Effects of Icariside II on Improving Erectile Function in Rats With Streptozotocin-Induced Diabetes

Authors


  • Supported by National Natural Science Foundation of China grant 30972993 and Peking National Natural Science Foundation grant Z080507030808011.

Correspondence to: Dr Zhong Cheng Xin, Andrology Center, Peking University First Hospital, Peking University, No. A59, Di'anmen W St, Xicheng District, Beijing 100009, China (e-mail: xinzc@bjmu.edu.cn).

Abstract

Abstract: Icariin and icariside II (ICA II), 2 active components isolated from herba epimedii, have a closely structural relationship. There is evidence that icariin may be useful in the treatment of erectile dysfunction (ED); however, the study on the therapeutic efficacy of ICA II on ED is currently scant. We investigated the effects of ICA II on improving erectile function of rats with streptozocin-induced diabetes. Fifty 8-week-old Sprague-Dawley rats were randomly distributed into normal control and diabetic groups. Diabetes was induced by a one-time intraperitoneal injection of streptozocin (60 mg/kg). Three days later, the diabetic rats were randomly divided into 4 groups including a saline-treated placebo arm and 3 ICA II-treated models (1, 5, and 10 mg/kg/d). After 3 months, penile hemodynamics was measured by cavernous nerve electrostimulation (CNE) with real time intracorporal pressure assessment. Penises were harvested with subsequent histological examination (picrosirius red stain, Hart elastin stain, and immunohistochemical stain) and Western blots to explore the expression of the nitric oxide-cyclic guanosine monophosphate (NO-cGMP) and transforming growth factor β1 (TGFβ1)/Smad2 signaling pathways. Diabetes significantly attenuated erectile responses to CNE. Diabetic rats had decreased corpus cavernosum smooth muscle/collagen ratio and endothelial cell content relative to the control group. The ratio of collagen I to III was significantly lower in the corpora of diabetic rats; furthermore, cavernous elastic fibers were fragmented in the diabetic animals. Neuronal nitric oxide synthase (nNOS), endothelial nitric oxide synthase, and vascular endothelial growth factor were expressed at lower levels in the diabetic group; ICA II-treated diabetic rats had higher expression in the penis relative to placebo-treated diabetic animals. Both the TGFβ1/Smad2/connective tissue growth factor (CTGF) signaling pathway and apoptosis were down-regulated in the penis from ICA II-treated rats. ICA II treatment attenuates diabetes-related impairment of penile hemodynamics, likely by increasing smooth muscle, endothelial function, and nNOS expression. ICA II could alter corpus cavernosum fibrous-muscular pathological structure in diabetic rats, which could be regulated by the TGFβ1/Smad2/CTGF and NO-cGMP signaling pathways.

About 52% of men between the age of 40 and 70 years suffer from erectile dysfunction (ED). Diabetes mellitus (DM) is a leading cause of ED; DM-induced ED is a common diabetic complication, and the prevalence of ED in men with DM is 3 times higher than in men without ED (Fedele, 2005). Pathological changes in diabetes-related ED are thought to include the damage of nerves, blood vesicles, corporal smooth muscle cells, and endothelial cells in the penis (Kandeel et al, 2001).

These tissues' pathological changes might be induced by a number of factors, including advanced glycosylation end products (AGE), oxidative stress by oxygen free radicals, or nitric oxide-cyclic guanosine monophosphate (NO-cGMP) pathway dysfunction, accompanied by inflammation, fibrosis, neuropathy, and hypogonadism (Hidalgo-Tamola and Chitaley, 2009). Recent data suggested that the up-regulation of the transforming growth factor β1/Smad (TGFβ1/Smad) signaling pathway may be a common final pathway for these various tissue insults (Hyman et al, 2002; Rodriguez-Vita et al, 2008; Zhang et al, 2010). Our previous study showed that fibrousmuscular structural changes in corpus cavernosum and deterioration of erectile function in streptozocin (STZ)-induced diabetic rats might have a relationship with up-regulation of the TGF-β1/Smad signaling pathway (Liu et al, 2011).

Oral inhibitors of phosphodiesterase type 5 (PDE5I) are currently the first-line treatments of choice for management of ED, but men with diabetes-related ED are less responsive to PDE5Is (Vickers and Satyanarayana 2002). Gene therapy and stem cell therapy are promising treatments that have shown some efficacy in preclinical studies. However, neither modality is currently available for wide clinical use; their ultimate role in management of ED in humans is yet to be determined (Lau et al, 2007; Bahk et al, 2010; Wong et al, 2010; Albersen et al, 2011; Jin et al, 2011).

Plants of the genus Epimedium (herba epimedii) have been utilized for the treatment of ED in traditional Chinese medicine for centuries. Icariin (C33H40O15, molecular weight 676.67 kd) and icariside II (ICA II; C27H32O10, molecular weight 514.54 kd), 2 flavonols isolated from herba epimedii, have a close structural relationship. Icariin is believed to be the principal active moiety of herba epimedii. Xin et al (2003) reported that icariin is a cGMP-specific phosphodiesterase type 5 (PDE5) inhibitor, with selective inhibitory activity on PDE5 that is more than 100 times that of PDE4 (PDE4/PDE5 of half maximal inhibitory concentration [IC50]). The PDE5 inhibitory activity of icariin is, however, 100 times lower than that of sildenafil.

Nitric oxide synthase (NOS; endothelial NOS [eNOS] and neuronal NOS [nNOS]) are important enzymes involved in the production of nitric oxide (NO) and thus regulate penile vascular homeostasis; molecular mechanisms involved in dysregulation of these NOS isoforms in the development of ED are essential to discovering the pathogenesis of ED in various disease states (Musicki et al, 2009). Despite a lower level of PDE5 inhibitory activity, preclinical studies have suggested that icariin may have utility in the management of ED. Chronic oral treatment with icariin in castrated rats led to significantly improved erectile function as well as decreased corporal fibrosis and increased NOS gene and protein expression (Liu et al, 2005). Shindel et al (2010) reported that icariin led to significantly greater neurite length of major pelvic ganglia fragments in vitro. In the same study, the authors reported improved penile hemodynamics and up-regulation of cavernous nNOS expression in icariin-treated rats relative to the neurotrophic effects. These tissue effects may be explained in part by data reported by Chung et al (2008), who found that icariin induced phosphorylation of eNOS as well as NO production by activating the Mek/Erk and PI3K/Akt/eNOS signaling pathways in human endothelial cells. Our previous study found that icariin could improve erectile function by preserving smooth muscle, endothelium content, and nNOS expression in the penis of STZ-induced diabetic rats; TGFβ1/Smad2 signaling pathway might play an important role in icariin improving erectile function in diabetic rats (Liu et al, 2011).

ICA II is another flavonol that could be isolated from herba epimedii. Although there is a small amount of ICA II in natural herba epimedii, ICA II could be transformed from icariin by intestinal bacteria after oraadministration (Xu et al, 2007). ICA II could be transformed from icariin with the action of β-glucosidase in vitro also (Park et al, 2008; Xia et al, 2010). Previous studies have indicated that ICA II inhibits PDE5A1 with an IC50 of 2 μM and shows at least 10-fold selectivity against other PDEs. It could bind to the active catalytic domain of PDE5A1 like sildenafil (Wang et al, 2006). It is logical to hypothesize based on these data that ICA II may be a more potent erectogenic agent than icariin. However, the therapeutic effect of ICA II on ED has not been investigated.

In this study we administered ICA II to diabetic rats as a daily supplement. Rats underwent functional testing of erectile hemodynamics during cavernous nerve stimulation as well as histological and molecular assessment of penile tissues for investigating the effects of ICAII on improving the corpus cavernosum fibrous-muscular change and TGFβ1/Smad2/connective tissue growth factor (CTGF) and NO-cGMP signaling pathways.

Materials and Methods

Preparation of ICA II

ICA II was biotransformed from icariin by the action of β-glucosidase and the optimum reaction conditions as follows: 50°C, 0.2 M disodium hydrogen phosphate and citric acid buffer system (pH 6.0); the ratio of icariin/enzyme was 1:1 and the reaction time was 5 hours (Xia et al, 2010). The result of purified biotransformed ICA II was 99.06% by high-performance liquid chromatography analysis.

Animals and Treatments

A total of 50 male 8-week-old Sprague-Dawley rats weighing 200–250 g were used in this study. The experiments performed were approved by the institutional animal care and use subcommittee of our university. Rats were fasted for 16 hours and injected intraperitoneally with freshly prepared STZ (60 mg/kg; Sigma Chemical Co, St Louis, Missouri) or vehicle (0.1 mol/L citrate-phosphate buffer, pH 4.5) according to the references (Usta et al, 2006; Chen et al, 2007). Blood glucose levels were monitored 72 hours later after STZ or vehicle injection, at regular intervals of every week throughout the study, and immediately prior to euthanasia. Blood samples were obtained by tail prick, and blood glucose concentration measured using a blood glucose meter (B. Braun, Melsungen, Germany). Seventy-two hours later, only rats with fasting glucose concentrations ≥300 mg/dL were included in the DM group. The diabetic rats were divided into 4 groups (n = 10 per group) and fed with saline or ICA II at the dose of 1, 5, or 10 mg/kg daily respectively (72 hours after STZ injection). Ten animals served as a control group and received no STZ and standard husbandry care. Twelve weeks later, we evaluated erectile function by cavernous nerve electrical stimulation. After functional testing the penis was harvested for histological examination (Masson trichrome stain, picrosirius red stain, Hart elastin stain, and immunohistochemical stain; each group contained 10 animals, and for each animal at least 5 slides from different areas of the organ were examined). Cavernous specimens were also used for Western blotting.

Measurement of Erectile Function

The rats from each DM group and their age-matched normal controls were anesthetized with 5% sodium pentobarbital intraperitoneally. The major pelvic ganglion, cavernous nerves, and pelvic organs were exposed. The skin overlying the penis was removed, and the penile crura were exposed by removing part of the overlying ischiocavernous muscle. Two 23-gauge needles connected to a PE-50 tube with heparinized saline (250 IU/mL) were carefully inserted into the crus and left carotid artery respectively. The other end of the PE-50 tube was connected to Biopac Systems MP150 multichannel physiological sienal recorder (Goleta, California). The cavernous nerve was exposed as described (Bella et al, 2007) and electrostimulation (12 Hz, pulse width 5 milliseconds, 1/2.5/5 V, duration 60 seconds) of the cavernous nerve was applied with a stainless steel bipolar hook electrode. The validity of this system was demonstrated previously (Martinez-Pineiro et al, 1994). Mean arterial pressure (MAP; calculated according to the formula diastolic blood pressure + [(systolic blood pressure 2 diastolic blood pressure)/3]) was measured concurrently. The ratio of maximal intracavernous pressure (ICP; mm Hg) to MAP (mm Hg) was calculated to normalize for variations in systemic blood pressure among subject rats.

Histochemistry

After functional testing, the penis was harvested and immersed in neutral buffered formalin containing 4% formaldehyde for a period of 6 hours, embedded in paraffin in vertical direction, and cut in cross section. Sections of 5-μm thickness were cut using a rotor microtome. The sections were stained for histological examination using picrosirius red stain and Hart elastin stain.

For picrosirius red F3BA (Sigma-Aldrich) (n = 10 per group), tissue sections (5 μm) were treated and incubated in 0.1% picrosirius red. Before dehydration, the slides were treated with 0.01 N hydrochloride and mounted. The picrosirius red staining was expressed as a ratio of the type I/type III collagen using Image Pro Plus software (version 6.0; Media Cybernetics, Bethesda, Maryland).

For elastic fiber staining (n = 10 per group), penile tissue sections (5 μm) were soaked in 0.25% potassium permanganate solution for 5 minutes, cleared in 5% oxalic acid, and soaked in resorcin-fuchsin solution (Poly Scientific, Bay Shore, New York) overnight. After washing, sections were counterstained with tartrazine. Hart-stained sections were captured with a digital camera and imported into Image Pro Plus software. The maximum length of elastic fibers was measured from different random fields of microscope.

Immunohistochemistry

For immunohistochemistry (n = 10 per group), penile tissue sections (5 μm) were deparaffinized and hydrated by sequential incubations in xylene and ethanol. After washing in 3× phosphate-buffered saline (PBS) for 5 minutes, the sections were blocked with 3% H2O2 for 10 minutes to quench endogenous peroxidase activity and with heat-induced epitope retrieval methods to perform the antigen unmasking (antigen retrieval solution: 0.01 M sodium citrate buffer, pH 6.0, 95°C). The tissue sections were incubated with antibody to alpha smooth muscle actin (α-SMA; rabbit polyclonal, 1:400; Abcam, Cambridge, Massachusetts), nNOS (1:400; rabbit polyclonal, Santa Cruz Biotechnology, Santa Cruz, California), eNOS (rabbit polyclonal, 1:300; Santa Cruz), vascular endothelial growth factor (VEGF; rabbit polyclonal, 1:600, located in the extracellular matrix; Abcam), CTGF (rabbit polyclonal, 1:800; located in the extracellular matrix; Abcam), TGFβ1 (mouse monoclonal, 1:400, located in the extracellular matrix; Abcam), Smad2 (rabbit monoclonal, 1:200, located in the cytolymph; Cell Signaling, Beverly, Massachusetts), phospho-Smad2 (rabbit polyclonal, 1:200, located in the cell nucleus; Cell Signaling) followed by the use of the Histostain-Plus Kit (Zymed Laboratories, San Francisco, California). The cell nucleus was stained with hematoxylin staining. Semiquantitative analysis was performed to evaluate the intensity of α-SMA, nNOS, eNOS, VEGF, CTGF, TGFβ1, Smad2, and phospho-Smad2 staining by the use of Image Pro Plus software. The intensity score for the evaluation of the grading intensity was 0 = no reactivity, 1 = weak reactivity, 2 = moderate reactivity, 3 = strong reactivity, and 4 = extremely strong reactivity.

For terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) staining (n = 10 per group), sections (5 μm) were deparaffinized and hydrated by sequential incubations in xylene and ethanol. The samples were then processed according to the instructions provided with the in situ cell death detection kit POD (Roche Diagnostic Corporation, Indianapolis, Indiana). The cell nucleus was performed with hematoxylin staining. Semiquantitative analysis was performed to evaluate the intensity by the use of Image Pro Plus software.

Western Blot

The cellular protein samples were prepared by homogenization of penile tissue in a lysis buffer containing 1% IGEPAL CA-630, 0.5% sodium deoxycholate, 0.1% sodium dodecyl sulfate, aprotinin (10 μg/mL), leupeptin (10 μg/mL), and PBS. Cell lysates containing 20 μg of protein were electrophoresed in sodium dodecyl sulfate-polyacrylamide gel electrophoresis and then transferred to a polyvinylidene fluoride membrane (Millipore Corp, Bedford, Massachusetts). The membrane was blocked with 5% skimmed milk for 1 hour at room temperature and incubated overnight at 4°C with primary antibody against β-actin (1:1000; Santa Cruz), α-SMA (1:1000; Abcam), nNOS (1:1000; Santa Cruz), eNOS (1:500; Santa Cruz), VEGF (1:1000; Abcam), Smad2 and phospho-Smad2 (1:1000; Cell Signaling), CTGF (1:2000; Abcam), and TGFβ1 (1:1000; Abcam). After the hybridization of secondary antibodies, the resulting images were analyzed with ChemiImager 4000 (Alpha Innotech Corporation, San Leandro, California) to determine the integrated density value of each protein band. Results were semiquantified by densitometry (n = 4 per group).

Statistical Analysis

The results are expressed as the means ± SD. One-way analysis of variance followed by Bonferroni multiple comparisons test was used to evaluate whether differences among groups were significant. All calculations were performed using SPSS statistical software (version 13.0; SPSS, Chicago, Illinois). Probability values of less than 5% were considered significant.

Results

Metabolic Variables and ICP/MAP Assessment

The data of metabolic variables and ICP/MAP are reported in the Table. The initial body weight and serum glucose concentrations among the 5 groups did not differ significantly. At 12 weeks after diabetes was induced, the fasting glucose concentrations of diabetic animals were significantly higher than the concentrations in the age-matched normal controls. Body weight was significantly lower in all diabetic animals compared with the normal control group. No significant difference in body weight or glucose concentration was found between the DM and ICA II-treated groups. Erectile function variables, such as ICP, ratio of ICP to MAP, and ratio of area under the curve to MAP were significantly lower in DM rats. ICA II-treated animals had superior penile hemodynamics compared with placebo-treated animals. No significant difference in MAP was noted among groups (Table; Figure 1A through C).

Table Table. . Metabolic and physiological variablesa
  
VariableDMICA II 1 mgICA II 5 mgICA II 10 mgNormal Control
  1. Abbreviations: AUC, area under the curve; DM, diabetes mellitus; ED, erectile dysfunction; ICA II, icariside II; ICP, intracavernous pressure; MAP, mean arterial pressure.

  2. a Values are the mean values (±SD) from n = 10 animals per group. Ratios of maximal ICP (mm Hg) and AUC to MAP (mm Hg) were calculated to normalize for variations in systemic blood pressure.

  3. b P < .01 compared with the normal control group.

  4. c P < .05 compared with the DM group.

  5. d P < .01 compared with the DM group.

Initial weight, g228.2 ± 12.1219.4 ± 10.4223.8 ± 18.3226.9 ± 23.9229.2 ± 28.1
Final weight, g328.3 ± 36.7b349.6 ± 39.5b338.7 ± 31.9b392.9 ± 32.7b625.5 ± 26.7
Initial fasting glucose, mg/dL107.2 ± 2.2113.1 ± 3.1104.1 ± 6.2107.2 ± 4.5111.2 ± 7.1
Final fasting glucose, mg/dL515.6 ± 46.7b531.2 ± 53.3.4b540.9 ± 45.3b496.8 ± 66.4b108.4 ± 16.1
MAP, mm Hg97.24 ± 7.35102.73 ± 6.74101.33 ± 7.35105.63 ± 6.38119.37 ± 5.24
ICP, mm Hg38.62 ± 5.3363.43 ± 6.53c79.42 ± 7.66d82.32 ± 7.58d102.70 ± 7.27d
AUC654 ± 1071453 ± 219d1759 ± 202d1823 ± 236d2040 ± 116d
Figure 1.

Increase of erectile function in icariside II (ICA II)-treated diabetic rats. (A) Representative intracavernous pressure (ICP) responses to electrical stimulation of the cavernous nerve for age-matched normal control, diabetic, and ICA II-treated (1, 5, or 10 mg/kg/d) diabetic groups. The stimulus interval (60 seconds) is indicated by a solid bar. (B, C)Erectile function presented as maximal ICP/mean arterial pressure (MAP) and total ICP/MAP (area under the curve/MAP) in each group. *, P < .05; †, P < .01 compared with the diabetes mellitus (DM) group.

Muscular Changes

We performed immunohistochemical staining with antibody to α-SMA to assess the smooth muscle content (Figure 2A and B). The penis tissue from diabetic rats showed significantly lower α-SMA-positive staining in the cavernous sinusoids relative to normal controls. Animals that had been treated with ICA II had greater density of α-SMA staining as the extent score. Also, the Western blots confirmed that α-SMA content was significantly lower in the DM group compared with the normal control group and significantly lower in the untreated diabetic group compared with the ICA II-treated group (Figure 2C and D).

Figure 2.

Increase of alpha smooth muscle actin (α-SMA) expression and collagen I and III fiber ratio in icariside II-treated (1, 5, or 10 mg/kg/d) diabetic rats' corpus cavernosum. (A) The expression of α-SMA by immunohistochemical staining in corpus cavernosum. (B) The analysis of α-SMA expression by immunohistochemical staining. (C) Western blot analysis of α-SMA expression. (D) The ratio of α-SMA to β-actin by Western blots. (E) The photographs of cavernosum sinus with picrosirius red stain. The red part shows the type I collagen and the green part shows the type III collagen. (F) Quantification of collagen fiber was assessed using Image Pro software (Media Cybernetics, Bethesda, Maryland) and expressed as a ratio of collagen I to collagen III. Each bar depicts the mean values (±SD) from n = 10 (immunohistochemical staining) and n = 4 (Western blots) per group. *, P < .05; †, P < .01 compared with the diabetes mellitus (DM) group.

Extracellular Matrix Changes

We performed histological examinations to assess the penile collagen fibers and elastic fibers. Picrosirius red staining revealed that the ratio of collagen I to III was lower in the diabetic group compared with the normal control and ICA II-treated groups (Figure 2E and F).

We also found morphologically that the elastic fibers were changed dramatically. The elastic fibers were long and well organized in the normal control group. In the diabetic group, these fibers appeared fragmented and disorganized. The maximum length of elastic fibers in the ICA II-treated groups was longer than in the diabetic group; the content percentage of elastic fibers in the ICA II-treated groups was higher than in the untreated diabetic group (Figure 3A through C).

Figure 3.

Assessment of elastic fiber in normal control, diabetic, and icariside II-treated (1, 5, or 10 mg/kg/d) diabetic rats' corpus cavernosum. (A)Representative photographs of cross sections of corpus cavernosum with Hart elastin stain. Elastic fiber manifested as black, connective tissue is pink-red, and other tissue elements are yellow. (B) Maximal elastic fiber length analysis results. (C) Quantification of elastic fiber was assessed using Image Pro software (Media Cybernetics, Bethesda, Maryland) and expressed as a percentage of the imaged area. Each bar depicts the mean values (±SD) from n = 10 per group. *, P < .05; †, P < .01 compared with the diabetes mellitus (DM) group.

Increase of nNOS expression by ICA II

We performed nNOS immunohistochemical staining to locate and quantify NOS-positive nerve fibers content in each group. The number of nNOS-positive nerve fibers in the dorsal penile nerves was significantly greater in the ICA II-treated group compared with the untreated diabetic animals (Figure 4A and B).

Figure 4.

Increase of neuronal nitric oxide synthase (nNOS), endothelial nitric oxide synthase (eNOS), and vascular endothelial growth factor (VEGF) expression in icariside II-treated (1, 5, or 10 mg/kg/d) diabetic rats' corpus cavernosum. (A) The expression of nNOS by immunohistochemical staining in corpus cavernosum. (B) The expression analysis of nNOS by immunohistochemical staining. (C) Western blot analysis of the nNOS expression. (D)The ratio of nNOS to β-actin by Western blots. (E) Western blot analysis of VEGF. (F) Data are presented as the relative density of VEGF compared with that of β-actin. (G) Western blot analysis of eNOS. (H) Data are presented as the relative density of eNOS compared with that of β-actin. Each bar depicts the mean values (±SD) from n = 10 (immunohistochemical staining) and n = 4 (Western blots) per group. *, P < .05; †, P < .01 compared with the diabetes mellitus (DM) group.

We performed Western blots to evaluate the expression of nNOS in the penis tissue in each group. nNOS protein expression was significantly lower in the DM group compared with the normal control group; however, the level of the protein expression was significantly greater in the ICA II-treated groups relative to the untreated diabetic group (Figure 4C and D).

Increase of VEGF and eNOS Expression by ICA II

We performed VEGF and eNOS Western blots to quantify endothelial function in each group. Penile tissue from diabetic rats showed significantly lower VEGF immunoexpression relative to normal controls. This was attenuated in part in the ICA II-treated groups (Figure 4E and F). eNOS expression was quantified by Western blots. The level of eNOS positivity was significantly lower in the diabetic group than in the normal control group; this decline was attenuated in part by ICA II (Figure 4G and H).

Down-Regulation of TGFβ1/Smad2/CTGF Signaling Pathway by ICA II

We performed immunohistochemical staining to localize TGFβ1 protein in the penis of each group. Diabetic animals had higher TGFβ1 immunoreactivity in the dorsal artery of the penis, the dorsal vena of the penis, and the corpora cavernosa compared with the normal control group. TGFβ1 expression was significantly lower in ICA II-treated diabetic groups compared with the DM group (Figure 5A and B).

Figure 5.

Decrease of transforming growth factor β1 (TGFβ1) and connective tissue growth factor (CTGF) expression in icariside II (ICA II)-treated (1, 5, or 10 mg/kg/d) diabetic rats' corpus cavernosum. (A)Immunohistochemical staining with antibody to TGFβ1 in artery, vein, and cavernosum sinus of normal, diabetic, and ICA II-treated rats' corpus cavernosum. (B) Semiquantitative analysis of TGFβ1 protein expression. (C) Western blot analysis of TGFβ1. (D) The ratio of TGFβ1 to β-actin. (E) Western blot analysis of CTGF expression. (F) The ratio of CTGF to β-actin. Each bar depicts the mean values (±SD) from n = 10 (immunohistochemical staining) and n = 4 (Western blots) per group. *, P < .05; †, P < .01 compared with the diabetes mellitus (DM) group.

Immunohistochemical staining for phospho-Smad2 and total Smad2 was performed to evaluate the expression of this cellular messenger in the corpus cavernosum (Figure 6A through C). The diabetic group showed higher phospho-Smad2 and total Smad2 immunoreactivity in the dorsal artery of the penis, the dorsal vein of the penis, and the corpora cavernosa compared with the normal control group. The levels of phospho-Smad2 and total Smad2 in ICA II-treated groups were significantly lower than those in the diabetic group.

Figure 6.

Decrease of total Smad2 and phospho-Smad2 expression in icariside II-treated (1, 5, or 10 mg/kg/d) diabetic rats' corpus cavernosum. (A)Immunohistochemical staining with antibody to phospho-Smad2 and total Smad2 in artery, vein, and cavernosum sinus. (B, C) Semiquantitative analysis of phospho-Smad2 and total Smad2 protein expression. (D) Western blot analysis of phospho-Smad2 and total Smad2 expression. (E) The ratio of phospho-Smad2 to total Smad2. Each bar depicts the mean values (±SD) from n = 10 (immunohistochemical staining) and n = 4 (Western blots) per group. *, P < .05; †, P < .01 compared with the diabetes mellitus (DM) group.

The expression of TGFβ1 by Western blots in the untreated diabetic group was higher than in the ICA II-treated groups (Figure 5C and D). The results of Western blots showed that the ratio of phospho-Smad2 to total Smad2 was significantly higher in the diabetic group than for the normal controls. This effect was lessened in ICA II-treated diabetic animals (Figure 6D and E).

We also performed Western blots to evaluate the expression of CTGF in each group. Diabetic animals showed higher CTGF expression than the control group. Expression of CTGF was less in the ICA II-treated groups compared with the untreated diabetic animals, especially in the 5 mg/kg/d and 10 mg/kg/d ICA II groups (Figure 5E and F).

Apoptotic Index

We measured apoptotic index in the corpus cavernosum by the TUNEL method. The apoptotic index was significantly higher in the diabetic groups relative to normal controls. The apoptotic index was lower in the ICA II-treated groups than in the untreated diabetic group, especially in the 5 and 10 mg/kg/d ICA II groups (Figure 7A and B).

Figure 7.

Decrease of apoptosis in icariside II-treated (1, 5, or 10 mg/kg/d) diabetic rats' corpus cavernosum. (A) Representative photos for terminal deoxynucleotidyl transferase dUTP nick end labeling staining of corpus cavernosum tissue (×400). (B) The statistical analysis of the apoptotic index. The apoptotic cells stained to be dark brown. Apoptotic index presented as the ratio of apoptotic nuclei to the total number of nuclei counted. Each bar depicts the mean values (±SD) for n = 4 per group. *, P < .05; †, P < .01 compared with the diabetes mellitus (DM) group.

Discussion

In the early course of diabetes, the intracellular hyperglycemia causes abnormalities in blood flow and increased vascular permeability. There is an overproduction of extracellular matrix induced by growth factors such as TGF; hyperglycemia may also decrease production of trophic factors for endothelial and neuronal cells. Together, these changes lead to neuropathy and endothelial lesions (Brownlee, 2001; Ceriello, 2009). DM-induced ED is a common diabetic complication, and ED can be an early sign of diabetes in men who have not yet been diagnosed. Means to prevent and treat pathological changes in diabetic complications such as ED are currently very important.

Icariin has numerous potential applications in different fields of medicine. However, it is best known for its putative role in potentiating sexual function by inhibition of PDE5 (Ning et al, 2006; Dell'Agli et al, 2008). ICA II is a metabolite of icariin and has at least 10-fold selectivity for PDE5I against other PDEs (Wang et al, 2006). In our study, we provided functional evidence that ICA II might have beneficial effects on erectile function in diabetic rats.

Fibrosis, accompanied with decreased smooth muscle, endothelial deterioration, and neuropathy, is an important pathological process in diabetic ED models (Pegge et al, 2006; Kovanecz et al, 2009). Collagen and elastic fibers of the tunica albuginea are key components of this compliant tissue and permit increase in girth and length during tumescence and rigidity of the penis (Hsu et al, 1994). Collagen fibers are composed of aggregations of tropocollagen molecules and are normally arranged in an undulating pattern in the flaccid state. Type I collagen forms stiff bands of fibrils, and type III collagen is found predominantly in distensible elastic tissue. Changes of elastic fibers or collagen types can provoke mechanical alterations of the penis, which may reduce its elasticity and compliance. In our study, the ratio of collagen I to III in the DM group was lower than that in the normal control group. A dramatic morphologic change was also noted in that penile elastic fibers from the DM group appeared fragmented and disorganized. Animals treated with ICA II demonstrated lesser fragmentation of elastin and a higher ratio of collagen I to collagen III. We also determined that penile tissues from diabetic rats showed decreased α-SMA, nNOS, and eNOS expression in the cavernous sinusoids relative to the normal control group, as well as up-regulation of the TGFβ1/Smad2/CTGF signaling pathway. Animals treated with ICA II demonstrated increased α-SMA, nNOS, and eNOS expression and down-regulation of the TGFβ1/Smad2/CTGF signaling pathway. The influence of ICA II on nNOS expression is of particular interest, as there is only a little published data to suggest that other treatment modalities for ED exert an influence on nNOS expression (Moore and Wang, 2006).

TGF-β1 increases collagen synthesis in human corpus cavernosal smooth muscle cells in culture (Moreland et al, 1995; Moreland, 1998). It is known to be a pleiotropic cytokine that increases collagen synthesis in cultured corpus cavernosum smooth muscle cells (Moreland, 1998). Zhang et al (2008, 2010) reported that up-regulation of the TGFβ1/Smad signaling pathway in the penis of diabetic rats may play an important role in diabetes-induced structural changes and deterioration of erectile function. Ahn et al (2005a,b) reported that diabetic rats showed a significant decrease in the smooth muscle and endothelial cell content with an increase of TGF-β1 expression. The PDE5 inhibitor DA-8159 partially prevented this reduction and ameliorated the over-expression of TGFβ1 and intracorporal fibrosis.

In our study, the expression of TGFβ1, Smad2, phospho-Smad2, and CTGF was higher in the diabetic groups relative to the normal control group, and the expression of these proteins was significantly lower in ICA II-treated groups compared with the untreated diabetic group. ICA II might have an effect on the TGFβ1/Smad2/CTGF signaling pathway and palliate the fibrotic process in the corpus cavernosum of diabetic animals. Increased expression of VEGF suggests that ICA II treatment might improve endothelial function. We further confirmed the apoptotic index was significantly lower in ICA II-treated DM groups. ICA II might reduce apoptosis of smooth muscle and endothelium in the corpus cavernosum of diabetic rats via inhibition of the TGFβ1/Smad2/CTGF signaling pathway.

To clarify the precise role of ICA II in diabetes-related erectile impairment, further investigation on mechanisms (eg, AGEs, inflammation, and neuropathy) of ICA II on diabetic ED are recommended.

In summary, ICA II shows promise in the treatment of diabetes-related ED. This activity is likely modulated by preservation/recovery of smooth muscle, endothelial function, and NOS activity. The TGFβ1/Smad2/CTGF signaling pathway may play a critical role in ICA II-related recovery of erectile capacity in diabetic rats.

Ancillary