The pericyte as a cellular regulator of penile erection and a novel therapeutic target for erectile dysfunction

Pericytes are known to play critical roles in vascular development and homeostasis. However, the distribution of cavernous pericytes and their roles in penile erection is unclear. Herein we report that the pericytes are abundantly distributed in microvessels of the subtunical area and dorsal nerve bundle of mice, followed by dorsal vein and cavernous sinusoids. We further confirmed the presence of pericytes in human corpus cavernosum tissue and successfully isolated pericytes from mouse penis. Cavernous pericyte contents from diabetic mice and tube formation of cultured pericytes in high glucose condition were greatly reduced compared with those in normal conditions. Suppression of pericyte function with anti-PDGFR-β blocking antibody deteriorated erectile function and tube formation in vivo and in vitro diabetic condition. In contrast, enhanced pericyte function with HGF protein restored cavernous pericyte content in diabetic mice, and significantly decreased cavernous permeability in diabetic mice and in pericytes-endothelial cell co-culture system, which induced significant recovery of erectile function. Overall, these findings showed the presence and distribution of pericytes in the penis of normal or pathologic condition and documented their role in the regulation of cavernous permeability and penile erection, which ultimately explore novel therapeutics of erectile dysfunction targeting pericyte function.


Animals and treatment
Male C57BL/6J mice were used and randomly grouped in this study. The experiments were approved by the institutional animal care and use subcommittee of our university. Diabetes was induced in 8-week-old mice by intraperitoneal injections of streptozotocin (50 mg/kg) for 5 days consecutively as we previously described 1 . At 8 weeks after the induction of diabetes, animals were anesthetized with ketamine (100 mg/kg) and xylazine (5 mg/kg) intramuscularly, and penis was exposed by use of sterile technique. The diabetic mice received repeated intracavernous injections of phosphate-buffered saline (PBS, days -3 and 0; 20 ml) or recombinant humanhepatocyte growth factor (rh-HGF) protein (days -3 and 0; 4.2 μg/20 ml of PBS), and normal mice received a single intracavernous injection of APB5 (1 µg, 5 µg, or 10 µg/20 ml of PBS), an anti-platelet-derived growth factor receptor-beta (PDGFR-β) blocking antibody. At 2 weeks after intracavernous injection of rh-HGF and 1 week after intracavernous injection of APB5, erectile function was measured during electrical stimulation of the cavernous nerve. The penis was then harvested for histologic examination. Fasting and postprandial blood glucose levels were determined with an Accu-Check blood glucose meter (Roche Diagnostics, Mannheim, Germany) before the mice were sacrificed (data not shown).

Immunohistochemistry and 3D reconstruction
The mouse or human penis tissue was fixed in 4% paraformaldehyde for 24 hours at 4 o C, and frozen tissue sections (7-μm [thin-cut] or 50-μm [thick-cut]) were incubated with antibodies to

Nerve-mediated erection studies
We evaluated erectile function (n = 6 per group) by electrical stimulation of the cavernous nerve 2 weeks after rh-HGF treatment and 1 week after APB5 treatment. Bipolar platinum wire electrodes were placed around the cavernous nerve. Stimulation parameters were 1 to 5 V at a frequency of 12 Hz, a pulse width of 1 ms, and a duration of 1 minute. During tumescence, the maximal intracavernous pressure (ICP) was recorded. The total ICP was determined by the area under the curve from the beginning of cavernous nerve stimulation to a point 20 seconds after stimulus termination. Systemic blood pressure was measured with a noninvasive tail-cuff system (Visitech Systems, Apex, NC, USA). The ratios of maximal ICP and total ICP to mean systolic blood pressure (MSBP) were calculated to normalize for variations in systemic blood pressure.

Cell culture
The mouse cavernous pericytes (MCPs) were prepared and maintained as previously described with minor modifications 2 . Briefly, eight-week-old C57BL/6J mice were used in this study. A schematic diagram of the procedure used to isolate the MCPs is illustrated in Figure 3A. were incubated with FITC-or TRITC-conjugated secondary antibodies (Zymed) for 2 hours at room temperature. We used human brain microvascular pericytes (HBMP) as a positive control.
Rat aorta smooth muscle cell line (A7r5) and mouse embryonic fibroblast cell line (NIH3T3) were used as negative controls. Signals were visualized and digital images were obtained with a confocal microscope (FV1000, Olympus, Tokyo, Japan).
The mouse cavernous endothelial cells (MCECs) were isolated and cultured as we previously described 3 .

In vitro Tube formation assay
The tube formation assay was performed as previously described 3 . About 50 µl of growth factor-reduced Matrigel (Collaborative Biomedical Products) was dispensed into 96-well tissue culture plates at 4°C. After gelling at 37°C for at least 30 minutes, the conditioned MCPs were seeded onto the gel at 2 ´ 10 4 cells/well in 200 µl of M199 medium. The assay was performed in a CO 2 incubator and the plates were incubated at 37°C for 24 hours. Images were obtained with a phase-contrast microscope and the numbers of tubes in each well of the plate were counted at a screen magnification of ´40. Only branch points were counted.

In vitro permeability assay
To examine the role of MCPs and MCECs on cavernous vascular permeability, MCPs and MCECs were co-cultured in the Transwell filters (1.0 µm pore size, Becton Dickinson Labware, Franklin Lakes, NJ) and the permeability was assayed by measuring the leakiness of Evans blue (Sigma-Aldrich) bound to bovine serum albumin (BSA, Bovogen Biologicals, VIC, Australia) as previously described 4 . MCPs at a density of 1 × 10 5 cells/well were seeded on the bottom side of the insert and grown to confluence in Transwell filters. After 2 days, MCECs (1 × 10 5 cells/well) were added to the upper inserts and grown to confluence for other 3 days, as previously described 5 .
To mimic an in vivo model for diabetes-induced pericyte-endothelial cell dysfunction, primary cultured MCPs-MCECs were serum-starved for 24 hours and were exposed to the normalglucose (5 mmol, Sigma-Aldrich) or high-glucose (30 mmol) condition for 48 hours. In order to examine the effect of rh-HGF on cavernous vascular permeability in high-glucose condition, the MCPs and MCECs were cultured and treated under the following conditions: the cells exposed to normal glucose condition (5 mmol), the cells exposed to the high-glucose condition (30 mmol), and the cells exposed to the high-glucose condition (30 mmol) and treated with rh-HGF (100 ng/ml). The permeability was measured at selected time points as previously described 4 .