• cell-based therapy
• predictive biomarkers
• proportional recovery
• stem cells
• stroke

Stroke is the third death cause in Italy and it affects 196’000 new patients per year. Around 70-80% of patients survive the ischemic attack, however around three fourths of them do not recover completely. Stroke is a heterogeneous condition: this makes the prediction of outcome and novel treatments development extremely challenging.
Many studies have showed how spontaneous recovery after stroke is variable and hard to predict. So far, the best factor for predicting long-term upper limb outcome is the initial severity of motor impairment. Interestingly, the amount of regained function is proportional to the initial deficit and corresponds to ~70%: this phenomenon is called proportional recovery. Unfortunately, this empirical rule holds true for just half of the highly severed patients, whereas the other half shows no substantial recovery.
For this reason, the field strongly needs to develop appropriate biomarkers to accurately predict long-term clinical outcome. Such biomarkers would allow to (i) stratify patients participating in clinical trials according to the expected outcome, and (ii) to judge the effectiveness of novel treatment approaches taking into account the amount of spontaneous recovery.
To address this point, we performed experiments on a preclinical mouse model of stroke, that consist in the permanent occlusion of the middle cerebral artery. We measured the extent of the lesion and the affected cortical regions via immunohistochemistry. Then, we performed electrophysiological recordings from primary forelimb motor cortex, which is perilesional to the stroke area, by means of chronic bipolar electrodes. More specifically, we recorded electrical activity in freely moving mice and during the execution of a motor task, i.e. the retraction of the forelimb on a robotic platform. From these recordings, we extracted several parameters (e.g. spectral bands) that we then correlated with the performance in several behavioral tests, namely the gridwalk test, the skilled-reaching task, the isometric force task and the quantitative parameters related to the retraction task onto the robotic platform. This way, we expect to discover electrophysiological markers that strongly correlate with the initial deficit and (more importantly) with spontaneous recovery, and thus have the strongest predictive power.

As we mentioned above, in ischemic stroke more than half of patients suffer significant residual deficits. Unfortunately, the currently available therapeutic strategies are aimed at the removal of the thrombi that occlude cerebral arteries, but no effective treatments which promote functional recovery exist in the post-ischemic phase. Stem cell-based approaches are very promising as potential novel therapies to restore brain function after ischemic stroke. This way, new cellular elements are added in the injured brain and they can be instructed to restore the lost functional circuitry.
For this purpose, we developed a protocol of neural progenitors transplantation to promote such a functional recovery in a photothrombotic model of cortical stroke. Our cells are derived from mouse embryonic stem cells, that are exposed to a double Wnt-BMP inhibition for 16 days in vitro. This protocol allows to obtain a population of post-mitotic cortical progenitors, that are not fully differentiated.
We grafted these cells two days after the stroke induction, and the gridwalk test was performed once a week up to 23 days post lesion. We performed immunohistochemistry analysis and we confirmed that the cells are able to survive and also extend long-range projections. From a behavioral point of view, the animals showed a recovery of motor function that reached the baseline level. Improvements in motor output were already apparent as early as 16 days after stroke, suggesting that bystander effects (i.e. modulation of inflammation, secretion of trophic factors, etc.) likely play a key role in the observed recovery.
As future perspective, we are planning to optimize the transplantation procedure and, more importantly, to implement a chemogenetic approach to switch on and off the new neurons. In particular, we will selectively stimulate them in phase with a motor task execution: due to Hebbian plasticity, this strategy is expected to guide their proper integration into spared cortical networks, thus improving functional recovery. At the same time, their acute silencing at the completion of training will allow the unambiguous assessment of their role in motor recovery.