Bile acids and their effects on diabetes

Abstract

Diabetes is a widespread, rapidly increasing metabolic disease that is driven by hyperglycemia. Early glycemic control is of primary importance to avoid vascular complications including development of retinal disorders leading to blindness, end-stage renal disease, and accelerated atherosclerosis with a higher risk of myocardial infarction, stroke and limb amputations. Even after hyperglycemia has been brought under control, “metabolic memory,” a cluster of irreversible metabolic changes that allow diabetes to progress, may persist depending on the duration of hyperglycemia. Manipulation of bile acid (BA) receptors and the BA pool have been shown to be useful in establishing glycemic control in diabetes due to their ability to regulate energy metabolism by binding and activating nuclear transcription factors such as farnesoid X receptor (FXR) in liver and intestine as well as the G-protein coupled receptor, TGR5, in enteroendocrine cells and pancreatic β-cells. The downstream targets of BA activated FXR, FGF15/21, are also important for glucose/insulin homeostasis. In this review we will discuss the effect of BAs on glucose and lipid metabolism and explore recent research on establishing glycemic control in diabetes through the manipulation of BAs and their receptors in the liver, intestine and pancreas, alteration of the enterohepatic circulation, bariatric surgery and alignment of circadian rhythms.

Keywords

bile acids / metabolic memory / diabetes / circadian rhythm / bariatric surgery

Introduction

Diabetes is a widespread and increasing global health problem that is driven by hyperglycemia. Intensive efforts to control hyperglycemia include administration of insulin or drugs that: (1) increase the endogenous release of insulin, (2) decrease release of glucose from the liver, (3) increase the use of glucose in both skeletal muscle and adipose tissue, (4) delay glucose absorption from food and, (5) upregulate the incretin system [¹,²]. The importance of early intensive glycemic control to reduce risk of micro- and macrovascular complications has been shown by many large studies [³–⁵]. As a result of hyperglycemia-induced vascular complications, diabetic patients may develop retinal disorders leading to blindness, end-stage renal disease, and accelerated atherosclerosis that leads to a higher risk of myocardial infarction, stroke and limb amputations [⁶].

Regulation of hepatic glucose metabolism is of key importance to overall energy metabolism because the liver orchestrates not only the production of glucose via gluconeogenesis and glycogenolysis but also its storage via glycogenesis and lipogenesis. Hyperglycemia can eventually exhaust glycogen storage and glucose will then be diverted into the production of lipids that are then exported as very low density lipoproteins which in turn, are converted into triglycerides in adipose tissue. Therefore increased blood glucose leads to increased lipidemia and obesity [⁷].

The early glycemic environment is proposed to be “remembered” and this “metabolic memory” is thought to be due to the accumulation of reactive species during hyperglycemia that persist after glycemic control is established. Four key pathways (Fig. 1) have been suggested to be involved in diabetic complications; (1) increased flux of the polyol pathway, (2) increased production of advanced glycosylation products (AGE), (3) increased amounts of dihydroxyacetone phosphate, and (4) increased flux of the hexosamine pathway (Fig. 1) [²,⁶]. Whereas, treating type I diabetes (T1D) with insulin as early as possible may have merit, the situation is different for type II diabetes (T2D) where insulin is usually not the first line treatment. Therefore a complete understanding of metabolic control of nutrients, in particular, glucose and lipid metabolism, is essential in order to be able to intervene aggressively, perhaps by means other than insulin administration, to combat hyperglycemia for T2D patients at its onset [¹]. In the following review we will explore recent research on the effect of bile acids (BAs) on glucose and lipid metabolism, the enteroendocrine system, as well as glycemic control via alteration of enterohepatic circulation of BAs, bariatric surgery and alignment of circadian rhythms. The articles evaluated in this review were chosen from recent articles published within the past 5 years in peer-reviewed journals. The keywords listed above were used to generate a search of the PUBMED database and representative articles pertaining to the involvement of BAs in the glycemic control of T2D were then selected.

Bile acid control of hepatic glucose and lipid metabolism via farnesoid X receptor

Bile acids (BAs) are synthesized in hepatocytes via cytochrome P450-mediated oxidation of cholesterol, which can occur through two pathways, the “classical” and “alternative” pathways. The “classical” pathway produces the primary BAs, cholic acid (CA) and chenodeoxycholic acid (CDCA) through the actions of the enzymes, CYP7A1 and CYP8B1 [¹¹]. CYP7A1 is the rate-limiting enzyme in the classical pathway and converts cholesterol to CDCA and the 12a-hydroxylase, CYP8B1 mediates the production of CA. Cholesterol will be converted to CDCA when CYP8B1 is deficient and thus CYP8B1 controls the CA/CDCA ratio [¹²]. The alternative pathway which involves CYP27A1 results in the production of CDCA. Notably, BA synthesis is under control of the farnesoid X receptor (FXR). Hepatic FXR via its effect on transcription of small heterodimer partner (SHP) suppresses the expression of CYP8B1 and to a lesser extent, CYP7A1, while intestinal FXR activation results in the production of FGF19 (15 in rodents) that upon binding to hepatic FGFR4/b-Klotho (bottom right, Fig. 3) causes ERK1/2 activation and suppresses both hepatic CYP7A1 and CYP8B1 expression [¹²]. After the primary BAs are synthesized they are then conjugated to either taurine (mainly in rodents) or to glycine (predominant in humans) by BA:CoA synthetase (BACS) and BA-CoA: amino acid N-acyltransferase (BAAT) enzymes to form TCA, TCDCA, GCA, and GCDCA. Next they are secreted from the liver into the bile caniculus via the canicular bile salt export pump (BSEP) where they are further modified via sulphation (sulfotransferase family 2A member 1 enzyme or SULT2A1) or via glucuronidation (UDP-glucuronosyltransferase or UGT enzymes) and secreted from the liver into the bile via the multidrug resistance-associated protein 2 transporter (MRP2) [¹³,¹⁴]. After a meal, the gallbladder contracts and releases bile salts in the form of mixed micelles which contain bile salts, cholesterol, and phospholipids, into the intestinal lumen, thus facilitating the emulsification and absorption of lipids in the small intestine [¹¹]. Once bile salts are in the small intestine, the gut microbiota use bile salt hydrolase enzymes (BSH) to deconjugate the bile salts and a portion of these are further transformed in the large intestine into the secondary BAs, i.e., CA is converted into deoxycholic acid (DCA) and CDCA is converted into lithocholic acid (LCA) which can be further metabolized by gut microbiota to ursodeoxychlic acid (UDCA) [¹⁵,¹⁶]. Both bile salts and unconjugated BAs are transported from the intestinal lumen into the enterocytes via the apical sodium-dependent transporter (ASBT) and passive diffusion, respectively. Upon entry into the enterocyte, BA species are coupled to the ileal BA carrier protein (IBABP) which transports them across the enterocyte [¹⁷] to the basolateral organic solute transporters a and b (OSTa/b) where they are then transported into the portal vein for recycling back to the liver [¹⁴,¹⁷,¹⁸]. From the portal vein, bile salts are taken up by the sodium-dependent taurocholate cotransporting polypeptide (NTCP) transporter and unconjugated BAs are taken up by the organic anion transporters-1,4 (OATP1,4) [¹⁸]. This cycling of BA species between the intestine and liver is termed, enterohepatic circulation [¹⁹].

BAs are thus the end products of cholesterol catabolism and provide a means for the elimination of excess cholesterol in the body. They also facilitate the absorption of nutrients and drugs. In addition, BAs play an important role in regulating energy metabolism [¹¹]. For example, hepatic gluconeogenesis, when excessive, can be a major cause of fasting hyperglycemia as well as increased postprandial hyperglycemia [²⁰,²¹]. BAs have been shown to regulate gluconeogenesis by repressing transcription of gluconeogenesis rate-limiting enzymes such as phosphoenolypyruvate carboxylase (PEPCK), glucose-6-phosphatase (G6Pase) and fructose-1,6-bis phosphatase (FBP1). The complete mechanism is detailed in the legend for Fig. 2 [²²]. When glucose enters the liver via the glucose transporter 2 (GLUT 2), it can be shunted either to glycolysis or to glycogenesis. The genes for the enzymes that promote these metabolic pathways are in turn, under the control of the transcription factor, liver X receptor (LXR) (see Fig. 2) [²³]. The two downstream targets of LXR, carbohydrate response binding protein (ChREBP) and sterol response element binding protein-1c (SREBP1c), as well as being responsible for transcription of genes for key glycolytic/glycogenesis enzymes, also act in a synergistic way to transcribe genes for enzymes that are important for the de novo lipogenesis (DNL) pathway [²³,²⁴]. BA activated FXR is an antagonist of LXR, causing transcription of peroxisome proliferator-activated receptor-a (PPARa) which, in turn, transcribes genes for substances that increase b-oxidation of fatty acids (FAs) [²⁵]. In this way, BAs are positive regulators of metabolism in the liver, via FXR. BA activated FXR leads to a decrease in the glycolysis → DNL → triglyceride (TG) axis, thus reducing hepatic lipid accumulation and also via increased transcription of PPARa, causing increased usage of lipids for energy via the b-oxidation pathway (Fig. 2).

In humans, an inverse relationship has been observed between metabolic disease and serum ceramides (Fig. 3) [²⁸,²⁹]. Intestinal FXR signaling has been found to be increased in distal ileum biopsies from obese individuals [³⁰]. In mice on a HFD and administered tempol, a gut microbiota modifying agent which reduces microbiota with high BSH activity, suggested that inhibition of FXR signaling was responsible for the resistance of these mice to obesity and insulin resistance as there were increased amounts of T-b-MCA observed (FXR antagonist) and decreased amounts of the unconjugated form, b-MCA (FXR agonist) [³¹]. Subsequently, a synthetic, intestine-specific small molecule, G-MCA, which cannot be hydrolyzed by gut microbial BSH, has been used to selectively inhibit intestinal FXR in mice and when exposed to HFD, these mice were protected from weight gain, insulin resistance, hyperglycemia and hepatic steatosis [³⁰]. The mechanism by which intestine selective inhibition of FXR signaling reduces diet-induced metabolic disease was shown to involve the modulation of ceramide synthesis as FXR targets genes in the ileum for the ceramide synthesis enzymes, sphingomyelin phosphodiesterase 3 (Smpd3) and serine palmitoyltransferase long-chain base subunit 2 (Sptlc2) (lower left, Fig. 3) [³⁰]. More direct evidence for the link between ceramide and metabolic disease came from studies of the WAT of obese human subjects which identified a positive correlation between ceramide synthase enzyme 6 (CerS6) with BMI and hyperglycemia and subsequent studies in CerS6 deficient mice that exhibited reduced C16:0 ceramides and were protected from diet-induced obesity and glucose intolerance [²⁹]. The effects of high serum ceramide (upper part of Fig. 3) include increased FA production in the liver via ceramide-induced activation of SREBP-1c in the nucleus and subsequent steatosis. Ceramide also activates JNK, IKK2, and PKCz (Fig. 3, upper left) and increases insulin resistance [²⁸].

In contrast, BA activation of intestinal FXR also leads to the production of the hormone fibroblast growth factor-19 (FGF19) (Fig. 3, upper right) that binds to the fibroblast growth factor receptor-4 (FGFR4) to inhibit BA synthesis, increase production of GLUT1 receptors allowing for increased glucose uptake into the liver ( Fig. 3, upper right) and increased synthesis of glycogen, thus increasing hepatic glucose storage [³⁰,³²]. These effects lower blood glucose and increase glycogenesis thus preventing usage of glucose in several other pathways that are responsible for metabolic memory which are described in detail in the legend for Fig. 1 and Table 1 [³³]. Briefly, increased O-GlcNAc of IRS-1,2 and AKT in the insulin signaling pathway diminishes their phosphorylation status and results in decreased insulin sensitivity and glucose uptake into the cell [³⁴]. O-GlcNAc also has an impact on the transcription factors that control gluconeogenesis and DNL. O-GlcNAc of FoxO1 causes upregulation of PEPCK and glucose-6-phosphate and thus increases gluconeogenesis [³⁵]. ChREBP can also be O-GlcNAc causing increased lipogenesis [¹⁰]. Therefore, increased O-GlcNAc due to hyperglycemia, causes a continuous, toxic cycle of increased glucose and lipids [⁸]. The final consequence of an enhanced polyol pathway activation are elevated levels of DHAP that can form methylglyoxal which, in turn, reacts with arginine and/or lysine to produce the advanced glycation end products, 5-hydro-5-methylimidazolone and N-carboxyethyl-lysine (AGEs)(Fig. 1) [⁹]. Higher plasma AGEs have been linked to increased cardiovascular events in T2D patients [⁹].

The contribution of fasting blood sugar levels (FBS) and post-prandial glucose levels (PPG) to the diurnal blood glucose concentration were studied in 290 T2D patients and the major findings were that the relative contribution of PPG was predominant in patients with glycemic control and a hemoglobin A1C (A1C)≤8.4% (68 mmol/mol) whereas, for patients with poor glycemic control (A1C>8.4% or 68 mmol/mol), FBS was the greater contributor to the diurnal hyperglycemia [³⁹]. Subsequently, a study was done that examined changes in postprandial BAs in T2D patients relative to non-diabetics (NGT) when subjected to 4 different meals, oral glucose tolerance test (OGTT), low, medium and high fat [⁴⁰]. BAs are released following a meal and are therefore in a position to regulate postprandial glucose homeostasis and overall glycemic control for T2D patients with A1C<8.4% (68 mmol/mol), via BA activation of FXR and Takeda G-protein coupled receptor 5 (TGR5) [¹¹,³⁹]. The results of this study were that postprandial serum concentrations of secondary BAs, cholic acid (CA) and total BAs were elevated in T2D vs. NGT patients that corresponded to lower serum FGF19 levels, a known inhibitor of BA synthesis. Total BAs were increased after the meal in a dose dependent way according to fat content with the highest fat content meal generating the highest level of BAs for both T2D and NGT. FGF19 levels corresponded with the total BA level for NGT but not T2D subjects which maintained a low, basal level of FGF19. Chenodeoxycholic acid (CDCA), the most potent FXR agonist was not significantly different between T2D and NGT but T2D patients had higher DCA (weaker FXR agonist), CA (not considered a strong FXR agonist) and higher UDCA which is considered an FXR antagonist [¹¹,⁴¹]. In conclusion, the investigators proposed a T2D-BA-FGF19 phenotype that results in altered FXR/FGF19 signaling in the intestine [⁴⁰]. Therefore, alteration of the BA pool in T2D may be a contributing factor for decreased glycemic control due to the decreased insulin-like action of FGF19 in the liver (Fig. 3).

The fact that for early T2D which has not progressed to an A1C of>8.4% (68 mmol/mol) it is the postprandial glucose that contributes more to the overall diurnal blood glucose concentrations and that BAs are peaked at postprandial times opens up a possibility for use of a particular BA or manipulation of the BA pool perhaps via adjustment of the gut microbiota as an option to achieve and maintain glycemic control and prevent metabolic memory for early onset T2D patients.

FGF21 results from BA activation of hepatic FXR and after secretion, FGF21 binds to the FGFR4 receptor in adipose causing GLUT1 upregulation [³⁷] and increased secretion of adiponectin [³⁶]. The FGF-19/21 pathway details are shown in Fig. 3.

Hepatic FGF21 has also been found to be crucial for PPARα agonist-induced improvement in energy and glucose metabolism in obese mice [⁴²]. BA activation of FXR causes increased expression of PPARα (Fig. 2). PPARα is highly expressed in liver, brown adipose tissue (BAT), heart and kidney. A recent study compared the effects of fenofibrate, a PPARα agonist on plasma FGF21 levels and improvements in glucose homeostasis in HFD-induced obesity in mice and compared results to lipodystophic A-Zip/F1 mice which lack WAT [⁴³]. The results showed that in DIO mice, plasma glucose and hyperinsulinemia were reduced under fenofibrate treatment relative to untreated controls. HFD feeding led to a reduction of FGF21 plasma levels that was greatly increased with fenofibrate treatment. When A-Zip/F1 mice, which exhibited severe glucose intolerance, hyperglycemia, hyperinsulinemia and insulin resistance, were treated with fenofibrate, there was improvement in hypertriglyceridemia but no improvement in hyperglycemia. Plasma levels of FGF21 were increased in the A-Zip/F1 mice as well as other genes targeted by PPARα but the lack of affect on hyperglycemia suggested that WAT was essential for the fenofibrate-induced reduction of glucose metabolism dysfunction [⁴²].

In the same study a comparison of FGF21 KO mice with WT controls was done to study the effects of fenofibrate treatment. Fenofibrate treatment lowered plasma triglycerides in both phenotypes, suggesting that this was not due to FGF21. The expression levels of PPARα target genes other than FGF21 were similar in both WT and FGF21 KO mice, suggesting that hepatic PPARα activation was not dependent on FGF21. In WAT, FGF21 expression was not induced by fenofibrate, suggesting that adipose tissue does not contribute to plasma FGF21 levels. In WT but not in FGF21 KO mice, fenofibrate induced genes related to brown adipocyte function such as Ucp1 and Pgc1α in WAT but not in BAT, suggesting that treatment-induced increases in FGF21 led to browning in white adipocytes (Fig. 3). HFD feeding of both WT and FGF21 KO mice, produced obesity in both phenotypes but treatment with fenofibrate reduced body weight and improved glucose tolerance/hyperinsulinemia only in the WT mice indicating that FGF21 was responsible for the induced increases in energy expenditure and anti-diabetic effects [⁴²]. Table 1 summarizes the metabolic effects of FGF19/21 hormones.

The effects of FGF15, FGF21 and an FGFR1/b-Klotho-activating antibody on the nervous system to regulate body weight and glycemia in mice were investigated very recently, employing genetic loss of function of the shared co-receptor, β-Klotho, in liver, adipose and the nervous system [¹⁹]. These tissue-specific β-Klotho knockout mice were generated in parallel with a floxed control. Initially, mice lacking β-Klotho in hepatocytes were generated (Klb^Alb mice) and tested for expression of the rate-limiting BA synthetic enzyme, CYP7A1, which was significantly elevated, thus confirming a previous study [⁴⁴] that showed that FGF15/β-Klotho-FGFR4 signaling was essential for physiologic regulation of BA homeostasis. Table 2 summarizes the important findings for this research article [¹⁹].

No significant correlations between FGF15/21 and ceramide concentrations were observed in this study although a modest decrease in adiponectin was seen in the control mice. Hepatic diacylglycerol (DAG) concentrations were significantly reduced by FGF15 in both control and Klb^adipo mice. Gene expression studies showed that FGFs acted directly on adipocytes by increasing expression of dual specificity phosphatase 4 (Dusp4) mRNA which is involved in the feedback regulation of ERK1/2, in both BAT and WAT in control mice only. The last analysis was in mice with loss of function for β-Klotho in the nervous system, Klb^Carnk2a mice. After 2 weeks of FGF15, only control mice showed weight loss and improvement in glycemic control. Sympathetic nerve activity was measured in BAT and was increased in a dose dependent manner with FGF15 administration indicating that FGF15 acts on the nervous system similar to the effects of FGF21 determined previously [⁴⁵]. Final confirmation for the longer term effects of FGF15/21 came from an experiment using an antibody, bFKB1, a selective activator of the FGFR1 receptor. Only the Klb^Camk2a mice were resistant to loss of body weight and improvement in glycemic control which mirrored the result of FGF15/21 in this mouse model. The final conclusion for this study was that longer term changes in weight and glycemic control require β-Klotho containing FGFR complexes in neurons and therefore, therapeutic strategies must consider access of the brain [¹⁹].

BA activated Takeda G-protein receptor 5 (TGR5) control of enteroendocrine cell signaling to regulate metabolism

It is well known that dietary fat is first broken down by pancreatic lipases that, in turn, degrade triglycerides into long chain fatty acids (LCFAs) that act as agonists for the Gαq-protein-coupled G-protein couples receptor 40 (GPR40) (shown in Fig. 4A). The remaining product of this lipase activity is 2-monoacylglycerol (2-MAG) which is an agonist for Gαs-coupled GPR119. All of these GPRs are expressed on enteroendocrine cells (EEC) that secrete glucagon-like peptide-1(GLP-1) [⁴⁶]. It is also well-known that ingestion of dietary fat causes BA secretion into the gut and that BAs are agonists for the Gαs-coupled BA responsive G-protein coupled receptor, TGR5, also causing secretion of glucagon-like protein-1 (GLP1) [⁴⁷]. Fig. 4 A shows the GPCR signaling pathway that leads to secretion of GLP-1 and other incretins such as GIP and PYY from EECs.

A recent study examined the relative importance of Gαs and Gαq signaling in GLP-1 secretion from EECs using primary crypt cultures, which allow cells to remain in a crypt organization rather than be plated out using single cell suspensions [⁴⁸]. The underlying hypothesis was that both Gαs and Gαq-coupled pathways were important for the stimulation of GPR40 to secrete GLP-1 and that the effect was synergistic and not simply additive. The study was prompted by the fact that GPR119 agonists (Gαs-coupled), although promising in animal models never achieved clinical improvement in glucose tolerance in phase II trials. GPR40 (Gαq-coupled) agonists also showed only modest effects on GLP-1 and insulin secretion. However, a second generation GPR40 agonist, AM-1638, produced a more robust response in terms of GLP-1 secretion. Selective pharmacological blockage of the Gαs signaling pathway using cholera toxin (CTX) and also of the Gαq signaling pathway using UBO-QIC and GPR40 receptor selective agonist, AM-1638, at low concentrations resulted in a decreased response in GLP-1 secretion with CTX (0.9-fold) and an even greater decrease (0.4–fold) with UBO-QIC. Addition of both CTX and UBO-QIC abolished GLP-1 secretion in response to AM-1638 stimulation completely. Therefore, the combined Gs+ Gq agonist, Am-1638, produced a more robust response and therefore, GPR40 was found to signal both by Gαs and Gαq pathways (Fig. 4A) explaining the lack of efficacy of the first generation GPR40 agonists, TAK-875 and MK-2305 that affect only the Gαq activation of the receptor.

Next, synergistic effects of GPAR40, GPR119 and TGR5 were probed for by administration of the Gαq-only GPR40 selective agonist, MK-2305 in combination with either the GPR119 selective AR246881 or the TGR5 selective Merck V agonists at low concentrations. When given alone, each had a small GLP-1 secretion response but when MK-2305 (GPR40) and Merck V (TGR5) were administered together, a synergistic effect was observed. A similar trend was observed for GPR119 but did not reach statistical significance. These results were confirmed in mice that were fasted and administered MK-2305, AR246881 and Merck V agonists by oral gavage. Therefore Gαq-coupled GPR40 acts in synergy with the Gαs-coupled TGR5 and GPR119 receptors with respect to GLP-1 secretion. The results support the notion that LCFAs stimulating GPR40 under physiological conditions act in synergy with 2-MAGs acting on GPR119 and BAs on TGR5 to jointly stimulate GLP-1 secretion. The high efficacy of the second generation GPR40 agonist AM-1638 and its likely mechanism of stimulating GLP-1 secretion not only through the Gαq pathway but also via Gαs, thus increasing cAMP in addition to IP3 and Ca²⁺, were also demonstrated (Fig. 4A). The enhanced GLP-1 secretion observed in this study via a combination of agonists that bind GPR40 and the BA sensitive TGR5 is therefore a step forward in glycemic control of T2D.

BAs also exert an effect on glucose-stimulated insulin secretion (GSIS) in pancreatic β-cells which is diagrammed in Fig. 4B. In particular, the BA, tauroursodeoxycholic acid (TUDCA), when incubated with mouse pancreatic islets in 11 mmol/L or higher glucose concentrations, enhanced GSIS [⁴⁹]. No differences were seen in total insulin between test and control cells and thus changes in insulin release were not related to changes in insulin synthesis. No differences were observed in glucose-induced NAD(P)H fluorescence levels between TUDCA treated and untreated cells, indicating that mitochondrial metabolism was unaffected by the BA. Diazoxide, a potent K_ATP channel opener, caused a decrease in GSIS at 11 mmol/L glucose, but in the presence of TUDCA, insulin secretion increased even in the presence of diazoxide, indicating that TUDCA action was via another mechanism than the effects of K_ATP closure. To further confirm this, the effect of TUDCA on glucose-induced Ca²⁺ signals were measured, and no differences were observed between BA treated and untreated cells at 11 mmol/L glucose stimulation. Therefore, the effect of TUDCA was not due to K_ATP channel-dependent mechanism or Ca²⁺ signals.

Next, the BA receptor TGR5 became a point of focus as TUDCA has poor affinity for FXR. TGR5 relies on the Gαs stimulatory subunit which leads to activation of adenyl cyclase → cAMP→ PKA (Fig. 4A). The Gα inhibitor, NF449, did not change GSIS at 11.1 or 22.2 mmol/L of glucose stimulation but did abolish TUDCA effects on GSIS. The GSIS stimulation was mimicked by the TGR5 selective agonist INT-777 at 11 mmol/L glucose concentration whereas it had no effect at lower glucose concentrations. Incubation with the FXR antagonist TβMCA did not alter the effect of TUDCA on GSIS, indicating that FXR was not involved. Therefore, TUDCA was proposed to increase GSIS at higher but not basal glucose concentrations via the TGR5 receptor. Lastly, to determine whether TUDCA modulated the cAMP/PKA pathway, the effects of the PKA inhibitor H89 and (Rp)-cAMP, a competitive inhibitor of PKA activation by cAMP, were examined. In both cases, the effect of TUDCA on GSIS was inhibited. TUDCA was also found to increase phosphorylation of PKA (increase its activation) and CREB phosphorylation (target protein of PKA). These effects lead to less loss of pancreatic β-cells along with increased protein translation and decreased endoplasmic reticulum (ER) stress. Therefore, a TUDCA →TGR5 → adenyl cyclase → cAMP → PKA → CREB → GSIS pathway was proposed as shown in Fig. 4B.

Thus, because TUDCA administration causes increased insulin secretion only under hyperglycemic conditions, a possibility for early therapeutic glycemic control using BA-induced secretion of endogenous insulin may be feasible for early onset T2D with the view of preventing metabolic memory.

Metabolic effect of altered enterohepatic circulation of BAs

BA sequestrants, cholestyramine, colevelam, colestimide and cholestipol, are resins that are not absorbed in the intestine and are able to bind negatively charged bile salts within the intestinal lumen. The sequestration process has the effect of removing those bile salts from the enterohepatic circulation and thus causes increased return of low-density lipoprotein cholesterol (LDL-C) to the liver and increased hepatic BA synthesis [⁵³,⁵⁴]. BA sequestration has also been shown to increase GLP1 secretion and improve glycemia in HFD-induced obese mice but not in TGR5^−/− mice, indicating that BAs activated TGR5 [⁵⁵]. Notably, a higher number of L-type enteroendocrine cells are found in more distal parts of the gut and BA sequestration allows BA transportation to these regions. Treatment of mice with colestilan was shown to cause increased secretion of GLP1 from the colon relative to duodenum or ileum [⁵⁶]. A meta-analysis of 15 different human trials, involving patients with T2D, of the effects of BA sequestrants revealed that high-density lipoprotein cholesterol (HDL-C), LDL-C, hemoglobin A1C (HbA1C), TGs and glycemia were all improved with this treatment [⁵⁷]. Another meta-analysis of 17 trials involving 2950 T2D patients that focused only on glycemic control showed significant improvement of A1C and FBS with the use of BA sequestration [⁵⁸]. The conclusion derived from both of these meta-analyses was that BA sequestrants were beneficial in glycemic control of T2D patients.

An alternative way to affect the enterohepatic circulation that has been proposed as a means to improve glycemic control in T2D patients is the inhibition of the apical sodium BA transporter (ASBT). A recent study of the ASBT inhibitor 264W94 was done on Zucker diabetic rats. Oral administration of the inhibitor for 2 weeks increased fecal BAs and elevated plasma GLP1 levels. There was significant improvement seen in A1C levels and glycemia [⁵⁹]. ASBT^−/− mice were found to display a similar metabolic phenotype to that described for BA sequestrants and ASBT inhibitors. For example, they have increased expression of hepatic CYP7A1 resulting in increased BA synthesis causing increased flux of BAs into the colon with potential for increased activation of TGR5 and increased GLP1 secretion. There was also suppression of hepatic SREBP1c and an improvement in TG metabolism [⁶⁰]. In summary, the adjustment of enterohepaic circulation may be another viable means for achieving early glycemic control for T2D subjects and avoidance of metabolic memory.

The metabolic effect of bariatric surgery on BAs and T2D

It has been hypothesized that postprandial variations of bile salts in the gut lumen and circulation could be important for the rapid establishment of glycemic control following bariatric surgery that cannot be accounted for by weight reduction. In a recent study, vertical sleeve gastrectomy (VSG) or sham surgery was performed on HFD Tgr5^+/+ and Tgr5^−/− mice [⁶¹]. VSG was found to decrease food intake and body weight, increase energy expenditure and circulating bile concentrations, improve fasting glucose, oral glucose tolerance (GTT), GSIS, increased GLP1 secretion and favorable shifts in gut microbiota for both mouse genotypes relative to sham WT controls. However, body weight independent improvements in fasting glucose, GTT, hepatic insulin signaling and islet morphology after VSG were attenuated in Tgr5^−/− relative to Tgr5^+/+ mice. The favorable BA alterations after VSG were also blunted in Tgr5^−/−mice. TGR5 signaling was found to contribute to long-term maintenance of body weight independent improvements in glucose tolerance and did not contribute to post-operative increases in GSIS after VSG. Reductions in fasting plasma insulin, leptin and cholesterol after VSG was partially attributed to TGR5 signaling as this was not found in Tgr5^−/− mice. VSG reduced hepatic monocyte chemotacticprotein-1 (MCP-1) and TNFα expression relative to sham-WM in Tgr5^+/+ but not Tgr5^−/−. MCP-1 and TNFα were elevated in VSG KO compared to VSG WT and thus it was concluded that TGR5 may contribute to VSG-induced improvements in GTT by promoting improved hepatic insulin signaling through decreases in inflammation [⁶¹].

At 4 months post surgery, fasting total BA concentrations were elevated in VSG-WT and VSG-KO relative to control groups and this was primarily due to an increase in conjugated BAs. An increase in the 12α-hydroxylated BA/non-12α-hydroxylated BA ratio has been associated with increased insulin resistance in humans [⁶²]. This ratio was 2-fold higher in VSG KO compared to VSG-WT and plasma DCA and TDCA were elevated 4-fold in VSG KO relative to VSG-WT. VSG resulted in a decreased BA hydrophobicity index for both WT and KO mice but this effect was blunted in KO mice. There were increased amounts of hydrophilic muricholates in VSG-WT compared to VSG-KO suggesting that TGR5 signaling contributes to this shift in BA profile.

Examination of hepatic protein expression was measured and whereas there were no differences between the mouse groups for SHP, Cyp7a1 and Cyp27a1, there was a significant decrease in the expression of Cyp8b1 in the VSG-WT group. Cyp8b1 is a critical enzyme for CA biosynthesis and therefore, TGR5-dependent reduction of Cyp8b1 in VSG-WT may be responsible for the increase in the 12α-OH:non-12α-OH ratio.

Assessment of the microbial populations at the phyla level revealed that the relative abundance of cyanobacteria was 10-fold higher in VSG-WT compared to sham controls. The relative abundance of cyanobacteria in VSG-KO remained unchanged relative to controls, suggesting that TGR5 contributes to this postoperative change in microbiota [⁶¹].

BAs and circadian rhythm dysregulation in diabetes

Feeding restriction, nutrient intake [⁶³] and the central circadian clock all control BA production, expression of metabolic hormones such as glucagon and insulin [⁶⁴] as well as lipid and glucose homeostasis [⁶⁵,⁶⁶]. Misalignment of the circadian clock has been associated with metabolic syndrome in humans [⁶⁷].

Molecular clocks are present in most cell types and they consist of autoregulated feedback loops of rhythmically expressed genes that oscillate within a 24 h period. A detailed description of hepatic circadian clock signaling is given in the legend of Fig. 5. From Fig. 5 it can be seen that the circadian rhythm of HMGCR via SREBP2 engenders corresponding changes in oxysterol production, which are known to regulate LXR and increase transcription of Cyp7a1. Thus, circadian control of cholesterol synthesis may be linked to the regulation of Cyp7a1 transcription [⁶⁸].

The circadian rhythm of BAs has been extensively studied in mice [¹⁵]. Primary, conjugated BAs exhibited two peaks in the dark phase corresponding to presumably two episodes of food ingestion. Unconjugated BAS peaked in the light phase, suggesting that active deconjugation of BAs by the gut microbiota occurred when mice were fasting [¹⁵]. Notably, BAs which are controlled by circadian rhythms also modulate activation of LXR via FXR to cause a decrease in transcription of Cyp7a1 (Fig. 2). RORα, in contrast, is a positive regulator of sterol-12α-hydroxylase (Cyp8b1) for CA synthesis [⁶⁹]. Both PPARα and the co-activator, peroxisome proliferators-activated receptor gamma co-activator 1-α (PGC1α) modulate Bmal1 transcription [⁶⁴,⁷⁰]. Pparα is also a target gene of BA modulated FXR (Fig. 2).

Sleep disruption is a well-documented cause of circadian clock misalignment [⁶⁶,⁷⁰]. In a recent study, B6xC57 mice were maintained on either normal chow or HFD and then sleep disrupted for 6 h/day for 5 days [⁷²]. Measurements were made of expression of genes for hepatic metabolism, core clock, and regulators of BAs. WAT weight was unchanged with sleep disruption in chow-fed mice and was significantly increased in HFD mice and further increased in sleep-disrupted HFD mice. Core clock genes, Clock, Bmal1 and Per1, 2 were suppressed at all times but did not change in phase. REV-ERBα was suppressed by sleep disruption. Expression of Cyp7a1 and an important transcriptional regulator of Cyp7a1, hepatocyte nuclear receptor 4α(Hnf4α) peaked at the same time but amplitudes were significantly reduced by sleep disruption. Arhythmic expression of Srebp1c was also observed with sleep disruption. Several enzymes involved in lipid metabolism were suppressed due to sleep disruption. These included HMGCR, FAS, and ACC. FFAs were elevated in both the liver and serum while TGs were decreased in the serum. Therefore, sleep disruption caused changes in lipid homeostasis. Liver BAs were decreased consistent with the depressed expression of Cyp7a1. However, intestinal BAs pool size was increased and the total BA pool was also increased. Therefore, sleep disruption altered BA homeostasis and enterohepatic circulation. Overall, in this study, even short-term circadian disruption was enough to impair BA and lipid homeostasis in mice.

Circadian rhythms are thought to affect the efficacy and/or toxicity of drugs [⁷³,⁷⁴] and a recent study identified soluble carrier protein family 22 member 4 (Slc22a4) which encodes the organic cation transporter novel type-1 (Octn1) as a PPARα-regulated gene and that its intestinal expression was regulated by BA circadian oscillation (middle of Fig. 5) [⁷¹]. BAs have been shown to repress the expression of PPARα target genes by interfering with the recruitment of the transcription factor coactivator, cAMP response element binding protein (CBP) and its analog p300 to PPARα/RXRα complexes in mice fed high CA diets [⁷⁵]. In the study described here, oligonucleotide microarray analysis using RNA isolates from CA treated intestinal cells from WT or PPARα null mice revealed 58 PPARα target genes that were repressed by the CA treatment and among these, the gene Slc22a4 for the Octn1 transporter was identified as being repressed by CA. This repressive effect of CA on the PPARα/RXRα-induced expression of Slc22a4 was verified by CA treatment of the intestinal cell line, Caco-2. A significant circadian oscillation of BAs were observed in both WT and PPARα^−/− mice with the greatest accumulation during the dark/feeding phase and an oscillation of mRNA for Slc22a4 was observed with a peak in the light phase and then decreased after the start of the dark phase, nearly anti-phase to BA accumulation. The expression of PPARα protein was detected at all time points with no obvious circadian oscillation. However, the proteins levels of Octn1 peaked before the start of the dark phase and exhibited circadian oscillation. PPARα mice had a similar BA circadian rhythm but the amplitude of the rhythms for Slc22a4 and Octn1 was significantly decreased, suggesting that PPARα acted as a positive regulator of Slc22a4 in intestinal epithelial cells and that PPARα-mediated expression of Slc22a4 is repressed in a BA dependent way. The next experiment was to test whether the circadian oscillation of Octn1 affected the pharmacokinetics of the anti-neuropathic pain medication often prescribed for diabetics, gabapentin. Incubation of WT jejunal segments with 5 µmol/L radiolabeled gabapentin showed time-dependent variations in drug uptake into the issue. Further verification of this was done using 39 serum concentrations of orally administered gabapentin from 39 different mice. Thus, in mice, increased accumulation of BAs due to feeding of CA influenced the uptake of drugs used to treat diabetic patients [⁷¹]. T2D patients have been reported to have increased amount of 12-α-hydroxylated BAs such as CA and its derivatives [⁴⁰].

Conclusions

In this review, we have discussed the role of BAs in glucose and lipid metabolism and ways in which they can be manipulated to achieve early glycemic control for T2D patients. Early glycemic control can be effectively initiated through the use of insulin treatment for T1D, but this approach has not been recommended for T2D. The importance of early glycemic control is due to the potential development of “metabolic memory,” a term used to describe the upregulation of glucose utilizing pathways that result in (1) the accumulation of AGEs which are linked to cardiovascular complications such as retinal disorders, end-stage renal disease, and accelerated atherosclerosis, (2) increased activity in the polyol pathway which produces increased amount of NADH/ROS in the mitochondria to cause dysfunction and also increased DHAP which leads to more insulin resistance, (3) increased amounts of fructose (also a product of the polyol pathway) that can enter the hexosamine pathway and produce excess O-GlcNAc products, including those which affect insulin signaling and lipogenesis. The changes associated with metabolic memory are not always reversible and can only be avoided by aggressive early glycemic control [²]. It is therefore important to seek out ways to achieve rapid control of hyperglycemia perhaps by utilizing some of the approaches discussed in this review which involve the use and/or manipulation of BA pool size, composition, transport, hormone signaling capabilities, bariatric surgery, and circadian rhythms.

While the use of animal models may show us a potential solution, there is also still a need for clinical trials to assess the true effectiveness of some of the approaches discussed in this review. In addition, the growing and evolving field of metabolomics will perhaps need to expand its repertoire to include quantification of the metabolites resulting from metabolic memory pathways such as glycosylated proteins and AGEs.

References

[1] Abramowicz M, Zuccotti G. Drugs for diabetes. Treat Guidel Med Lett 2005; 3(36): 57–62

[2] Ceriello A. The emerging challenge in diabetes: the “metabolic memory”. Vascul Pharmacol 2012; 57(5-6): 133–138

[3] Ceriello A, Esposito K, Piconi L, Ihnat MA, Thorpe JE, Testa R, Boemi M, Giugliano D. Oscillating glucose is more deleterious to endothelial function and oxidative stress than mean glucose in normal and type 2 diabetic patients. Diabetes 2008; 57(5): 1349–1354

[4] Ceriello A, Esposito K, Ihnat M, Thorpe J, Giugliano D. Long-term glycemic control influences the long-lasting effect of hyperglycemia on endothelial function in type 1 diabetes. J Clin Endocrinol Metab 2009; 94(8): 2751–2756

[5] Nathan DM, Cleary PA, Backlund JY, Genuth SM, Lachin JM, Orchard TJ, Raskin P, Zinman B; Diabetes Control and Complications Trial/Epidemiology of Diabetes Interventions and Complications (DCCT/EDIC) Study Research Group. Intensive diabetes treatment and cardiovascular disease in patients with type 1 diabetes. N Engl J Med 2005; 353(25): 2643–2653

[6] Brownlee M. Biochemistry and molecular cell biology of diabetic complications. Nature 2001; 414(6865): 813–820

[7] Foufelle F, Ferré P. New perspectives in the regulation of hepatic glycolytic and lipogenic genes by insulin and glucose: a role for the transcription factor sterol regulatory element binding protein-1c. Biochem J 2002; 366(Pt 2): 377–391

[8] Banerjee PS, Lagerlöf O, Hart GW. Roles of O-GlcNAc in chronic diseases of aging. Mol Aspects Med 2016; 51: 1–15

[9] Hanssen NM, Beulens JW, van Dieren S, Scheijen JL, van der A DL, Spijkerman AM, van der Schouw YT, Stehouwer CD, Schalkwijk CG. Plasma advanced glycation end products are associated with incident cardiovascular events in individuals with type 2 diabetes: a case-cohort study with a median follow-up of 10 years (EPIC-NL). Diabetes 2015; 64(1): 257–265

[10] Guinez C, Filhoulaud G, Rayah-Benhamed F, Marmier S, Dubuquoy C, Dentin R, Moldes M, Burnol AF, Yang X, Lefebvre T, Girard J, Postic C. O-GlcNAcylation increases ChREBP protein content and transcriptional activity in the liver. Diabetes 2011; 60(5): 1399–1413

[11] Li T, Chiang JY. Bile acid signaling in metabolic disease and drug therapy. Pharmacol Rev 2014; 66(4): 948–983

[12] Kong B, Wang L, Chiang JY, Zhang Y, Klaassen CD, Guo GL. Mechanism of tissue-specific farnesoid X receptor in suppressing the expression of genes in bile-acid synthesis in mice. Hepatology 2012; 56(3): 1034–1043

[13] Thomas C, Pellicciari R, Pruzanski M, Auwerx J, Schoonjans K. Targeting bile-acid signalling for metabolic diseases. Nat Rev Drug Discov 2008; 7(8): 678–693

[14] Trauner M, Boyer JL. Bile salt transporters: molecular characterization, function, and regulation. Physiol Rev 2003; 83(2): 633–671

[15] Zhang YK, Guo GL, Klaassen CD. Diurnal variations of mouse plasma and hepatic bile acid concentrations as well as expression of biosynthetic enzymes and transporters. PLoS One 2011; 6(2): e16683

[16] Ridlon JM, Kang DJ, Hylemon PB. Bile salt biotransformations by human intestinal bacteria. J Lipid Res 2006; 47(2): 241–259

[17] Dawson PA, Lan T, Rao A. Bile acid transporters. J Lipid Res 2009; 50(12): 2340–2357

[18] Meier PJ, Stieger B. Bile salt transporters. Annu Rev Physiol 2002; 64(1): 635–661

[19] Lan T, Morgan DA, Rahmouni K, Sonoda J, Fu X, Burgess SC, Holland WL, Kliewer SA, Mangelsdorf DJ. FGF19, FGF21, and an FGFR1/β-Klotho-activating antibody act on the nervous system to regulate body weight and glycemia. Cell Metab 2017; 26(5): 709–718.e3

[20] Rizza RA. Pathogenesis of fasting and postprandial hyperglycemia in type 2 diabetes: implications for therapy. Diabetes 2010; 59(11): 2697–2707

[21] Chung ST, Hsia DS, Chacko SK, Rodriguez LM, Haymond MW. Increased gluconeogenesis in youth with newly diagnosed type 2 diabetes. Diabetologia 2015; 58(3): 596–603

[22] Yamagata K, Daitoku H, Shimamoto Y, Matsuzaki H, Hirota K, Ishida J, Fukamizu A. Bile acids regulate gluconeogenic gene expression via small heterodimer partner-mediated repression of hepatocyte nuclear factor 4 and Foxo1. J Biol Chem 2004; 279(22): 23158–23165

[23] Bechmann LP, Hannivoort RA, Gerken G, Hotamisligil GS, Trauner M, Canbay A. The interaction of hepatic lipid and glucose metabolism in liver diseases. J Hepatol 2012; 56(4): 952–964

[24] Dentin R, Girard J, Postic C. Carbohydrate responsive element binding protein (ChREBP) and sterol regulatory element binding protein-1c (SREBP-1c): two key regulators of glucose metabolism and lipid synthesis in liver. Biochimie 2005; 87(1): 81–86

[25] Ip E, Farrell GC, Robertson G, Hall P, Kirsch R, Leclercq I. Central role of PPARα-dependent hepatic lipid turnover in dietary steatohepatitis in mice. Hepatology 2003; 38(1): 123–132

[26] Denechaud PD, Dentin R, Girard J, Postic C. Role of ChREBP in hepatic steatosis and insulin resistance. FEBS Lett 2008; 582(1): 68–73

[27] Zhang Y, Lee FY, Barrera G, Lee H, Vales C, Gonzalez FJ, Willson TM, Edwards PA. Activation of the nuclear receptor FXR improves hyperglycemia and hyperlipidemia in diabetic mice. Proc Natl Acad Sci USA 2006; 103(4): 1006–1011

[28] Gonzalez FJ, Jiang C, Patterson AD. An Intestinal microbiota-farnesoid X receptor axis modulates metabolic disease. Gastroenterology 2016; 151(5): 845–859

[29] Turpin SM, Nicholls HT, Willmes DM, Mourier A, Brodesser S, Wunderlich CM, Mauer J, Xu E, Hammerschmidt P, Brönneke HS, Trifunovic A, LoSasso G, Wunderlich FT, Kornfeld JW, Blüher M, Krönke M, Brüning JC. Obesity-induced CerS6-dependent C16:0 ceramide production promotes weight gain and glucose intolerance. Cell Metab 2014; 20(4): 678–686

[30] Jiang C, Xie C, Lv Y, Li J, Krausz KW, Shi J, Brocker CN, Desai D, Amin SG, Bisson WH, Liu Y, Gavrilova O, Patterson AD, Gonzalez FJ. Intestine-selective farnesoid X receptor inhibition improves obesity-related metabolic dysfunction. Nat Commun 2015; 6(1): 10166

[31] Li F, Jiang C, Krausz KW, Li Y, Albert I, Hao H, Fabre KM, Mitchell JB, Patterson AD, Gonzalez FJ. Microbiome remodelling leads to inhibition of intestinal farnesoid X receptor signalling and decreased obesity. Nat Commun 2013; 4(1): 2384

[32] Kir S, Beddow SA, Samuel VT, Miller P, Previs SF, Suino-Powell K, Xu HE, Shulman GI, Kliewer SA, Mangelsdorf DJ. FGF19 as a postprandial, insulin-independent activator of hepatic protein and glycogen synthesis. Science 2011; 331(6024): 1621–1624

[33] Stefano GB, Challenger S, Kream RM. Hyperglycemia-associated alterations in cellular signaling and dysregulated mitochondrial bioenergetics in human metabolic disorders. Eur J Nutr 2016; 55(8): 2339–2345

[34] Whelan SA, Dias WB, Thiruneelakantapillai L, Lane MD, Hart GW. Regulation of insulin receptor substrate 1 (IRS-1)/AKT kinase-mediated insulin signaling by O-linked β-N-acetylglucosamine in 3T3-L1 adipocytes. J Biol Chem 2010; 285(8): 5204–5211

[35] Housley MP, Rodgers JT, Udeshi ND, Kelly TJ, Shabanowitz J, Hunt DF, Puigserver P, Hart GW. O-GlcNAc regulates FoxO activation in response to glucose. J Biol Chem 2008; 283(24): 16283–16292

[36] Holland WL, Adams AC, Brozinick JT, Bui HH, Miyauchi Y, Kusminski CM, Bauer SM, Wade M, Singhal E, Cheng CC, Volk K, Kuo MS, Gordillo R, Kharitonenkov A, Scherer PE. An FGF21-adiponectin-ceramide axis controls energy expenditure and insulin action in mice. Cell Metab 2013; 17(5): 790–797

[37] Ge X, Chen C, Hui X, Wang Y, Lam KS, Xu A. Fibroblast growth factor 21 induces glucose transporter-1 expression through activation of the serum response factor/Ets-like protein-1 in adipocytes. J Biol Chem 2011; 286(40): 34533–34541

[38] Copple BL, Li T. Pharmacology of bile acid receptors: evolution of bile acids from simple detergents to complex signaling molecules. Pharmacol Res 2016; 104: 9–21

[39] Monnier L, Lapinski H, Colette C. Contributions of fasting and postprandial plasma glucose increments to the overall diurnal hyperglycemia of type 2 diabetic patients: variations with increasing levels of HbA(1c). Diabetes Care 2003; 26(3): 881–885

[40] Sonne DP, van Nierop FS, Kulik W, Soeters MR, Vilsbøll T, Knop FK. Postprandial plasma concentrations of individual bile acids and FGF-19 in patients with type 2 diabetes. J Clin Endocrinol Metab 2016; 101(8): 3002–3009

[41] Mueller M, Thorell A, Claudel T, Jha P, Koefeler H, Lackner C, Hoesel B, Fauler G, Stojakovic T, Einarsson C, Marschall HU, Trauner M. Ursodeoxycholic acid exerts farnesoid X receptor-antagonistic effects on bile acid and lipid metabolism in morbid obesity. J Hepatol 2015; 62(6): 1398–1404

[42] Goto T, Hirata M, Aoki Y, Iwase M, Takahashi H, Kim M, Li Y, Jheng HF, Nomura W, Takahashi N, Kim CS, Yu R, Seno S, Matsuda H, Aizawa-Abe M, Ebihara K, Itoh N, Kawada T. The hepatokine FGF21 is crucial for peroxisome proliferator-activated receptor-α agonist-induced amelioration of metabolic disorders in obese mice. J Biol Chem 2017; 292(22): 9175–9190

[43] Reitman ML, Gavrilova O. A-ZIP/F-1 mice lacking white fat: a model for understanding lipoatrophic diabetes. Int J Obes Relat Metab Disord 2000; 24 (Suppl 4): S11–14

[44] Katafuchi T, Esterházy D, Lemoff A, Ding X, Sondhi V, Kliewer SA, Mirzaei H, Mangelsdorf DJ. Detection of FGF15 in plasma by stable isotope standards and capture by anti-peptide antibodies and targeted mass spectrometry. Cell Metab 2015; 21(6): 898–904

[45] Owen BM, Ding X, Morgan DA, Coate KC, Bookout AL, Rahmouni K, Kliewer SA, Mangelsdorf DJ. FGF21 acts centrally to induce sympathetic nerve activity, energy expenditure, and weight loss. Cell Metab 2014; 20(4): 670–677

[46] Psichas A, Reimann F, Gribble FM. Gut chemosensing mechanisms. J Clin Invest 2015; 125(3): 908–917

[47] van Nierop FS, Scheltema MJ, Eggink HM, Pols TW, Sonne DP, Knop FK, Soeters MR. Clinical relevance of the bile acid receptor TGR5 in metabolism. Lancet Diabetes Endocrinol 2017; 5(3): 224–233

[48] Hauge M, Ekberg JP, Engelstoft MS, Timshel P, Madsen AN, Schwartz TW. Gq and Gs signaling acting in synergy to control GLP-1 secretion. Mol Cell Endocrinol 2017; 449: 64–73

[49] Vettorazzi JF, Ribeiro RA, Borck PC, Branco RC, Soriano S, Merino B, Boschero AC, Nadal A, Quesada I, Carneiro EM. The bile acid TUDCA increases glucose-induced insulin secretion via the cAMP/PKA pathway in pancreatic β cells. Metabolism 2016; 65(3): 54–63

[50] Thomas C, Gioiello A, Noriega L, Strehle A, Oury J, Rizzo G, Macchiarulo A, Yamamoto H, Mataki C, Pruzanski M, Pellicciari R, Auwerx J, Schoonjans K. TGR5-mediated bile acid sensing controls glucose homeostasis. Cell Metab 2009; 10(3): 167–177

[51] Miyamoto J, Hasegawa S, Kasubuchi M, Ichimura A, Nakajima A, Kimura I. Nutritional signaling via free fatty acid receptors. Int J Mol Sci 2016; 17(4): 450

[52] Meloni AR, DeYoung MB, Lowe C, Parkes DG. GLP-1 receptor activated insulin secretion from pancreatic β-cells: mechanism and glucose dependence. Diabetes Obes Metab 2013; 15(1): 15–27

[53] Insull W Jr. Clinical utility of bile acid sequestrants in the treatment of dyslipidemia: a scientific review. South Med J 2006; 99(3): 257–273

[54] Sonne DP, Hansen M, Knop FK. Bile acid sequestrants in type 2 diabetes: potential effects on GLP1 secretion. Eur J Endocrinol 2014; 171(2): R47–R65

[55] Potthoff MJ, Potts A, He T, Duarte JA, Taussig R, Mangelsdorf DJ, Kliewer SA, Burgess SC. Colesevelam suppresses hepatic glycogenolysis by TGR5-mediated induction of GLP-1 action in DIO mice. Am J Physiol Gastrointest Liver Physiol 2013; 304(4): G371–G380

[56] Harach T, Pols TW, Nomura M, Maida A, Watanabe M, Auwerx J, Schoonjans K. TGR5 potentiates GLP-1 secretion in response to anionic exchange resins. Sci Rep 2012; 2(1): 430

[57] Mazidi M, Rezaie P, Karimi E, Kengne AP. The effects of bile acid sequestrants on lipid profile and blood glucose concentrations: a systematic review and meta-analysis of randomized controlled trials. Int J Cardiol 2017; 227: 850–857

[58] Hansen M, Sonne DP, Mikkelsen KH, Gluud LL, Vilsbøll T, Knop FK. Bile acid sequestrants for glycemic control in patients with type 2 diabetes: a systematic review with meta-analysis of randomized controlled trials. J Diabetes Complications 2017; 31(5): 918–927

[59] Chen L, Yao X, Young A, McNulty J, Anderson D, Liu Y, Nystrom C, Croom D, Ross S, Collins J, Rajpal D, Hamlet K, Smith C, Gedulin B. Inhibition of apical sodium-dependent bile acid transporter as a novel treatment for diabetes. Am J Physiol Endocrinol Metab 2012; 302(1): E68–E76

[60] Ferrebee CB, Dawson PA. Metabolic effects of intestinal absorption and enterohepatic cycling of bile acids. Acta Pharm Sin B 2015; 5(2): 129–134

[61] McGavigan AK, Garibay D, Henseler ZM, Chen J, Bettaieb A, Haj FG, Ley RE, Chouinard ML, Cummings BP. TGR5 contributes to glucoregulatory improvements after vertical sleeve gastrectomy in mice. Gut 2017; 66(2): 226–234

[62] Haeusler RA, Astiarraga B, Camastra S, Accili D, Ferrannini E. Human insulin resistance is associated with increased plasma levels of 12α-hydroxylated bile acids. Diabetes 2013; 62(12): 4184–4191

[63] Manoogian ENC, Panda S. Circadian rhythms, time-restricted feeding, and healthy aging. Ageing Res Rev 2017; 39: 59–67

[64] Froy O. Circadian aspects of energy metabolism and aging. Ageing Res Rev 2013; 12(4): 931–940

[65] Adamovich Y, Rousso-Noori L, Zwighaft Z, Neufeld-Cohen A, Golik M, Kraut-Cohen J, Wang M, Han X, Asher G. Circadian clocks and feeding time regulate the oscillations and levels of hepatic triglycerides. Cell Metab 2014; 19(2): 319–330

[66] Kalsbeek A, la Fleur S, Fliers E. Circadian control of glucose metabolism. Mol Metab 2014; 3(4): 372–383

[67] Ribas-Latre A, Eckel-Mahan K. Interdependence of nutrient metabolism and the circadian clock system: importance for metabolic health. Mol Metab 2016; 5(3): 133–152

[68] Le Martelot G, Claudel T, Gatfield D, Schaad O, Kornmann B, Lo Sasso G, Moschetta A, Schibler U. REV-ERBα participates in circadian SREBP signaling and bile acid homeostasis. PLoS Biol 2009; 7(9): e1000181

[69] Pathak P, Li T, Chiang JY. Retinoic acid-related orphan receptor α regulates diurnal rhythm and fasting induction of sterol 12α-hydroxylase in bile acid synthesis. J Biol Chem 2013; 288(52): 37154–37165

[70] Bass J, Takahashi JS. Circadian integration of metabolism and energetics. Science 2010; 330(6009): 1349–1354

[71] Wada E, Koyanagi S, Kusunose N, Akamine T, Masui H, Hashimoto H, Matsunaga N, Ohdo S. Modulation of peroxisome proliferator-activated receptor-α activity by bile acids causes circadian changes in the intestinal expression of Octn1/Slc22a4 in mice. Mol Pharmacol 2015; 87(2): 314–322

[72] Ferrell JM, Chiang JY. Short-term circadian disruption impairs bile acid and lipid homeostasis in mice. Cell Mol Gastroenterol Hepatol 2015; 1(6):664–677

[73] Lis CG, Grutsch JF, Wood P, You M, Rich I, Hrushesky WJ. Circadian timing in cancer treatment: the biological foundation for an integrative approach. Integr Cancer Ther 2003; 2(2): 105–111

[74] Zhang YK, Yeager RL, Klaassen CD. Circadian expression profiles of drug-processing genes and transcription factors in mouse liver. Drug Metab Dispos 2009; 37(1): 106–115

[75] Sinal CJ, Yoon M, Gonzalez FJ. Antagonism of the actions of peroxisome proliferator-activated receptor-α by bile acids. J Biol Chem 2001; 276(50): 47154–47162

RIGHTS & PERMISSIONS

Higher Education Press and Springer-Verlag GmbH Germany, part of Springer Nature