Intestinal Lipoprotein Secretion: Incretin-Based Physiology and Pharmacology Beyond Glucose

The incretin axis, mostly glucagon-like peptide 1 (GLP-1), is a well-acknowledged and -applied science of glucose homeostasis. Importantly, modifying this physiology with GLP-1 receptor (GLP-1R) agonists or inhibiting the enzymatic breakdown of GLP-1 with dipeptidyl peptidase-4 (DPP-4) inhibitors experienced increasing success in achieving glycemic control in patients with type 2 diabetes (T2D) (1). Yet, the role of gut peptides on lipid and lipoprotein metabolism has been relatively unnoticed, and the favorable changes in fasting and nonfasting lipids with incretin-directed pharmacotherapy may be an important pathway to promote cardiovascular health beyond glucose lowering alone (2).

The most typical lipid and lipoprotein disorder in T2D is hypertriglyceridemia with a reduced plasma level of HDL cholesterol, mechanisms related to increased VLDL triglyceride (TG) production and secretion, and reduced clearance of TG-rich particles by lipoprotein lipase (3). Moreover, increases in postprandial lipid excursion in patients with T2D are also seen, an alteration related to similar mechanisms in the setting of chylomicron assembly and secretion by the intestine. Herein is where the evolving science of incretins is worthy of attention.

In this issue of Diabetes, the Perspective by Xiao et al. (4) is timely and state of the art. As documented by the authors, treatment of patients with T2D with GLP-1R agonists has variable reductions in fasting TG, an effect typically more pronounced with GLP-1R agonists than DPP-4 inhibitors. A reduction mediated by GLP-1R agonists and/or DPP-4 inhibitors in postprandial TG excursion may also be beneficial in reducing the exposure of the vasculature to proatherogenic chylomicron remnants (5); however, an important question is whether this effect is mediated entirely within the intestine and/or through the central nervous system? Although studies in rodents indicate that GLP-1R agonists reduce chylomicron size, it remains unclear as to whether these particles when secreted may be cleared more efficiently. The GLP-1R– and DPP-4–related reduction in markers of oxidative stress and inflammation are also encouraging. In addition, GLP-2 looks like an exciting molecule for further investigation; the effect to enhance the processing and release of intestinally stored lipids and lipoproteins has implications for physiology, disease, and therapeutics to follow. The fact that GLP-2 receptors are not found on epithelial cells, but on neuroendocrine cells, and that the effect of GLP-2 on TG absorption may be mediated by increases in intestinal blood flow are very provocative observations.

There are a number of issues related to the effects of the GLP-1R agonists and DPP-4 inhibitors on intestinal lipid and lipoprotein metabolism that need to be addressed. How much of the GLP-1R agonist and DPP-4 effects on TG relate to improved glycemia (6,7) and less likely weight loss (8) rather than to the incretin axis alone? The greater effect of GLP-1R agonists versus DPP-4 inhibitors would support dependent effects. In an attempt to examine time-independent effects of a GLP-1R agonist, 14 normal-weight male subjects were infused intravenously over 390 min with either GLP-1(7-36) amide or saline with a standard 251 kcal meal with 30% fat fed at time 0 (9). Although the GLP-1R agonist totally blocked the postprandial TG rise, with saline the increase in TG at 3 h was modest, from ∼80 to ∼115 mg/dL. More importantly, when examined in a randomized cross-over manner in subjects with impaired glucose tolerance or T2D and hypertriglyceridemia (fasting plasma TG 2.6 mmol/L), a single subcutaneous injection of exenatide (10 μg) or saline just prior to an extremely high-calorie (1,286 kcal), high-fat (45% of kcal) breakfast meal reduced almost entirely the postprandial elevation of TG, apolipoproteins B-48 and CIII, remnant lipoprotein cholesterol, and remnant lipoprotein TG (10). Here, the rise in postprandial TG was more impressive, from a baseline TG of ∼3.0 mmol/L to 4.5 mmol/L with placebo; however, of interest, 8-h postfeeding plasma TG levels were still elevated and similar for exenatide- and placebo-treated subjects at ∼3.5 mmol/L. Thus, in the setting of an incredibly high caloric and fat load, the independent incretin effect of exenatide was clear, a scenario that has not been routinely experienced in most studies carried out over intervals of 1 to 16 weeks, a period over which glycemia and weight were variably reduced (see Table 1 in ref. 4). Moreover, the baseline TG data in Table 1 of Xiao et al. were not provided for many of the studies cited.

Xiao et al. (4) claim that there is no relationship between fasting TG and postprandial TG excursion; yet, substantial evidence to the contrary across the spectrum of normotriglyceridemia to moderate hypertriglyceridemia exists (11–14). Moreover, the mechanism of the GLP-1R agonist and DPP-4 effect to reduce postprandial TG excursion is unclear (Fig. 1). Yes, an insulin-mediated effect on free fatty acid flux to the intestine may be important, but if this results in reduced chylomicron assembly and secretion, what happens to an equal amount of dietary fat in the absence of steatorrhea? Presumably and hopefully, the reduction in postprandial chylomicron excursion will result in less chylomicron remnant production and less atherosclerosis, a theory originally propagated by Zilversmit (15) and maintained in recent studies (16,17). However, this hypothesis remains controversial, and, at present, convincing evidence for modifying fasting TG and/or postprandial TG and reduction of cardiovascular disease events is lacking, although post hoc analyses of fibrate trials suggest such a benefit in hypertriglyceridemic subjects with lower HDL cholesterol concentrations (18,19). At present, numerous ongoing trials of GLP-1R agonists and DPP-4 inhibitors are being conducted to assess predominantly cardiovascular disease safety, not benefit (20,21).

In summary, GLP-1R agonists and DPP-4 inhibitors are important options for the treatment of T2D. The glycemic benefits are well documented and the effects on intestinal lipid and lipoprotein processing and systemic metabolism are documented for GLP-1R agonists and DPP-4 inhibitors; however, in general these effects are modest and appear to relate partly to improved glycemia (GLP-1R and DPP-4 inhibitors) and perhaps less likely weight reduction (GLP-1R agonists), although the combination of both may be even more important. Nevertheless, the field is ripe for additional insight, including the apparent divergent effects of GLP-1 from GLP-2 on intestinal chylomicron assembly and secretion, and determination of how this science relates to T2D and the link between lipid and lipoprotein metabolism and macrovascular complications—a big task!