ICA was procured by three-fold ethanol extraction of the aerial part of Epimedii herba (specifically, E. koreanum, E. sagittatum, and/or E. brevicornum
). The product was dried with a vacuum. The dried extract was suspended in water and partitioned successively with n-hexane, CHCl3, and n-butyl alcohol. The n-butyl alcohol fraction was subjected to silica gel column chromatography to isolate ICA, which was further purified by repeated recrystallization with methanol. The final product was 98.8% ICA as determined by high-performance liquid chromatography analysis [11
Fifty-eight healthy 12-week-old male Sprague Dawley rats (Harlan Laboratories, Indianapolis, Indiana, USA) were obtained and housed two per cage in standard rodent cages. All animals were given standard rat chow and tap water ad libitum and maintained in a temperature- and humidity-controlled room on 12-hour light/dark cycles. Live animals were handled in accordance with our institution's policy on animal husbandry; approval for all procedures on live animals was obtained from our Institutional Animal Care and Use Committee.
Cavernous nerve exposure with or without injury was performed as previously described [12
]. Briefly, animals were anesthetized with isoflurane by inhalation. A laparotomy was performed and the cavernous nerves identified in the periprostatic space in all rats. In positive control animals (n = 12) no further intervention was performed. In all other animals the cavernous nerve was crushed distal to the major pelvic ganglion with a specially designated needle driver for a period of 2 minutes; this procedure was repeated on the contralateral side. After cavernous nerve crush the laparotomy was closed in two layers with absorbable suture. All animals received an opioid and an nonsteroidal anti-inflammatory drug (NSAID) by intraperito-neal (IP) injection in the perioperative period.
Starting on the day of nerve injury and continuing for 4 weeks, rats were given daily gavage feedings of a 50:50 mix of normal saline and DMSO in which was dissolved ICA at concentrations of 0 (positive and negative control groups, n = 12 each), 1 mg/mL (n = 10), 5 mg/mL (n = 12), and 10 mg/mL (n = 12). Animals were treated with the appropriate solutions at ICA doses of 0 (sham and negative control) 1 mg/kg, 5 mg/kg, or 10 mg/kg. Animals were weighed weekly and dosing was adjusted as appropriate. To facilitate gavage feedings animals were briefly exposed to isoflurane 2% in an induction chamber; after the animal was mildly sedated an olive tip 16-gauge feeding needle was advanced atraumatically into the animal's esophagus and the prescribed treatment was administered. To assess the effect of a single dose of ICA 10 mg/kg on penile hemodynamics, a separate group of six rats underwent cavernous nerve crush and received no treatment other than a single dose of ICA 10 mg/kg 2 hours prior to sacrifice and cavernous nerve stimulation.
At the 4-week time point all rats were anesthetized with ketamine and midazolam (100 and 5 mg/kg, respectively) by IP injection. All rats received a dose of saline/DMSO ± ICA at the appropriate concentration 2 hours prior to anesthesia. A repeat laparotomy was performed and the cavernous nerves were identified. The penis was denuded of overlying skin and cannulated with a heparinized 23-gauge needle connected to a real-time continuous pressure transducer. The cavernous nerves were then stimulated with a bipolar steel electrode; stimulation parameters were 50-second continuous trains at 20 Hz, 1.5 mAmp. Real-time response of the erectile tissue was determined by change in ICP. The maximum change in ICP was utilized for further analysis; additionally, the area under the curve (AUC) for all ICP assessments was quantitated by determining the number of pixels below the ICP curve during cavernous nerve stimulation using Image-Pro Plus v 5.1 (Media Cybernetics, Bethesda, MD, USA).
After functional testing, systemic blood pressure was measured via aortic cannulation. Mean arterial pressure (MAP) was calculated by the formula MAP = (diastolic blood pressure + [(systolic blood pressure—diastolic blood pressure)/3]). Serum samples were obtained at the time of aortic cannulation. Whole blood was aspirated into a collection tube and allowed to clot for 30 minutes. Clotted blood was then centrifuged at 1,000 rpm for 15 minutes. Serum was decanted from the collection tube and stored at -80 centigrade until use. Serum testosterone levels in eight randomly selected rats from each group were assayed using the Parameter Testosterone Assay (R&D Systems, Minneapolis, MN, USA) according to the manufacturer's instructions. Serum LH levels from the same rats were assayed using a rat LH ELISA kit (Cusabio, Newark, DE, USA) according to the manufacturer's instructions.
After aortic cannulation, animals were euthanized with IP pentobarbital (200 mg/kg) and bilateral thoracotomy. Penile tissues were harvested for immunohistochemistry (stored in 3% formaldehyde with 0.05% picric acid for 4 hours followed by 30% sucrose in phosphate buffered saline) at 4°C until use.
A segment of the penis from all subject rats was fixed in optimal cutting temperature fixative and subsequently sectioned at 5 microns in a cryostat. Tissue sections were stained with mouse anti-nNOS (1:800; BD Transduction Laboratories, Franklin Lakes, NJ, USA), mouse anti-eNOS (1:11,000; Sigma-Aldrich, St. Louis, MO, USA), rabbit anti-calponin antibody (1:500; Abcam, Cambridge, MA, USA) and terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL; Roche Diagnostics Corporation, Indianapolis, IN, USA) using standard technique. Slides were examined by an observer blinded to treatment group. nNOS positivity was ascertained by counting positive fibers in four random high-power fields (hpf; 400×) of the midline dorsal neurovascular bundle of the penis. eNOS positivity was quantified by selection of four random 200× fields of the corporal bodies. Calponin positivity was ascertained by selection of four random 200× fields of the corporal bodies. TUNEL positivity was calculated by counting positive nuclei in four random hpf fields of the corporal bodies. All staining intensity was measured using Image-Pro Plus v 5.1 (Media Cybernetics, Bethesda, MD, USA).
For Western blot analysis, cellular protein samples from the penis were prepared by homogenization of cells in a lysis buffer containing 1% IGEPAL CA-630, 0.5% sodium deoxycholate, 0.1% sodium docecyl sulfate, aprotinin (10 μg/mL), leupeptin (10 μg/mL), and phosphate buffered saline. Cell lysates containing 20 mcg of protein were electrophoresed in sodium docecyl sulfate polyacrylamide gel electrophoresis and then transferred to a polyvinylidene fluoride membrane (Millipore Corp, Bedford, MA, USA). The membrane was stained with Ponceau S to verify the integrity of the transferred proteins and to monitor the unbiased transfer of all protein samples. Detection of target proteins on the membranes was performed with an electrochemiluminescence kit (Amersham Life Sciences Inc, Arlington Heights, IL, USA) with the use of primary antibodies for nNOS, eNOS, Calponin, and Caspase3 (all antibodies at 1:500 concentration, Abcam Inc, Cambridge, MA, USA) After hybridization of secondary antibodies, the resulting images were analyzed with ChemiImager 4,000 (Alpha Innotech Corporation, San Leandro, CA, USA) to determine the integrated density value of each protein band.
A separate group of twelve 12-week-old Sprague-Dawley rats underwent isoflurane anesthesia for harvest of the major pelvic ganglia (MPG) for the purpose of cell culture as previously described [13
]. After harvest rats were euthanized with IP pentobarbital (200 mg/kg) and bilateral thoracotomy. Each dorsocaudal region of the MPG (from which the cavernous nerve originates) of subject rats was sectioned into three pieces and plated on a sterile cell culture well in a matrigel medium for a total of 72 specimens. All MPG fragments were incubated in Dulbecco's Modified Eagle medium (Sigma-Aldrich) in a humidified chamber with 5% CO2. To each culture well was added ICA or sildenafil at 10 nM, 100 nM, and 10 uM. Control specimens were incubated with phosphate buffered saline. At 48 hours, MPG fragments were examined under the dissecting microscope for neurite outgrowth. The longest neurite from each segment was measured; mean neurite length for each treatment group was calculated.
All data are reported as mean ± standard deviation. For statistical analysis, one-way anova was performed followed by Bonferroni multiple comparison's test for between-group differences. All calculations were performed using Prism version 4.2 (Graphpad Software, La Jolla, CA, USA). Statistical significance was set at P < 0.05.