β-Arrestin-2 enhances endoplasmic reticulum stress-induced glomerular endothelial cell injury by activating transcription factor 6 in diabetic nephropathy

Figure 1 β-Arrestin-2 is increased significantly in renal biopsies from patients with diabetic nephropathy and glomerular endothelial cells from mice with diabetic nephropathy. A: Images of immunohistochemical (IHC) staining to detect the expression of β-arrestin-2 in paraffin sections of human renal tissue from normal controls and patients with diabetic nephropathy (DN) (black bars = 10 μm, n = 5); B: Relative mRNA levels of β-arrestin-2 in the renal cortex of DN mice (mean ± SD, ^cP < 0.001 vs control, n = 8); C: The expression of β-arrestin-2 in the renal cortex of DN mice was analyzed by immunoblotting (^cP < 0.001 vs control, n = 8); D: Representative images of IHC staining to detect the expression of β-arrestin-2 in paraffin sections of the kidneys from control and DN mice (red bars = 20 μm, n = 8); E: Detection of β-arrestin-2 expression in glomerular endothelial cells (GENCs) in streptozotocin (STZ)-induced DN mouse model by immunofluorescence double labeling: Endothelin (red, mark protein in GENCs) and β-arrestin-2 (green) (white bars = 50 μm, n = 8); F: Quantifications showing expression of β-arrestin-2 in the kidneys from different groups of mice (^bP < 0.01 vs control, n = 8).

Figure 2 β-Arrestin-2 is upregulated in glomerular endothelial cells under high glucose levels and other stimuli in vitro. A: Representative images and summarized data showing increased expression of β-arrestin-2 in glomerular endothelial cells (GENCs) with high glucose (HG) (concentrations of 20 and 40 mmol/L; 40 mmol/L mannitol was used as osmolarity control) stimulation for 24 h by Western blotting (^aP < 0.05 vs control, ^bP < 0.01 vs control, n = 6); B: Representative images and summarized data showing upregulation of β-arrestin-2 in GENCs treated with advanced glycation end products (0, 50, 100, and 200 μg/mL) for 24 h by immunoblotting (^bP < 0.01 vs control, ^cP < 0.001 vs control, n = 6); C: Representative images and summarized data showing increased expression of β-arrestin-2 in GENCs treated with tumor necrosis factor-α (0, 20, 40, and 80 ng/mL) for 24 h by Western blotting (^aP < 0.05 vs control, ^bP < 0.01 vs control, n = 6); D: Representative images and summarized data showing increased β-arrestin-2 expression in angiotensin II (Ang II: 0, 10^-7, 10^-6, and 10^-5 mol/L) treated GENCs for 24 h by immunoblotting (^aP < 0.05 vs control, ^bP < 0.01 vs control, n = 6); E: Images and summarized data reflecting upregulation of β-arrestin-2 in GENCs treated with transforming growth factor-β1 (0, 2, 4, and 8 ng/mL) for 24 h by Western blotting (^aP < 0.05 vs control, ^bP < 0.01 vs control, n = 6). TGF-β1: Transforming growth factor-β1; HG: High glucose; AGE: Advanced glycation end product; TNF-α: Tumor necrosis factor-α.

Figure 3 Knockdown of β-arrestin-2 by siRNA ameliorates high glucose induced glomerular endothelial cell injury and apoptosis. A: Images showing efficiency of silencing β-arrestin-2 with small interfering RNA (siRNA) by Western blotting (^aP < 0.05 vs control, n = 6); B: Representative images showing effect of β-arrestin-2 knockdown on expression of occludin, ZO-1, and eNOS in high glucose (HG)-treated glomerular endothelial cell (GENCs) by Western blotting (^bP < 0.01 vs scramble/control, ^dP < 0.05 vs scramble/HG treatment, n = 6); C: Immunoblotting images showing effect of β-arrestin-2 knockdown on expression of apoptosis related proteins Bcl-2 and Bax in HG-treated GENCs (^aP < 0.05 vs scramble/control, ^cP < 0.001 vs scramble/control, ^dP < 0.05 vs scramble/HG treatment, n = 6); D: Immunoblotting images showing effect of β-arrestin-2 knockdown on expression of cleaved caspase 3 and caspase 3 in GENCs with HG treatment (^cP < 0.001 vs scramble/control, ^eP < 0.01 vs scramble/HG treatment, n = 6); E: Flow cytometric images showing decreased apoptosis of GENCs with HG treatment by knockdown of β-arrestin-2; F: Summarized flow cytometric data showing reduced apoptosis of GENCs with HG treatment by silencing β-arrestin-2 (^bP < 0.01 vs scramble/control, ^dP < 0.05 vs scramble/HG treatment, n = 6). HG: High glucose; siRNA: Small interfering RNA.

Figure 4 Knockdown of β-arrestin-2 reduces endoplasmic reticulum stress and apoptosis of glomerular endothelial cells in high glucose treatments by decreasing expression of activating transcription factor 6. A: Representative images and summarized data showing that knockdown of β-arrestin-2 inactivated endoplasmic reticulum (ER) stress by downregulating expression of BiP and CHOP in HG-treated glomerular endothelial cells (GENCs) by Western blotting (^aP < 0.05 vs scramble/control, ^bP < 0.01 vs scramble/control, ^dP < 0.05 vs scramble/HG treatment, n = 6); B: Flow cytometric analysis showing that the ER stress activators tunicamycin (TM) and thapsigargin (TG) induced GENC apoptosis and the inhibitor 4-phenyl butyric acid (4-PBA) reduced GENC apoptosis (^bP < 0.01 vs scramble/control, ^fP < 0.05 vs TM/TG treatment, n = 5); C: Summarized flow cytometric data showing apoptosis of GENCs under different stimuli (^aP < 0.05 vs scramble/control, ^bP < 0.01 vs scramble/control, ^dP < 0.05 vs scramble/HG treatment, ^fP < 0.05 vs TM/TG treatment, n = 4); D: Western blotting images showing effects of β-arrestin-2 knockdown on expression of ATF6 in GENCs under HG treatment (^bP < 0.01 vs scramble/control, ^dP < 0.05 vs scramble/HG treatment, n = 6). HG: High glucose; siRNA: Small interfering RNA; TM: Tunicamycin; TG: Thapsigargin; 4-PBA: 4-Phenyl butyric acid.

Figure 5 Overexpression of β-Arrestin-2 aggravates glomerular endothelial cell injury and apoptosis by activating endoplasmic reticulum stress through upregulation of activating transcription factor 6. A: Representative images showing that overexpression of β-arrestin-2 enhanced expression of BiP and CHOP in glomerular endothelial cells (GENCs), which could be blocked by knockdown of activating transcription factor 6 (ATF6) (^aP < 0.05 vs scramble/control, ^gP < 0.05 vs pCDNA-β-arrestin-2 treatment, n = 6); B: Representative images showing that overexpression of β-arrestin-2 reduced expression of ZO-1 and Occludin in GENCs, which could be recovered by knockdown of ATF6 in GENCs (^aP < 0.05 vs scramble/control, ^bP < 0.01 vs scramble/control, ^hP < 0.01 vs pCDNA-β-arrestin-2 treatment, n = 6); C: Representative images showing that overexpression of β-arrestin-2 promoted apoptosis related protein expression in GENCs, which was blocked by knockdown of ATF6 (^aP < 0.05 vs scramble/control, ^bP < 0.01 vs scramble/control, ^gP < 0.05 vs pCDNA-β-arrestin-2 treatment, n = 6); D: Summarized flow cytometric data showing that overexpression of β-arrestin-2 aggravated apoptosis of GENCs, which was inhibited by knockdown of ATF6 (^aP < 0.05 vs scramble/control, ^gP < 0.05 vs pCDNA-β-arrestin-2 treatment, n = 6); E: Representative Western blot images showing that overexpression of β-arrestin-2 upregulated expression of ATF6, but knockdown of ATF6 did not affect expression of β-arrestin-2 (^aP < 0.05 vs scramble/control, ^bP < 0.01 vs scramble/control, ^gP < 0.05 vs pCDNA-β-arrestin-2 treatment, n = 6). ATF6: Activating transcription factor 6.

Figure 6 β-Arrestin-2 enhances expression of activating transcription factor 6 and promotes activating transcription factor 6 translocation into the nucleus of glomerular endothelial cells treated with high glucose. A: Representative images and summarized data showing that knockdown of β-arrestin-2 reduced protein levels of BiP and CHOP, which was reversed by overexpression of activating transcription factor 6 (ATF6) [^aP < 0.05 vs scramble/control, ^bP < 0.01 vs scramble/control, ^dP < 0.05 vs scramble/high glucose (HG) treatment, ⁱP < 0.05 vs Si-β-arrestin-2/HG treatment]; B: Representative images and summarized data showing that knockdown of β-arrestin-2 recovered protein levels of ZO-1 and Occludin, which was reversed by overexpression of ATF6 (^aP < 0.05 vs scramble/control, ^dP < 0.05 vs scramble/HG treatment, ⁱP < 0.05 vs Si-β-arrestin-2/HG treatment); C: Flow cytometric analysis showing that knockdown of β-arrestin-2 reduced apoptosis of glomerular endothelial cells treated with HG was reversed by overexpression of ATF6; D: Representative images and summarized data showing that knockdown of β-arrestin-2 reduced apoptosis related protein expression, which was reversed by overexpression of ATF6. Silencing of β-arrestin-2 decreased expression of ATF6, but overexpression of ATF6 did not change expression of β-arrestin-2 (^aP < 0.05 vs scramble/control, ^dP < 0.05 vs scramble/HG treatment, ⁱP < 0.05 vs Si-β-arrestin-2/HG treatment); E: Representative Western blot images and summarized data showing expression of cleaved ATF6 in the nucleus of cells under different stimuli (^bP < 0.01 vs scramble/control, ^cP < 0.001 vs scramble/control, ^dP < 0.05 vs scramble/HG treatment, ^gP < 0.05 vs pCDNA-β-arrestin-2 treatment, ⁱP < 0.05 vs Si-β-arrestin-2/HG treatment); F: Relative mRNA levels of GRP78 (^aP < 0.05 vs scramble/control, ^bP < 0.01 vs scramble/control, ^dP < 0.05 vs scramble/HG treatment, ^fP < 0.05 vs TM/TG treatment, ^gP < 0.05 vs pCDNA-β-arrestin-2 treatment, ⁱP < 0.05 vs Si-β-arrestin-2/HG treatment). ATF6: Activating transcription factor 6; HG: High glucose; TM: Tunicamycin; 4-PBA: 4-Phenyl butyric acid.

Figure 7 Adeno-associated virus injection induced gene silencing of β-arrestin-2 ameliorates renal injury in streptozotocin-induced diabetic nephropathy mice. A: Relative mRNA levels of β-arrestin-2 in the renal cortex of mice after adeno-associated virus (AAV)-null (scramble) and AAV-shRNA-β-arrestin-2 (shRNA-β-arrestin-2) tail injection (mean ± SD, ^aP < 0.05 vs scramble/control, ^bP < 0.01 vs scramble/control, ^dP < 0.05 vs scramble/streptozotocin (STZ) treatment, n = 8); B: Western blot images showing protein level of β-arrestin-2 in the renal cortex of mice after AAV-null (scramble) and AAV-shRNA-β-arrestin-2 (shRNA-β-arrestin-2) tail injection (mean ± SD, ^aP < 0.05 vs scramble/control, ^bP < 0.01 vs scramble/control, ^dP < 0.05 vs scramble/STZ treatment, n = 8); C: Urinary albumin-to-creatinine ratio (UACR) in different groups of mice (mean ± SD, ^aP < 0.05 vs scramble/control, ^bP < 0.01 vs scramble/control, ^dP < 0.05 vs scramble/STZ treatment, n = 8); D: Periodic-acid Schiff staining showing changes of glomerular structure in different groups of mice (black bars = 20 μm, n = 8); E: TUNEL staining showing apoptosis of cells in the kidneys in different groups of mice (white bars = 50 μm, n = 8); F: Representative images showing protein levels of eNOS, ZO-1, and Occludin in the renal cortex of different groups of mice (mean ± SD, ^aP < 0.05 vs scramble/control, ^dP < 0.05 vs scramble/STZ treatment, n = 8); G: Western blot analysis showing expression of endoplasmic reticulum stress related proteins BiP, CHOP, and activating transcription factor 6 in the renal cortex of different groups of mice (mean ± SD, ^aP < 0.05 vs scramble/control, ^dP < 0.05 vs scramble/STZ treatment, n = 8). ATF6: Activating transcription factor 6; STZ: Streptozotocin.