Regulation of high glucose-induced apoptosis of brain pericytes by mitochondrial CA VA: A specific target for prevention of diabetic cerebrovascular pathology

Abstract

Events responsible for cerebrovascular disease in diabetes are not fully understood. Pericyte loss is an early event that leads to endothelial cell death, microaneurysms, and cognitive impairment. A biochemical mechanism underlying pericyte loss is rapid respiration (oxidative metabolism of glucose). This escalation in respiration results from free influx of glucose into insulin-insensitive tissues in the face of high glucose levels in the blood. Rapid respiration generates superoxide, the precursor to all reactive oxygen species (ROS), and results in pericyte death. Respiration is regulated by carbonic anhydrases (CAs) VA and VB, the two isozymes expressed in mitochondria, and their pharmacologic inhibition with topiramate reduces respiration, ROS, and pericyte death. Topiramate inhibits both isozymes; therefore, in the earlier studies, their individual roles were not discerned. In a recent genetic study, we showed that mitochondrial CA VA plays a significant role in regulation of reactive oxygen species and pericyte death. The role of CA VB was not addressed. In this report, genetic knockdown and overexpression studies confirm that mitochondrial CA VA regulates respiration in pericytes, whereas mitochondrial CA VB does not contribute significantly. Identification of mitochondrial CA VA as a sole regulator of respiration provides a specific target to develop new drugs with fewer side effects that may be better tolerated and can protect the brain from diabetic injury. Since similar events occur in the capillary beds of other insulin-insensitive tissues such as the eye and kidney, these drugs may also slow the onset and progression of diabetic disease in these tissues.

Keywords: Apoptosis; Brain peicytes; Diabetes; Mitochondrial carbonic anhydrases; Reactive oxygen species.

Figure 1. Immunoblot analysis of mitochondrial CA VA and CA VB proteins in brain pericytes

A)Representative Western blots for mitochondrial CA VA and tubulin in BPC and CA VA KD-BPC. B) Quantification of mitochondrial CA VA protein. C) Representative Western blots for mitochondrial CA VB and tubulin in BPC and CA VB KD-BPC. D) Quantification of mitochondrial CA VB protein. Tubulin is the loading control. BPC, brain pericytes; CA VA KD-BPC, CA VA knockdown brain pericytes; CA VB KD-BPC, CA VB knockdown brain pericytes.

Figure 2. Comparison of mitochondrial CA VB protein between BPC and CA VB-BPC

A) Light microscopic image of mitochondrial CA VB-BPC. B) Representative Western blots for mitochondrial CA VB and tubulin in BPC and CA VB-BPC. C) Quantification of mitochondrial CA VB protein. Tubulin is the loading control. BPC, brain pericytes; CA VB-BPC, CA VB overexpressing brain pericytes.

Figure 3. Effect of mitochondrial CA VA and CA VB knockdown on high glucose-induced intracellular ROS in pericytes

A) ROS. B) Cell viability. Results are presented as percentage of BPC treated with HG. Data are shown as mean ± SEM (n=4–8). The graphs are representative of three independent experiments. *p<0.05, **p<0.01. BPC, brain pericytes; CA VA KD-BPC, CA VA knockdown brain pericytes; CA VB KD-BPC, CA VB knockdown brain pericytes; HG, high glucose; NS, not significant.

Figure 4. Effect of mitochondrial CA VA and CA VB knockdown on apoptosis in pericytes

Percentages of apoptotic cells with respect to total DAPI (blue)-stained cells were calculated from 5 independent fields of 100 cells each. The values are expressed as mean ± SEM (n=4–5). The graphs are representative of three independent experiments. Evaluation was performed by a researcher blinded to the experimental protocol. *** p<0.001. BPC, brain pericytes; CA VA KD-BPC, CA VA knockdown brain pericytes; CA VB KD-BPC, CA VB knockdown brain pericytes; HG, high glucose; NS, not significant.

Figure 5. Effect of mitochondrial CA VB overexpression on high glucose-induced ROS in pericytes

A) ROS. B) Cell viability. Insets are the results published previously for CA VA overexpression on ROS and cell viability. Results are presented as percentage of BPC treated with NG. Data are shown as mean ± SEM (n=3–7). The graphs are representative of three independent experiments. *p<0.05. BPC, brain pericytes; CA VA-BPC, CA VA overexpressing brain pericytes; CA VB-BPC, CA VB overexpressing brain pericytes; HG, high glucose; NG, normal glucose.

Figure 6. Effect of mitochondrial CA VB overexpression on apoptosis

Percentages of apoptotic cells with respect to total DAPI (blue)-stained cells were calculated from 5 independent fields of 100 cells. The values are expressed as mean ± SEM (n=4–8). The graphs are representative of three independent experiments. Evaluation was performed by a researcher blinded to the experimental protocol. *p<0.05. BPC, brain pericytes; CA VA-BPC, CA VA overexpressing brain pericytes; CA VB-BPC, CA VB overexpressing brain pericytes; HG, high glucose; NG, normal glucose.

Figure 7

Role of mitochondrial CA in ROS production and apoptosis in brain pericytes.