DPP-4 inhibition enhanced renal tubular and myocardial GLP-1 receptor expression decreased in CKD with myocardial infarction

All animal experiments were approved by the Institutional Animal Research Board of Chungbuk National University (CBNUA-1051-17-02). Male Sprague Dawley rats (n = 90) approximately 200 to 250 g weight (Daehan Biolink, Chungbuk, Korea) were randomized into 15 groups with 6 animals in each group: control, CKD 2, 4, and 8 weeks with/without linagliptin, MI-R 1, 3, and 7 days with/without linagliptin, CKD 8 weeks + MI-R with/without linagliptin. We established three experimental designs for CKD, acute MI-R, and CKD with MI-R (Fig. 1). Sample size was calculated by Mead’s resource equation, and we added animals against incidental peri-operative animal death. Blood pressure was measured via a femoral arterial pressure monitoring sensor (MLT844, Memscap AS, Skoppum, Norway), which was supported by a bridge amp (ML112, AD Instruments, Sydney, Australia) and Powerlab/4sp (ML750, AD Instruments). Urine was collected, with volume measured for a 24 h period. Creatinine clearance was assessed via a metabolic cage with mineral oil treatment on the day of organ harvesting (the last day). Blood samples were obtained via femoral venous sampling, centrifuged, and stored at − 80 °C. Body weight, serum glucose, creatinine, and albumin levels were measured using a Nova Stat Profile M critical care analyzer (Nova Biomedical Co., Waltham, MA, USA) and compared between groups. The pH of fresh urine was measured with a pH meter (Orion 3 star plus, Thermo, Beverly, MA, USA).

In all CKD groups, the lower and upper thirds of the left kidneys were resected, and after 1 week, the right kidney was removed (5/6 nephrectomy); a sham operation was performed in the control group. For general anesthesia, 50 mg/kg of tiletamine plus zolazepam (Zoletil) and 10 mg/kg of xylazine (Rompun) were mixed and injected into the thigh muscle. Rats were divided into seven groups: control, CKD 2, 4, 8 weeks with or without DPP-4 inhibitor-treated. All DPP-4 inhibitor-treated groups were orally administered 5 mg/kg of linagliptin (Trajenta, Boehringer-Ingelheim, Columbus, OH, USA) mixed in water, once daily for 8 weeks beginning the day after 5/6 nephrectomy.

Using the same anesthesia method described above, the left fourth intercostal space was dissected to expose the heart and an endotracheal catheter with rodent ventilator (model 683, Harvard Apparatus, South Natick, MA, USA) application was inserted. The left anterior descending branch of coronary artery (LAD) was then ligated with 6–0 silk for 30 min and released after ischemia (ischemia-reperfusion) via same method published by our college laboratory [13]. Acute GLP-1R changes among the groups were assessed 1, 3, and 7 days with or without DPP-4 inhibitor-treated after myocardial infarction. CKD-MI model were divided into five groups (n = 6 each): control, CKD 8 weeks, CKD 8 weeks with DPP-4 inhibitor (CKD-Lina), CKD 8 weeks with myocardial ischemia/reperfusion (CKD-MI), and CKD-MI with DPP-4 inhibitor (CKD-MI-Lina). For the myocardial infarction model, animals were treated with linagliptin for 8 weeks, after which the heart was harvested for immunoblotting and histochemical staining. The survival rate was analyzed by SPSS (SPSS 24.0; SPSS Inc., Chicago, IL, USA) and displayed by Kaplan-Meier survival curves. After the end of all experiments, animals were euthanized by ether inhalation without pain or stress which was based on the Institutional animal care and use committee (IACUC) standard operating guideline and approval (CBNUA-1051-17-02).

Rat kidney proximal tubule (NRK-52E) cells (American Type Culture Collection, Manassas, VA, USA) were cultured in Dulbecco’s Modified Eagle Medium (DMEM; Welgene, Kyeongbuk, Korea) supplemented with 5% fetal bovine serum (FBS; Gibco, Grand Island, NY, USA), 100 U/mL of penicillin, and 100 mg/mL of streptomycin (Gibco). Primary cultures of cardiomyocytes (CMCs) were isolated by modification of reported protocols [14‐16]. Obtained ventricles were washed three times with cold ADS buffer, chopped, and digested three times for 20 min with collagenase (0.5 mg/mL) and pancreatin (0.3 mg/mL). The resulting cells were collected and enriched by differential centrifugation through a discontinuous Percoll (Amersham Pharmacia Biotech, Piscataway, NJ, USA) gradient of densities 1.050, 1.062, and 1.082 g/mL [17]. The band which collected at the 1.062/1.082 density interface was collected and used as the source of CMCs. The CMCs were washed and suspended in DMEM supplemented with medium 199 (Welgene), 10% horse serum (Gibco), 5% FBS, 5000 U/L streptomycin, and 5000 U/L penicillin plus 10% (v/v), and seeded at a density of 5 × 10⁵ mononuclear cells/well on 2% gelatin (Sigma-Aldrich, St. Louis, MO, USA) coated on 60 mm Primarian 6-well plastic culture plates (Becton Dikenson and company, Hunt Valley, MD, USA). Media changes were initiated on day 1 via addition of serum free DMEM/M199 medium (4:1). NRK-52E cells and CMCs were incubated in a humidified atmosphere with 37 °C and 5% CO₂. For hypoxic conditions, NRK-52E cells cultured 2, 6, 12, 24, and 48 h and CMCs cultured for 1, 3 and 5 days were placed in a GasPakTM pouch system (Becton Dickinson, Franklin Lakes, NJ, USA) in conditions of 1% O₂ and 5% CO₂.

Serum DPP-4 enzyme activity was measured using the DPP-4 Activity Assay Kit (Biovision, Milpitas, CA, USA). Briefly, 50 μL of serum was diluted with 48 μL of DPP IV assay buffer and 2 μL substrate Gly-Pro-7-Amino-4-Methylcoumarin (AMC), and incubated at 37 °C for 30 min. The AMC density of the substrate was then measured by spectrophotometer at excitation and emission wavelengths of 360 and 460 nm, respectively.

Renal and myocardial GLP-1R expression was measured in renal cortex and heart tissue by WB using specific antibodies (Bioss, Boston, MA, USA) with β-actin (Sigma-Aldrich) and glyceraldehyde 3-phosphate dehydrogenase (GAPDH, Young-In Frontier, Seoul, Korea) as a loading control. Proteins were extracted with protein extraction solution (PRO-PREP, Intron, Korea) and measured by spectrophotometer. Samples were loaded in 10% polyacrylamide-sodium dodecyl sulfate mini gels and transferred to polyvinylidene fluoride (PVDF) membranes. Membranes were blocked for 2 h in tris-buffered saline-0.1% plus Tween20 (TBS-T) containing 5% non-fat dry milk, treated with primary antibodies against GLP-1R, phospho ERK1/2, total ERK1/2, Bcl-2 (Santa Cruz Biotechnology, Santa Cruz, CA, USA), and β-actin for 2 h in TBS-T, followed by the secondary antibody (goat anti-rabbit IgG-horseradish peroxidase; Santa Cruz Biotechnology, Santa Cruz, CA, USA). Immunohistochemical staining was used to assess changes in renal tubular expression between groups, and to assess the effects of renal impairment and DPP-4 inhibition on protein expression. WB band densities were quantified using the Multi Gauge 3.1 program and expressed as a percentage relative to the control group.

The kidneys and heart were harvested from all experimental and control groups. The renal cortex, outer medulla, and inner medulla were separated and fixed in 8% periodate-lysine-paraformaldehyde (PLP) solution for 8 h at room temperature, stored at 4 °C overnight, and embedded in paraffin. For IHC processing, tissues sections were rinsed in xylene to remove paraffin, then rehydrated using a gradient of 100 to 70% ethanol. Endogenous peroxidase activity was inhibited by treatment with 3% H₂O₂ at 4 °C for 45 min. Normal goat serum (Vector Laboratories, Inc., Burlingame, CA, USA) treated slides were exposed to primary antibodies at 4 °C overnight, then exposed to biotinylated goat anti-rabbit IgG (MACH2 Rabbit HRP Polymer; Concord, Biocare Medical, Burlingame, CA, USA) at room temperature for 30 min. After antibody exposure, heart tissue sections were treated with peroxidase 3, 3′-diaminobenzidine (DAB) substrate, and mounted after rinsing with xylene. These paraffin blocks were sectioned with 4 μm thickness and stained with Masson’s-Trichrome for the evaluation of extent of infarcted area.

Seventy-day-survival of the different animal groups were 100% (control, CKD and CKD-Li), and 50% (CKD-MI/R, CKD-MI/R-Li) (Fig. 2). Serum creatinine and urine output were significantly increased during CKD progression compared with the controls 2–8 weeks after 5/6 nephrectomy (p < 0.01); blood pressure was also increased (p < 0.01). Animals with 5/6 nephrectomy also showed reduced body weight during CKD progression, relative to sham-operated controls (p < 0.05), with no difference between CKD and linagliptin-treated CKD. Serum albumin was also decreased after CKD progression (p < 0.05). Blood pressure was shown to be increased after CKD development (78.0 ± 9.8 mmHg in control vs. 137.4 ± 10.0 mmHg in CKD at 2 weeks, p < 0.05), but not during CKD progression (CKD, 2 vs. 4 weeks, p = 0.291; 4 vs. 8 weeks, p = 0.758). Urine pH was elevated after 4 weeks of CKD, relative to controls (p < 0.01); however, these changes were not evident in early CKD (control vs. CKD, 2 weeks, p = 0.492; Table 1).

Serum DPP-4 level was increased in the CKD group, relative to control (2 weeks, 21.1 ± 4.8; 4 weeks, 22.0 ± 6.1; and 8 weeks, 26.4 ± 1.8; vs. 12.6 ± 3.1 for controls; p < 0.05). Although DPP-4 levels were highest in 8-week CKD animals, these differences were not statistically significant compared with earlier time points (CKD, 2 vs. 4 weeks, p = 0.808; CKD, 4 vs. 8 weeks, p = 0.226). Serum DPP-4 levels were markedly decreased following linagliptin treatment, with the largest differences seen at 8 weeks (26.4 ± 1.8 vs. 2.8 ± 1.8 for CKD and CKD-Lina 8 weeks, respectively; p < 0.01; Fig. 3). Differences between earlier time points were not statistically significant (2 vs. 4 weeks, p = 0.34).

Systolic blood pressure was increased after CKD development, though these changes were not evident during CKD progression. In contrast, blood pressure was significantly decreased at all three CKD time points after linagliptin treatment (p < 0.05; Fig. 4a). The volume of 24 h urine was markedly increased 2 weeks after 5/6 nephrectomy, with no differences evident between the 2 and 4 weeks points for either CKD or CKD with linagliptin treatment (CKD, 2 weeks vs. CKD-Lina, 2 weeks, p = 0.513; CKD, 4 weeks vs. CKD-Lina, 4 weeks, p = 0.274). However, this trend was not evident at later time points, with urine output significantly decreased in the CKD-Lina group 8 weeks after surgery (43.0 ± 12.3 in CKD, 8 weeks vs. 12.1 ± 0.1 mL/day in CKD-Lina, 8 weeks, p < 0.05; Fig. 4b). Although serum urea nitrogen levels exhibited no change after treatment (Fig. 4c), serum creatinine was significantly decreased both 4 and 8 weeks after surgery in linagliptin-treated animals relative to the CKD group (1.52 ± 0.06 in CKD, 4 weeks vs. 0.75 ± 0.06 in CKD-Lina, 4 weeks, p < 0.01; 1.59 ± 0.10 in CKD, 8 weeks vs. 0.66 ± 0.01 in CKD-Lina, 8 weeks, p < 0.01; Fig. 4d). Improvements in creatinine clearance were evident by 8 weeks of 5/6 nephrectomy (Fig. 4f). Serum albumin levels were also increased in the CKD-Lina group (p < 0.05; Fig. 4e).

Under hypoxic conditions, NRK-52E cellular GLP-1R expression was the highest 6 h after initiation of hypoxia, with decreases in expression thereafter. By 48 h post-hypoxia, GLP-1R levels had risen to 62% of the control (Fig. 5a). The density of the renal cortical GLP-1R band on WB was markedly increased 2 weeks after 5/6 nephrectomy; however, it decreased to control levels 4 weeks after surgery. Eight weeks after CKD onset, GLP-1R expression was 61% of that of the control group (Fig. 5b). IHC showed similar results to those of WB (Fig. 5c). The peak intensity of GLP-1R expression was noted 2 weeks post-CKD onset; expression started to decrease after 4 weeks and decreased markedly 8 weeks after surgery.

Linagliptin treatment enhanced GLP-1R expression in the renal cortex. GLP-1R was increased 2 weeks after CKD onset in the linagliptin treatment group, with incremental increases in GLP-1R density on WB evident at 4 and 8 weeks (Fig. 5b). Similar results were seen with IHC, with staining intensity increasing incrementally after linagliptin treatment during each stage of CKD (2 weeks, 154.4 ± 3.0 vs. 178.1 ± 0.1; 4 weeks, 97.0 ± 0.4 vs. 127.8 ± 7.4; 8 weeks, 61.0 ± 5.0 vs. 101 ± 13.8, without and with treatment, respectively; p < 0.05, relative to control; Fig. 6A). Linagliptin treatment significantly enhanced tubular GLP-1R expression, with higher levels of GLP-1R staining evident in the proximal tubular area (Fig. 6B), which recovered to control levels even at 8 weeks after CKD onset.

WB revealed significant decreases in renal tubular GLP-1R expression 8 weeks after CKD and CKD with myocardial ischemia. GLP-1R protein levels were significantly increased in both CKD and CKD with myocardial ischemia groups after linagliptin treatment (Fig. 7A-B). IHC analyses revealed similar results as WB, with increased GLP-1R expression in the renal cortical tubule after linagliptin treatment (Fig. 7C).

Cellular GLP-1R expression was decreased after acute ischemic damage to CMCs. After ischemic injury, WB analysis of CMCs revealed gradual decreases in GLP-1R over 72 h, though these changes were not statistically significant (Fig. 8A). IHC analysis of myocardial tissues showed increased GLP-1R in the ischemic area after LAD ligation, relative to control, though differences between days 1, 3, and 7 post-myocardial ischemia were not statistically significant (Fig. 8B). Ischemic myocardium showed significant wall thinning and fibrotic change by Masson’s trichrome stain 3 and 7 days after ischemia, which explained myocardial remodeling (Fig. 8C; d and f). Interestingly, this fibrotic area and wall thinning were markedly attenuated after linagliptin treatment (Fig. 8C; e and g).

WB of myocardial GLP-1R was significantly increased 8 weeks after CKD onset, with the strongest changes seen in CKD with myocardial ischemia. After linagliptin treatment, the WB density of GLP-1R was significantly increased in both the CKD and CKD with myocardial ischemia groups (Fig. 9A and B). IHC showed similar results as WB, with increased expression evident in the border of the ischemic area of the myocardium (Fig. 9C). Masson’s trichrome stain showed histological change increased ischemic area in myocardium by LAD ligation. Ischemic area was decreased by linagliptin treatment compared with CKD-MI/R group (Fig. 9D).

Renal cortical expression of Bcl-2 and phosphorylated ERK1/2 (p-ERK1/2) were all decreased in the CKD and CKD with MI/R rat, with these decreases markedly attenuated following linagliptin treatment (Fig. 10a-c).

Myocardial p-ERK1/2 and Bcl-2 expression were compared among control, CKD, CKD-Lina, and CKD with MI/R with and without linagliptin treatment. All groups exhibited decreased protein expression after 5/6 nephrectomy, in combination with increased expression after linagliptin treatment (Fig. 11a). The largest difference was seen in the CKD with MI/R group (Fig. 11b–c).