Wageningen Academic  P u b l i s h e r s

Beneficial Microbes, June 2010; 1(2): 189-196

Probiotics lower plasma glucose in the high-fat fed C57BL/6J mouse

http://www.wageningenacademic.com/doi/pdf/10.3920/BM2009.0036 - Wednesday, January 25, 2017 4:59:39 PM - IP Address:37.44.207.174

U. Andersson1, C. Bränning2, S. Ahrné3, G. Molin3, J. Alenfall4, G. Önning4, M. Nyman2 and C. Holm1 1Lund University, Department of Experimental Medical Science, Division of Diabetes, Metabolism and Endocrinology, BMC,

221 84 Lund, Sweden; 2Lund University, Department of Food Technology, Engineering and Nutrition, Division of Applied Nutrition and Food Chemistry, P.O. Box 124, 221 00 Lund, Sweden; 3Lund University, Department of Food Technology, Engineering and Nutrition, Division of Food Hygiene, P.O. Box 124, 221 00, Lund, Sweden; 4Probi AB, Lund, 223 70, Sweden; [email protected] Received: 31 July 2009 / Accepted: 17 October 2009 © 2010 Wageningen Academic Publishers

Abstract Today, the gut microbiota is considered a key organ in host nutritional metabolism and recent data have suggested that alterations in gut microbiota contribute to the development of type 2 diabetes and obesity. Accordingly, a whole range of beneficial effects relating to inflammation and gut health have been observed following administration of probiotics to both humans and different animal models. The objective of this study was to evaluate the metabolic effects of an oral probiotic supplement, Lactobacillus plantarum DSM 15313, to high-fat diet (HFD) fed C57BL/6J mice, a model of human obesity and early diabetes. The mice were fed the experimental diets for 20 weeks, after which the HFD had induced an insulin-resistant state in both groups compared to the start of the study. The increase in body weight during the HFD feeding was higher in the probiotic group than in the control group, however, there were no significant differences in body fat content. Fasting plasma glucose levels were lower in the group fed the probiotic supplement, whereas insulin and lipids were not different. Caecal levels of short-chain fatty acids were not significantly different between the groups. An oral glucose tolerance test showed that the group fed probiotics had a significantly lower insulin release compared to the control group, although the rate of glucose clearance was not different. Taken together, these data indicate that L. plantarum DSM 15313 has anti-diabetic properties when fed together with an HFD. Keywords: Lactobacillus plantarum, insulin sensitivity, glucose tolerance, body weight, short-chain fatty acids

1. Introduction Probiotics are defined as live microorganisms that survive passage through the gastro-intestinal tract and have beneficial effects on the host (Fuller, 1989). The effects that probiotics exert on gastrointestinal tract function and disease have been extensively studied (Guarner et al., 2003). Recently the gut microbiota was identified as an environmental factor involved in the control of body weight and energy metabolism (Backhed et al., 2004, 2007; Ley et al., 2005, 2006; Turnbaugh et al., 2006). Lactobacilli are part of the healthy gastrointestinal tract and several species are used as probiotics due to the health promoting effects attributed to these bacteria. In view of the global epidemic

of obesity and type 2 diabetes, there is a growing need for preventive actions of the disease. A few studies have pointed to the potential of different species of Lactobacillus to reduce the risk of diabetes onset in different animal models of non-insulin dependent diabetes. Yadav et al. (2007) recently reported that a diet supplemented with dahi, a fermented dairy product containing Lactobacillus acidophilus and Lactobacillus casei, significantly delayed the progression of high fructose-induced glucose intolerance, hyperglycaemia, hyperinsulinaemia, dyslipidemia and oxidative stress in rats. Also, Matsuzaki et al. (1997) have shown that oral administration of L. casei significantly decreased plasma glucose levels in KK-Ay mice. Tabuchi et al. (2003) reported that feeding of Lactobacillus rhamnosus

ISSN 1876-2833 print, ISSN 1876-2891 online, DOI 10.3920/BM2009.0036

189

http://www.wageningenacademic.com/doi/pdf/10.3920/BM2009.0036 - Wednesday, January 25, 2017 4:59:39 PM - IP Address:37.44.207.174

U. Andersson et al.

GG improved glucose tolerance in rats with streptozotocininduced diabetes. A strain of Lactobacillus plantarum (299v) has, in a placebo-controlled pilot-study, been shown to significantly reduce the systolic blood pressure and the plasma concentrations of insulin, leptin, fibrinogen, F2-isoprostanes and IL-6 (Naruszewicz et al., 2002). The mechanisms whereby lactobacilli exert these beneficial metabolic effects are poorly understood, but may be related to immunological effects and/or antagonistic effects towards pro-inflammatory components of the gut microbiota, e.g. producers of lipopolysaccharides (endotoxins) as Enterobacteriaceae and Bacteroides. Furthermore, different strains of L. plantarum have been shown to improve the mucosal barrier effects in animal models (Fak et al., 2008; Mao et al., 1996; Osman et al., 2007) as well as in humans (Klarin et al., 2008). Hence, the risk of translocation of gut bacteria and adverse gut-components such as LPS can be decreased by probiotic administration. Fermentation of carbohydrates in the colon is a major source of energy for bacterial growth. The main end-products of fermentation are short-chain fatty acids (SCFA) that may be absorbed by the host, utilised by the colonic mucosa or excreted in the faeces (Wolin, 1981). The aim of this study was to evaluate the long-term metabolic effects of feeding a supplement of L. plantarum DSM 15313 (Lp DSM 15313) using the high-fat diet (HFD) fed C57BL/6J mouse, a frequently used mouse model of human obesity and insulin resistance, as experimental model (Surwit et al., 1988; Winzell et al., 2004).

2. Materials and methods Animals and study design Female C57BL/6J mice 6-8 weeks old, weighing 19.1±0.1 g were purchased from Taconic (Skensved, Denmark). The animals were maintained in a temperature-controlled room on a 12-h light-dark cycle (light on at 07:00). The study was approved by the Animal Ethics Committee (Lund, Sweden) and is in accordance with the Council of Europe Convention (ETS 123). After one week of acclimatisation the mice were randomly divided into two groups (n=20/ group, 10 mice per cage) and shifted to either a control HFD (ctrl) or an HFD supplemented with Lp DSM 15313. The mice were fed the experimental diets ad libitum for 20 weeks. Body weights were monitored throughout the study period. An oral glucose tolerance test (OGTT) was performed after 4 weeks in 7 mice from each group. After 18 weeks 9 mice from each group underwent an intravenous glucose tolerance test (IVGTT) and another 9 mice underwent an OGTT. Blood samples were taken before the start of the study, after 12 weeks and at the end of the study from over-night fasted (12 h) anesthetised mice to measure basal plasma levels of glucose, insulin, total 190

cholesterol, triacylglycerol (TAG) and non-esterified fatty acids (NEFA). Serum amyloid A (SAA), a marker of systemic inflammation, IL-6 and adiponectin were measured in plasma at the end of the study. Body composition was determined with dual-energy X-ray absorptiometry (DEXA) at the start of the study, at week 5, 12 and at the end of the study using a Lunar PIXImus (Brommage, 2003). At the end of the study islets of Langerhans were isolated for in vitro insulin secretion analysis and the caecal content was collected for carboxylic acid (CA) analysis.

Diets The administered probiotic was L. plantarum DSM 15313 (strain = HEAL 19) which originates from the gastrointestinal mucosa of a healthy human subject (Probi AB, Lund, Sweden). The production process was built as a sequential fermentation, where one step served as inoculum for the next, and the final product was concentrated by filtration. The filtrate was mixed with bacterial suspension medium (4 mmol/l K2HPO4, 1.4 mmol/l KH2PO4, 2 mmol/l Na-citrate, 0.9 mmol/l MgSO4 × 7 H2O, 12% glycerol), portioned and frozen at -80 °C. The experimental diet was prepared by thoroughly mixing the bacterial suspension with the HFD. The bacterial suspension medium alone was used in the control diet (Table 1). Six batches of each diet were prepared during the 20-week study period. The diets were stored at -20 °C in single portion packs. New portions of food were provided to the mice every day and the left over food was discarded. Food intake was monitored by weighing the left over food in each cage and was measured on a cage-by-cage basis. The diets from each batch were analysed for viable count of lactobacilli. Table 1. Composition (g/kg) of experimental diets. Ingredient

Casein DL-methionine Maltodextrin Sucrose Lard Cellulose Mineral mix Sodium bicarbonate Potassium citrate Vitamin mix Choline bitartrate Total Energy (kJ/g) Bacterial suspension Bacterial suspension medium

ctrl

Lp DSM 15313

157.3 1.4 117.3 311.2 255.1 112 27.5 7.2 2.7 6.9 1.4

157.3 1.4 117.3 311.2 255.1 112 27.5 7.2 2.7 6.9 1.4

1000 1.83 120

1000 1.83 40 80

Beneficial Microbes 1(2)



http://www.wageningenacademic.com/doi/pdf/10.3920/BM2009.0036 - Wednesday, January 25, 2017 4:59:39 PM - IP Address:37.44.207.174

PCR Colonies suspected to be Lp DSM 15313 were counted and representative colonies were selected, purified on Rogosa agar and dissolved in distilled water. In brief, isolates were cultivated overnight in MRS broth (Biokar Diagnostics, Beauvais, France) supplemented with 1% glucose. Cells from 100 μl of culture were collected by centrifugation and washed twice in 1 ml of sterile distilled water. The cells were resuspended in 0.25 ml of sterile distilled water and then disintegrated by vigorous shaking with glass beads using an IKA Vibrax VXR shaker (IKA Labortechnik, Staufen, Germany) at maximum speed for 20 min at 4 °C. The disrupted cells were pelleted by centrifugation and 1 μl of the supernatant was used for the PCR. The identification was performed using Randomly Polymorphic DNA (RAPD) according to the protocol of Johansson et al. (1995). The PCR was carried out in a DNA Thermal Cycler 480 (Eppendorf, Hamburg, Germany) using a 9-mer primer (ACGCGCCCT; Scandinavian Gene syntheses AB, Köping, Sweden) at a concentration of 478 μg/ml and Master mix (TaqPCR Master Mix; Qiagen Nordic, Solna, Sweden). Gel electrophoresis was carried out as described by Johansson et al. (1995), and the DNA fragment profiles of the isolates were compared visually with the fragment profile of Lp DSM 15313 that was run as a control on all gels. Faeces collected from the cages at three different time points were analysed for lactobacilli as above.

OGTT and IVGTT In the OGTT 12 h fasted mice were anesthetised with 0.5 mg fluanisone, 0.02 mg fentanyl (Hypnorm; Janssen, Beerse, Belgium) and 0.25 mg Midazolam (Dormicum; HoffmanLaRoche, Basel, Switzerland) per mouse. D-glucose (75 mg in 0.5 ml) was given by intragastric gavage. Blood samples were drawn from the retrobulbar, intraorbital, capillary plexus at 0, 15, 30, 60 and 120 min after glucose administration. After immediate centrifugation plasma was collected and stored at -20 °C until analyses of glucose and insulin. In the IVGTT, D-glucose (1g/kg) was injected in a tail vein in 12-h fasted anesthetised mice and blood samples were drawn at 0, 1, 5, 10, 20, 50 and 75 min after glucose administration and handled as above.

Assays of plasma samples Plasma was prepared as above. Glucose, total cholesterol, TAG (Infinity, Thermo Electron Melbourne, Australia) and NEFA (NEFA C; Wako Chemicals, Neuss, Germany) were measured enzymatically. Insulin and adiponectin were measured radioimmunochemically (RIA; Linco Research, St. Charles, MI, USA). SAA was measured with ELISA (Tridelta Development Ltd., Maynooth, Ireland). IL-6 was measured using a Meso Scale Discovery assay (MSD, Gaithersburg, MA, USA) and analysed on a SECTOR 2400 Beneficial Microbes 1(2)

Antidiabetic actions of probiotics

instrument. SAA and IL-6 were analysed in an aliquot of plasma that had been snap-frozen in liquid nitrogen and stored at -80 °C.

Isolation of islets and insulin secretion assay Islets were isolated by collagenase digestion and hand picked under a stereo microscope. The islets were first allowed to recover in HEPES balanced salt solution (HBSS, 114 mmol/l NaCl, 4.7 mmol/l KCl, 1.2 mmol/l KH2PO4, 1.16 mmol/l MgSO4, 20 mmol/l HEPES, 2.5 mmol/l CaCl2, 25.5 mmol/l NaHCO3, 0.1% BSA; pH 7.2) containing 5.6 mmol/l glucose for 60 min in an incubator at 37 °C and 5% CO2. Batches (n=8) of three islets were transferred to a 96-well plate containing 200 μl of the same buffer per well with an addition of 3.3, 8.6 or 16.7 mmol/l glucose. After another 60 min of incubation a sample was removed from each well for measurement of insulin by RIA. Total insulin content in islets was measured after sonication of islets in acid ethanol (2% H2SO4) followed by three freeze/thaw cycles and centrifugation for 5 min at 10,000g. Insulin was measured in the supernatant by RIA.

CA analysis A gas liquid chromatography (GLC) method (Richardson et al., 1989) was used to analyse the amount of CAs, i.e. acetic, propionic, butyric, succinic, lactic, isobutyric, isovaleric, valeric, caproic and heptanoic (the last five acids are referred to as minor in Table 3) acid in the caecal content. The caecal contents were homogenised with an Ultra Turrax® T25 basic (IKA®-Werke, Staufen, Germany) after adding internal standard (2-ethylbutyric acid; Sigma Chemical Company, St. Louis, MO, USA). Hydrochloric acid was added to protonise the CAs to make the extraction of the CAs in diethyl ether possible. After being silylated with n-(tert-butyldimethylsilyl)-n-methyl trifluoroacetamide (MTBSTFA; Sigma Chemical Company), the samples were left to stand for 48 h to complete derivatisation. The samples were then injected onto an HP-5 column (GLC, HP 6890; Hewlett Packard, Wilmington, DE, USA), and Chem Station software (Hewlett Packard) was used for the analysis.

Statistics Data are presented as means with standard error of the mean (SEM). To compare body weights, body fat content, glucose, insulin, cholesterol, TAG and NEFA during the study 2-way ANOVA with Bonferroni post-test was used. For the SAA, adiponectin and area under the curve (AUC) comparisons Student’s t-test was used. The above analyses were performed using GraphPad Prism version 4.00 for Windows (GraphPad Software, San Diego, CA, USA). CA comparison was performed using 1-way ANOVA with Bonferroni post-test (Minitab 14.3 software; Minitac Inc., State College, PA, USA). 191

50

Diets

40

The overall energy intake during the study was not significantly different between the two groups (49.6±0.6) kJ vs. 50.9±0.6) kJ). The concentration of lactobacilli in the six batches of diets ranged from 1.2×109 to 5.6×109 cfu/g diet with a mean of 2.4×109 cfu/g. The mice had a mean intake of 7×109 cfu/day, ranging from 3×109 to 2×1010 cfu/day. The concentration in the ctrl diet was below 100 cfu lactobacilli/g. The concentration of Lp DSM 15313 in faeces from the mice was measured at week 2, 8 and 16 and was approximately 8.8×108, 1.3×109 and 4.8×108 cfu/g, respectively, in the Lp DSM 15313 group. Faeces from mice fed the control diet contained below 500 cfu/g at all time-points.

Body weight and body fat content Body weight gain during the study period was more pronounced in the group fed a supplement of Lp DSM 15313 compared to the ctrl group (21.5±1.1 g vs. 17.1±1.1 g, P