Combination therapy in ischaemic stroke

Stroke is the 3rd leading cause of death worldwide, second to heart disease and cancer, and is the leading cause of acquired adult long-term disability. Despite substantial advancement in the understanding of the pathogenesis of ischaemic stroke, development of treatments for stroke patients has been largely unsuccessful. As a result, administration of the thrombolytic, recombinant tissue plasminogen activator (rt-PA), in the acute stages following onset of ischaemia, remains the only clinically approved therapy. However, the 4.5 h “door-to-needle” time window observed with ischaemic stroke in addition to rt-PA’s potential haemorrhage risk, results in only 2 – 5 % of all stroke patients receiving this therapy. Stroke is therefore a disease with largely unmet clinical needs. Following its discovery in 2000, neuroglobin (Ngb) has been reported to mediate neuroprotection via a number of different pathways in both in vitro models of hypoxia ± reoxygenation and pre-clinical in vivo models of stroke, when upregulated from endogenous levels. c-Jun N-terminal Kinase (JNK) is a well-documented downstream mediator of apoptosis. Inhibition of JNK, by a number of different strategies, has been reported to be neuroprotective following hypoxia ± reoxygenation in vitro and ischaemia ± reperfusion in vivo. Although a regularly utilised strategy in the treatment of other cardiovascular diseases (CVD) and cancers, combination therapy is not currently employed in the treatment of stroke. Therefore it was the aim of this PhD to assess the effect of gene delivery-mediated upregulation of Ngb in combination with pharmacological pan-inhibition of JNK, using SP600125. The benefit of these therapies would first be assessed in an in vitro model of hypoxia / reoxygenation to provide the proof-of-concept required to perform a comprehensive in vivo study. In vivo analysis would include randomised and blinded assessment of neurological deficit longitudinally across a recovery period of 14 days and final lesion size in a rat strain exhibiting a number of the co-morbidities of CVD, the stroke-prone spontaneously hypertensive rat (SHRSP). It was first necessary to determine the optimal viral vector for use in both the in vitro and in vivo intervention studies. Canine adenovirus serotype 2 (CAV2) had been reported to travel with great efficiency from the injection site following stereotactic injection into the striatum, with substantial reporter-gene expression observed in afferent structures such as the cortex. As the cortex is widely accepted as the location of the penumbra following ischaemia / reperfusion in the present middle cerebral artery occlusion (MCAO) model, the level of cortical expression was of significant importance and as such compared following both striatal and cortical injection of CAV2-GFP 7 days post-injection in male SHRSPs. Significantly greater cortical transduction was observed following direct injection into the cortex with 3 x 109 vp CAV2-GFP in comparison to striatal injection. To ensure CAV2 was the best possible viral vector for use in this study, an additional comparative analysis with a GFP-expressing lentivirus was assessed. 4 x 107 vp lenti-GFP was administered stereotactically into the cortex over two injection sites, spanning the perceived region of penumbral tissue - owing to its previously reported inefficient transduction from injection site. Analysis at 7 days post-injection demonstrated a significantly reduced level of transduction in comparison to CAV2-GFP. Therefore a Ngb-expression CAV2 was generated. In vitro analysis of CAV2 demonstrated an inability to transduce the B50 rat neuronal cell line being utilised for the in vitro model. Therefore, a Ngb-expressing lentivirus was generated to mediate Ngb overexpression in the in vitro model with CAV2-Ngb utilised in vivo. The effect of pre-treatment with MOI 5 vp / cell lenti-Ngb in combination with 20 µM SP600125 was examined in an in vitro model of 9 h hypoxia and serum deprivation with 24 h reoxygenation in complete medium using B50 rat neuronal cells. The results showed that pre-treatment with lenti-Ngb in combination with SP600125 lowered levels of oxidative stress and apoptosis to that equivalent of normoxic control cells, assessed all across four assays of cell viability. The four assays included two oxidative stress assays [Electron paramagnetic resonance (EPR) to assess superoxide (SO) production and a bioluminescent malondialdehyde (MDA) assay as a marker of lipid peroxidation], and two apoptosis assays (a cell death ELISA and caspase-3 immunocytochemistry). Interestingly, although it had initially been perceived to act as an anti-oxidant, Ngb upregulation was shown to significantly mediate neuroprotection through both anti-oxidant and anti-apoptotic parameters. Additionally, SP600125 pre-treatment alone was shown to have no effect on SO levels but potentiated Ngb’s action when administered in combination, assessed by EPR. Furthermore, the anti-apoptotic agent, SP600125, completely abolished levels of lipid peroxidation, used as a marker of oxidative stress, when administered alone. A pilot in vivo study of 45 min transient middle cerebral artery occlusion (tMCAO), using the intraluminal thread model, with 14 days recovery was performed in 12 SHRSP in order to assess the reproducibility of infarct in addition to assessing the sensitivity of the desired neurological tests. Results demonstrated a substantial variation in infarct size between animals, with 4 out of 12 animals being excluded from the study as they did not exhibit an infarct. Average results from neurological tests in remaining animals were not significantly different from baseline at day 14. Additionally, one neurological test (the cylinder test) was discontinued mid-study as animals at baseline and post-tMCAO failed to rear onto hind legs at any point making analysis impossible. This pilot in vivo study allowed for procedural surgical issues to be ‘flagged’ and for refinement of technique prior to initiation of the intervention study. In addition, the pilot study allowed for experience to be gained in outcome assessments including neurological scoring and ex vivo infarct analysis, to ensure minimal user variation following initiation of the intervention study. Further methodological refinement for the tMCAO model was implemented as a result of high mortality rates observed within control animals following initiation of the in vivo intervention study. It was observed that following stereotactic administration of control virus (CAV2-GFP), although there was no significant difference in neurological score between CAV2-GFP and control animals, mortality was significantly reduced in CAV2-GFP treated animals. Subsequent analysis determined no significant difference in lesion size between control animals and CAV2-GFP treated animal. This led to the hypothesis that the administration of a small (1 mm diameter) craniectomy (although immediately sealed with dental cement following injection) and subsequent piercing of the dura attributed to the stereotactic procedure perhaps allowed for a reduction in intracranial pressure in the acute stages following tMCAO improving survival. Additionally, it has been reported that pre-conditioning animals to isoflurane may mediate neuroprotection, as is observed in the present study to carry out a pre-tMCAO surgical procedure. However, although it is likely this would have been reflected in final lesion size, the present study may be weighted to only observe animals exhibiting a lesion within a certain threshold. Finally, an in vivo intervention study was performed to determine the effect of pre-treatment with 3 x 109 vp CAV2-Ngb in combination with 1 mg / kg SP600125 in a model of tMCAO in male SHRSP with 14 days recovery. CAV2-Ngb or CAV2-GFP was administered into the cortex in a total of 2 µl 5 days prior to tMCAO, and SP600125 was administered i.v. 15 min pre- and 3 h post-tMCAO onset. Results of the intervention study showed that combined treatment of CAV2-Ngb and SP600125 reduced final infarct greater than either therapy alone, by 58 % in comparison to untreated controls. Treatment with SP600125 alone reduced infarct by 36 % and CAV2-Ngb treatment resulted in a reduction of 38 %. This was reflected in neurological outcome where combined treatment mediated an improvement in neurological recovery assessed by 32-point neurological (where a healthy animal scores 32) from 18 in control animals to 26 in those receiving the combined intervention at day 14. Treatment with either SP600125 or CAV2-Ngb alone improved 32-point neurological score to 20 at day 14. Additionally, percentage footfalls were significantly improved from 45 % of all steps taken in control animals to ~ 28 % in animals administered with either single or combined treatment at day 14, no further improvement was noted with combined treatment assessed by this outcome. This study represents the first anti-oxidant and anti-apoptotic combination study of cerebral ischaemia and additionally the first time a CAV2 vector has been utilised in a model of experimental stroke. In conclusion, research presented in this thesis demonstrates the combined neuroprotective action of Ngb upregulation in combination with JNK inhibition in an in vitro model of hypoxia / reoxygenation improves neuronal cell viability greater than either therapy alone. Subsequently, a robust pre-clinical in vivo model of tMCAO utilising the SHRSP, demonstrated a substantial reduction in lesion size and improvement neurological score, mediated by combined therapy greater than singular treatment