Abstract

Background: Aneurysmal subarachnoid haemorrhage (SAH) is a variant of haemorrhagic stroke with a striking 50% mortality rate. In addition to the initial insult, secondary delayed brain injury may occur days after the initial ischemic insult and is associated with vasospasms leading to delayed cerebral ischemia. We have previously shown that the MEK1/2 inhibitor U0126 improves neurological assessment after SAH in rats. Aim: The purpose of the present study was to analyse the impact of a broad selection of high potency MEK1/2 inhibitors in an organ culture model and use the IC50 values obtained from the organ culture to select highly potent inhibitors for pre-clinical in vivo studies. Results: Nine highly potent mitogen activated protein kinase kinase (MEK1/2) inhibitors were screened and the two most potent inhibitors from the organ culture screening, trametinib and PD0325901, were tested in an in vivo experimental rat SAH model with intrathecal injections. Subsequently, the successful inhibitor trametinib was administered intraperitoneally in a second in vivo study. In both regimens, trametinib treatment caused significant reductions in the endothelin-1 induced contractility after SAH, which is believed to be associated with endothelin B receptor up-regulation. Trametinib treated rats showed improved neurological scores, evaluated by the ability to traverse a rotating pole, after induced SAH. Conclusion: The PD0325901 treatment did not improve the neurological score after SAH, nor showed any beneficial therapeutic effect on the contractility, contrasting with the reduction in neurological deficits seen after trametinib treatment. These data show that trametinib might be a potential candidate for treatment of SAH.

Introduction

Aneurysmal subarachnoid haemorrhage (SAH) is a variant of haemorrhagic stroke causing around 5% of all stroke incidents, however with a striking 50% mortality rate [1]. Survivors will often have cognitive impairments and reduced quality of life, making it a very debilitating disease [2,3]. SAH is usually caused by the burst of an aneurysm, leading to a rapid leakage of arterial blood into the subarachnoid space, followed by a dramatic rise of the intracranial pressure (ICP) and a drop in the cerebral blood flow (CBF) [1]. This leads to oxygen and glucose deprivation of the brain resulting in cerebral ischemia and brain damage, often referred to as early brain injury [4]. Delayed cerebral ischemia (DCI) is associated with secondary delayed brain injury and is comprised of various pathophysiological changes, including inflammation [5], oedema [6] and blood–brain barrier disruption. We have postulated that the initiation of DCI is associated with remodelling and narrowing of the cerebral arteries, and especially the vascular hyperactivity that are often referred to as delayed cerebral vasospasm (CVS), of which there is currently few treatment options [4,7].

Vascular contractility has therefore been the focus of many clinical and pre-clinical studies attempting to prevent the following DCI. This includes recent attempts to modulate acute vascular contractility, for example with endothelin receptor antagonists including the specific endothelin A (ETA) and also endothelin B (ETB) receptor antagonist, clazosentan. These attempts have unfortunately not been successful [8–10].

We have used an alternative approach to prevent DCI, by limiting the enhanced vascular contractility seen after SAH. It was originally shown that the cessation of blood flow lead to up-regulation of contractile receptors in the cerebral vasculature, both seen in an organ culture (OC) model and following in vivo induction of experimental stroke in rodents [11,12]. Our current working hypothesis is that the prevention of phenotypical changes and subsequent effects on vascular contractility would reduce the following DCI. This is exemplified by a study, wherein clazosentan and the mitogen activated protein kinase kinase (MEK1/2) inhibitor U0126 were compared by their efficacy to prevent pathological modulation of the vasculature, after experimental SAH [13]. In that study, U0126 reduced the up-regulation of contractile receptor subtypes and improved early neurological assessment scores compared with vehicle (DMSO) or clazosentan treatment [13].

The OC of cerebral arteries has been used as a screening method by our group to explore targets involved in phenotypical vascular changes [11]. Arteries cultured for 24–48 h encounter cessation of flow, similar to an ischemic event. This leads to up-regulation of contractile receptors, e.g. 5-hydroxytrymptamine 1B (5-HT1B) receptors and the ETB receptor. Particularly in regard to the ETB receptors, the highly specific agonist (10000-fold selective for ETB over ETA) sarafotoxin 6C (S6c) has been proven a useful tool [14]. The receptor up-regulation on both protein level and functional level could be reduced/prevented by incubation with 10 µM U0126 in the OC [15,16]. This ex vivo method has proven valuable as a predictive method to screen for compounds, before further investigation in vivo.

The purpose of the present study was (I) to analyse the impact of a broad selection of potent inhibitors of the MEK1/2 pathway and compare the data to the established MEK1/2 inhibitor U0126, on the contractility to the specific ETB receptor agonist S6c in the OC model; (II) to use the IC50 values to select highly potent inhibitors for pre-clinical in vivo studies. The primary outcome was evaluated by changes in contractions to endothelin-1 (ET-1) and improved neurological assessment scores (as an early indicator of potential DCI) following experimental SAH in rat.

Methods and materials

Husbandry, housing and ethics

One hundred and ten Sprague-Dawley rats (NTac:SD), obtained from Taconic (Denmark), were maintained at a 12/12-h light-dark cycle (with light beginning at 7 a.m.) and housed at a constant temperature (22 ± 2°C) and humidity (55 ± 10%), with food and water ad libitum. Rats were generally housed in Eurostandard cages (Type VI with 123-Lid) 2–6 together and single housed (Type III with 123-Lid) after the surgical procedure. Fifty-two male Sprague-Dawley rats (298–370 g) were used for surgical procedures and were approved by the Danish Animal Experimentation Inspectorate (license no. 2016-15-0201-00940). The animal work was performed at the Glostrup Research Institute, Rigshospitalet-Glostrup, Denmark.

Harvest and organ culture of cerebral arteries (ex vivo model)

Rats were sedated with O2/CO2 (30/70%) and killed by decapitation. Brains were gently removed and chilled in a cold oxygenated buffer solution of the following composition: 119 mM NaCl, 4.6 mM KCl, 1.5 mM CaCl2, 1.2 mM MgCl2, 1.2 mM NaH2PO4, 15 mM NaHCO3 and 5.5 mM Glucose; pH 7.4. The basilar artery (BA) was carefully dissected from the brain in a physiological buffer solution, followed by either OC (naïve animals) or directly mounted in a wire myograph (arteries from rats which had undergone surgical procedures). The OC method is described elsewhere [17]. Segments were incubated for 48 h at 37°C in a humidified incubator with 5% CO2 in DMEM, supplemented with streptomycin and penicillin and with inhibitors or vehicle (DMSO) added. Culture media was changed after 24 h.

Rat subarachnoid haemorrhage model (in vivo model)

The experimental SAH procedure was induced as previously described in detail [18,19]. Sham-operated rats went through the same procedure, except for the intracisternal blood injection of 300 µl. At the end of the procedure, a PinPort (PNP3F22, Instech, US) was placed at the end of the ICP catheter to provide access for intrathecal (i.t.) injectable treatments by a PinPort injector (PNP-3M, Instech, US). 24 h post-surgery, the rats received subcutaneous (s.c.) injections of Carprofen (Norodyl, 5 mg/kg) (Scanvet, Denmark) for analgesia.

In vivo treatment regimens

The concentrations and doses of trametinib and PD0325901 were based on the ex vivo data from the present study and data from previous in vivo and ex vivo studies investigating the MEK1/2 inhibitor U0126 [15,16,20]. All treatments were blinded throughout the study and all treatment regimens can be found in Table 1.

Table 1
Regimens for intrathecal and intraperitoneal treatment.
GroupsConcentrationDrug administration (time points)Dose volume (µl/kg)Single dose (µg/kg)Total dose (µg/kg)
Intrathecal treatment 
Sham + vehicle i.t. 0.5% Cremophor® EL in Elliotts B (artificial cerebrospinal fluid) i.t. (4, 10 and 24 h) 50 – – 
SAH – – – – – 
SAH + vehicle i.t. 0.5% Cremophor® EL in Elliotts B (artificial cerebrospinal fluid) i.t. (4, 10 and 24 h) 50 – – 
SAH + trametenib i.t. 1 µM in 0.5% Cremophor® EL i.t. (4, 10 and 24 h) 50 0.031 0.092 
SAH + PD0325901 i.t. 1 µM in 0.5% Cremophor® EL i.t. (4, 10 and 24 h) 50 0.024 0.072 
Intraperitoneal treatment 
SAH + vehicle i.p. 10% Cremophor® EL and 10% PEG400 in NaCl i.p. (6 and 24 h) 750 – – 
SAH + trametinib i.p. 1 mM in 10% Cremophor® EL and 10% PEG400 in NaCl i.p. (6 and 24 h) 750 285 570 
GroupsConcentrationDrug administration (time points)Dose volume (µl/kg)Single dose (µg/kg)Total dose (µg/kg)
Intrathecal treatment 
Sham + vehicle i.t. 0.5% Cremophor® EL in Elliotts B (artificial cerebrospinal fluid) i.t. (4, 10 and 24 h) 50 – – 
SAH – – – – – 
SAH + vehicle i.t. 0.5% Cremophor® EL in Elliotts B (artificial cerebrospinal fluid) i.t. (4, 10 and 24 h) 50 – – 
SAH + trametenib i.t. 1 µM in 0.5% Cremophor® EL i.t. (4, 10 and 24 h) 50 0.031 0.092 
SAH + PD0325901 i.t. 1 µM in 0.5% Cremophor® EL i.t. (4, 10 and 24 h) 50 0.024 0.072 
Intraperitoneal treatment 
SAH + vehicle i.p. 10% Cremophor® EL and 10% PEG400 in NaCl i.p. (6 and 24 h) 750 – – 
SAH + trametinib i.p. 1 mM in 10% Cremophor® EL and 10% PEG400 in NaCl i.p. (6 and 24 h) 750 285 570 

Intrathecal treatment

The intrathecal (i.t.) treatment volumes were estimated for a cerebrospinal fluid (CSF) volume of approximately 90 µl [21] and the total dose was administered as three treatments (4, 10 and 24 h) through the PinPort in the ICP catheter, which was placed in the cisterna magna during the surgical procedure. The first and third injections were given under fixation of the rat. Since the treatment 10 h post-surgery was administered by a single researcher, the rats were briefly anaesthetised with isoflurane 3.5–4% (maintained at 1.75–2%) in atmospheric air/O2 (70%/30%) using a facemask, to prevent sudden movements of the rat. Animals were given 2.5 ml isotonic saline s.c. to avoid dehydration, immediately after surgery and in conjunction with the 10 and 24 h treatments. All in vivo treatments were dissolved in 0.5% cremophor EL (Kolliphor EL) in Elliott’s B (artificial CSF): NaCl 125 mM, NaHCO3 23 mM, Dextrose 4 mM, MgSO4 1 mM, KCl 4 mM, CaCl21 mM and Na2HPO4 1 mM.

Surgical parameters – i.t. groups

In all forty-one rats, mean arterial blood pressure (MABP), pH, pCO2, pO2, ICP (138.4 ± 6.9 mmHg) and temperature were within acceptable physiologic limits during surgery (Supplementary Table S1). There was one example of post-surgery mortality at 24 h after the SAH in the SAH + vehicle group.

Intraperitoneal treatment

For the intraperitoneal (i.p.) treatment the compound was dissolved in 10% cremophor EL (Kolliphor EL) and 10% PEG400 in NaCl, which also served as vehicle. The total dose was administered as two treatments (6 and 24 h). Animals were given 2.5 ml isotonic saline s.c. to avoid dehydration, immediately after surgery and in conjunction with the treatments.

Surgical parameters – i.p. groups

In all eleven rats, MABP, pH, pCO2, pO2, ICP (124.2 ± 6.3 mmHg) and temperature were within acceptable physiologic limits during surgery (Supplementary Table S1).

Neurological assessment – rotating pole test

Gross sensorimotor function was evaluated using a rotating pole test [22] as previously described [23], except for an additional scoring day before the SAH surgery (pre-SAH). Briefly, movement across a 10-rpm rotating pole (45 mm diameter, 150 cm length) was evaluated, with the rats receiving motivation in the form of a cage at the end, which contained bedding material from the home-cage (‘smells like home’). Rat performance was scored according to the following definitions: Low, the animal is unable to cross the pole without falling off; High, the animal can traverse the entire pole without falling off. All animals were trained to traverse the pole before surgery. Pre-SAH and on day 1 and 2 after surgery, each animal was scored twice for left and right rotation, respectively, i.e. 4 counts per animal. Animals were graded by personnel blinded to the experimental groups of the animals. Data are shown as % (high score count/total score count).

Myograph – ex vivo pharmacology (OC and in vivo)

For contractility measurements, both incubated BAs (ex vivo) and BAs from rats having undergone the surgical procedure (in vivo) were cut into segments and mounted on a pair of steel wires (40 μm) in a myograph bath. One wire was attached to a micrometer screw which allows for fine adjustments of the distance between the wires, controlling the vascular tone. The second wire was connected to a force displacement transducer, paired together with an analogue–digital converter (AD Instruments, Oxford, U.K.).

The segments were equilibrated in physiological buffer aerated with 95% O2/5% CO2, pH of 7.4, with the temperature set at 37°C and the wires separated for isometric pretension at 2 N m−1. The arterial segments were exposed two or three times with 60 mM K+, by exchanging the buffer. To maintain equal osmolarity, a proportional amount of Na+ had been removed from this buffer. An absolute cut-off was set at 2.0 mN K+max for inclusion of arterial segments from rats that underwent the surgical procedure. Endothelium function was evaluated with the addition of 5-HT (3 × 10−7 M) followed by carbachol (10−5 M). For arteries from the OC the protocol was as follows: (i) a cumulative concentration–response curve to S6c (10−14 to 10−7 M), (ii) a cumulative concentration–response curve to ET-1 (10−14 to 10−7 M), (iii) at peak ET-1 (10−7 M), the buffer was changed to a Ca2+-free buffer containing 10−7 M ET-1 and nimodipine (L-type voltage-dependent Ca2+ channel entry blocker, 10−7 M), (iv) a cumulative concentration–response curve to Ca2+ (0.0125–3 mM). When vasodilation was investigated, arteries were precontracted with U46619 (1–3 × 10−7 M) or K+ (41 mM) and cumulative concentration–response curves were performed by adding calcitonin gene-related peptide (CGRP, 10−12 to 10−7 M), carbachol (10−10 to 10−5 M) or SNP (10−11 to 10−4 M).

Two arterial segments from each operated animal were selected for either a cumulative concentration–response curve to ET-1 (10−14 to 10−7 M) or an ET-1 precontracted (10−7 M) Ca2+ concentration–response curve (0.0125–3 mM) in the presence of nimodipine (10−7 M). If there was only one curve above the cut-off, the concentration–response curve to ET-1 was prioritized. Segments with the highest K+ response were generally allocated to the concentration–response curve to ET-1. Concentration–response curves to Ca2+ were performed by adding increasing volumes of CaCl2, from a 125 mM stock solution, to a Ca2+-free buffer solution. The Ca2+-free buffer solution had a similar composition as above, but 1.5 mM CaCl2 was exchanged with 0.03 mM EDTA.

Statistics and data acquisition

The OC data for 1 µM U0126 and the DMSO vehicle control are also included in a separate study by our group [24]. Contractile responses of each segment were adjusted according to the length of the artery and are expressed as mN/mm (N m−1). If the responses to 60 mM K+ were not significantly different, the contractility data were shown as the percentage contraction of the individual 60 mM K+max plateau contraction. To compare ET-1 sensitivity, arteries were normalized to the percentage of the individual ET-1max. When the artery reached maximal contraction before the last concentration was added, the curves were constrained to the max contraction. The relative log IC50/log EC50 is the concentration corresponding to a response midway between the estimates of the lower and upper plateaus. Emax is the maximal contraction in the concentration response curve and ET-1max is the maximal contraction to ET-1. All quantitative data are presented as mean ± standard error of the mean (SEM), unless otherwise stated.

K+ and endothelium-dependent responses were statistically compared by one-way ANOVA followed by Holm-Sidak’s multiple comparison test (all groups compared with each other). Concentration–response curves were statistically compared by a two-way repeated measures ANOVA with Holm-Sidak’s multiple comparison test and Geisser–Greenhouse correction for sphericity was used for the normalised ET-1max data. A competitive curve fit was performed by comparing a biphasic vs. a non-linear regression curve fit, log (agonist) – variable slope. For all the ET-1 curves the biphasic regression model was accepted as the best curve fit. Significance of neurological assessment scores was evaluated with a two-tailed Fischer’s exact test. Statistical analyses were done using Graphpad 8.02 software and significant P-values were defined as: *P<0.05, **P<0.01, ***P<0.001.

Reagents

S6c was from PolyPeptide Group (Sweden), ET-1 was from Bachem (Germany) and CGRP from Tocris (U.K.). All MEK1/2 inhibitors except for U0126 were obtained from Selleckchem and dissolved in DMSO. U0126 monoethanolate (U120), DMSO (Sigma D2650) and all other chemicals were obtained from Sigma–Aldrich.

Results

Organ culture

OC was performed on rat BAs isolated from 58 rats (Sprague-Dawley male rats, ∼320 g) and each BA was divided into four segments. We initially tested a panel of MEK1/2 inhibitors at a concentration of 1 µM, which is just below the established threshold for U0126 efficacy (10 µM) [15,16]. Figure 1A shows the contraction induced by the highly specific ETB agonist S6c, relative to the contraction induced by 60 mM K+. Inhibitors were divided into four groups following the initial screening. The groups were as follows: (I) Not effective at 1 µM (black): Vehicle (DMSO) and U0126. (II) Some effect (blue): Binimetinib, selumetinib and RO5126766. (III) Borderline effective (dark green): Refametinib. (IV) Highly effective (light green and red): Cobimetinib, TAK-733, trametinib and PD0325901. The latter group was further divided in two groups (light green and red) based on the cell free IC50 (Supplementary Table S3), as the two subgroups could not be distinguished from each other at 1 µM.

Comparison of inhibitory capacity of nine different MEK1/2 inhibitors (1 µM), after 48 h organ culture of the basilar artery

Figure 1
Comparison of inhibitory capacity of nine different MEK1/2 inhibitors (1 µM), after 48 h organ culture of the basilar artery

(A) Concentration–response curves to the ETB specific agonist S6c, following incubation with 1 µM of different MEK1/2 inhibitors. (B) Maximal contraction to 60 mM K+ following incubation with 1 µM of different MEK1/2 inhibitors. There is no significant difference between the vehicle (DMSO) and any of the antagonists. (C) Endothelium-induced dilation with significant increase in segments incubated TAK-733, trametinib and PD0325901 compared with vehicle (DMSO). Data are shown as mean ± SEM with statistics by two-way ANOVA followed by a Holm-Sidak’s multiple comparison test. *P<0.05, **P<0.01 compared with DMSO. n = 4–15. The data for U0126 and DMSO have also been used in [24].

Figure 1
Comparison of inhibitory capacity of nine different MEK1/2 inhibitors (1 µM), after 48 h organ culture of the basilar artery

(A) Concentration–response curves to the ETB specific agonist S6c, following incubation with 1 µM of different MEK1/2 inhibitors. (B) Maximal contraction to 60 mM K+ following incubation with 1 µM of different MEK1/2 inhibitors. There is no significant difference between the vehicle (DMSO) and any of the antagonists. (C) Endothelium-induced dilation with significant increase in segments incubated TAK-733, trametinib and PD0325901 compared with vehicle (DMSO). Data are shown as mean ± SEM with statistics by two-way ANOVA followed by a Holm-Sidak’s multiple comparison test. *P<0.05, **P<0.01 compared with DMSO. n = 4–15. The data for U0126 and DMSO have also been used in [24].

There were no significant differences in the depolarization-induced contraction between arterial segments incubated with DMSO (vehicle) or the MEK1/2 inhibitors (Figure 1B). We also investigated the endothelial function of the arteries. This was tested by applying 10−5 M carbachol to arteries precontracted with 3 × 10−7 M of 5-HT. OC with DMSO or U0126 were both associated with poor endothelial response to carbachol (Figure 1C). Interestingly, there were significant differences between some of the groups, which appear to follow the same tendency as observed for the inhibition of S6c-induced contraction (Figure 1A). Worth noting, is that the increase in endothelium is not linked to the reduction in S6c-induced contractile response (Supplementary Figure S1). Three of the MEK1/2 inhibitors, TAK-733 (P = 0.0065), trametinib (P = 0.0026) and PD0325901 (P = 0.0474) had significantly better endothelium function compared with the DMSO vehicle.

Concentration–response curves to a selection of MEK1/2 inhibitors

We further determined the IC50 values for the six selected candidates (based on the Emax for S6c) using whole log concentrations ranging from no inhibition till near maximal inhibition of S6c-induced contraction (Figure 2A). Binimetinib (pIC50 5.17 ± 0.46) and RO5126766 (pIC50 5.68 ± 0.26) were the least potent inhibitors as expected from the data in Figure 1. When analysing the concentration–response curves to the inhibitors, it is evident that TAK-733 (pIC50 6.89 ± 0.31) and cobimetinib (pIC50 7.25 ± 0.51) are less potent than trametinib (pIC50 7.68 ± 0.32) and PD0325901 (pIC50 7.71 ± 0.29) (Figure 2A), which was not detectable in the data in Figure 1.

Concentration–response curves of selected highly potent MEK1/2 inhibitors
Figure 2
Concentration–response curves of selected highly potent MEK1/2 inhibitors

Concentration–response curves of selected inhibitors based on the (A) Emax of S6c, (B) maximal contraction to 60 mM K+ and (C) endothelium function evaluated by the addition of 10−5 M carbachol on arteries precontracted with 3 × 10−7 M 5 HT. Data are shown as mean ± SEM, n = 4.

Figure 2
Concentration–response curves of selected highly potent MEK1/2 inhibitors

Concentration–response curves of selected inhibitors based on the (A) Emax of S6c, (B) maximal contraction to 60 mM K+ and (C) endothelium function evaluated by the addition of 10−5 M carbachol on arteries precontracted with 3 × 10−7 M 5 HT. Data are shown as mean ± SEM, n = 4.

We further investigated if there were any effects on depolarization (60 mM K+) induced contractility of the arteries (Figure 2B). There were no significant differences between the groups and no obvious concentration-dependent effect was observed. The initial screen (Figure 1C) showed interesting effects of the MEK1/2 inhibitors on the endothelium function. Figure 2C shows a concentration-dependent effect of the MEK1/2 inhibitors on the endothelium function in response to carbachol, with a similar trend for the inhibitors observed in the concentration–response curves for the Emax of S6c (Figure 2A).

Effect of potent MEK1/2 inhibitors on pathways regulating vasomotion after 48 h organ culture of basilar artery

Since the arteries incubated with the most potent MEK1/2 inhibitors indicated preserved endothelium function, we further investigated the possible cause of this improvement. Arteries after OC with 1 µM of trametinib or PD0325901 were precontracted with the thromboxane A2 agonist, U46619 (1–3 × 10−7 M) or K+ (41 mM). There were no significant differences in the level of precontraction between the groups (data not shown). In this new set of experiments, we confirmed the improved vasodilation in response to carbachol (10−10 to 10−5 M, Figure 3A). This effect of carbachol could either be caused by improved endothelial NO release or changes in vascular smooth muscle cell (VSMC) NO sensitivity. As shown in Figure 3B there were significant differences in the Emax following the addition of the NO donor sodium nitroprusside (SNP, 10−11 to 10−4 M). This indicates changes in the cGMP and NO sensitive signalling in the VSMCs, as the cause of the increased apparent endothelium function following incubation with MEK1/2 inhibitors. Since both cGMP and cAMP are important for regulating vasomotion in cerebral arteries, we also investigated a signalling pathway that is linked to the generation of cAMP, but we did not observe any differences in the response to CGRP (10−12 to 10−7 M, Figure 3C).

Effect of trametinib and PD0325901 on pathways regulating vasomotion after 48 h organ culture of the basilar artery
Figure 3
Effect of trametinib and PD0325901 on pathways regulating vasomotion after 48 h organ culture of the basilar artery

Arteries from OC in the presence of DMSO control (n = 5), 1 µM of trametinib (n = 6) or PD0325901 (n = 5) were precontracted with U46619 (1–3×10−7 M) or K+ (41 mM) and cumulative concentration–response curves was performed by adding (A) carbachol (10−10 to 10−5 M), (B) SNP (10−11 to 10−4 M) or (C) CGRP (10−12 to 10−7 M). (D) Cumulative concentration–response curves with Ca2+ (0.0125–3 mM) after an ET-1 precontraction (10−7 M) following OC with DMSO control (n = 15), 1 µM of trametinib (n = 4) or PD0325901 (n = 4). Data are shown as mean ± SEM with statistics by two-way ANOVA, followed by Holm-Sidak’s multiple comparison test */# P<0.05, **/## P<0.01, ***/### P<0.001. For normalised data the Geisser–Greenhouse correction for sphericity was applied.

Figure 3
Effect of trametinib and PD0325901 on pathways regulating vasomotion after 48 h organ culture of the basilar artery

Arteries from OC in the presence of DMSO control (n = 5), 1 µM of trametinib (n = 6) or PD0325901 (n = 5) were precontracted with U46619 (1–3×10−7 M) or K+ (41 mM) and cumulative concentration–response curves was performed by adding (A) carbachol (10−10 to 10−5 M), (B) SNP (10−11 to 10−4 M) or (C) CGRP (10−12 to 10−7 M). (D) Cumulative concentration–response curves with Ca2+ (0.0125–3 mM) after an ET-1 precontraction (10−7 M) following OC with DMSO control (n = 15), 1 µM of trametinib (n = 4) or PD0325901 (n = 4). Data are shown as mean ± SEM with statistics by two-way ANOVA, followed by Holm-Sidak’s multiple comparison test */# P<0.05, **/## P<0.01, ***/### P<0.001. For normalised data the Geisser–Greenhouse correction for sphericity was applied.

It was recently shown that the cerebral arteries had an VDCC- (voltage-dependent calcium channel) independent contraction after cerebral ischemia [25] and SAH [23], which is not present in the fresh arteries. To further characterize changes in the cellular pathways leading to enhanced contraction, we performed a concentration–response curve to Ca2+, by adding Ca2+ to BAs precontracted with ET-1 in Ca2+ free buffer containing 10−7 M nimodipine (L-type VDCC inhibitor). Figure 3D shows the VDCC-independent contraction of the BAs, where only trametinib (P = 0.019) significantly prevented this increased VDCC-independent contraction.

Experimental SAH

We investigated the two most potent MEK1/2 inhibitors identified in the OC studies further in a rat model of SAH. Trametinib and PD0325901 had the highest potency, meaning that they potentially can be applied in smaller volumes than the current drug of choice, U0126. Following a 1 µM, 15 µl i.t. injection into CSF volume of the rat (assumed to be 90 µl), we predicted a CSF concentration of approximately 10−7 M after dilution. This corresponds to an estimated 75% inhibitor effect in our ex vivo OC study (Figure 2A).

Effect of i.t. trametinib and PD0325901 treatment on contractile responses to 60 mM K+ and on the endothelium function

The individual segment lengths (range 0.9–1.2 mm; 1.0 ± 0.1 mm) were not significantly different between the groups. The contractile responses (N m−1) of all BAs to 60 mM K+ (including vessels below the cut-off) showed a significantly higher contractile response in the SAH + i.t. trametinib group (5.00 ± 0.29 N m−1) compared with both the sham + i.t. vehicle group (3.30 ± 0.45 N m−1) and SAH + i.t. vehicle group (3.45 ± 0.22 N m−1) (Figure 4A). The SAH group had a slightly lower mean of endothelium function compared with the other groups, but it was not significantly different (SAH vs. i.t. trametinib, P = 0.1382) (Figure 4B).

Effect of trametinib and PD0325901 treatment on contractile responses to 60 mM K+ and on the endothelium function after SAH and sham surgery in rat
Figure 4
Effect of trametinib and PD0325901 treatment on contractile responses to 60 mM K+ and on the endothelium function after SAH and sham surgery in rat

(A) Emax,K+ (N m−1) induced by 60 mM K+. (B) Endothelium function. Data include basilar arteries below 2.0 mN cut-off (orange dots) for comparison of all arteries; sham + vehicle (n = 10), SAH (n = 11), SAH + vehicle (n = 27), trametinib (n = 13) and PD0325901 (n = 12). Horizontal dotted black line shows Totalmean of all n (n = 73). Data are shown as mean ± SEM, in this figure ‘n’ equals individual arterial segments. Statistics was done by one-way ANOVA followed by Holm-Sidak’s multiple comparison test. *P<0.05, **P<0.01.

Figure 4
Effect of trametinib and PD0325901 treatment on contractile responses to 60 mM K+ and on the endothelium function after SAH and sham surgery in rat

(A) Emax,K+ (N m−1) induced by 60 mM K+. (B) Endothelium function. Data include basilar arteries below 2.0 mN cut-off (orange dots) for comparison of all arteries; sham + vehicle (n = 10), SAH (n = 11), SAH + vehicle (n = 27), trametinib (n = 13) and PD0325901 (n = 12). Horizontal dotted black line shows Totalmean of all n (n = 73). Data are shown as mean ± SEM, in this figure ‘n’ equals individual arterial segments. Statistics was done by one-way ANOVA followed by Holm-Sidak’s multiple comparison test. *P<0.05, **P<0.01.

Effect of i.t. trametinib and PD0325901 treatment on contractile responses to ET-1

Since the K+ responses were different, we initially used non-normalized data. BAs from all treatment regimens (Table 1) were compared by cumulative concentration–response curves to ET-1 (10−14 to 10−7 M). No significant differences between curves were observed (Figure 5A). We also investigated the ET-1 sensitivity of the arteries by normalising the curves to their own ET-1max. For the ET-1max normalised data, the contraction at low ET-1 concentrations (10−12.5 to 10−11.0 M) was significantly reduced in the SAH + i.t. trametinib group compared with the SAH + vehicle group (Figure 5B). A competitive curve fit was performed by comparing a biphasic vs. a four-parameter variable slope regression. For all groups the biphasic regression model was accepted as the best curve fit (Table 2). The log EC50 (1) 95% confidence intervals (CI) for the SAH group (−12.74 to −12.17) had significantly higher sensitivity compared with the sham + i.t. vehicle group (−11.91 to −10.07) and SAH + i.t. trametinib group (−11.04 to −9.936). At log EC50 (2), the SAH + i.t. trametinib group (−8.964 to −8.862) was significantly less sensitive than the SAH + i.t. vehicle group (−9.228 to −9.050). The log EC50 (1) and log EC50 (2) values were similar for both the absolute curves (N m−1) and the ET-1max normalised curves (Figure 5A,B). All the log EC50 (1) values, log EC50 (2) values and 95% CI values can be found in Table 2.

Effect of trametinib and PD0325901 treatment on contractile responses to endothelin-1 after SAH or sham surgery in rat
Figure 5
Effect of trametinib and PD0325901 treatment on contractile responses to endothelin-1 after SAH or sham surgery in rat

Cumulative concentration–response curves to ET-1 (10−14 to 10−7 M) of basilar arteries. (A) ET-1 (N m−1) and (B) ET-1 (% of ET-1max), for sham + vehicle (n = 5), SAH (n = 5), SAH + vehicle (n = 12), SAH + trametinib (n = 6) and SAH + PD0325901 (n = 6). Data are shown as mean ± SEM, with statistics by two-way ANOVA followed by Holm-Sidak’s multiple comparison test. *P≤0.05. ET-1max normalised data applied Geisser–Greenhouse correction for sphericity and ET-1 curves are a biphasic non-linear regression curve fit.

Figure 5
Effect of trametinib and PD0325901 treatment on contractile responses to endothelin-1 after SAH or sham surgery in rat

Cumulative concentration–response curves to ET-1 (10−14 to 10−7 M) of basilar arteries. (A) ET-1 (N m−1) and (B) ET-1 (% of ET-1max), for sham + vehicle (n = 5), SAH (n = 5), SAH + vehicle (n = 12), SAH + trametinib (n = 6) and SAH + PD0325901 (n = 6). Data are shown as mean ± SEM, with statistics by two-way ANOVA followed by Holm-Sidak’s multiple comparison test. *P≤0.05. ET-1max normalised data applied Geisser–Greenhouse correction for sphericity and ET-1 curves are a biphasic non-linear regression curve fit.

Table 2
Curve fits and comparison of EC50 values
% of ET-1maxCurve fit competition
GroupsBiphasic versus log(agonist) vs. response – variable slope
Biphasic P-valuesMean log EC50(1); 95% confidence interval (CI) (asymptotic)log EC50(1) CI assessmentMean log EC50(2); 95% confidence interval (CI) (asymptotic)log EC50(2) CI assessment
Intrathecal treatment 
Sham + i.t. vehicle 0.0002 −10.99; −11.91 to −10.07 Sham + vehicle < SAH −8.978; −9.128 to −8.827  
SAH 0.0002 −12.46: −12.74 to −12.17 SAH > Sham + vehicle, SAH + trametinib −8.868; −9.002 to −8.734 SAH < SAH + vehicle 
SAH + i.t. vehicle <0.0001 −11.58; −12.19 to −10.97  −9.139; −9.228 to −9.050 SAH + vehicle > SAH + trametinib, SAH 
SAH + i.t. trametinib <0.0001 −10.49; −11.04 to −9.936 SAH + trametinib < SAH −8.913; −8.964 to −8.862 SAH + trametinib < SAH + vehicle 
SAH + i.t. PD0325901 0.004 −11.19; −13.69 to −8.682  −8.993; −9.217 to −8.769  
Intraperitoneal treatment 
SAH + i.p. vehicle 0.0023 −11.92; −12.84 to −10.36  −9.56; −9.64 to −9.43  
SAH + i.p. trametinib 0.003 9.91; −10.91 −to 9.45  −8.99; −9.10 to −8.90  
% of ET-1maxCurve fit competition
GroupsBiphasic versus log(agonist) vs. response – variable slope
Biphasic P-valuesMean log EC50(1); 95% confidence interval (CI) (asymptotic)log EC50(1) CI assessmentMean log EC50(2); 95% confidence interval (CI) (asymptotic)log EC50(2) CI assessment
Intrathecal treatment 
Sham + i.t. vehicle 0.0002 −10.99; −11.91 to −10.07 Sham + vehicle < SAH −8.978; −9.128 to −8.827  
SAH 0.0002 −12.46: −12.74 to −12.17 SAH > Sham + vehicle, SAH + trametinib −8.868; −9.002 to −8.734 SAH < SAH + vehicle 
SAH + i.t. vehicle <0.0001 −11.58; −12.19 to −10.97  −9.139; −9.228 to −9.050 SAH + vehicle > SAH + trametinib, SAH 
SAH + i.t. trametinib <0.0001 −10.49; −11.04 to −9.936 SAH + trametinib < SAH −8.913; −8.964 to −8.862 SAH + trametinib < SAH + vehicle 
SAH + i.t. PD0325901 0.004 −11.19; −13.69 to −8.682  −8.993; −9.217 to −8.769  
Intraperitoneal treatment 
SAH + i.p. vehicle 0.0023 −11.92; −12.84 to −10.36  −9.56; −9.64 to −9.43  
SAH + i.p. trametinib 0.003 9.91; −10.91 −to 9.45  −8.99; −9.10 to −8.90  

Effect of i.t. trametinib and PD0325901 treatment on VDCC-independent contractions

To investigate if the in vivo treatment affected the VDCC-independent contractility, as previously seen after U0126 treatment [23], we performed cumulative concentration–response curves to Ca2+ (0.0125–3 mM) in ET-1 (10−7 M) pre-contracted BAs and in the presence of 10−7 M nimodipine. The SAH group (2.8 ± 0.5 N m−1) had significantly higher nimodipine insensitive ET-1max contractions compared with the sham + i.t. vehicle group (0.6 ± 1.2 N m−1), SAH + i.t. vehicle group (1.2 ± 0.2 N m−1) and SAH + i.t. PD0325901 group (1.0 ± 0.3 N m−1), while there was no significant difference between the SAH + i.t. trametinib group (1.8 ± 0.5 N m−1) and the other groups (Figure 6).

Effect of trametinib and PD0325901 treatment on VDCC-independent ET-1-induced contractions after SAH or sham surgery in rat
Figure 6
Effect of trametinib and PD0325901 treatment on VDCC-independent ET-1-induced contractions after SAH or sham surgery in rat

Basilar arteries were precontracted with ET-1 (10–7 M) followed by cumulative concentration–response curves with Ca2+ (N m−1); sham + vehicle (n = 5), SAH (n = 5), SAH + vehicle (n = 13), SAH + trametinib (n = 7) and SAH + PD0325901 (n = 6). Data are shown as mean ± SEM, with statistics by two-way ANOVA followed by Holm-Sidak’s multiple comparison test. */#/$ P≤0.05.

Figure 6
Effect of trametinib and PD0325901 treatment on VDCC-independent ET-1-induced contractions after SAH or sham surgery in rat

Basilar arteries were precontracted with ET-1 (10–7 M) followed by cumulative concentration–response curves with Ca2+ (N m−1); sham + vehicle (n = 5), SAH (n = 5), SAH + vehicle (n = 13), SAH + trametinib (n = 7) and SAH + PD0325901 (n = 6). Data are shown as mean ± SEM, with statistics by two-way ANOVA followed by Holm-Sidak’s multiple comparison test. */#/$ P≤0.05.

Neurological assessment of i.t. trametinib and PD0325901 treatment effects

Our rotating pole test is not a pure motor function test, as it does require training of the rats in advance. In addition to the learning aspect, the fact that the pole is rotating also leads to motivation being a factor of success. Therefore memory, motivation and attention are involved in a successful score [26]. The rats were scored at three time points: Pre-SAH, 24 h and 48 h post-SAH. (Data are shown as % (high score count/total score count).)

All rats scored 100% in the rotating pole test prior to surgery (Figure 7A). Neurological score deficits were seen for all groups when comparing 24 h post-SAH with pre-SAH, except for the sham + i.t. vehicle group (Figure 7A–C, Supplementary Table S2). Rats demonstrated significantly worsened neurological score 48 h after experimental SAH, when compared with pre-SAH (Scorepre 100% to Score48h 75%, P = 0.0022). At the 48 h endpoint the rats in the SAH + i.t. trametinib (Score48h 96%), sham + i.t. vehicle (Score48h 100%) and SAH + i.t. vehicle (Score48h 94%) groups had a significantly higher neurological score than the rats in the SAH group (Score48h 75%) (Figure 7C). See Supplementary Table S2 for all score percentages.

Effect of trametinib and PD0325901 treatment on neurologic function
Figure 7
Effect of trametinib and PD0325901 treatment on neurologic function

Rotating pole test at 10 rpm for (A) pre-SAH, (B) 24 h post-SAH and (C) 48 h post-SAH. Scored with four counts per animal, i.e. two scores for left- and right rotation, respectively; Low = Unable to traverse in one try; High = Able to traverse in one try. Sham + vehicle (n = 5), SAH (n = 9), SAH + vehicle (n = 14), SAH + trametinib (n = 7) and SAH + PD0325901 (n = 6). All animals, excluding sham + vehicle group, were exposed to experimental SAH. Statistics were done by a Fischer’s exact test, two-sided, 95% CI. *P<0.05; **P<0.01; ***P<0.001.

Figure 7
Effect of trametinib and PD0325901 treatment on neurologic function

Rotating pole test at 10 rpm for (A) pre-SAH, (B) 24 h post-SAH and (C) 48 h post-SAH. Scored with four counts per animal, i.e. two scores for left- and right rotation, respectively; Low = Unable to traverse in one try; High = Able to traverse in one try. Sham + vehicle (n = 5), SAH (n = 9), SAH + vehicle (n = 14), SAH + trametinib (n = 7) and SAH + PD0325901 (n = 6). All animals, excluding sham + vehicle group, were exposed to experimental SAH. Statistics were done by a Fischer’s exact test, two-sided, 95% CI. *P<0.05; **P<0.01; ***P<0.001.

Effect of i.p. trametinib treatment on contractile responses to 60 mM K+, ET-1 and neurological assessment

Proceeding from the i.t. proof of concept study, we opted to test trametinib using an i.p. injection treatment protocol. Animals were exposed to SAH and treated with i.p. injections of trametinib or vehicle at 6 and 24 h post-SAH. 48 h after induced experimental SAH, arteries were isolated for the wire myograph. The individual segment lengths (range 0.9–1.2 mm; 1.13 ± 0.02 mm) were not significantly different between the SAH + i.p. vehicle or SAH + i.p. trametinib groups. The contractile responses (N m−1) of all BAs to 60 mM K+ did not show any differences when comparing the SAH + i.p. trametinib group (3.03 ± 0.19 N m−1) with the SAH + i.p. vehicle group (3.23 ± 0.27 N m−1) (Figure 8A). This contrasts with the SAH + i.t. treatment (Figure 4A).

Effect of i.p. trametinib treatment after SAH surgery in rat
Figure 8
Effect of i.p. trametinib treatment after SAH surgery in rat

(A) Emax, for K+ (Nm−1) induced by 60 mM K+ for SAH + i.p. vehicle (n = 10), i.p. trametinib (n = 12). Data are shown as mean ± SEM. In this figure ‘n’ equals individual arterial segments. (B) Cumulative concentration–response curves to ET-1 (10−14 to 10−7 M) of basilar arteries normalized to 60 mM K+ from animals treated with SAH + i.p. vehicle (n = 5) or SAH + i.p. trametinib (n = 6). Data are shown as mean ± SEM, with statistics by two-way ANOVA followed by Holm-Sidak’s multiple comparison test. ***P≤0.001. ET-1 curves are a biphasic non-linear regression curve fit. The right panel show the rotating pole tests at 10 rpm for neurological function at (C) 24 h post-SAH and (D) 48 h post-SAH. Scored with four counts per animal, i.e. two scores for left and right rotation, respectively; Low = Unable to traverse in one try; High = Able to traverse in one try. Statistics were done by a Fischer’s exact test, two-sided, 95% CI. *P<0.05.

Figure 8
Effect of i.p. trametinib treatment after SAH surgery in rat

(A) Emax, for K+ (Nm−1) induced by 60 mM K+ for SAH + i.p. vehicle (n = 10), i.p. trametinib (n = 12). Data are shown as mean ± SEM. In this figure ‘n’ equals individual arterial segments. (B) Cumulative concentration–response curves to ET-1 (10−14 to 10−7 M) of basilar arteries normalized to 60 mM K+ from animals treated with SAH + i.p. vehicle (n = 5) or SAH + i.p. trametinib (n = 6). Data are shown as mean ± SEM, with statistics by two-way ANOVA followed by Holm-Sidak’s multiple comparison test. ***P≤0.001. ET-1 curves are a biphasic non-linear regression curve fit. The right panel show the rotating pole tests at 10 rpm for neurological function at (C) 24 h post-SAH and (D) 48 h post-SAH. Scored with four counts per animal, i.e. two scores for left and right rotation, respectively; Low = Unable to traverse in one try; High = Able to traverse in one try. Statistics were done by a Fischer’s exact test, two-sided, 95% CI. *P<0.05.

The SAH + i.p. trametinib group and SAH + i.p. vehicle group were compared by cumulative concentration–response curves to ET-1 (10−14 to 10−7 M). There was a significant decrease in contractility (at 10−9.5 M) for the SAH + i.p. trametinib group compared with the SAH + i.p. vehicle group. A competitive curve fit was performed by comparing a biphasic vs. a four-parameter variable slope regression. For both groups, the biphasic regression model was accepted as the best curve fit. The log EC50 (1) values, log EC50 (2) values and 95% CI values can be found in Table 2. The animals had been tested for neurological deficits by the rotating pole test (Figure 8C,D), prior to the myograph study. The rats were scored at two time points: 24 and 48 h post-SAH. (Data are shown as % (high score count/total score count).) At the 48 h endpoint the rats in the SAH + i.p. trametinib group (Score48h 100%), scored significantly better (P = 0.0143) than the SAH + i.p. vehicle group (Score48h 71%). See Supplementary Table S2 for all score percentages.

Discussion

In this study, nine highly potent MEK1/2 inhibitors were screened in vitro and the two most potent MEK1/2 inhibitors (trametinib and PD0325901) were tested in an in vivo experimental rat SAH model. Both i.t. and i.p. trametinib treatment caused a significant reduction in the ET-1-induced contractility after SAH, which is believed to be associated with the ETB receptor up-regulation shown previously by molecular methods [15,16]. The effect of the i.t. trametinib treatment was limited, as it significantly increased the contractility to 60 mM K+, which was not observed with the i.p. trametinib treatment.

Ex vivo data from organ culture

Previous studies of cerebral arteries treated with 10 µM U0126 in 48 h OC showed an inhibitory effect on the ETB specific S6c-induced contractility [15,16,24]. In the current study, 1 µM U0126 showed no significant effect, whereas trametinib and PD0325901 at 1 µM almost completely inhibited the contractile response after 48 h OC (Figure 1A). The data from the OC experiments illustrate a strong connection between MEK1/2 inhibition potency and the ability to prevent the emergence of S6c-inducible contractions (associated with ETB receptors). The cell free IC50 values (S3 Table) also correlate well with the IC50 values for the inhibition of the ETB-dependent S6c-induced contraction (Figure 2A). This supports the link between MEK1/2 inhibition and functional receptor changes in the cerebral artery. We can therefore conclude that the emergence of ETB receptor specific S6c-inducible contractility ex vivo can be completely prevented by inhibiting a signalling pathway that includes MEK1/2.

Effect on K+ contractility and endothelium function

The use of novel MEK1/2 inhibitors with higher potency than U0126 allowed us for the first time to observe concentration-dependent preservation of apparent endothelium function (Figure 2C). A similar pattern was observed for both the endothelium function and for the S6c-induced contraction (Figure 2A). Therefore, the MEK1/2 pathway appears to be involved in the disruption of endothelial and VSMC signalling in response to the reduction in blood flow through the artery. This is supported by the current understanding in the field [27]. This contrasts with the neuronal vasodilation signalling, exemplified by CGRP, for which we did not observe any changes (Figure 3C). We did not observe significant changes in the endothelium function for the animals treated in vivo with neither trametinib nor PD0325901 due to relatively high variability, although both groups had higher mean values than the SAH and SAH + i.t. vehicle groups (Figure 4B).

In contrast with the effect of trametinib or PD0325901 on the S6c-induced contractility and apparent endothelium function, we did not see any concentration-dependent effect of the inhibitors on the K+ responses after OC (Figure 2B). However, rats treated with i.t. trametinib, but not i.t. PD0325901, after experimental SAH showed higher K+ responses compared with the treatment with i.t. vehicle (Figure 4A). Vessels incubated with i.t. trametinib or i.t. PD0325901 had K+ responses in the higher end of the compounds tested in the OC model (Figure 1B). We cannot exclude that our sample size in the initial screening was too small to detect changes in K+ contractile properties.

In vivo data from the rat SAH model

We did not observe increased neurological scores in rats subjected to SAH and treated with PD0325901, when compared with SAH rats (Figure 7). Furthermore, these rats had contractile responses to ET-1 that were comparable to those of the untreated and i.t. vehicle treated SAH groups (Figure 5). The only effect was a minor impact on the functional increase in nimodipine insensitive ET-1max contractions compared with the untreated SAH group (Figure 6). Therefore, PD0325901 treatment did not appear to be a promising candidate for improving neurological assessment scores after SAH. We postulate that this could be due to hydrolysation of PD0325901, as it contains a hydroxamate ester bond [28]. Esterases can be found in the arachnoid mater and are present in CSF [29,30] and the amount could be increased with the presence of blood. Another possibility is differences in the MEK1/2 signalling caused by trametinib and PD0325901. It was shown in an in vivo comparison between these two antagonists, that besides strongly reducing ERK1/2 phosphorylation (the target of MEK1/2), PD0325901 also increased the phosphorylation of STAT5a and STAT5b, whereas trametinib did not [31]. Interestingly, STAT5b has been shown to play a role in thrombin-induced VSMC growth and motility [32]. The implications of an increase in STAT5b signalling were made apparent in a different study, by applying the balloon injury model on carotid arteries of rats. It was shown that adenovirus-mediated expression of dnSTAT-5B (which blocks STAT5b signalling) attenuated balloon injury-induced media to intima VSMC migration and VSMC proliferation in the intima, resulting in reduced neointima formation [33]. Hence, the absent effect of PD0325901 could partially be caused by differences in intracellular signalling pathways.

Compared with data from the less specific and less potent MEK1/2 inhibitor U0126, the trametinib treatment showed similar neurological improvements and beneficial characteristics on its ability to modify the contractile responses of cerebral arteries to ET-1. Nevertheless, trametinib did not exhibit any ET-1max lowering effect as seen for U0126 in other studies [23]. This is most likely explained by the increase in the contractile response to 60 mM K+ in the trametinib treated rats (Figure 4A). We do not yet understand this increase in the contractile response and this deserves further investigations. In contrast with the i.t. trametinib treatment, the i.p. trametinib treatment had no effects on the 60 mM K+Emax. These data suggest that i.t. trametinib treatment could have some local effects on the arteries.

I.t. trametinib treatment only had a minor inhibitory effect on the VDCC-independent ET-1max contraction, but a stronger inhibitory effect was observed in the SAH + i.t. vehicle group. It is important to note that the difference in VDCC-independent ET-1max of the trametinib treated group is smaller when normalising the ET-1 response to the K+ response, as the group exhibited a higher response to K+ (Figure 4A).

Selectivity

It is evident that applying i.t. trametinib did not lead to the exact same arterial changes as observed for i.t. U0126. We did however observe improved neurological assessment scores (rotating pole test) with trametinib (and not with the pharmacokinetically unstable PD0325901), which further supports the involvement of the MEK1/2 pathway in the phenotypical modulation observed in VSMCs following SAH. It is worth noting, that the cremophor EL vehicle has also been shown not to be a fully inert vehicle and potentially has a minor PKC inhibitor function [34]. In addition, we have shown that cremophor EL has a positive effect on neurological assessment after SAH, possibly by affecting the emergence of the S6c-inducible contraction [23,24]. Though, this effect is not unique to cremophor EL as a vehicle, since DMSO has been shown to have a positive effect on stroke outcome [35,36]

What may differentiate trametinib from U0126? Trametinib is a highly selective MEK1/2 inhibitor with little or no effects on other targets [37]. On the contrary, U0126 has been reported to have other targets, such as ERK5 [38,39] and AMP-activated protein kinase [40], as well as a range of MEK/ERK-independent activities, including activation of hepatocytic cell lines [41], K+ channel inactivation in Chinese hamster ovary cells [42] and inhibition of combretastatin A4 inactivation in human hepatocellular carcinoma cell lines [43]. We hypothesize that some of these other targets could explain this observed difference and should be the focus of future studies.

Conclusion and clinical relevance

The i.t. PD0325901 treatment did not improve the neurological score deficit associated with SAH, nor show any beneficial therapeutic effect on the contractility when compared with the SAH and SAH + i.t. vehicle groups, in contrast with the reduction in neurological deficits seen after i.t. trametinib treatment. We went further and investigated trametinib as a systemic i.p. treatment. Here we observed a reduction in contractility and improved neurological scores. This treatment also minimised any local or combinational effects of the i.t. vehicle and i.t. trametinib treatments on the cerebral vessels, thereby avoiding any direct effects on general contractility. A future study investigating some of the unspecific inhibitory targets of U0126 as potential SAH treatment targets is warranted.

Clinical perspectives

  • Vascular contractility has been the focus of many clinical and pre-clinical studies designed to prevent DCI, including recent attempts to modulate acute vascular contractility. Our current working hypothesis is that the prevention of phenotypical changes and subsequent effects on vascular contractility could reduce the subsequent DCI. The MEK1/2 inhibitor U0126 has been shown to improve DCI and neurological assessment after SAH. Our current aim is to examine more potent MEK1/2 inhibitors.

  • Intrathecal treatment of trametinib improved the neurological score after SAH and showed beneficial therapeutic effect on the contractility compared with the SAH and SAH + intrathecal vehicle groups. In contrast, no effect was shown with PD0325901.

  • Intraperitoneal treatment with trametinib also improved the neurological score and reduced arterial contractility to ET-1. Intraperitoneal trametinib did not affect the potassium induced contractility, suggesting that systemic injections could be the optimal route of administration.

Funding

This work was supported by the Lundbeck foundation, Lundbeck Grant of excellence [grant number R59-A5404]; the Swedish Heart Lung Foundation [grant number 20130271]; and the KA Wallenberg Foundation [grant number KAW 2016.0081]. The funders had no role in study design, data collection and analysis, and decision to publish or preparation of the manuscript.

Competing interests

The authors declare that there are no competing interests associated with the manuscript.

Author contribution

S.T.C., K.A.H., K.W., L.E. and S.E.J. conceived and the study; S.T.C., K.A.H., S.S., A.S.G., L.E. and S.E.J. designed experiments; S.T.C., K.A.H., S.S., A.S.G. and S.E.J. performed experiments; L.E. provided tools and reagents; S.T.C., K.A.H. and S.E.J. analysed data; S.T.C. and K.A.H. wrote the manuscript; S.S., A.S.G., K.W., L.E. and S.E.J. made further critical manuscript revisions. All authors read and commented on the final manuscript.

Abbreviations

     
  • BA

    basilar artery

  •  
  • CSF

    cerebrospinal fluid

  •  
  • DCI

    delayed cerebral ischemia

  •  
  • ET-1

    endothelin-1

  •  
  • ETA and ETB receptor

    endothelin A and B receptor

  •  
  • ICP

    intracranial pressure

  •  
  • i.p.

    intraperitoneal

  •  
  • i.t.

    intrathecal

  •  
  • MABP

    mean arterial blood pressure

  •  
  • MEK1/2

    mitogen activated protein kinase kinase

  •  
  • OC

    organ culture

  •  
  • S6c

    sarafotoxin 6C

  •  
  • SAH

    subarachnoid haemorrhage

  •  
  • s.c.

    subcutaneous

  •  
  • VSMC

    vascular smooth muscle cell

References

References
1.
Macdonald
R.L.
(
2014
)
Delayed neurological deterioration after subarachnoid haemorrhage
.
Nat. Rev. Neurol.
10
,
44
58
[PubMed]
2.
Suarez
J.I.
,
Tarr
R.W.
and
Selman
W.R.
(
2006
)
Aneurysmal subarachnoid hemorrhage
.
N. Engl. J. Med.
354
,
387
396
[PubMed]
3.
van Gijn
J.
,
Kerr
R.S.
and
Rinkel
G.J.
(
2007
)
Subarachnoid haemorrhage
.
Lancet
369
,
306
318
[PubMed]
4.
Geraghty
J.R.
and
Testai
F.D.
(
2017
)
Delayed cerebral ischemia after subarachnoid hemorrhage: beyond vasospasm and towards a multifactorial pathophysiology
.
Curr. Atheroscler. Rep.
19
,
50
[PubMed]
5.
Lucke-Wold
B.P.
,
Logsdon
A.F.
,
Manoranjan
B.
,
Turner
R.C.
,
McConnell
E.
,
Vates
G.E.
et al.
(
2016
)
Aneurysmal subarachnoid hemorrhage and neuroinflammation: a comprehensive review
.
Int. J. Mol. Sci.
17
,
497
[PubMed]
6.
Stokum
J.A.
,
Gerzanich
V.
and
Simard
J.M.
(
2016
)
Molecular pathophysiology of cerebral edema
.
J. Cereb. Blood Flow Metab.
36
,
513
538
[PubMed]
7.
Findlay
J.M.
,
Nisar
J.
and
Darsaut
T.
(
2015
)
Cerebral vasospasm: a review
.
Can. J. Neurol. Sci.
1
18
[PubMed]
8.
Kirkpatrick
P.J.
,
Turner
C.L.
,
Smith
C.
,
Hutchinson
P.J.
and
Murray
G.D.
(
2014
)
Simvastatin in aneurysmal subarachnoid haemorrhage (STASH): a multicentre randomised phase 3 trial
.
Lancet Neurol.
13
,
666
675
[PubMed]
9.
Brathwaite
S.
and
Macdonald
R.L.
(
2014
)
Current management of delayed cerebral ischemia: update from results of recent clinical trials
.
Transl. Stroke Res.
5
,
207
226
[PubMed]
10.
Laban
K.G.
,
Vergouwen
M.D.
,
Dijkhuizen
R.M.
,
Sena
E.S.
,
Macleod
M.R.
,
Rinkel
G.J.
et al.
(
2015
)
Effect of endothelin receptor antagonists on clinically relevant outcomes after experimental subarachnoid hemorrhage: a systematic review and meta-analysis
.
J. Cereb. Blood Flow Metab.
35
,
1085
1089
[PubMed]
11.
Edvinsson
L.I.
and
Povlsen
G.K.
(
2011
)
Vascular plasticity in cerebrovascular disorders
.
J. Cereb. Blood Flow Metab.
31
,
1554
1571
[PubMed]
12.
Edvinsson
L.
,
Larsen
S.S.
,
Maddahi
A.
and
Nielsen
J.
(
2014
)
Plasticity of cerebrovascular smooth muscle cells after subarachnoid hemorrhage
.
Transl. Stroke Res.
5
,
365
376
[PubMed]
13.
Povlsen
G.K.
and
Edvinsson
L.
(
2015
)
MEK1/2 inhibitor U0126 but not endothelin receptor antagonist clazosentan reduces upregulation of cerebrovascular contractile receptors and delayed cerebral ischemia, and improves outcome after subarachnoid hemorrhage in rats
.
J. Cereb. Blood Flow Metab.
35
,
329
337
[PubMed]
14.
Davenport
A.P.
,
Hyndman
K.A.
,
Dhaun
N.
,
Southan
C.
,
Kohan
D.E.
,
Pollock
J.S.
et al.
(
2016
)
Endothelin
.
Pharmacol. Rev.
68
,
357
418
[PubMed]
15.
Henriksson
M.
,
Xu
C.B.
and
Edvinsson
L.
(
2004
)
Importance of ERK1/2 in upregulation of endothelin type B receptors in cerebral arteries
.
Br. J. Pharmacol.
142
,
1155
1161
[PubMed]
16.
Sandhu
H.
,
Ansar
S.
and
Edvinsson
L.
(
2010
)
Comparison of MEK/ERK pathway inhibitors on the upregulation of vascular G-protein coupled receptors in rat cerebral arteries
.
Eur. J. Pharmacol.
644
,
128
137
[PubMed]
17.
Blixt
F.W.
,
Haanes
K.A.
,
Ohlsson
L.
,
Tolstrup Christiansen
A.
,
Warfvinge
K.
and
Edvinsson
L.
(
2018
)
Increased endothelin-1-mediated vasoconstriction after organ culture in rat and pig ocular arteries can be suppressed with MEK/ERK1/2 inhibitors
.
Acta Ophthalmol.
96
,
e619
e625
[PubMed]
18.
Prunell
G.F.
,
Mathiesen
T.
and
Svendgaard
N.A.
(
2002
)
A new experimental model in rats for study of the pathophysiology of subarachnoid hemorrhage
.
Neuroreport
13
,
2553
2556
[PubMed]
19.
Johansson
S.E.
,
Abdolalizadeh
B.
,
Sheykhzade
M.
,
Edvinsson
L.
and
Sams
A.
(
2019
)
Vascular pathology of large cerebral arteries in experimental subarachnoid hemorrhage: vasoconstriction, functional CGRP depletion and maintained CGRP sensitivity
.
Eur. J. Pharmacol.
846
,
109
118
[PubMed]
20.
Larsen
C.C.
,
Povlsen
G.K.
,
Rasmussen
M.N.
and
Edvinsson
L.
(
2011
)
Improvement in neurological outcome and abolition of cerebrovascular endothelin B and 5-hydroxytryptamine 1B receptor upregulation through mitogen-activated protein kinase kinase 1/2 inhibition after subarachnoid hemorrhage in rats
.
J. Neurosurg.
114
,
1143
1153
[PubMed]
21.
Pardridge
W.M.
(
2011
)
Drug transport in brain via the cerebrospinal fluid
.
Fluids Barriers CNS
8
,
7
[PubMed]
22.
Ohlsson
A.L.
and
Johansson
B.B.
(
1995
)
Environment influences functional outcome of cerebral infarction in rats
.
Stroke
26
,
644
649
[PubMed]
23.
Christensen
S.T.
,
Johansson
S.E.
,
Radziwon-Balicka
A.
,
Warfvinge
K.
,
Haanes
K.A.
and
Edvinsson
L.
(
2019
)
MEK1/2 inhibitor U0126, but not nimodipine, reduces upregulation of cerebrovascular contractile receptors after subarachnoid haemorrhage in rats
.
PLoS One
14
,
e0215398
[PubMed]
24.
Christensen
S.T.
,
Grell
A.S.
,
Johansson
S.E.
,
Andersson
C.M.
,
Edvinsson
L.
and
Haanes
K.A.
(
2019
)
Synergistic effects of a cremophor EL drug delivery system and its U0126 cargo in an ex vivo model
.
Drug Deliv.
26
,
680
688
[PubMed]
25.
Johansson
S.E.
,
Andersen
X.E.
,
Hansen
R.H.
,
Povlsen
G.K.
and
Edvinsson
L.
(
2015
)
Cerebrovascular endothelin-1 hyper-reactivity is associated with transient receptor potential canonical channels 1 and 6 activation and delayed cerebral hypoperfusion after forebrain ischaemia in rats
.
Acta Physiol. (Oxf.)
214
,
376
389
[PubMed]
26.
Jeon
H.
,
Ai
J.
,
Sabri
M.
,
Tariq
A.
,
Shang
X.
,
Chen
G.
et al.
(
2009
)
Neurological and neurobehavioral assessment of experimental subarachnoid hemorrhage
.
BMC Neurosci.
10
,
103
[PubMed]
27.
Frismantiene
A.
,
Philippova
M.
,
Erne
P.
and
Resink
T.J.
(
2018
)
Smooth muscle cell-driven vascular diseases and molecular mechanisms of VSMC plasticity
.
Cell. Signal.
52
,
48
64
[PubMed]
28.
Barrett
S.D.
,
Bridges
A.J.
,
Dudley
D.T.
,
Saltiel
A.R.
,
Fergus
J.H.
,
Flamme
C.M.
et al.
(
2008
)
The discovery of the benzhydroxamate MEK inhibitors CI-1040 and PD 0325901
.
Bioorg. Med. Chem. Lett.
18
,
6501
6504
[PubMed]
29.
Buckell
M.
(
1968
)
Enzymes in the cerebrospinal fluid
.
Proc. Assoc. Clin. Biochemists
5
,
33
34
30.
Siegelman
J.
,
Fleit
H.B.
and
Peress
N.S.
(
1987
)
Characterization of immunoglobulin G-Fc receptor activity in the outflow system of the cerebrospinal fluid
.
Cell Tissue Res.
248
,
599
605
[PubMed]
31.
Paulo
J.A.
,
McAllister
F.E.
,
Everley
R.A.
,
Beausoleil
S.A.
,
Banks
A.S.
and
Gygi
S.P.
(
2015
)
Effects of MEK inhibitors GSK1120212 and PD0325901 in vivo using 10-plex quantitative proteomics and phosphoproteomics
.
Proteomics
15
,
462
473
[PubMed]
32.
Cao
H.
,
Dronadula
N.
,
Rizvi
F.
,
Li
Q.
,
Srivastava
K.
,
Gerthoffer
W.T.
et al.
(
2006
)
Novel role for STAT-5B in the regulation of Hsp27-FGF-2 axis facilitating thrombin-induced vascular smooth muscle cell growth and motility
.
Circ. Res.
98
,
913
922
[PubMed]
33.
Kundumani-Sridharan
V.
,
Wang
D.
,
Karpurapu
M.
,
Liu
Z.
,
Zhang
C.
,
Dronadula
N.
et al.
(
2007
)
Suppression of activation of signal transducer and activator of transcription-5B signaling in the vessel wall reduces balloon injury-induced neointima formation
.
Am. J. Pathol.
171
,
1381
1394
[PubMed]
34.
Zhao
F.K.
,
Chuang
L.F.
,
Israel
M.
and
Chuang
R.Y.
(
1989
)
Cremophor EL, a widely used parenteral vehicle, is a potent inhibitor of protein kinase C
.
Biochem. Biophys. Res. Commun.
159
,
1359
1367
[PubMed]
35.
Shimizu
S.
,
Simon
R.P.
and
Graham
S.H.
(
1997
)
Dimethylsulfoxide (DMSO) treatment reduces infarction volume after permanent focal cerebral ischemia in rats
.
Neurosci. Lett.
239
,
125
127
[PubMed]
36.
Bardutzky
J.
,
Meng
X.
,
Bouley
J.
,
Duong
T.Q.
,
Ratan
R.
and
Fisher
M.
(
2005
)
Effects of intravenous dimethyl sulfoxide on ischemia evolution in a rat permanent occlusion model
.
J. Cereb. Blood Flow Metab.
25
,
968
977
[PubMed]
37.
Yamaguchi
T.
,
Kakefuda
R.
,
Tajima
N.
,
Sowa
Y.
and
Sakai
T.
(
2011
)
Antitumor activities of JTP-74057 (GSK1120212), a novel MEK1/2 inhibitor, on colorectal cancer cell lines in vitro and in vivo
.
Int. J. Oncol.
39
,
23
31
[PubMed]
38.
Kamakura
S.
,
Moriguchi
T.
and
Nishida
E.
(
1999
)
Activation of the protein kinase ERK5/BMK1 by receptor tyrosine kinases. Identification and characterization of a signaling pathway to the nucleus
.
J. Biol. Chem.
274
,
26563
26571
[PubMed]
39.
Mody
N.
,
Leitch
J.
,
Armstrong
C.
,
Dixon
J.
and
Cohen
P.
(
2001
)
Effects of MAP kinase cascade inhibitors on the MKK5/ERK5 pathway
.
FEBS Lett.
502
,
21
24
[PubMed]
40.
Dokladda
K.
,
Green
K.A.
,
Pan
D.A.
and
Hardie
D.G.
(
2005
)
PD98059 and U0126 activate AMP-activated protein kinase by increasing the cellular AMP:ATP ratio and not via inhibition of the MAP kinase pathway
.
FEBS Lett.
579
,
236
240
[PubMed]
41.
Andrieux
L.
,
Langouet
S.
,
Fautrel
A.
,
Ezan
F.
,
Krauser
J.A.
,
Savouret
J.F.
et al.
(
2004
)
Aryl hydrocarbon receptor activation and cytochrome P450 1A induction by the mitogen-activated protein kinase inhibitor U0126 in hepatocytes
.
Mol. Pharmacol.
65
,
934
943
[PubMed]
42.
Yuan
L.L.
,
Chen
X.
,
Kunjilwar
K.
,
Pfaffinger
P.
and
Johnston
D.
(
2006
)
Acceleration of K+ channel inactivation by MEK inhibitor U0126
.
Am. J. Physiol. Cell Physiol.
290
,
C165
C171
[PubMed]
43.
Quan
H.
,
Liu
H.
,
Li
C.
and
Lou
L.
(
2009
)
1,4-Diamino-2,3-dicyano-1,4-bis(methylthio)butadiene (U0126) enhances the cytotoxicity of combretastatin A4 independently of mitogen-activated protein kinase kinase
.
J. Pharmacol. Exp. Ther.
330
,
326
333
[PubMed]
44.
Selleckem. Available at: https://www.selleckchem.com/MEK.html accessed on June 20th, 2019