Ahmed, Jewel
(2020)
Refining behavioural analysis to promote the use of a milder model of stroke.
PhD thesis, University of Nottingham.
Abstract
The intraluminal filament model is the most commonly used model of stroke and has been imperative to our understanding of this disease. Nonetheless, this model has several disadvantages, such as high variability and poor animal welfare. An alternative model that could be useful is the distal middle cerebral artery occlusion (dMCAO) of stroke, which produces far more reproducible lesions and better animal welfare. This model could be useful for preclinical stroke studies whilst addressing the 3R’s; the refinement, reduction and replacement of animals for scientific purposes. The caveat is that the dMCAO model is mild and primarily affects the somatosensory cortex (including the barrel cortex, primary forelimb and hindlimb areas and the secondary somatosensory cortex) whilst leaving the motor cortex mostly spared. Therefore, detection of functional deficits is difficult and because of this the dMCAO model is seldom used. The overall aim of this thesis was to develop methods to improve the detection of functional deficits in the dMCAO model.
A meta-analysis was conducted with the aim of systematically identifying the different functional assessments used following dMCAO to assess which tests could consistently detect functional deficits. Relevant publications were identified and behavioural and infarct data were extracted. Following this a meta-analysis was conducted using a random effects model to estimate the effect sizes of the various functional assessments. In mice, the adhesive removal test, cylinder test and ladder test produced large effect sizes. In rats, the staircase paw reaching task produced large effect sizes.
In addition assessment of a number of tests was carried out within the laboratory. A pilot study was conducted with the aim of clarifying which were the most and least useful tests in the dMCAO model from tests that are commonly use in the field. In addition, several novel tests were piloted. 15 mice underwent either dMCAO (n = 8) or sham surgery (n = 7) before undergoing testing with a battery of tests, including: grip strength, corner test, cylinder test, adhesive removal test, peri-oral stimulation, rotarod and staircase paw reaching. In addition, we piloted two novel tests; whisker discrimination and lickometry. Testing was conducted from 24 hours to 9 weeks post-surgery. Deficits were only seen with the adhesive removal test (at 24 hours and 7 days post-surgery) and staircase paw-reaching (4-7 weeks post-surgery). However, there were other promising results could have been useful with optimisation, such as lickometry. Therefore the following experiment focused on optimising these.
A larger scale study was then carried out (dMCAO, n = 13; sham, n = 12), focusing on the most promising tasks (adhesive removal, paw-reaching and lickometry) and those commonly used in the field (cylinder test, grip strength and neuroscore). Additional optimisations to these tasks were examined. Testing was conducted from 24 hours to 14 weeks post-surgery. However, no deficits were seen in any of the tests, except the adhesive removal test at 28 days post-surgery.
With these findings we began to question whether refining behavioural analysis was the most appropriate approach and instead investigated if the model itself could be refined to suit the current behavioural assessment methods by increasing the lesion volume. The approach used combined dMCAO with either permanent ipsilateral common carotid artery occlusion (CCAO) or permanent ipsilateral internal carotid artery occlusion (ICAO). Lesion volume was assessed at 24 hours post-surgery. A total of 25 mice were used (dMCAO only, n = 9, dMCAO + CCAO, n = 8, dMCAO + ICAO, n = 8) and this data was compared to historical data in our laboratory from mice that underwent middle cerebral artery occlusion via the intraluminal filament model (n = 8). We found that dMCAO with tandem occlusion of either the CCA or ICA resulted in significantly increased infarct volume when compared to dMCAO only. There was no significant difference in infarct volume between the dMCAO + CCAO model and dMCAO + ICAO occlusion model. However, the dMCAO + CCAO model produced more reproducible infarcts with a co-efficient of variation of 17.75% compared to 25.35% in the dMCAO + ICAO model. The next step was to characterise functional outcome in the dMCAO + CCAO model, whilst providing a more comprehensive characterisation of welfare.
For the final experiment, we carried out dMCAO + permanent ipsilateral CCAO (n = 18) or sham surgery (n = 15) before assessing welfare by neuroscore, mouse grimace scoring and assessment of 7-day post-surgery weight-loss and recovery. In addition, we assessed functional outcome with a large battery of tests, mostly including those tests used in experiments 1 and 2 with the addition of “whisker tracking” as part of the CRACKIT challenge. Whisker tracking uses high-speed cameras to track whiskers to detect subtle changes in whisking behaviours. Testing was conductd from 24 hours to 9 weeks post-surgery. We found no significant differences between dMCAO + CCAO and sham mice for neuroscore and grimace score. For weight-loss, both groups lost weight at 24 hours before recovering to pre-surgery weight at 48 hours and beyond, suggesting weight-loss was independent of the infarct. No functional deficits were seen with any of the tests, except lickometry; consumption per 1000 licks was significantly lower in the dMCAO + CCAO group at 1-month post-surgery.
Overall, the results demonstrated that there is good welfare with the dMCAO model even when infarct volume is increased by tandem occlusion of the CCA. However, functional deficits are very difficult to detect in this model even with the use of novel and sensitive behavioural assays, such as lickometry and whisker tracking. With the good welfare seen in the dMCAO model, other stroke researchers should adopt this model when functional outcome is not being assessed in their studies. However, when functional outcome being assessed, researchers should look to other models, such as the photothrombotic model or intraluminal filament model. Future research can focus on developing more sensitive functional assessments that are appropriate for mild models of stroke, further exploration of the dMCAO + CCAO model (i.e. by combination with aged or co-morbid animals) and improving welfare and reproducibility in the intraluminal filament model.
Actions (Archive Staff Only)
 |
Edit View |
Tools
Tools
![[img]](/style/images/fileicons/application_pdf.png)