Obesity has now attained pandemic proportions in many parts of the world. There is a recent interest in studying the gut microbiota of obese and lean individuals, with differences in composition being suggested. As this is the case, the microbiota is susceptible to change through the diet. Probiotics are live microbes which when administered in adequate amounts confer health benefits to the host(1). This occurs through targeting the gut microbiota. Lactobacilli and bifidobacteria are the most widely used probiotic species. For probiotics to have an effect on metabolism, it is necessary that they should reach distal parts of the intestine in sufficient numbers to exert effects therein. For this purpose, they must be able to bypass gastric acidity and possess bile-tolerant properties to ensure survival in the lower gut. The potential role of probiotics in metabolic diseases has recently gained wide attention owing to rising evidence of the importance of the gut microbiota balance in energy homeostasis(Reference Backhed, Manchester and Semenkovich2). Firmicutes and Bacteroidetes represent dominant microbial phylum members of gut ecology and their ratio has been linked to obesity in mice(Reference Ley, Backhed and Turnbaugh3) as well as in human subjects(Reference Gill, Pop and Deboy4). Recent studies have indicated the anti-obesity roles of different probiotics in terms of lowering of food intake and body weights, though the possible mechanism of action is not yet clear.
In the Indian population, there is an increased prevalence (10 %) of overweight individuals, of which the majority are from urban areas consuming energy-dense food products(Reference Yadav and Krishnan5). ‘Dahi’ is a fermented milk product of the Indian subcontinent having low glycaemic index and produced by multiple starter strains of Lactococcus. It is perceived as a ‘healthy’ food product among the Indian population. It differs from European yogurt in having a higher diacetyl content produced by multiple strains of Lactococcus, which also gives a characteristic flavour to the product(Reference Yadav, Jain and Sinha6). Therefore, addition of probiotics to dahi may act as an important vehicle for probiotic administration and may produce a physiological response in the host. Some earlier reports have indicated the anti-diabetic potential of probiotic dahi containing Lactobacillus acidophilus and L. casei (Reference Yadav, Jain and Sinha6–Reference Yadav, Jain and Sinha9).
L. acidophilus NCDC 13 is an indigenous probiotic strain isolated from fermented milk sources in India. The strain was selected from twenty different lactobacilli strains and has been shown to exhibit the best probiotic properties such as acid tolerance, bile tolerance and surface hydrophobicity as assessed under in vitro conditions(Reference Kushal, Anand and Chander10). The probiotic has also been used in animal experiments and been shown to survive in the gastrointestinal tract of mice(Reference Kushal, Anand and Chander11, Reference Kushal, Anand and Chander12). The present study was designed to evaluate the effects of L. acidophilus NCDC 13 supplementation on intestinal microbiota modulation and on the progression of obesity under high-fat dietary conditions in diet-induced obese C57BL/6 mice. We investigated the anti-obesity effects of the probiotic by supplementing it in a fermented milk product, dahi, compared with the control dahi alone.
Materials and methods
Cultures
L. acidophilus NCDC 13 and control dahi culture NCDC 167 (Lactococcus lactis ssp. lactis, L. lactis ssp. cremoris, L. lactis ssp. lactis biovar. diacetylactis) were procured from the National Collection of Dairy Cultures, Dairy Microbiology Division, National Dairy Research Institute, Karnal, India.
Preparation of control and probiotic dahi
The methodology has been reported previously(Reference Yadav, Jain and Sinha6) and demonstrated to give good viable counts. Semi-skimmed milk (1·5 %, w/v fat) was purchased for the preparation of dahi. The control dahi (mixed L. lactis sp.) and probiotic (L. acidophilus NCDC 13) cultures were propagated and maintained in M17 and de Man–Rogosa–Sharpe broth, respectively. The cultures were activated in sterile skimmed milk before their usage for the preparation of dahi. For the present study, two types of dahi (control and probiotic) were prepared. Control dahi was prepared by inoculating milk with starter dahi culture at the 1 % (v/v) level. Probiotic dahi was prepared by inoculating milk with both starter dahi culture and probiotic (L. acidophilus NCDC 13) cultures, each at the 1 % (v/v) level. Dahi is a fermented milk product having specific textural properties (set-type, semi-solid). Adding the probiotic after fermentation would result in the alteration of the texture of the product, resulting in the control dahi and probiotic dahi with different textures. Therefore, it is recommended to add the probiotic before dahi fermentation. There were no other nutritional differences between the two preparations other than the presence of the probiotic in the latter. Incubations were done at 30°C in both cases for 12 h and the final products so formed were stored at 4°C. The dahi products were prepared every week and bacterial enumeration was performed repeatedly. Viable counts of the control and probiotic cultures in dahi preparations were determined by serially diluting the aliquots and plating on M17 and de Man–Rogosa–Sharpe agar plates, respectively, and are expressed as colony-forming units (cfu)/ml. The viable counts of total Lactococcus in control dahi as determined on M17 agar were in the range of 3 × 107–4 × 107 cfu/ml. The viable counts of L. acidophilus in probiotic dahi as determined on de Man–Rogosa–Sharpe agar were in the range of 5 × 107–9 × 107 cfu/ml.
Animals and treatments
All animal procedures were performed in accordance with the UK Animals Scientific Procedures Act (1986). A total of twenty-four male C57BL/6 mice (6–8 weeks old; Charles River) were singly housed under controlled temperature (21–23°C) and light conditions (12 h light–12 h dark cycle; lights on at 07.00 hours). After acclimatisation for 1 week on a normal chow diet, they were then randomised according to their body weights and divided into two groups (control dahi and probiotic dahi), each having twelve mice. Both groups were fed on a 21 % high-fat diet (Table 1) and their respective dahi products.
Composition of the high-fat diet
AIN, American Institute of Nutrition.
* Vitamin mixture, g/kg (#Teklad 40 060): vitamin A palmitate, 3·9; vitamin D3 (cholecalciferol), 0·4; vitamin E (α-tocopheryl acetate), 24·2; vitamin K3, 4·9; thiamin hydrochloride, 2·2; riboflavin, 2·2; niacin, 9·9; folic acid, 0·19; pyridoxine hydrochloride, 2·2; vitamin B12, 2·9; inositol, 11·0; p-aminobenzoic acid, 11·0; biotin, 0·04; choline dihydrogen citrate, 349·6; calcium pantothenate, 6·6; vitamin C, 101·6; mixed in maize starch.
Both the control and probiotic dahi were provided in special hoppers to prevent spillage, for a brief period (about 6 h) during the daytime. During this period, the high-fat diet was removed from the cages to ensure that the animals in both groups consumed their respective dahi products. Animals were resumed on the high-fat diet after dahi supplementation. Water was given ad libitum. The feeding schedule was followed for 8 weeks.
Whole-body MRI and localised 1H magnetic resonance spectroscopy for intrahepatocellular (IHCL) and intramyocellular (IMCL) lipid levels were performed after 8 weeks, at the end of the dietary intervention. Faecal pellets were collected at week 0 before the start of the dietary intervention and at week 8 before the start of the scanning experiments and kept frozen at − 20°C until further analysis. Following MRI procedures, mice were culled, epididymal fat collected and stored on ice. Caeca were precisely excised and the contents were frozen.
Fluorescent in situ hybridisation to assess microbial changes
The protocol was followed as described previously(Reference Rycroft, Jones and Gibson13) with some modifications. Briefly, frozen faecal and caecal samples were diluted in ten volumes of PBS (pH 7·2; Oxoid). The samples were homogenised and the supernatant fraction (375 μl) containing bacterial cells was fixed with 4 % (w/v) paraformaldehyde in a ratio of 1:3 at 4°C. The bacterial cells were subjected to centrifugation (1000 g; 3 min), washed twice with PBS and finally suspended in an equal volume of PBS and ethanol (150 μl each). The cells were appropriately diluted and 20 μl of the suspension were applied to Teflon poly-l-lysine-coated six-well glass slides (Tekdon, Inc.). Permeabilisation of the cells was carried out in an ethanol series (50, 80 and 96 %) for 3 min each and hybridised with group-specific Cy-3-labelled (at the 5′-end) probes for 4 h at 50°C. The probes used were LAB158(Reference Harmsen, Elfferich and Schut14), BIF164(Reference Langendijk, Schut and Jansen15), EREC482(Reference Franks, Harmsen and Raangs16) and MIB663(Reference Salzman, de Jong and Paterson17) to enumerate total Lactobacillus–Enterococcus, bifidobacteria, Eubacterium rectale–Clostridium coccoides and mouse intestinal bacteria belonging to the Bacteroides subgroup in the phylum Cytophaga–Flavobacter–Bacteroides, respectively. The slides were subsequently washed, mounted in ProLong® Gold antifade reagent and stored at 4°C under dark conditions. The fluorescent cells were manually counted in fifteen random fields of view per well. Total cell counts were obtained with 4′,6-diamidino-2-phenylindole dihydrochloride staining.
Magnetic resonance studies
Whole-body 1H magnetic resonance spectroscopy
Mice were fasted for 16 h and anaesthetised with a 1–2 % isoflurane–oxygen mix which was maintained throughout the scan. The animals were scanned on a 4.7 Tesla Varian INOVA imaging system (Varian, Inc.) using pulse sequence with the following parameters: repetition time, 10 s; pulse angle, 45°; averages, 4. The spectra obtained were analysed and body adiposity calculated as described earlier(Reference So, Yu and Kuo18).
Whole-body MRI
Whole-body MRI was carried out to determine the amount of visceral and subcutaneous adipose tissue. Consecutive 2 mm-thick slices were acquired using a spin-echo sequence with the following parameters: repetition time, 2·2 s; echo time), 20 ms; matrix size, 256 × 192; field of view), 45 × 45 mm; averages, 2. The slices/images were then subjected to segmentation analysis (SliceOmatic™, Tomovision®) by an observer blinded to the experimental groups and the masses of different adipose tissue depots calculated.
Localised 1H magnetic resonance spectroscopy
The IHCL and IMCL lipid content was assessed by placing a voxel of 2 × 2 × 2 mm3 on liver and muscle images. A point resolved spectroscopy sequence with the following parameters, repetition time = 10 s, echo time = 9 ms and averages = 64, was applied on the voxel to obtain the spectra and the relative percentage of lipid determined, by integration of the lipid peak(Reference So, Yu and Kuo18).
Adipocyte cell size and number
White adipose tissue (epididymal depot) was collected and finely minced in Dulbecco's modified Eagle's medium solution supplemented with 4 % (w/v) glucose and collagenase (1 mg/ml). The tissue suspension was incubated in a vibrating water-bath set at 140 cycles/min for 45 min at 37°C. The suspension was filtered through a polypropylene mesh (400 μm) after complete digestion of the tissue. This was followed by washing twice using Dulbecco's modified Eagle's medium solution containing 4 % (w/v) bovine serum albumin and 1 mg/40 ml trypsin inhibitor. An aliquot of the adipocytes was collected, diluted with trypan blue and the cells were counted in a haemocytometer. Cell number and size were calculated from the images acquired using software Cellprofiler after setting an appropriate threshold.
Statistical analysis
All data are presented as means with their standard errors. The data were statistically analysed using GraphPad Software (GraphPad Software). Two-way ANOVA with Bonferroni post hoc tests were used for the analysis of body weights, food intake and fluorescent in situ hybridisation data using time (d/weeks) and treatment as two independent variables. Other parameters such as whole body, liver and muscle lipid levels, adipocyte size and number were analysed by an unpaired t test to compare the two groups. P values < 0·05 were considered significant.
Results
Body weight, cumulative food and dahi intake
The feeding of the probiotic brought no significant changes in body weight and cumulative body-weight gains of mice over control dahi supplementation. Similarly, no differences in cumulative food and dahi intake were observed between the two groups (Figs. 1–4).
Effect of control and probiotic dahi on body weight in mice fed a high-fat diet. Mice were fed either control (◇) or probiotic (■) dahi along with a high-fat diet and their body weight was measured daily over the 8-week intervention period. Values are means corresponding to twelve mice, with standard errors represented by vertical bars.
Effect of control and probiotic dahi on cumulative body-weight gain in mice fed a high-fat diet. Mice were fed either control (◇) or probiotic (■) dahi along with a high-fat diet and their body weight was measured daily. Cumulative body-weight gain was calculated by subsequently adding the daily weight gains of mice over the 8-week intervention period. Values are means corresponding to twelve mice, with standard errors represented by vertical bars.
Effect of control and probiotic dahi on cumulative food intake in mice fed a high-fat diet. Mice were fed either control (◇) or probiotic (■) dahi along with a high-fat diet and their food intake was measured daily. Cumulative food intake was calculated by subsequently adding the daily food intake of mice over the 8-week intervention period. Values are means corresponding to twelve mice, with standard errors represented by vertical bars.
Effect of control and probiotic dahi on cumulative dahi intake in mice fed a high-fat diet. Mice were fed either control (◇) or probiotic (■) dahi along with a high-fat diet and their dahi intake was measured daily. Cumulative dahi intake was calculated by subsequently adding the daily dahi intake of mice over the 8-week intervention period. Values are means corresponding to twelve mice, with standard errors represented by vertical bars.
Whole-body fat composition
Measurements of whole-body adiposity and fat deposition in visceral and subcutaneous fat depots by MRI revealed no differences between the control and probiotic dahi-fed mice. Lipid levels in the IHCL and IMCL fluids also differed non-significantly between the two groups (Table 2).
Effect of control and probiotic dahi on tissue weights, body fat composition, and adipocyte size and number in mice fed a high-fat diet (Mean values with their standard errors, n 12)
WAT, white adipose tissue.
Adipocyte size and number
The average adipocyte size was measured to be 78·65 (sem 12·31) and 69·96 (sem 12·41) μm for the control and probiotic dahi-fed groups, respectively (Table 2). Adipocyte size, though apparently smaller in the case of the probiotic dahi-fed group, was statistically non-significant. Adipocyte number also differed non-significantly between the two groups (Table 2).
Microbial modulation
Data on different bacterial groups in the faecal and caecal samples, expressed as log10 cells/g, are shown in Figs. 5 and 6. The baseline number of total bacteria as measured by 4′,6-diamidino-2-phenylindole dihydrochloride and other bacterial groups targeted by different probes in the faecal samples were similar between the control and probiotic dahi-fed groups before the start of the dietary intervention. At week 8, a significant difference was observed in bifidobacterial numbers (P < 0·05) between the faecal samples of the control and probiotic dahi-fed groups (7·59 (sem 0·12) v. 8·04 (sem 0·10) log10 cells/g). However, no difference was observed in the other bacterial groups such as total bacteria, E. rectale–C. coccoides and mouse intestinal bacteria. A similar trend was observed in the caecal contents collected terminally, with a significant difference in bifidobacterial numbers. The feeding of a high-fat diet along with control yogurt resulted in a significant decrease in the faecal Lactobacillus–Enterococcus counts at week 8 when compared with baseline. However, probiotic (L. acidophilus) feeding was found to maintain a similar count of Lactobacillus–Enterococcus after 8 weeks under high-fat dietary conditions.
Effect of control and probiotic dahi on faecal bacterial groups in mice fed a high-fat diet. Mice were fed either control or probiotic dahi along with a high-fat diet. Faecal samples were collected either at baseline (
, control; 
, probiotic) or after 8 weeks at the end of dietary intervention (□, control; 
, probiotic). The different bacterial groups belonging to mouse intestinal bacteria (MIB), Eubacterium rectale–Clostridium coccoides (EREC), and total bifidobacteria (BIF) and Lactobacillus–Enterococcus (LAB) were measured using fluorescent in situ hybridisation. Total bacteria were determined by 4′,6-diamidino-2-phenylindole dihydrochloride (DAPI) staining. Values are means corresponding to six mice, with standard errors represented by vertical bars. *P < 0·05, ***P < 0·001.
Effect of control and probiotic dahi on caecal bacterial groups in mice fed a high-fat diet. Mice were fed either control or probiotic dahi along with a high-fat diet. Caecal contents were collected after 8 weeks at the end of dietary intervention from both control (
) or probiotic (□) dahi-fed mice. The different bacterial groups belonging to mouse intestinal bacteria (MIB), Eubacterium rectale–Clostridium coccoides (EREC), and total bifidobacteria (BIF) and Lactobacillus–Enterococcus (LAB) were measured using fluorescent in situ hybridisation. Total bacteria were determined by 4′,6-diamidino-2-phenylindole dihydrochloride (DAPI) staining. Values are means corresponding to six mice, with standard errors represented by vertical bars. * P < 0·05.
Discussion
The study aimed to investigate the effects of feeding a probiotic, L. acidophilus NCDC 13, on intestinal microbiota modulation and the progression of obesity in diet-induced obese mice under a high-fat dietary environment. The probiotic was fed in the form of a dahi preparation and the results were compared with the control dahi-fed animals. The feeding of the probiotic dahi resulted in no significant differences in body weight, cumulative body-weight gain, cumulative food and dahi intake in comparison with the control dahi-fed group of mice. The present results are consistent with the observations of Esposito et al. (Reference Esposito, Iacono and Bianco19) who reported no change in the body weight and food intake of rats fed a high-fat diet supplemented with a commercial probiotic preparation, VSL#3, containing multiple strains of Lactobacillus, Bifidobacterium and Streptococcus thermophilus. Similarly, no change in body weight as well as food intake has been reported in rats fed high-cholesterol diets supplemented with L. plantarum (Reference Wang, Xu and Xi20). However, body-weight gain was rather high with no change in food consumption in rats fed non-fermented milk produced with L. gasseri (Reference Usman and Hosono21). On the contrary, Lee et al. (Reference Lee, Park and Seok22) observed that administration of L. rhamnosus, a conjugated linoleic acid-producing bacterium, resulted in the lowering of body-weight gain in mice with no change in average energy intake. In another study, the probiotic strain, L. paracasei, reduced the body-weight gain in rats fed a high-fat diet with no change in food consumption(Reference Tanida, Shen and Maeda23). Similarly, no effect on food intake was observed, despite an increase in the plasma levels of peptide YY and neuropeptide Y, the hormones implicated in central appetite regulation, when rats were fed a mixture of probiotics (L. delbrueckii and B. lactis) and a prebiotic (inulin)(Reference Lesniewska, Rowland and Cani24). The efficacy of a probiotic is highly strain-specific, which could account for variations in the outcome of studies employing different probiotic strains.
We measured whole-body adiposity and lipid levels in the IHCL and IMCL fluids, and lipid deposition in visceral and subcutaneous fat depots in both groups of mice using MRI/S imaging techniques. However, we found no significant differences in the aforementioned parameters, which are in agreement with the observed non-significant difference in the adipocyte size of mice. There are recent reports which demonstrated the effects of different probiotic strains in the lowering of adipose tissue mass(Reference Esposito, Iacono and Bianco19, Reference Lee, Park and Seok22, Reference Choi, Bae and Kang25) and adipocyte size(Reference Sato, Uzu and Yoshida26–Reference Hamad, Sato and Uzu28). This discrepancy may be explained by the difference in the mode of the probiotic administration. In the present study, the probiotic was supplemented in the form of a fermented milk preparation along with dahi starter cultures, whereas in other studies, the probiotic was administered alone either by oral administration or was incorporated in the skimmed milk. Recent elegant studies in axenic mice harbouring simplified gut bacterial members have shown that the presence of two different bacterial species alters the metabolic activity of each other(Reference Mahowald, Rey and Seedorf29, Reference Samuel and Gordon30). Therefore, it is also possible that though the multiple starter strains of Lactococcus sp. used in the preparation of dahi did not affect the viability of the probiotic in the product, they might have hindered the efficacy of the probiotic culture in the gut of the animals. Also, it is demonstrated that permanent colonisation of species from common probiotic drinks has proven difficult(Reference Lee, Ho and Low31). There could be interactions between two cultures during fermentation in vitro but the interaction in the gut in vivo is far more complex. The desired property is the proliferation of the probiotic in the presence of the starter culture in the product so that the probiotic reaches the gut in good amounts.
In the present study, the number of total Lactobacillus–Enterococcus recovered in the faeces and caecal contents of mice after 8 weeks of probiotic supplementation did not differ significantly when compared with the control dahi group, which lacked the probiotic bacteria. Nevertheless, the probiotic supplementation maintained the total Lactobacillus–Enterococcus count (at week 8) similar to the initial baseline count (at week 0), while in the control dahi group, a significant reduction in the total lactobacilli counts was observed at week 8 under a high-fat diet. This may be indicative of the inability of the probiotic strain used in the present study to colonise the intestinal walls and proliferate in the intestinal environment under high-fat dietary conditions. L. acidophilus NCDC 13 has been shown to exhibit probiotic properties in terms of acid, and bile tolerance under in vitro conditions by other workers(Reference Kushal, Anand and Chander10). However, it is generally stated that in vitro studies are not often predictive of the in vivo situation, since conditions are so vastly different(Reference Sanders and Klaenhammer32). On the contrary, Kushal et al. (Reference Kushal, Anand and Chander11) reported an increase in total Lactobacillus counts in the faeces of mice (maintained on a normal chow diet) after 8 d of L. acidophilus NCDC 13 supplementation. However, a conventional plating technique in selective media was used for the enumeration of total faecal Lactobacillus when compared with fluorescent in situ hybridization hybridisation used in the present study. The fluorescent in situ hybridisation hybridisation technique relies upon specific hybridisation of the 16 s rRNA-based probes in the target organism and provides a better measure of the bacterial number accurately. It also alleviates the laborious cultivation and purification tasks(Reference Amann, Ludwig and Schleifer33). The dose given may also make a difference in the potential to colonise the intestinal walls. Kushal et al. (Reference Kushal, Anand and Chander12) used sterile skimmed milk to provide 1 × 1010 cfu/ml of L. acidophilus NCDC 13 per mouse per d, while in the present study, mice consumed about 2 g of the probiotic dahi, which provided 1–2 × 108 cfu of L. acidophilus NCDC 13 per mouse per d. The difficulty in achieving colonisation has been highlighted by Lee et al. (Reference Lee, Ho and Low31) who demonstrated decreased replication rates of probiotic organisms in the mouse gut. The addition of prebiotic carbohydrates has been shown to enhance the survival and proliferation of probiotics both under in vitro (Reference Donkor, Nilmini and Stolic34) and in vivo conditions(Reference Bielecka, Biedrzycka and Majkowska35). Therefore, it would be interesting to supplement L. acidophilus NCDC 13 with suitable prebiotics to enhance the survivability of the probiotic in the gut and to study its potential anti-obesity role.
We have reported here a significant increase in the number of Bifidobacterium in both the caecal contents and faeces of mice fed the probiotic dahi preparation. This is in contrast to the expected rise in Lactobacillus numbers. Dinoto et al. (Reference Dinoto, Suksomcheep and Ishizuka36) also reported similar results where supplementation of a probiotic, B. breve, into the diet of rats produced no change in total bifidobacteria counts as well as no recovery of the probiotic strain in the caecal contents, whereas the number of total lactobacilli increased. Some other reports have also revealed an increase in bifidobacterial numbers upon administration of probiotic lactobacilli strains in different animal models(Reference Zhang, Xu and Liu37, Reference Nanda Kumar, Balamurugan and Jayakanthan38). Thus, certain unexplained metabolic interactions are undergoing in the colonic environment upon supplementation of dahi and probiotic cultures. With regard to anti-obesogenesis, an increase in bifidobacterial number is desirable and has been positively correlated with weight loss(Reference Cani, Neyrinck and Fava39) and the lean phenotype in rodents(Reference Waldram, Holmes and Wang40).
Thus, it is concluded that supplementation of L. acidophilus NCDC 13 did not result in the lowering of body weight and food intake. However, supplementation of L. acidophilus NCDC 13 in the diet of high-fat fed mice increased bifidobacterial numbers in the caecal contents as well as in the faeces. Further investigation with the probiotic could provide further insights into the mechanism of probiotic functioning and its potential anti-obesity role.
Acknowledgements
This study was funded by grants from the MRC, BBSRC, NIHR, an Integrative Mammalian Biology (IMB) Capacity Building Award, an FP7-HEALTH-2009-241 592 EurOCHIP grant and funding from the NIHR Imperial Biomedical Research Centre Funding Scheme. T. A. was supported by a Commonwealth split-site fellowship and G. F. was supported by an NIHR Senior investigator award. T. A., R. S. and G. F. conceived the study. T. A., J. A., G. G., K. T., R. S., J. B. and G. F. developed the study protocol. T. A., J. A. and K. T. carried out the studies. T. A., J. A., G. G., K. T., R. S., J. B. and G. F. analysed the data and wrote the manuscript. The authors state that there is no conflict of interest.
