The Journal of Nutrition Nutrition and Disease
Exopolysaccharide-Producing Probiotic Lactobacilli Reduce Serum Cholesterol and Modify Enteric Microbiota in ApoE-Deficient Mice1,2 Lis EE London,3,6 Arun HS Kumar,4 Rebecca Wall,3 Pat G Casey,3,5 Orla OÕSullivan,6 Fergus Shanahan,3 Colin Hill,3,5 Paul D Cotter,3,6 Gerald F Fitzgerald,3,5 R Paul Ross,3,5 Noel M Caplice,4 and Catherine Stanton3,6* 3 Alimentary Pharmabiotic Centre, 4Centre for Research in Vascular Biology, and 5Department of Microbiology, University College Cork, Cork, Ireland; and 6Teagasc, Food Research Centre Moorepark, Fermoy, Cork, Ireland
Abstract death and disability. Objectives: The aim of this study was to assess the impact of dietary administration of exopolysaccharide-producing probiotic Lactobacillus cultures on lipid metabolism and gut microbiota in apolipoprotein E (apoE)–deficient mice. Methods: First, we examined lipid metabolism in response to dietary supplementation with recombinant b-glucan–producing Lactobacillus paracasei National Food Biotechnology Centre (NFBC) 338 expressing the glycosyltransferase (Gtf) gene from Pediococcus parvulus 2.6 (GTF), and naturally exopolysaccharide-producing Lactobacillus mucosae Dairy Product Culture Collection (DPC) 6426 (DPC 6426) compared with the non–b-glucan-producing isogenic control strain Lactobacillus paracasei NFBC 338 (PNZ) and placebo (15% wt: vol trehalose). Second, we examined the effects on the gut microbiota of dietary administration of DPC 6426 compared with placebo. Probiotic Lactobacillus strains at 1 3 109 colony-forming units/d per animal were administered to apoE2/2 mice fed a high-fat (60% fat)/ high-cholesterol (2% wt:wt) diet for 12 wk. At the end of the study, aortic plaque development and serum, liver, and fecal variables involved in lipid metabolism were analyzed, and culture-independent microbial analyses of cecal content were performed. Results: Total cholesterol was reduced in serum (P < 0.001; ;33–50%) and liver (P < 0.05; ;30%) and serum triglyceride concentrations were reduced (P < 0.05; ;15–25%) in mice supplemented with GTF or DPC 6426 compared with the PNZ or placebo group, respectively. In addition, dietary intervention with GTF led to increased amounts of fecal cholesterol excretion (P < 0.05) compared with all other groups. Compositional sequencing of the gut microbiota revealed a greater prevalence of Porphyromonadaceae (P = 0.001) and Prevotellaceae (P = 0.001) in the DPC 6426 group and lower proportions of Clostridiaceae (P < 0.05), Peptococcaceae (P < 0.001), and Staphylococcaceae (P < 0.01) compared with the placebo group. Conclusion: Ingestion of exopolysaccharide-producing lactobacilli resulted in seemingly favorable improvements in lipid metabolism, which were associated with changes in the gut microbiota of mice. J Nutr 2014;144:1956–62. Keywords: probiotics, lactobacilli, exopolysaccharide, cholesterol, gut microbiota, lipid metabolism
Introduction Cardiovascular disease (CVD)7 (1) is a leading cause of death and disability globally (1). The human intestinal microbiota has been implicated in the development of obesity (2), insulin resistance (3), and more recently, CVD risk (4, 5). Patients with 1
Supported by the Irish Government, Department of Agriculture Food and Marine (DAFM); the Science Foundation of Ireland–funded Centre for Science, Engineering, and Technology; the Alimentary Pharmabiotic Centre and Enterprise Ireland Commercialisation Fund Contract (reference: CF/2013/3030A/B); the European Union Seventh Framework Programme–funded FIBEBIOTICS: Dietary Fibers Supporting Gut and Immune Function (EC proposal 289517). LEE London received a Teagasc Walsh Fellowship (2007–2010). * To whom correspondence should be addressed. E-mail: catherine.stanton@ teagasc.ie.
1956
symptomatic atherosclerosis were shown to harbor an altered gut metagenome compared with healthy controls (6), whereas a 2
Author disclosures: LEE London, AHS Kumar, R Wall, PG Casey, O OÕSullivan, F Shanahan, C Hill, PD Cotter, GF Fitzgerald, RP Ross, NM Caplice, and C Stanton, no conflicts of interest. 7 Abbreviations used: Cd36, cluster of differentiation 36; CVD, cardiovascular disease; Cyp7a1, cholesterol 7a-hydroxylase; DPC, Dairy Product Culture Collection; DPC 6426, Lactobacillus mucosae Dairy Product Culture Collection 6426; Fas, FA synthase; GTF, Lactobacillus paracasei National Food Biotechnology Center 338 expressing the glycosyltransferase gene from Pediococcus parvulus 2.6; Gtf, glycosyltransferase; Hmg-coA, 3-hydroxy-3-methylglutaryl-coenzyme A; MRS, de Man, Rogasa, and Sharpe; NFBC, National Food Biotechnology Center; PNZ, Lactobacillus paracasei National Food Biotechnology Center 338; rRNA, ribosomal RNA; Srebp-1c, sterol regulatory element-binding protein 1; sVCAM-1, soluble vascular cell adhesion molecule 1; TC, total cholesterol; TMAO, trimethylamine-N-oxide.
ã 2014 American Society for Nutrition. Manuscript received February 11, 2014. Initial review completed March 26, 2014. Revision accepted September 15, 2014. First published online October 15, 2014; doi:10.3945/jn.114.191627.
Downloaded from jn.nutrition.org by guest on December 4, 2017
Background: Probiotic bacteria have been associated with a reduction in cardiovascular disease risk, a leading cause of
Methods Experimental animals and diet. All mouse studies were undertaken in accordance with the Department of Health and Children of the Irish Government. A license and permission for the study were obtained from the Department of Health, Ireland. Male Apoetm1Unc/J mice, which
develop atherosclerostic lesions similar to the process of atherogenesis in humans (22), that were between 10 and 12 wk old at the beginning of the experiment were obtained from JAX (The Jackson Laboratory through Charles River Laboratories International). The mice were caged either individually, in pairs, or in groups of 4–6 per cage. Group housing differences were distributed into the different feeding groups. All mice consumed a basal diet (23) ad libitum for 2 wk to stabilize all metabolic conditions, with free access to water at all times. After the acclimatization period, mice were divided into 4 groups (n = 9) and fed a high-fat (60% kcal from fat), high-cholesterol (2% wt:wt) diet (23). Mice were subjected to the following daily dietary interventions for 12 wk: the GTF group were administered ;109 CFU/(mouse d) of b-glucan–producing L. paracasei NFBC 338 expressing the Gtf gene from P. parvulus 2.6 in drinking water, the PNZ group were administered ;109 CFU/(mouse d) of L. paracasei NFBC 338 in drinking water, the DPC 6426 group were administered ;109 CFU/(mouse d) of DPC 6426 in drinking water, and the placebo group were administered resuspended freeze-dried powder containing trehalose (15% wt:vol) as a placebo control in drinking water. Mice were housed in an isolator station, exposed to a 12-h light/ dark cycle, and maintained at a constant temperature of 25°C. Body weight and food intake were monitored weekly. After 12 wk of being fed experimental diets, serum was collected by retro-orbital blood sampling from mice not deprived of food and stored at 4°C for 30 min, followed by centrifugation at 10,000 3 g for 5 min, and divided into aliquots and stored at 220°C until processing and mice were killed by cervical dislocation. Tissues were removed, blotted dry, weighed, and frozen in liquid nitrogen and stored at 280°C until use. Preparation and administration of lactobacilli. Chloramphenicolresistant and recombinant cultures of L. paracasei NFBC 338 were activated and propagated 3 times in de Man, Rogasa, and Sharpe (MRS) medium (Difco Laboratories) containing 5 mg/L chloramphenicol (Sigma-Aldrich) at 37°C anaerobically (21). Rifampicin-resistant variants of L. mucosae DPC 6426 (originally isolated from bovine feces at Teagasc, Moorepark Food Research Centre) were activated and propagated 3 times in MRS containing 5 mg/L rifampicin (Sigma-Aldrich) at 37°C anaerobically. Fermentation of L. mucosae DPC 6426 was undertaken in MRS broth containing 5% (wt:vol) sucrose (SigmaAldrich) at 37°C anaerobically for 20 h and 18 h for both L. paracasei NFBC 338 (GTF and PNZ) cultures. After fermentation, the cultures were washed twice in PBS (Sigma-Aldrich) and resuspended at a concentration of ;2 3 1010 cells/mL in 15% (wt:vol) trehalose (SigmaAldrich). Aliquots were freeze-dried (VirTis AdVantage Freeze Dryer; SP Industries) as outlined by Wall et al. (24). Viable cells in freezedried powders (produced weekly) containing L. paracasei NFBC 338 and L. mucosae DPC 6426 were confirmed weekly for every batch of powder produced and involved enumeration of the strains on MRS agar supplemented with appropriate antibiotic after anaerobic incubation for 72 h at 37°C. Individual mice consumed ;1 3 109 live microorganisms/d. This was achieved by resuspending appropriate quantities of freeze-dried powder in the water that mice consumed ad libitum. Microbial analysis. Fresh fecal samples were taken from experimental mice weekly and analyzed for the presence of b-glucan–producing and non–b-glucan-producing L. paracasei NFBC 338 and exopolysaccharidesproducing L. mucosae DPC 6426. Microbial analysis of the fecal samples involved enumeration of the strains on MRS agar supplemented with appropriate antibiotic after anaerobic incubation for 72 h at 37°C. Preparation and analysis of aortas. The whole length of the aorta was removed and fixed in a 4% (wt:vol) paraformaldehyde (Sigma-Aldrich) solution in PBS. Quantification of the ‘‘en face’’ plaque area was performed as previously described (25, 26) and inspected under the microscope for quantification of atherosclerotic lesions. Lipid-containing plaque area was determined as a percentage of Oil Red O-stained area from the total aortic surface area by using AxioVision 4.6 (Carl Zeiss Vision) or NIH Image J software. Measurement of serum variables. A commercially available murine ELISA kit (Quantikine; R&D Systems) was used to quantify serum Bacterial polysaccharides reduce serum cholesterol
1957
Downloaded from jn.nutrition.org by guest on December 4, 2017
link between intestinal microbial production of proatherosclerotic metabolites from dietary phospholipids and atherosclerosis risk was also recently reported (4). Moreover, a mechanistic link has been established between changes in intestinal microbiota (via oral treatment with vancomycin in rats) and myocardial infarction. Intestinal microbial metabolism of L-carnitine, abundant in red meat, to proartherogenic trimethylamine-N-oxide (TMAO) has recently been postulated as a mechanism by which red meat ingestion may be linked to CVD risk (7). In a study by Wang et al. (4), it was also reported that dietary administration of a probiotic mixture comprising Lactobacillus plantarum (Lp299v) and Bifidobacterium lactis (Bi-07) was cardioprotective. Several dietary fibers were shown to significantly decrease serum cholesterol concentrations and thereby reduce risk of CVD (8). b-Glucan is a polysaccharide found in yeast, mushrooms, oats, barley, and bacteria; and b-glucan from whole oats has been recommended by the U.S. FDA (9) as a safe and practical dietary approach for blood cholesterol reduction. Cholesterol-lowering effects of oats, barley, and yeast are well established (10); and b-glucan from oats and yeast has been shown to improve lipid profiles (11). Some lactic acid bacteria produce exopolysaccharides, a wide diversity of carbohydrates varying in structure and monosaccharide composition, which, when ingested, may confer health benefits, including cardioprotective properties such as hypocholesterolemic, antioxidant, and immunomodulatory activities, in addition to prebiotic, antitumor effects, and affect gastrointestinal health (12–15). However, there are little data available from in vivo studies in humans and animal models and in vitro studies that demonstrate beneficial effects of bacterial b-glucan, and exopolysaccharides. In an animal model, the ingestion of bacterial b-glucan led to increased concentrations of serum HDL cholesterol and decreased TG and total cholesterol (TC) concentrations (16). Interestingly, the ingestion of isolated exopolysaccharides enhanced serum cholesterol metabolism (17), reduced TC in serum, LDLcholesterol concentrations in the liver (18), and size of atherosclerotic lesions in rabbits (19). In a human study, the ingestion of nondairy, oat-based foods containing the b-glucan–producing strain Pediococcus parvulus 2.6 resulted in decreased serum cholesterol concentrations (20). We previously reported the functional expression of the pediococcal glycosyltransferase gene (Gtf) in Lactobacillus paracasei National Food Biotechnology Center (NFBC) 338 (21) resulted in b-glucan production. The objectives of the present study were as follows: 1) to compare the impact of dietary supplementation of the b-glucan–producing recombinant L. paracasei NFBC 338 (GTF) with its non–b-glucan-producing isogenic control (PNZ) and a placebo control on lipid metabolism in apolipoprotein E (apoE)–deficient mice fed a high-fat/high-cholesterol diet for 12 wk, 2) to compare the impact of dietary supplementation of the exopolysaccharidesproducing Lactobacillus mucosae Dairy Product Culture Collection (DPC) 6426 (DPC 6426) with a placebo control on lipid metabolism in apoE-deficient mice fed a high-fat/highcholesterol diet for 12 wk, and 3) to investigate the impact on the gut microbial community of the host after ingestion of DPC 6426.
soluble vascular cell adhesion molecule 1 (sVCAM-1) following the manufacturerÕs protocol. Serum b-glucan concentrations were determined with Glucatell, a 1,3-b-D-glucan detection kit (Associates of Cape Cod Inc.) for mice that received b-glucan–producing and non–b-glucanproducing L. paracasei NFBC 338. Serum TC was determined by using the EnzyChrom Cholesterol Assay kit (ECCH-100; BioAssay Systems). Determination of HDL cholesterol in serum samples was undertaken by using the EnzyChrom HDL and LDL/VLDL Assay kit (EHDL-100; BioAssay Systems). TGs in serum were determined by using the LabAssay Triglyceride kit (Wako Diagnostics). All samples and standards were analyzed in duplicate. Lipid extraction from feces and liver. Livers and feces were stored at 280°C before analysis. Liver and fecal lipids were extracted with chloroform:methanol 2:1 (vol:vol) according to the method of Folch et al. (27). The extracted fat samples were dried and dissolved in assay buffer. Liver cholesterol and TG and fecal cholesterol concentrations were determined by using the same kits as used for blood lipid analysis. Extraction of total bile acids from feces was performed as described by Van der Meer et al. (28). Dried samples were reconstituted in 1 mL assay buffer, and total bile acid concentrations were determined by using the Colorimetric Total Bile Acids Assay kit (Diazyme Laboratories). Samples were analyzed in duplicate.
Culture-independent microbial analysis. Cecal content was used for analysis of the microbial community composition. Total bacterial DNA was extracted from the cecal content of 2 groups of mice that were supplemented with either DPC 6426 or the placebo control by using the QIAamp DNA stool mini kit (Qiagen), according to the manufacturerÕs instructions. The 16S ribosomal RNA (rRNA) bacterial gene amplicons flanking the V4–V5 region were generated with a view to high-throughput sequencing by using the Roche genome sequencer FLX platform at the Teagasc 454 sequencing facility. Amplicons were generated by using
Statistical analysis. Data are presented as per-group means 6 SEMs. Statistical analysis was performed by ANOVA, and data were analyzed by 1-factor ANOVA followed by TukeyÕs multiple-comparison test. Differences were considered significant at P < 0.05 unless otherwise stated. For pyrosequencing, Minitab release 15.1.1.0 was used to perform nonparametric statistical analysis (Mann-Whitney test) to compare 2 specific subject groups to determine the impact of dietary intervention with DPC 6426 on the microbiota (data not normally distributed; hence, the choice of a nonparametric test). SPSS software package version 18 was used to calculate Kruskal-Wallis and Mann-Whitney tests to find significant differences in microbial taxa and a diversity.
Results Weekly quantification of fecal bacterial numbers of individual mice confirmed gastrointestinal transit and recovery of administrated strains. There was no significant difference in fluid intake between the groups and all mice received expected doses (1 3 109 CFU/d). Recovery of GTF was 1 3 107 CFU/g feces, for PNZ was 3 3 107 CFU/g feces, and for DPC 6426 was 2 3 107 CFU/g feces. Body mass, relative organ, and tissue weights of experimental mice. There were no significant differences in food intake between groups (data not shown). As expected, body weight increased gradually over time for all groups and
TABLE 1 Serum profile of mice after administration of live polysaccharide-producing lactobacilli/placebo and a high-fat/highcholesterol diet for 12 wk1 b-Glucan, mg/L GTF PNZ DPC 6426 P
5.4 3 1024 6 1.1 3 1024 5.8 3 1024 6 1.2 3 1024 — —
sVCAM-1, mg/L 0.53b 0.53b 0.52b 1.16a
6 0.1 6 0.1 6 0.1 6 0.1
TC, mg/L 1700c 6 3100b 6 2100c 6 3900a 6
100.0 180.0 130.0 260.0
HDL cholesterol, mg/L 500.0 700.0 500.0 800.0
6 120.0 6 100.0 6 100.0 6 200.0
Ratio HDL-C/TC 0.10 0.20 0.40 0.10
6 6 6 6
0.00 0.10 0.10 0.00
TG, mg/L 2600c 6 3100b 6 3000b 6 3500a 6
90.00 890.0 160.0 140.0
1 Values are means 6 SEMs, n = 9/group. Labeled means in a column with superscripts without a common letter differ, P , 0.05. DPC 6426, Lactobacillus mucosae Dairy Product Culture Collection 6426, GTF, Lactobacillus paracasei National Food Biotechnology Center 338 expressing the glycosyltransferase (Gtf) gene; HDL-C, HDL cholesterol; P, placebo control; PNZ, isogenic control strain Lactobacillus paracasei National Food Biotechnology Center 338; sVCAM-1, soluble vascular cell adhesion molecule 1; TC, total cholesterol.
1958
London et al.
Downloaded from jn.nutrition.org by guest on December 4, 2017
RNA preparation and real-time fluorescence monitoring RT-PCR. Total RNA was extracted from liver samples by using the Qiagen RNase mini kit according to the manufacturerÕs protocol. Total RNA was reversetranscribed into cDNA by using the cDNA Synthesis kit (Bioline) according to the manufacturerÕs protocol. To quantify mRNA expression, PCR was performed by using a fluorescence temperature cycler (LightCycler System; Roche Diagnostics). The oligonucleotide primers for b-actin, cholesterol 7a-hydroxylase (Cyp7a1) and 3-hydroxy-3-methylglutaryl-coenzyme A (Hmg-coA) reductase, were designed on the basis of published nucleotide sequences for named genes in apoE-deficient mice (29). The oligonucleotide primers for FA synthase (Fas; 5# TCCACCTTTAAGTTGCCCTG 3#,5# TCTGCTCTCGTCATGTCACC 3#), sterol regulatory element-binding protein 1c (Srebp-c1; 5# CTCCAGCTCATCAACAACCAAGAC,5# AGAGGAGGCCAGAGAAGCAGAAGA 3#), and cluster of differentiation 36 (Cd36; 5# TGATACTATGCCCGCCTCTCC 3#,5# TTTCCCACACTCCTTTCTCCTCTA 3#) were newly designed on the basis of gene sequences available on the gene network service at the National Center for Biotechnology Information (http://www.ncbi.nlm.nih.gov/). Basic relative quantification of expression was determined by using the comparative 2-DDCt method.
one forward primer, F1 (5#-AYTGGGYDTAAAGNG), and a combination of 4 reverse primers, R1 (5#-TACCRGGGTHTCTAATCC), R2 (5#-TACCAGAGTATCTAATTC), R3 (5#-CTACDSRGGTMTCTAATC), and R4 (5#-TACNVGGGTATCTAATC). PCR was undertaken as previously described (30), and PCR products were cleaned by using an AMPure kit (Beckman Coulter Genomics) as per the manufacturerÕs protocol. Emulsion-based clonal amplification was completed as part of the 454-pyrosequencing process. Raw sequencing reads were quality trimmed as previously described (31). The SILVA 16S rRNA (version 106) database was used to BLAST (Basic Local Alignment Search Tool) the trimmed fasta sequence files using default parameters. BLAST output was parsed by using MEGAN (32), which assigns reads to National Center for Biotechnology Information taxonomies by using the Lowest Common Ancestor algorithm. A bit score of 86 was selected as previously used for 16S rRNA sequencing data (33). Clustering of sequence reads into operational taxonomical units at a 97% identity level was achieved by using Qiime. The ChimeraSlayer program was used to remove chimeras from aligned operational taxonomical units and the FastTreeMP tool generated a phylogenetic tree (34–36). a Diversity indices (Shannon diversity, Chao1, Simpson, Phylogenetic diversity, and observed species) and rarefaction curves were generated with the use of Qiime. b Diversities were also calculated on the sequence reads on the basis of weighted and unweighted Unifrac and bray Curtis distance matrices; subsequently, principal coordinate analysis was implemented.
FIGURE 1 Liver profile and fecal cholesterol excretion in mice after administration of live polysaccharide-producing lactobacilli/ placebo and high-fat/high-cholesterol diets for 12 wk. (A) Total liver lipids; (B) liver TGs; (C) total liver cholesterol; and (D) total fecal cholesterol excretion. Values are means 6 SEMs, n = 9/group. Labeled means without a common letter differ, P , 0.05. DPC 6426, Lactobacillus mucosae Dairy Product Culture Collection 6426; GTF, Lactobacillus paracasei National Food Biotechnology Center 338 expressing the glycosyltransferase (Gtf) gene; P, placebo control; PNZ, isogenic control strain Lactobacillus paracasei National Food Biotechnology Center 338.
Quantification of atherosclerotic lesions. All 4 experimental groups developed atherosclerotic lesions after 12 wk (data not shown). However, there were no statistical differences observed for plaque area and plaque density (data not shown). Effect of dietary administration of L. paracasei NFBC 338 on b-glucan concentration in serum. No differences in serum b-glucan concentrations were found after ingestion of GTF and PNZ (Table 1). Effect of administered lactobacilli on sVCMA-1 in serum. All 3 groups that received lactobacilli (GTF, PNZ, and DPC 6426) had significantly decreased serum concentrations of sVCAM1 compared with the placebo control group, which received no lactobacilli (Table 1). Serum lipid profiles. Total serum cholesterol concentration was decreased (by ;50%) for the GTF and DPC 6426 group compared with the placebo (P < 0.001) and PNZ (P < 0.01) control group (Table 1). In addition, total serum cholesterol was significantly lower for GTF and DPC 6426 groups (P < 0.01) than the PNZ group and, furthermore, the PNZ group was significantly lower (P < 0.01) when compared with the placebo group. There were no significant differences in serum HDLcholesterol concentrations between the GTF and DPC 6426 groups after 12 wk of dietary intervention (Table 1). Furthermore, the ratio of HDL cholesterol to TC was not significantly altered in any of the groups (Table 1). The serum TG concentration was significantly lower in the GTF (P < 0.001) and DPC 6426 (P < 0.05) groups than the placebo group (P < 0.05; Table 1). However, there was no significant difference between the GTF group and the PNZ group (Table 1). Liver lipids. Liver lipid concentrations were significantly decreased for the GTF (P < 0.01) and DPC 6426 (P < 0.01) groups compared with PNZ and placebo groups (Figure 1A), which relates to the findings of fatty livers (data not shown). TG accumulation in the liver was similar for the PNZ group and the placebo group and was significantly decreased (P < 0.05) for both GTF and DPC 6426 groups (Figure 1B). Hepatic total
cholesterol concentrations were similar between GTF, DPC 6426, and placebo groups but were found to be increased for the PNZ group relative to the GTF and DPC 6426 groups (P < 0.05; Figure 1C). Total bile acid and cholesterol in feces. Dietary intervention with GTF led to increased fecal cholesterol excretion compared with all other groups (P < 0.05; Figure 1D). Furthermore, there was no significant difference between fecal bile acid excretions among the groups (data not shown). Expression of hepatic key enzymes involved in fat and cholesterol metabolism. The mRNA expression of Cyp7a1 was significantly increased in the GTF group compared with all other groups (P < 0.05; Figure 2). In addition, the mRNA expression of Cd36 significantly decreased (P < 0.05) in the GTF and DPC 6426 groups compared with the PNZ and placebo control groups (Figure 2). There were no significant differences in mRNA expression of either Hmg-coA, Srebp-1c, or Fas (Figure 3). Effects on gut microbiota composition after dietary intervention with DPC 6426. Bioinformatic analysis of 16S rRNA sequence data, generated for DPC 6426 and placebo
FIGURE 2 Relative gene expression of hepatic mRNA in mice after administration of live polysaccharide-producing lactobacilli/placebo and high-fat/high-cholesterol diets for 12 wk. Values are means 6 SEMs, n = 9/group. Labeled means for each enzyme/protein without a common letter differ, P , 0.05. Cd36, cluster of differentiation 36; Cyp7a1, cholesterol 7a-hydroxylase; DPC 6426, Lactobacillus mucosae Dairy Product Culture Collection 6426; Fas, FA synthase; GTF, Lactobacillus paracasei National Food Biotechnology Center 338 expressing the glycosyltransferase (Gtf) gene; Hmg-coA, 3-hydroxy3-methylglutaryl coenzyme A; P, placebo control; PNZ, isogenic control strain Lactobacillus paracasei National Food Biotechnology Center 338; Srebp-1c, sterol regulatory element-binding protein 1. Bacterial polysaccharides reduce serum cholesterol
1959
Downloaded from jn.nutrition.org by guest on December 4, 2017
there were no significant differences in final change in body mass between groups. There were no significant differences in the masses of various adipose tissues (subcutaneous adipose tissue, epididymal adipose tissue, and mesenteric adipose tissue), liver, and cecum (data not shown).
FIGURE 3 Phylum-level distributions of the microbial communities in the cecal contents, expressed as a percentage of total population of assignable tags, in mice after administration of DPC 6426 (n = 9) or a placebo control (n = 9) and a high-fat/high-cholesterol diet for 12 wk. Labeled means for each phylum without a common letter differ, P , 0.05. DPC 6426, Lactobacillus mucosae Dairy Product Culture Collection 6426; P, placebo control.
FIGURE 4 Family-level taxonomic distribution of the microbial communities in cecal contents, expressed as a percentage of total tags assignable at the family level, in mice after administration of DPC 6426 (n = 9) or a placebo control (n = 9) and a high-fat/high-cholesterol diet for 12 wk. *Significant difference, P , 0.05. DPC 6426, Lactobacillus mucosae Dairy Product Culture Collection 6426; P, placebo control. 1960
London et al.
Discussion The present study shows that lipid metabolism in high-fat/highcholesterol–fed apoE-deficient mice is positively influenced by oral administration of GTF and DPC 6426 compared with PNZ and placebo controls, respectively. We showed that dietary intervention with GTF and DPC 6426 are both associated with significantly decreased serum cholesterol (by ;57% and ;47%, respectively), compared with the placebo control. Hong et al. (16) reported similar results for dietary bacterial b-glucan, which decreased serum total cholesterol in Sprague-Dawley rats. In our study, administration of PNZ was associated with significantly decreased TC concentration in serum (by ;32%) compared with the placebo control. These data suggest that the hypocholesterolemic effect seen for both exopolysaccharides-producing bacteria is not solely attributable to the b-glucan and exopolysaccharides but to the lactobacilli used in this study in general. This is consistent with evidence for the cholesterol-reducing abilities of other lactobacilli previously reported (38). In addition, both GTF and DPC 6426 are further associated with prevention of hypercholesterolemia by a decreased hepatic accumulation of cholesterol. However, only GTF led to the induction of hepatic Cyp7a1 mRNA levels, indicating that cholesterol homeostasis is regulated differently for GTF and DPC 6426. In some species, the excess consumption of cholesterol results in a homeostatic response that includes the induction of Cyp7a1 mRNA levels (39). This would suggest that DPC 6426 might prevent intestinal absorption of dietary cholesterol. On the contrary, we did not find significantly increased fecal cholesterol for DPC 6426 but we did for GTF. This finding is important because it indicates that the hypocholesterolemic effect of DPC 6426 and GTF might not be solely caused by a decrease in cholesterol absorption in the host, but by modifying cholesterol homeostasis through synthesis and conjugation of cholesterol to bile acids. Our study also shows that hepatic lipid contents are decreased by dietary intervention with GTF and DPC 6426 in the apoE mouse model but not by PNZ and the suppression of hepatic Cd36 mRNA levels in mice that received either GTF or DPC 6426, indicating that the uptake of FAs from circulation and intracellular transport of long-chain FAs was also altered. Furthermore, dietary intervention with GTF and DPC 6426 resulted
Downloaded from jn.nutrition.org by guest on December 4, 2017
groups, revealed a total of 306,282 reads, averaging at 18,017 reads per mouse. Rarefaction curves for each data set were parallel or approaching parallel with the x axis, demonstrating that the depth of sequencing was sufficient (data not shown). At the 97% similarity level, the Shannon Index indicated good overall biodiversity within the samples with a mean value of 4.3 (data not shown). Chao1 (an indicator of species richness) also indicated a sufficient level of overall diversity with a mean value of 703 (data not shown). Taxonomy-based analysis at the phylum level showed that the mouse gut microbiota was dominated by Firmicutes and Bacteroidetes (Figure 3), which is in agreement with previous findings (37). The most abundant families were Peptostreptococcaceae, Ruminococcaceae, and Lachnospiraceae (Figure 4); and the most abundant genera were Ruminococcaceae Incertae Sedis, Peptostreptococcaceae Incertae Sedis, Lachnospiraceae Incertae Sedis, Clostridium, Blautia, Anaerotruncus, and Allobaculum (Figure 5). The composition of the microbiota of cecal content of mice supplemented with DPC 6426 was significantly altered at family and genus levels but remained similar at the phylum level compared with placebo controls. Dietary intervention with DPC 6426 resulted in significantly less Clostridiaceae
(P < 0.05), Peptococcaceae (P < 0.001), and Staphylococcaceae (P < 0.01) in cecum compared with placebo controls (Figure 4), and was associated with the presence of Porphyromonadaceae and Prevotellaceae, whereas these families were not found in the placebo group (Figure 4). At the genus level, there were a few populations present only in the cecal microbiota of the placebo group (Figure 5); i.e., Coprococcus (P < 0.001), Eubacterium (P < 0.001), and Erysipelotrichales Incertae Sedis (P < 0.001). Furthermore, dietary administration of DPC 6426 was associated with the presence of Alkaliphilus (P < 0.05), Peptococcus (P < 0.01), Paludibacter (P < 0.05), Porphyromonas (P < 0.001), and Prevotella (P < 0.001) in cecal contents (Figure 5). Other observed differences at the genus level after dietary administration of DPC 6426 included a significant increase in cecal Allobaculum (P < 0.05), Clostridium (P < 0.01), Lachnospiraceae (P < 0.001), and Peptostreptococcaceae (P < 0.05), whereas Oscillibacter (P < 0.001) were significantly decreased (Figure 5). Principal coordinate analysis generated by using Bray Curtis, unweighted Unifrac, and weighted Unifrac distances showed that mice did not cluster into distinct groups based on whether they received DPC 6426 or not (data not shown).
FIGURE 5 Genus-level taxonomic distributions of the microbial communities present in cecal contents, expressed as a percentage of total tags assignable at the genus level, in mice after administration of DPC 6426 (n = 9) or a placebo control (n = 9) and a high-fat/highcholesterol diet for 12 wk. *Significant difference, P , 0.05. B-B-C, Brenneria-Buttiauxella-Cedecea-Citrobacter-Cronobacter-Dickeya-Edwardsiella-Enterobacter-Erwinia-Escherichia-Ewingella-Hafnia-Klebsiella-KluyveraLeclercia-Levinea-Pantoea-PectobacteriumRahnella-Raoultella-Serratia-Yersinia; DPC 6426, Lactobacillus mucosae Dairy Product Culture Collection 6426; P, placebo control.
view of the effects of dietary intervention with DPC 6426 on the entire microbial population. In conclusion, we show that dietary intervention with a b-glucan–producing probiotic strain results in modulation of lipid metabolism in a mouse model of atherosclerosis. We also show that b-glucan production by a probiotic strain conferred an advantage in terms of reduced serum cholesterol and TG concentrations and reduced hepatic lipid concentrations compared with the use of a non–b-glucan-producing isogenic control strain. Furthermore, dietary intervention with exopolysaccharides-producing DPC 6426 resulted in modulation of lipid metabolism in the mouse model of atherosclerosis, including significantly lower serum cholesterol and TG concentrations, and was associated with changes in the gut microbial community. These results confirm that dietary intervention with exopolysaccharides-producing (probiotic) strains is a feasible strategy for managing lipid metabolism. Acknowledgments GFF, RPR, NMC, and CS designed the research; LEEL, AHSK, RW, and PGC conducted the research; LEEL, AHSK, PDC, and OO analyzed the data; FS and CH helped with the manuscript and provided significant advice; LEEL and CS wrote the manuscript; and NMC and CS had primary responsibility for the final content. All authors read and approved the final manuscript.
References 1.
2. 3.
4.
FAO/WHO. Interim summary of conclusions and dietary recommendations on total fat & fatty acids. From the Joint FAO/WHO Expert Consultation on Fats and Fatty Acids in Human Nutrition, 10–14 November 2008, WHO, Geneva. Available from: http://www.who.int/ nutrition/topics/FFA_summary_rec_conclusion.pdf?ua=1. Ley RE, Turnbaugh PJ, Klein S, Gordon JI. Microbial ecology—human gut microbes associated with obesity. Nature 2006;444:1022–3. Dumas ME, Barton RH, Toye A, Cloarec O, Blancher C, Rothwell A, Fearnside J, Tatoud R, Blanc V, Lindon JC, et al. Metabolic profiling reveals a contribution of gut microbiota to fatty liver phenotype in insulin-resistant mice. Proc Natl Acad Sci USA 2006;103:12511–6. Wang Z, Klipfell E, Bennett BJ, Koeth R, Levison BS, Dugar B, Feldstein AE, Britt EB, Fu XM, Chung YM, et al. Gut flora metabolism of phosphatidylcholine promotes cardiovascular disease. Nature 2011; 472:57–63.
Bacterial polysaccharides reduce serum cholesterol
1961
Downloaded from jn.nutrition.org by guest on December 4, 2017
in significantly decreased serum TG concentrations compared with PNZ and placebo controls. This apparent discrepancy might provide a novel mechanism for how exopolysaccharidesproducing lactobacilli could affect lipid metabolism. We posit that the ingestion of GTF and DPC 6426 alters cholesterol homeostasis/lipid metabolism because of their production of b-glucan/exopolysaccharides (21), although the mechanism need to be further elucidated. In this study, all mouse groups exhibited atherosclerosis to a certain degree, a finding that is in agreement with the animal model (40). However, studies addressing the effect of lactobacilli on atherosclerosis are limited. Ingested Lactobacillus acidophilus American Type Culture Collection 4356 was recently associated with a decrease in atherosclerotic lesion size by reducing vascular oxidative stress and mitigating the inflammatory response rather than by regulating cholesterol homeostasis (41). Although we did not observe any significant atheroprotective effect after consumption of lactobacilli nor associated changes in the HDL cholesterol concentration or HDL cholesterol:TC ratio, we found that serum sVCAM-1 significantly decreased in all Lactobacillus-supplemented groups, indicating that dietary intervention with lactobacilli may positively influence the inflammation process of atherosclerosis (42, 43). Our results indicate that dietary intervention with DPC 6426 has an effect on the composition of the murine gut microbiota and that administration of this single strain can play a role in the modulation of the gut microbiota composition in vivo. Indeed, the microbiota was linked to complex disease phenotypes such as obesity and insulin resistance (44, 45). Moreover, after chronic L-carnitine supplementation, an increase in Prevotella was recently linked to a microbiota-dependent trimethylamine production and subsequent hepatic TMAO production in the host, leading to proatherosclerosis phenotypes by impairing reverse cholesterol transport, whereas an increase in Bacteroidetes led to less trimethylamine and TMAO production (7). Surprisingly, we found that dietary intervention with DPC 6426 led to an apparent enrichment in Prevotellaceae and Prevotella at taxa and genus levels, which were both absent in the cecum of placebo mice. However, we did not see any reverse effect on cholesterol homeostasis and atherosclerosis as seen before. We accept that pyrosequencing of the 16S rRNA genes does not provide quantitative population data, but it provides an over-
5.
6.
7.
8.
9. 10.
11.
12.
14.
15.
16.
17. 18.
19.
20.
21.
22.
23.
24.
25.
1962
London et al.
26. Koga M, Kai H, Yasukawa H, Kato S, Yamamoto T, Kawai Y, Kusaba K, Seki Y, Kai M, Egashira K, et al. Postnatal blocking of interferon-gamma function prevented atherosclerotic plaque formation in apolipoprotein E-knockout mice. Hypertens Res 2007;30:259–67. 27. Folch J, Lees M, Sloane Stanley GH. A simple method for the isolation and purification of total lipides from animal tissues. J Biol Chem 1957; 226:497–509. 28. Van der Meer R, De Vries HT, Glatz JFC. t-Butanol extraction of feces: a rapid procedure for enzymatic determination of fecal bile acids. Beynen AC, Geelen MJH, Katan MB, Schouten JA, eds. Cholesterol metabolism in health and disease. Wageningen: Ponsen and Looyen; 1985:113–9. 29. Han SH, Chung MJ, Lee SJ, Rhee C. Digestion-resistant fraction from soybean [Glycine max (L.) Merrill] induces hepatic LDL receptor and CYP7A1 expression in apolipoprotein E-deficient mice. J Nutr Biochem 2006;17:682–8. 30. Fouhy F, Guinane CM, Hussey S, Wall R, Ryan CA, Dempsey EM, Murphy B, Ross RP, Fitzgerald GF, Stanton C, et al. High-throughput sequencing reveals the incomplete, short-term recovery of infant gut microbiota following parenteral antibiotic treatment with ampicillin and gentamicin. Antimicrob Agents Chemother 2012;56:5811–20. 31. Claesson MJ, O’Sullivan O, Wang Q, Nikkila J, Marchesi JR, Smidt H, de Vos WM, Ross RP, O’Toole PW. Comparative analysis of pyrosequencing and a phylogenetic microarray for exploring microbial community structures in the human distal intestine. PLoS ONE 2009;4:e6669. 32. Huson DH, Auch AF, Qi J, Schuster SC. MEGAN analysis of metagenomic data. Genome Res 2007;17:377–86. 33. Urich T, Lanzen A, Qi J, Huson DH, Schleper C, Schuster SC. Simultaneous assessment of soil microbial community structure and function through analysis of the meta-transcriptome. PLoS ONE 2008;3:e2527. 34. Caporaso JG, Kuczynski J, Stombaugh J, Bittinger K, Bushman FD, Costello EK, Fierer N, Pena AG, Goodrich JK, Gordon JI, et al. QIIME allows analysis of high-throughput community sequencing data. Nat Methods 2010;7:335–6. 35. Haas BJ, Gevers D, Earl AM, Feldgarden M, Ward DV, Giannoukos G, Ciulla D, Tabbaa D, Highlander SK, Sodergren E, et al. Chimeric 16S rRNA sequence formation and detection in Sanger and 454-pyrosequenced PCR amplicons. Genome Res 2011;21:494–504. 36. Price MN, Dehal PS, Arkin AP. FastTree 2—approximately maximumlikelihood trees for large alignments. PLoS ONE 2010;5:e9490. 37. Wall R, Marques TM, O’Sullivan O, Ross RP, Shanahan F, Quigley E, Dinan TG, Kiely B, Fitzgerald GF, Cotter PD, et al. Contrasting effects of Bifidobacterium breve NCIMB 702258 and Bifidobacterium breve DPC 6330 on the composition of murine brain fatty acids and gut microbiota. Am J Clin Nutr 2012;95:1278–87. 38. Wang J, Zhang H, Chen X, Chen Y, Menghebilige, Bao Q. Selection of potential probiotic lactobacilli for cholesterol-lowering properties and their effect on cholesterol metabolism in rats fed a high-lipid diet. J Dairy Sci 2012;95:1645–54. 39. Dueland S, Drisko J, Graf L, Machleder D, Lusis AJ, Davis RA. effect of dietary-cholesterol and taurocholate on cholesterol 7-alphahydroxylase and hepatic Ldl receptors in inbred mice. J Lipid Res 1993;34:923–31. 40. Meir KS, Leitersdorf E. Atherosclerosis in the apolipoprotein E-deficient mouse—a decade of progress. Arterioscler Thromb Vasc Biol 2004;24: 1006–14. 41. Chen L, Liu WE, Li YM, Luo S, Liu QX, Zhong YM, Jian ZJ, Bao MH. Lactobacillus acidophilus ATCC 4356 attenuates the atherosclerotic progression through modulation of oxidative stress and inflammatory process. Int Immunopharmacol 2013;17:108–15. 42. Cockerill GW, Rye KA, Gamble JR, Vadas MA, Barter PJ. High-density lipoproteins inhibit cytokine-induced expression of endothelial cell adhesion molecules. Arterioscler Thromb Vasc Biol 1995;15:1987–94. 43. De Backer J, Mak R, De Bacquer D, Van Renterghem L, Verbraekel E, Kornitzer M, De Backer G. Parameters of inflammation and infection in a community based case-control study of coronary heart disease. Atherosclerosis 2002;160:457–63. 44. Turnbaugh PJ, Hamady M, Yatsunenko T, Cantarel BL, Duncan A, Ley RE, Sogin ML, Jones WJ, Roe BA, Affourtit JP, et al. A core gut microbiome in obese and lean twins. Nature 2009;457:480–4. 45. Turnbaugh PJ, Ley RE, Mahowald MA, Magrini V, Mardis ER, Gordon JI. An obesity-associated gut microbiome with increased capacity for energy harvest. Nature 2006;444:1027–31.
Downloaded from jn.nutrition.org by guest on December 4, 2017
13.
Lam V, Su JD, Koprowski S, Hsu A, Tweddell JS, Rafiee P, Gross GJ, Salzman NH, Baker JE. Intestinal microbiota determine severity of myocardial infarction in rats response. FASEB J 2012;26:4388–9. Karlsson FH, Fak F, Nookaew I, Tremaroli V, Fagerberg B, Petranovic D, Backhed F, Nielsen J. Symptomatic atherosclerosis is associated with an altered gut metagenome. Nat Commun 2012;3:1245. Koeth RA, Wang Z, Levison BS, Buffa JA, Org E, Sheehy BT, Britt EB, Fu X, Wu Y, Li L, et al. Intestinal microbiota metabolism of L-carnitine, a nutrient in red meat, promotes atherosclerosis. Nat Med 2013;19:576–85. Marlett JA, McBurney MI, Slavin JL. Position of the American Dietetic Association: health implications of dietary fiber. J Am Diet Assoc 2002; 102:993–1000. FDA. A rule by the Food and Drug Administration. Food labeling: health claims: oat and coronary heart disease. Federal Register. 2010;62:15. Wilson TA, Nicolosi RJ, Delaney B, Chadwell K, Moolchandani V, Kotyla T, Ponduru S, Zheng GH, Hess R, Knutson N, et al. Reduced and high molecular weight barley beta-glucans decrease plasma total and non-HDL-cholesterol in hypercholesterolemic Syrian golden hamsters. J Nutr 2004;134:2617–22. Nicolosi R, Bell SJ, Bistrian BR, Greenberg I, Forse RA, Blackburn GL. Plasma lipid changes after supplementation with beta-glucan fiber from yeast. Am J Clin Nutr 1999;70:208–12. Ganzle M, Michael G, Schwab C. Exopolysaccharide production by intestinal lactobacilli. In: Tannock GW, editor. Probiotics & Prebiotics: Scientific Aspects. Norfolk: Caister Academic Press. 2005. Garai-Ibabe G, Duen˜as MT, Irastorza A, Sierra-Filardi E, Werning ML, Lopez P, Corb´ı AL, Fernandez de Palencia P. Naturally occurring 2´ ´ substituted (1,3)-beta-D-glucan producing Lactobacillus suebicus and Pediococcus parvulus strains with potential utility in the production of functional foods. Bioresour Technol 2010;101:9254–63. Kim Y, Oh S, Kim SH. Released exopolysaccharide (r-EPS) produced from probiotic bacteria reduce biofilm formation of enterohemorrhagic Escherichia coli O157:H7. Biochem Biophys Res Commun 2009; 379:324–9. Rodr´ıguez C, Medici M, Rodriguez AV, Mozzi F, de Valdez GF. Prevention of chronic gastritis by fermented milks made with exopolysaccharideproducing Streptococcus thermophilus strains. J Dairy Sci 2009;92: 2423–34. Hong K, Jang KH, Lee JC, Kim S, Kim MK, Lee IY, Kim SM, Lim YH, Kang SA. Bacterial beta-glucan exhibits potent hypoglycemic activity via decrease of serum lipids and adiposity, and increase of UCP mRNA expression. J Microbiol Biotechnol 2005;15:823–30. Nakajima H, Suzuki Y, Kaizu H, Hirota T. Cholesterol lowering activity of ropy fermented milk. J Food Sci 1992;57:1327–9. Maeda H, Zhu X, Suzuki S, Suzuki K, Kitamura S. Structural characterization and biological activities of an exopolysaccharide kefiran produced by Lactobacillus kefiranofaciens WT-2B(T). J Agric Food Chem 2004;52:5533–8. Uchida M, Ishii I, Inoue C, Akisato Y, Watanabe K, Hosoyama S, Toida T, Ariyoshi N, Kitada M. Kefiran reduces atherosclerosis in rabbits fed a high cholesterol diet. J Atheroscler Thromb 2010;17:980–8. Martensson O, Biorklund M, Lambo AM, Duenas-Chasco M, Irastorza A, Holst O, Norin E, Welling G, Oste R, Onning G. Fermented, ropy, oat-based products reduce cholesterol levels and stimulate the bifidobacteria flora in humans. Nutr Res 2005;25:429–42. Stack HM, Kearney N, Stanton C, Fitzgerald GF, Ross RP. Association of beta-glucan endogenous production with increased stress tolerance of intestinal lactobacilli. Appl Environ Microbiol 2010;76:500–7. Nakashima Y, Plump AS, Raines EW, Breslow JL, Ross R. Apoedeficient mice develop lesions of all phases of atherosclerosis throughout the arterial tree. Arterioscler Thromb 1994;14:133–40. Dhar MS, Sommardahl CS, Kirkland T, Nelson S, Donnell R, Johnson DK, Castellani LW. Mice heterozygous for Atp10c, a putative amphipath, represent a novel model of obesity and type 2 diabetes. J Nutr 2004; 134:799–805. Wall R, Ross RP, Shanahan F, O’Mahony L, O’Mahony C, Coakley M, Hart O, Lawlor P, Quigley EM, Kiely B, et al. Metabolic activity of the enteric microbiota influences the fatty acid composition of murine and porcine liver and adipose tissues. Am J Clin Nutr 2009;89:1393–401. Koga M, Kai H, Yasukawa H, Yamamoto T, Kawai Y, Kato S, Kusaba K, Kai M, Egashira K, Kataoka Y, et al. Inhibition of progression and stabilization of plaques by postnatal interferon-gamma function blocking in apoE-knockout mice. Circ Res 2007;101:348–56.
Exopolysaccharide-Producing Probiotic Lactobacilli Reduce Serum Cholesterol and Modify Enteric Microbiota in ApoE-Deficient Mice1,2 Lis EE London,3,6 Arun HS Kumar,4 Rebecca Wall,3 Pat G Casey,3,5 Orla OÕSullivan,6 Fergus Shanahan,3 Colin Hill,3,5 Paul D Cotter,3,6 Gerald F Fitzgerald,3,5 R Paul Ross,3,5 Noel M Caplice,4 and Catherine Stanton3,6* 3 Alimentary Pharmabiotic Centre, 4Centre for Research in Vascular Biology, and 5Department of Microbiology, University College Cork, Cork, Ireland; and 6Teagasc, Food Research Centre Moorepark, Fermoy, Cork, Ireland
Abstract death and disability. Objectives: The aim of this study was to assess the impact of dietary administration of exopolysaccharide-producing probiotic Lactobacillus cultures on lipid metabolism and gut microbiota in apolipoprotein E (apoE)–deficient mice. Methods: First, we examined lipid metabolism in response to dietary supplementation with recombinant b-glucan–producing Lactobacillus paracasei National Food Biotechnology Centre (NFBC) 338 expressing the glycosyltransferase (Gtf) gene from Pediococcus parvulus 2.6 (GTF), and naturally exopolysaccharide-producing Lactobacillus mucosae Dairy Product Culture Collection (DPC) 6426 (DPC 6426) compared with the non–b-glucan-producing isogenic control strain Lactobacillus paracasei NFBC 338 (PNZ) and placebo (15% wt: vol trehalose). Second, we examined the effects on the gut microbiota of dietary administration of DPC 6426 compared with placebo. Probiotic Lactobacillus strains at 1 3 109 colony-forming units/d per animal were administered to apoE2/2 mice fed a high-fat (60% fat)/ high-cholesterol (2% wt:wt) diet for 12 wk. At the end of the study, aortic plaque development and serum, liver, and fecal variables involved in lipid metabolism were analyzed, and culture-independent microbial analyses of cecal content were performed. Results: Total cholesterol was reduced in serum (P < 0.001; ;33–50%) and liver (P < 0.05; ;30%) and serum triglyceride concentrations were reduced (P < 0.05; ;15–25%) in mice supplemented with GTF or DPC 6426 compared with the PNZ or placebo group, respectively. In addition, dietary intervention with GTF led to increased amounts of fecal cholesterol excretion (P < 0.05) compared with all other groups. Compositional sequencing of the gut microbiota revealed a greater prevalence of Porphyromonadaceae (P = 0.001) and Prevotellaceae (P = 0.001) in the DPC 6426 group and lower proportions of Clostridiaceae (P < 0.05), Peptococcaceae (P < 0.001), and Staphylococcaceae (P < 0.01) compared with the placebo group. Conclusion: Ingestion of exopolysaccharide-producing lactobacilli resulted in seemingly favorable improvements in lipid metabolism, which were associated with changes in the gut microbiota of mice. J Nutr 2014;144:1956–62. Keywords: probiotics, lactobacilli, exopolysaccharide, cholesterol, gut microbiota, lipid metabolism
Introduction Cardiovascular disease (CVD)7 (1) is a leading cause of death and disability globally (1). The human intestinal microbiota has been implicated in the development of obesity (2), insulin resistance (3), and more recently, CVD risk (4, 5). Patients with 1
Supported by the Irish Government, Department of Agriculture Food and Marine (DAFM); the Science Foundation of Ireland–funded Centre for Science, Engineering, and Technology; the Alimentary Pharmabiotic Centre and Enterprise Ireland Commercialisation Fund Contract (reference: CF/2013/3030A/B); the European Union Seventh Framework Programme–funded FIBEBIOTICS: Dietary Fibers Supporting Gut and Immune Function (EC proposal 289517). LEE London received a Teagasc Walsh Fellowship (2007–2010). * To whom correspondence should be addressed. E-mail: catherine.stanton@ teagasc.ie.
1956
symptomatic atherosclerosis were shown to harbor an altered gut metagenome compared with healthy controls (6), whereas a 2
Author disclosures: LEE London, AHS Kumar, R Wall, PG Casey, O OÕSullivan, F Shanahan, C Hill, PD Cotter, GF Fitzgerald, RP Ross, NM Caplice, and C Stanton, no conflicts of interest. 7 Abbreviations used: Cd36, cluster of differentiation 36; CVD, cardiovascular disease; Cyp7a1, cholesterol 7a-hydroxylase; DPC, Dairy Product Culture Collection; DPC 6426, Lactobacillus mucosae Dairy Product Culture Collection 6426; Fas, FA synthase; GTF, Lactobacillus paracasei National Food Biotechnology Center 338 expressing the glycosyltransferase gene from Pediococcus parvulus 2.6; Gtf, glycosyltransferase; Hmg-coA, 3-hydroxy-3-methylglutaryl-coenzyme A; MRS, de Man, Rogasa, and Sharpe; NFBC, National Food Biotechnology Center; PNZ, Lactobacillus paracasei National Food Biotechnology Center 338; rRNA, ribosomal RNA; Srebp-1c, sterol regulatory element-binding protein 1; sVCAM-1, soluble vascular cell adhesion molecule 1; TC, total cholesterol; TMAO, trimethylamine-N-oxide.
ã 2014 American Society for Nutrition. Manuscript received February 11, 2014. Initial review completed March 26, 2014. Revision accepted September 15, 2014. First published online October 15, 2014; doi:10.3945/jn.114.191627.
Downloaded from jn.nutrition.org by guest on December 4, 2017
Background: Probiotic bacteria have been associated with a reduction in cardiovascular disease risk, a leading cause of
Methods Experimental animals and diet. All mouse studies were undertaken in accordance with the Department of Health and Children of the Irish Government. A license and permission for the study were obtained from the Department of Health, Ireland. Male Apoetm1Unc/J mice, which
develop atherosclerostic lesions similar to the process of atherogenesis in humans (22), that were between 10 and 12 wk old at the beginning of the experiment were obtained from JAX (The Jackson Laboratory through Charles River Laboratories International). The mice were caged either individually, in pairs, or in groups of 4–6 per cage. Group housing differences were distributed into the different feeding groups. All mice consumed a basal diet (23) ad libitum for 2 wk to stabilize all metabolic conditions, with free access to water at all times. After the acclimatization period, mice were divided into 4 groups (n = 9) and fed a high-fat (60% kcal from fat), high-cholesterol (2% wt:wt) diet (23). Mice were subjected to the following daily dietary interventions for 12 wk: the GTF group were administered ;109 CFU/(mouse d) of b-glucan–producing L. paracasei NFBC 338 expressing the Gtf gene from P. parvulus 2.6 in drinking water, the PNZ group were administered ;109 CFU/(mouse d) of L. paracasei NFBC 338 in drinking water, the DPC 6426 group were administered ;109 CFU/(mouse d) of DPC 6426 in drinking water, and the placebo group were administered resuspended freeze-dried powder containing trehalose (15% wt:vol) as a placebo control in drinking water. Mice were housed in an isolator station, exposed to a 12-h light/ dark cycle, and maintained at a constant temperature of 25°C. Body weight and food intake were monitored weekly. After 12 wk of being fed experimental diets, serum was collected by retro-orbital blood sampling from mice not deprived of food and stored at 4°C for 30 min, followed by centrifugation at 10,000 3 g for 5 min, and divided into aliquots and stored at 220°C until processing and mice were killed by cervical dislocation. Tissues were removed, blotted dry, weighed, and frozen in liquid nitrogen and stored at 280°C until use. Preparation and administration of lactobacilli. Chloramphenicolresistant and recombinant cultures of L. paracasei NFBC 338 were activated and propagated 3 times in de Man, Rogasa, and Sharpe (MRS) medium (Difco Laboratories) containing 5 mg/L chloramphenicol (Sigma-Aldrich) at 37°C anaerobically (21). Rifampicin-resistant variants of L. mucosae DPC 6426 (originally isolated from bovine feces at Teagasc, Moorepark Food Research Centre) were activated and propagated 3 times in MRS containing 5 mg/L rifampicin (Sigma-Aldrich) at 37°C anaerobically. Fermentation of L. mucosae DPC 6426 was undertaken in MRS broth containing 5% (wt:vol) sucrose (SigmaAldrich) at 37°C anaerobically for 20 h and 18 h for both L. paracasei NFBC 338 (GTF and PNZ) cultures. After fermentation, the cultures were washed twice in PBS (Sigma-Aldrich) and resuspended at a concentration of ;2 3 1010 cells/mL in 15% (wt:vol) trehalose (SigmaAldrich). Aliquots were freeze-dried (VirTis AdVantage Freeze Dryer; SP Industries) as outlined by Wall et al. (24). Viable cells in freezedried powders (produced weekly) containing L. paracasei NFBC 338 and L. mucosae DPC 6426 were confirmed weekly for every batch of powder produced and involved enumeration of the strains on MRS agar supplemented with appropriate antibiotic after anaerobic incubation for 72 h at 37°C. Individual mice consumed ;1 3 109 live microorganisms/d. This was achieved by resuspending appropriate quantities of freeze-dried powder in the water that mice consumed ad libitum. Microbial analysis. Fresh fecal samples were taken from experimental mice weekly and analyzed for the presence of b-glucan–producing and non–b-glucan-producing L. paracasei NFBC 338 and exopolysaccharidesproducing L. mucosae DPC 6426. Microbial analysis of the fecal samples involved enumeration of the strains on MRS agar supplemented with appropriate antibiotic after anaerobic incubation for 72 h at 37°C. Preparation and analysis of aortas. The whole length of the aorta was removed and fixed in a 4% (wt:vol) paraformaldehyde (Sigma-Aldrich) solution in PBS. Quantification of the ‘‘en face’’ plaque area was performed as previously described (25, 26) and inspected under the microscope for quantification of atherosclerotic lesions. Lipid-containing plaque area was determined as a percentage of Oil Red O-stained area from the total aortic surface area by using AxioVision 4.6 (Carl Zeiss Vision) or NIH Image J software. Measurement of serum variables. A commercially available murine ELISA kit (Quantikine; R&D Systems) was used to quantify serum Bacterial polysaccharides reduce serum cholesterol
1957
Downloaded from jn.nutrition.org by guest on December 4, 2017
link between intestinal microbial production of proatherosclerotic metabolites from dietary phospholipids and atherosclerosis risk was also recently reported (4). Moreover, a mechanistic link has been established between changes in intestinal microbiota (via oral treatment with vancomycin in rats) and myocardial infarction. Intestinal microbial metabolism of L-carnitine, abundant in red meat, to proartherogenic trimethylamine-N-oxide (TMAO) has recently been postulated as a mechanism by which red meat ingestion may be linked to CVD risk (7). In a study by Wang et al. (4), it was also reported that dietary administration of a probiotic mixture comprising Lactobacillus plantarum (Lp299v) and Bifidobacterium lactis (Bi-07) was cardioprotective. Several dietary fibers were shown to significantly decrease serum cholesterol concentrations and thereby reduce risk of CVD (8). b-Glucan is a polysaccharide found in yeast, mushrooms, oats, barley, and bacteria; and b-glucan from whole oats has been recommended by the U.S. FDA (9) as a safe and practical dietary approach for blood cholesterol reduction. Cholesterol-lowering effects of oats, barley, and yeast are well established (10); and b-glucan from oats and yeast has been shown to improve lipid profiles (11). Some lactic acid bacteria produce exopolysaccharides, a wide diversity of carbohydrates varying in structure and monosaccharide composition, which, when ingested, may confer health benefits, including cardioprotective properties such as hypocholesterolemic, antioxidant, and immunomodulatory activities, in addition to prebiotic, antitumor effects, and affect gastrointestinal health (12–15). However, there are little data available from in vivo studies in humans and animal models and in vitro studies that demonstrate beneficial effects of bacterial b-glucan, and exopolysaccharides. In an animal model, the ingestion of bacterial b-glucan led to increased concentrations of serum HDL cholesterol and decreased TG and total cholesterol (TC) concentrations (16). Interestingly, the ingestion of isolated exopolysaccharides enhanced serum cholesterol metabolism (17), reduced TC in serum, LDLcholesterol concentrations in the liver (18), and size of atherosclerotic lesions in rabbits (19). In a human study, the ingestion of nondairy, oat-based foods containing the b-glucan–producing strain Pediococcus parvulus 2.6 resulted in decreased serum cholesterol concentrations (20). We previously reported the functional expression of the pediococcal glycosyltransferase gene (Gtf) in Lactobacillus paracasei National Food Biotechnology Center (NFBC) 338 (21) resulted in b-glucan production. The objectives of the present study were as follows: 1) to compare the impact of dietary supplementation of the b-glucan–producing recombinant L. paracasei NFBC 338 (GTF) with its non–b-glucan-producing isogenic control (PNZ) and a placebo control on lipid metabolism in apolipoprotein E (apoE)–deficient mice fed a high-fat/high-cholesterol diet for 12 wk, 2) to compare the impact of dietary supplementation of the exopolysaccharidesproducing Lactobacillus mucosae Dairy Product Culture Collection (DPC) 6426 (DPC 6426) with a placebo control on lipid metabolism in apoE-deficient mice fed a high-fat/highcholesterol diet for 12 wk, and 3) to investigate the impact on the gut microbial community of the host after ingestion of DPC 6426.
soluble vascular cell adhesion molecule 1 (sVCAM-1) following the manufacturerÕs protocol. Serum b-glucan concentrations were determined with Glucatell, a 1,3-b-D-glucan detection kit (Associates of Cape Cod Inc.) for mice that received b-glucan–producing and non–b-glucanproducing L. paracasei NFBC 338. Serum TC was determined by using the EnzyChrom Cholesterol Assay kit (ECCH-100; BioAssay Systems). Determination of HDL cholesterol in serum samples was undertaken by using the EnzyChrom HDL and LDL/VLDL Assay kit (EHDL-100; BioAssay Systems). TGs in serum were determined by using the LabAssay Triglyceride kit (Wako Diagnostics). All samples and standards were analyzed in duplicate. Lipid extraction from feces and liver. Livers and feces were stored at 280°C before analysis. Liver and fecal lipids were extracted with chloroform:methanol 2:1 (vol:vol) according to the method of Folch et al. (27). The extracted fat samples were dried and dissolved in assay buffer. Liver cholesterol and TG and fecal cholesterol concentrations were determined by using the same kits as used for blood lipid analysis. Extraction of total bile acids from feces was performed as described by Van der Meer et al. (28). Dried samples were reconstituted in 1 mL assay buffer, and total bile acid concentrations were determined by using the Colorimetric Total Bile Acids Assay kit (Diazyme Laboratories). Samples were analyzed in duplicate.
Culture-independent microbial analysis. Cecal content was used for analysis of the microbial community composition. Total bacterial DNA was extracted from the cecal content of 2 groups of mice that were supplemented with either DPC 6426 or the placebo control by using the QIAamp DNA stool mini kit (Qiagen), according to the manufacturerÕs instructions. The 16S ribosomal RNA (rRNA) bacterial gene amplicons flanking the V4–V5 region were generated with a view to high-throughput sequencing by using the Roche genome sequencer FLX platform at the Teagasc 454 sequencing facility. Amplicons were generated by using
Statistical analysis. Data are presented as per-group means 6 SEMs. Statistical analysis was performed by ANOVA, and data were analyzed by 1-factor ANOVA followed by TukeyÕs multiple-comparison test. Differences were considered significant at P < 0.05 unless otherwise stated. For pyrosequencing, Minitab release 15.1.1.0 was used to perform nonparametric statistical analysis (Mann-Whitney test) to compare 2 specific subject groups to determine the impact of dietary intervention with DPC 6426 on the microbiota (data not normally distributed; hence, the choice of a nonparametric test). SPSS software package version 18 was used to calculate Kruskal-Wallis and Mann-Whitney tests to find significant differences in microbial taxa and a diversity.
Results Weekly quantification of fecal bacterial numbers of individual mice confirmed gastrointestinal transit and recovery of administrated strains. There was no significant difference in fluid intake between the groups and all mice received expected doses (1 3 109 CFU/d). Recovery of GTF was 1 3 107 CFU/g feces, for PNZ was 3 3 107 CFU/g feces, and for DPC 6426 was 2 3 107 CFU/g feces. Body mass, relative organ, and tissue weights of experimental mice. There were no significant differences in food intake between groups (data not shown). As expected, body weight increased gradually over time for all groups and
TABLE 1 Serum profile of mice after administration of live polysaccharide-producing lactobacilli/placebo and a high-fat/highcholesterol diet for 12 wk1 b-Glucan, mg/L GTF PNZ DPC 6426 P
5.4 3 1024 6 1.1 3 1024 5.8 3 1024 6 1.2 3 1024 — —
sVCAM-1, mg/L 0.53b 0.53b 0.52b 1.16a
6 0.1 6 0.1 6 0.1 6 0.1
TC, mg/L 1700c 6 3100b 6 2100c 6 3900a 6
100.0 180.0 130.0 260.0
HDL cholesterol, mg/L 500.0 700.0 500.0 800.0
6 120.0 6 100.0 6 100.0 6 200.0
Ratio HDL-C/TC 0.10 0.20 0.40 0.10
6 6 6 6
0.00 0.10 0.10 0.00
TG, mg/L 2600c 6 3100b 6 3000b 6 3500a 6
90.00 890.0 160.0 140.0
1 Values are means 6 SEMs, n = 9/group. Labeled means in a column with superscripts without a common letter differ, P , 0.05. DPC 6426, Lactobacillus mucosae Dairy Product Culture Collection 6426, GTF, Lactobacillus paracasei National Food Biotechnology Center 338 expressing the glycosyltransferase (Gtf) gene; HDL-C, HDL cholesterol; P, placebo control; PNZ, isogenic control strain Lactobacillus paracasei National Food Biotechnology Center 338; sVCAM-1, soluble vascular cell adhesion molecule 1; TC, total cholesterol.
1958
London et al.
Downloaded from jn.nutrition.org by guest on December 4, 2017
RNA preparation and real-time fluorescence monitoring RT-PCR. Total RNA was extracted from liver samples by using the Qiagen RNase mini kit according to the manufacturerÕs protocol. Total RNA was reversetranscribed into cDNA by using the cDNA Synthesis kit (Bioline) according to the manufacturerÕs protocol. To quantify mRNA expression, PCR was performed by using a fluorescence temperature cycler (LightCycler System; Roche Diagnostics). The oligonucleotide primers for b-actin, cholesterol 7a-hydroxylase (Cyp7a1) and 3-hydroxy-3-methylglutaryl-coenzyme A (Hmg-coA) reductase, were designed on the basis of published nucleotide sequences for named genes in apoE-deficient mice (29). The oligonucleotide primers for FA synthase (Fas; 5# TCCACCTTTAAGTTGCCCTG 3#,5# TCTGCTCTCGTCATGTCACC 3#), sterol regulatory element-binding protein 1c (Srebp-c1; 5# CTCCAGCTCATCAACAACCAAGAC,5# AGAGGAGGCCAGAGAAGCAGAAGA 3#), and cluster of differentiation 36 (Cd36; 5# TGATACTATGCCCGCCTCTCC 3#,5# TTTCCCACACTCCTTTCTCCTCTA 3#) were newly designed on the basis of gene sequences available on the gene network service at the National Center for Biotechnology Information (http://www.ncbi.nlm.nih.gov/). Basic relative quantification of expression was determined by using the comparative 2-DDCt method.
one forward primer, F1 (5#-AYTGGGYDTAAAGNG), and a combination of 4 reverse primers, R1 (5#-TACCRGGGTHTCTAATCC), R2 (5#-TACCAGAGTATCTAATTC), R3 (5#-CTACDSRGGTMTCTAATC), and R4 (5#-TACNVGGGTATCTAATC). PCR was undertaken as previously described (30), and PCR products were cleaned by using an AMPure kit (Beckman Coulter Genomics) as per the manufacturerÕs protocol. Emulsion-based clonal amplification was completed as part of the 454-pyrosequencing process. Raw sequencing reads were quality trimmed as previously described (31). The SILVA 16S rRNA (version 106) database was used to BLAST (Basic Local Alignment Search Tool) the trimmed fasta sequence files using default parameters. BLAST output was parsed by using MEGAN (32), which assigns reads to National Center for Biotechnology Information taxonomies by using the Lowest Common Ancestor algorithm. A bit score of 86 was selected as previously used for 16S rRNA sequencing data (33). Clustering of sequence reads into operational taxonomical units at a 97% identity level was achieved by using Qiime. The ChimeraSlayer program was used to remove chimeras from aligned operational taxonomical units and the FastTreeMP tool generated a phylogenetic tree (34–36). a Diversity indices (Shannon diversity, Chao1, Simpson, Phylogenetic diversity, and observed species) and rarefaction curves were generated with the use of Qiime. b Diversities were also calculated on the sequence reads on the basis of weighted and unweighted Unifrac and bray Curtis distance matrices; subsequently, principal coordinate analysis was implemented.
FIGURE 1 Liver profile and fecal cholesterol excretion in mice after administration of live polysaccharide-producing lactobacilli/ placebo and high-fat/high-cholesterol diets for 12 wk. (A) Total liver lipids; (B) liver TGs; (C) total liver cholesterol; and (D) total fecal cholesterol excretion. Values are means 6 SEMs, n = 9/group. Labeled means without a common letter differ, P , 0.05. DPC 6426, Lactobacillus mucosae Dairy Product Culture Collection 6426; GTF, Lactobacillus paracasei National Food Biotechnology Center 338 expressing the glycosyltransferase (Gtf) gene; P, placebo control; PNZ, isogenic control strain Lactobacillus paracasei National Food Biotechnology Center 338.
Quantification of atherosclerotic lesions. All 4 experimental groups developed atherosclerotic lesions after 12 wk (data not shown). However, there were no statistical differences observed for plaque area and plaque density (data not shown). Effect of dietary administration of L. paracasei NFBC 338 on b-glucan concentration in serum. No differences in serum b-glucan concentrations were found after ingestion of GTF and PNZ (Table 1). Effect of administered lactobacilli on sVCMA-1 in serum. All 3 groups that received lactobacilli (GTF, PNZ, and DPC 6426) had significantly decreased serum concentrations of sVCAM1 compared with the placebo control group, which received no lactobacilli (Table 1). Serum lipid profiles. Total serum cholesterol concentration was decreased (by ;50%) for the GTF and DPC 6426 group compared with the placebo (P < 0.001) and PNZ (P < 0.01) control group (Table 1). In addition, total serum cholesterol was significantly lower for GTF and DPC 6426 groups (P < 0.01) than the PNZ group and, furthermore, the PNZ group was significantly lower (P < 0.01) when compared with the placebo group. There were no significant differences in serum HDLcholesterol concentrations between the GTF and DPC 6426 groups after 12 wk of dietary intervention (Table 1). Furthermore, the ratio of HDL cholesterol to TC was not significantly altered in any of the groups (Table 1). The serum TG concentration was significantly lower in the GTF (P < 0.001) and DPC 6426 (P < 0.05) groups than the placebo group (P < 0.05; Table 1). However, there was no significant difference between the GTF group and the PNZ group (Table 1). Liver lipids. Liver lipid concentrations were significantly decreased for the GTF (P < 0.01) and DPC 6426 (P < 0.01) groups compared with PNZ and placebo groups (Figure 1A), which relates to the findings of fatty livers (data not shown). TG accumulation in the liver was similar for the PNZ group and the placebo group and was significantly decreased (P < 0.05) for both GTF and DPC 6426 groups (Figure 1B). Hepatic total
cholesterol concentrations were similar between GTF, DPC 6426, and placebo groups but were found to be increased for the PNZ group relative to the GTF and DPC 6426 groups (P < 0.05; Figure 1C). Total bile acid and cholesterol in feces. Dietary intervention with GTF led to increased fecal cholesterol excretion compared with all other groups (P < 0.05; Figure 1D). Furthermore, there was no significant difference between fecal bile acid excretions among the groups (data not shown). Expression of hepatic key enzymes involved in fat and cholesterol metabolism. The mRNA expression of Cyp7a1 was significantly increased in the GTF group compared with all other groups (P < 0.05; Figure 2). In addition, the mRNA expression of Cd36 significantly decreased (P < 0.05) in the GTF and DPC 6426 groups compared with the PNZ and placebo control groups (Figure 2). There were no significant differences in mRNA expression of either Hmg-coA, Srebp-1c, or Fas (Figure 3). Effects on gut microbiota composition after dietary intervention with DPC 6426. Bioinformatic analysis of 16S rRNA sequence data, generated for DPC 6426 and placebo
FIGURE 2 Relative gene expression of hepatic mRNA in mice after administration of live polysaccharide-producing lactobacilli/placebo and high-fat/high-cholesterol diets for 12 wk. Values are means 6 SEMs, n = 9/group. Labeled means for each enzyme/protein without a common letter differ, P , 0.05. Cd36, cluster of differentiation 36; Cyp7a1, cholesterol 7a-hydroxylase; DPC 6426, Lactobacillus mucosae Dairy Product Culture Collection 6426; Fas, FA synthase; GTF, Lactobacillus paracasei National Food Biotechnology Center 338 expressing the glycosyltransferase (Gtf) gene; Hmg-coA, 3-hydroxy3-methylglutaryl coenzyme A; P, placebo control; PNZ, isogenic control strain Lactobacillus paracasei National Food Biotechnology Center 338; Srebp-1c, sterol regulatory element-binding protein 1. Bacterial polysaccharides reduce serum cholesterol
1959
Downloaded from jn.nutrition.org by guest on December 4, 2017
there were no significant differences in final change in body mass between groups. There were no significant differences in the masses of various adipose tissues (subcutaneous adipose tissue, epididymal adipose tissue, and mesenteric adipose tissue), liver, and cecum (data not shown).
FIGURE 3 Phylum-level distributions of the microbial communities in the cecal contents, expressed as a percentage of total population of assignable tags, in mice after administration of DPC 6426 (n = 9) or a placebo control (n = 9) and a high-fat/high-cholesterol diet for 12 wk. Labeled means for each phylum without a common letter differ, P , 0.05. DPC 6426, Lactobacillus mucosae Dairy Product Culture Collection 6426; P, placebo control.
FIGURE 4 Family-level taxonomic distribution of the microbial communities in cecal contents, expressed as a percentage of total tags assignable at the family level, in mice after administration of DPC 6426 (n = 9) or a placebo control (n = 9) and a high-fat/high-cholesterol diet for 12 wk. *Significant difference, P , 0.05. DPC 6426, Lactobacillus mucosae Dairy Product Culture Collection 6426; P, placebo control. 1960
London et al.
Discussion The present study shows that lipid metabolism in high-fat/highcholesterol–fed apoE-deficient mice is positively influenced by oral administration of GTF and DPC 6426 compared with PNZ and placebo controls, respectively. We showed that dietary intervention with GTF and DPC 6426 are both associated with significantly decreased serum cholesterol (by ;57% and ;47%, respectively), compared with the placebo control. Hong et al. (16) reported similar results for dietary bacterial b-glucan, which decreased serum total cholesterol in Sprague-Dawley rats. In our study, administration of PNZ was associated with significantly decreased TC concentration in serum (by ;32%) compared with the placebo control. These data suggest that the hypocholesterolemic effect seen for both exopolysaccharides-producing bacteria is not solely attributable to the b-glucan and exopolysaccharides but to the lactobacilli used in this study in general. This is consistent with evidence for the cholesterol-reducing abilities of other lactobacilli previously reported (38). In addition, both GTF and DPC 6426 are further associated with prevention of hypercholesterolemia by a decreased hepatic accumulation of cholesterol. However, only GTF led to the induction of hepatic Cyp7a1 mRNA levels, indicating that cholesterol homeostasis is regulated differently for GTF and DPC 6426. In some species, the excess consumption of cholesterol results in a homeostatic response that includes the induction of Cyp7a1 mRNA levels (39). This would suggest that DPC 6426 might prevent intestinal absorption of dietary cholesterol. On the contrary, we did not find significantly increased fecal cholesterol for DPC 6426 but we did for GTF. This finding is important because it indicates that the hypocholesterolemic effect of DPC 6426 and GTF might not be solely caused by a decrease in cholesterol absorption in the host, but by modifying cholesterol homeostasis through synthesis and conjugation of cholesterol to bile acids. Our study also shows that hepatic lipid contents are decreased by dietary intervention with GTF and DPC 6426 in the apoE mouse model but not by PNZ and the suppression of hepatic Cd36 mRNA levels in mice that received either GTF or DPC 6426, indicating that the uptake of FAs from circulation and intracellular transport of long-chain FAs was also altered. Furthermore, dietary intervention with GTF and DPC 6426 resulted
Downloaded from jn.nutrition.org by guest on December 4, 2017
groups, revealed a total of 306,282 reads, averaging at 18,017 reads per mouse. Rarefaction curves for each data set were parallel or approaching parallel with the x axis, demonstrating that the depth of sequencing was sufficient (data not shown). At the 97% similarity level, the Shannon Index indicated good overall biodiversity within the samples with a mean value of 4.3 (data not shown). Chao1 (an indicator of species richness) also indicated a sufficient level of overall diversity with a mean value of 703 (data not shown). Taxonomy-based analysis at the phylum level showed that the mouse gut microbiota was dominated by Firmicutes and Bacteroidetes (Figure 3), which is in agreement with previous findings (37). The most abundant families were Peptostreptococcaceae, Ruminococcaceae, and Lachnospiraceae (Figure 4); and the most abundant genera were Ruminococcaceae Incertae Sedis, Peptostreptococcaceae Incertae Sedis, Lachnospiraceae Incertae Sedis, Clostridium, Blautia, Anaerotruncus, and Allobaculum (Figure 5). The composition of the microbiota of cecal content of mice supplemented with DPC 6426 was significantly altered at family and genus levels but remained similar at the phylum level compared with placebo controls. Dietary intervention with DPC 6426 resulted in significantly less Clostridiaceae
(P < 0.05), Peptococcaceae (P < 0.001), and Staphylococcaceae (P < 0.01) in cecum compared with placebo controls (Figure 4), and was associated with the presence of Porphyromonadaceae and Prevotellaceae, whereas these families were not found in the placebo group (Figure 4). At the genus level, there were a few populations present only in the cecal microbiota of the placebo group (Figure 5); i.e., Coprococcus (P < 0.001), Eubacterium (P < 0.001), and Erysipelotrichales Incertae Sedis (P < 0.001). Furthermore, dietary administration of DPC 6426 was associated with the presence of Alkaliphilus (P < 0.05), Peptococcus (P < 0.01), Paludibacter (P < 0.05), Porphyromonas (P < 0.001), and Prevotella (P < 0.001) in cecal contents (Figure 5). Other observed differences at the genus level after dietary administration of DPC 6426 included a significant increase in cecal Allobaculum (P < 0.05), Clostridium (P < 0.01), Lachnospiraceae (P < 0.001), and Peptostreptococcaceae (P < 0.05), whereas Oscillibacter (P < 0.001) were significantly decreased (Figure 5). Principal coordinate analysis generated by using Bray Curtis, unweighted Unifrac, and weighted Unifrac distances showed that mice did not cluster into distinct groups based on whether they received DPC 6426 or not (data not shown).
FIGURE 5 Genus-level taxonomic distributions of the microbial communities present in cecal contents, expressed as a percentage of total tags assignable at the genus level, in mice after administration of DPC 6426 (n = 9) or a placebo control (n = 9) and a high-fat/highcholesterol diet for 12 wk. *Significant difference, P , 0.05. B-B-C, Brenneria-Buttiauxella-Cedecea-Citrobacter-Cronobacter-Dickeya-Edwardsiella-Enterobacter-Erwinia-Escherichia-Ewingella-Hafnia-Klebsiella-KluyveraLeclercia-Levinea-Pantoea-PectobacteriumRahnella-Raoultella-Serratia-Yersinia; DPC 6426, Lactobacillus mucosae Dairy Product Culture Collection 6426; P, placebo control.
view of the effects of dietary intervention with DPC 6426 on the entire microbial population. In conclusion, we show that dietary intervention with a b-glucan–producing probiotic strain results in modulation of lipid metabolism in a mouse model of atherosclerosis. We also show that b-glucan production by a probiotic strain conferred an advantage in terms of reduced serum cholesterol and TG concentrations and reduced hepatic lipid concentrations compared with the use of a non–b-glucan-producing isogenic control strain. Furthermore, dietary intervention with exopolysaccharides-producing DPC 6426 resulted in modulation of lipid metabolism in the mouse model of atherosclerosis, including significantly lower serum cholesterol and TG concentrations, and was associated with changes in the gut microbial community. These results confirm that dietary intervention with exopolysaccharides-producing (probiotic) strains is a feasible strategy for managing lipid metabolism. Acknowledgments GFF, RPR, NMC, and CS designed the research; LEEL, AHSK, RW, and PGC conducted the research; LEEL, AHSK, PDC, and OO analyzed the data; FS and CH helped with the manuscript and provided significant advice; LEEL and CS wrote the manuscript; and NMC and CS had primary responsibility for the final content. All authors read and approved the final manuscript.
References 1.
2. 3.
4.
FAO/WHO. Interim summary of conclusions and dietary recommendations on total fat & fatty acids. From the Joint FAO/WHO Expert Consultation on Fats and Fatty Acids in Human Nutrition, 10–14 November 2008, WHO, Geneva. Available from: http://www.who.int/ nutrition/topics/FFA_summary_rec_conclusion.pdf?ua=1. Ley RE, Turnbaugh PJ, Klein S, Gordon JI. Microbial ecology—human gut microbes associated with obesity. Nature 2006;444:1022–3. Dumas ME, Barton RH, Toye A, Cloarec O, Blancher C, Rothwell A, Fearnside J, Tatoud R, Blanc V, Lindon JC, et al. Metabolic profiling reveals a contribution of gut microbiota to fatty liver phenotype in insulin-resistant mice. Proc Natl Acad Sci USA 2006;103:12511–6. Wang Z, Klipfell E, Bennett BJ, Koeth R, Levison BS, Dugar B, Feldstein AE, Britt EB, Fu XM, Chung YM, et al. Gut flora metabolism of phosphatidylcholine promotes cardiovascular disease. Nature 2011; 472:57–63.
Bacterial polysaccharides reduce serum cholesterol
1961
Downloaded from jn.nutrition.org by guest on December 4, 2017
in significantly decreased serum TG concentrations compared with PNZ and placebo controls. This apparent discrepancy might provide a novel mechanism for how exopolysaccharidesproducing lactobacilli could affect lipid metabolism. We posit that the ingestion of GTF and DPC 6426 alters cholesterol homeostasis/lipid metabolism because of their production of b-glucan/exopolysaccharides (21), although the mechanism need to be further elucidated. In this study, all mouse groups exhibited atherosclerosis to a certain degree, a finding that is in agreement with the animal model (40). However, studies addressing the effect of lactobacilli on atherosclerosis are limited. Ingested Lactobacillus acidophilus American Type Culture Collection 4356 was recently associated with a decrease in atherosclerotic lesion size by reducing vascular oxidative stress and mitigating the inflammatory response rather than by regulating cholesterol homeostasis (41). Although we did not observe any significant atheroprotective effect after consumption of lactobacilli nor associated changes in the HDL cholesterol concentration or HDL cholesterol:TC ratio, we found that serum sVCAM-1 significantly decreased in all Lactobacillus-supplemented groups, indicating that dietary intervention with lactobacilli may positively influence the inflammation process of atherosclerosis (42, 43). Our results indicate that dietary intervention with DPC 6426 has an effect on the composition of the murine gut microbiota and that administration of this single strain can play a role in the modulation of the gut microbiota composition in vivo. Indeed, the microbiota was linked to complex disease phenotypes such as obesity and insulin resistance (44, 45). Moreover, after chronic L-carnitine supplementation, an increase in Prevotella was recently linked to a microbiota-dependent trimethylamine production and subsequent hepatic TMAO production in the host, leading to proatherosclerosis phenotypes by impairing reverse cholesterol transport, whereas an increase in Bacteroidetes led to less trimethylamine and TMAO production (7). Surprisingly, we found that dietary intervention with DPC 6426 led to an apparent enrichment in Prevotellaceae and Prevotella at taxa and genus levels, which were both absent in the cecum of placebo mice. However, we did not see any reverse effect on cholesterol homeostasis and atherosclerosis as seen before. We accept that pyrosequencing of the 16S rRNA genes does not provide quantitative population data, but it provides an over-
5.
6.
7.
8.
9. 10.
11.
12.
14.
15.
16.
17. 18.
19.
20.
21.
22.
23.
24.
25.
1962
London et al.
26. Koga M, Kai H, Yasukawa H, Kato S, Yamamoto T, Kawai Y, Kusaba K, Seki Y, Kai M, Egashira K, et al. Postnatal blocking of interferon-gamma function prevented atherosclerotic plaque formation in apolipoprotein E-knockout mice. Hypertens Res 2007;30:259–67. 27. Folch J, Lees M, Sloane Stanley GH. A simple method for the isolation and purification of total lipides from animal tissues. J Biol Chem 1957; 226:497–509. 28. Van der Meer R, De Vries HT, Glatz JFC. t-Butanol extraction of feces: a rapid procedure for enzymatic determination of fecal bile acids. Beynen AC, Geelen MJH, Katan MB, Schouten JA, eds. Cholesterol metabolism in health and disease. Wageningen: Ponsen and Looyen; 1985:113–9. 29. Han SH, Chung MJ, Lee SJ, Rhee C. Digestion-resistant fraction from soybean [Glycine max (L.) Merrill] induces hepatic LDL receptor and CYP7A1 expression in apolipoprotein E-deficient mice. J Nutr Biochem 2006;17:682–8. 30. Fouhy F, Guinane CM, Hussey S, Wall R, Ryan CA, Dempsey EM, Murphy B, Ross RP, Fitzgerald GF, Stanton C, et al. High-throughput sequencing reveals the incomplete, short-term recovery of infant gut microbiota following parenteral antibiotic treatment with ampicillin and gentamicin. Antimicrob Agents Chemother 2012;56:5811–20. 31. Claesson MJ, O’Sullivan O, Wang Q, Nikkila J, Marchesi JR, Smidt H, de Vos WM, Ross RP, O’Toole PW. Comparative analysis of pyrosequencing and a phylogenetic microarray for exploring microbial community structures in the human distal intestine. PLoS ONE 2009;4:e6669. 32. Huson DH, Auch AF, Qi J, Schuster SC. MEGAN analysis of metagenomic data. Genome Res 2007;17:377–86. 33. Urich T, Lanzen A, Qi J, Huson DH, Schleper C, Schuster SC. Simultaneous assessment of soil microbial community structure and function through analysis of the meta-transcriptome. PLoS ONE 2008;3:e2527. 34. Caporaso JG, Kuczynski J, Stombaugh J, Bittinger K, Bushman FD, Costello EK, Fierer N, Pena AG, Goodrich JK, Gordon JI, et al. QIIME allows analysis of high-throughput community sequencing data. Nat Methods 2010;7:335–6. 35. Haas BJ, Gevers D, Earl AM, Feldgarden M, Ward DV, Giannoukos G, Ciulla D, Tabbaa D, Highlander SK, Sodergren E, et al. Chimeric 16S rRNA sequence formation and detection in Sanger and 454-pyrosequenced PCR amplicons. Genome Res 2011;21:494–504. 36. Price MN, Dehal PS, Arkin AP. FastTree 2—approximately maximumlikelihood trees for large alignments. PLoS ONE 2010;5:e9490. 37. Wall R, Marques TM, O’Sullivan O, Ross RP, Shanahan F, Quigley E, Dinan TG, Kiely B, Fitzgerald GF, Cotter PD, et al. Contrasting effects of Bifidobacterium breve NCIMB 702258 and Bifidobacterium breve DPC 6330 on the composition of murine brain fatty acids and gut microbiota. Am J Clin Nutr 2012;95:1278–87. 38. Wang J, Zhang H, Chen X, Chen Y, Menghebilige, Bao Q. Selection of potential probiotic lactobacilli for cholesterol-lowering properties and their effect on cholesterol metabolism in rats fed a high-lipid diet. J Dairy Sci 2012;95:1645–54. 39. Dueland S, Drisko J, Graf L, Machleder D, Lusis AJ, Davis RA. effect of dietary-cholesterol and taurocholate on cholesterol 7-alphahydroxylase and hepatic Ldl receptors in inbred mice. J Lipid Res 1993;34:923–31. 40. Meir KS, Leitersdorf E. Atherosclerosis in the apolipoprotein E-deficient mouse—a decade of progress. Arterioscler Thromb Vasc Biol 2004;24: 1006–14. 41. Chen L, Liu WE, Li YM, Luo S, Liu QX, Zhong YM, Jian ZJ, Bao MH. Lactobacillus acidophilus ATCC 4356 attenuates the atherosclerotic progression through modulation of oxidative stress and inflammatory process. Int Immunopharmacol 2013;17:108–15. 42. Cockerill GW, Rye KA, Gamble JR, Vadas MA, Barter PJ. High-density lipoproteins inhibit cytokine-induced expression of endothelial cell adhesion molecules. Arterioscler Thromb Vasc Biol 1995;15:1987–94. 43. De Backer J, Mak R, De Bacquer D, Van Renterghem L, Verbraekel E, Kornitzer M, De Backer G. Parameters of inflammation and infection in a community based case-control study of coronary heart disease. Atherosclerosis 2002;160:457–63. 44. Turnbaugh PJ, Hamady M, Yatsunenko T, Cantarel BL, Duncan A, Ley RE, Sogin ML, Jones WJ, Roe BA, Affourtit JP, et al. A core gut microbiome in obese and lean twins. Nature 2009;457:480–4. 45. Turnbaugh PJ, Ley RE, Mahowald MA, Magrini V, Mardis ER, Gordon JI. An obesity-associated gut microbiome with increased capacity for energy harvest. Nature 2006;444:1027–31.
Downloaded from jn.nutrition.org by guest on December 4, 2017
13.
Lam V, Su JD, Koprowski S, Hsu A, Tweddell JS, Rafiee P, Gross GJ, Salzman NH, Baker JE. Intestinal microbiota determine severity of myocardial infarction in rats response. FASEB J 2012;26:4388–9. Karlsson FH, Fak F, Nookaew I, Tremaroli V, Fagerberg B, Petranovic D, Backhed F, Nielsen J. Symptomatic atherosclerosis is associated with an altered gut metagenome. Nat Commun 2012;3:1245. Koeth RA, Wang Z, Levison BS, Buffa JA, Org E, Sheehy BT, Britt EB, Fu X, Wu Y, Li L, et al. Intestinal microbiota metabolism of L-carnitine, a nutrient in red meat, promotes atherosclerosis. Nat Med 2013;19:576–85. Marlett JA, McBurney MI, Slavin JL. Position of the American Dietetic Association: health implications of dietary fiber. J Am Diet Assoc 2002; 102:993–1000. FDA. A rule by the Food and Drug Administration. Food labeling: health claims: oat and coronary heart disease. Federal Register. 2010;62:15. Wilson TA, Nicolosi RJ, Delaney B, Chadwell K, Moolchandani V, Kotyla T, Ponduru S, Zheng GH, Hess R, Knutson N, et al. Reduced and high molecular weight barley beta-glucans decrease plasma total and non-HDL-cholesterol in hypercholesterolemic Syrian golden hamsters. J Nutr 2004;134:2617–22. Nicolosi R, Bell SJ, Bistrian BR, Greenberg I, Forse RA, Blackburn GL. Plasma lipid changes after supplementation with beta-glucan fiber from yeast. Am J Clin Nutr 1999;70:208–12. Ganzle M, Michael G, Schwab C. Exopolysaccharide production by intestinal lactobacilli. In: Tannock GW, editor. Probiotics & Prebiotics: Scientific Aspects. Norfolk: Caister Academic Press. 2005. Garai-Ibabe G, Duen˜as MT, Irastorza A, Sierra-Filardi E, Werning ML, Lopez P, Corb´ı AL, Fernandez de Palencia P. Naturally occurring 2´ ´ substituted (1,3)-beta-D-glucan producing Lactobacillus suebicus and Pediococcus parvulus strains with potential utility in the production of functional foods. Bioresour Technol 2010;101:9254–63. Kim Y, Oh S, Kim SH. Released exopolysaccharide (r-EPS) produced from probiotic bacteria reduce biofilm formation of enterohemorrhagic Escherichia coli O157:H7. Biochem Biophys Res Commun 2009; 379:324–9. Rodr´ıguez C, Medici M, Rodriguez AV, Mozzi F, de Valdez GF. Prevention of chronic gastritis by fermented milks made with exopolysaccharideproducing Streptococcus thermophilus strains. J Dairy Sci 2009;92: 2423–34. Hong K, Jang KH, Lee JC, Kim S, Kim MK, Lee IY, Kim SM, Lim YH, Kang SA. Bacterial beta-glucan exhibits potent hypoglycemic activity via decrease of serum lipids and adiposity, and increase of UCP mRNA expression. J Microbiol Biotechnol 2005;15:823–30. Nakajima H, Suzuki Y, Kaizu H, Hirota T. Cholesterol lowering activity of ropy fermented milk. J Food Sci 1992;57:1327–9. Maeda H, Zhu X, Suzuki S, Suzuki K, Kitamura S. Structural characterization and biological activities of an exopolysaccharide kefiran produced by Lactobacillus kefiranofaciens WT-2B(T). J Agric Food Chem 2004;52:5533–8. Uchida M, Ishii I, Inoue C, Akisato Y, Watanabe K, Hosoyama S, Toida T, Ariyoshi N, Kitada M. Kefiran reduces atherosclerosis in rabbits fed a high cholesterol diet. J Atheroscler Thromb 2010;17:980–8. Martensson O, Biorklund M, Lambo AM, Duenas-Chasco M, Irastorza A, Holst O, Norin E, Welling G, Oste R, Onning G. Fermented, ropy, oat-based products reduce cholesterol levels and stimulate the bifidobacteria flora in humans. Nutr Res 2005;25:429–42. Stack HM, Kearney N, Stanton C, Fitzgerald GF, Ross RP. Association of beta-glucan endogenous production with increased stress tolerance of intestinal lactobacilli. Appl Environ Microbiol 2010;76:500–7. Nakashima Y, Plump AS, Raines EW, Breslow JL, Ross R. Apoedeficient mice develop lesions of all phases of atherosclerosis throughout the arterial tree. Arterioscler Thromb 1994;14:133–40. Dhar MS, Sommardahl CS, Kirkland T, Nelson S, Donnell R, Johnson DK, Castellani LW. Mice heterozygous for Atp10c, a putative amphipath, represent a novel model of obesity and type 2 diabetes. J Nutr 2004; 134:799–805. Wall R, Ross RP, Shanahan F, O’Mahony L, O’Mahony C, Coakley M, Hart O, Lawlor P, Quigley EM, Kiely B, et al. Metabolic activity of the enteric microbiota influences the fatty acid composition of murine and porcine liver and adipose tissues. Am J Clin Nutr 2009;89:1393–401. Koga M, Kai H, Yasukawa H, Yamamoto T, Kawai Y, Kato S, Kusaba K, Kai M, Egashira K, Kataoka Y, et al. Inhibition of progression and stabilization of plaques by postnatal interferon-gamma function blocking in apoE-knockout mice. Circ Res 2007;101:348–56.
