Early systemic microvascular damage in pigs with atherogenic diabetes mellitus coincides with renal angiopoietin dysbalance

Abstract

Background: Diabetes mellitus (DM) is associated with a range of microvascular complications including diabetic nephropathy (DN). Microvascular abnormalities in the kidneys are common histopathologic findings in DN, which represent one manifestation of ongoing systemic microvascular damage. Recently, sidestream dark-field (SDF) imaging has emerged as a noninvasive tool that enables one to visualize the microcirculation. In this study, we investigated whether changes in the systemic microvasculature induced by DM and an atherogenic diet correlated spatiotemporally with renal damage.

Methods: Atherosclerotic lesion development was triggered in streptozotocin-induced DM pigs (140 mg/kg body weight) by administering an atherogenic diet for approximately 11 months. Fifteen months following induction of DM, microvascular morphology was visualized in control pigs (n = 7), non-diabetic pigs fed an atherogenic diet (ATH, n = 5), and DM pigs fed an atherogenic diet (DM+ATH, n = 5) using SDF imaging of oral mucosal tissue. Subsequently, kidneys were harvested from anethesized pigs and the expression levels of well-established markers for microvascular integrity, such as Angiopoietin-1 (Angpt1) and Angiopoietin-2 (Angpt2) were determined immunohistochemically, while endothelial cell (EC) abundance was determined by immunostaining for von Willebrand factor (vWF).

Results: Our study revealed an increase in the capillary tortuosity index in DM+ATH pigs (2.31±0.17) as compared to the control groups (Controls 0.89±0.08 and ATH 1.55±0.11; p<0.05). Kidney biopsies showed marked glomerular lesions consisting of mesangial expansion and podocyte lesions. Furthermore, we observed a disturbed Angpt2/Angpt1 balance in the cortex of the kidney, as evidenced by increased expression of Angpt2 in DM+ATH pigs as compared to Control pigs (p<0.05).

Conclusion: In the setting of DM, atherogenesis leads to the augmentation of mucosal capillary tortuosity, indicative of systemic microvascular damage. Concomitantly, a dysbalance in renal angiopoietins was correlated with the development of diabetic nephropathy. As such, our studies strongly suggest that defects in the systemic microvasculature mirror the accumulation of microvascular damage in the kidney.

Fig 1. Early atherogenic DM leads to increased capillary tortuosity.

A. Sidestream darkfield images of the oral mucosa visualizing the microvascular capillaries of a representative pig in the Controls, ATH and DM+ATH group. B. Mean tortuosity index of microvascular capillaries in the Controls (n = 7), ATH (n = 5) and DM+ATH pigs (n = 5). C. Mean vessel density (capillaries/mm²) in Controls (n = 7), ATH (n = 5) and DM+ATH (n = 5) pigs. Data shown are mean±SEM. *P<0.05 compared to Controls or ATH pigs. Mean vessel density (capillaries/mm²) was calculated by evaluation of number of vessels per selected microcirculatory image. Subsequently, tortuosity of capillary loops was assessed according to a validated scoring system (0: no twists to 4: four or more twists) and the average of assessed capillary tortuosity was used to calculate mean tortuosity index per patient.

Fig 2. Early stages of atherogenic DM leads to renal damage.

A: Representative illustration of PAS stained glomeruli from a DM+ATH pig, showing mesangial proliferation and matrix expansion with capillary loops lying around the mesangium as a corona, reminiscent of a beginning Kimmelstiel-Wilson nodule (left panel; thin black arrow). Dilated capillary loops with red cell fragments show intense PAI-1 staining on consecutive slides (right panel; thick black arrow). B: Mesangial expansion index in Controls (n = 7), ATH (n = 5) and DM+ATH (n = 5) pigs. C. Electron microscopy images illustrating a normal GBM architecture (left panel; thick arrow) of the Controls pig. In ATH, there is slight effacement of the podocyte pedicles (middle panel; thick arrow). In DM+ATH, marked lipid deposits were found (right panel). Data are shown as mean ± SEM. *P<0.05 compared to Controls or ATH pigs. Original magnification of A: x400 and C: x8000.

Fig 3. The Angpt2/Angpt1 balance is disturbed in atherogenic DM pigs.

A. Representative illustrations of kidney sections stained with Angpt1 (upper panels; arrow: glomerulus; arrowhead: peritubular area) or Angpt2 (lower panels; arrow: glomerulus; arrowhead: tubular staining) in Controls, ATH, and DM+ATH pigs. B. Immunofluorescent double staining of representative kidney sections for desmin (green)/Angpt1 (red;left panel) and vWF (green/Angpt2 (red; middle/right panel). Insets: double positivity for vWF/Angpt2 staining in yellow. C: Quantitative analysis of renal expression of Angpt1, Angpt2 and Angpt2/Angpt1 ratio. D. Relative mRNA expression of Angpt1 and Angpt2. Data are shown as mean ± SEM. *P<0.05 compared to Controls or ATH pigs. Original magnification of A and B: x400.

Fig 4. No difference in renal vWf and VEGF-A expression.

A. Representative illustrations of kidney sections stained with endothelial marker vWF (arrow: glomerulus; arrowhead: peritubular area) in Controls, ATH, and DM+ATH pigs. B. Representative images of kidney sections stained with VEGF in Controls, ATH, and DM+ATH pigs showing expression in podocytes (arrow head), parietal epithelial cells (thin arrow) and tubuli (asterix). Original magnification of A: x 200 and B: x400.

Fig 5. Correlation between capillary tortuosity and Angpt2/Angpt1balance and creatinine levels.

Scatter plot showing the correlation of renal protein expression of Angpt1(A), Angpt2 (B), Angpt2/Angpt1ratio (C).