Introduction
Type 2 diabetes mellitus (T2DM) accounts for the majority of all diabetes cases. It affected approximately 171 million people worldwide in 2000, and this number is estimated to grow to 366 million people by 2030 [
1,
2]. As a systemic, chronic and progressive disease, T2DM has its roots in genetics but is also heavily influenced by the environment. Despite the relative wealth of knowledge about T2DM obtained by intensive research during the past two decades, current medical and conventional interventions cannot reverse the disease progression and its high incidence [
3,
4]. The time of T2DM diagnosis is called the time of “no return,” which means patients are compelled to take medications for the rest of their lives [
5].
Bariatric surgeries were first designed to treat morbid obesity. Obesity is another rapidly growing health care challenge, and it usually occurs hand in hand with T2DM [
6]. Bariatric surgeries have emerged as the most effective and successful long-term methods for the treatment of morbid obesity, as compared with traditional weight loss therapies [
7-
9]. In particular, Roux-en-Y gastric bypass (RYGB) is considered by surgeons worldwide as the “gold standard” procedure [
10]. Surprisingly, RYGB has been repeatedly shown to dramatically improve T2DM, resulting in normal blood glucose and glycosylated hemoglobin levels, without any blood glucose controlling medications [
11-
17]. This remission is also persistent, with rare recurrences in more than 10 years, as reported by a long-term follow-up clinical study [
7]. Since obesity itself causes some degree of insulin resistance, which is a major cause of T2DM, weight loss caused by the surgery was proposed to be the underlying mechanism of T2DM remission after RYGB [
11]. However, a return to euglycemia is observed within days after surgery, long before any significant weight loss occurs, suggesting that weight loss alone cannot entirely explain the improvement of T2DM by surgery [
16].
In the past several years, clinical and laboratory studies, using different animal models, have been conducted to provide new insight into the molecular mechanisms by which RYGB dramatically arrests T2DM progression and reshapes the host’s metabolism. Increasing evidence suggests that the gut may play an important role in the amelioration of T2DM after gastric bypass surgery. In this report, we critically review the recent data and clinical studies addressing the changes in the gut related to T2DM remission after RYGB. We summarize the potential relationships between the gut and T2DM in the order from macro to micro, including gut anatomical rearrangement, microbial composition changes, altered gut cells, and gut hormone modulations (Fig. 1). Understanding the connections between gut and metabolism is critical to shift the treatment modes from bariatric surgeries to metabolic surgeries.
Gut anatomical changes from gastric bypass surgeries
The most effective procedures in terms of T2DM remission among gastric bypass surgeries, as well as bariatric surgeries, are RYGB and biliopancreatic diversion (BPD) [
18,
19]. The remission rates reported for RYGB and BPD are 80%-83% and 96%-100%, respectively [
20]. Except for a two/third distal gastrectomy, BPD is, for all intents and purposes, a long-limb RYGB. As the most effective procedures, they share two common gut anatomic changes: bypass of the duodenal and upper jejunum, and fast delivery of undigested food to an ileal channel [
18]. Thus, the “foregut” and “hindgut” hypotheses have been proposed [
19].
The foregut hypothesis holds that the exclusion of nutrients from the duodenum reduces the release of factors relevant to insulin resistance, which contributes to improved glucose tolerance [
21-
28]. The hindgut hypothesis emphasizes that the improved metabolism after gastric bypass is caused by factors, such as glucagon-like peptide 1(GLP-1), released from the hindgut after exposure to increased undigested nutrients [
22]. Both hypotheses have been supported in animal and clinical studies by specifically designed surgeries, duodenal-jejunal bypass (DJB) and ileal transposition (IT), respectively [
21,
22]. The duodenal factors contributing to improved insulin resistance and T2DM remain unidentified, and the role of the foregut in the early improvements of glucose tolerance and insulin sensitivity following bypass surgery is still under debate. Indeed, neuronal and metabolic feedback systems can be activated by the small intestine to counteract energy excess and glucose imbalance [
23-
28]. Breen
et al. reported that intrajejunal nutrient administration by DJB rapidly lowers plasma glucose concentrations in normal rats, streptozotocin-induced diabetic rats and autoimmune type 1 diabetic rats. The metabolic improvement is not associated with body weight and plasma insulin levels, but disappears if the jejunal glucose uptake or the formation of long chain fatty acyl-CoA is blocked. Thus, they concluded that the jejunal nutrient-sensing is revitalized by enhanced nutrient redirection by the anatomic changes of DJB, and decreased hepatic glucose production leads to early improvements in glycemic control [
29]. On the contrary, Hansen
et al. recently conducted a prospective crossover study, in which a mixed meal was delivered to either the stomach or jejunum of nine subjects, before and after RYGB. Each subject underwent two tests, receiving food by either oral administration, which enters the jejunum first, or gastrostomy tubing, which enters the stomach first. They found that the exclusion of nutrients from the foregut with RYGB does not improve either glucose tolerance or insulin sensitivity. These data argued against the bypass of nutrient exposure to the foregut as the primary mechanism in the immediate improvements in glucose tolerance following gastric bypass [
30], and suggested that the hindgut may be more important in diabetes remission after gastric bypass.
The experimental surgical technique IT has provided important insights into the role of the hindgut, independent of decreased food intake and body weight loss [
31]. The fast delivery of food to the ileum by IT with sleeve gastrectomy induces weight loss and T2DM remission, in a similar manner to gastric bypass in human [
32]. IT also dramatically improved glucose tolerance after oral glucose challenges in diet-induced obese rats, but the beneficial effects disappeared if the GLP-1 receptor was blocked by extendin 9-39 [
31]. Increased GLP-1 receptor signaling and exaggerated GLP-1 secretion from the L cells, as well as an increase in circulating GLP-1 levels after RYGB, have been reported [
31-
34]. However, the exact mechanism or stimulus that promotes GLP-1 secretion remains unidentified [
26,
35,
36], given all the possibilities of nutrient-rich chyme changes after RYGB. Some scientists proposed a potential role of bile acids, key enzymes in lipid metabolism, in stimulating GLP-1 secretion post-gastric bypass [
37-
40]. The anatomic changes caused by gastric bypass inevitably affect bile acid circulation, and thus more bile acids accumulate in the hindgut [
38,
39]. Comparison studies of gastric bypass and gastric banding in human, canine and rodent models revealed increased concentrations of bile acids, incretion and satiety gut hormone concentrations in blood after gastric bypass [
40].
In vitro experiments provided evidence that the G-protein coupled bile acid receptor (TGR5) is linked to GLP-1 secretion [
39]. The drainage of endogenous bile into the terminal ileum achieved the same gut hormone responses [
40,
41]. The beneficial effects of gut hormones in regulating glucose homeostasis are well documented, and will be described in the section of
Hormone release. These data strengthen the idea that the hindgut contributes to the enhanced incretin response that improves metabolism after RYGB.
The alimentary limb, which connects to small portion of stomach and passes food from stomach to the common channel of RYGB, digests and absorbs most of the starches and proteins, due to the actions of brush-border enzymes. Beyond the functions of glucose and peptide absorption, the middle gut can also produce glucose and release it into the portal blood, in a process called intestinal gluconeogenesis. The key enzymes of gluconeogenesis, Glc6Pase and phosphoenolpyruvate carboxykinase (PEPCK), are synthesized in the small intestine of both rat and human. The portal sensing of intestinal gluconeogenesis, by the infusion of glucose into the portal vein, reportedly decreased food intake and restored glucose homeostasis [
42]. Troy
et al. speculated that the above mechanism may explain the modulated whole-body glucose disposal after bariatric surgery. They performed entero-gastro anastomosis (EGA) and gastric banding experiments in C57Bl6 mice on a high-fat diet, and detected increased intestinal gluconeogenesis, which activated the hepatoportal sensor and inhibited glucose release from the liver, in the EGA group. The beneficial effect disappeared when they performed the EGA procedure in GLUT2 knockout or portal vein denervated mice, which indicated that the middle gut of the gastric bypass procedure can somewhat modify glucose homeostasis by increasing intestinal gluconeogenesis via a GLUT2-hepatoportal sensor pathway [
43]. This conclusion was subsequently tested in humans by Kashyap
et al. [
44]. However, another clinical study conducted by Hayes
et al. found no significant difference in the glucose levels in the fasting portal venous blood and the central venous blood of 28 patients, before and six days after RYGB surgery [
45]. These results did not support the proposal that the fast T2DM resolution after RYGB is due to intestinal gluconeogenesis from the midgut. However, since both the foregut and midgut presented gluconeogenesis as an important factor, further research is needed to study this hypothesis as a potential mechanism to explain the early improvement in glucose homeostasis observed following RYGB.
Microbe composition
Microbes reside in places where the human body interacts with the environment, such as the skin, gut and vagina, throughout life. In healthy individuals, the microbial communities, mainly gut flora, are stable and help to maintain a healthy environment, while in disease conditions, the microbial population tends to change, with an increase in opportunistic pathogens that contribute to disease progression [
46]. Studies from mouse models suggested that gut microbe disorders may contribute to the development of many complex diseases [
47-
53]. On the other hand, body changes could also affect the gut microbial composition. For example, the lack of Toll-like receptor 5 (TLR5), a component of the innate immune system that is expressed in the gut mucosa and helps defend against infection, and changes in fatty acid synthase (FAS), which is required to maintain the integrity of the intestinal mucosal layer, will shift the gut microbe composition [
54,
55]. In recent years, 16S rRNA [
56] and whole-genome shotgun (WGS) sequencing [
57] revealed an overall picture of the microbial communities. Researchers discovered that there are mainly two types of microbial flora in the human gut, the Bacteroidetes and the Firmicutes. Most of them reside in the large intestine [
57].
A tightly coordinated connection exists between the gut microbes and the host’s metabolism, energy utilization, and fat storage [
49-
53,
58-
61]. To gain detailed information about gut microbial compositional changes and their associated impact on T2DM, a metagenome-wide association study was performed, based on deep next-generation shotgun sequencing of DNA extracted from stool samples from a total of 345 Chinese T2DM patients and nondiabetic controls. The results revealed a decrease in metabolically beneficial butyrate-producing bacteria and an increase in various opportunistic pathogens, in patients with T2DM. An increase in opportunistic pathogens is associated with a variety of diseases. The researchers proposed that it is not the particular microbial species, but rather the “functional dysbiosis” that is linked with T2DM pathophysiology [
62].
RYGB profoundly changes the digestive tract, and thus not surprisingly affects the microbial environment. Zhang
et al. examined 184 094 sequences of microbial 16S rRNA genes using 454-pyrosequencing technology, in stool samples from 9 individuals before and after gastric bypass surgery. They concluded that RYGB uniquely changed the intestinal microbial community, with a shift toward a large increment in Bacteriaceae, a proportional decrease in Firmicutes, and a total loss of methanogens [
52,
63]. However, further studies are needed to determine whether such modifications in the microbial composition are related to the consequences of weight loss or T2DM remission after gastric bypass surgery.
Cell differentiation direction
The development of the pancreas begins with a region of localized primitive gut epithelium [
64-
68]. Neurog3-expressing (Neurog3
+ ) endocrine progenitors are located in the pancreas and intestine, where they further develop into various cell types that produce different hormones during embryonic development [
69-
72]. For example, in the pancreas, the progenitors give rise to all pancreatic islet cell types, while in the gut, they generate most of the cells in the enteroendocrine system [
73,
74]. Pancreatic endocrine progenitors exist in the pancreas of adult mice [
75,
76], but are not activated to differentiate into functional βcells except under extreme circumstances, such as partial duct ligation [
77]. However, enteroendocrine progenitors maintain the population of gut endocrine cells by continual differentiation and proliferation after birth [
74].
Since the Neurog3
+ progenitors found in the pancreas and gut system are from the same endodermal origin [
75], researchers are testing the possibilities of converting enteroendocrine progenitors into functional β cells. Forkhead box protein O1 (Foxo1) is co-expressed with Neurog3 in the embryonic pancreas [
78]. As a master transcription factor, Foxo1 affects cell fate [
79]. Studies have shown that the knockdown of the Foxo1 protein increases the numbers of Neurog3
+ cells in human fetal pancreatic epithelium [
80]. To investigate the Foxo1 function in the gut, mice with a somatic Foxo1 deletion in Neurog3
+ enteroendocrine progenitor cells were generated. They confirmed that Foxo1 ablation causes the expansion of Neurog3
+ enteric progenitors in the gut. These progenitors then differentiated into mature β cells, which were able to secrete insulin in response to changing blood glucose levels and to maintain glucose homeostasis in diabetic mice [
81]. When the knockout animals were treated with the pancreatic β cell toxin streptozotocin, the gut Ins
+ cells spontaneously regenerated and secreted insulin, reversing the hyperglycemia in mice without the need for insulin therapy [
81]. The differentiation of enteroendocrine progenitors after RYGB has not been reported yet, but it is a very attractive and interesting area which may explain the enhanced insulin response in the glucose-stimulated manner and the improved T2DM after gastric bypass surgery.
Hormone release
Most clinical studies have focused on hormone secretion after gastric bypass. However, given the complexity of gut hormones and the paucity of basic research information about them, it is not known their relative importance in the metabolic improvements among the known hormones and whether a new “miracle” gut hormone that can cure T2DM exists.
(1) Ghrelin. Ghrelin is a 28 amino acids peptide [
82] secreted predominantly by X/A-like enteroendocrine cells, with the highest content in the fundus of the stomach [
83]. Until now, it is the only known appetite-stimulating hormone [
84]. Ghrelin has been shown to increase the levels of growth hormone (GH), cortisol, and epinephrine, three hormones that increase blood glucose and oppose insulin action [
85-
87]. Furthermore, ghrelin could influence insulin secretion via the ghrelin receptor GHSR1-a, which is expressed in pancreatic β cells [
88,
89]. Decreases of insulin secretion in both human and animal studies were reported after intraperitoneal ghrelin administration [
90,
91]. Therefore, the suppression of post-prandial ghrelin secretion is associated with improved glucose homeostasis. No clear conclusions have yet been drawn from publications about ghrelin changes after RYGB: most reported an increase [
92-
97]; however, no changes were also found [
98,
99].
(2) The gut incretin hormones. Glucose-dependent insulinotropic polypeptide (GIP) and GLP-1 are gastrointestinal incretin hormones secreted by neuroendocrine K cells and L cells, respectively [
22,
100-
104]. The physiological effects of the incretins are responsible for ~50% of post-prandial insulin secretion [
105,
106], and are blunted in patients with T2DM [
107]. The beneficial effects of GLP-1 on pancreatic β cells and glucose homeostasis have been established, and enhanced amounts of post-prandial circulating GLP-1 were repeatedly found after RYGB [
92,
108-
116]. Regarding GIP, the results are less consistent, with some researchers finding elevated [
111,
114] or decreased [
92,
113] levels. Moreover, studies suggested that the incretin changes after gastric bypass are independent of weight loss [
117].
(3) Other gastrointestinal hormones. In addition to GLP-1, the secretion of other products from neuroendocrine L cells, peptide YY (PYY) and oxyntomodulin (OXM), is also enhanced after RYGB [
109,
115,
118-
122]. (i) OXM, a proglucagon-derived 37 amino acids peptide hormone, reduced food intake and body weight [
103,
123-
125], and improved glucose homeostasis in human and rodents [
124,
126-
128]. A chemically modified OXM resistant to dipeptidylpeptidase-IV (DPP-IV) degradation exhibited a wide range of beneficial actions on β cells, insulin resistance, satiety and body weight [
129]. (ii) PYY is a 36 amino acids peptide that belongs to the pancreatic polypeptide family [
130], and is particularly abundant in the distal intestine [
130,
131]. PYY dysfunction is implicated in the pathogenesis of human and rodent obesity [
132-
134]. In addition, a blunted postprandial PYY concentration in serum is associated with the development of T2DM [
135-
137].
The gut-brain-axis (GBA), which is a bidirectional neurohumoral communication system that integrates brain and gastrinal functions, is also implicated in the pathophysiology of functional gastrointestinal disorders. The communication between gut hormones and the brain, in the modulation of food intake and energy expenditure after gastric bypass, has been excellently reviewed by Dr. Berthoud [
138], and thus we will not discuss this area.
Conclusions
The gastrointestinal tract is an intimate place where our bodies communicate with the environment. As an endocrine organ, the gut transmits information about the nutrient load of a meal as well as the microbial responses into neurohumoral signals to other tissue, thereby modulating metabolic control [
2-
4]. Changes in its anatomy predictably affect the above neurohumoral signals. After gastric bypass surgeries, gradual changes from the cellular to molecular levels, in the bowel-wall and intestinal lumen occur. Those changes may not happen simultaneously. The great opportunities offered by gastric bypass in understanding and treating diabetes are novel. However, more research focusing on the various changes in the gut post-gastric bypass is required for a comprehensive appraisal of the mechanisms underlying T2DM and its control. These advances will surely provide valuable knowledge about T2DM and its treatment.
Higher Education Press and Springer-Verlag Berlin Heidelberg