Prebiotic fiber modulation of the gut microbiota improves risk factors for obesity and the metabolic syndrome

Abstract

Prebiotic fibers are non-digestible carbohydrates that promote the growth of beneficial bacteria in the gut. Prebiotic consumption may benefit obesity and associated co-morbidities by improving or normalizing the dysbiosis of the gut microbiota. We evaluated the dose response to a prebiotic diet on the gut microbiota, body composition and obesity associated risk factors in lean and genetically obese rats. Prebiotic fibers increased Firmicutes and decreased Bacteroidetes, a profile often associated with a leaner phenotype. Bifidobacteria and Lactobacillus numbers also increased. Changes in the gut microbiota correlated with energy intake, glucose, insulin, satiety hormones, and hepatic cholesterol and triglyceride accumulation. Here we provide a comprehensive analysis evaluating the results through the lens of the gut microbiota. Salient, new developments impacting the interpretation and significance of our data are discussed. We propose that prebiotic fibers have promise as a safe and cost-effective means of modulating the gut microbiota to promote improved host:bacterial interactions in obesity and insulin resistance. Human clinical trials should be undertaken to confirm these effects.

Introduction

The human gastrointestinal (GI) tract contains an estimated 10¹⁴ microbial cells, primarily inhabiting the colon, and representing over 1,000 bacterial types.¹ Historically, the profile of the gut microbiota has been highly scrutinized, as it is intricately related to physical wellbeing; with some strains promoting health and others disease. In addition to its role in establishing and maintaining normal intestinal health, the gut microbiota has been found to improve or exacerbate a myriad of diseases ranging from colorectal cancer to autoimmune and allergic diseases.² Certain gut microbes have been found to be protective, for example, the higher levels of Faecalibacterium prausnitzii found in healthy subjects compared with patients with colitis are important given the protective effect of F. prausnitzii on gut mucosa.³ Additionally, colonization of germ free mice with Bacteroides fragilis produces polysaccharide A (PSA), which positively modulates the host’s T-cell dependent immune response.⁴ Recently, however, the role of the gut microbiota in the development of obesity and its associated co-morbidities has come to the forefront. Dysbiosis of the gut microbiota is present in obesity with a reduction in Bacteroidetes and an increase in Firmicutes typically reported^5,6 although not consistently.^7,8 Ley et al.⁶ also report that weight loss increases the proportion of Bacteroidetes relative to the Firmicutes. Within the Archaea, obesity is associated with a decrease in Methanobrevibacter smithii.⁹ Whether the gut microbiota promotes obesity or the changes occur as a result of obesity requires further research. A prospective study in children found that Bifidobacterium spp numbers were higher and Staphylococcus aureus were lower in children that were normal weight at the outset of 7 y,¹⁰ providing some support for the former hypothesis. Conversely, a high-fat, Western diet has been found to increase Firmicutes and decrease Bacteroidetes in the absence of weight gain.¹¹ It is entirely possible that both the obese phenotype and diet affect the gut microbiota. Furthermore, it has been proposed that changes at the microbial community level impact obesity, however, less data are available on these groups and this would be a fruitful area for future exploration.

The role of the gut microbiota, as it relates to obesity development, has not been fully defined; however, is evidenced by the elegant experiments of Turnbaugh et al.¹² who demonstrated colonization of gnotobiotic mice with microbiota extracted from obese animals resulted in increased fat mass compared with those colonized with microbiota from lean animals. It has been suggested that the gut microbiota from, an obese phenotype, extract the energy from foods more efficiently, resulting in increased adiposity. The energy extraction hypothesis is based on comparisons between germ free and conventionalized mice. Here the gut microbiota increased monosaccharide uptake in the gut resulting in increased production of short chain fatty acids (SCFA). One of the chief SCFA is acetate, which has been found to stimulate hepatic de novo lipogenesis, and consequently increase adipocyte fatty acid storage.¹³ SCFA may also promote fat storage by acting as signaling molecules for the G-protein coupled receptors (GPR) 41 and 43. GPR43 is expressed in intestine, adipocytes, and immune cells and its deficiency protects mice from high fat diet-induced obesity.^14,15 Furthermore, germ free GPR41 knockout mice are leaner and weigh less than their wild type littermates, but only when colonized with a model fermentative community composed of B. thetaiotaomicron and M. smithii.¹⁶ The gut microbiota has also been proposed to increase adipocyte fatty acid storage through the suppression of intestinal fasting-induced adipose factor (FIAF). Suppression of FIAF in the gut epithelium increases lipoprotein lipase (LPL) activity which, in turn, promotes triglyceride storage in adipocytes.¹⁷ Conversely, a hypothesis has been proposed whereby the gut microbiota suppresses AMP-activated protein kinase (AMPK) and consequently fatty acid oxidation resulting in increased adiposity.¹⁸ It should be noted that these two mechanisms are not mutually exclusive.

The gut microbiota is highly active and produces many metabolic byproducts resulting in both local and systemic effects. Research into the role of these metabolic byproducts on obesity and the associated inflammation is still in its infancy; however, it has been shown that lipopolysaccharide (LPS) levels negatively correlate with bifidobacteria numbers.¹⁹ LPS is secreted by gram negative bacteria and has been found to increase levels of tumor necrosis factor α (TNF-α), a potent inflammatory cytokine linked to obesity and type 2 diabetes. This has been dubbed metabolic endotoxemia.¹⁹ Importantly, the relationship between bifidobacteria and reduced LPS levels requires additional investigation to determine the specific mechanism(s) involved, however, it may be linked to improvements in gut mucosal barrier functions.¹⁹ On the other hand, antibiotic induced changes in the gut microbiota reduced metabolic endotoxemia in genetically obese and diet-induced, obese mice.²⁰

Bacterial colonization begins at, or possibly before birth,²¹ and can be influenced throughout the lifespan by a variety of factors, of which diet is a key contributor.²² Prebiotic fibers have been defined by Roberfroid as “non-digestible food ingredients that benefit the host by selectively stimulating the growth or activity of one or a limited number of bacteria in the colon.”²³ The full effects of prebiotic fibers on the microbiota have yet to be determined; however, increases in lactobacilli and bifidobacteria are consistently reported.^24,25 Increases in Bifidobacterium are relevant to obesity, as high fat feeding reduced Bifidobacterium spp, and Eubacterium rectal-Clostridium coccoides groups in mice.²⁶

Given the abnormalities associated with the obese gut microbiota and the ability of prebiotic fibers to alter its composition, we set out to examine the effects of increasing doses of prebiotic fiber, as compared with a control diet, on the gut microbiota and physiological parameters related to obesity. This is important given a growing need to enhance our understanding of how dietary choices can positively modify the gut microbiota and improve weight loss outcomes. Considering the complexity of achieving and maintaining weight loss, adding another tool to the arsenal is crucial.

Modulation of the Gut Microbiota by Prebiotic Fibers

Our article, published in the British Journal of Nutrition,²⁷ identifies several changes in the gut microbiota, of animals exposed to prebiotic fibers, and links these changes to alterations in physiology. We designed three diets: control 0% prebiotic, 10% prebiotic, and 20% prebiotic fiber by weight. In humans, doses greater than 30 g/day, which equates to an approximate 5% dose by weight, produce unfavorable gastrointestinal symptoms.^28,29 The majority of animal studies have evaluated a 10% dose by weight. By examining a range from 0–20% in our study, our goal was to determine dose-response effects and whether there exists a threshold above which no further benefits are identified. All diets provided an equal contribution of protein and fat to the total energy value. Subsequently, lean and genetically obese rats were exposed to the diets for 10 weeks. Cecal samples were obtained from the rats, total bacterial DNA was extracted and total bacteria, Bacteroides/Prevotella, Clostridium coccoides, Clostridium leptum, Lactobacillus, Bifidobacterium, and Enterobacteriaceae were quantified by qPCR.

This work supports previous reports of decreased numbers of Bacteroides and increased Firmicutes with obesity.⁶ However, the Firmicutes phylum includes at least 250 genera and the Bacteroidetes at least 20; therefore, the impact of obesity at the division level is of interest³⁰ and requires further exploration. To this end, Million et al.⁹ found lower levels of B. animalis in obesity and alterations in various strains of Lactobacillus compared with lean. Here we report that the presence of obesity, without any dietary intervention, reduced total bacteria, Clostridium leptum and Enterobacteriaceae numbers as compared with the lean animals. Others work supports reduced Clostridum leptum in humans with obesity.⁸ The implications of reduced Enterobacteriaceae in obesity remain to be elucidated; however, these results are in opposition to others that found increased Enterobacteriaceae and Escherichia coli in overweight, pregnant women compared with normal weight, pregnant women.³¹ The significance of these results to non-pregnant populations, however, may be limited.

Correlational relationships between body weight and body fat, as determined by dual energy X-ray absorptiometry (DXA), revealed Bacteriodes and total bacterial were negatively correlated with percent body fat and body weight; whereas, levels of Lactobacillus were positively correlated with body weight and fat. Other accounts of increased Lactobacillus with obesity exist,⁹ however, the genus Lactobacillus contains many different species with potentially differing effects. A recent study looked at seven different strains of Lactobacillus, in the lean and obese state. L. paracasei and L. plantarum were associated with a lean phenotype whereas L. reuteri was increased in obesity.⁹ The implications of increased Lacto-bacillus are unclear, however, given that probiotic administration of Lactobacilli gaseri SBT2055 has been found to decrease visceral and subcutaneous fat mass in overweight adults (BMI 24.2–30.7 kg/m²).³² On the other hand, it has been suggested that L. reuteri can enhance nutrient absorption and processing by the intestine.^9,33 L. reuteri will also produce reuterin in the presence of glycerol which could impact obesity and inflammation given reuterins’ antimicrobial properties and potential for modifying gut microbiota.³⁴ As evidenced by the Million et al.⁹ study, the effects likely depend on the specific species involved.

In our experiment, we were also able to evaluate how the obese microbiota responds to a dietary intervention compared with the response by lean microbiota. The lean and obese animals in the 10% prebiotic fiber group, had lower levels of total bacteria, however, Clostridium leptum and Enterobacteriaceae numbers normalized, with no difference between the lean and obese groups. With the 20% prebiotic fiber group, levels of Clostridium coccoides, C. leptum, Lactobacillus, Bifidobacterium, and Enterobacteriaceae were all increased with obesity compared with the lean animals. Limited data are available regarding the Clostridium group in obesity. One study found a trend (p = 0.088) toward increased C. coccoides in obese pregnant women compared with lean³¹ and another study showed no change with overweight or obesity.⁸ Faecalibacterium prausnitzii, of the class Clostridia and member of the C. leptum cluster, were also found to be increased in obese children living in south India.³⁵ Importantly, however, in the absence of the dietary intervention we found reduced levels of Clostridium leptum and Enterobacteriaceae numbers. This discrepancy highlights the importance of the interaction between diet and physiology and the complexity of the gut microbiota. Others also report increased Bifidobacterium levels with prebiotic fibers in obesity.¹⁹ Increases in Bifidobacterium are noteworthy as levels have been reported to be low in obesity⁸ and bifidobacteria are associated with health improvements due to a reduction in inflammatory cytokines.¹⁹ It is likely that changes at the smaller community level have a significant impact on disease progression and this remains relatively unexplored. With respect to our results, it is interesting to report that the response to dietary interventions is altered by the host physiology (i.e., genetically lean or obese).

When we evaluate the effects of diet independently of genetic grouping, supplementation with prebiotic fiber decreased the Firmicutes equally in the 10% and 20% dose; however, this pattern was reversed in the Bacteroidetes groups. Recently, Everard et al.³⁶ conducted a comprehensive evaluation of the effects of prebiotic fibers on the gut microbiota, in genetically obese and diet-induced, leptin-resistant mice. They support our results with increased Bacteroidetes and decreased Firmicutes. It is important to note that although the lean and obese animals were exposed to the same diets, the amount of food consumed by the obese animals was significantly greater than the lean animals. For example, the average food intake, over the ten weeks, for the obese 20% group was 1.82 kg of which 356 g was prebiotic fiber; whereas, in the lean 20% group the average food intake was 1.08 kg of which 210 g was prebiotic fiber. Thus, an obese animal on a 20% prebiotic diet is consuming a greater absolute amount of fiber because of increased overall food consumption. This makes it difficult to distinguish if the effects are related to the dose or the disease state or a combination of the two. A further limitation is that the energy values of the prebiotic fiber diets are lower due to the lower energy value of the fiber as compared with the cornstarch.

Overall, these results support an aberrant gut microbial profile in obesity and improvements with prebiotic fiber supplementation. The effects of prebiotic fibers were significantly altered by both the prebiotic dose and the disease state. This suggests the lean and obese physiology will respond differently to treatments designed to modulate the gut microbiota and this should be considered in future research and clinical practice. Furthermore, the effective dose and the feasibility of using prebiotic fibers as a treatment should be determined via human clinical trials.

Gut Microbiota and Enteroendocrine Hormone Secretion

The gut is an active enteroendocrine organ that secretes several hormones in response to food stimulus or lack of food. The foods eaten can alter the expression of these hormones which, in turn, act on the brain to signal hunger or satiety. To evaluate the effects of prebiotic fibers on satiety hormones, we provided the animals with 5 g of their diet and subsequently measured plasma levels of satiety hormones over 90 min. Glucagon-like peptide-1 (GLP-1) is an anorexigenic, appetite suppressing hormone, proposed to reduce GI transit time by acting as an ileal break.³⁷ Furthermore, GLP-1 stimulates insulin secretion from the pancreatic β cells and inhibits glucagon secretion.³⁸ In our study, postprandial GLP-1 levels increased with the 20% prebiotic fiber diet. Given that we also measured select gut bacteria, we were interested to determine if changes in satiety hormones could be linked to alterations in the gut microbiota. Of note, Enterobacteriaceae increased in conjunction with GLP-1 total area under the curve (tAUC). Further research into the relationship between satiety hormones and the gut microbiota is needed; however, it has been reported that GI transit time can alter the profile of the gut microbiota.^39–41 In support of this theory, we report both dysbiosis of the gut microbiota and increased gastric emptying in the obese rats as compared with the lean rats. We did not find a significant reduction in GI transit time with prebiotic supplementation, however, our sample size was small (n = 6). We do report increased GLP-1 and an upregulation of peptide YY (PYY) gene expression with prebiotic supplementation. Given that PYY and GLP-1 slow GI transit time,^37,42 this might provide another mechanism whereby prebiotic fibers affect the composition of the gut microbiota. Prebiotic fibers also modulate the gut microbiota in a cyclic manner. When the bacteria metabolize prebiotic fibers,⁴³ they produce SCFA, which lowers the luminal pH. This affects the composition of the gut microbiota, as not all species can thrive in this environment.⁸ Additionally, the gut microbiota influences satiety hormone production. Metabolism of non-digestible carbohydrates by the gut bacteria results in the production of SCFA, which have been shown to upregulate gene expression of proglucagon, the precursor to GLP-1,^44,45 and PYY⁴³ in the intestinal tract.

Ghrelin is an orexigenic hormone that stimulates appetite.⁴⁶ Here we found decreased plasma levels of des-acyl ghrelin with prebiotic fiber supplementation in the lean group. Ghrelin levels in the obese group did not change, however, others note that ghrelin levels are reduced in obesity and the meal response is attenuated, both of which we also observed.^47,48 In our study, Bacteroides and total bacteria were positively correlated with ghrelin tAUC. The interplay between gut endocrine functions and the microbiota and their metabolic byproducts requires further investigation.

We also assessed the gene expression of select hormones throughout the GI tract. Proglucagon is the precursor for GLP-1, discussed above, but also for glucagon-like peptide-2 (GLP-2). In this experiment, proglucagon levels increased with prebiotic supplementation. GLP-2 is a gut trophic hormone, involved in maintaining the gut epithelial barrier through the tight-junction integrity.^49,50 In mice, a high fat diet resulted in increased gut permeability and disruption of tight-junction proteins. This was associated with an increase in metabolic endotoxemia.^20,50 Prebiotic treatment, on the other hand, improved the gut barrier and reduced LPS levels.⁵⁰ Prebiotic-stimulated GLP-2 secretion may improve host health by improving the gut barrier and allowing for tighter regulation of the transport of bacterial metabolic byproducts; thereby, reducing the systemic inflammation associated with obesity and related metabolic disorders.^2,50 This remains to be demonstrated in humans, however, and a study by Brignardello et al.⁵¹ reports no differences in intestinal permeability in obese vs. lean individuals.

Gut Microbiota and Glycemic Regulation

Aberrant microbiota has also been linked to type 2 diabetes, with reduced numbers of the Firmicutes, Clostridia, Bacteroides vulgates, and Bifidobacteria in adults with type 2 diabetes.^2,52,53 Furthermore, the ratio of Bacteroidetes to Firmicutes is positively correlated with plasma glucose concentrations.⁵³ In support of gut microbial dysbiosis in type 2 diabetes development, we found Bacteroides and total bacteria were negatively correlated with fasting insulin and insulin incremental area under the curve (iAUC); whereas, Enterobacteriaceae increased in conjunction with glucose iAUC. Others report the ratios of Bacteroides/Prevotella to C. coccoides–E. rectal are positively correlated with plasma glucose concentrations.⁵³ The results of the few human clinical trials involving prebiotic fibers or probiotics on patients with overweight or diabetes mellitus has recently been reviewed by Delzenne et al.⁵⁴ The role of prebiotic fibers in glucoregulation is uncertain, as improvements, in humans, are not consistently reported⁵⁴ and the role of the gut microbiota is unknown, as these studies did not measure the profile of the gut microbiota. Several animal studies report improvements in glycemic regulation with prebiotic supplementation,^55,56 however, no changes in glucose or insulin iAUC were noted with prebiotic supplementation in our animals. Conversely, Lactobacillus acidophilus NCFM administration improved insulin sensitivity in a subset of patients with type 2 diabetes, impaired glucose tolerance, and normal glucose tolerance as compared with a placebo.⁵⁷ This data suggests modification of the gut microbial profile could reduce the risk of type 2 diabetes. The discrepancy in outcomes with prebiotic fiber supplementation between human and animal studies may be a result of dosage, as human studies typically administer a lower dose due to gastrointestinal discomfort experienced by patients with higher dosages. Differences between pre and probiotic effects are likely due to the differences in bacterial strains. The effects of prebiotics are broader than supplementation with a single probiotic and would be affected by the host’s gut microbiota.

Gut Microbiota and Lipid Metabolism

The experiments to assess the effects of prebiotic fibers on obesity and associated risk factors were published in two separate papers. The second paper focused on the effects of prebiotic fibers on lipid metabolism.⁵⁸ Given that the same animals were used for both experiments we can report in this addendum the interactions between the gut microbiota and lipid metabolism. The gut microbiota stimulate monosaccharide uptake and transfer to the liver resulting in subsequent stimulation of de novo lipogenesis.¹⁷ Furthermore, the gut microbiota may affect hepatic fatty acid storage through changes to bile acids and SCFA production.⁵⁹ In our study, Bacteroides numbers were negatively correlated with cecal triglyceride (TG), liver total cholesterol (TC), serum TC, and serum TG. Lactobacillus positively correlated with liver TC and serum TG and C. coccoides with liver TC whereas C. leptum was positively correlated with liver TC. Enterobacteriaceae was negatively correlated with serum TG. The role of the gut microbiota in response to liver disease has recently been reviewed.⁶⁰ There is significant evidence that the endotoxins produce by microbial metabolism play a role in non-alcoholic fatty liver diseases (NAFLD) as plasma endotoxin levels are elevated in NAFLD, a phenomenon linked to intestinal overgrowth and induction of toll-like receptor 4 (TLR4) in the hepatocytes.^61,62 The gut microbiota has been clearly linked to NAFLD, however, dietary modulation of the gut microbiota, via prebiotics, for the treatment of NAFLD remains an untested possibility.

Conclusion

Dysbiosis of the gut microbiota is present in a number of metabolic disorders including obesity, type 2 diabetes, and NAFLD. Prebiotic fibers have the ability to alter the gut microbiota in a positive manner, indicating their promise as a dietary treatment. Several animal studies, including our recent publications, have been undertaken and provide promising results. Human clinical trials are now necessary to determine if these benefits are translatable. One major limitation to prebiotic fiber usage in humans is the effective dose. The current dose of 10% used in the majority of animal studies is not feasible in humans due to increased GI side-effects. Future research should evaluate the physiological benefits of prebiotic fibers, in human patients, with a variety of diseases associated with the metabolic syndrome and their relationship to the gut microbiota.

Abbreviations

AMPK

AMP-activated protein kinase

DXA

dual energy X-ray absorptiometry

FIAF

fasting-induced adipose factor

GI

gastrointestinal

GLP-1

glucagon-like peptide-1

GLP-2

glucagon-like peptide-2

GPR 41

G-protein coupled receptor 41

GPR43

G-protein coupled receptor 43

iAUC

incremental area under the curve

LPL

lipoprotein lipase

LPS

lipopolysaccharide

NAFLD

non-alcoholic fatty liver disease

PYY

peptide YY

SCFA

short chain fatty acids

TLR4

Toll-like receptor 4

tAUC

total area under the curve

TC

total cholesterol

TG

triglyceride

TNFα

tumor necrosis factor α