This was a randomized, single-blinded controlled interventional study, and our study protocol conformed to the ethical guidelines of the 1975 Declaration of Helsinki, as reflected by the prior approval of the institutional review board of The Ethic Committee of Kurume University. All experiments were performed in accordance with relevant guidelines and regulations. All subjects provided written informed consent to participate in this study.

A total of 5 male patients diagnosed with NAFLD were enrolled in this study. The following patient inclusion criteria were used: Patients with i) NAFLD, ii) age >20 years, and iii) written informed consent. The exclusion criteria were as follows: Patients with i) hemoglobin A1c >8.0%, ii) insulin or α-glucosidase inhibitor treatment, iii) liver cirrhosis, iv) renal failure, v) a history of cardiovascular disease, or vi) participation in any other clinical trial.

The subjects were randomly assigned into the sucrose (Control) or isomaltulose group. After a 14-day wash-out term, each subject was assigned to the other group.

NAFLD was diagnosed according to the Clinical Practice Guidelines for NAFLD/NASH as follows (25): i) hepatic steatosis evaluated by liver biopsy, ultrasonography, computed tomography, or magnetic resonance imaging; ii) ethanol intake <30 g/day; and iii) exclusion of other liver diseases, including chronic hepatitis B, chronic hepatitis C, autoimmune hepatitis, drug-induced liver disease, primary biliary cholangitis, primary sclerosing cholangitis, biliary obstruction, Wilson's disease, and hemochromatosis.

After a 12-hour overnight fast, the subjects consumed 20 g isomaltulose in 200 ml of water or 20 g sucrose (Control) in 200 ml of water in 1 min at 8:30 a.m. During the procedure, the subjects were instructed to rest on a bed and fast. Venous blood samples were collected before and 15 min after administration. For plasma GIP and GLP-1 measurements, blood samples were collected into a BD™ P100 Blood Collection System containing protein stabilizers (BD Biosciences, Franklin Lakes, NJ, USA). Blood samples were centrifuged at 3,000 × g for 15 min at 4°C and stored at −20°C until analysis.

Serum levels of aspartate aminotransferase (AST), alanine aminotransferase (ALT), lactate dehydrogenase (LDH), γ-glutamyl transpeptidase (GGT), total bilirubin, total cholesterol, low-density lipoprotein (LDL)-cholesterol, blood urea nitrogen (BUN), creatinine, nonesterified fatty acid, C-peptide immunoreactivity (CPR), and blood glucose level were measured using standard clinical methods (SRL Inc., Tokyo, Japan) as previously described (4,26,27).

As previously described (28), plasma GIP and GLP-1 levels were measured by an enzyme-linked immunosorbent assay using a Human GIP ELISA kit (YK253; Yanaihara Institute Inc., Fujimiya, Japan) and a Human GLP1 (7–36) ELISA kit (ab184857; Abcam, Cambridge, UK), respectively, according to each manufacturer's instructions.

Metabolomic analysis was performed twice using serum samples taken before (0 min) and 15 min after isomaltulose or sucrose treatment.

Serum samples taken before and 15 min after isomaltulose or sucrose treatment from a representative subject were applied to a metabolomic analysis. Metabolome measurements were performed at a service facility of LSI Medience Corporation (Tokyo, Japan). Briefly, serum (200 µl) was added to methanol (800 µl) and then mixed for 15 min with a shaker at room temperature. After it was centrifuged by 10,000 × g for 10 min, the supernatant was dried up with nitrogen gas, and the residue was dissolved with 10% acetonitrile aqueous solution (200 µl). After adding internal standards, they were analyzed with both liquid chromatography mass spectrometry (LC-MS) and capillary electrophoresis coupled to mass spectrometry (CE-MS). Tuning and calibration were performed with a standard solution provided by Agilent Technology, and the resolution errors were controlled within 3 ppm. The order of measurement was randomized to minimize the specific error in each group. Quality control samples were prepared by pooling samples and were analyzed every 5 samples to verify the measurement accuracy.

LC-MS datasets were acquired on a liquid chromatography system (Agilent HP1200; Agilent Technologies, Inc., Santa Clara, CA, USA) equipped with a C18 column (2 µm, 50×2.0 mm ID, CAPCELL PAK C18 IF; Shiseido, Tokyo, Japan) and coupled with an electrospray ionization quadrupole TOFMS (Agilent 6520; Agilent Technologies, Inc.). Solvent A was composed of a 5 mM ammonium acetate aqueous solution, while solvent B was acetonitrile. Metabolites were eluted at a flow rate of 0.2 ml/min at 40°C with a linear gradient of 10–90% solvent B over 10 min, followed by a further 5 min hold at 100% solvent B. The mass spectrometer was operated in positive and negative scan mode (m/z 60 to 1,200) with a capillary voltage of 3,500 V. The nebulizing gas pressure was 40 psi, and the dry gas flow was 8 l/min at 350°C.

Ionic metabolites were measured in the positive mode of CE-TOFMS (Agilent CE-TOFMS 6520; Agilent Technologies, Inc.,). Metabolites were separated in a fused-silica capillary (50 µm i.d.x100 cm total length; GL Science, Tokyo, Japan) filled with 1 mol/l formic acid aqueous solution (cation mode), or 20 mM ammonium formate and 20 mM ammonium acetate aqueous solution (pH 10, anion mode) as an electrolyte. The sample solution was injected at 5 kPa for 15 sec (approximately 15 nl), and a voltage of 30 kV was applied. The temperatures of capillary and sample trays were maintained at room temperature and 5°C, respectively. The sheath liquid was methanol/water (50%v/v) containing 5 mM ammonium acetate. CE-TOFMS was operated in both positive and negative scan mode (m/z 60 to 1,200). The capillary voltage was set at 3,500 V, and the nitrogen gas (heater temperature 250°C) flow rate was set at 10 l/min.

A data file of mass spectrometry was converted to csv format with an Agilent csv converter. All peak positions (retention time and m/z) and areas were calculated by Markeranalysis (LSI Medience, Tokyo, Japan) (29). All peak areas were aligned into one data sheet, and the errors of peak intensities were corrected by internal standards. Noise peaks were deleted compared with the peaks detected in blank samples. Metabolites were identified by comparing the retention time and m/z with standard data set established by LSI Medience Corporation.

Insulin resistance was evaluated by an increment of serum CPR as previously described (30). ΔCPR was the change of serum CPR level from baseline to 15 min after sucrose or isomaltulose administration. Similarly, changes in GIP and GLP-1 secretions were evaluated by ΔGIP and ΔGLP-1, which were changes of plasma GIP and GLP-1 levels from baseline to 15 min after sucrose or isomaltulose administration, respectively. The effects of isomaltulose on metabolites were also evaluated by change of each metabolite level from baseline (0 min) to 15 min after sucrose or isomaltulose administration (Δ each metabolite). Variable that increased or decreased after the administration was shown as positive or negative expression level (∆ variable or -∆ variable) (Fig. 1).

Data are expressed as numbers or means ± standard deviations. Differences between the two groups were analyzed using the Wilcoxon signed-rank test. In metabolome analysis, mea n fold-change and t-test for all detected peaks were performed between the two groups. P<0.05 was considered to indicate a statistically significant difference.

The baseline patient characteristics are summarized in Table I. No significant difference was seen in the serum levels of AST, ALT, GGT, or total bilirubin between the sucrose (Control) and Isomaltulose groups. There was no significant difference in fasting blood glucose or serum CPR levels between the 2 groups (Table I).

The effect of isomaltulose on blood glucose and serum CPR levels was evaluated by changes in these variables from baseline to 15 min after sucrose or isomaltulose administration. Although there was no significant difference in changes of blood glucose levels, the serum CPR level was significantly decreased in the Isomaltulose group compared to that in the Control group (Fig. 2).

The effect of isomaltulose on plasma GIP and GLP-1 levels was evaluated by changes in these variables from baseline to 15 min after sucrose or isomaltulose administration. There was no significant difference in changes of plasma GIP and GLP-1 levels between the Control and Isomaltulose groups (Table II).

With metabolomic analysis, the effects of isomaltulose on 201 metabolite levels were evaluated by changes in these variables from baseline to 15 min after sucrose or isomaltulose administration. A significant alteration was seen in 52 metabolites between the Control and Isomaltulose groups (Table III). Many of these 52 altered metabolites were categorized as bile acid (6 metabolites), fatty acid (6 metabolites), or glycine/serine metabolism (4 metabolites) (Fig. 3).

A representative change of metabolite in each category was an increased taurodeoxycholic acid level in the bile acid category, a decreased arachidonic acid level in the fatty acid category, and an increased betaine level in the glycine/serine category after isomaltulose treatment (Fig. 4A-C). In addition, a significant decrease was seen in the serum L-arginine level in the isomaltulose group compared to that in the Control group (Fig. 4D), while the serum L-ornithine level was significantly increased in the isomaltulose group compared to that in the Control group (Fig. 4E).