HYA ameliorated postprandial hyperglycemia in type 1 diabetes model rats with bolus insulin treatment

Fatty acids are a source of energy that are absorbed in the small intestine after digestion of pancreatic lipase. Some fatty acids are sensed by G protein-coupled receptors in the intestinal mucosa [1], and long-chain fatty acids are agonists of G protein-coupled receptor (GPR) 40 and/or GPR120 [2]. Oral administration of linoleic acid, included in long-chain fatty acid, two hours before an oral glucose tolerance test (OGTT) reportedly ameliorated postprandial hyperglycemia by promotion of insulin secretion via GPR40 pathway in mice [3]. We previously demonstrated that orally-administered linoleic acid immediately before an OGTT slowed the gastric motility through promoting GLP-1 secretion via GPR120 pathway, which ameliorated the postprandial hyperglycemia in normal rats and slowed the elevation of postprandial glucose level in type 1 diabetes model rats [4]. Linoleic acid is also converted to lipid mediators of the inflammatory responses including prostaglandins and thromboxane through arachidonic acid cascade [5]. The concentration of prostaglandin E2 is increased in the ileum and white adipose tissue of the mice by the administration of linoleic acid. Long-term use of linoleic acid to ameliorate postprandial hyperglycemia might therefore be inappropriate because of the production of lipid mediators.

Lactobacillus plantarum was reported to hydroxylate a double bond in polyunsaturated fatty acids and to ketonize the hydroxyl group, and it produces various intermediary metabolites including 10-hydroxy-cis−12-octadecenoic acid (HYA) through these conversions [6]. HYA is a stronger agonist of GPR40 and GPR120 than linoleic acid [3, 7], and it attenuates inflammatory responses in CACO-2 and BV-2 cells [7, 8]. Converting dietary linoleic acid to HYA by gut microbiota suppresses the production of inflammatory mediators [3]. HYA might therefore be superior to linoleic acid at the point of the agonistic activity of GPR40 and GPR120 and inflammatory responses.

Previous reports have indicated that the oral administration of HYA 2 h before OGTT in mice ameliorated postprandial hyperglycemia as well as linoleic acid, depending on the promotion of insulin secretion via GPR40 pathway [3]. Although the oral administration of HYA immediately before OGTT was expected to reveal the amelioration of postprandial hyperglycemia by the same mechanism as linoleic acid, as reported in our previous study, the result was unclear [4]. For clarification, we now investigate the effect of orally-administration of HYA immediately before OGTT on postprandial glucose level in rats.

Five-week-old male Sprague-Dawley rats were from CLEA Japan, Inc. (Tokyo, Japan). Rats were housed in plastic cages (24.7 × 40.9 × 19.7 cm) with free access to laboratory animal chow (Oriental Yeast Co., Ltd., Tokyo, Japan) and tap water under a 12 h light/dark cycle (lights on at 8:00 AM) at 25 ± 1 ºC and 50–60% humidity. All animals were used after a one-week acclimation period. The protocol of animal experiments was approved by the Wakayama Medical University Animal Experiment Committee on July 1, 2019 (No. 883).

CACO-2 cells were from Deutsche Sammlung für Mikroorganismen und Zellkulturen (Braunschweig, Germany). They were maintained in Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum, 1% non-essential amino acids, and 100 IU/mL penicillin-100 µg/mL streptomycin. Cells were grown under 5% CO₂−95% air at 37 °C.

OGTT was performed as in the previous reports [4, 9]. Rats were fasted overnight for performance of OGTT (2 g/kg body weight), and they were administered HYA solution (2.5 g/5 mL) or olive oil (each 5 mL/kg bodyweight) immediately before OGTT. Blood was taken from the tail vein prior to the OGTT and then 30, 60 and 120 min after the OGTT. Blood glucose levels were directly measured from blood samples with Freestyle Libre (Abbott, Alameda, CA). Concentration of insulin in plasma was measured using rat insulin measurement kits (Morinaga Institute of Biological Science, Kanagawa, Japan).

[¹⁴C]α-MDG uptake study was performed as previously reported [4], the procedures are summarized below. CACO-2 cells were seeded at a cell density of 2.0 × 10⁵ cells/well on 24-well plastic plates. The culture medium was refreshed every other day, and the cell monolayers were used for the uptake experiments 14 days after plating. The culture medium was removed and then all wells were washed and pre-incubated with glucose-free Hank's balanced salt solution (HBSS) buffer, and then 0.5 mL of HBSS buffer containing 0.4 µM (0.1 µCi/mL). [¹⁴C] α-MDG in the absence or presence of various compounds was added and incubated for 30 min at 37 °C. Each cell monolayer was rapidly washed twice with 1.0 mL ice-cold HBSS buffer at the end of the incubation period. To quantify the radioactivity of [¹⁴C] α-MDG, the cells were solubilized in SDS buffer, then to estimate the radioactivity, the remaining sample was mixed with 3 ml of scintillation cocktail. Uptake values were corrected for protein content, and the protein concentration was determined using a Pierce^® BCA Protein Assay Kit (Thermo Fisher Scientific, Waltham, MA, USA).

Type 1 diabetes model rats were made using previously-reported methods [4]. To prepare the type 1 diabetes model rats, six-week-old male Sprague-Dawley rats were intraperitoneally injected with 60 mg/kg streptozocin on the first day and 100 mg/kg streptozocin on the second day. The following week, the concentrations of plasma glucose and insulin were measured before and 30 min after oral glucose load (2 g/kg body weight). We excluded the rats according to two criteria in this experiment: blood glucose level < 500 mg/dL 30 min after glucose load, and the concentration of plasma insulin increased 1.5-fold 30 min after glucose load compared with that before glucose load. OGTT and the measurement of glucose level were also performed according to the above protocol in type 1 diabetes rats. The type 1 diabetes model rats given bolus insulin were also subjected to OGTT following the subcutaneous injection of Humulin^® R (0.1 U/kg).

Oral administration of HYA immediately before OGTT slowed the elevation of postprandial glucose levels in normal rats (Fig. 1A, Supplemental Fig. 1) as well as oral administration of linoleic acid [4]. This beneficial effect was associated without promotion of insulin secretion (Figs. 1B and 4A) but with inhibition of the absorption of glucose in the small intestine (Fig. 2A). HYA inhibited SGLT1-mediated transport in CACO-2 cells (Fig. 2B). It also slowed the gastric motility following the promotion of secretion of GLP-1 and CCK in normal rats (Fig. 3). Furthermore, HYA ameliorated postprandial hyperglycemia in type 1 diabetes model rats injected with bolus insulin immediately before OGTT (Fig. 4B). These results suggest that the beneficial effect of HYA might be associated with the inhibition of glucose uptake via SGLT1 in the mucosa of small intestine, and the slowing of transportation of glucose from the stomach to the small intestine via promotion of GLP-1 and CCK secretion.

SGLT1 is expressed on the brush-border membrane of epithelial cells from the small intestine to take up the glucose in the intestinal tract [10, 11]. HYA inhibited glucose uptake via SGLT1 in CACO-2 cells, although the inhibition was weaker than phlorizin, an SGLT1 inhibitor (Fig. 2B). LX2761 and LC-4211, SGLT1 inhibitors, inhibited the elevation of blood glucose level after OGTT through inhibiting the absorption of glucose in the small intestine mucosa [12, 13]. In patients with type 1 diabetes, sotagliflozin, a dual SGLT1 and SGLT2 inhibitor, improved glycemic control [14]. The amelioration of postprandial hyperglycemia in HYA might therefore be partially associated with SGLT1 inhibition.

In the current study, approximately double the amount of glucose remained 30 min after glucose load in the stomachs of the HYA group compared with in the control group (Fig. 3A). GLP-1 and CCK, which slow the gastric motility, are secreted in the small intestine via GPR120 [15, 16]. HYA acted as an agonist of GPR120 and promoted the secretion of GLP-1 [3]. The plasma levels of GLP-1 and CCK 30 min after the OGTT were increased by administration of HYA (Fig. 3B and C). Gastric motility slowed by HYA might therefore be associated with the promotion of secretion of GLP-1 and CCK via GPR120 in normal rats. The peak of postprandial hyperglycemia was lower in the patients with diabetes whose gastric emptying was lower [17], and GLP-1 and CCK were secreted after a meal in the patients with type 1 diabetes [18, 19]. This beneficial effect of HYA on postprandial hyperglycemia might therefore be associated with slowing gastric motility through promotion of the secretion of GLP-1 and CCK.

Several oral diabetes agents are known to promote the secretion of insulin in the pancreas, but they can cause hypoglycemia [20, 21]. Novel diabetes drugs have recently been developed, and the mechanisms of these drugs are enhancing insulin secretion responding to postprandial hyperglycemia through GLP-1 signaling [22, 23] and inhibition of glucose absorption via SGLT2 [24, 25]. These types of drugs have lower risks of hypoglycemia, because the secretion of insulin is not directly induced. Oral-administration of HYA without OGTT did not decrease the blood glucose level in normal rats (Supplemental Fig. 1B). This study demonstrated that HYA might be a diabetes agent without accompanying a risk of hypoglycemia.

Orally-administered HYA immediately before OGTT slowed the elevation of postprandial glucose level in type 1 diabetes model rats (Fig. 4A). In this study, we used type 1 diabetes model rats injected with bolus insulin. This is similar to the situation before a meal for patients with type 1 diabetes. The peak of postprandial glucose level was lower in the HYA group than in the control group in this type 1 diabetes model (Fig. 4B). Inhibition of hyperglycemia is important because it is associated with cardiovascular disorders, peripheral neuropathy, and diabetic retinopathy [26 –29]. The administration of rapid-acting insulin analogs 15–20 min before meals reportedly ameliorated postprandial hyperglycemia compared with administration immediately before meals in patients with type 1 diabetes [30, 31]. Slowing of gastric motility induced by HYA might be a useful proxy for the administration of bolus insulin 15–20 min before a meal. This might have possible application as a novel treatment for patients with type 1 diabetes whose postprandial hyperglycemia would be difficult to control.

Orally-administered HYA before OGTT slowed the elevation of blood glucose level after OGTT due to inhibiting glucose absorption via SGLT1 in small intestine mucosa and slowing the transportation of glucose from the stomach to the duodenum via promotion of GLP-1 and CCK secretion in normal rats. HYA also ameliorated postprandial hyperglycemia in type 1 diabetes model rats injected with bolus insulin before OGTT.

Below is the link to the electronic supplementary material.

Supplementary file1 (TIFF 896 KB) Effect of HYA on blood glucose levels after glucose load in dose-dependent manner and in rats not loading glucose. The postprandial hyperglycemia was ameliorated by HYA in a dose-dependent manner (A). The transition of blood glucose level without glucose load after oral HYA administration was not decreased (B). Values indicates mean ± SD. n = 4 to 6.