatment method when strains were cultivated in MRS broth compared with commercial media where the differences in adhesion were distinguishable. In addition, the studied strains have been cultured and isolated from canine faeces using MRS medium thus providing a growth advantage on MRS medium.
Taken together, our findings indicate that manufacturing process conditions such as growth media and pretreatment methods may significantly affect the adhesive ability of the tested strains. These should be considered when designing new probiotics, including their nonviable forms, and planning in vivo studies, and also for quality control of probiotic commercial products for dogs.
This study was supported by the Vetcare Ltd. Shea Beasley is employed by Vetcare Ltd. None of the other authors had any conflict of interest to the study probiotics. We thank Satu Tölkkö and Hanna Lehmussola for the technical assistance and Jaakko Matomaki for statistical consultation during the data analysis. Markus Peurla and Erica Nyman, Turku University Laboratory of Electron Microscopy, are acknowledged for preparing TEM samples. The author's responsibilities were as follows: ŁG, MCC, SS, SB planned and coordinated the study; ŁG was responsible for the laboratory experiments regarding adhesion studies. AE and LJP interpreted the TEM pictures. All authors participated in analysis of results and in writing and revising the manuscript.