Alterations in the phosphodiesterase type 5 pathway and oxidative stress correlate with erectile function in spontaneously hypertensive rats

Abstract To explore how alterations in the phosphodiesterase type 5 (PDE5) signalling pathway and oxidative stress correlate with changes in the expression of relaxation and contraction molecules and erectile dysfunction (ED) in the corpus cavernosum smooth muscle (CCSM) of spontaneously hypertensive rats (SHR). In this study, SHR and Wistar‐Kyoto (WKY) rats were used. Erectile function was determined by apomorphine test and electrical stimulation (ES) of cavernous nerve. Masson's trichrome staining and confocal microscopy were performed. Nitric oxide synthase (NOS), PDE5, phosphorylated‐PDE5 and α1‐adrenergic receptor (α1AR) were determined by RT‐PCR and Western blotting while oxidative stress in CC was determined by colorimetric analysis. SHR exhibited obvious ED. CC of SHR showed less SM but more collagen fibres. The expression of NOS isoforms in SHR was significantly decreased while all α1AR isoforms were increased. In addition, PDE5 and phosphorylated‐PDE5 were down‐regulated and its activity attenuated in the hypertensive rats. Meanwhile, the SHR group suffered oxidative stress, which may be modulated by endoplasmic reticulum stress and NADPH oxidase up‐regulation. Dysregulation of NOS and α1AR, histological changes and oxidative stress in CC may be associated with the pathophysiology of hypertension‐induced ED. In addition, PDE5 down‐regulation may lead to the decreased efficacy of PDE5 inhibitors in some hypertensive ED patients and treatment of oxidative stress could be used as a new therapeutic target for this type of ED.


| INTRODUC TI ON
Penile erection is a complex psycho-physiological process involving a series of neural and vascular activities in which contraction and relaxation of corpus cavernosum smooth muscle (CCSM) play an important role. It is well known that the nitric oxide/cyclic guanosine monophosphate (NO/cGMP) pathway predominantly modulates CCSM relaxation and penile erection. 1 Specifically, NO is produced by nitric oxide synthase (NOS) using l-arginine and oxygen. NO increases the production of cGMP (the second messenger), which relaxes the CCSM. 2 The cGMP is degraded by phosphodiesterase type 5 (PDE5). Currently, the most effective drugs for treating erectile dysfunction (ED) are PDE5 inhibitors (PDE5is). PDE5is can block the activity of the PDE5 enzyme to increase the cGMP level, leading to the relaxation of CCSM and erection. In corpus cavernosum (CC), NOS exists as three isoforms: eNOS (endothelial NOS), nNOS (neuronal NOS) and iNOS (inducible NOS). The nNOS isoform is preferentially expressed in neurons or nerves and considered to mediate both the initiation and maintenance of penile erection, while the eNOS isoform is mainly expressed in endothelial cells and considered responsible for the maintenance of erection. [3][4][5] iNOS is expressed in almost all cell types. Although recent evidence suggested that iNOS in the penis exhibits an antifibrotic role and improves erectile function in diabetes mellitus-induced ED, 6,7 the functions of iNOS in CC were not fully explored. On the other hand, the CCSM spends the majority of its time in the contracted state, which is in contrast with other smooth muscles. 8 Although the relaxation of CCSM has been extensively studied throughout the last couple of decades, the contraction of CCSM is also important but less studied. Many molecules and pathways are involved in the process of CCSM contraction with adrenergic neurotransmission the most important, 9 including three major isoforms of α1-receptors (1a, 1b and 1d) identified in CC. 10 In general, changes in the expression and function of molecules in the relaxation and contraction pathways may cause an imbalance between the contraction and relaxation of CCSM, eventually resulting in ED.
The risk factors for ED include ageing, psychiatric/psychological disorders, smoking, medications, hormonal factors and some chronic diseases like diabetes mellitus and hypertension. 11,12 It is well-known that hypertension is an important worldwide health problem which is more common in elderly people and it can be an independent risk factor for ED. 13,14 CC is a part of the vascular system therefore ED is closely related to cardiovascular diseases such as hypertension.
Indeed, the high prevalence of ED in hypertensive patients was well defined in previous study. 15 Interestingly, ED is an early marker of hypertension and other cardiovascular diseases. 16,17 Although ED and hypertension share many similar underlying pathological mechanisms, including endothelial dysfunction, inflammation and atherosclerosis, 17,18 the mechanism of ED induced by hypertension remains controversial.
Hypertension can impair normal erectile function from both a 'functional' and a 'structural' aspect. In the hypertensive state, blood vessels often undergo remodelling, including in the penis which is also a vascular organ. In the hypertensive rat, studies showed that the penis exhibited morphological changes and tissue remodelling. [19][20][21] With regard to functional activity, a number of vasodilators and vasoconstrictors, such as NO, hydrogen sulphide (H 2 S), angiotensin II (AngII), endothelin-1 (ET-1), were dysregulated or dysfunctional in CC. 22,23 As the strongest vasoconstrictor, ET-1 may also constrict the internal pudendal artery (the major supplying blood vessel of the penis) and reduce blood flow to penis. 24 Moreover, the activation of the RhoA/ROCK pathway was found to contribute to hypertensive ED. 25 In recent years, more studies demonstrated that oxidative stress related to hypertension may act as a pathophysiological insult. 26 The occurrence of oxidative stress is due to an imbalance between the reactive oxygen species (ROS) level and antioxidant activity. ROS includes superoxide, hydrogen peroxide and others. Nicotinamide adenine dinucleotide phosphate (NADPH) oxidase (Nox) is reported to be a major source of ROS in the vascular wall. 27 The excessive ROS will be cleared by antioxidants in the body, such as superoxide dismutase (SOD) and catalase. Otherwise, they will have adverse effects on cells. Recent evidence showed that Nox subunits were up-regulated in vascular smooth muscle cells (VSMC) cultured from the spontaneously hypertensive rat (SHR), which is involved in endoplasmic reticulum (ER) stress. 28,29 However, the role of oxidative stress in CC dysfunction remains unknown.
Currently, the most commonly used PDE5is include sildenafil, vardenafil and tadalafil, which has less side effect and can be delivered on-demand. Although PDE5is are the first-line treatment for ED patients, almost 35% of patients show low or no response to PDE5is, 30 which has perplexed clinicians and patients alike. Recent studies also revealed that PDE5is have less of an effect in hypertensive patients, especially in patients older than 65 years old. 31,32 Actually, responsiveness to PDE5is is dependent on NO/cGMP pathway integrity and PDE5 enzyme expression. Our previous studies reported that castrated rats 33 and diabetes mellitus rats 34 exhibited hypo-responsiveness to PDE5is because PDE5 protein was down-regulated in those rat models. Hence, it will be intriguing to determine the activity of PDE5 in the hypertensive CC.
Thus, the mechanism of ED induced by hypertension remains controversial. Our current study aims to explore the effects of hypertension on CC with emphasis on the NO/cGMP/PDE5 axis and oxidative stress.

| Experimental animal
In total, thirty 12-week-old male spontaneously hypertensive rats (SHR) were used as a hypertension rat model and 30 age-matched male Wistar-Kyoto rats (WKY) were used as a normotensive control. All animals were specific-pathogen-free (SPF) grade. After purchase from Beijing Vital River Laboratory Animal Technology Co., Ltd. (Beijing, China), all animals were kept fed normal chow with a 12 hours day/night light cycle for more than one week to adapt to a new environment. Animal experiments were conducted at the Animal Center of Zhongnan Hospital of Wuhan University and all animal protocols were approved by the Medical Ethics Committee for Experimental animals of Zhongnan Hospital of Wuhan University.
The suffering of experimental rats was kept to a minimum.
According to Fabrizio Sanna study, 35 after a 30-minute habituation period, rats were treated with APO (dissolved in saline, 0.2 mL/rat) by subcutaneous injection. After treatments, rats were observed for 30 minutes in order to count the times of penile erection and yawning episodes. Penile erections were scored when the penis emerged from the penile sheath, which was usually accompanied by penile grooming and hip flexions. Yawnings were scored when the mouth open at least 1-3 seconds of duration, which occasionally accompanied by stretching. Both behavioural responses were recorded by an observer who was not aware of the treatments done.

| Evaluation of erectile response in the rat
Erectile function of all animals was evaluated. As previously reported, 36 rats were anaesthetized with pentobarbital 35 mg/kg intraperitoneal injection. Mean arterial pressure (MAP) was continuously monitored via the carotid artery. MAP and intracavernous pressure (ICP) were recorded through pressure transducers connected to a PowerLab 4/30 data acquisition system (ADInstruments), which was connected in turn to a computer for real-time monitoring of pressure changes. Pressure transducers were calibrated to water before each experiment. The measurement of erectile response elicited by ES of cavernous nerve was performed as previously described. 37

| Tissue preparation
After anaesthetization by isoflurane (in 100% oxygen, 5% for induction and 1.5% for maintenance) at a flow rate of 1 L/min, rats were killed by cervical dislocation and their penes were quickly obtained. The skin, urethra, superficial blood vessels and nerves were removed. Tissue from 10 SHR rats and 10 WKY rats were used for histological experiments, RNA extractions and protein extractions.
Each tissue was cut into three pieces. One piece was placed into 10% formalin for histological studies and another two pieces were snapfrozen in liquid nitrogen and stored at −80°C for RNA extraction and protein extraction. Moreover, tissues from another 10 SHR rats and 10 WKY rats were used for detection of ROS level and antioxidant capacity. Thus, each tissue was cut into five pieces for each different assay. Additionally, tissues from 10 SHR and WKY rats were used to determine the in vitro activity of PDE5 enzyme.

| Total RNA extraction and real-time RT-PCR
Total RNA was extracted from the frozen tissues using TRIzol reagent (Invitrogen) according to the manufacturer's protocol and quantitated by a Nanodrop spectrophotometer (Bio-Rad). Next, 2 μg of RNA was used to perform reverse transcription using the SuperScript II First-Strand Synthesis System (Invitrogen) according to the manufacturer's protocol. For each sample, 100 ng of cDNA was used to perform RT-PCR using a Bio-Rad CFX96 system based on SYBR green incorporation and fluorescence and each determination was repeated independently three times for analysis. For rat CC tissue, the following targets were amplified: PDE5A, NOS isoforms and α1-adrenoreceptor isoforms (α1aAR, α1bAR and α1dAR). Primer TA B L E 1 Primer sequences used to amplify target genes by realtime RT-PCR

Target gene
Primer sequence sequences are shown in Table 1. For relative quantification, gene expression was normalized to expression of the β-actin housekeeping gene and compared by the 2 −ΔΔCT method.

| SDS-PAGE and western blotting analysis
As previously described, 40

| Masson's trichrome staining
As previous described, 40  Specifically, we took the whole area (sinusoidal space not included) of the CC as 100%, and then calculated the per cent area of each component.

| Immunofluorescence microscopy
Rat CC were embedded in Tissue-Tec OCT compound (SakuraFinetek Japan) and snap-frozen. Then, the tissue was sectioned into 10 μm thick slices, thawed and then mounted onto glass slides using a cry-

| PDE5 activity assay
The CC tissue was immediately frozen in liquid nitrogen and crushed into a powder with grinding rods. Then, the powder was placed into a centrifuge tube containing lysis buffer and vortexed for 30 seconds.
After storage at 4°C for 30 minutes, the homogenate was centrifuged for 15 minutes at 12 000 g; then, the supernatant was used for protein quantification using the BCA method and lysates containing 5 mg of total protein were used for the following PDE5 activity assay.

| Statistical analysis
Results are expressed as mean ± SEM for n experiments. Statistical analysis used either the Student's t test with Excel software (two sample treatments compared). P < .05 was considered significant.

| RE SULTS
As described in Table 3, the MAP of the SHR rats was significantly higher than that of the WKY rats, which was (180.60 ± 9.83) mm Hg vs (112.64 ± 9.57) mm Hg (P < .01). The bodyweight of SHR rats was lower than that of WKY rats, which was (257.8 ± 7.5) g vs (293.5 ± 10.5) g (P < .05). The baseline ICP was found to be not significantly different between the two groups.
SHR rats were found to exhibit obvious ED, as demonstrated in  Figure 1A,B), when compared with WKY rats. In addition, the AUC of the ICP curve was decreased at the stimulation frequency of 8 Hz (P < .05), 16 Hz and 32 Hz (P < .01) ( Figure 1C).

Masson's trichrome staining revealed histological changes in SHR
CC. As demonstrated in Figure 2, the percentage of smooth muscle (P < .01) and epithelia (P < .01) were significantly decreased in the CC of SHR rats, while the percentage of collagen fibres (P < .01) were relatively increased. In line with histological results, α-SMA (smooth muscle marker) protein and collagen I (collagen fibre marker) protein were down-regulated (P < .05) and up-regulated (P < .05), respectively. The localization of PDE5 in rat CC was determined using confocal microscopy. As shown in Figure 3, PDE5 was present both in SM and endothelial cells. Moreover, the staining was less in SHR than that in WKY rats. A negative control was performed by omitting the primary antibody ( Figure 3C). Rat lung tissue was used as positive control for PDE5 ( Figure 3D).
Next, the expression of important molecules in the relaxation and contraction pathways of CCSM was determined. As shown in Figure 4A, the mRNA levels of α 1A , α 1B and α 1D were elevated with hypertension by 2.1-fold (P < .05), 3.2-fold (P < .05) and 3.8-fold (P < .01), respectively. In contrast, the mRNA levels of PDE5 (P < .05), eNOS (P < .01) and nNOS (P < .05) were attenuated with higher blood pressure. Consistently, protein levels of α 1A (P < .01), α 1B (P < .01) and α 1D (P < .01) were significantly increased while eNOS (P < .01), nNOS (P < .05) and PDE5 (P < .01) were significantly decreased ( Figure 4B,C) in hypertensive animals. Moreover, the protein level of phosphorylated-PDE5 (Ser-92, P-PDE5) was determined and it was observed that this active form of PDE5 was also decreased (P < .05) in SHR CC tissue ( Figure 4D,E). Correspondingly, the in vitro activity of PDE5 in CC tissue was observed attenuated in hypertensive rats ( Additionally, the ROS level and antioxidant capacity in CC tissue were quantitated. As shown in Figure 5A

| D ISCUSS I ON
The current study demonstrates for the first time that PDE5 expression was down-regulated and its activity was attenuated in hypertensive rat CC. Our novel data also showed that CC tissue from hypertensive rats exhibited an imbalance between ROS and antioxidants in the penis, which may be associated with the occurrence and development of ED induced by high blood pressure. Our study suggested that hypertensive ED patients may be less responsive to TA B L E 3 Bodyweight, MAP and baseline ICP in study rats    47 It has also been reported that α1-receptor isoform distribution is changed in the iliac artery of SHR. 48 Similar to Yono et al, 49 our study showed that all α1-adrenergic receptor isoforms were up-regulated at both the mRNA and protein levels. Therefore, higher expression of α1-adrenergic receptors may lead to higher contractile responsiveness of CCSM and hence attenuated erection.

TA B L E 4 Times of erection and yawning after apomorphine treatment
Interestingly, our current study found that there was a significantly lower expression of PDE5 in SHR CC both at the mRNA and protein level. Similar to a previous report, 33 our immunolocalization study observed PDE5 was almost entirely confined to the endothelial cells and smooth muscle cells of blood vessels and cavernous spaces. In recent decades, the modulation of PDE5 was widely studied. The Burnett lab suggested eNOS could modulate PDE5 expression. 50 They found that in mice with the eNOS gene  was up-regulated in SHR using Western blotting and immunohistochemistry. 55 However, the above two studies only demonstrated the up-regulation of PDE5 at protein expression level. As PDE5 is an enzyme that hydrolyses cGMP, and its phosphorylation form (P-PDE5, at Ser-92) exhibited higher activities, we further determined that the level of P-PDE5 decreased ( Figure 4D,E) and in vitro activity of PDE5 attenuated (Table 5) in SHR, which analyses the alteration of PDE5 in CC tissue of hypertensive rats more comprehensively. Nevertheless, it will be intriguing and more convincing to explore the in vivo functional activity of PDE5 in SHR CC in the future.
It is also noteworthy that CC from SHR exhibited oxidative stress due to an imbalance between excessive ROS levels and impaired antioxidant activity. A previous study has suggested that Nox was a major source of ROS production in penis under the hypertension state. 56 Our present study found that Nox1 and Nox4 were up-regulated in the SHR group ( Figure 6), accounting for an elevated ROS level in CC. BiP (glucose-regulated protein 78/ binding immunoglobulin protein) is a key regulator of ER stress and its up-regulation was often used as a marker of ER stress.
Our current study found hypertensive animals exhibited ER stress with BiP up-regulation ( Figure 6), which was similar to previous reports. 57 (Figure 2), which suggested that CCSM may succumb to apoptosis or fibrosis under high blood pressure. Indeed, CHOP, an apoptosis-related protein, was up-regulated and correlated with ER stress (Figure 2). Increased CHOP could lower the anti-apoptotic protein Bcl-2 and modify the redox process of the cell, making cells tend towards apoptosis. 60 We did show that Bcl-2 decreased but BAX increased in the SHR group ( Figure 6). In general, an imbalance between excessive ROS and impaired antioxidant activity leads to persistent oxidative stress, which might play an important role in the pathophysiology of hypertensive ED.
In summary, severe ED was observed in hypertensive ani- Interestingly, PDE5 was found attenuated in CCSM of hypertensive rats. It is possible that the reduction of NO could be compensated by the decrease of PDE5 to maintain a certain level of cGMP and ultimately maintain partial or satisfactory erectile function, which may explain why not all hypertensive men develop ED. On the other hand, the down-regulation of PDE5 could be contribute to less responsiveness to PDE5is for some hypertensive ED patients.

ACK N OWLED G EM ENTS
We thank the staff at Zhongnan Hospital of Wuhan University for their help in completing the study.

CO N FLI C T O F I NTE R E S T
No potential conflict of interest was reported by the authors.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data used to support the findings of this study are available from the corresponding author upon request.