Introduction

The therapeutic action of phosphodiesterase V inhibitors involves blocking the degradation of guanosine 3', 5'-cyclic monophosphate (cGMP) to guanosine 5' monophosphate (GMP) leading to enhanced relaxation of the corpus cavernosum smooth muscle (CCSM). This molecular mechanism alone has proven sufficient to restore good erectile function to millions of previously impotent men. However, there remains a large percentage of men with erectile dysfunction (ED) for whom this drug is not as effective in restoring erectile function. One group over-represented in these nonresponders is men with diabetes.^1,2 Even the newer more potent phosphodiesterase V inhibitors being developed do not appear significantly more efficacious than Viagra in treating impotence in the diabetic population,^3,4,5 suggesting that separate distinct pathological alteration(s) in diabetic men may exist. The clinical relevance of this problem is highlighted by the fact that about 50% of men with diabetes also have ED⁶ and both type I (insulin-dependent) and type II (noninsulin-dependent) diabetes is associated with ED, with type I men reporting ED slightly more frequently than type II men.⁷

One molecule that has been receiving much attention recently in the diabetic research community is endothelin.⁸ The endothelins (ETs) are a family of 21-amino acid peptides consisting of ET-1, ET-2 and ET-3,⁹ each the product of a separate gene and differing from one another by only a few amino acids.¹⁰ Relative expression of the ET isoforms varies in different tissues,¹¹ with the biological actions of the ETs being determined by their relative binding to ET receptor subtypes. ET-1, the most well-characterized and predominant ET in normal plasma,¹² is synthesized by endothelial cells.⁹ Thus, in the CCSM, ET-1 could be released from the endothelium lining the CCSM itself as well as from the lining of the arteries within the cavernous bodies. Recent studies have suggested that ET-1 plays an important role as a modulator of erectile physiology and dysfunction.^13,14 Interestingly, ET-1 has been shown to cause both relaxation and contraction of vascular smooth muscle.

The two main subtypes of ET receptors are referred to as ETA and ETB and are encoded by separate genes.^15,16 Activation of the ETB receptor has been shown to cause a transient vasodilation,¹⁷ while activation of either the ETA or ETB receptor can cause a sustained contraction of smooth muscle.¹⁸ Thus, the relative expression of these ET receptors may be crucial for defining the contractile state of the CCSM. Although both ETA and ETB receptors exist in mammalian CCSM¹⁹ including human,¹⁴ ET-1-induced CCSM contraction appears to be mediated predominantly by ETA receptors.

In vitro radioligand binding studies initially identified ET-1 and the ETA receptor in rat CCSM, and showed increased binding of these molecules in diabetic CCSM, but ETB receptor binding was not observed.¹⁹ Using the same technique, this same group reported that ETB, but not ETA or ET-1, was significantly upregulated in diabetic rabbits.²⁰ In a separate study, they reported a decrease in ETB receptor binding sites in the cavernosal tissue of hypercholesterolemic rabbits.²¹ However, none of these studies examined whether there were any changes in the physiological response of the CCSM to ET-1 or in the downstream regulators of ET-1-induced contraction.

It has been shown that ET-1 stimulates the small GTPase 'RhoA',²² suggesting that the molecular mechanism by which ET-1 causes smooth muscle contraction may involve the Rho-kinase (ROK) pathway. This hypothesis is supported by the fact that ET-1-induced contraction of stellate cells and portal vein constriction can be almost completely inhibited by the ROK-selective inhibitor Y-27632.²³ ROK is a downstream effector of Rho and has been implicated in calcium sensitization of smooth muscles by lowering the Ca²⁺ concentration required for myosin light-chain phosphorylation, a prerequisite for force generation.²⁴ ROK has been classified into two isoforms known as ROK and ROK.^25,26

In this study, we report that CCSM from diabetic rabbits exhibits an overexpression of both the ETA and ETB receptors and an increased sensitivity to ET-1-induced force generation. In addition, we also provide evidence that ET-1-induced contraction of CCSM is largely mediated by ROK and that the ROK isoform is selectively upregulated in the diabetic rabbit CCSM. These results suggest that a sensitization of CCSM to ET-1 may be a key factor involved in the pathogenesis of ED in diabetics and could at least partly explain the decreased efficacy of Viagra in treating ED in the male diabetic population.

Materials and methods

Diabetic rabbit model. All studies involving animals were approved by the University of Pennsylvania Animal Use Committee. Diabetes was induced by injection of alloxan (100 mg/kg body weight) into the ear vein of 12-week-old male New Zealand White rabbits weighing approximately 6 lbs. The blood glucose level was checked 1 week after the alloxan injection and prior to euthanasia after 6 months and only rabbits with blood glucose levels of 300 mg/dl or higher were used for this study. Age-matched normal rabbits served as controls.

Isolation of rabbit corpus cavernosum. Corpora cavernosa were removed from both normal and diabetic rabbits, cleaned of the adjacent tissue as previously described²⁷ and placed in organ baths containing Tyrode's buffer equilibrated with 95% O₂ and 5% CO₂ for physiological studies or immediately snap frozen in liquid nitrogen for subsequent mRNA and protein analyses. Some corpora with an intact tunica albuginea were also placed in fixative for histological analysis.

Force measurements. The CCSM (one strip per corpora of 50 mg) was prepared for physiological force measurements and L₀ determined as previously described.²⁷ The response to ET-1 (human/porcine from Sigma Chemicals; St Louis, MO, USA) was then determined over the concentration range 0.41–33.3 nM. For studies examining the role of ROK in ET-mediated CCSM contraction, the CCSM was preincubated with Y-27632 (5 M) or vehicle for 15 min prior to stimulation with ET-1.

RNA extraction and RT-PCR. RNA was extracted from frozen CCSM of normal and diabetic rabbits using TRIZOL reagent (Invitrogen; Carlsbad, CA, USA) according to the manufacturer's protocol. The quality of the RNA for each sample was monitored by electrophoresis through formaldehyde containing agarose gels. Reverse transcription (RT) and polymerase chain reaction (PCR) were carried out as previously described²⁸ using primer pairs that specifically amplify ET-1, ETA, ETB, ROK and ROK. In all reactions, -actin was amplified as an internal control.

The sequence of the primers used for ET-1 RT–PCR and the predicted product sizes were: ET-1, upstream 5'GGGCAGCGTACCACAAGAAC-3', downstream 5'CACCCGGCTGGAAGAAGATA3' (442 bp); ETA, upstream 5' TCACCACTCATCGACCCACCAAT3', downstream 5'TGCAGAGAAATACGCCAAAGTCA3'(330 bp); ETB, upstream 5'GCGAGATCCTGACACCTTCC3', downstream 5'CTCCAAATGGCCAGTCTTCT3' (378 bp); ROK, upstream 5'GTGATGGTTACTATGGGCGAGAAT3', downstream 5'GTTAAGAAGGCACAGATGAGAT3' (202 bp); ROK, upstream 5'AAGTAGTTCTTGCATTGG3', downstream 5'TATCATCAGGGAAGGTAAGTG3' (366 bp).The sequences of all primers were based on the published rabbit sequences with the exception of ROK, which had not been cloned, and was desig-ned based upon the known human, mouse, and rat sequences. Quantification of the resulting PCR pro-ducts was performed by scanning densitometry using a Bio Rad GS-700 (Hercules, CA, USA) densitometer as described previously.²⁹ A volume analysis of the bands was performed and the raw data expressed as mean OD unitsarea. All PCR products were sequenced to confirm their identities.²⁹

SDS-polyacrylamide gel electrophoresis (PAGE) and Western blot analysis. Total extractable protein was isolated from frozen corpora of both normal and 6-month diabetic rabbits as previously described.²⁷ Next, equivalent amounts of total protein from normal and diabetic rabbit CCSM were loaded onto 7.5% or 12% SDS-PAGE gels and transferred to Immobilon-P membrane.²⁷ After blocking with 5% fat-free milk for 1 h, the membrane was incubated with a 1 : 1000 dilution of primary antibody ((anti-ETA, cat. # 324758, Calbiochem; San Diego, CA, USA) or (anti-ETB, cat. # 324755, Calbiochem)) or 1 : 2000 dilution ((anti-ROK, cat. # R54520, Clone 21, Transduction Laboratories; Lexington, KY, USA) or (anti-ROK, cat. # R81520, Clone C-19, Transduction Laboratories)) for 2 h at room temperature. The blot was then washed, and further incubated with secondary antibody at 1 : 4000 ((ETA: anti-rabbit Ig, cat. # NA934, Amersham Biosciences; Piscataway, NJ, USA), (ETB: anti-sheep IgG, cat. # A3415, Sigma), (ROK or ROK: anti-mouse Ig, cat. # NA931, Amersham Biosciences)) for 1 h at room temperature. Membranes were washed thoroughly with PBS containing 0.05% Tween 20 between incubations. Antibody reactivity was detected using an enhanced chemiluminescence kit (ECL) from Amersham Biosciences and quantitated by scanning densitometry as described above.

Histochemistry. Paraffin sections were prepared from normal and diabetic rabbit corpora cavernosa as previously described. Briefly, tissue sections were placed in histoclear solution and then washed in descending grades of ethanol and finally with 1 PBS (phosphate buffered saline). Hematoxylin & eosin (H & E) staining and immunostaining were performed essentially as previously described.²⁹ For immunostaining, the slides were incubated for 30 min in 1% BSA to block nonspecific binding and then the sections were incubated for 1–2 h at room temperature with different primary antibodies ((anti-ET-1, cat. # CP44, Clone TR.ET.48.5, 1 : 400, Calbiochem), anti-ETA (1 : 200) or anti-ETB (1 : 200)). Next, the sections were washed 3 in PBS and then treated with secondary antibody ((ET-1: anti-mouse IgG-Cy3, C-2181, Sigma), (ETA: anti-rabbit IgG-FITC, F-0511, Sigma), (ETB: anti-sheep IgG-FITC, F-7634, Sigma)) all at a dilution of 1 : 400 for 1 h, washed 3 with PBS, and mounted with a drop of mounting medium (Aqua-Mount, Lerner Labs; Pittsburgh, PA, USA). Sections were then examined under a fluorescence microscope (Leitz) equipped for epifluorescence illumination. Negative controls were performed in which the primary antibody was omitted.

Statistical analysis. All data are expressed as the means.e. of the mean (s.e.m.) with <0.05 considered statistically significant. The nonpaired Student's t-test was applied using SigmaStat Version 2.03 (SPSS, Chicago, IL, USA).

Results

Comparison of ET-1-induced contraction. Stimulation of normal CCSM by ET-1 induced a characteristically slow generation of force and a sustained contraction (Figure 1a). Compared to normal CCSM, the diabetic CCSM reached maximum contraction much more quickly in response to ET-1 (Figure 1b). However, there was no significant change in the force produced by CCSM from diabetic rabbits in response to 125 mM KCl compared to CCSM from normal rabbits (Figure 1c). An ET-1 dose response curve (0.41–33.3 nM) demonstrated that at each concentration of ET-1 tested, the diabetic CCSM generated more force per milligram tissue than CCSM from normal rabbits with the force at each concentration tested between 20 and 50% greater (Figure 2). In addition, it was determined from the dose response curve that the diabetic CCSM was much more sensitive (EC₅₀: 30.04 vs 90.22 nM) to ET-1 (n=5) (Table 1). In contrast, the sensitivity to the ₁-adrenergic agonist phenylephrine (n=5) was not altered significantly (Table 1).

Figure 1.

Force generation by CCSM in response to stimulation by ET-1 or KCl. Strips of CCSM from normal and diabetic rabbits were equilibrated with 95% O₂–5% CO₂ in Tyrode's buffer at 37°C for 30 min and then L₀ determined. After equilibration at L_o for an additional 15 min, the response to 33.3 nM ET-1 (administered at arrow) was recorded. (a) Normal CCSM, (b) Diabetic CCSM. The tracings are from one representative typical experiment, but similar observations were made in three other pairs of CCSM. The response of CCSM from diabetic and normal rabbits to 125 mM KCl was also determined (c).

Full figure and legend (50K)

Figure 2.

CCSM dose–response curve for ET-1. Strips of CCSM from normal and diabetic rabbits were stimulated with 0.41, 1.23, 3.7, 11.1 and 33.3 nM ET-1. The weight of the strips was determined and the force expressed as grams per 100 mg tissue. Each data point represents the average of results from five different rabbits. The symbol indicates a P value <0.01 and the ^* symbol indicates a P-value <0.05.

Full figure and legend (24K)

Table 1 - EC₅₀ for normal and diabetic CCSM determined from dose response curves in Figure 2.

Full table

Expression of ET-1 and ETA/ETB receptors. Total RNA isolated from normal and diabetic corpora cavernosa was reverse transcribed and PCR performed using primers specific for ET-1 and the ETA and ETB receptors. The expression of ET-1 was not significantly different between normal and diabetic corpora (Figures 3a and 3d) (n=6). The -actin, amplified as an internal control, demonstrated equal amplification. In contrast to ET-1, the ET receptor ETA was expressed at a much higher level (4-fold higher) in CCSM from diabetic rabbits than in the CCSM of normal controls (n=6), while the amplification of -actin was not significantly different (Figures 3b and 3d). The expression of ETB receptor was also increased in diabetic CCSM compared to normal CCSM (n=6) but to a much lesser degree (60% higher) than the change in ETA receptor (Figures 3c and 3d).

Figure 3.

RT–PCR analysis of ET-1, ETA and ETB expression. Representative ethidium bromide-stained agarose gels from RT–PCR using primers specific for ET-1 (a), ETA (b) or ETB (c) performed on total RNA isolated from corpora from normal and diabetic rabbits. The -actin cDNA was amplified as an internal control. Results from two normal rabbits (lanes 1 and 2) and two diabetic rabbits (lanes 3 and 4) are shown. The PCR fragments migrate to the predicted positions of 442 bp for ET-1, 330 bp for ETA, 378 bp for ETB and 96 bp for -actin. PCR products were all sequenced to confirm their identity. The bar graph in (d) summarizes the densitometric analysis of the PCR gels (n=6). The symbol indicates a P value <0.01 and the ^* symbol indicates a P-value <0.05.

Full figure and legend (32K)

Western blot analyses using antibodies specific for the ETA or ETB receptor demonstrated similar increases in expression of these receptors at the protein level as was found at the mRNA level. As shown in Figures 4a and 4c), the expression of ETA was increased about four-fold (n=4) while the expression of ETB was on average increased only about 75% (n=4) (Figures 4b and 4c). Interestingly, the Western blot for ETA also produced a minor band that ran just below the major band and was consistently found in all samples checked. The intensity of this band did not change in response to diabetes.

Figure 4.

Western blot analysis of ETA and ETB. Total protein was extracted from corpora and protein concentration of the supernatants were determined as described in the Methods section. Equal amounts of total extractable protein from two corpora each from normal (lanes 1 and 2) and diabetic rabbits (lanes 3 and 4) were then loaded onto a mini 12% SDS-polyacrylamide gel, separated by electrophoresis and then transferred to Immobilon-P membrane and probed with antibody to ETA (a) or ETB (b) as described in the Methods. The antibodies each reacted with bands of approximately 49 kDa, the sizes predicted from the cloned cDNAs. The bar graph in (c) summarizes the densitometric analysis of the Western blots (n=4). The symbol indicates a P-value <0.01 and the ^* symbol indicates a P-value <0.05.

Full figure and legend (31K)

Immunofluorescence microscopy for ET-1 and ETA/ETB receptors. Histological analysis of H & E stained sections showed no major differences between normal (left column) and diabetic (right column) corpora (Figure 5a). Immunofluorescence microscopy of paraffin sections of normal rabbit corpus cavernosum with ET-1 antibody revealed positive reactions not only to the endothelial cells, but also to the CCSM cells (Figure 5b). Corpus cavernosum from diabetic rabbits showed a similar pattern of staining to the normal with no readily apparent difference in the localization or intensity of ET-1 staining (Figure 5b).

Figure 5.

Histochemical staining of corpora. Representative microscopic fields are shown for H&E staining and immunostaining of paraffin sections from corpora of normal (left column) and diabetic rabbits (right column) for ET-1, ETA and ETB. (a) H & E staining, (b) ET-1 staining, (c) ETA staining, (d) ETB staining. Small white arrows point toward small arteries.

Full figure and legend (223K)

Immunostaining of normal corpora using specific antibodies to the ETA and ETB ET receptors are depicted in Figure 5c and d, respectively. Smooth muscle cells and the endothelium lining the cavernosal spaces reacted to these antibodies. Although the distribution of ETA and ETB did not change in diabetic compared to normal corpus cavernosum, the intensity of ETA (Figure 5c) and ETB (Figure 5d) staining was increased in CCSM from diabetic compared to CCSM from normal rabbits. The increases in ET receptor staining appeared more pronounced in the CCSM than in the arteries (indicated by white arrows).

ET-induced contraction of CCSM involves RhoA/ROK pathway. Since it has been shown that rabbit CCSM expresses significant amounts of ROK³⁰ and also that ET-induced smooth muscle contraction may involve the RhoA-ROK pathway, we pretreated CCSM with the ROK selective inhibitor Y-27632 prior to stimulation by ET-1. After 15 min pretreatment with Y-27632, the corporal tissue was exposed to 33.3 nM ET-1 and force determined. As can be seen in Figure 6, the force per 100 mg tissue elicited by ET-1 was reduced approximately 95% (n=3) by Y-27632 indicating that ET-induced contraction of CCSM is largely mediated by the ROK pathway.

Figure 6.

Effect of Y-27632 on ET-1 stimulated CCSM contraction. Strips of CCSM were equilibrated and L_o determined as described in Figure 1. The CCSM was then preincubated for 15 min with either 5 M Y-27632 or vehicle and then stimulated to contract by addition of 33.3 nM ET-1. The maximum force produced is expressed as grams per 100 mg tissue. The results are an average of five different pairs of rabbits.

Full figure and legend (33K)

Expression of ROK. Since ROK inhibition had such a profound effect on ET-1-induced CCSM contraction, any alteration in the expression of ROK in diabetic CCSM could also be involved in the mechanism of increased ET-1 sensitivity. Therefore, we examined the expression of the two major isoforms of ROK (ROK and ROK) in normal and diabetic corporal tissue. RT–PCR showed that ROK expression was increased approximately three-fold in the diabetic corporal tissue compared to the normal, although -actin expression was not altered (Figures 7b and 7c). In contrast, the expression of the ROK isoform was not significantly altered (Figures 7a and 7c). A similar (2.5-fold) upregulation of the ROK isoform (n=4) was also found at the protein level by Western blotting (Figures 8b and 8c). Similar to our message level finding, no significant change in the expression of the ROK isoform was noted at the protein level (Figures 8a and 8c).

Figure 7.

RT–PCR analysis of ROK and ROK. Representative ethidium bromide-stained agarose gels from RT–PCR for ROK (a) and ROK (b). Results from the corpora of two normal rabbits (lanes 1 and 2) and two diabetic rabbits (lanes 3 and 4) are shown. The PCR fragments appear to migrate to the predicted positions of 202 bp for ROK and 336 bp for ROK. PCR products were all sequenced to confirm their identity. The bar graph in (c) summarizes the densitometric analysis of the PCR gels (n=6). The symbol indicates a P-value <0.01.

Full figure and legend (27K)

Figure 8.

Western blot analysis of ROK and ROK expression. Protein was extracted from corpora and protein concentration of the supernatants were determined as described in the Methods section. Equal amounts of total extractable protein from a corpora of a normal rabbit (lane 1) and a corpora from a diabetic rabbit (lane 2) were then loaded onto a mini 7.5% SDS-polyacrylamide gel, separated by electrophoresis and then transferred to Immobilon-P membrane and probed with antibody to ROK (a) or ROK (b) as described in the Methods. The antibodies each reacted with bands of approximately 160 kDa, the sizes predicted from the cloned cDNAs. The bar graph in (c) summarizes the densitometric analysis of the Western blots (n=6). The symbol indicates a P-value <0.01.

Full figure and legend (34K)

Discussion

The results of this study show that the CCSM from diabetic rabbits exhibits increased contractility in response to ET-1 compared to CCSM from normal rabbits. The CCSM from diabetic rabbits was more sensitive to ET-1 (EC₅₀: 3 vs 9 nM for normal) and generated greater force (20–50%) across all concentrations of ET-1 tested (0.41–33.3 nM) compared to CCSM from normal rabbits. This increased sensitivity appeared specific for the ET-mediated contractile pathway since no change in the sensitivity of force production was reported for phenylephrine. Furthermore, the maximum force, along with time to maximum force, in response to 125 mM KCl was not significantly different between CCSM from diabetic and normal rabbits.

To our knowledge, this is the only study that has compared the contractility of CCSM from diabetic animals to normal ones in response to ET-1. However, our data are in contrast to an earlier report by Christ et al¹⁴ in humans showing no significant differences between diabetic and nondiabetic men with organic ED in the response of CCSM to ET-1. One possible explanation for this discrepancy is that, unlike our rabbit population, their ED patient population may have had other pre-existing etiologies, in addition to diabetes, which complicated the conclusions. Further studies on CCSM from ED patients are required to clarify this.

The concentrations of ET-1 required to cause smooth muscle contraction in our studies, although similar to what has been reported for human CCSM and other types of smooth muscle,^14,31 are well above the normal circulating plasma levels of ET, which are generally in the low picomolar range.³² However, it is now being accepted that plasma ET levels are relatively meaningless, since the majority of ET-1 is secreted abluminally toward the vessel wall.³³ Furthermore, it has been suggested based upon receptor concentrations and the equilibrium dissociation constant of ET receptor complexes that in vivo ET probably binds stoichiometrically to its receptor and thus, most of the ET is probably bound and not in the plasma.³⁴ The strong staining for ET we observed throughout the corpora seems to support this hypothesis. In our present study, we did not measure plasma levels of ET for these reasons, but rather the expression of ET-1. Our results showed no detectable change in the mRNA expression level of ET-1 in CCSM from diabetic rabbits compared to normal CCSM.

The expression of the ETA receptor subtype was upregulated four-fold at both the mRNA and protein levels. Since binding of ET-1 to the ETA receptor results in smooth muscle contraction, overexpression of ETA in the diabetic CCSM could be the mechanism by which the diabetic CCSM becomes sensitized to ET. This observation supports the finding of Bell et al¹⁹ who, using in vitro radioligand binding studies, demonstrated an approximately 20% increase in ETA receptor binding by the CCSM of 3-month streptozotocin-induced diabetic rats compared to age-matched controls. However, this same group later showed that there was no significant difference in ETA receptor binding between 3- or 6-month alloxan-induced diabetic rabbits and controls.²⁰ One possible explanation for the discrepancy between this latter study and our study is that, if as suggested by Frelin and Guedin,³⁴ most of the endogenous ET-1 in the CCSM is bound to the receptor, coupled with the fact that once bound, ET is very hard to wash off, there may be a limited number of unoccupied receptor sites available for additional binding by the selective ET receptor radioligands.

The expression of the ETB receptor subtype was also increased in the rabbit CCSM in response to diabetes (70%), but to a lesser extent than the increase found for ETA. In the diabetic rat study referenced above, the authors were unable to detect any appreciable binding to the ETB receptor in CCSM from either normal or diabetic rats,¹⁹ while the above rabbit study found a three-fold increase in ETB radioligand binding.²⁰

Recent pharmacologic evidence is consistent with the existence of two different populations of ETB receptors, activation of one causing smooth muscle relaxation (ETB1) and the other causing smooth muscle contraction (ETB2).^35,36 These two ETB receptor populations are thought to correspond to the previously described ETB1 and ETB2 alternatively spliced variants of the ETB gene.³⁷ Thus, overexpression of ETB could also contribute to the increased sensitivity and contractility in response to ET found in this study depending upon which isoform is overexpressed. We plan to address this in future studies by stimulating with ET-1 in the presence of ETB1 or ETB2 selective antagonists. It has also now been proposed that there may be two isoforms of the ETA receptor.³⁸ Interestingly, our Western blots with the anti-ETA antibody did show reactivity with a second faster migrating protein, although the relative amount was much less compared to the main reacting band and did not vary in response to diabetes.

Although it is possible that upregulation of the ET receptors alone may be sufficient to account for the increased ET sensitivity of the CCSM from diabetic rabbits, other alterations in the ET signaling pathway downstream of the receptor could also be involved. One likely candidate is the RhoA/ROK pathway, since it has been shown that the ROK selective inhibitor Y-27632 attenuates ET-1-induced increases in portal vein constriction²³ and in vivo inhibits vasoconstriction in the penis.³⁹ Our study confirms the above in vivo studies by showing that pretreatment of rabbit CCSM with 5 M Y-27632 almost completely abolishes ET-1-induced CCSM contraction. Since the ET-1-mediated pathway of CCSM contraction is largely dependent upon the RhoA/ROK pathway, alterations in the expression of ROK could also be involved in the mechanism of the increased sensitivity of CCSM from diabetic rabbits.

RT–PCR performed on corporal tissue from diabetic and normal rabbits using primers that specifically amplify ROK or ROK, the two known isoforms of ROK, revealed that although the expression of ROK was not altered, the expression of ROK was increased about three-fold in the diabetic corpora. Similarly, Western blot analysis revealed a 2.5-fold upregulation of ROK at the protein level. Interestingly, all but a few of the studies conducted on ROK have examined the expression of the ROK isoform. However, the two ROK isoforms, although encoded by different genes, appear structurally very similar. They are both approximately 160 kDa and possess highly conserved pleckstrin homology and kinase domains.²⁵ Since ROK has been shown to increase the phosphorylation level of smooth muscle myosin (necessary for contraction) either by deactivating myosin phosphatase (by phosphorylating it)⁴⁰ or by direct phosphorylation of the smooth muscle myosin,⁴¹ overexpression of ROK in CCSM from diabetic rabbits may also contribute to the increased sensitivity of CCSM from diabetic rabbits to ET.

In conclusion, data from the present study clearly show an increase in sensitivity and force generation of CCSM from diabetic rabbits compared to CCSM from normal rabbits in response to ET-1, the mechanism of which involves an upregulation of ET receptor expression. Furthermore, the ET-1-induced CCSM contractile pathway is largely dependent upon the RhoA/ROK pathway and the ROK isoform is upregulated in diabetic CCSM. Thus, an increased CCSM tone, modulated by sensitization of the ET -mediated contractile pathway, may be a key component of the molecular mechanism of diabetes-induced ED and may at least partially explain the decreased efficiency of phosphodiesterase V inhibitors in treating ED in the diabetic population.

Although the potential clinical role of ET-1 antagonists in treating ED appears promising,⁴² a preliminary study by Kim et al⁴³ found that, in a pilot study using the ETA receptor selective antagonist BMS-193884, this compound failed to enhance erectile function in men with mild to moderate ED. Although this trial found no significant effect of BMS-193884 on ED, this does not rule out the possibility that ET antagonists can have therapeutic use for ED for a number of reasons.

First, this study did not examine the effect of the ET antagonist specifically in diabetic men with ED. Our study only reports the increase in ET sensitivity in diabetic rabbits. Thus, nondiabetic patients with ED would not be expected to benefit from this treatment. Second, ET antagonists by themselves may not be able to compensate for any cGMP deficiency but rather when used in combination with a PDE5 inhibitor may have an effect. Third, the ET antagonist used in this study was selective for ETA. Since the ETB2 receptors also can mediate contraction, they may compensate for the decrease in ETA binding. Fourth, the BMS-193884 antagonist appears to be more selective for blocking ET-2 binding than ET-1 binding and ET-1 is the physiologically most relevant form of ET in human CCSM. Interestingly, a preliminary report shows that in vivo chronic ET receptor antagonism using RO-48-5695 (a combined ETA/ETB receptor antagonist) can preserve erectile function in diabetic rats.⁴⁴

References

1. McMahon CG. Erectile dysfunction. Med J Aust 2000; 173: 492–497. | PubMed |
2. Merrick GS et al. Efficacy of sildenafil citrate in prostate brachytherapy patients with erectile dysfunction. Urology 1999; 53: 1112–1116. | Article | PubMed | ChemPort |
3. Porst H. IC351 (tadalafil, Cialis): update on clinical experience. Int J Impot Res 2002; 14: S57–S63. | Article | PubMed | ISI |
4. Porst H et al. The efficacy and tolerability of vardenafil, a new, oral, selective phosphodiesterase type 5 inhibitor, in patients with erectile dysfunction: the first at-home clinical trial. Int J Impot Res 2001; 13: 192–199. | Article | PubMed | ISI | ChemPort |
5. Goldstein I et al. Vardenafil, a new highly selective PDE5 inhibitor, improves erectile function in patients with diabetes mellitus. Diabetes 2001; 50: A114.
6. Krane RJ, Goldstein I, Saenz IdT. Impotence. N Engl J Med 1989; 321: 1648–1659. | PubMed | ChemPort |
7. Fedele D et al. Erectile dysfunction in type 1 and type 2 diabetics in Italy. On behalf of Gruppo Italiano Studio Deficit Erettile nei Diabetici. Int J Epidemiol 2000; 29: 524–531. | Article | PubMed | ISI | ChemPort |
8. Hopfner RL, Gopalakrishnan V. Endothelin: emerging role in diabetic vascular complications. Diabetologia 1999; 42: 1383–1394. | Article | PubMed | ISI | ChemPort |
9. Yanagisawa M et al. A novel potent vasoconstrictor peptide produced by vascular endothelial cells. Nature 1988; 332: 411–415. | Article | PubMed | ISI | ChemPort |
10. Luscher TF, Barton M. Endothelins and endothelin receptor antagonists: therapeutic considerations for a novel class of cardiovascular drugs. Circulation 2000; 102: 2434–2440. | PubMed | ISI | ChemPort |
11. Firth JD, Ratcliffe PJ. Organ distribution of the three rat endothelin messenger RNAs and the effects of ischemia on renal gene expression. J Clin Invest 1992; 90: 1023–1031. | PubMed | ISI | ChemPort |
12. Usuki S, Kondoh K, Kubo T. Plasma endothelin and LH-RH, LH, FSH, prolactin, progesterone, 17alpha-hydroxyprogesterone, estrone, 17beta-estradiol, delta4-androstenedione, testosterone, active renin, angiotensin-II and ANP levels in blood and LH, estrone and 17beta-estradiol and pregnanediol levels in urine of normal cycling women. J Cardiovasc Pharmacol 2000; 36: S21–S427.
13. Saenz IdT et al. Endothelin: localization, synthesis, activity, and receptor types in human penile corpus cavernosum. Am J Physiol 1991; 261: H1078–H1085. | PubMed | ChemPort |
14. Christ GJ, Lerner SE, Kim DC, Melman A. Endothelin-1 as a putative modulator of erectile dysfunction: I. Characteristics of contraction of isolated corporal tissue strips. J Urol 1995; 153: 1998–2003. | PubMed | ISI | ChemPort |
15. Sakamoto A et al. Cloning and functional expression of human cDNA for the ETB endothelin receptor. Biochem Biophys Res Commun 1991; 178: 656–663. | Article | PubMed | ChemPort |
16. Nakamuta M et al. Cloning and sequence analysis of a cDNA encoding human non-selective type of endothelin receptor. Biochem Biophys Res Commun 1991; 177: 34–39. | PubMed | ISI | ChemPort |
17. Matsuda H et al. Involvement of cyclo-oxygenase-generated vasodilating eicosanoid(s) in addition to nitric oxide in endothelin-1-induced endothelium-dependent vasorelaxation in guinea pig aorta. Heart Vessels 1993; 8: 121–127. | Article | PubMed | ISI | ChemPort |
18. Sumner MJ et al. Endothelin ETA and ETB receptors mediate vascular smooth muscle contraction. Br J Pharmacol 1992; 107: 858–860. | PubMed | ChemPort |
19. Bell CR et al. The density and distribution of endothelin 1 and endothelin receptor subtypes in normal and diabetic rat corpus cavernosum. Br J Urol 1995; 76: 203–207. | PubMed | ISI | ChemPort |
20. Sullivan ME et al. Alterations in endothelin B receptor sites in cavernosal tissue of diabetic rabbits: potential relevance to the pathogenesis of erectile dysfunction. J Urol 1997; 158: 1966–1972. | Article | PubMed | ISI | ChemPort |
21. Sullivan ME et al. Down-regulation of endothelin-B receptor sites in cavernosal tissue of hypercholesterolaemic rabbits. Br J Urol 1998; 81: 128–134. | Article |
22. Sakurada S et al. Rho activation in excitatory agonist-stimulated vascular smooth muscle. Am J Phys 2001; 281: C571–C578.
23. Kawada N, Seki S, Kuroki T, Kaneda K. ROCK inhibitor Y-27632 attenuates stellate cell contraction and portal pressure increase induced by endothelin-1. Biochem Biophys Res Commun 1999; 266: 296–300. | Article | PubMed | ISI | ChemPort |
24. Adelstein RS, Eisenberg E. Regulation and kinetics of the actin–myosin–ATP interaction. Annu Rev Biochem 1980; 49: 921–956. | Article | PubMed | ISI | ChemPort |
25. Leung T, Chen XQ, Manser E, Lim L. The p160 RhoA-binding kinase ROK alpha is a member of a kinase family and is involved in the reorganization of the cytoskeleton. Mol Cell Biol 1996; 16: 5313–5327. | PubMed | ISI | ChemPort |
26. Matsui T et al. Rho-associated kinase, a novel serine/threonine kinase, as a putative target for small GTP binding protein Rho. EMBO J 1996; 15: 2208–2216. | PubMed | ISI | ChemPort |
27. Chang S et al. Enhanced force generation by corpus cavernosum smooth muscle in rabbits with partial bladder outlet obstruction. J Urol 2002; 167: 2636–2644. | Article | PubMed | ChemPort |
28. DiSanto ME, Cox RH, Wang Z, Chacko S. NH₂-terminal-inserted myosin II heavy chain is expressed in smooth muscle of small muscular arteries. Am J Physiol 1997; 272: C1532–C1542. | PubMed | ISI | ChemPort |
29. DiSanto ME et al. Expression of myosin isoforms in smooth muscle cells in the corpus cavernosum penis. Am J Physiol 1998; 275: C976–C987. | PubMed | ChemPort |
30. Chang S et al. Expression of myosin light chain kinase and Rho-associated kinase in corpus cavernosum smooth muscle. Biophys J 2000; 78: 139. | Article |
31. Olsen UB, Weis J. Rat gastric relaxation induced by stimulation of endothelin-1 selective receptors. Regul Pept 1992; 39: 113–119. | Article |
32. Sarman B et al. Circulating plasma endothelin-1, plasma lipids and complications in Type 1 diabetes mellitus. Diabetes Nutr Metab 2000; 13: 142–148.
33. Morganti A et al. Plasma endothelin levels: a meaningless number?. J Cardiovasc Pharmacol 2000; 35: S21–S23. | Article | PubMed | ChemPort |
34. Frelin C, Guedin D. Why are circulating concentrations of endothelin-1 so low? Cardiovasc Res 1994; 28: 1613–1622. | PubMed | ISI | ChemPort |
35. Mazzoni MR et al. Suc-[Glu9,Ala11,15]-endothelin-1 (8-21), IRL 1620, identifies two populations of ET(B) receptors in guinea-pig bronchus. Br J Pharmacol 1999; 127: 1406–1414. | Article |
36. Zuccarello M, Boccaletti R, Rapoport RM. Does blockade of endothelinB1-receptor activation increase endothelinB2/endothelinA receptor-mediated constriction in the rabbit basilar artery? J Cardiovasc Pharmacol 1999; 33: 679–684. | Article |
37. Shyamala V, Moulthrop TH, Stratton-Thomas J, Tekamp-Olson P. Two distinct human endothelin B receptors generated by alternative splicing from a single gene. Cellular Mol Biol Res 1994; 40: 285–296.
38. Nishiyama M et al. Pharmacological heterogeneity of both endothelin ETA- and ETB-receptors in the human saphenous vein. Jpn J Pharmacol 1995; 69: 391–398.
39. Mills TM et al. Effect of Rho-kinase inhibition on vasoconstriction in the penile circulation. J Appl Physiol 2001; 91: 1269–1273. | PubMed | ISI | ChemPort |
40. Kimura K et al. Regulation of myosin phosphatase by Rho and Rho-associated kinase (Rho-kinase). Science 1996; 273: 245–248. | Article | PubMed | ISI | ChemPort |
41. Amano M et al. Phosphorylation and activation of myosin by Rho-associated kinase (Rho-kinase). J Biol Chem 1996; 271: 20246–20249. | Article | PubMed | ISI | ChemPort |
42. Khan MA et al. Normal and pathological erectile function: the potential clinical role of endothelin-1 antagonists. Curr Drug Targets 2000; 1: 247–260.
43. Kim NN et al. Pilot study of the endothelin-A receptor selective antagonist BMS-193884 for the treatment of erectile dysfunction. J Androl 2002; 23: 76–83.
44. Yang D et al. The role of endothelin antagonism for the preservation of erectile function in diabetic rats. J Urol 2001; 165: 230.

Acknowledgements

This work was supported by Grants DK55529 and DK55042 from the National Institutes of Health.