Phosphodiesterase 5 inhibition as a therapeutic target for ischemic stroke: A systematic review of preclinical studies

Accepted Manuscript Phosphodiesterase 5 inhibition as a therapeutic target for ischemic stroke: A systematic review of preclinical studies

Joakim N.E. Ölmestig, Ida R. Marlet, Atticus H. Hainsworth, Christina Kruuse PII: DOI: Reference:

S0898-6568(17)30172-9 doi: 10.1016/j.cellsig.2017.06.015 CLS 8944

To appear in:

Cellular Signalling

Received date: Revised date: Accepted date:

3 March 2017 10 June 2017 20 June 2017

Please cite this article as: Joakim N.E. Ölmestig, Ida R. Marlet, Atticus H. Hainsworth, Christina Kruuse , Phosphodiesterase 5 inhibition as a therapeutic target for ischemic stroke: A systematic review of preclinical studies, Cellular Signalling (2017), doi: 10.1016/j.cellsig.2017.06.015

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

Phosphodiesterase 5 inhibition as a therapeutic target for ischemic stroke: a

PT

systematic review of preclinical studies.

Neurovascular Research Unit, Department of Neurology, Herlev Gentofte Hospital, University of

SC

a

RI

Joakim N.E. Ölmestiga, Ida R. Marleta, Atticus H. Hainsworthb, Christina Kruusea

[email protected], [email protected],

b

MA

[email protected]

NU

Copenhagen, Herlev Ringvej 75, 2730 Herlev, Denmark.

Clinical Neuroscience, Molecular & Clinical Sciences Research Institute, St George’s University

PT E

[email protected]

D

of London, Cranmer Terrace, London SW17 0RE, UK

Correspondence to Christina Kruuse, MD, DMSc, PhD, Neurovascular Research Unit, Department

Herlev, Denmark.

CE

of Neurology, Herlev Gentofte Hospital, University of Copenhagen, Herlev Ringvej 75, 2730

AC

Email: [email protected]

Telephone no.: + 45 38 68 12 33

ACCEPTED MANUSCRIPT

Abstract Phosphodiesterase 5 inhibitors (PDE5i), such as sildenafil (Viagra®) are widely used for erectile dysfunction and pulmonary hypertension. Preclinical studies suggest that PDE5i may improve functional outcome following ischemic stroke. In this systematic review we aimed to evaluate the

PT

effects of selective PDE5i in animal models of brain ischaemia. A systematic search in Medline, Embase, and The Cochrane Library was performed including studies in English assessing the

RI

effects of selective PDE5i. 32 publications were included describing outcome in 3646 animals.

SC

Neuroprotective effects of PDE5i were dependent on the NO-cGMP-PKG-pathway. These included

NU

reduced neuronal apoptosis (n=3 studies), oxidative stress (n=5), and neuroinflammation (n=2). PDE5i increased angiogenesis and elevated regional cerebral blood flow in the ischemic penumbra,

MA

and improved functional recovery. Some studies found that PDE5i treatment reduced lesion volume (n=9), others found no effect (n=9). Treatment was effective when administered within 24 hours

D

post-ischemia, though treatment delayed to seven days improved outcome in one study. This review

PT E

demonstrates both neuroprotective and neurorestorative effects of PDE5i in animal models of stroke, though the specific underlying signalling pathways relating to PDE5 inhibition and cGMP

CE

may remain serendipitous in some studies. There is currently limited evidence on the effects of selective PDE5i in human stroke patients, hence translation of preclinical results into clinical trials

Key words

AC

may be warranted.

Phosphodiesterase 5 inhibitor; PDE5 inhibitor; cerebrovascular disorder; stroke translational models

ACCEPTED MANUSCRIPT

Abbreviations Akt, protein kinase B; Bax, bcl-2-like protein 4; BCCAO, bilateral common carotid artery occlusion; Bcl-xL, B-cell lymphoma-extra large; bcl-2, B-cell lymphoma 2; MAPK, mitogen-

PT

activated protein kinase; MCAO, middle cerebral artery occlusion; PDE5i, phosphodiesterase 5

RI

inhibitors; TUNEL, terminal deoxynucleotidyl transferase dUTP nick end labeling; 4-VO, 4-vessel

AC

CE

PT E

D

MA

NU

SC

occlusion

ACCEPTED MANUSCRIPT

1. Introduction Acute ischemic stroke is a major cause of mortality and morbidity worldwide [1]. While many therapeutic candidates have been developed for treatment of stroke, with efficacy in preclinical data, few have shown efficacy in humans [2]. The only acute treatment available today,

PT

thrombolysis, aims to restore blood flow to the ischemic area. Thrombolysis has a relatively short therapeutic time-window (up to 4.5 h) and carries risk of severe adverse effects due to haemorrhage

RI

[3]. New treatments with different mechanisms of action and wider therapeutic window are

SC

required.

NU

Phosphodiesterase 5 (PDE5) is a cyclic nucleotide phosphodiesterase which is found in multiple tissues, including neurons [4], smooth muscle cells [5], and the vascular wall in the brain [6]. It

MA

inactivates the second messenger cGMP by hydrolysis to GMP [7-9]. Via protein kinase G (PKG), cGMP activates downstream intracellular mechanisms, e.g. transcription factor CREB, involved in

D

cell survival and synaptic plasticity [10, 11]. High potency selective PDE5i, such as sildenafil

PT E

(licensed as Viagra®) and tadalafil (Cialis®) inhibit PDE5 and therefore prolong the half-life of endogenous cGMP [9]. PDE5i are widely prescribed “as required” treatment for sexual or erectile

CE

dysfunction and as daily treatment for pulmonary hypertension, where it has the ability to relax vascular smooth muscle cells, leading to dilation of the vessels thus increasing downstream blood

AC

flow [12]. In preclinical studies, PDE5i have been reported to be neuroprotective, antiinflammatory, and cognition-enhancing in animal models of Alzheimer’s disease [13, 14], multiple sclerosis [15], and hepatic encephalopathy [16]. PDE5i also ameliorate ischemic damage in animal models of myocardial infarction [17, 18], renal ischemia [19-21], hepatic ischemia [22], testicular torsion [23, 24], and ovarian ischemia [25] through anti-apoptotic and anti-oxidative stress mechanisms.

ACCEPTED MANUSCRIPT

The aim of this study was to review the molecular effects of PDE5i on neurons and cerebrovascular reactivity in animal models of ischemic stroke. In addition, we aimed to assess functional outcome, cerebral blood flow, and stroke lesion volume.

2. Methods

PT

2. 1 Search strategy

RI

We searched Medline, Embase, and the Cochrane Library on August 16, 2016, with a search

SC

strategy to identify preclinical interventional in vivo animal studies investigating the effect of selective PDE5i on experimental stroke. The review protocol followed the Cochrane Handbook for

NU

Systematic Reviews and Interventions (version 5.1.0, March 2011). See supplementary material 1 for search string used. Two investigators (JNEÖ and IRM) screened all titles and abstracts for

MA

eligibility. Full texts of potentially eligible studies were obtained and the same two investigators applied inclusion and exclusion criteria to each article. Disagreement was resolved through

D

discussion or by a third investigator (CK). Reference lists of all included articles were screened for

PT E

further eligible studies by one investigator (JNEÖ). No effort was made to pool quantitative data

CE

from different studies, due to the diversities in methodology.

2.2 Inclusion and exclusion criteria

AC

Inclusion criteria for the screened publications were: in vivo animal studies; induced ischemic stroke, brain ischemia, or brain hypoxia; animals treated with a pharmacologically-specific PDE5i; studies that used a control group; the study is an original publication. Applied exclusion criteria were: human studies; in vitro or in situ studies; no induced ischemic stroke; cerebral haemorrhagic animal models; treatment with a non-specific PDE5i (e.g. dipyridamole and icariin); studies presented in form of a review, conference abstracts, letters, and abstracts; full-text not in English.

ACCEPTED MANUSCRIPT

2.3 Data extraction We recorded author, publication year, type of animal (species, strain, sex, and weight), number of animals per treatment group, ischemia model, PDE5i used (dose, time of administration, and administration route) and outcome. Outcomes were noted under the following categories:

PT

neurogenesis/neuronal protection, synaptogenesis/synapse protection, functional outcome (e.g. memory performance, beam walking score, and rotating rod test) stroke lesion

RI

volume/neurodegeneration, angiogenesis/cerebral vessel protection, cerebral blood flow, apoptosis,

SC

oxidative stress, inflammation within brain tissue, and mortality. Significance were considered

NU

when p < 0.05.

2.4 Quality assessment

MA

Risk of bias was assessed using checklists derived from the Animal Research: Reporting of In Vivo Experiments (ARRIVE) guidelines [26] and Collaborative Approach to Meta-Analysis and Review

3. Results

CE

3.1 Study selection

PT E

D

of Animal Data from Experimental Studies (CAMARADES) [27].

482 non-duplicate publications were identified. 405 publications were excluded after title and

AC

abstract screening and 45 publications were further excluded after full-text screening. No publications were added from reference lists from already included articles. The remaining 32 studies were included in this systematic review, assessing outcome in a total of 3646 animals (2396 rats, 1146 mice, and 104 gerbils).

ACCEPTED MANUSCRIPT

3.2 Neurogenesis and neuronal protection Several studies show a PDE5i-dependent increase in surviving neurons and preserved neuronal structure after experimental stroke [28-37]. Cell proliferative markers, such as Bromodeoxy-uridine (BrdU) labelling, demonstrate increased neurogenesis in PDE5i treated rats, following permanent

PT

middle cerebral artery occlusion (MCAO), in the dentate gyrus [38], subventricular zone [38, 39], and ipsilateral striatum [40]. By use of diffusion tension imaging (DTI) MRI, water movement can

RI

be visualized in the brain, where movements are faster along white matter tracts and not

SC

perpendicular to it. This method makes it possible to image white matter tracts and by that the

NU

axonal remodelling after ischemia [41]. Sildenafil (10 mg/kg) increased axonal remodelling in the ischemic penumbra in rats exposed to permanent MCAO, which was histologically confirmed [42,

MA

43].

Rats treated with yonkenafil and sildenafil starting two hours after temporary MCAO (with two

D

hours occlusion) exhibited increased number of surviving neurons [28, 30]. The cellular Akt

PT E

pathway, related to neurite outgrowth and cell survival, were augmented in these PDE5i-treated animals through cGMP dependent pathways (Figure 2) [28, 30]. Co-administration of inhibitors of

CE

soluble guanylyl cyclase (sGC), PKG, or protein kinase A (PKA) together with PDE5i reduced

AC

activation of the Akt pathway and the positive effects on cell survival [28, 30].

3.3 Synaptogenesis and synapse protection PDE5i showed protective effects on synapse damage, including post-stroke axonal swelling and demyelination [28, 30]. After ischemic stroke, an increase in coupling of postsynaptic density protein 95 (PSD-95) to nNOS is seen. PSD-95 is a post synaptic membrane bound protein which functions as a scaffold for clustering of receptors, ion channels, and signalling proteins and is thus important for synaptic plasticity [44]. Reduced coupling of PSD-95 to nNOS post-ischemia can

ACCEPTED MANUSCRIPT

therefore be interpreted as a reduced synapse damage. Yonkenafil and sildenafil (8, 10, 16, and 32 mg/kg) administered to rats 2-24 hours post temporary and permanent MCAO, reduced loss of synaptophysin (a marker for number of synapses) and decreased coupling of postsynaptic density protein 95 (PSD-95) to neuronal NOS (nNOS), effects seen together with preserved synapse

PT

structure [28, 30, 45]. PDE5i significantly increased brain concentration of the growth factors tropomyosin-related kinase B (TrkB), brain-derived neurotrophic factor (BDNF), tropomyosin-

RI

related kinase A (TrkA), and nerve growth factor (NGF) involved in synapse functioning in rats

NU

3.4 Angiogenesis and cerebral blood flow

SC

subjected to temporary MCAO [28, 30].

The density of cerebral blood vessels following stroke was histologically augmented in rats exposed

MA

to permanent MCAO and treated with PDE5i [39, 45, 46]. Sildenafil (5 and 10 mg/kg) reduced the loss of capillary density and the abundance of TUNEL-labeled endothelium, a method used to

D

detect DNA damage, and reduced extravasation of endogenous IgG from cerebral vessels following

PT E

stroke, expressing preserved blood brain barrier in rats exposed to unilateral common carotid artery occlusion and hypoxia [35]. Further, it increased BrdU-labeled endothelial cells in the ischemic

CE

penumbra in rats with induced permanent MCAO, corresponding to increased formation of new endothelial cells and angiogenesis [39, 47]. One study found no difference in length of

AC

immunohistochemically labelled capillaries between PDE5i- and control treated mice exposed to temporary MCAO [48]. By use of T2* MRI, extravascular blood and blood vessels can be visualized in the brain and used as an estimate for angiogenesis. In rats with induced permanent MCAO, sildenafil (10 mg/kg) was shown to increase angiogenesis measured with T2* MRI which was immunohistochemically confirmed in the ischemic penumbra by endothelial barrier antigen (EBA) labeling [42, 43, 46]. An angiogenetic effect was also seen in the ischemic penumbra in

ACCEPTED MANUSCRIPT

control treated rats with induced permanent MCAO, however PDE5i could accelerate and enhance this effect [49]. By use of Doppler ultrasound, the mean blood flow velocity (mBFV) was measured in the internal carotid arteries and basilar trunk as an estimate of global cerebral blood flow (CBF) [35, 50]. One

PT

study found no change in mBFV following sildenafil treatment in mice with permanent MCAO

RI

[50], while another study observed a temporary increase in mBFV over one hour in the basilary trunk and contralateral internal carotid artery after sildenafil treatment in rats with induced

SC

unilateral common carotid artery occlusion and hypoxia [35]. Laser-Doppler flow measurement

NU

showed a short temporary increase in relative CBF following zaprinast treatment in permanent

MA

MCAO rats [51].

MRI can be used to asses CBF in vivo in animals and humans. A well-known, implemented method is arterial spin labelling (ASL) MRI which is commonly used in human trials to assess CBF. This

PT E

D

method involves labelling of endogenous water protons spins in the arterial blood before entering the brain, and then observe the labelled water protons as they flow into the brain, making it possible to calculate CBF [52]. Sildenafil treatment (10 mg/kg) significantly increased CBF by

CE

approximately 8 % measured with ASL MRI in aged rats (18 months old) six weeks post permanent

AC

MCAO [42]. One study using flow sensitive alternating inversion recovery MRI, an ASL-MRI method, showed no difference in CBF in vardenafil treated mice (10 mg/kg twice daily) compared to control treated mice exposed to temporary MCAO [53]. Two other studies using ASL MRI showed that sildenafil treatment in rats (10 mg/kg) increased CBF by approximately 17%, respectively 36% in the ischemic penumbra six weeks after permanent MCAO [43, 46].

ACCEPTED MANUSCRIPT

3.5 Apoptosis, oxidative stress, and inflammation Ligation of both common carotid arteries and both vertebral arteries (4-vessel occlusion (4-VO)) and bilateral common carotid artery occlusion (BCCAO), two different methods used to induce severe hypoperfusion to the brain [54], cause significant cell death in the hippocampal areas [31, 34,

PT

36, 37] and cortex [34, 37]. Tadalafil administered after the first occlusion stage in the 4-VO rat model and tadalafil administered one day after BCCAO in gerbils reduced neuronal apoptosis in

SC

RI

these brain areas [31, 34, 36, 37].

Treatment with tadalafil and yonkenafil reduced the expression of pro-apoptotic factors known to

NU

initiate apoptotic cell death. Similarly, PDE5i increased concentration of anti-apoptotic factors poststroke (such as the heat shock protein hsp70, see figure 2) further to the cGMP-related mechanisms

MA

is not given. These findings were seen together with a significant reduction in neuronal cell death following temporary MCAO in rats [28, 31]. Neuronal apoptosis was, in one study using

D

intrastriatal malonate injections to induce focal chemical hypoxia in rats, assessed by measuring the

PT E

concentration of the important anti-apoptotic factors bcl-2 and Bcl-xL and the pro-apoptotic factor Bax. The study also investigated activation of calpain and cyclin-dependent kinase 5 (cdk5) in the

CE

brain post-stroke. Pretreatment with sildenafil (1.5 mg/kg) significantly increased Bcl-2/Bax ratio and Bcl-xL/Bax ratio. Moreover, it reduced malonate-induced pathological activation of calpain and

AC

cdk5 which are factors known to induce tau-hyperphosphorylation and inactivation of the cell survival promoting transcription factor myocyte enhancer factor 2 (MEF2) and therefore cause cell death. Pretreatment with sildenafil prevented malonate-induced tau hyperphosphorylation and inactivation of MEF2 [32]. These findings were accompanied by reduced neuronal apoptosis (see figure 2), but which of the downstream cGMP related mechanisms that are involved is not known [32]. In another study intrastriatal malonate injection in rats reduced inhibitory phosphorylation of apoptosis signal-regulating kinase 1 (ASK1) which led to increased phosphorylation of MAPK

ACCEPTED MANUSCRIPT

Kinase 3/6 and MAPK Kinase 7, an effect significantly decreased by approximately 50% by sildenafil (1.5 mg/kg) pretreatment [33]. Sildenafil pretreatment further led to reduced downstream activation and translocation of p38-MAPK to the neuronal nucleus and reduced c-jun N-terminal kinase (JNK) phosphorylation and activation of its downstream effector, c-Jun [33]. These effects

PT

were seen together with reduced neuronal apoptosis [33]. p38-MAPK and c-Jun activation has

RI

previously shown to be involved in neuronal apoptosis (see figure 2) [55, 56].

Pretreatment with tadalafil (2, 5, 10, and 20 mg/kg), in rats and mice exposed to BCCAO,

SC

attenuated ischemia-induced oxidative stress markers including acetylcholine esterase (AChE)

NU

activity, malondialdehyde (MDA), and thiobarbituric acid reactive substances (TBARS). Similarly, tadalafil increased the anti-oxidative stress markers glutathione (GSH) and superoxide dismutase

MA

(SOD) in the ischemic brain [29, 57, 58]. L-NAME, an inhibitor of NOS, abolished the PDE5iinduced positive effects on oxidative stress markers [57, 58]. Inconsistent results were found

D

regarding the effect of PDE5i on ischemia-induced elevation of nitrite/nitrate levels [29, 57, 58].

PT E

Two studies reported that PDE5i had no effect on oxidative stress in mice exposed to BCCAO and that PDE5i abolished the anti-oxidative effects achieved by antidepressants alone post-stroke with

CE

no further suggestions to the underlying signaling mechanisms involved [59, 60].

AC

Following permanent MCAO in neonatal mice or unilateral common carotid artery occlusion and hypoxia in neonatal rats, an increase in microglial activation and density were seen [35, 50]. Sildenafil (10 mg/kg) lowered microglial activation and reduced the number of microglia in the ischemic penumbra [50] and other areas of the brain [35]. Moreover, PDE5i had in general the ability to upregulate the neuroprotective M2-microglia and downregulate the neurotoxic M1microglia in the ischemic penumbra with no specification of the downstream cGMP effects [50].

ACCEPTED MANUSCRIPT

3.6 Functional outcome, memory, and mortality PDE5i significantly improved functional outcome post-stroke measured with e.g. beam walking test, rotating rod test, and lateral push test. PDE5i also improved memory performance measured with Morris water maze and aversive radial maze post-stroke [28, 31, 35, 38-40, 42, 45, 46, 57, 58,

PT

61, 62]. The positive effects of PDE5i on functional outcome were abolished when inhibitors of NOS [57, 58], sGC, PKG, or PKA, [28, 30], were co-administered with the PDE5i. Five studies

RI

found no improvement of functional outcome and/or memory performance after PDE5 inhibition

SC

[34, 37, 48, 53, 59]. PDE5 inhibition reduced mortality in two studies [36, 63] but had no effect in

NU

five studies [34, 37, 45, 48, 53]. See table 1 for details regarding PDE5i, animal model, and test.

3.7 Stroke lesion volume and neurodegeneration

MA

Nine studies showed a significant reduction in stroke lesion volume after PDE5 inhibition [28, 30, 32, 33, 35, 50, 51, 57, 58] whilst nine studies demonstrated no significant effect [38-40, 42, 45, 46,

D

48, 63, 64]. By applying a NOS inhibitor (L-NAME, 1.5 or 3.0 mg/kg) 30 to 90 minutes prior to

PT E

BCCAO, the positive results of tadalafil (10 mg/kg) on lesion volume were abolished [57, 58]. Inhibitors of sGC (ODQ, 3.5 mg/kg), PKA (IP-20, 0.34 mg/kg), or PKG (DT-3, 1 mg/kg)

CE

administered intraperitoneal immediately after MCAO, likewise eradicated the reduced lesion

AC

volume achieved by sildenafil or yonkenafil (both 16 mg/kg) [28, 30]. One study demonstrated, by use of MRI, that the stroke volume was reduced in PDE5i-treated mice exposed to temporary MCAO, though this could not be verified histologically [53]. See table 1 for details.

3.8 Study quality The median number of ARRIVE criteria scored was 21 out of 38 (interquartile range: 20-24; full range: 17-28), and CAMARADES score was four out of ten (interquartile range: 3.5-5; full range: 2-6). Some of the most commonly reported ARRIVE criteria, except for title and abstract, were

ACCEPTED MANUSCRIPT

sufficient scientific background, details of how the procedure was carried out, and interpretation of results. 15 out of 32 studies (47%) reported randomization and only 22% reported blinded assessment of outcome. None of the studies reported a sample size calculation and only 66% provided a statement regarding conflict of interest. See supplementary material 2.

PT

4. Discussion

RI

In this systematic review, the effects of selective PDE5i as a potential therapeutic target in stroke,

SC

have been evaluated. The reported neuroprotective effects are diverse, including reduced neuronal apoptosis, oxidative stress and brain inflammation. Increased neurogenesis is seen in different areas

NU

of the brain, together with preserved synapse function, increased angiogenesis and augmented cerebral blood flow to the ischemic penumbra. These results suggest that PDE5i pass the blood

MA

brain barrier and directly affect brain tissue, in accordance with previous findings [14, 65]. The neuroprotective effects manifest as improved functional outcome and, in some studies, as reduced

D

stroke lesion volume, at doses relevant for human use. PDE5i were in general successful when

PT E

administered within the first 24 hours of stroke and one study even showed that delayed treatment up to seven days post-ischemia was effective in promoting neurogenesis and improving functional

CE

outcome.

AC

In ischemic stroke, regional cerebral blood flow is reduced to the brain area supplied by the occluded artery or arteriole. In the ischemic core, neurons are lost, however the surrounding ischemic penumbra can be salvaged if sufficient blood flow to the area is achieved in a timely fashion. The studies in this review showed that treatment with PDE5i in animals post-stroke increased formation of new blood vessel in the ischemic penumbra, protected the existing vessels, and increased blood flow to the ischemic penumbra [39, 42, 43, 45-47, 49]. These discoveries highlight a likely beneficial effect of PDE5i on the brain vascular system in animals. The ability of

ACCEPTED MANUSCRIPT

PDE5i to dilate vessels, and increase blood flow in peripheral vascular beds, is well known from their effects in human erectile dysfunction and pulmonary hypertension [12]. PDE5i effects on the vascular system post-stroke support the hypothesis that increasing local blood flow to the penumbral zone protects neuronal tissue and improves functional outcome. An increase in perfusion

PT

of neuronal ischemic tissue, in addition to the direct anti-apoptotic effect of PDE5i on neurons, is most likely involved in the reduced stroke volume and improvement of functional outcome shown

RI

by some studies after PDE5 inhibition. Further, improved functional outcome may also result from

SC

increased perfusion in neuronal tissue outside the ischemic area, thus improving neuronal network

NU

connectivity.

The medications applied in the included studies are selective PDE5i, which specifically inhibit the

MA

intracellular cGMP-specific phosphodiesterase type 5 enzyme. Some of the PDE5i may to a lesser degree also inhibit other phosphodiesterase isoenzymes. For example, tadalafil also inhibits PDE11

D

(IC50 = 37nM) which degrades both cAMP and cGMP [66]. Due to the selectivity of the

PT E

medication, the neuroprotective and neurorestorative actions of PDE5i in animal models of stroke are most likely due to an increase in brain cGMP concentration with subsequent upstream or down-

CE

stream effects. Some studies assessed whether a concomitant increase in brain cGMP concentration was seen. Following cerebral ischemia (MCAO and BCCAO) PDE5i-treated animals had a

AC

significantly higher concentration of cGMP in brain tissue homogenates compared to control treated animals [31, 38, 39, 45, 50, 67]. As further confirmation, neurorestorative effects seen following treatment with PDE5i were abolished when inhibitors of downstream or upstream targets of cGMP were co-administered. Simultaneous treatment with inhibitors of NOS, sGC, PKG, or PKA eliminated the neurorestorative effect seen by PDE5i alone [28, 30, 57, 58]. This indicates strongly that the effects seen by PDE5i are dependent on the NO-cGMP-PKG-pathway and PKA is further involved. cGMP is known to regulate various cellular processes, including cell growth,

ACCEPTED MANUSCRIPT

inflammation, neuronal plasticity, and cardiovascular homeostasis [68]. Thus, PDE5i, via an increase in cGMP, most likely affect numerous intracellular signalling pathways post-stroke depending on the cell type and tissue involved. The cellular signalling pathways behind the neuroprotective and neurorestorative effects seen by PDE5i in animal models of stroke are still to

PT

be discovered.

RI

Several of the included studies investigated different cellular pathways that could be involved in the neuroprotective and neurorestorative effects seen by PDE5i in animal models of stroke. PDE5i

SC

induced activation of the Akt pathway and the increase and activation of the growth factors TrkB-

NU

BDNF and TrkA-NGF could play important roles in mediating increased protection and functioning of neurons and synapses post-stroke, since these signalling pathways were augmented in the PDE5i

MA

treated animals [28, 30, 69-73]. Of note, PDE5i was shown to reduce oxidative stress, reduce activation of the intracellular apoptotic cascade, and reduce different cellular signaling pathways

D

involved in promoting cell death, which is of importance to increase cell survival post-stroke [28,

PT E

29, 31-33, 57, 58]. Oxidative stress is one of the most prominent contributors to the range of tissue damage after stroke. It oxidizes and causes dysfunction of cellular macromolecules, leading to

CE

apoptosis [74]. One of the important antioxidants increased by PDE5i post-stroke is SOD which is involved in preventing activation of the apoptotic p38-MAPK pathway [75] and interestingly it can

AC

endorse the Akt pathway [76].

In general, the included studies were unambiguous regarding positive effects of PDE5i in stroke. However, some contradictory findings were described, which could be caused by the variability in stroke models, treatment regimens, and animals used. In regards to angiogenesis and cerebral blood flow, one study described no difference in capillary length as an estimate for angiogenesis [48], and two studies showed no difference in CBF [50, 53]. Common for these three studies were the use of

ACCEPTED MANUSCRIPT

mice instead of rats, the latter was the animal of choice in the studies showing increased angiogenesis and CBF. Thus, a species difference cannot be excluded. In addition, one of the studies applied a lower dose of sildenafil compared to the remaining [48]. Inconsistent results were also found regarding PDE5i effect on nitrate/nitrite levels (oxidative stress marker) [29, 57, 58]

PT

post-stroke; two studies showed no effect of PDE5i on oxidative stress post-stroke [59, 60]. The cause of variability in results is not clear but could involve variations in treatments and doses.

RI

Major discrepancy in results were the reports on stroke volume after PDE5i compared to control

SC

treatment in ischemic animals. Half of the included studies, which assessed stroke volume, showed a decrease in stroke volume after PDE5i compared to controls, whereas the others did not. No

NU

studies showed an increase in stroke volume after PDE5i. The studies showing reduced stroke

MA

volume after PDE5i tended to initiate the PDE5i-treatment early (within few hours), or even as a pre-treatment before induced ischemia, compared to the studies with no effect on stroke volume.

D

This finding is in line with the fact that neuronal tissue is highly sensitive to ischemia, and time is a

PT E

major factor in stroke outcome. Functional outcome shows high consistency in results with most studies reporting positive effects of PDE5 inhibition post-stroke compared to control groups. The reason for some studies not showing effect on functional outcome is not clear as the studies had

CE

very similar designs to the ones showing positive effects, though minor but important study

AC

differences may still prevail.

The methodological scores on the ARRIVE and CAMERADES scales were generally low for the included studies which may bias results. While the exclusion of non-English articles is a potential risk of bias, only one publication in Chinese was found using the search string described. To minimize bias, the search string used free text search, not limited to MeSH terms.

ACCEPTED MANUSCRIPT

Few trials have studied the effects of selective PDE5i in human stroke patients [77-80]. Preclinical ischemic stroke models do not necessarily represent the pathophysiology of ischemic stroke in humans, in part because experimental animals are usually young and male, without comorbidities. Further, PDE5i was applied pre-ischemia in many studies, a treatment regimen not feasible for

PT

human ischemic stroke.

RI

5. Conclusion

SC

In animal models of ischemic stroke, PDE5i demonstrate neuroprotective effects dependent on cGMP downstream mechanisms. PDE5i improved functional outcome and, in some studies,

NU

reduced stroke lesion volume. Outcomes were related to reduced brain inflammation, oxidative stress, and neuronal apoptosis and formation of new neurons and blood vessels in ischemic brain

MA

tissue. There is limited evidence on PDE5i effects in clinical stroke, though all data suggest that currently licensed PDE5i are safe to use in stroke patients. New treatment targets are warranted, not

D

only restoring perfusion but also aiming at neurorestoration of lesioned tissue post-ischemia. PDE5

PT E

may be such a treatment target and well-characterized, specific PDE5i are available to test this.

CE

6. Authors’ contributions

JNEÖ screened all publications, organized data extraction, scored included publications for quality,

AC

and prepared the outline of the paper. IRM repeated the search and screened publications for eligibility, and contributed to the paper. AHH contributed to the writing of the paper. CK initiated the study, and contributed to the outline, design, and writing of the paper. The authors approved this version of the article.

7. Acknowledgement This work was supported by a grant from the Capital Region Research Fund (no grant number), by an internationalization scholarship from University of Copenhagen (no grant number), and by a

ACCEPTED MANUSCRIPT

scholarship from Lundbeck Foundation (no grant number). AHH and CK gratefully acknowledge funding from the UK Alzheimer’s Society and from Alzheimer's Drug Discovery Foundation (grant ref. 20140901). The funding sources had no involvement in study design, analysis or writing of this article.

PT

8. Conflict of interest

AC

CE

PT E

D

MA

NU

SC

RI

The authors declare no conflict of interest.

ACCEPTED MANUSCRIPT

References 1.

Krishnamurthi, R.V., V.L. Feigin, M.H. Forouzanfar, G.A. Mensah, M. Connor, D.A. Bennett, A.E. Moran, R.L. Sacco, L.M. Anderson, T. Truelsen, M. O'Donnell, N. Venketasubramanian, S. Barker-Collo, C.M. Lawes, W. Wang, Y. Shinohara, E. Witt, M.

PT

Ezzati, M. Naghavi, and C. Murray, Global and regional burden of first-ever ischaemic and

2010. Lancet Glob Health, 2013. 1(5): p. e259-81.

O'Collins, V.E., M.R. Macleod, G.A. Donnan, L.L. Horky, B.H. van der Worp, and D.W.

SC

2.

RI

haemorrhagic stroke during 1990-2010: findings from the Global Burden of Disease Study

NU

Howells, 1,026 experimental treatments in acute stroke. Ann Neurol, 2006. 59(3): p. 46777.

Siket, M.S., Treatment of Acute Ischemic Stroke. Emerg Med Clin North Am, 2016. 34(4):

MA

3.

p. 861-882.

Teich, A.F., M. Sakurai, M. Patel, C. Holman, F. Saeed, J. Fiorito, and O. Arancio, PDE5

D

4.

PT E

Exists in Human Neurons and is a Viable Therapeutic Target for Neurologic Disease. J Alzheimers Dis, 2016. 52(1): p. 295-302. Rybalkin, S.D., I.G. Rybalkina, R. Feil, F. Hofmann, and J.A. Beavo, Regulation of cGMP-

CE

5.

specific phosphodiesterase (PDE5) phosphorylation in smooth muscle cells. J Biol Chem,

6.

AC

2002. 277(5): p. 3310-7.

Kruuse, C., T.S. Khurana, S.D. Rybalkin, S. Birk, U. Engel, L. Edvinsson, and J. Olesen, Phosphodiesterase 5 and effects of sildenafil on cerebral arteries of man and guinea pig. Eur J Pharmacol, 2005. 521(1-3): p. 105-14.

7.

Francis, S.H., T.M. Lincoln, and J.D. Corbin, Characterization of a novel cGMP binding protein from rat lung. J Biol Chem, 1980. 255(2): p. 620-6.

8.

Arnold, W.P., C.K. Mittal, S. Katsuki, and F. Murad, Nitric oxide activates guanylate

ACCEPTED MANUSCRIPT

cyclase and increases guanosine 3':5'-cyclic monophosphate levels in various tissue preparations. Proc Natl Acad Sci U S A, 1977. 74(8): p. 3203-7. 9.

Bender, A.T. and J.A. Beavo, Cyclic nucleotide phosphodiesterases: molecular regulation to clinical use. Pharmacol Rev, 2006. 58(3): p. 488-520. Ciani, E., S. Guidi, R. Bartesaghi, and A. Contestabile, Nitric oxide regulates cGMP-

PT

10.

dependent cAMP-responsive element binding protein phosphorylation and Bcl-2 expression

RI

in cerebellar neurons: implication for a survival role of nitric oxide. J Neurochem, 2002.

11.

SC

82(5): p. 1282-9.

Lu, Y.F., E.R. Kandel, and R.D. Hawkins, Nitric oxide signaling contributes to late-phase

Ghofrani, H.A., I.H. Osterloh, and F. Grimminger, Sildenafil: from angina to erectile

MA

12.

NU

LTP and CREB phosphorylation in the hippocampus. J Neurosci, 1999. 19(23): p. 10250-61.

dysfunction to pulmonary hypertension and beyond. Nat Rev Drug Discov, 2006. 5(8): p.

Zhu, L., J.Y. Yang, X. Xue, Y.X. Dong, Y. Liu, F.R. Miao, Y.F. Wang, H. Xue, and C.F.

PT E

13.

D

689-702.

Wu, A novel phosphodiesterase-5 Inhibitor: Yonkenafil modulates neurogenesis, gliosis to improve cognitive function and ameliorates amyloid burden in an APP/PS1 transgenic mice

Garcia-Barroso, C., A. Ricobaraza, M. Pascual-Lucas, N. Unceta, A.J. Rico, M.A. Goicolea,

AC

14.

CE

model. Mech Ageing Dev, 2015. 150: p. 34-45.

J. Salles, J.L. Lanciego, J. Oyarzabal, R. Franco, M. Cuadrado-Tejedor, and A. Garcia-Osta, Tadalafil crosses the blood-brain barrier and reverses cognitive dysfunction in a mouse model of AD. Neuropharmacology, 2013. 64: p. 114-23. 15.

Nunes, A.K., C. Raposo, R.L. Luna, M.A. Cruz-Hofling, and C.A. Peixoto, Sildenafil (Viagra(R)) down regulates cytokines and prevents demyelination in a cuprizone-induced MS mouse model. Cytokine, 2012. 60(2): p. 540-51.

ACCEPTED MANUSCRIPT

16.

Hernandez-Rabaza, V., A. Agusti, A. Cabrera-Pastor, S. Fustero, O. Delgado, L. TaoroGonzalez, C. Montoliu, M. Llansola, and V. Felipo, Sildenafil reduces neuroinflammation and restores spatial learning in rats with hepatic encephalopathy: underlying mechanisms. J Neuroinflammation, 2015. 12: p. 195. Koka, S., A. Das, F.N. Salloum, and R.C. Kukreja, Phosphodiesterase-5 inhibitor tadalafil

PT

17.

attenuates oxidative stress and protects against myocardial ischemia/reperfusion injury in

Wang, G., Q. Zhang, W. Yuan, J. Wu, and C. Li, Sildenafil Protects against Myocardial

SC

18.

RI

type 2 diabetic mice. Free Radic Biol Med, 2013. 60: p. 80-8.

Ischemia-Reperfusion Injury Following Cardiac Arrest in a Porcine Model: Possible Role

Kucuk, A., M. Yucel, N. Erkasap, M. Tosun, T. Koken, M. Ozkurt, and S. Erkasap, The

MA

19.

NU

of the Renin-Angiotensin System. Int J Mol Sci, 2015. 16(11): p. 27015-31.

effects of PDE5 inhibitory drugs on renal ischemia/reperfusion injury in rats. Mol Biol Rep,

Medeiros, P.J., A. Villarim Neto, F.P. Lima, I.M. Azevedo, L.R. Leao, and A.C. Medeiros,

PT E

20.

D

2012. 39(10): p. 9775-82.

Effect of sildenafil in renal ischemia/reperfusion injury in rats. Acta Cir Bras, 2010. 25(6): p. 490-5.

Zahran, M.H., A.M. Hussein, N. Barakat, A. Awadalla, S. Khater, A. Harraz, and A.A.

CE

21.

AC

Shokeir, Sildenafil activates antioxidant and antiapoptotic genes and inhibits proinflammatory cytokine genes in a rat model of renal ischemia/reperfusion injury. Int Urol Nephrol, 2015. 47(11): p. 1907-15. 22.

Bektas, S., K. Karakaya, M. Can, B. Bahadir, B. Guven, N. Erdogan, and S.O. Ozdamar, The effects of tadalafil and pentoxifylline on apoptosis and nitric oxide synthase in liver ischemia/reperfusion injury. Kaohsiung J Med Sci, 2016. 32(7): p. 339-47.

23.

Yildiz, H., A.S. Durmus, H. Simsek, and M. Yaman, Dose-dependent protective effect of

ACCEPTED MANUSCRIPT

sildenafil citrate on testicular injury after torsion/detorsion in rats. Andrologia, 2012. 44 Suppl 1: p. 300-6. 24.

Erol, B., H. Tokgoz, V. Hanci, S. Bektas, B. Akduman, F. Yencilek, G. Mungan, and A. Mungan, Vardenafil reduces testicular damage following ischemia/reperfusion injury in

25.

PT

rats. Kaohsiung J Med Sci, 2009. 25(7): p. 374-80. Arikan, D.C., V. Bakan, E.B. Kurutas, H. Sayar, and A. Coskun, Protective effect of

RI

tadalafil on ischemia/reperfusion injury of rat ovary. J Pediatr Surg, 2010. 45(11): p. 2203-

26.

SC

9.

Kilkenny, C., W.J. Browne, I.C. Cuthill, M. Emerson, and D.G. Altman, Improving

NU

bioscience research reporting: the ARRIVE guidelines for reporting animal research. PLoS

27.

MA

Biol, 2010. 8(6): p. e1000412.

Macleod, M.R., T. O'Collins, D.W. Howells, and G.A. Donnan, Pooling of animal

D

experimental data reveals influence of study design and publication bias. Stroke, 2004.

28.

PT E

35(5): p. 1203-8.

Chen, X., N. Wang, Y. Liu, Y. Liu, T. Zhang, L. Zhu, Y. Wang, C. Wu, and J. Yang, Yonkenafil: a novel phosphodiesterase type 5 inhibitor induces neuronal network

CE

potentiation by a cGMP-dependent Nogo-R axis in acute experimental stroke. Exp Neurol,

29.

AC

2014. 261: p. 267-77.

Altas, M., M. Aras, S. Meydan, E. Nacar, K.T. Ulutas, Y. Serarslan, and N. Yilmaz, Effects of tadalafil on ischemia/reperfusion injury in rat brain. Acta Neurol Belg, 2014. 114(1): p. 33-40.

30.

Chen, X.M., N.N. Wang, T.Y. Zhang, F. Wang, C.F. Wu, and J.Y. Yang, Neuroprotection by sildenafil: neuronal networks potentiation in acute experimental stroke. CNS Neurosci Ther, 2014. 20(1): p. 40-9.

ACCEPTED MANUSCRIPT

31.

Ko, I.G., M.S. Shin, B.K. Kim, S.E. Kim, Y.H. Sung, T.S. Kim, M.C. Shin, H.J. Cho, S.C. Kim, S.H. Kim, K.H. Kim, D.H. Shin, and C.J. Kim, Tadalafil improves short-term memory by suppressing ischemia-induced apoptosis of hippocampal neuronal cells in gerbils. Pharmacol Biochem Behav, 2009. 91(4): p. 629-35. Barros-Minones, L., D. Martin-de-Saavedra, S. Perez-Alvarez, L. Orejana, V. Suquia, B.

PT

32.

Goni-Allo, I. Hervias, M.G. Lopez, J. Jordan, N. Aguirre, and E. Puerta, Inhibition of

RI

calpain-regulated p35/cdk5 plays a central role in sildenafil-induced protection against

33.

SC

chemical hypoxia produced by malonate. Biochim Biophys Acta, 2013. 1832(6): p. 705-17. Barros-Minones, L., L. Orejana, B. Goni-Allo, V. Suquia, I. Hervias, N. Aguirre, and E.

NU

Puerta, Modulation of the ASK1-MKK3/6-p38/MAPK signalling pathway mediates sildenafil

MA

protection against chemical hypoxia caused by malonate. Br J Pharmacol, 2013. 168(8): p. 1820-34.

Godinho, J., J.N. de Oliveira, E.D. Ferreira, G.G. Zaghi, C.C. Bacarin, R.M. de Oliveira,

D

34.

PT E

and H. Milani, Cilostazol but not sildenafil prevents memory impairment after chronic cerebral hypoperfusion in middle-aged rats. Behav Brain Res, 2015. 283: p. 61-8. 35.

Charriaut-Marlangue, C., T. Nguyen, P. Bonnin, A.P. Duy, P.L. Leger, Z. Csaba, J. Pansiot,

CE

T. Bourgeois, S. Renolleau, and O. Baud, Sildenafil mediates blood-flow redistribution and

36.

AC

neuroprotection after neonatal hypoxia-ischemia. Stroke, 2014. 45(3): p. 850-6. Romanini, C.V., A.P. Schiavon, E.D. Ferreira, R.M. de Oliveira, and H. Milani, Sildenafil prevents mortality and reduces hippocampal damage after permanent, stepwise, 4-vessel occlusion in rats. Brain Res Bull, 2010. 81(6): p. 631-40. 37.

Dias Fiuza Ferreira, E., C. Valerio Romanini, P.E. Cypriano, R.M. Weffort de Oliveira, and H. Milani, Sildenafil provides sustained neuroprotection in the absence of learning recovery following the 4-vessel occlusion/internal carotid artery model of chronic cerebral

ACCEPTED MANUSCRIPT

hypoperfusion in middle-aged rats. Brain Res Bull, 2013. 90: p. 58-65. 38.

Zhang, R., Y. Wang, L. Zhang, Z. Zhang, W. Tsang, M. Lu, L. Zhang, and M. Chopp, Sildenafil (Viagra) induces neurogenesis and promotes functional recovery after stroke in rats. Stroke, 2002. 33(11): p. 2675-80. Zhang, L., Z. Zhang, R.L. Zhang, Y. Cui, M.C. LaPointe, B. Silver, and M. Chopp,

PT

39.

Tadalafil, a long-acting type 5 phosphodiesterase isoenzyme inhibitor, improves

RI

neurological functional recovery in a rat model of embolic stroke. Brain Res, 2006. 1118(1):

40.

SC

p. 192-8.

Zhang, R.L., Z. Zhang, L. Zhang, Y. Wang, C. Zhang, and M. Chopp, Delayed treatment

NU

with sildenafil enhances neurogenesis and improves functional recovery in aged rats after

41.

MA

focal cerebral ischemia. J Neurosci Res, 2006. 83(7): p. 1213-9. Alexander, A.L., J.E. Lee, M. Lazar, and A.S. Field, Diffusion tensor imaging of the brain.

Ding, G., Q. Jiang, L. Li, L. Zhang, Z. Zhang, M. Lu, Q. Li, S. Gu, J. Ewing, and M. Chopp,

PT E

42.

D

Neurotherapeutics, 2007. 4(3): p. 316-29.

Longitudinal magnetic resonance imaging of sildenafil treatment of embolic stroke in aged rats. Stroke, 2011. 42(12): p. 3537-41. Ding, G., Q. Jiang, L. Li, L. Zhang, Z.G. Zhang, K.A. Ledbetter, S. Panda, S.P. Davarani, H.

CE

43.

AC

Athiraman, Q. Li, J.R. Ewing, and M. Chopp, Magnetic resonance imaging investigation of axonal remodeling and angiogenesis after embolic stroke in sildenafil-treated rats. J Cereb Blood Flow Metab, 2008. 28(8): p. 1440-8. 44.

Xu, W., PSD-95-like membrane associated guanylate kinases (PSD-MAGUKs) and synaptic plasticity. Curr Opin Neurobiol, 2011. 21(2): p. 306-12.

45.

Zhang, L., R.L. Zhang, Y. Wang, C. Zhang, Z.G. Zhang, H. Meng, and M. Chopp, Functional recovery in aged and young rats after embolic stroke: treatment with a

ACCEPTED MANUSCRIPT

phosphodiesterase type 5 inhibitor. Stroke, 2005. 36(4): p. 847-52. 46.

Li, L., Q. Jiang, L. Zhang, G. Ding, Z. Gang Zhang, Q. Li, J.R. Ewing, M. Lu, S. Panda, K.A. Ledbetter, P.A. Whitton, and M. Chopp, Angiogenesis and improved cerebral blood flow in the ischemic boundary area detected by MRI after administration of sildenafil to rats

47.

PT

with embolic stroke. Brain Res, 2007. 1132(1): p. 185-92. Zhang, R., L. Wang, L. Zhang, J. Chen, Z. Zhu, Z. Zhang, and M. Chopp, Nitric oxide

RI

enhances angiogenesis via the synthesis of vascular endothelial growth factor and cGMP

48.

SC

after stroke in the rat. Circ Res, 2003. 92(3): p. 308-13.

Wang, Q.M., Y. Wei, Y. Zheng, and C. Waeber, Efficacy of combined atorvastatin and

NU

sildenafil in promoting recovery after ischemic stroke in mice. Am J Phys Med Rehabil,

49.

MA

2013. 92(2): p. 143-50.

Ding, G., Q. Jiang, L. Li, L. Zhang, Z.G. Zhang, K.A. Ledbetter, L. Gollapalli, S. Panda, Q.

D

Li, J.R. Ewing, and M. Chopp, Angiogenesis detected after embolic stroke in rat brain using

50.

PT E

magnetic resonance T2*WI. Stroke, 2008. 39(5): p. 1563-8. Moretti, R., P.L. Leger, V.C. Besson, Z. Csaba, J. Pansiot, L. Di Criscio, A. Gentili, L. Titomanlio, P. Bonnin, O. Baud, and C. Charriaut-Marlangue, Sildenafil, a cyclic GMP

CE

phosphodiesterase inhibitor, induces microglial modulation after focal ischemia in the

51.

AC

neonatal mouse brain. J Neuroinflammation, 2016. 13(1): p. 95. Gao, F., M. Sugita, and H. Nukui, Phosphodiesterase 5 inhibitor, zaprinast, selectively increases cerebral blood flow in the ischemic penumbra in the rat brain. Neurol Res, 2005. 27(6): p. 638-43. 52.

Williams, D.S., J.A. Detre, J.S. Leigh, and A.P. Koretsky, Magnetic resonance imaging of perfusion using spin inversion of arterial water. Proc Natl Acad Sci U S A, 1992. 89(1): p. 212-6.

ACCEPTED MANUSCRIPT

53.

Royl, G., M. Balkaya, S. Lehmann, S. Lehnardt, K. Stohlmann, U. Lindauer, M. Endres, U. Dirnagl, and A. Meisel, Effects of the PDE5-inhibitor vardenafil in a mouse stroke model. Brain Res, 2009. 1265: p. 148-57.

54.

Jiwa, N.S., P. Garrard, and A.H. Hainsworth, Experimental models of vascular dementia

PT

and vascular cognitive impairment: a systematic review. J Neurochem, 2010. 115(4): p. 814-28.

Behrens, A., M. Sibilia, and E.F. Wagner, Amino-terminal phosphorylation of c-Jun

RI

55.

SC

regulates stress-induced apoptosis and cellular proliferation. Nat Genet, 1999. 21(3): p. 326-9.

Zhao, S., L. Zhang, G. Lian, X. Wang, H. Zhang, X. Yao, J. Yang, and C. Wu, Sildenafil

NU

56.

MA

attenuates LPS-induced pro-inflammatory responses through down-regulation of intracellular ROS-related MAPK/NF-kappaB signaling pathways in N9 microglia. Int

Gulati, P. and N. Singh, Tadalafil enhances the neuroprotective effects of ischemic

PT E

57.

D

Immunopharmacol, 2011. 11(4): p. 468-74.

postconditioning in mice, probably in a nitric oxide associated manner. Can J Physiol Pharmacol, 2014. 92(5): p. 418-26. Gulati, P. and N. Singh, Neuroprotective effect of tadalafil, a PDE-5 inhibitor, and its

CE

58.

AC

modulation by L-NAME in mouse model of ischemia-reperfusion injury. J Surg Res, 2014. 186(1): p. 475-83. 59.

Gaur, V. and A. Kumar, Behavioral, biochemical and cellular correlates in the protective effect of sertraline against transient global ischemia induced behavioral despair: possible involvement of nitric oxide-cyclic guanosine monophosphate study pathway. Brain Res Bull, 2010. 82(1-2): p. 57-64.

60.

Gaur, V. and A. Kumar, Protective effect of desipramine, venlafaxine and trazodone against

ACCEPTED MANUSCRIPT

experimental animal model of transient global ischemia: possible involvement of NO-cGMP pathway. Brain Res, 2010. 1353: p. 204-12. 61.

Menniti, F.S., J. Ren, D.K. Sietsma, A. Som, F.R. Nelson, D.T. Stephenson, B.A. Tate, and S.P. Finklestein, A non-brain penetrant PDE5A inhibitor improves functional recovery after

62.

PT

stroke in rats. Restor Neurol Neurosci, 2012. 30(4): p. 283-9. Menniti, F.S., J. Ren, T.M. Coskran, J. Liu, D. Morton, D.K. Sietsma, A. Som, D.T.

RI

Stephenson, B.A. Tate, and S.P. Finklestein, Phosphodiesterase 5A inhibitors improve

SC

functional recovery after stroke in rats: optimized dosing regimen with implications for mechanism. J Pharmacol Exp Ther, 2009. 331(3): p. 842-50. Novitzky, I., N.J. Marianayagam, S. Weiss, O. Muhsinoglu, M. Fridman, T.A. Leibovitch,

NU

63.

MA

N. Goldenberg-Cohen, and S. Michowiz, Comparison of Neuroprotective Effect of Bevacizumab and Sildenafil following Induction of Stroke in a Mouse Model. Biomed Res

Zhang, R.L., M. Chopp, C. Roberts, M. Wei, X. Wang, X. Liu, M. Lu, and Z.G. Zhang,

PT E

64.

D

Int, 2016. 2016: p. 3938523.

Sildenafil enhances neurogenesis and oligodendrogenesis in ischemic brain of middle-aged mouse. PLoS One, 2012. 7(10): p. e48141. Gomez-Vallejo, V., A. Ugarte, C. Garcia-Barroso, M. Cuadrado-Tejedor, B. Szczupak, I.G.

CE

65.

AC

Dopeso-Reyes, J.L. Lanciego, A. Garcia-Osta, J. Llop, J. Oyarzabal, and R. Franco, Pharmacokinetic investigation of sildenafil using positron emission tomography and determination of its effect on cerebrospinal fluid cGMP levels. J Neurochem, 2016. 136(2): p. 403-15. 66.

Bischoff, E., Potency, selectivity, and consequences of nonselectivity of PDE inhibition. Int J Impot Res, 2004. 16 Suppl 1: p. S11-4.

67.

Kim, K.T., K.J. Chung, H.S. Lee, I.G. Ko, C.J. Kim, Y.G. Na, and K.H. Kim,

ACCEPTED MANUSCRIPT

Neuroprotective effects of tadalafil on gerbil dopaminergic neurons following cerebral ischemia. Neural Regen Res, 2013. 8(8): p. 693-701. 68.

Francis, S.H., J.L. Busch, J.D. Corbin, and D. Sibley, cGMP-dependent protein kinases and cGMP phosphodiesterases in nitric oxide and cGMP action. Pharmacol Rev, 2010. 62(3): p.

69.

PT

525-63. Song, G., G. Ouyang, and S. Bao, The activation of Akt/PKB signaling pathway and cell

Tonges, L., J.C. Koch, M. Bahr, and P. Lingor, ROCKing Regeneration: Rho Kinase

SC

70.

RI

survival. J Cell Mol Med, 2005. 9(1): p. 59-71.

Inhibition as Molecular Target for Neurorestoration. Front Mol Neurosci, 2011. 4: p. 39. Greenberg, M.E., B. Xu, B. Lu, and B.L. Hempstead, New insights in the biology of BDNF

NU

71.

72.

MA

synthesis and release: implications in CNS function. J Neurosci, 2009. 29(41): p. 12764-7. Cui, X., M. Chopp, A. Zacharek, C. Roberts, B. Buller, M. Ion, and J. Chen, Niacin

D

treatment of stroke increases synaptic plasticity and axon growth in rats. Stroke, 2010.

73.

PT E

41(9): p. 2044-9.

Jang, S.W., M. Okada, I. Sayeed, G. Xiao, D. Stein, P. Jin, and K. Ye, Gambogic amide, a selective agonist for TrkA receptor that possesses robust neurotrophic activity, prevents

Loh, K.P., S.H. Huang, R. De Silva, B.K. Tan, and Y.Z. Zhu, Oxidative stress: apoptosis in

AC

74.

CE

neuronal cell death. Proc Natl Acad Sci U S A, 2007. 104(41): p. 16329-34.

neuronal injury. Curr Alzheimer Res, 2006. 3(4): p. 327-37. 75.

Nito, C., H. Kamada, H. Endo, K. Niizuma, D.J. Myer, and P.H. Chan, Role of the p38 mitogen-activated protein kinase/cytosolic phospholipase A2 signaling pathway in bloodbrain barrier disruption after focal cerebral ischemia and reperfusion. J Cereb Blood Flow Metab, 2008. 28(10): p. 1686-96.

76.

Noshita, N., T. Sugawara, A. Lewen, T. Hayashi, and P.H. Chan, Copper-zinc superoxide

ACCEPTED MANUSCRIPT

dismutase affects Akt activation after transient focal cerebral ischemia in mice. Stroke, 2003. 34(6): p. 1513-8. 77.

Lorberboym, M., I. Mena, J. Wainstein, M. Boaz, and Y. Lampl, The effect of sildenafil citrate (Viagra) on cerebral blood flow in patients with cerebrovascular risk factors. Acta

78.

PT

Neurol Scand, 2010. 121(6): p. 370-6. Lorberboym, M., E. Makhline, and Y. Lampl, Regional cerebral blood flow following

RI

single-dose and continuous-dose tadalafil after stroke. Acta Neurol Scand, 2014. 130(6): p.

79.

SC

380-6.

Silver, B., S. McCarthy, M. Lu, P. Mitsias, A.N. Russman, A. Katramados, D.C. Morris,

NU

C.A. Lewandowski, and M. Chopp, Sildenafil treatment of subacute ischemic stroke: a

Di Cesare, F., J. Mancuso, P. Woodward, M.M. Bednar, and P.T. Loudon,

D

Phosphodiesterase-5 Inhibitor PF-03049423 Effect on Stroke Recovery: A Double-Blind,

PT E

Placebo-Controlled Randomized Clinical Trial. J Stroke Cerebrovasc Dis, 2016. 25(3): p.

CE

642-9.

AC

80.

MA

safety study at 25-mg daily for 2 weeks. J Stroke Cerebrovasc Dis, 2009. 18(5): p. 381-3.

ACCEPTED MANUSCRIPT Table 1: Overview of the included studies and the results of selective PDE5 inhibitors in animal models of stroke. Reference

Species

Ischemia duration

Temporary middle cerebral artery occlusion (tMCAO) [53] C57BL/6 mice, 7 45 min. weeks old, male.

[63]

C57Bl/6 mice, adult, male.

60 min.

[48]

Mice (wild-type C57 black/6, adult (2-3 months), male.

60 min.

[28]

[30]

Sprague-Dawley rats, 8 weeks old, male.

Sprague-Dawley rats, unknown age, male.

Treatment

Effects of PDE5i

Functional outcome and stroke volume

Vardenafil 10 mg/kg p.o. twice daily starting 3 h post ischemia and continued for 14 days.

→ CBF.

→ Modified Bederson test score, wire hanging test score, and pole test score.

Sildenafil, 24 µg i.p., singledose immediately after reperfusion. Sildenafil 0.3 mg/kg/day p.o. starting 24 h after MCAO and continued for 6 days.

T P

I R

C S U

→ Cerebral capillary length.

N A

D E

PT

2 h.

E C

C A 2 h.

Permanent middle cerebral artery occlusion (pMCAO)

M

Yonkenafil 4, 8, 16, 32 mg/kg i.p. or i.v. starting 2, 4, or 6 h after reperfusion.

→ Mortality. → Stroke volume. ↓ Mortality. → Stroke volume. → Adhesive removal test score, cylinder test score, and neurological score. → Mortality. → Body weight. → Stroke volume.

↓ Loss of neurons.

Improved beam walking test score and rotarod test score.

↑ Synaptic preservation. ↓ Neurological deficits. ↓ Apoptosis.

Sildenafil 4, 8, 16, 32 mg/kg i.p. or i.v. starting 2, 4, or 6 h after reperfusion.

↓ Loss of neurons.

↓ Stroke volume. Improved beam walking test score and rotarod test score.

↑ Synaptic preservation. ↓ Stroke volume.

ACCEPTED MANUSCRIPT [50]

C57Bl/6 mice pups.

Permanent.

Sildenafil 0.5, 2.5, 10, 15 mg/kg i.p. 5 min after MCAO induction.

↓ Inflammation.

↓ Stroke volume.

↑ cGMP brain concentration. → mBFV.

[64]

[61]

[42]

[62]

Nestin-CreERT2; R26R-stop-YFP mice, 14 months old, male.

Permanent.

Sprague-Dawley rats, unknown age, male.

Permanent.

Wistar rats, 18 months old, male.

Sprague-Dawley rats, unknown age, male.

Permanent.

Sildenafil 10 mg/kg/day s.c. for 7 days starting one day after MCAO.

↑ Neurogenesis.

PF-5 (1.0 mg/kg) or PF489,791 (10 mg/kg) twice daily s.c. starting 24 h after MCAO and continued for 7 days.

PF-5 crosses BBB. PF-489,791 does not cross BBB.

Improved limb placement test scores and body swing test score (effect by both PDE5i).

↑ Axonal remodeling.

Improved neurological severity score.

↓ Ventricle volume.

→ Adhesive removal test score, footfault test score.

D E

↑ Oligodendrogenesis.

C A

Permanent.

T P

I R

C S U

N A

M

Sildenafil 10 mg/kg/day s.c. starting 24 h after MCAO and continued for 7 days.

PT

E C

→ Stroke volume.

↑ Angiogenesis. → Stroke volume. ↑ CBF 6 weeks post-stroke.

Sildenafil (10 mg/kg) or PF-5 (0.1, 1, 10 mg/kg) s.c. Administration at different time-points.

Improved fore limp placement test score and body swing test score. PDE5i was effective when administration was given 72 h after MCAO. No effect of PDE5i when treatment was delayed to 14 days after MCAO. No difference in effectiveness between 7 or 28 days treatment.

ACCEPTED MANUSCRIPT [43]

Wistar rats, adult (812 weeks old), male.

Permanent.

[49]

Wistar rats, adult (812 weeks old), male.

Permanent.

[46]

Wistar rats, adult (812 weeks old), male.

Permanent.

[39]

Wistar rats, unknown age, male.

Tadalafil 2, 10 mg/kg p.o. starting 24 h after MCAO and continued every other day for 6 days (3 administrations).

↑ Axonal remodeling/density. ↑ Angiogenesis. ↑ CBF 6 weeks post-stroke. ↑ Angiogenesis.

Permanent.

[51]

Wistar rats, unknown age, male.

Permanent.

PT

E C

C A

Permanent.

D E

M

Sildenafil 3 mg/kg i.p. starting 7 days after MCAO and continued for 7 days.

Zaprinast 10 mg/kg i.v. starting 10 min before MCAO. Second i.v. infusion in infarct volume evaluation after 110 min from first i.v. infusion. Sildenafil 2 mg/kg p.o. or 10 mg/kg s.c. Administration started 24 h after MCAO and continued daily for additionally 6 days.

T P

↑ Angiogenesis.

I R

↑ CBF 6 weeks post-stroke.

C S U

↑ Neurogenesis.

Improved neurological severity score and foot fault test score. → Stroke volume.

↑ Angiogenesis.

Improved foot-fault test score, neurological severity score, and adhesive removal test score.

↑ cGMP brain concentration.

→ Stroke volume.

N A

Wistar rats, 3-4 months (young adults) and 18 months (aged), male.

Wistar rats, 8-12 weeks (young) and 18 months (aged), male.

Sildenafil 10 mg/kg s.c. starting 24 h after MCAO and continued for 7 days. Sildenafil 10 mg/kg s.c. starting 24 h after MCAO and continued for 7 days.

Permanent.

[40]

[45]

Sildenafil 10 mg/kg s.c. starting 24 h after MCAO and continued for 7 days.

→ cAMP brain concentration. ↑ Neurogenesis.

↑ rCBF.

Improved neurological severity score, adhesive removal test score, and footfault test score. → Stroke volume. ↓ Stroke volume.

↑ cGMP brain concentration.

↑ Synaptic preservation/synaptogenesis.

Improved neurological severity score, adhesive removal test score, foot-fault test score, and corner test score.

↑ Angiogenesis. → Mortality. ↑ cGMP brain concentration. → Stroke volume.

ACCEPTED MANUSCRIPT [47]

[38]

Wistar rats, unknown age, male.

Wistar rats, unknown age, male.

Permanent.

Sildenafil 2 mg/kg p.o. started 24 h after MCAO and continued for additionally 6 days. Sildenafil 2 or 5 mg/kg p.o. starting either 2 h or 24 h after MCAO and continued for additionally 6 days.

Permanent.

Bilateral common carotid artery occlusion (BCCAO) [67] Mongolian gerbils, 15 5 min. weeks old, male.

Tadalafil 0.1, 1, 10 mg/kg/day p.o. once daily for 7 days, starting 1 day after BCCAO.

[60]

Laca mice, unknown age, unknown sex.

5 min twice with 10 min interval.

[59]

Laca mice, unknown age, unknown sex.

5 min twice with 10 min interval.

[31]

Mongolian gerbils, 12-14 weeks old, male.

7 min.

A

↑ VEGF brain concentration. ↑ Neurogenesis.

Improved foot-fault test score and adhesive removal test score.

↑ cGMP brain concentration. ↓ Weight loss.

T P

I R

↑ cGMP brain concentration.

→ Stroke volume.

C S U

↓ Loss of D2 receptor concentration in striatum and substantia nigra. ↓ Increase in thyrosin hydroxylase expression in striatum and substantia nigra. → Oxidative stress. PDE5i reversed oxidative stress reduction by antidepressants.

N A

M

Sildenafil 5 mg/kg i.p. starting on day of surgery and continued for 4 days postsurgery. Sildenafil 5 mg/kg starting 2 h after BCCAO and continued to day 4. No information about route of administration.

D E

T P E

C C

↑ Angiogenesis.

Tadalafil, 0.1, 1, 10 mg/kg p.o. daily for 7 days, starting one day after BCCAO.

→ Oxidative stress and mitochondrial enzyme complex. PDE5i reversed oxidative stress reduction by sertraline.

→ Lateral push, forced swim test.

PDE5i suppressed ischemia-induced neurogenesis in the dentate gyrus.

Improved memory.

↓ Loss of neurons. ↓ Apoptosis. ↑ cGMP brain concentration.

ACCEPTED MANUSCRIPT [57]

Swiss mice, unknown age, male.

12 min.

Tadalafil 5, 10, 20 mg/kg p.o. 60 min prior to BCCAO.

↓ Oxidative stress.

Improved rotarod test score, beam walking test score, and lateral push test score. Improved memory.

[58]

Swiss mice, unknown age, male

12 min.

Tadalafil 5, 10, 20 mg/kg p.o. 60 min prior to BCCAO.

↓ Oxidative stress.

T P

I R

[29]

Wistar rats, unknown age, male.

20 min.

Tadalafil 2 mg/kg/day, administered once daily for 4 days before BCCAO. No information about route of administration.

D E

[36]

Wistar rats, young adult (3-4 months old), male.

C C

T P E

2 stages.

A

Right common carotid artery occlusion and hypoxia

SC

Sildenafil 0.75, 1.5, 3.0 mg/kg/day p.o. starting after first occlusion stage (3-VO) and continued for 7 days.

Improved memory. ↓ Stroke volume.

↓ Loss of neurons.

U N

↓ Oxidative stress.

A M

4-vessel occlusion/internal carotid artery model (4-VO/ICA) [34] Wistar rats, middle 3 stages. Sildenafil, 3.0 mg/kg/day p.o. aged (12-15 months starting 45 min after first old), male. occlusion stage and continued for 15 days. [37] Wistar rats, middle 3 stages. Sildenafil, 3.0 mg/kg/day p.o. aged (12-15 months starting shortly after first step old), male. in 4-VO/ICA and continued for 9 days.

↓ Stroke volume. Improved rotarod test score, beam walking test score, and lateral push test score.

↓ Loss of hippocampal neurons.

→ Memory.

→ Surviving neurons in cerebral cortex. ↓ Loss of hippocampal and cortical neurons.

→ Mortality. → Memory. → Mortality

↓ Loss of hippocampal neurons.

↓ Mortality.

ACCEPTED MANUSCRIPT [35]

Sprague-Dawley rat pups, both sexes.

Right common carotid artery occlusion followed by 120 min of hypoxia (FiO2 = 8%).

Sildenafil 5 or 10 mg/kg i.p. after hypoxia-ischemia.

↑ Blood vessel preservation.

Improved motor behaviour.

↓ BBB leakiness.

↓ Stroke volume.

↑ mBFV in basilar trunk and contralateral internal carotid artery. ↓ Inflammation.

T P

↓ Loss of hippocampal and cortical neurons. Chemical hypoxia stroke model [32] Wistar rats, unknown age, male.

[33]

Wistar rats, unknown age, male.

Intrastriatal malonate injection.

Intrastriatal malonate injection.

Sildenafil 1.5 mg/kg p.o. or i.v. Administration at – 30 min, + 10 min, + 1 h, + 3 h, and + 6 h compared to hypoxia induction. Sildenafil 1.5 mg/kg p.o. Administration 30 min before hypoxia induction.

C S U

↓ Loss of neurons.

D E

↓ Stroke volume.

↓ Apoptosis.

N A

M

I R

↓ Loss of neurons.

↓ Stroke volume.

Publications are listed according to the experimental model used. ↓ = decreased; ↑ = increased; and → = no change. Abbreviation list for table 1: BBB, blood brain barrier; cAMP, cyclic adenosine monophosphate; cGMP, cyclic guanosine monophosphate; CBF, cerebral blood flow; i.p., intra peritoneal; mBFV, mean blood flow velocity; PDE5i, phosphodiesterase 5 inhibitors; p.o., per os; rCBF, regional cerebral blood flow; s.c., subcutaneous; VEGF, vascular endothelial growth factor

T P E

A

C C

ACCEPTED MANUSCRIPT

Figure Captions: Figure 1: Flowchart of the literature selection process.

Figure 2: Proposed cellular mechanisms of PDE5i in stroke models.

PT

The central dashed box indicates the pharmacological effect of selective PDE5i. The main effect is proposed to be through cGMP and PKG activation. The following graphic explanations apply;

RI

Increased concentration/activity after cerebral ischemia, (upward red arrow concentration/activity after cerebral ischemia (downward red arrow

). Decreased

). Increased

SC

concentration/activity after PDE5i following cerebral ischemia (upward green arrow

). Decreased

concentration/activity after PDE5i following cerebral ischemia (downward green arrow

line with arrow

) and inhibits (line with bar

NU

Stimulates (line with arrow

) and blocks indirectly, (dashed line with bar

).

). Stimulates indirectly (dashed ).

MA

Abbreviations: Akt, Protein kinase B; apaf-1, apoptotic protease activating factor 1; Arg, arginine; ASK1, apoptosis signal-regulating kinase 1; Bax, bcl-2-like protein 4; Bcl-2, B-cell lymphoma 2 protein; BDNF, brain-derived neurotrophic factor; CBF, cerebral blood flow; cdk5, cyclin-

D

dependent kinase 5; cGMP, cyclic guanosine monophosphate; eNOS, endothelial nitric oxide

PT E

synthase; GSH, glutathione; GSSG, glutathione disulphide; GTP, guanosine triphosphate; hsp70, 70 kilodalton heat shock proteins; iNOS, inducible nitric oxide synthase; JNK, c-Jun N-terminal kinases; MDA, malondialdehyde; MKK3/6, mitogen-activated protein kinase kinase 3/6; MKK7,

CE

mitogen-activated protein kinase kinase 7; NGF, nerve growth factor; nNOS, neuronal nitric oxide synthase; NO, nitric oxide; Nogo-R, reticulon 4 receptor; NOX, NADPH oxidase; PDE5i,

AC

phosphodiesterase 5 inhibitors; PTEN, phosphatase and tensin homolog; PSD-95, postsynaptic density protein 95; p38, p38 mitogen-activated protein kinases; RhoA, Ras homolog gene family, member A; ROCK, Rho-associated protein kinase; sGC, soluble guanylyl cyclase; SOD, superoxide dismutase; synaptophysin, major synaptic vesicle protein p38; TBARS, thiobarbituric acid reactive substances; TrkA, Tropomyosin receptor kinase A; TrkB, Tropomyosin receptor kinase B

AC

CE

PT E

D

MA

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

Fig. 1

AC

Fig. 2

CE

PT E

D

MA

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT

Highlights: PDE5 inhibitors provide neuroprotection in animal models of stroke.



Treatment is dependent on the NO-cGMP-PKG pathway.



PDE5 inhibitors may reduce stroke volume and improve functional outcome.



PDE5 may be used as a therapeutic target against ischemic stroke in humans.

AC

CE

PT E

D

MA

NU

SC

RI

PT

