Andrology
Phosphodiesterase type 5 in men with
vasculogenic and post-radical prostatectomy
erectile dysfunction: is there a molecular
difference?
Serkan Karakus , Biljana Musicki and Arthur L. Burnett
Department of Urology, Johns Hopkins School of Medicine, The James Buchanan Brady Urological Institute, Baltimore,
MD, USA
Objectives
To clarify the molecular basis of penile erection at the human
level and distinguish the mechanisms underlying vasculogenic
and post-radical prostatectomy (RP) erectile dysfunction (ED)
subtypes.
Patients and Methods
Erectile tissue was obtained from men without history of ED
who underwent penile surgery for Peyronie’s disease (control
group, n = 5) and from men with ED who underwent penile
prosthesis implantation (n = 17). ED was categorized into
vasculogenic (n = 8) and post-RP (n = 9) subtypes. Penile
erectile tissue samples were collected for molecular analyses of
protein expressions of neuronal and endothelial isoforms of
nitric oxide synthase (nNOS and eNOS, respectively), phospho-
nNOS (Ser-1412), phospho-eNOS (Ser-1177), phospho-protein
kinase B (Ser-473), phosphodiesterase type 5 (PDE5), a-smooth
muscle actin, phospho-myosin phosphatase target subunit 1,
RhoA/Rho-associated protein kinase (ROCK)-a, ROCK-b,
4-hydroxy-2-nonenal, and nNOS and eNOS uncoupling by
Western blot.
Results
Vasculogenic ED was characterized by decreased eNOS
protein expression and eNOS and nNOS phosphorylation
on their activatory sites (Ser-1177 and Ser-1412,
Introduction
Erectile dysfunction (ED) affects >18 million men in the USA
and is predicted to rise to 322 million men worldwide by the
year 2025 [1]. Vasculogenic ED involves ~80% of ED cases
and is commonly associated with vascular disorders such as
hypertension, cardiovascular disease, diabetes mellitus,
smoking and dyslipidaemia [2]. Studies in men with
vasculogenic ED have shown decreased endothelium-
mediated and neurogenic relaxation, increased adrenergic
BJU Int 2018; 122: 1066–1074
wileyonlinelibrary.com
respectively), uncoupled eNOS, upregulated PDE5 protein
expression, increased ROCK activity, and increased
oxidative stress in erectile tissue. Post-RP ED was
characterized by decreased nNOS protein expression,
increased nNOS phosphorylation on its activatory site
(Ser-1412), uncoupled nNOS, downregulated PDE5
protein expression, and increased oxidative stress in
erectile tissue.
Conclusion
The mechanisms of vasculogenic and post-RP ED in the
human penis involve derangements in constitutive nitric
oxide synthase function, PDE5 protein expression and
ROCK activity, and increased oxidative stress, which
conceivably provide a molecular basis for chronically
reduced nitric oxide bioavailability and increased smooth
muscle contraction contributing to erectile impairment.
Selective differences in PDE5 protein expression suggest
distinct molecular mechanisms are in play for these ED
subtypes.
Keywords
PDE-5 inhibitors, nitric oxide synthase, phosphorylation,
uncoupling, oxidative stress, ROCK protein kinases,
#erectiledysfunction
contraction, apoptosis and fibrosis, and decreased smooth
muscle/collagen ratio in the erectile tissue [3,4], and increased
oxidative stress in the peripheral blood [5]. Neurogenic ED
represents 10–19% of all causes of ED and, at the peripheral
level, is mostly a consequence of pelvic surgeries such as
radical prostatectomy (RP) [6]. Post-RP ED is typically
attributed to injury to the cavernous nerve [2], although
damage to accessory or aberrant pudendal arteries can add a
vasculogenic component [7]. Penile denervation causes
decreased neurogenic relaxation, increased adrenergic
© 2018 The Authors
BJU International © 2018 BJU International | doi:10.1111/bju.14433
Published by John Wiley & Sons Ltd. www.bjui.org
 Mechanisms of vasculogenic vs post-radical prostatectomy erectile dysfunction
contraction, apoptosis and fibrosis in human erectile tissue
[3,4].
Regardless of ED aetiology, phosphodiesterase type 5 (PDE5)
inhibitors are commonly prescribed as first-line ED
treatment; however, clinical trials with PDE5 inhibitors have
not shown that they have a major impact on erectile function
recovery after RP, in contrast to their more beneficial effect in
vasculogenic ED [8]; thus, the management of post-RP ED
requires the implementation of adjunctive neuroprotective or
neuroregenerative therapies that will be defined and
developed by ongoing investigative studies of neurogenic
molecular pathways.
The nitric oxide (NO)-dependent cGMP pathway is the major
signalling mechanism responsible for smooth muscle
relaxation in the corpora cavernosa. The RhoA/Rho-
associated protein kinase (ROCK) contractile pathway closely
interacts with the NO pathway and contributes to the
regulation of erection responses [2]; however, at the human
level, the mechanism of ED and, specifically, the mechanisms
underlying vasculogenic and post-RP ED subtypes are
unclear. Clarifying the molecular basis of penile erection at
the human level may serve not only to distinguish ED
subtypes but also to guide further cause-specific therapies for
ED.
In the present study, we hypothesized that oxidative stress,
dysregulation of the NO pathway, and derangements in PDE5
signalling and the RhoA/ROCK pathway are major
determinants of ED in men. We performed an extensive
molecular evaluation of these major erection regulatory
factors in erectile tissue of men with vasculogenic and post-
RP ED.
Patients and Methods
Patient Population and Tissue Collection
Tissue collection and clinical history review were performed
with approval from the institutional review board of the
Johns Hopkins Medical Institutions, and written informed
consent was obtained from all patients. A total of 22 human
erectile tissue samples were selected. Erectile tissue was
obtained from men without histories of ED (Sexual Health
Inventory for Men [SHIM] score >21 and verified by penile
duplex ultrasonography) who underwent penile surgery for
Peyronie’s disease (control group) and with ED (SHIM score
<21) who underwent penile prosthesis implantation. ED
patients were categorized based on ED aetiology: vasculogenic
ED and ED secondary to RP (post-RP ED), the latter being
used to model neurogenic ED. For the vasculogenic group, we
selected patients without diabetes mellitus (which is
associated with both vasculogenic and neurogenic ED), and
for the post-RP ED group, we selected patients without
diabetes mellitus and cardiovascular disease who developed
ED only after undergoing RP. All patients in vasculogenic
and post-RP ED groups who underwent penile prosthesis
surgery had PDE5 inhibitor failures and the therapies were
stopped at least 3 months before penile implantation and
tissue collection, lessening the possibility of a PDE5 inhibitor
effect on PDE5 protein expression. Cavernous nerve-sparing
surgery was reportedly performed in seven out of nine
patients with post-RP ED, of whom five had both cavernous
nerves spared; however, we assumed some element of
neuropraxia had occurred in these patients. The tissues were
collected at a mean (range) of 73.8  12.8 (21–130) months
after RP surgery. Patient characteristics are summarized in
Table 1. All penile prosthesis surgeries were performed by the
same surgeon (A.L.B.). Penile erectile tissue samples (0.5–
1 cm) were obtained bilaterally from the proximal corpus
cavernosum, at a region separate from fibrotic tunica in
patients with Peyronie’s disease. Samples were collected
during surgery, immediately placed in cold saline, and then
frozen in liquid nitrogen for molecular studies.
Western Blot
Penile tissue was homogenized as reported previously [9–11].
NO synthases (NOSs) were partially purified as described
previously [9–11]. Membranes with partially purified NOSs
were probed with anti-phospho-neuronal (n)NOS (Ser-1412)
antibody (polyclonal rabbit, kindly provided by Dr Solomon
Snyder, Johns Hopkins University, Baltimore, MD, USA) at
1:6 000 dilution or anti-phospho-endothelial (e)NOS (Ser-
1177) antibody (polyclonal rabbit; Cell Signaling Technology,
Beverly, MA, USA) at 1:450 dilution (for phospho-nNOS and
phospho-eNOS analyses) [9]. After probing for
phosphoforms, membranes were stripped and probed with
antibodies against nNOS (polyclonal rabbit, 1:9 000 dilution;
Dr Solomon Snyder) or eNOS (monoclonal mouse, 1:500
dilution; Santa Cruz Biotechnology, Santa Cruz, CA, USA,).
phospho-nNOS (Ser-1412) and phospho-eNOS (Ser-1177)
densities were normalized relative to those of nNOS and
eNOS, respectively, in partially purified samples. Dimeric and
monomeric forms of nNOS and eNOS were measured by low
temperature SDS electrophoresis, as described previously
[9–11], using anti-nNOS antibody (polyclonal rabbit, 1:1 000
dilution; Cell Signaling Technology) and anti-eNOS antibody
(monoclonal mouse, 1:500 dilution; Santa Cruz
Biotechnology). For analysis of total nNOS (polyclonal rabbit
antibody, 1:1 000 dilution; Cell Signaling Technology), total
eNOS (monoclonal mouse antibody, 1:500 dilution; Santa
Cruz Biotechnology), ROCKa (monoclonal mouse antibody,
1:1 000 dilution; BD Transduction Laboratories, San Jose, CA,
USA), ROCKb (monoclonal mouse antibody, 1:1 000 dilution;
Santa Cruz Biotechnology), phospho-myosin phosphatase
target subunit 1 (MYPT1; Thr-696; polyclonal rabbit
antibody, 1:1 000 dilution; Santa Cruz Biotechnology), PDE5
(monoclonal mouse antibody, 1:450 dilution; BD
© 2018 The Authors
BJU International © 2018 BJU International 1067
Karakus et al.

Table 1 Patient group characteristics.

No ED Vasculogenic ED Post-RP ED P

n 5 8 9
Age, years (range) 59.6  0.9 (58–63) 59.6  1.6 (53–64) 61.5  2.5 (50–71) 0.74
Diabetes, n (%) 0 (0) 0 (0) 0 (0) –
Hypertension, n (%) 0 (0) 5 (62.5) 0 (0) –
Hyperlipidaemia, n (%) 3 (60) 3 (37.5) 2 (22.2) –
Cardiovascular disease, n (%) 0 (0) 3 (37.5) 0 (0) –
BMI, kg/m2 27.9  1.1 27.6  1.1 25.7  0.6 0.19
Smoking, n (%) 0 (0) 1 (12.5) 0 (0) –
SHIM score (range) 21.8  1.3 (21–24) 9.1  1.6 (2–14)* 2.4  0.8 (1–6)*,** 0.001
Preoperative SHIM score (range) – – 22  0.8 (18–25)*** 0.001
Penile duplex ultrasonography results
Peak systolic velocity, cm/s (range) 64.2  6 (29–104) 21.3  1.5 (11–30)* – 0.001
Resistive index (range) 0.94  0.03 (0.81–1.13) 0.78  0.04 (0.4–1.29)* – 0.01

ED, erectile dysfunction; RP, radical prostatectomy; SHIM, Sexual Health Inventory for Men. *P < 0.05 vs No ED, **P < 0.05 vs Vasculogenic ED, and ***P < 0.05 vs post-RP ED.
No-ED: Patients with Peyronie’s disease without history of ED.

Transduction Laboratories), 4-hydroxy-2-nonenal (4-HNE;
polyclonal rabbit antibody, 1:2 000 dilution; Alpha Diagnostic
International, San Antonio, TX, USA), phospho-Akt (Ser-473;
polyclonal rabbit antibody, 1:1 000 dilution; Cell Signaling
Technology), and Akt (polyclonal rabbit antibody, 1:1 000
dilution; Cell Signaling Technology), a separate set of
homogenates (70–100 lg) was used without purification, and
standardized to glyceraldehyde 3-phosphate dehydrogenase
(GAPDH; monoclonal mouse antibody, 1:1 000 dilution;
Santa Cruz Biotechnology) [9–11]. Band densities were
quantified using NIH Image 1.29. The analysis of 4-HNE was
a densitometric composite of all proteins in each lane. nNOS
and eNOS uncoupling were represented inversely as a ratio of
functional nNOS and eNOS dimers to non-functional nNOS
and eNOS monomers, respectively. The ratio was determined
in terms of arbitrary units and expressed relative to the ratio
for controls.
compared with erectile tissue in the control group (Fig. 2A).
Similarly, Akt phosphorylation on Ser-473, which directly
regulates phospho-eNOS (Ser-1177) expression [13], was
reduced (P < 0.05) in vasculogenic ED erectile tissue and not
changed in post-RP ED erectile tissue, compared with the
control group (Fig. 2B). Protein expression of total eNOS was
similarly decreased (P < 0.05) in vasculogenic ED erectile
tissue, and not changed in post-RP ED erectile tissue,
compared with erectile tissue in the control group (Fig. 2C).
The ratio of nNOS dimer (functional nNOS)/monomer (non-
functional nNOS) was decreased (P < 0.05) in post-RP ED
erectile tissue, and not changed in vasculogenic ED erectile
tissue, compared with erectile tissue in the control group
(Fig. 3A). The ratio of eNOS dimer (functional eNOS)/
monomer (non-functional eNOS) was decreased (P < 0.05) in
vasculogenic ED erectile tissue, and not changed in post-RP
ED erectile tissue, compared with the control group (Fig. 3B).

Statistical Analysis Protein expression of 4-HNE, a marker of oxidative stress

Data were expressed as means  SEM [12]. Statistical analysis
was performed using a modified Student’s t-test. P values
<0.05 were taken to indicate statistical significance.
Results
Phosphorylation of nNOS on its stimulatory (Ser-1412)
regulatory site was decreased (P < 0.05) in vasculogenic ED
erectile tissue and increased (P < 0.05) in post-RP ED erectile
tissue, compared with erectile tissue in the control group
(Fig. 1A). Protein expression of total nNOS was decreased (P
< 0.05) in post-RP ED erectile tissue, and not changed in
vasculogenic ED erectile tissue, compared with the control
group (Fig. 1B).
Phosphorylation of eNOS on its stimulatory (Ser-1177)
regulatory site was decreased (P < 0.05) in vasculogenic ED
erectile tissue and not changed in post-RP ED erectile tissue,
[11], was increased (P < 0.05) in both vasculogenic and post-
RP ED erectile tissues, compared with the control group
(Fig. 4).
Protein expression of phospho-MYPT1 (Thr-696), reflecting
RhoA/ROCK activity [14], was increased (P < 0.05) in
vasculogenic ED erectile tissue, and not changed in post-RP
ED erectile tissue, compared with erectile tissue in the control
group (Fig. 5A). Protein expressions of ROCKa and ROCKb
in vasculogenic and post-RP ED erectile tissues did not differ
from values in the control group (Fig. 5B,C).
Protein expression of PDE5 was increased (P < 0.05) in
vasculogenic ED erectile tissue and decreased (P < 0.05) in
post-RP ED erectile tissue compared with erectile tissue in
the control group (Fig. 6A). Protein expression of a-smooth
muscle actin, an indicator of smooth muscle content [15],
was not different in post-RP ED or vasculogenic ED erectile
tissue, compared with the control group (Fig. 6B).

© 2018 The Authors

1068 BJU International © 2018 BJU International
 Mechanisms of vasculogenic vs post-radical prostatectomy erectile dysfunction

Fig. 1 Protein expression of phospho-neuronal nitric oxide synthase (nNOS; Ser-1412) (A) is decreased in the erectile tissue of men with vasculogenic
erectile dysfunction (ED) and increased in the erectile tissue of men with post-radical prostatectomy (RP) ED, compared with that of the control group.
Protein expressions of total nNOS (B) is decreased in the erectile tissue of men with post-RP ED, compared with that of the control group. Upper panels
are representative Western immunoblots and lower panels are densitometric analyses of phospho-nNOS (Ser-1412)/nNOS and total nNOS/
glyceraldehyde 3-phosphate dehydrogenase (GAPDH) in the penis of men in the control group, men with vasculogenic ED, and men with post-RP ED.
Each bar represents the mean  SEM of five to nine patients. *P < 0.05 vs control group.

A kDa B kDa

nNOS 160
Phospho-nNOS 160
nNOS 160 GAPDH 37

*
Phospho-nNOS Ser-1412/nNOS

150
100

nNOS/GAPDH
(% of Control)

(% of Control)
100 *
*
50
50

0 0

ED sc.
.

ED sc.

nt
nt

ED -RP
ED -RP

Co

Va
Co

Va

st
st

Po
Po

Fig. 2 Protein expressions of phospho-endothelial nitric oxide synthase (eNOS; Ser-1177) (A), phospho-Akt (Ser-473) (B) and total eNOS (C) are
decreased in the erectile tissue of men with vasculogenic erectile dysfunction (ED), compared with that of the control group. Upper panels are
representative Western immunoblots and lower panels are densitometric analyses of phospho-eNOS (Ser-1177)/eNOS, phospho-Akt (Ser-473)/Akt, and
total eNOS/glyceraldehyde 3-phosphate dehydrogenase (GAPDH) in the penis of men in the control group, men with vasculogenic ED, and men with
post-RP ED. Each bar represents the mean  SEM of five to nine patients. *P < 0.05 vs control group.

A kDa B kDa C kDa

Phospho-eNOS 140 Phospho-Akt 60 eNOS 140

eNOS 140 Akt 60 GAPDH 37

Phospho-eNOS Ser-1177/eNOS

eNOS/GAPDH (% of Control)

100 100 100

Phospho-Akt Ser-473/Akt

* * *
(% of Control)

(% of Control)

50 50 50

0 0 0
ED -RP

ED -RP

ED -RP
ED sc.

ED sc.

ED sc.
.

.
nt

nt

nt
Va

Va

Va
Co

Co

Co
st

st

st
Po

Po

Po

selectively decreased constitutive NOS expressions,

Discussion functionally inactivated constitutive NOSs, increased oxidative
The results of the present study indicate that mechanisms stress, and derangements in PDE5 protein expression and
associated with vasculogenic and post-RP ED in men involve ROCK activity in the erectile tissue. Selective differences in

© 2018 The Authors

BJU International © 2018 BJU International 1069
Karakus et al.

Fig. 3 Neuronal nitric oxide synthase (nNOS) uncoupling (A) is increased (decreased nNOS dimers) in the erectile tissue of men with post-radical
prostatectomy (RP) erectile dysfunction (ED), compared with that of the control group. endothelial nitric oxide synthase (eNOS) uncoupling (B) is
increased (decreased eNOS dimers) in the erectile tissue of men with vasculogenic ED, compared with that of the control group. Upper panels are
representative Western immunoblots and lower panels are densitometric analyses of nNOS dimers/nNOS monomers and eNOS dimers/eNOS
monomers in the penis of men in the control group, men with vasculogenic ED, and men with post-RP ED. Each bar represents the mean  SEM of five to
nine patients. *P < 0.05 vs control group.

A kDa B kDa
nNOS Dimers 320 eNOS Dimers 280
nNOS Monomers 160 eNOS Monomers 140

nNOS Dimers/Monomers

eNOS Dimers/Monomers
100 100
* *
(% of Control)

(% of Control)
50 50

0 0

Post-RP
Cont.
Post-RP

Vasc.
Cont.
Vasc.

ED
ED

ED
ED

Fig. 4 Protein expression of 4-hydroxy-2-nonenal (4-HNE) is increased in
the erectile tissue of men with vasculogenic erectile dysfunction (ED) and
post-radical prostatectomy (RP) ED compared with that of the control
group. Upper panels are representative Western immunoblots and lower
panels is a densitometric analysis of 4-HNE/glyceraldehyde 3-phosphate
dehydrogenase (GAPDH) in the penis of men in the control group, men
with vasculogenic ED, and men with post-RP ED. Each bar represents the
mean  SEM of five to nine patients. *P < 0.05 vs control group.
4-HNE
GAPDH
150
* lasts longer than the short-term calcium transient,
* contributing to both the initiation and maintenance phases of
4-HNE/GAPDH
(% of Control)
protein expression, decreased eNOS and nNOS
phosphorylation on their activation sites (Ser-1177 and Ser-
1412, respectively), uncoupled eNOS, upregulated PDE5
protein expression, and increased ROCK activity, while post-
RP ED involves decreased total nNOS protein expression,
increased nNOS phosphorylation on its activatory site (Ser-
1412), uncoupled nNOS, and downregulated PDE5 protein
expression. Common mechanisms for both vasculogenic and
kDa post-RP ED include increased oxidative stress in the erectile
250 tissue, conceivably originating in some part from
dysfunctional, uncoupled, eNOS and nNOS.
50
The major molecular mechanisms underlying the function of
constitutive NOS isoenzymes are phosphorylation and
20 uncoupling. Phosphorylation of eNOS and nNOS on
activatory sites (Ser-1177 and Ser1412, respectively) activate
37 the enzyme’s catalytic functions by reducing their calcium
requirement and facilitating electron transfer [16,17]. This
phospho-modification on both constitutive NOS isoenzymes

penile erection [13–18]. Functional dimerization of NOSs is

100
required for NO-producing activity of the enzymes. Under
pathological conditions resulting in a loss of the functional
50 enzyme’s dimerization, NOSs can transform into prooxidants,
generating predominantly superoxide anion rather than NO
(‘NOS uncoupling’) [19,20]. In the present study, we found
0 that erectile tissue of men with vasculogenic ED exhibited
ED sc.
.

ED - RP

downregulated phospho-eNOS (Ser-1177), conceivably

nt

Va
Co

st

through downregulation of its main regulatory Akt pathway

Po

[13], and also downregulated phospho-nNOS (Ser-1412),

implying the impairment of both NOS isoenzymes involved
molecular profiles for vasculogenic and post-RP ED suggest in vasodilation [20,21]; a further decrease in endothelial NO
there are distinct molecular mechanisms for these ED bioavailability was achieved by a loss of functional eNOS
subtypes: vasculogenic ED involves decreased total eNOS dimerization. By contrast, in post-RP ED erectile tissue,

© 2018 The Authors

1070 BJU International © 2018 BJU International
 Mechanisms of vasculogenic vs post-radical prostatectomy erectile dysfunction

Fig. 5 Protein expression of phospho- myosin phosphatase target subunit 1 (MYPT1) (A) is increased in the erectile tissue of men with vasculogenic
erectile dysfunction (ED); protein expressions of RhoA/Rho-associated protein kinase (ROCK)a (B) and ROCKb (C) are not different between erectile
tissues of men with vasculogenic ED and post-radical prostatectomy (RP) ED compared with that of the control group. Upper panels are representative
Western immunoblots and lower panels are densitometric analyses of phospho-MYPT1/glyceraldehyde 3-phosphate dehydrogenase (GAPDH),
ROCKa/GAPDH, and ROCKb/GAPDH in the penis of men in the control group, men with vasculogenic ED, and men with post-RP ED. Each bar represents
the mean  SEM of five to nine patients. *P < 0.05 vs control group.

A B C
kDa kDa kDa
Phospho-MYPT1 140 ROCKα 160 ROCKβ 160

GAPDH 37 GAPDH 37 GAPDH 37

ROCKβ/GAPDH (% of Control)
150
Phospho-MYPT1/GAPDH

100
200
ROCKα/GAPDH
(% of Control)

(% of Control) 100

50
100
50

0 0 ED sc. 0

ED sc.
.

.
ED sc.

nt

nt
.

ED -RP
nt

ED -RP
ED -RP

Va

Va
Co

Co
Va
Co

st
st
st

Po
Po
Po

Fig. 6 Protein expression of phosphodiesterase type 5 (PDE5) (A) is
increased in the erectile tissue of men with vasculogenic erectile
dysfunction (ED) and decreased in the erectile tissue of men with post-
radical prostatectomy (RP) ED; protein expression of a-smooth muscle
actin (B) is not different in the erectile tissue of men with vasculogenic ED
and post-RP ED compared with that of the control group. Upper panels
are representative Western immunoblots and lower panels are
densitometric analyses of PDE5/glyceraldehyde 3-phosphate
dehydrogenase (GAPDH) and a-smooth muscle actin/GAPDH in the penis
of men in the control group, men with vasculogenic ED, and men with
post-RP ED. Each bar represents the mean 
< 0.05 vs control group. SMA, smooth muscle actin.
A kDa B
PDE5 100 α-SMA
GAPDH 37 GAPDH
150 * to oxidation of a critical NOS cofactor tetrahydrobiopterin
α-SMA/GAPDH
PDE5/GAPDH
nNOS phosphorylation on Ser-1412 was increased and nNOS
was uncoupled. The latter data mimic our recent findings in
the rat penis 3 days after cavernous nerve damage, which,
similarly to men with post-RP ED, showed increased nNOS
phosphorylation on Ser-1412 and nNOS uncoupling [9].
While the mechanism of increased nNOS phosphorylation on
its activation site in the neuropathic erectile tissue is
unknown, it conceivably indicates a deleterious effect of over-
activated nNOS, shown previously in the central and
SEM of five to nine patients. *P peripheral nervous systems [22,23]; overactivated, but
uncoupled, nNOS further contributes to decreased NO
bioavailability. In line with our findings, nNOS uncoupling
kDa
has recently been demonstrated in the major pelvic ganglia,
42
the site of cavernous nerve origin, of rats following cavernous
37 nerve damage [24]. The precise mechanism by which nNOS
and eNOS become uncoupled is unclear but may be related
100 (BH4), depletion of the NOS substrate L-arginine, or S-

glutathionylation of NOSs [25,26]. Downregulation of eNOS

(% of Control)

(% of Control)

100 * and nNOS protein expressions in vasculogenic and post-RP

50 ED, respectively, conceivably contributes to a long-term
50 decrease in NO bioavailability.
Oxidative stress has an important role in the pathogenesis of
0 0
ED, shown previously in both animal and human ED studies
ED sc.

ED sc.
.

.
ED -RP

ED -RP
nt

nt
Va

Va
Co

Co

[10,11]. In the present study, we confirmed these previous

st

st
Po

Po

findings and showed that increased oxidative stress (4-HNE

© 2018 The Authors

BJU International © 2018 BJU International 1071
Karakus et al.

expression) is common to both vasculogenic and neurogenic
ED. 4-HNE, a marker of oxidative stress, is an end-product
of lipid peroxidation which binds to lysine, cysteine and
histidine, causing the modification and malfunction of
multiple proteins [11]. It is conceivable that dysfunctional,
uncoupled eNOS and nNOS in the erectile tissue of men with
vasculogenic and post-RP ED, respectively, are sources of
oxidative stress. Further studies, measuring other markers of
oxidative stress, such as nitrotyrosine, and superoxide itself,
would conceivably further substantiate increased oxidative
stress in the erectile tissue of the two patient groups.
In vascular smooth muscle cells, including those within the
penis, an inverse functional relationship exists between the
NO/cGMP/PKG and RhoA/ROCK contractile signalling
pathways. In the rat penis, RhoA/ROCK suppresses eNOS
gene expression and activity and erectile function [27]. The
RhoA/ROCK signalling pathway inhibits MYPT1 activity by
phosphorylating the regulatory subunit of myosin light chain
phosphatase at Thr-696, promoting smooth muscle
contraction at low calcium levels [13]. In the present study,
we showed upregulation of the RhoA/ROCK pathway in
erectile tissue of men with vasculogenic ED, conceivably
further contributing to decreased vasorelaxation and ED.
While the mechanism for RhoA/ROCK upregulation is not
known, decreased NO signaling may increase ROCK activity,
as shown in other vascular tissues [28,29]. Several animal
studies showed that inhibition of the RhoA/ROCK pathway
potentiated erectile function in vasculogenic ED associated
with diabetes, hypertension and aging [27]. Increased RhoA/
ROCK activity in the erectile tissue of men with vasculogenic

Fig. 7 Proposed integrative model of derangements in erection regulatory mechanisms for vasculogenic and post-RP ED. (A) Vascular disease is
associated with constitutive nitric oxide synthase (NOS) inactivation through decreased phosphorylation of neuronal nitric oxide synthase (nNOS; Ser-
1412), decreased phosphorylation of endothelial nitric oxide synthase (eNOS; Ser-1177), and increased eNOS uncoupling. Uncoupled eNOS increases
oxidative stress (ROS) in the penis. Increased phosphodiesterase type 5 (PDE5) protein expression and upregulated RhoA/Rho-associated protein
kinase (ROCK) activity together with impaired nitric oxide (NO) signaling contribute to vasculogenic erectile dysfunction (ED). (B) Radical
prostatectomy (RP) is associated with nNOS inactivation through excessive phosphorylation of nNOS on Ser-1412 and increased nNOS uncoupling.
Uncoupled nNOS increases oxidative stress in the penis. Decreased PDE5 protein expression together with impaired NO signaling contribute to post-RP
ED. ROS, reactive oxygen species.

A B
Vascular Disease Post Radical Prostatectomy

Sexual Stimulus Sexual Stimulus

eNOS monomer Phospho-nNOS nNOS monomer

Phospho-nNOS
eNOS monomer (Ser-1412) nNOS dimer eNOS dimer
(Ser-1412) nNOS monomer

ROS ROS
NO/cGMP NO/cGMP

Initiation of relaxation Initiation of relaxation

Blood flow RhoA/Rho-kinase
Blood flow

PI3-kinase PI3-kinase

Phospho-Akt Continued Continued PDE5

Phospho-Akt
(Ser-473) relaxation relaxation
(Ser-473)
PDE5
NO/cGMP
NO/cGMP
Phospho-eNOS Phospho-eNOS
(Ser-1177) (Ser-1177)

Vasculogenic ED Post-RP ED

© 2018 The Authors

1072 BJU International © 2018 BJU International
 Mechanisms of vasculogenic vs post-radical prostatectomy erectile dysfunction
ED may suggest that treating this patient population with
RhoA/ROCK inhibitors may improve the effect of PDE5
inhibitors in the long term. In contrast to vasculogenic ED,
post-RP ED is not associated with aberrant RhoA/ROCK
molecular signalling. Our findings are in line with a study
showing that ROCK activity was increased in the rat major
pelvic ganglia shortly after cavernous nerve injury, but
unchanged at later times [24].
Human studies indicate that ED originating from post-RP ED
is less successfully treated with PDE5 inhibitors compared
with ED of vasculogenic origin [8]. We hypothesized that the
lack of erection ability enhancement by PDE5 inhibitors in
post-RP ED patient groups may relate to decreased PDE5
protein expression in the erectile tissue. Indeed, we found
decreased protein expression of PDE5 in the erectile tissue of
men with post-RP ED. Our results are consistent with
previous studies in rats, which demonstrated reduced PDE5
mRNA and protein levels in the penis after cavernous nerve
injury [30]. Because the presence of functional, inhibitable
PDE5 is required to enhance erectile function in response to
PDE5 inhibitors, our findings may partially explain why men
with post-RP ED are refractory to PDE5 inhibitor treatment.
In line with our findings, a recent study showed that
neurogenic relaxation in human corpus cavernosum strips
from men with vasculogenic ED, but not post-RP ED, was
potentiated by the PDE5 inhibitor tadalafil [31]. By the same
logic, increased PDE5 protein expression in the erectile tissue
of men with vasculogenic ED may allow a greater level of
inhibition by PDE5 inhibitors, resulting in a better erectile
response. PDE5 protein expression does not appear to be
influenced by smooth muscle content in the erectile tissue, as
a-smooth muscle actin, expressed in vascular smooth muscle
cells [32], was not affected in any of the patient groups. While
the mechanism of PDE5 upregulation in the erectile tissue of
men with vasculogenic ED is not known, it is conceivable that
upregulated RhoA/ROCK signalling augments PDE5 activity,
as shown previously in gastric smooth muscle cells [33]. By
contrast, decreased protein expression of PDE5 in the erectile
tissue of men with post-RP ED is conceivably a consequence
of severely impaired NO signalling; downregulated NOS
function and resulting decreased accumulation of intracellular
cGMP may decrease PDE5 protein expression through cGMP-
responsive sequences in the PDE5 promoter [32]. These topics
require further investigation.
The present study has several limitations. First, the study
population comprised a small number of patients, partially
because of our careful selection of patient groups. Second, we
acknowledge that a post-RP patient group with minimal or no ED
would have represented an interesting comparative group;
however, it is not feasible in practice to obtain erectile tissue from
patients without ED. Third, SHIM scores differ between patients
with vasculogenic and those with post-RP ED (consistent with
previous studies [33,34]), which reflects the extent of disease.
Finally, the limited amount of tissue precluded
immunohistochemical studies, which would have provided
illustrative images of evaluated molecular markers. An integrative
model of molecular derangements in the erectile tissue of men
with vasculogenic ED and post-RP ED is presented in Fig. 7.
In conclusion, the present study provides evidence that the
mechanisms of vasculogenic ED and post-RP ED in the
human penis involve derangements in constitutive NOS
function, PDE5 protein expression, and RhoA/ROCK activity,
along with increased oxidative stress, which conceivably
provide a molecular basis for chronically reduced NO
bioavailability and smooth muscle contraction in these
conditions of erectile impairment. Selective differences in
PDE5 protein expression suggest there are distinct molecular
mechanisms for these ED subtypes.
Acknowledgements
This work was supported by grant from the National
Institutes of Health, USA (NIH/NIDDK grant R01DK067223
to Arthur L. Burnett).
Conflict of Interest
None declared.
References
1 Aytac IA, McKinlay JB, Krane RJ. The likely worldwide increase in
erectile dysfunction between 1995 and 2025 and some possible policy
consequences. BJU Int 1999; 84: 50–6
2 Dean RC, Lue TF. Physiology of penile erection and pathophysiology of
erectile dysfunction. Urol Clin North Am 2005; 32: 379–95
3 Angeloni NL, Bond CW, McVary KT, Podlasek CA. Sonic hedgehog
protein is decreased and penile morphology is altered in prostatectomy
and diabetic patients. PLoS ONE 2013; 14: 8
4 Mart
ınez-Salamanca JI, La Fuente JM, Fern
andez A et al. Nitrergic
function is lost but endothelial function is preserved in the corpus
cavernosum and penile resistance arteries of men after radical
prostatectomy. J Sex Med 2015; 12: 590–9
5 Barassi A, Colpi GM, Piediferro G, Dogliotti G, D’Eril GV, Corsi MM.
Oxidative stress and antioxidant status in patients with erectile
dysfunction. J Sex Med 2009; 6: 2820–5
6 Shridharani AN, Brant WO. The treatment of erectile dysfunction in
patients with neurogenic disease. Transl Androl Urol 2016; 5: 88
7 Walz J, Burnett AL, Costello AJ et al. A critical analysis of the current
knowledge of surgical anatomy related to optimization of cancer control
and preservation of continence and erection in candidates for radical
prostatectomy. Eur Urol 2010; 57: 179
8 Sopko NA, Burnett AL. Erection rehabilitation following prostatectomy–
current strategies and future directions. Nat Rev Urol 2016; 13: 216–25
9 Karakus S, Musicki B, La Favor DJ, Burnett AL. cAMP-dependent post-
translational modification of neuronal nitric oxide synthase neuroprotects
penile erection in rats. BJU Int 2017; 120: 861–72
10 Musicki B, Liu T, Lagoda GA et al. Hypercholesterolemia-induced
erectile dysfunction: endothelial nitric oxide synthase (eNOS) uncoupling
in the mouse penis by NAD(P)H oxidase. J Sex Med 2010; 7: 3023
11 Musicki B, Burnett AL. Constitutive NOS uncoupling and NADPH
oxidase upregulation in the penis of type 2 diabetic men with erectile
dysfunction. Andrology 2017; 5: 294
© 2018 The Authors
BJU International © 2018 BJU International 1073
Karakus et al.
12 Altman DG, Bland JM. Standard deviations and standard errors. BMJ
2005; 331: 903
13 Hurt KJ, Musicki B, Palese MA et al. Akt-dependent phosphorylation of
endothelial nitric-oxide synthase mediates penile erection. Proc Natl Acad
Sci USA 2002; 99: 4061
14 Feng J, Ito M, Ichikawa K et al. Inhibitory phosphorylation site for Rho-
associated kinase on smooth muscle myosin phosphatase. J Biol Chem
1999; 274: 37385
15 Liu L, Li E, Li F et al. Effect of testosterone on the phenotypic
modulation of corpus cavernosum smooth muscle cells in a castrated rat
model. Urology 2017; 103: 273
16 McCabe TJ, Fulton D, Roman LJ, Sessa WC. Enhanced electron flux and
reduced calmodulin dis-sociation may explain “calcium-independent”eNOS
activation by phosphorylation. J Biol Chem 2000; 275: 6123
17 Adak S, Santolini J, Tikunova S, Wang Q, Johnson JD, Stuehr DJ.
Neuronal nitric-oxide synthase mutant (Ser-1412 –> Asp) demonstrates
surprising connections between heme reduction, NO complex formation,
and catalysis. J Biol Chem 2001; 276: 1244
18 Hurt KJ, Sezen SF, Lagoda GF et al. Cyclic AMP-dependent
phosphorylation of neuronal nitric oxide synthase mediates penile
erection. Proc Natl Acad Sci USA 2012; 109: 16624–9
19 Chen PF, Tsai AL, Wu KK. Cysteine 99 of endothelial nitric oxide
synthase (NOS-III) is critical for tetrahydrobiopterin-dependent NOS-III
stability and activity. Biochem Biophys Res Commun 1995; 215: 1119–29
20 Forstermann U, Sessa WC. Nitric oxide synthases: regulation and
function. Eur Heart J 2012; 33: 829–37
21 Melikian N, Seddon MD, Casadei B, Chowienczyk PJ, Shah AM.
Neuronal nitric oxide synthase and human vascular regulation. Trends
Cardiovasc Med 2009; 19: 256–62
22 Rameau GA, Tukey DS, Garcin-Hosfield ED et al. Biphasic coupling of
neuronal nitric oxide synthase phosphorylation to the NMDA receptor
regulates AMPA receptor trafficking and neuronal cell death. J Neurosci
2007; 27: 3445–55
23 S
anchez-Bl
azquez P, Rodr
ıguez-Mu~ noz M, Garz
on J. Mu-opioid
receptors transiently activate the Akt-nNOS pathway to produce sustained
potentiation of PKC-mediated NMDAR-CaMKII signaling. PLoS ONE
2010; 5: e11278
24 Hannan JL, Matsui H, Sopko NA et al. Caspase-3 dependent nitrergic
neuronal apoptosis following cavernous nerve injury is mediated via
RhoA and ROCK activation in major pelvic ganglion. Sci Rep 2016; 6:
29416
25 Hoang HH, Padgham SV, Meininger CJ. L-arginine, tetrahydrobiopterin,
nitric oxide and diabetes. Curr Opin Clin Nutr Metab Care 2013;
16: 76–82
26 De Pascali F, Hemann C, Samons K, Chen CA, Zweier JL. Hypoxia and
reoxygenation induce endothelial nitric oxide synthase uncoupling in
© 2018 The Authors
1074 BJU International © 2018 BJU International
endothelial cells through tetrahydrobiopterin depletion and
S-glutathionylation. Biochemistry 2014; 53: 3679–88
27 Sopko NA, Hannan JL, Bivalacqua TJ. Understanding and targeting
the Rho kinase pathway in erectile dysfunction. Nat Rev Urol 2014; 11:
622–8
28 Chitaley K, Webb RC. Nitric oxide induces dilation of rat aorta via
inhibition of rho-kinase signaling. Hypertension 2002; 39: 438–42
29 Sawada N, Itoh H, Miyashita K et al. Cyclic GMP kinase and RhoA
Ser188 phosphorylation integrate pro- and antifibrotic signals in blood
vessels. Mol Cell Biol 2009; 29: 6018–32
30 Vignozzi L, Filippi S, Morelli A et al. Effect of chronic tadalafil
administration on penile hypoxia induced by cavernous neurotomy in the
rat. J Sex Med 2006; 3: 419
31 Mart
ınez-Salamanca JI, La Fuente JM, Mart
ınez-Salamanca E et al. a1A-
Adrenergic Receptor Antagonism Improves Erectile and Cavernosal
Responses in Rats With Cavernous Nerve Injury and Enhances
Neurogenic Responses in Human Corpus Cavernosum From Patients
With Erectile Dysfunction Secondary to Radical Prostatectomy. J Sex Med
2016; 13: 1844–57
32 Lin CS. Phosphodiesterase type 5 regulation in the penile corpora
cavernosa. J Sex Med 2009; 6: 203
33 Murthy KS. Contractile agonists attenuate cGMP levels by stimulating
phosphorylation of cGMP-specific PDE5; an effect mediated by RhoA/
PKC-dependent inhibition of protein phosphatase 1. Br J Pharmacol 2008;
153: 1214–24
34 Raina R, Lakin MM, Thukral M et al. Long-term efficacy and compliance
of intracorporeal (IC) injection for erectile dysfunction following radical
prostatectomy: SHIM (IIEF-5) analysis. Int J Impot Res 2003; 15: 318–22
Correspondence: Arthur L. Burnett, The James Buchanan
Brady Urological Institute, Johns Hopkins School of
Medicine, 600 North Wolfe Street, Marburg 407, Baltimore,
MD 21287-2101, USA.
e-mail: aburnet1@jhmi.edu
Abbreviations: 4-HNE4-hydroxy-2-nonenal; ED, erectile
dysfunction; eNOS, endothelial isoform of nitric oxide synthase;
GAPDH, glyceraldehyde 3-phosphate dehydrogenase; MYPT1,
myosin phosphatase target subunit 1; nNOS, neuronal isoform
of nitric oxide synthase; NO, nitric oxide; NOS, nitric oxide
synthase; PDE5, phosphodiesterase type 5; ROCK, RhoA/Rho-
associated protein kinase; RP, radical prostatectomy; SHIM,
Sexual Health Inventory for Men.