Diet is an important factor in maintaining energy homeostasis, and it is taken orally by animals. Fat, a major nutrient within a diet, is digested into fatty acids and glycerol by pancreatic lipase in the small intestine (Porsgaard et al., 2005). Fatty acid is absorbed as nutrition in the small intestine, and particular fatty acids are ligands of fatty acid receptors which detect concentration of fatty acid in the gastrointestinal tract to secrete hormones in intestinal mucosa (Yu et al., 2020). Long chain fatty acids are major fatty acids digested from dietary fat, and some long chain fatty acids are ligands of the fatty acid-specific G-protein-coupled receptors (GPRs), such as GPR40 and GPR120 (Briscoe et al., 2003; Itoh et al., 2003; Hirasawa et al., 2005; Briscoe et al., 2006; Kimura et al., 2020).

GPR120 is expressed in endocrine cells from the small intestine to the large intestine (Anbazhagan et al., 2016). In mice, a high-fat diet containing fatty acids including GPR120 ligands for 5 weeks reportedly reduced insulin resistance through the inhibitory effect of transforming growth factor-β activated kinase 1 (Oh et al., 2010). GPR120 is expressed in the enteroendocrine cells, and STC-1 cells, an enteroendocrine cell line, secrete glucagon-like peptide 1 (GLP-1) and cholecystokinin (CCK) in response to long chain fatty acid recognized by GPR120 (Hirasawa et al., 2005; Tanaka et al., 2008). Oral pre-administration of GPR120 ligand (TUG-891) tends to ameliorate hyperglycemia after glucose load in rats (Li et al., 2017). GPR40 is highly expressed in the ileum, especially in L cells, which secrete GLP-1 (Edfalk et al., 2008; Little et al., 2014). Beta cells in the human pancreas express GPR40 (Tomita et al., 2005) and GLP-1 receptor, which enhance the secretion of insulin in response to the elevation of blood glucose level (Luo et al., 2012). In rats, the oral administration of GPR40 ligand, fasiglifam (TAK-875) and MR1704 an hour or more before glucose load reportedly decreased the postprandial glucose level after the glucose load by promoting insulin secretion in response to elevation of blood glucose level (Tsujihata et al., 2011; Tsuda et al., 2017). Long-term administration of GPR120 agonists and pre-administration of GPR40 agonists is said to allow the possibility of development of drugs for type 2 diabetes via insulin-dependent pathway (Christiansen et al., 2008; Tikhonova et al., 2008; Oh et al., 2010). However, when administered immediately before glucose load, the effects of GPR40 and GPR120 agonists in rodents or humans have been unclear.

Gastric emptying is controlled by various mechanisms including the endocrine system and/or nervous system, and it regulates the sending of gastric content in the gastric cavity to the small intestine (Goyal et al., 2019). Postprandial glycemia is attenuated by gastric emptying being slowed in healthy subjects (Deane et al., 2010). Endocrine cells in the small intestine sense the nutritional environment of their cavity, and they secrete CCK and GLP-1, which slows the gastric emptying (Deane et al., 2010; Kimura et al., 2020). Linoleic acid (LA), a dual agonist of GPR40 and GPR120, is of the triglyceride type in nature and is contained in vegetable oils, including olive oil. However, LA contained in triglycerides might not reveal the agonist activity of GPR40 and GPR120 immediately after taking triglycerides containing LA. This is because agonist activity of GPR40 and GPR120 is revealed in LA as a fatty acid, but not LA contained in triglycerides. Moreover, time is needed to digest the triglyceride from the daily diet to fatty acid and glycerol in the duodenum by pancreatic lipase. In this sense, oral administration of LA as fatty acid was different from that of LA contained in triglyceride regarding the immediacy of the effect. Oral administration of LA, a free fatty acid, appears to promote GLP-1 and/or CCK secretions immediately after LA administration. This could result in slowing the elevation of postprandial blood glucose level.

This study investigates whether the oral administration of LA taken immediately before glucose load slows the elevation of postprandial blood glucose level. We also examine whether the beneficial effect is induced by GLP-1 or CCK, and whether their secretions are associated with the GPR40 or GPR120 pathway.

Particular long-chain fatty acids are the agonist of GPR40 and GPR120, which are associated with gastrointestinal endocrine system, as well as nutrition. GPR40 agonists were previously reported to promote the secretion of insulin in response to the elevating glucose level when rats were administered the agonists one or 2 h before glucose load, and to decrease the postprandial glucose level and plasma glucose AUC _0–120 min after glucose load (Tsujihata et al., 2011; Tsuda et al., 2017). GPR120 agonists improved insulin resistance in long-term administration (≥8 weeks) (Oliveira et al., 2015; Jia and Liu, 2022), but did not ameliorate postprandial hyperglycemia by single oral administration 30 min before glucose load in mice (Li et al., 2017). This study indicates that LA, an agonist of both GPR40 and GPR120, slowed the elevation of postprandial blood glucose level when rats were administered LA immediately before the glucose load. This mechanism was slowing of gastric motility, and it is associated with promoting secretion of GLP-1 via GPR120 signaling.

Oral administration of LA immediately before glucose load did not decrease postprandial blood glucose levels by 120 min after glucose load, but it did decrease them up to 30 min after glucose load (Figure 1A). This effect is different from the effect of peroral administration of GPR40 agonist on postprandial glucose levels (Tsujihata et al., 2011), because the pretreatment of GPR40 agonist promotes insulin secretion in response to the elevating glucose levels, then decreases the postprandial glucose levels compared with a control. This GPR40 agonist effect may therefore be the superior to the LA effect in healthy individuals or patients with type 2 diabetes. However, oral administration of LA immediately before glucose load did not promote insulin secretion in response to the elevation of postprandial glucose level in normal rats (Figure 1C), and it decreased the postprandial glucose levels by 30 min after glucose load in the most severe T1DM model rats, whose ability to secrete insulin in response to the elevating glucose level was impaired by streptozocin (Figure 1D). LA could therefore slow the elevation of postprandial glucose level in patients with T1DM who take LA immediately before a meal. Previous clinical studies of T1DM indicated that the administration of rapid-acting insulin analogs immediately before meals resulted in the postprandial blood glucose level being higher than 180 mg/dL, and this administration 15 or 20 min before meals kept postprandial blood glucose levels under 180 mg/dL (Cobry et al., 2010; Luijf et al., 2010). Time-in-range has been proposed, it is the percentage of time patients spend with blood glucose level between 70 and 180 mg/dL (American Diabetes Association, 2021). Time-in-range has been reported to be negatively correlated with hemoglobin A1c (Valenzano et al., 2021). This administration of LA may slow the elevation of postprandial blood glucose level and extend the time when the postprandial blood glucose level is peak in patients with T1DM. The extension may be preferable to the recommended 15–20 min delay before being able to eat after the injection of insulin. Thus, administration of LA and insulin immediately before meals may improve hemoglobin A1c in T1DM. The oral administration of GPR40 agonists does not ameliorate the postprandial glucose level in patients with T1DM whose ability to secrete insulin is impaired. Future studies will examine the effect of oral administration of LA immediately before meals on the elevating postprandial glucose levels in patients with T1DM including during the honeymoon period of T1DM and in slowly progressive insulin-dependent T1DM.

We demonstrated that the oral administration of TriLA esterized with LA and glycerin immediately before glucose load and did not decrease the postprandial glucose levels up to 30 min after glucose load (Figure 1B). This result suggests that LA contained in triglycerides including olive oil and TriLA did not slow the elevation of postprandial glucose levels in rats administered them immediately before glucose load. Fatty acids, but not those in triglyceride form, are signaling molecules recognized by free fatty acids receptors including GPR40 and GPR120 (Hara et al., 2013). In order to indicate the agonist activity of GPR40 and GPR120, triglyceride taken orally is needed to be driven to the duodenum and digested by pancreatic lipase (Zoller et al., 2022). Oral administration of TriLA immediately before glucose load may therefore be insufficient to have the GPR40 and/or GPR120-agonisticeffect immediately after TriLA load. However, the interaction between LA and compounds (e.g., polyphenols, vitamin E, phytosterols and carotenoids) included in olive oil was not excluded in this study because olive oil was used as control and solvent of fatty acid and TriLA to minimize the difference of nutrition intake between the control and other groups.

Glucose is transported by SGLT 1 and 2, which are located on the brush border lumen of the small intestine and the proximal tubule (Du et al., 2018), and SGLT1 is expressed in the small intestine, especially the jejunum, but SGLT2 is not (Balen et al., 2008; Wright et al., 2011). We measured the blood glucose in the portal vein to examine whether LA also inhibited glucose uptake in the small intestine. LA decreased the postprandial glucose level in the portal vein 15 min after glucose load (Figure 2A), so LA might inhibit glucose uptake via SGLT1 in the small intestine. α-MDG is a selective substrate of SGLTs (Mitsuoka et al., 2016), and phlorizin is an antagonist of SGLT1 and SGLT2 (Niederberger et al., 2020). CACO-2 cells from DSMZ were previously shown to efficiently express SGLT1, and α-MDG uptake in CACO-2 cells from DSMZ was at maximum efficiency 13 or more days after seeding (Steffansen et al., 2017). CACO-2 cells from DSMZ were used to examine whether LA was an antagonist of SGLT1, and LA did not inhibit uptake of α-MDG (Figure 2B). LA did not therefore directly inhibit glucose uptake in lumen of the small intestine via SGLT1.

Gastric motility is the delivery of gastric contents into the duodenum and is slowed by several factors, including hormones (GLP-1 and CCK) (Goyal et al., 2019). LA might decrease the rate of gastric emptying, and we measured the remaining amount of glucose in the gastric tract. The amount of loaded glucose was reduced to 50% 15 min after glucose load, and to 30% 30 min after glucose load in control group (Figure 3A). LA had slowed this gastric motility at 15 min after glucose load, and the motility had almost disappeared between 15 and 30 min after glucose load. Oral administration of LA increased the plasma GLP-1 concentration, but not CCK (Figures 3B, C). Oral administration of LA might therefore slow gastric motility quickly via GLP-1 signaling, and inhibit movement of glucose from the stomach to the duodenum, because GLP-1 slows the gastric emptying (Deane et al., 2010). However, it was unclear whether oral administration of LA immediately before glucose load directly promoted GLP-1 secretion via GPR120 pathway and slowed gastric empty by GLP-1. Future studies should clarify this concern using GLP-1 antagonist or GPR120 knockout mice.

LA is a dual ligand of both GPR40 and GPR120 (Kimura et al., 2020), but previous studies indicated the effect of postprandial glucose level on pretreatment of GPR40 or GPR120 agonists (Tsujihata et al., 2011; Li et al., 2017). It was unclear, however, whether the postprandial glucose level was improved by administration of these agonists immediately before glucose load. TUG-891, a GPR120 agonist, decreased the postprandial glucose level by 30 min after glucose load compared with the olive oil group, but GPR40 agonist did not show beneficial effects (Figure 4A). Pretreatment of GPR120 antagonist partially canceled the beneficial effects induced by LA (Figure 4B). Alpha-LA has high affinity of ligands in both GPR40 and 120, but oleic acid had high affinity of ligand in GPR40 and not in GPR120 (Kimura et al., 2020). The oral administration of α-LA immediately before glucose load slowed the elevation of postprandial glucose level, but oleic acid did not reveal the beneficial effects (Figures 4C, D). Slowing the elevation of postprandial glucose level in oral administration of LA may therefore be associated with GPR120 signaling.

This study used olive oil as control and solvent of fatty acids or TriLA to minimize the difference of nutrition intakes. However, olive oil contains polyphenols, vitamin E, phytosterols, and carotenoids derived from olives as well as triglyceride, and these substances might affect the postprandial glucose levels. In this study, the solvent in the LA (625 mg/kg) group containing 87.5% olive oil and 12.5% LA also slowed the postprandial glucose levels (Supplementary Figure S2), but the solvent in the TriLA group containing 50% olive oil and 50% TriLA did not slow the postprandial glucose levels (Figure 1B). Thus, the concern was almost minimized, but was not fully solved in this study.

This study indicated that the oral administration of GPR120 agonist immediately before glucose load decreased the postprandial blood glucose levels up to 120 min after glucose load (Figure 4A). In case of LA administration, the blood glucose level 120 min after glucose load was high compared with in the olive oil group. The higher blood glucose levels were also observed in α-LA and oleic acid group but not in the olive oil or the TriLA group (Figures 1A, B, 4C, D). This maintained higher glucose level was also observed in TAK-875 group in a dose-dependent manner (Supplementary Figure S3). Thus, the oral administration of fatty acids immediately before glucose load might therefore inhibit decrease of the postprandial glucose level 120 min after glucose load, and this mechanism might be associated with GPR40 pathway. TUG-891, a GPR120 agonist, did not indicate this effect, so a GPR120 agonist could ameliorate the postprandial hyperglycemia. The EC50 of GPR120 is 75.3 nM, and that of GPR40 is 52.6 μM in TUG-891 (Li et al., 2017). The EC50 of GPR120 is more than 10 μM, and that of GPR40 is 33.6 nM in TAK-875 (Li et al., 2017). The EC50 of GPR120 is 1 μM, and that of GPR40 is 2 μM in LA (Kimura et al., 2020). These selective GPR120 or GPR40 agonists have 10 or 20-fold affinity with each receptor compared with LA. Thus, GPR120 agonist, which has high affinity and selectivity of GPR120, might decrease postprandial glucose levels in animals and humans immediately before glucose load or meals. Further studies will examine the oral administration of selective GPR120 agonists which decrease the postprandial blood glucose level immediately before glucose load in a dose-dependent manner, with the ultimate aim of developing drugs for individuals with type 1 as well as type 2 diabetes.

Oral administration of free LA, a GPR120 agonist, immediately before glucose load improved postprandial glucose levels independent from insulin. LA contained in dietary fat may therefore improve postprandial hyperglycemia in patients with T1DM.