Adult male C57BL/6J mice (age, 6 weeks; weight, 15–18 g) were obtained from Qingdao Institute Food and Drug Control and housed in standard rodent cages. Animals were housed in a temperature-controlled (23±2°C) and humidity-controlled (55±20%) animal room (illumination from 7:00 to 19:00) with free access to standard food and tap water for at least 1 week to adapt to their surroundings. The experimental protocols were approved by the Qingdao University Animal Care and Use Committee and Animal Welfare Committee (approval no. QYFY WzLL26594).

After 1 week of acclimatization, the 7-week-old mice were fed a high-fat-diet (HFD) consisting of 59% basic mice feed, 20% sugar, 18% lard and 3% egg yolk.

The 24 adult male C57BL/6J mice were randomly divided into the following four groups: i) HFD (n=4); ii) HFD+HBOT (n=4); iii) T2DM (n=8); and iv) T2DM+HBOT (n=8). Following the contrast principle and the single variable principle, the T2DM and T2DM+HBOT groups were used as the experimental groups, whereas the HFD and HFD+HBOT groups were used as the control groups. As the expected success rate of T2DM modeling was unknown at first, the sample size of the T2DM group was doubled in order to reduce sampling error and improve statistical efficiency. As the results showed that the rate of successful T2DM model establishment was 100%, the sample sizes of the HFD and T2DM groups were 4 and 8, respectively.

From the 7th week, all mice were fed a HFD until sacrifice. HFD feeding was combined with a subsequent injection of low-dose streptozotocin (STZ, 60 mg/kg; i.p.; cat. no. S0130; Sigma-Aldrich; Merck KGaA) to model T2DM. Young adult mice are fed a HFD to elicit insulin resistance, multiple injections with low-dose STZ then elicit partial loss of β-cells, which results in hypoinsulinemia and hyperglycemia (15). T2DM mice were intraperitoneally injected with low-dose STZ once daily for 3 days after 1-week HFD feeding. The age-matched HFD mice were injected with the same frequency citrate buffer as STZ alone (15,16). Mice with fasting blood glucose levels >12 mmol/l were defined as successful T2DM models (13).

At the 14th week, HFD+HBOT and T2DM+HBOT groups received 1-h HBOT daily from 5:00 to 6:00 p.m. for 7 days, whereas the other two groups received normobaric oxygen in the same chamber. HBOT was administered in an animal HBO chamber (Moon Environmental Technology Co., Ltd.), operating at high pressure (2 ATA) with 100% oxygen, including a 5-min pressure rise adaptation stage, 50-min stabilization stage and 5-min re-pressurization stage (17). After the final HBOT, an intraperitoneal glucose tolerance test (IPGTT) was performed. Subsequently, the mice were sacrificed by exsanguination following an intraperitoneal injection of 50 mg/kg sodium pentobarbital. Blood samples were obtained from the orbital sinus and stored at −20°C or −80°C until further use (18,19).

As mice are primarily nocturnal and their feeding activity typically occurs during the night, nocturnal food intake was assessed (20). During HBOT, the 12-h nocturnal food intake was measured daily to investigate the effects of HBOT on nocturnal food intake (20). An electronic precision scale (TE412-L; Sartorius AG) was used to weigh the food at 8:00 a.m and 8:00 p.m. The 12-h nocturnal food intake was defined as the weight of the food at 8:00 p.m. minus the weight of the food at 8:00 a.m. the following morning. An electric balance (PL1501-S; Mettler Toledo) was used to weigh the mice at 8:00 a.m. after weighing the food.

To determine the success of T2DM model establishment, non-fasting blood glucose (non-FBG) was measured after 42 days of STZ injection by a tail vein prick and a glucometer.

Before HBOT, fasting blood glucose (FBG) levels were measured in the four groups and then FBG levels were assessed again after 7-day HBOT to observe the effects of HBOT on FBG.

FBG was measured at 9:00 a.m. after a 12-h fast and non-FBG was measured at 2:00 p.m. without fasting.

The IPGTT was performed in the T2DM and T2DM+HBOT groups after 7-day HBOT. The mice were allowed to fast for 14 h before the IPGTT and then intraperitoneally administered glucose (2 g/kg body weight). Using a glucometer, blood glucose concentrations were measured at 0, 15, 30, 60, 90 and 120 min after IPGTT using tail vein blood.

Fasting insulin (FI) levels of the blood samples obtained from sacrificed mice were measured using an ultrasensitive ELISA Kit (cat. no. CEA448Ra, Wuhan USCN Business Co., Ltd.). The homeostasis model assessment for insulin resistance and β-cell function (HOMA-IR and HOMA-β, respectively) indices were also measured (21). The HOMA-IR was calculated as follows: FI (mU/ml) × FBG (mmol/l)/22.5. The HOMA-β was calculated as follows: 20 × FI/(FBG-3.5) (22).

Electronic precision scales were used to measure fat mass, including inguinal adipose tissue, epididymal adipose tissue and pancreas weight. The sum of the inguinal white adipose tissue (iWAT) weight and epididymal white adipose tissue (eWAT) weight was calculated to assess the total weight of the fat pad, reflecting the balance of fat accumulation and metabolism. The current study calculated the pancreas weight coefficient as follows: Pancreas weight/body weight on the 8th day.

Pancreatic and liver tissues were fixed in 4% formaldehyde at 4°C for 24 h. After dehydration with various ethanol concentrations (75, 80, 85, 95 I, 95 II, 100 I and 100% II, 5 min each at room temperature) and clarifying with xylene (xylene I and xylene II, 10 min each at room temperature), the tissues were embedded in soft paraffin, hard paraffin and mixed paraffin (1 h for each) before sectioning, the temperature was 2–3°C above the melting point, at last the tissues were solidified in the mixed paraffin at room temperature. After sectioning into 4-6-µm sections using a paraffin slicing machine (RM2016; Leica Microsystems, Inc.), the tissue sections were stained using a hematoxylin-eosin staining kit (cat. no. C0105; Beyotime Institute of Biotechnology). The light microscope (CX31; Olympus Corporation) was used to observe staining results. Cell numbers were counted using ImageJ software (version 1.8.0; National Institutes of Health).

Paraffin-embedded sections (thickness, 4-6-µm) were used to determine the effects of HBOT on insulin resistance. Pancreatic tissues were fixed in 4% formaldehyde at 4°C for 24 h. After dehydration with various ethanol concentrations (5 min each at room temperature) and clarifying with xylene (xylene I and xylene II, 10 min each at room temperature), the tissues were embedded in soft paraffin, hard paraffin and mixed paraffin (1 h for each) before sectioning, the temperature was 2–3°C above the melting point, lastly the tissues were solidified in the mixed paraffin at room temperature. After sectioning, the tissues were dewaxed by xylene for (xylene I and xylene II) 10 min each at room temperature and placed in different ethanol concentrations (100 I, 100 II, 95 I, 95 II, 85, 80 and 75%) for 5 min each at room temperature. Following antigen retrieval by heating at 95°C in citrate buffer for 1 h, the sections were quenched with 0.3% hydrogen peroxide at room temperature for 30 min. After three washes in 0.01 M PBS, the sections were blocked with 1% BSA (cat. no. 1213G057; Beijing Solarbio Science & Technology Co., Ltd.) in 0.01 M PBS for 1 h at 37°C. Subsequently, the sections were incubated with a rabbit anti-insulin primary antibody (1:200; cat. no. 4590; Cell Signaling Technology, Inc.) overnight at 4°C. The sections were then incubated with a HRP-conjugated secondary antibody (cat. no. ZB2301; Zsbio; http://www.zsbio.com/product/ZB-2301) for 3 h at 37°C. Finally, 3,3′diaminobenzidine (DAB) staining was performed using a kit (cat. no. ZLI-9018; Zsbio) according to the manufacturer's protocol. The nuclei were stained for 30 sec to 1 min, then the sections were rinsed with running water for 10 min. Differentiation: 70, 90, 100 I, 100% II ethanol, xylene I and xylene II, 2 min each at room temperature. The slides were removed from xylene and neutral gum was used to seal the slides. The morphology was assessed using a CX31 light microscope (Olympus Corporation). The β-cell area ratio was assessed using Image-Pro Plus software (Version 6.0, Media Cybernetics, Inc.). The β-cell mass was calculated by the β-cell area ratio and pancreas weight (23,24): β-cell mass=β-cell area ratio × pancreas weight.

Paraffin-embedded sections (thickness, 4-6-µm) were used to determine the effects of HBOT on β-cell apoptosis. Pancreatic tissues were fixed in 4% formaldehyde at 4°C for 24 h. After dehydration with various ethanol concentrations (5 min each at room temperature) and clarifying with xylene (xylene I and xylene II, 10 min each at room temperature), the tissues were embedded in soft paraffin, hard paraffin and mixed paraffin (1 h for each) before sectioning, the temperature was 2–3°C above the melting point, lastly the tissues were solidified in the mixed paraffin at room temperature. After sectioning, the tissues were dewaxed by xylene (xylene I and xylene II) for 20 min each and placed in different ethanol concentrations (100 I, 100 II, 95 I, 95 II, 85, 80 and 75%) for 5 min each at room temperature, then the tissues were incubated in 0.1% Triton for 10 min. After four washes with PBS (2 min per wash), the sections were incubated with protease K (20 µg/ml in 10 mM Tris-HCl, pH 8.0) and digested for 15 min. Subsequently, the sections were incubated in labeling buffer for 2 h at 37°C. After rinsing three times with PBS, the sections were stained using a TUNEL Cell Apoptosis Detection kit III (FITC, cat. no. MK1023; Boster Biological Technology) according to the manufacturer's protocol. Sections were also immunostained with anti-insulin antibody to identify β-cells, they were incubated with a rabbit anti-insulin primary antibody (1:200; cat. no. 4590; Cell Signaling Technology, Inc.) overnight at 4°C and incubated with a HRP-conjugated secondary antibody (cat. no. ZB2301; Zsbio) for 3 h at 37°C. The slides were rinsed in 0.01 M PBS buffer and mounted in fluorescent mounting medium (Dako; Agilent Technologies, Inc.). Differentiation: 70, 90, 100 I, 100% II ethanol, xylene I and xylene II, 2 min each at room temperature. The slides were removed from xylene and neutral gum was used to seal the slides. The stained sections were visualized using a fluorescence microscope and apoptotic cell numbers were counted using ImageJ software (version 1.8.0; National Institutes of Health). In total, 20 islets were chosen at random from each group of individual mice.

Total protein was isolated from pancreatic tissues using RIPA buffer (cat. no. P0013B; Beyotime Institute of Biotechnology) supplemented with protease inhibitors (1:100; cat. no. P1005; Beyotime Institute of Biotechnology). After centrifugation at 14,000 × g for 10 min (4°C), protein concentrations were determined using a BCA assay (cat. no. P0012; Beyotime Institute of Biotechnology). Proteins were denatured in sodium dodecyl sulfate sampling buffer (95°C for 5 min), then separated using 12% polyacrylamide gel electrophoresis. After transferring to PVDF membranes (cat. no. IPVH00010, MilliporeSigma) over 2 h, the membranes were blocked with 5% FBS (cat. no. 9048-46-8, Beijing Solarbio Science & Technology Co., Ltd.) for 2 h at room temperature. The mass of protein was 30 µg/12 µl per lane. Subsequently, the membranes were incubated at 4°C overnight with primary antibodies targeted against the following: Bax (rabbit IgG; cat. no. 2772; 1:2,000; Cell Signaling Technology, Inc.), Bcl-2 (rabbit IgG; cat. no. ab196495; 1:2,000; Abcam), caspase-3 (Casp-3, rabbit IgG; cat. no. ab184787; 1:2,000; Abcam), cleaved Casp-3 (rabbit IgG; cat. no. ab214430; 1:2,000; Abcam), poly(ADP-ribose) polymerase (PARP; rabbit IgG; cat. no. 9532; 1:2,000; Cell Signaling Technology, Inc.), cleaved PARP (rabbit IgG; cat. no. 5625; 1:2,000; Cell Signaling Technology, Inc.) and β-actin (rabbit IgG; cat. no. 4967; 1:4,000; Cell Signaling Technology, Inc.). The membranes were then incubated with a goat anti-rabbit IgG H&L HRP-conjugated secondary antibody (cat. no. ab6721; 1:2,000; Abcam) at room temperature for 1 h. Protein bands were visualized using Immobilon western chemiluminescent substrate (cat. no. WBKLS0100; 200 µl; MilliporeSigma) and a UVP 810 gel-imager (Analytik Jena AG). Protein expression was semi-quantified using ImageJ software (version 1.8.0, National Institutes of Health).

The experiments were repeated >2 times. Data are presented as the mean ± SEM. Comparisons between two groups were analyzed using the unpaired Student's t-test. Comparisons among multiple groups were analyzed using one-way ANOVA followed by Tukey's post hoc test. GraphPad Prism (GraphPad Software, Inc, version 7.0.) and SPSS (SPSS Inc, version 22.0) software were used to create graphs and perform statistical analyses. P<0.05 was considered to indicate a statistically significant difference.

Non-FBG was measured for 2 consecutive days after 42 days injecting low-dose STZ (Fig. 1A). FBG was measured before the HBO intervention (Fig. 1B). Both the non-FBG (22.81±1.88 vs. 12.23±0.73 mmol/l; P<0.01) and FBG (12.30±1.14 vs. 7.71±0.79 mmol/l; P<0.01) levels of the T2DM group were significantly higher compared with those in the HFD group, which indicated successful establishment of the T2DM model. No mice died during the model establishment and the successful T2DM model establishment rate was 100%.

During HBOT, all mice were fed separately for 7 days and their 12-h nocturnal feeding and body weight were assessed daily. There were no marked differences among the HBOT groups in 12-h nocturnal food intake (P>0.05; Fig. 2A). The body weight appeared to increase at first, then decrease; however, there were no significant differences among the four groups (P>0.05; Fig. 2B). Body weight gain after HBO intervention was determined on the 8th day, and the results indicated that HBOT did not reduce body weight gain in both the HFD (1.292±0.021 vs. 0.922±0.19 g; P>0.05) and T2DM (1.164±0.102 vs. 0.891±0.059 g; P>0.05) groups (Fig. 2C).

HBOT notably reduced FBG in T2DM mice (11.63±1.26 vs. 8.11±0.74 mmol/l; P>0.05; Fig. 3A). The glucose tolerance tests showed that HBOT significantly reduced peak blood glucose levels at 15 min (21.48±2.88 vs. 14.26±2.51; P<0.05; Fig. 3B) and the area under the blood glucose curve within 120 min of IPGTT (1,628±185 vs. 1,425±72; P<0.05; Fig. 3C) in T2DM mice, suggesting that HBOT reduced FBG levels and improved glucose tolerance in T2DM.

To assess insulin resistance and β-cell function, HOMA-IR and HOMA-β were calculated. The HOMA-IR of T2DM mice was significantly higher compared with that of HFD mice (8.98±0.55 vs. 2.96±0.27; P<0.001). There were no significant differences between the two HFD groups (2.96±0.272 vs. 3.25±0.12; P>0.05). Compared with the T2DM group, the HOMA-IR was significantly reduced in the T2DM+HBOT group (8.98±0.55 vs. 6.17±0.21; P<0.01; Fig. 3D). The HOMA-β of T2DM mice was significantly lower compared with that in the HFD group (0.362±0.84 vs. 1.084±0.102; P<0.01). Moreover, there was no significant difference between the two HFD groups or between the two T2DM groups (P>0.05; Fig. 3E).

Visceral adipose tissue and subcutaneous adipose tissue are considered as distinct types of white fat. Typically, the inguinal fat weight is measured to reflect subcutaneous adipose tissue and the epididymal fat weight is measured to reflect visceral adipose tissue (25,26). Therefore, the weights of visceral epididymal fat and inguinal subcutaneous fat were measured in the present study.

The weight of the inguinal subcutaneous adipose tissue was significantly lower in the T2DM+HBOT group compared with that in the T2DM group (0.311±0.019 vs. 0.451±0.034 g; P<0.01; (Fig. 4A), although there was no significant difference in the weight of the visceral epididymal adipose tissue between these two T2DM groups (P>0.05; Fig. 4B). HBOT significantly decreased the sum of iWAT and eWAT weight in T2DM mice (0.592±0.049 vs. 0.856±0.067 g; P<0.05; Fig. 4C). Compared with the T2DM group, the pancreas weight (0.206±0.012 vs. 0.143±0.004 g; P<0.01) and the pancreas weight coefficient were significantly increased in the T2DM+HBOT group (0.849±0.053 vs. 0.582±0.026; P<0.01; Fig. 4D and E).

The islets were divided into large and small islets; those with >10 islet cells were defined as large islets and those with ≤10 cells as small islets.

H&E staining showed that the number of small islets was higher, whereas the number of large islets was lower in the T2DM group compared with the HFD group. After HBOT, the number of large islets increased significantly in T2DM mice (0.52±0.016 vs. 0.42±0.019; P<0.01; Fig. 5A).

The β-cell area in the T2DM group was significantly smaller compared with that in the HFD (3.77±0.38 vs. 6.097±0.88; P<0.05) and HFD+HBOT (3.77±0.38 vs. 8.77±1.18; P<0.05) groups. After HBOT, the β-cell area in T2DM mice was significantly increased (3.77±0.38 vs. 7.68±1.68; P<0.05), albeit smaller compared with that in the HFD and HFD+HBOT groups (Fig. 5B). In the T2DM mice, the β-cell mass was significantly lower in pancreatic tissue compared with the HFD group (0.0103±0.0019 vs. 0.005463±0.000691; P<0.05). However, HBOT significantly increased β-cell mass by ~2-fold without HBOT in T2DM mice (0.005463±0.000691 vs. 0.015201±0.001261; P<0.01; Fig. 5C).

The TUNEL results showed that HBOT significantly reduced the β-cell apoptotic rate in T2DM mice (0.3588±0.01237 vs. 0.1550±0.00898; P<0.001; Fig. 6A).

To further investigate the mechanism underlying HBO-mediated inhibition of β-cell apoptosis, the expression levels of the apoptosis-related proteins Bax/Bcl-2, caspase-3 and downstream PARP were measured in pancreatic tissue. After HBOT, the expression levels of Bax were markedly decreased, whereas Bcl-2 expression levels were notably increased, resulting in a significantly decreased Bax/Bcl-2 ratio in T2DM mice. The expression levels of cleaved caspase-3 and cleaved PARP were significantly lower in the T2DM+HBOT group compared with those in the T2DM group (Fig. 6B).

H&E staining showed that the arrangement cords of hepatocytes appeared more orderly after HBOT, and there were fewer intracytoplasmic vacuoles and inflammatory cells. The nuclei were larger and located in the center, and there was less shrinkage and necrosis than in the T2DM group (Fig. 7A). HBOT increased hepatic glycogen storage in T2DM mice (Fig. 7B).