Impact of Dietary Fibers on Nutrient Management and Detoxification Organs: Gut, Liver, and Kidneys^,

Abstract

Increased dietary fiber (DF) intake elicits a wide range of physiologic effects, not just locally in the gut, but systemically. DFs can greatly alter the gut milieu by affecting the gut microbiome, which in turn influences the gut barrier, gastrointestinal immune and endocrine responses, and nitrogen cycling and microbial metabolism. These gut-associated changes can then alter the physiology and biochemistry of the body’s other main nutrient management and detoxification organs, the liver and kidneys. The molecular mechanisms by which DF alters the physiology of the gut, liver, and kidneys is likely through gut-localized events (i.e., bacterial nitrogen metabolism, microbe-microbe, and microbe–host cell interactions) coupled with specific factors that emanate from the gut in response to DF, which signal to or affect the physiology of the liver and kidneys. The latter may include microbe-derived xenometabolites, peptides, or bioactive food components made available by gut microbes, inflammation signals, and gut hormones. The intent of this review is to summarize how DF alters the gut milieu to specifically affect intestinal, liver, and kidney functions and to discuss the potential local and systemic signaling networks that are involved.

Introduction

The consumption of dietary fiber (DF)⁸ can positively affect gut health (1) as well as non-gastrointestinally related conditions such as diabetes (2), cardiovascular disease (3), nonalcoholic fatty liver disease (NAFLD) (4), and chronic kidney disease (CKD) (5). DF has a variety of physiologic effects (6–8), such as fostering the growth of select gut microbes (9), altering the production of host factors such as hormones (10) and cytokines (11), as well as the production of microbe-derived metabolites (xenometabolites) (12). With respect to specific target organs, poor gut health is increasingly recognized as an important contributor in regulating the physiology and biochemistry of nutrient management and detoxification; this concept has given rise to terms such as the gut-liver axis (13) and the gut-kidney axis (14). The intent of this review is to focus on these systems 1) by considering how various DFs affect the gut milieu to alter intestine, liver, and kidney function and 2) to provide examples of how “omics”-based technologies can be leveraged to gain novel insights into potential mechanisms and the therapeutic potential of DF. To place these topics into proper context, a brief overview of DF is presented.

Defining and classifying DF.

The definitions and nutritional aspects of DF have been comprehensively reviewed elsewhere (7, 15), and thus only the key highlights are described herein. The term “dietary fiber” encompasses a wide range of nondigestible carbohydrates. Several definitions and classification systems for DF exist, as the highly varied nature of DF has made it difficult to define and classify. The Institute of Medicine divides fiber into 2 categories: 1) DF, which consists of nondigestible carbohydrates and lignin that are intrinsic and intact in plants, and 2) functional fiber, which consists of isolated, nondigestible carbohydrates that have beneficial physiologic effects in humans. Total fiber is defined as the sum of DF and functional fiber (16). DF has been categorized on the basis of solubility, viscosity, susceptibility to fermentation by gastrointestinal bacteria, and whether the fiber occurs naturally in plants or is isolated or synthetic. The term “viscosity” (gel-forming ability) is preferred over solubility because it is a better predictor of physiologic outcomes than is solubility (7). Establishing consistent viscosity values has been difficult because viscosity can differ on the basis of the concentration of fiber, diet matrix, pH, and temperature (17). The susceptibility of fermentation by the gut microbiota has also been used to classify DF as fermentable or nonfermentable. The degree of fermentation is often assessed by the microbial production of SCFAs or the disappearance of fiber from the feces (18). Each classification system includes fibers that vary greatly in composition and structure (19), and these differences in monosaccharide content, glycosidic linkages, degree of polymerization (length of molecule), degree of substitution (side chains), and fiber preparation can contribute to the different physiologic outcomes associated with certain fiber types (20). It is important to acknowledge these differences in order to attribute health outcomes with fiber type; this information can then be leveraged to develop more specific fiber recommendations to achieve desired outcomes (i.e., reduced hepatic lipid accumulation in NAFLD, lowered serum creatinine to ameliorate CKD, etc.). A list of common fiber types, structure, and food sources is shown in Table 1.

TABLE 1.

Common fiber types, structure, and food sources

Fiber type	Structure	Sources
Lignin	Cross-linked aromatic rings (21)	Ubiquitous in plant cell walls
Cellulose	β-(1,4)-Linked glucose units (22)	Ubiquitous in plant cell walls
Arabinoxylan	β-(1,4)-Linked xylose backbone with arabinose side chains (23)	Cereal grains
Inulin	β-(2,1)-Linked fructose units typically with terminal glucose ends (24)	Onions, Jerusalem artichokes, and chicory root isolates added to processed foods to increase fiber content (25)
β-Glucan	β-(1,3)-Linked glucose units (26)	Cereals and mushrooms
Guar gum	β-(1,4)-Linked mannose residues with α-(1,6)-linked galactose side chains (27)	Guar bean
Gum acacia (gum arabic)	β-(1,3)-Linked galactose backbone with highly branched arabinose and rhamnose side chains and glycoproteins (28)	Hardened Acacia tree sap
Pectin	Complex chemical structures generally consisting of an α-(1,4)-linked galacturonic acid backbone with arabinose, galactose, and/or xylose side chains (29)	Apples, pears, peaches, and cherries (30)
Psyllium	β-(1,4)-Linked xylose backbone with arabinose and xylose side chains (31)	Seeds from the genus Plantago
Fructo-oligosaccharides	Two to 10 β-(1,2)-linked fructose units (32, 33)	Inulin degradation or transfructosylation of sucrose
Resistant starch (5 types)	α-(1,4)-Linked glucose molecules (34, 35)	Type 1: whole kernel grains
		Type 2: green bananas, high-amylose corn starch
		Type 3: cooked then cooled potatoes and rice
		Type 4: chemically cross-linked
		Type 5: lipid interactions

DF alters gut microbiota and xenometabolites.

A well-known effect of DF is the alteration of the gut microbiota. The term “gut microbiota” refers to all of the archaea, bacteria, eukaryotes (i.e., fungi and parasites), and viruses present in the gut. Bacteria are the most intensely studied and characterized; however, emerging evidence indicates that other gut microbes (fungi, viruses, and yeasts) are also important modulators of host phenotype (36, 37). The human gut microbiome is characterized by trillions of microbes that possess 150-fold more protein-coding genes than the human genome (38, 39). These microbial genes greatly expand the metabolic potential of the host by providing enzymes the host lacks, such as those that degrade various DFs (40). Unlike the human genome, which is largely fixed, the gut microbiome is plastic and can be affected by diet (12), past and present diseases (41), lifestyle factors such as exercise (42) and stress (43), or environmental exposures (44). These factors contribute to the high interindividual variation observed in the gut microbiota (45), which has made it difficult to establish consistent DF-induced bacterial changes, at least in humans.

Fermenter systems that mimic human digestion in vitro have been used to overcome the challenges of interindividuality. One such study compared the fermentation of inulin and apple pectin and found that apple pectin gave rise to a more diverse bacterial community (46). This is likely due to the complex structural and chemical nature of pectin (47). Thus, DF complexity plays an important role in microbial diversity. This is further supported by the finding that resistant starch (RS) decreases microbial diversity (48). RS has a simple structure and chemical composition [composed solely of α-(1,4)-linked glucose molecules] (49) and may therefore select for a more homogeneous microbiome than a fiber with a complex chemical structure, such as pectin. Typically, diets that consist of a variety of fiber-rich foods give rise to a more diverse gut microbiota (50, 51), and this is generally associated with better health outcomes (52, 53). It has been proposed that individuals with more diverse gut microbiota are more adept at responding to environmental challenges (54), such as resisting colonization of gut pathogens by competitive exclusion (i.e., commensal bacteria take over niches and/or consume substrates to inhibit the growth of pathogenic bacteria) (55). Interestingly, mice fed low-fiber diets showed decreased microbial diversity, which could be recovered after the introduction of a high-fiber diet; however, after generations of feeding a low-fiber diet, microbial diversity could not be recovered after the re-introduction of fiber (56). This finding may have implications to our current population, because DF consumption has decreased since the industrial revolution (57).

In summary, DF encompasses a wide variety of carbohydrates that vary greatly in chemical composition and structure. This inherent variability along with preparation method and differences in resident host microbiota contribute to the range of responses observed with the consumption of different fiber types. In general, increased DF consumption has been attributed to improved health outcomes, especially in relation to gut health.

Impact of DF on the Gut, the Gatekeeper of the Body

The gut has the dual and opposing roles of allowing nutrients to enter the body while excluding the entry of harmful substances. Both gut barrier function and nutrient absorption have been shown to be altered by DF. One example of DF-induced changes to the gut barrier is an increase in mucins and the cells that produce them, goblet cells (58, 59). Mucins are large glycoproteins that, along with water, ions, proteins, lipids, antibodies, antimicrobial peptides, and bacteria, form what is known as mucus (60). Mucus acts to protect the gut epithelium from mechanical stress, to lubricate the intestine to ease transit of digested material, and to prevent the translocation of harmful substances. A study comparing a standard rodent diet (fiber from wheat, corn, and oats comprising 4.3% of the diet by weight) with a diet devoid of any fiber showed that mice fed the fiber-deficient diet had a thinner mucus layer, thus allowing microbes to come in closer proximity to the gut epithelium (61). Without sufficient amounts of DF in the gut, bacteria may degrade the host mucus layer in order to provide themselves with the substrates necessary to survive, thus breaking down one of the host’s physical barriers.

SCFAs resulting from the fermentation of DF have been shown to bolster gut barrier function by increasing gut cell proliferation and differentiation (62). SCFAs decrease intestinal pH, which can alter the gut microbiota by inhibiting the growth of pathogens and reduce the expression of microbial virulence genes (63). Recently, it was shown that epithelial cell lines metabolize the SCFA butyrate (and to a lesser extent propionate and acetate), resulting in oxygen reduction that leads to stabilization of the transcription factor, hypoxia-inducible factor 1α (Hif-1α) (64). In the intestine, this transcription factor has been implicated in gut barrier function by regulating inflammation (65) and apoptosis (66). A microarray study found increased levels of Hif-1α expression along with increases in genes related to cell growth, proliferation, differentiation, and apoptosis in the cecal tissue of rats supplemented with 30% RS compared with rats fed an equivalent amount of energy from a low-fiber diet (67). Another component of the gut barrier affected by DF is that of tight junction proteins. One study found that feeding a standard rodent diet supplemented with 10% fructo-oligosaccharides (FOSs) increased gene expression of the jejunal tight junction proteins occludin and ZO1, reduced intestinal permeability, and lowered plasma LPS concentrations. With regard to mechanism of action, these changes were ablated by injections of glucagon-like peptide 2 (GLP-2) antagonist over 4 wk (68). GLP-2 has been shown to regulate both transcellular and paracellular gut permeability (69, 70), increase epithelial cell proliferation (71), and promote intestinal wound healing through a TGF-β–mediated mechanism (72). Notably, rats fed 2.5% pectin for 2 wk showed increased cecal SCFAs and increased plasma GLP-2 (73). Butyrate, in a Caco-2 cell culture model, was also shown to activate AMP-activated protein kinase, resulting in tight junction protein assembly and improved barrier function indicated by increased transepithelial electrical resistance (TEER) (74). Another study that used Caco-2 cells found that butyrate increased lipoxygenase activity by inhibiting histone deacetylation, resulting in increased TEER (75). The importance of fiber has also been recognized in critical care settings, because the use of total enteral or parenteral diets that lack fiber were found to induce gut atrophy and to increase gut permeability; this could be recovered with the addition of fiber or SCFAs (76, 77). Together, these studies highlight the importance of microbe derived SCFAs in bolstering the physical components of the gut barrier (mucus, cellularity, and tight junctions) through the regulation of specific cell signal pathways and transcription factors.

In addition to affecting physical barriers, DF can also alter gut immune factors. The gut is the largest immune organ in the body (78), harboring 70–80% of the body’s immune cells (79), and has been implicated as a major source of inflammation suggested to contribute to diseases such as NAFLD (80, 81) and CKD (82). Several studies have shown immunomodulatory activities for a variety of DFs, including FOSs (83), arabinoxylans (84), and β-glucans (85). FOSs (0.06% in the diet for 15 d) have been shown to increase the production of the immunoglobulin IgA in the cecum of rodents. Efficacy was dependent on FOS chain length, with shorter chain lengths resulting in higher cecal IgA concentrations (86). Shorter chain lengths resulted in higher viscosity and enhanced microbial fermentation (7, 87). IgA plays an important role in maintaining gut barrier function by binding to microbes and preventing adhesion and translocation of bacteria across the gut barrier (88). Mice supplemented with a 150-mM mix of SCFAs in the drinking water daily for 2 wk showed increased intestinal regulatory T cells (89), which are responsible for limiting intestinal inflammation. One way in which SCFAs have been shown to increase colonic regulatory T cells is by reducing histone deacetylase 6 (HDAC6) and HDAC9 gene expression, thereby increasing histone acetylation which allows for increased gene transcription. This process required the presence of the SCFA receptor, FFA receptor 2 (GPCR 43) (90). DF has been recognized as a potential dietary treatment for inflammatory bowel diseases because fiber can favorably affect gut microbe and gut immune factors found to be altered in diseases such as Crohn disease and ulcerative colitis (91). In summary, DF can bolster the gut barrier by maintaining host physical barriers (mucosal layer and cellular tight junctions) as well as by altering host immune factors. Such outcomes serve to minimize systemic proinflammatory insults that would otherwise gain access to tissues such as liver and kidneys.

In addition to altering physical barriers and intestinal immune function to minimize harm from microbe-derived proinflammatory factors, DF can also protect key organs such as the liver and kidney from metabolic insults. It has long been recognized that the consumption of nondigestible carbohydrates, in lieu of rapidly digestible carbohydrates, reduces increases in blood glucose and insulin. Another carbohydrate regulatory pathway affected by DF consumption was described: intestinal gluconeogenesis (92). Intestinal production of glucose is thought to increase glucose sensing in the portal vein, leading to decreases in hepatic glucose production and altered signaling to the brain, resulting in increased satiation. Fiber is thought to play a role via microbial fermentation of DF to propionate, which can then serve as a gluconeogenic precursor (93). One study found that mice supplemented with FOSs (10% by weight of the diet) for ∼2 wk showed increased mRNA expression of intestinal gluconeogenic enzymes [glucose-6-phosphatase catalytic subunit (G6pc), phosphoenolpyruvate carboxykinase 1 (Pck1)] and these changes were concurrent with reductions in body weight gain and improved glucose and insulin sensitivity despite no change in food intake; furthermore, these changes were ablated when FOSs were fed to intestine-specific G6pc (I-G6pc) knockout mice (93). Mice lacking I-G6pc are unable to convert propionate into glucose in the intestine; instead, the propionate is converted to glucose in the liver. The authors proposed that glucose production in the liver, rather than in the intestine, bypasses the gut-brain glucose-sensing system, ultimately resulting in impaired glucose and insulin homeostasis and increased adiposity in the I-G6pc knockout mice. Maintaining proper glucose and insulin homeostasis and preventing the accumulation of advanced glycation end-products is an important component for delaying disease progression in both NAFLD (94, 95) and CKD (96, 97). As we will see, beyond carbohydrate regulation through gut-derived events and signals, DF also plays an important role in fat and protein metabolism relevant to liver and kidneys.

Liver Responses to DF

The liver receives blood from the gut through the portal vein, and therefore this organ is a logical target of gut-derived factors influenced by diet and microbiome shifts. Indeed, DF is being considered as a potential treatment option for nongastrointestinal diseases, such as NAFLD (98). It is likely that the hepatic effects of DF involve alteration of microbiome ecology and hence gut permeability, systemic inflammation, and circulating gut-derived hormone and metabolite signals. Supporting the link between liver and gut health, patients with NAFLD have been found to exhibit an altered gut microbiota (80) and increased gut permeability (99), and several studies have found detectable levels of bacterial DNA in the serum (100) and in ascites fluid (excessive fluid accumulation in peritoneal cavity) of patients with cirrhosis (101). DFs have been shown to reduce translocation of bacterial products such as LPS (102); this would serve to reduce hepatic exposure to LPS and other microbe-derived proinflammatory products. This might reduce the likelihood of fatty liver progressing to the inflammatory form known as non-alcoholic steatohepatitis (NASH). The transition from fatty liver to NASH is thought to occur in 2 stages and is referred to as the “2-hit hypothesis.” The “first hit” is the accumulation of fat in the liver, making the liver more vulnerable to the “second hit,” which induces hepatic inflammation. The “second hit” is thought to come from a variety of sources, including bacterial overgrowth (103). In addition to affecting the gut barrier, DFs have also been shown to decrease erythrocyte lipid peroxidation and increase antioxidant enzyme activity (i.e., hepatic and erythrocyte superoxide dismutase and catalase) (104) and alter detoxifying enzymes in the liver (i.e., increase protein expression of cytochrome p450 1A2) (105, 106). Reduced concentrations of cytochrome p450 1A2 have been observed in human and NAFLD animal models (106). Increasing the activity of antioxidant and detoxification enzymes may be useful in preventing the transition from fatty liver to NASH. Animal models that examined the effects of fiber in NAFLD have shown promising results (98); however, to our knowledge, to date there have been no randomized controlled trials to determine the effectiveness of fiber on NAFLD in humans.

DF may also affect liver metabolism by altering bile acid pools. Bile acids aid in the absorption of dietary fat and fat-soluble vitamins (107) as well as serve as signaling molecules (108). Primary bile acids are made by the host in the liver and secondary bile acids are generated by the gut microbiota. Examples of gut microbiota–derived modulations to bile acids include the following: dehydration, deconjugation, desulfation, epimerization, and oxidation (109–111). Patients with cirrhosis showed lower fecal concentrations of secondary bile acids (lithocholic and deoxycholic acid decreased by an average of 23% and 68%, respectively) than did healthy controls (112). Decreased concentrations of secondary bile acids may be protective because secondary bile acids can destabilize membranes, potentially increasing intestinal permeability (113). The primary bile acid chenodeoxycholic acid serves as the strongest ligand for the farnesoid X receptor (FXR) (108). Bile acid activation of FXR and another bile acid receptor, TGR5, has been shown to decrease hepatic lipid accumulation and inflammation (114). A study in humans found decreased expression of FXR and increased expression of LXR, SREBP-1, and FAS proteins in the liver of patients with NAFLD compared with healthy controls (115). Not all bile acids serve as agonists for FXR: tauro-conjugated β- and α-muricholic acids serve as FXR antagonists and their generation is dependent on the types of gut microbes present (116). These observations support the idea that specific DFs may be a useful tool for fostering the growth of desired gut microbes to generate the types of bile acids that can favorably modulate metabolism (117, 118). However, more work needs to be done to determine how DF affects bile acid signaling pathways and to associate any observed changes with overall host phenotype (i.e., liver TG accumulation and inflammation).

DF, Kidney Function, and Nitrogen Metabolism

The kidney is another important organ affected by DF. For instance, DF may reduce nitrogen burden and systemic inflammatory insult in CKD. Just as with NAFLD, patients and animal models with CKD often exhibit an altered gut microbiota (119, 120), increased intestinal permeability (121), intestinal inflammation (122, 123), and increased blood concentrations of microbe-derived metabolites (e.g., indoxyl sulfate and p-cresol sulfate) (124). Epidemiologic studies have shown that increased DF intake reduces all-cause mortality in patients with CKD (125). The mechanisms are not clear, but one likely scenario involves maintenance of substrate delivery to the lower gut, which modifies bacterial metabolism. If sufficient amounts of nondigestible carbohydrates do not reach the colon, then other substrates such as amino acids will be fermented, resulting in the production of potentially harmful metabolites such as indoles and p-cresol, which stress the kidney (93, 126). Yet, patients with CKD are often advised to limit their consumption of many common fiber-rich foods to prevent the blood accumulation of potassium and phosphorus, minerals that can lead to cardiac arrhythmias and bone mineral disorders, respectively (127, 128).

Studies have begun to test if regimens that increase DF intake without increasing potassium and phosphorous load (i.e., through DF supplementation) can improve kidney function through the alteration of bacteria that metabolize uremic retention solutes and other kidney-relevant metabolites. A study of the fecal microbiota in patients with end-stage renal disease found increases in bacterial families possessing the enzymatic capacity to produce indole, p-cresol, urease, and uricase and decreases in families capable of producing butyrate (129). Microbial metabolism of urea by the enzyme urease generates ammonia, which can be further converted to ammonium hydroxide. Ammonium hydroxide increases intestinal pH and can lead to the disruption and loss of the intestinal tight junction proteins and thereby increase gut permeability (130, 131). Indoxyl sulfate and p-cresol sulfate are derived from microbial metabolite derivatives of tryptophan and tyrosine (indole and cresol), respectively, and have been associated with increased cardiovascular disease and all-cause mortality (132, 133). A study in hemodialysis patients supplemented with 15 g RS/d for 6 wk found reductions in plasma concentrations of indoxyl sulfate (134). Another study conducted by our group found that supplementing rats with adenine-induced CKD with 59% high-amylose RS (by weight of the diet) for 3 wk significantly improved kidney function and gut permeability indexes [i.e., decreased serum creatinine, increased creatinine clearance, improved tubulo-interstitial injury score, and restored colonic tight junction proteins (occludin and claudin-1)] (135). This treatment also markedly altered the systemic metabolome, including reducing serum concentrations of toxic metabolites known to accumulate in the blood of patients with CKD (e.g., indoxyl sulfate, p-cresol-sulfate) (135). Indole and p-cresol metabolites undergo O-sulfonation in the liver to enhance excretion and aid in detoxification (136). Excessive amounts of these metabolites may reduce the liver’s capacity to detoxify other metabolites, as evidenced by a pharmacometabolomic study in humans that found that individuals with higher urinary concentrations of p-cresol sulfate had a reduced capacity to sulfonate the common drug acetaminophen (137). Another study found that supplementing patients with chronic renal failure with 50 g guar gum/d for 4 wk increased fecal nitrogen excretion and decreased serum urea; these changes were not found with an equivalent amount of pectin supplementation (5). Recently, it was shown that 12 mo of combined probiotic and prebiotic treatment improved glomerular filtration rate better than a low protein diet (138). A reduced glomerular filtration rate can lead to the accumulation of toxins in the body (139).

One mechanism underlying kidney effects of DF may involve a decrease in the nitrogen load on the liver and the kidneys by increasing microbial biomass, which serves to sequester nitrogen in the gut and reduce the amount that enters the portal circulation (140). A study that compared germ-free and conventional mice found lower concentrations of amino acids entering the hepatic portal vein in the conventional mice; this decrease in portal amino acids was attributed to increased nitrogen demands for microbial synthesis (141). Increased microbial demand for nitrogen may also divert urea away from the liver into the intestine where it can be used for microbial synthesis (142). In addition to nitrogen sequestration in the gut, another novel mechanism by which DF could affect kidney function involves SCFAs. It has been proposed that SCFAs can modulate kidney blood flow through the activation of olfactory receptor 78, a G-coupled protein receptor located in the renal juxtaglomerular apparatus that increases renin secretion, which is responsible for controlling blood pressure (143). The modulation of blood pressure control is an important component of managing the progression of CKD (144). Taken together, these studies show that DF can improve CKD outcomes and kidney function by causing shifts in the microbiome that enhance or maintain the gut barrier (thus reducing bacterial translocation and subsequent inflammation), altering microbial nitrogen and uremic solute metabolism, and possibly influencing renal blood flow.

Systems Biology Approaches Expand Our Understanding of DF-Induced Changes in Host Metabolism

The molecular signals involved in DF-associated alterations in host systems remain largely unknown, and the integrated networks that modulate diet-microbiome-host crosstalk remain to be fully elaborated. Some aspects of signaling were discussed above for well-known molecules (i.e., SCFAs, GLP-2, and proinflammatory factors such as LPS), but there are certainly many more that remain to be discovered. Exploring the comprehensive effects of DF-induced outcomes has been made possible by the advent and widespread adoption of “omics”-based technologies (e.g., transcriptomics, proteomics, and metabolomics). These tools have allowed researchers to move beyond measuring a few classic biomarkers and to begin to explore, in an unbiased manner, how DF affects body-wide systems and the molecular events in specific tissues. This approach will prove valuable in uncovering new and unanticipated DF-related mechanisms of action and potential therapeutic targets (145).

Transcriptomics has highlighted that fibers elicit differential metabolic effects on the gut, depending on DF type. For instance, one study compared the colon mucosal transcriptome of mice fed 5 different fibers (arabinoxylan, FOSs, inulin, guar gum, or RS at 10% by weight of the diet) for 10 d. Some unique properties associated with each were as follows: arabinoxylan increased tryptophan metabolism gene expression, FOSs increased the unfolded protein response transcripts, inulin increased β-oxidation pathway mRNAs, and guar gum increased cholesterol and arachidonic acid metabolism gene expression (146). These fibers increased PPAR-γ, which is known to affect gut inflammation (147). Thus, the molecular pathways engaged by DF differ significantly depending on the specific fiber used. The disparate effects of specific DFs could be leveraged, in theory, for comparative “omics” studies to identify DF-specific microbes or microbe-derived metabolites that correlate with molecular phenotype outcomes. These would serve as candidates that could explain DF specificity on outcomes in the intestine and other tissues.

With respect to the liver, a study that supplemented rats with an inulin-rich fiber (10% of diet by weight for 4 wk) on the background of a high-fructose diet found decreased liver TGs along with differential expression of 147 hepatic genes including genes related not only to lipid metabolism but also to fibrosis and inflammation (148). For example, inulin supplementation downregulated the expression of connective tissue growth factor and decorin, which are known to play a role in fibrosis (148). These findings may shed light on new hepatic regulators relevant to NASH and NAFLD that are influenced by DF. Another study that supplemented high fat–fed mice with the nonfermentable viscous DF hydroxypropyl-methylcellulose (6% by weight of the diet for 5 wk) also found decreases in hepatic lipid accumulation and altered expression of hepatic genes involved in glucocorticoid metabolism, steroid metabolism, androgen and estrogen hormone synthesis, methylation, and oxidation reduction (149). Although the specific signals that link DF-associated changes in the gut to liver fat metabolism and gene expression remain to be determined, these experiments have identified new potential downstream targets sensitive to DF feeding.

With the use of a multi-omics approach, our group recently discovered that mice fed different forms of DF [enzyme-treated wheat bran (ETWB) or high-amylose-maize RS type 2 (HAMRS2)] at 20% of the diet exhibited marked changes in the liver transcriptome and metabolome (150, 151). For example, metabolomics revealed almost-uniformly reduced hepatic amino acid concentrations in HAMRS2-fed mice despite normal blood concentrations, suggestive of a DF-associated shift in liver amino acid intermediary metabolism. Several bacterial taxa tended to shift along with changes in liver amino acid abundances (i.e., Ruminococcaceae was negatively correlated with many liver amino acids, whereas Lachnospiraceae was positively correlated). These unexpected findings point to new biology associated with DF feeding; namely, hepatic nitrogen and amino acid metabolism can be dramatically altered in response to DF. Another unexpected observation was that livers in ETWB-fed mice also showed significantly altered metabolite and gene expression patterns that, in some ways, mimicked the fasting state. For instance, the rate-limiting enzyme for gluconeogenesis, Pck1, was upregulated in ETWB-fed mice; this enzyme is increased during fasting (152). Liver and blood concentrations of the ketone body β-hydroxybutyrate were also increased in ETWB-fed mice. As with the HAMRS2 effects in the liver, many of these changes were significantly correlated with DF-associated alterations in specific gut microbiota, suggesting that factors emanating from these microbes were involved.

Metabolomics has been used to identify metabolite shifts that occur in CKD (153). Recently, this approach was extended to uncover the effects of different rhubarb extracts in rats with CKD. The authors found that treatment with various rhubarb extracts restored urine metabolite abnormalities (i.e., increased creatinine, decreased pyrimidine) and improved renal function and kidney histopathology (154). This study did not study the effects of DF; however, this same approach can be used to determine which metabolic pathways are affected by CKD and whether these perturbations are normalized by treatment. Note that many sources of DF are accompanied by phytochemicals and it may be the phytochemicals, and not the DF per se, that have an impact. The administration of the intact fiber source and isolated phytochemicals should prove useful in untangling the specific beneficial effects of DF.

Summary and Future Directions

“Omics” studies have begun to successfully catalog the changes in hundreds of variables in response to DF (i.e., DF-associated alterations in microbes, metabolites, and transcripts). Some of these variables are significantly correlated with biological phenotypes. However, the field is rapidly moving beyond this descriptive phase, and future efforts will consider how the identified variables affect target organs at a molecular level. For example, it is increasingly appreciated that simply characterizing the types of bacteria present may not provide adequate evidence to draw reliable conclusions with regard to gut microbiota influences on host phenotype; it may be more biologically relevant to identify the metabolic activities of the gut microbiota via meta-transcriptomics (bacterial gene expression), rather than simply quantifying the bacteria that are present. Stated another way, what the bacteria are doing is more important than which bacteria are there (155). Meta-transcriptomics will likely aid in the identification of bacterial species that are responsible for producing specific xenometabolites. Effects of the identified xenometabolites can then be studied in host cell culture systems as well as in germ-free and humanized animal models to unravel how the gut microbiota communicates with the intestines, liver, kidney, and other organs.

These technologies can also be used to identify patterns of interindividual variability. Fiber supplementation studies in humans have often generated inconsistent results that may be accounted for by differences in resident gut microbes. The ability to classify people as responders or nonresponders by assessing variables such as fecal SCFAs or breath hydrogen can aid in identifying gut microbe communities that are necessary to elicit a response to DF supplementation (156). These microbes can then be supplied to nonresponders along with the DF to determine if supplementing the “missing microbe” does indeed result in an enhanced response to the DF intervention.

In conclusion, DFs alter the gut environment by a number of mechanisms, including fostering the growth of select bacteria, which leads to altered microbial metabolite production and host immune response. An overview of fiber-induced changes in gut, liver, and kidney is shown in Figure 1. Fiber-induced gut changes can result in enhanced gut barrier function that protects the liver and kidney from translocation of proinflammatory bacteria and bacterial products. This could allow the liver and kidneys to devote more capacity to metabolism-associated processes rather than controlling inflammation (104). In addition, DF increases microbial sequestration of nitrogen in the gut, resulting in increased fecal nitrogen excretion and reduced concentrations of nitrogenous metabolites in the blood. Reduced nitrogenous burden on the kidneys is desired for the treatment of diseases such as CKD. However, not all fibers behave the same way, with differences in fiber structure and preparation resulting in varied outcomes; fiber from a diverse range of sources is likely to provide the most health benefits. Unfortunately, most Americans do not consume enough fiber (157), prompting the 2015 Dietary Guidelines Advisory Committee to name fiber a nutrient of concern (158). Fiber intake has declined over time as has the diversity of our microbiomes (159), and this decreased diversity is generally associated with poor health outcomes (54). Increasing the amount of fiber in the diet is likely one way to increase diversity and ameliorate many diseases beyond the gut. An in-depth understanding of the specific microbes and signals that are altered in response to fibers (and the host tissue molecular targets) will support the development of evidence-based strategies to improve health and thwart diseases such as gastrointestinal disorders, NAFLD, and CKD.

FIGURE 1.

Schematic overview of the major mechanisms by which dietary fiber affects gut, liver, and kidneys.