Repeated administration in the present study did not confer neuroprotection from photothrombotic ischemia. Focal ischemia in rats selectively up-regulates AREmediated gene expression. ARE mediated gene expression itself is mediated by Nrf2. This in conjunction with other ischemic cascade processes such as mitochondrial inhibition, oxidative stress and glucose deprivation all promote the activity of Nrf2. We hypothesize that the sulforaphane treatment starting 15 min after ischemia induction did not provide any further stimulation of the Nrf2 system in addition to the activation triggered by the ischemia alone. Functional disabilities following ischemic stroke are often profound and therefore it is of fundamental importance to study changes in animal behaviour after ischemia. Specific tests on forepaw motor function demonstrated that photothrombotic infarction results in animals losing fine motor control of their paw. The mice spontaneously recovered use of their forepaw quickly after infarction. In our study, single and repeated administration of sulforaphane did not affect the functional deficit at 24 h or the speed of functional recovery. However, as only two time points for the evaluation were used in our study and all the mice fully recovered by 72 h, we cannot exclude that the sulforaphane treatment affected the functional deficit between these time points. The photothrombotic infarct is associated with extensive glial cell activation and recruitment of different neural cells in the areas surrounding the ischemic region. Reactive gliosis constitutes a complex cellular response to different types of injuries and stress to the CNS and the evolving cellular reactions can be followed through the up-regulation of GFAP in LY2157299 molecular weight astrocytes and Iba1 in microglia. Interestingly, involvement of the Nrf2 system in the underlying mechanisms of glial cell activation has been proposed since Nrf2-deficient mice demonstrate widespread astrogliosis throughout the CNS. However, in the present study, activation of the Nrf2 system by repeated sulforaphane treatment did not alter the number of BrdU, GFAP and Iba1 positive cells compared to vehicle injection, suggesting that at the time points and concentrations used, sulforaphane did not interfere with the mechanisms involved in the activation of glial cells or cell proliferation. The photothrombotic model used in the present study induces a much smaller penumbra compared to the transient MCAO model where the collateral blood flow is more prominent. This, together with the small volume of infarcted cortical tissue resulting from the photothrombosis injury, might result in the generation of relatively smaller amounts of ROS compared to the MCAO and may explain why we did not observe any neuroprotection after sulforaphane administration. The photothrombotic stroke model was chosen for the current study for the reason that it is highly reproducible regarding location and size, generating small infarcts limited to the cortex with a similar cellular response as the MCAO model. The model generates an irreversibly damaged ischemic core surrounded by a small penumbral region. Importantly, neuroprotection has previously been achieved in the photothrombotic model by i.e. the free radical scavenger 21- aminosteroid, NMDA and GABAA receptor antagonists, the allosteric modulators lubeluzole and chlormethiazole and the calcium channel blocker flunarizine. In contrast, MK801 and nimodipine were shown to be neuroprotective in other models of cerebral ischemia but failed to be neuroprotective in photothrombotic ischemia. Based on our findings, sulforaphane is another compound the neuroprotective effects of which are model dependent.