Convolutional neural network models applied to neuronal responses in macaque V1 reveal limited nonlinear processing.

Computational models of the primary visual cortex (V1) have suggested that V1 neurons behave like Gabor filters followed by simple nonlinearities. However, recent work employing convolutional neural network (CNN) models has suggested that V1 relies on far more nonlinear computations than previously thought. Specifically, unit responses in an intermediate layer of VGG-19 were found to best predict macaque V1 responses to thousands of natural and synthetic images. Here, we evaluated the hypothesis that the poor performance of lower layer units in VGG-19 might be attributable to their small receptive field size rather than to their lack of complexity per se. We compared VGG-19 with AlexNet, which has much larger receptive fields in its lower layers. Whereas the best-performing layer of VGG-19 occurred after seven nonlinear steps, the first convolutional layer of AlexNet best predicted V1 responses. Although the predictive accuracy of VGG-19 was somewhat better than that of standard AlexNet, we found that a modified version of AlexNet could match the performance of VGG-19 after only a few nonlinear computations. Control analyses revealed that decreasing the size of the input images caused the best-performing layer of VGG-19 to shift to a lower layer, consistent with the hypothesis that the relationship between image size and receptive field size can strongly affect model performance. We conducted additional analyses using a Gabor pyramid model to test for nonlinear contributions of normalization and contrast saturation. Overall, our findings suggest that the feedforward responses of V1 neurons can be well explained by assuming only a few nonlinear processing stages.

Improved modeling of human vision by incorporating robustness to blur in convolutional neural networks.

Whenever a visual scene is cast onto the retina, much of it will appear degraded due to poor resolution in the periphery; moreover, optical defocus can cause blur in central vision. However, the pervasiveness of blurry or degraded input is typically overlooked in the training of convolutional neural networks (CNNs). We hypothesized that the absence of blurry training inputs may cause CNNs to rely excessively on high spatial frequency information for object recognition, thereby causing systematic deviations from biological vision. We evaluated this hypothesis by comparing standard CNNs with CNNs trained on a combination of clear and blurry images. We show that blur-trained CNNs outperform standard CNNs at predicting neural responses to objects across a variety of viewing conditions. Moreover, blur-trained CNNs acquire increased sensitivity to shape information and greater robustness to multiple forms of visual noise, leading to improved correspondence with human perception. Our results provide multi-faceted neurocomputational evidence that blurry visual experiences may be critical for conferring robustness to biological visual systems.

Convolutional neural network models of neuronal responses in macaque V1 reveal limited non-linear processing.

Computational models of the primary visual cortex (V1) have suggested that V1 neurons behave like Gabor filters followed by simple non-linearities. However, recent work employing convolutional neural network (CNN) models has suggested that V1 relies on far more non-linear computations than previously thought. Specifically, unit responses in an intermediate layer of VGG-19 were found to best predict macaque V1 responses to thousands of natural and synthetic images. Here, we evaluated the hypothesis that the poor performance of lower-layer units in VGG-19 might be attributable to their small receptive field size rather than to their lack of complexity per se. We compared VGG-19 with AlexNet, which has much larger receptive fields in its lower layers. Whereas the best-performing layer of VGG-19 occurred after seven non-linear steps, the first convolutional layer of AlexNet best predicted V1 responses. Although VGG-19's predictive accuracy was somewhat better than standard AlexNet, we found that a modified version of AlexNet could match VGG-19's performance after only a few non-linear computations. Control analyses revealed that decreasing the size of the input images caused the best-performing layer of VGG-19 to shift to a lower layer, consistent with the hypothesis that the relationship between image size and receptive field size can strongly affect model performance. We conducted additional analyses using a Gabor pyramid model to test for non-linear contributions of normalization and contrast saturation. Overall, our findings suggest that the feedforward responses of V1 neurons can be well explained by assuming only a few non-linear processing stages.

Improved modeling of human vision by incorporating robustness to blur in convolutional neural networks.

Whenever a visual scene is cast onto the retina, much of it will appear degraded due to poor resolution in the periphery; moreover, optical defocus can cause blur in central vision. However, the pervasiveness of blurry or degraded input is typically overlooked in the training of convolutional neural networks (CNNs). We hypothesized that the absence of blurry training inputs may cause CNNs to rely excessively on high spatial frequency information for object recognition, thereby causing systematic deviations from biological vision. We evaluated this hypothesis by comparing standard CNNs with CNNs trained on a combination of clear and blurry images. We show that blur-trained CNNs outperform standard CNNs at predicting neural responses to objects across a variety of viewing conditions. Moreover, blur-trained CNNs acquire increased sensitivity to shape information and greater robustness to multiple forms of visual noise, leading to improved correspondence with human perception. Our results provide novel neurocomputational evidence that blurry visual experiences are very important for conferring robustness to biological visual systems.

Longitudinal evaluation of the functional connectivity changes in the secondary somatosensory cortex (S2) of the monkey brain during acute stroke.

Background: Somatosensory deficits are frequently seen in acute stroke patients and may recover over time and affect functional outcome. However, the underlying mechanism of function recovery remains poorly understood. In the present study, progressive function alteration of the secondary somatosensory cortex (S2) and its relationship with regional perfusion and neurological outcome were examined using a monkey model of stroke. Methods and materials: Rhesus monkeys (n = 4) were induced with permanent middle cerebral artery occlusion (pMCAo). Resting-state functional MRI, dynamic susceptibility contrast perfusion MRI, diffusion-weighted, T1 and T2 weighted images were collected before surgery and at 4-6, 48, and 96 h post stroke on a 3T scanner. Progressive changes of relative functional connectivity (FC), cerebral blood flow (CBF), and CBF/Tmax (Time to Maximum) of affected S2 regions were evaluated. Neurological deficits were assessed using the Spetzler approach. Results: Ischemic lesion was evidently seen in the MCA territory including S2 in each monkey. Relative FC of injured S2 regions decreased substantially following stroke. Spetzler scores dropped substantially at 24 h post stroke but slightly recovered from Day 2 to Day 4. Relative FC progressively increased from 6 to 48 and 96 h post stroke and correlated significantly with relative CBFand CBF/Tmax changes. Conclusion: The present study revealed the progressive alteration of function connectivity in S2 during acute stroke. The preliminary results suggested the function recovery might start couple days post occlusion and collateral circulation might play a key role in the recovery of somatosensory function after stroke insult. The relative function connectivity in S2 may provide additional information for prediction of functional outcome in stroke patients.

Rapid adaptation of primate LGN neurons to drifting grating stimulation.

The visual system needs to dynamically adapt to changing environments. Much is known about the adaptive effects of constant stimulation over prolonged periods. However, there are open questions regarding adaptation to stimuli that are changing over time, interrupted, or repeated. Feature-specific adaptation to repeating stimuli has been shown to occur as early as primary visual cortex (V1), but there is also evidence for more generalized, fatigue-like adaptation that might occur at an earlier stage of processing. Here, we show adaptation in the lateral geniculate nucleus (LGN) of awake, fixating monkeys following brief (1 s) exposure to repeated cycles of a 4-Hz drifting grating. We examined the relative change of each neuron's response across successive (repeated) grating cycles. We found that neurons from all cell classes (parvocellular, magnocellular, and koniocellular) showed significant adaptation. However, only magnocellular neurons showed adaptation when responses were averaged to a population response. In contrast to firing rates, response variability was largely unaffected. Finally, adaptation was comparable between monocular and binocular stimulation, suggesting that rapid LGN adaptation is monocular in nature.NEW & NOTEWORTHY Neural adaptation can be defined as reduction of spiking responses following repeated or prolonged stimulation. Adaptation helps adjust neural responsiveness to avoid saturation and has been suggested to improve perceptual selectivity, information transmission, and predictive coding. Here, we report rapid adaptation to repeated cycles of gratings drifting over the receptive field of neurons at the earliest site of postretinal processing, the lateral geniculate nucleus of the thalamus.

A sculpting effect of reading on later representational quality of phonology revealed by multi-voxel pattern analysis in young children.

Using univariate analysis, a previous study by Wang et al. (2020) found a scaffolding effect of earlier phonological representation in superior temporal gyrus (STG) on later reading skill but failed to observe a sculpting effect of earlier reading on later phonological representation. The current study applied multi-voxel pattern analysis (MVPA) to examine if both scaffolding and sculpting effects were present in young children. We found that better initial reading skill predicted higher decoding coefficient of brain activity patterns for phonological representations in STG. This sculpting effect was present only for decoding small grain sizes (phonemes) and in younger children (6- to 7.5-year-olds), as we did not find any effects for large grain sizes (rhymes) or older children (7.5- to 9.5-year-olds). Although a scaffolding effect was not observed, the current study provides the first neural evidence of how earlier reading sculpts later phonological awareness in beginning readers.

Spikiness and animacy as potential organizing principles of human ventral visual cortex.

Considerable research has been devoted to understanding the fundamental organizing principles of the ventral visual pathway. A recent study revealed a series of 3-4 topographical maps arranged along the macaque inferotemporal (IT) cortex. The maps articulated a two-dimensional space based on the spikiness and animacy of visual objects, with "inanimate-spiky" and "inanimate-stubby" regions of the maps constituting two previously unidentified cortical networks. The goal of our study was to determine whether a similar functional organization might exist in human IT. To address this question, we presented the same object stimuli and images from "classic" object categories (bodies, faces, houses) to humans while recording fMRI activity at 7 Tesla. Contrasts designed to reveal the spikiness-animacy object space evoked extensive significant activation across human IT. However, unlike the macaque, we did not observe a clear sequence of complete maps, and selectivity for the spikiness-animacy space was deeply and mutually entangled with category-selectivity. Instead, we observed multiple new stimulus preferences in category-selective regions, including functional sub-structure related to object spikiness in scene-selective cortex. Taken together, these findings highlight spikiness as a promising organizing principle of human IT and provide new insights into the role of category-selective regions in visual object processing.

Aneurysmal recurrence after successful flow-diversion embolization.

Aneurysmal recurrence after successful flow-diversion embolization is exceptionally rare. The rarity of recurrence has called into question the yield of interval surveillance imaging. Here we report the case of a recurrent intracranial aneurysm despite complete angiographic resolution after flow-diversion therapy with a Pipeline embolization device (PED). Given the absence of poor wall apposition, endoleak, and device migration, how this aneurysm recurred remains unclear, particularly given the recurrence was at a timepoint at which complete reendothelialization of the device would be expected. The patient ultimately underwent interval treatment with a second device placed across the neck of the aneurysm. Although rare, reports of aneurysmal recurrences support the use of interval non-invasive imaging surveillance to ensure successful embolization in this patient population.

Delayed Presentations and Worse Outcomes After Aneurysmal Subarachnoid Hemorrhage in the Early COVID-19 Era.

BACKGROUND: The early phase of the COVID-19 pandemic led to significant healthcare avoidance, perhaps explaining some of the excess reported deaths that exceeded known infections. The impact of the early COVID-19 era on aneurysmal subarachnoid hemorrhage (aSAH) care remains unclear. OBJECTIVE: To determine the impact of the early phase of the COVID-19 pandemic on latency to presentation, neurological complications, and clinical outcomes after aSAH. METHODS: We performed a retrospective cohort study from March 2, 2012, to June 30, 2021, of all patients with aSAH admitted to our center. The early COVID-19 era was defined as March 2, 2020, through June 30, 2020. The pre-COVID-19 era was defined as the same interval in 2012 to 2019. RESULTS: Among 499 patients with aSAH, 37 presented in the early COVID-19 era. Compared with the pre-COVID-19 era patients, patients presenting during this early phase of the pandemic were more likely to delay presentation after ictus (median, interquartile range; 1 [0-4] vs 0 [0-1] days, respectively, P < .001). Radiographic-delayed cerebral ischemia (29.7% vs 10.2%, P < .001) was more common in the early COVID-19 era. In adjusted analyses, presentation in the early COVID-19 era was independently associated with increased inhospital death or hospice disposition (adjusted odds ratio 3.29 [1.02-10.65], P = .046). Both latency and adverse outcomes returned to baseline in 2021. CONCLUSION: aSAH in the early COVID-19 era was associated with delayed presentation, neurological complications, and worse outcomes at our center. These data highlight how healthcare avoidance may have increased morbidity and mortality in non-COVID-19-related neurosurgical disease.

Bridging thrombolysis in atrial fibrillation stroke is associated with increased hemorrhagic complications without improved outcomes.

BACKGROUND: Atrial fibrillation (AF) associated ischemic stroke is associated with worse functional outcomes, less effective recanalization, and increased rates of hemorrhagic complications after intravenous thrombolysis (IVT). Conversely, AF is not associated with hemorrhagic complications or functional outcomes in patients undergoing mechanical thrombectomy (MT). This differential effect of MT and IVT in AF associated stroke raises the question of whether bridging thrombolysis increases hemorrhagic complications in AF patients undergoing MT. METHODS: This international cohort study of 22 comprehensive stroke centers analyzed patients with large vessel occlusion (LVO) undergoing MT between June 1, 2015 and December 31, 2020. Patients were divided into four groups based on comorbid AF and IVT exposure. Baseline patient characteristics, complications, and outcomes were reported and compared. RESULTS: 6461 patients underwent MT for LVO. 2311 (35.8%) patients had comorbid AF. In non-AF patients, bridging therapy improved the odds of good 90 day functional outcomes (adjusted OR (aOR) 1.29, 95% CI 1.03 to 1.60, p=0.025) and did not increase hemorrhagic complications. In AF patients, bridging therapy led to significant increases in symptomatic intracranial hemorrhage and parenchymal hematoma type 2 (aOR 1.66, 1.07 to 2.57, p=0.024) without any benefit in 90 day functional outcomes. Similar findings were noted in a separate propensity score analysis. CONCLUSION: In this large thrombectomy registry, AF patients exposed to IVT before MT had increased hemorrhagic complications without improved functional outcomes, in contrast with non-AF patients. Prospective trials are warranted to assess whether AF patients represent a subgroup of LVO patients who may benefit from a direct to thrombectomy approach at thrombectomy capable centers.

Diffuse Angiogram-Negative Subarachnoid Hemorrhage is Associated with an Intermediate Clinical Course.

BACKGROUND: The cerebral angiography result is negative for an underlying vascular lesion in 15-20% of patients with nontraumatic subarachnoid hemorrhage (SAH). Patients with angiogram-negative SAH include those with perimesencephalic SAH and diffuse SAH. Consensus suggests that perimesencephalic SAH confers a more favorable prognosis than diffuse SAH. Limited data exist to contextualize the clinical course and prognosis of diffuse SAH in relation to aneurysmal SAH in terms of critical care complications, neurologic complications, and functional outcomes. Here we compare the clinical course and functional outcomes of patients with perimesencephalic SAH, diffuse SAH, and aneurysmal SAH to better characterize the prognostic implications of each SAH subtype. METHODS: We conducted a retrospective cohort study that included all patients with nontraumatic SAH admitted to a tertiary care referral center between January 1, 2012, and December 31, 2017. Bleed patterns were radiographically adjudicated, and patients were assigned to three groups: perimesencephalic SAH, diffuse SAH, and aneurysmal SAH. Patient demographics, complications, and clinical outcomes were reported and compared. RESULTS: Eighty-six patients with perimesencephalic SAH, 174 with diffuse SAH, and 998 with aneurysmal SAH presented during the study period. Patients with aneurysmal SAH were significantly more likely to be female, White, and active smokers. There were no significant differences between patients with diffuse SAH and perimesencephalic SAH patterns. Critical care complications were compared across all three groups, with significant between-group differences in hypotension and shock (3.5% vs. 16.1% vs. 38.4% for perimesencephalic SAH vs. diffuse SAH vs. aneurysmal SAH, respectively; p < 0.01) and endotracheal intubation (0% vs. 26.4% vs. 48.8% for perimesencephalic SAH vs. diffuse SAH vs. aneurysmal SAH, respectively; p < 0.01). Similar trends were noted with long-term supportive care with tracheostomy and gastrostomy tubes and length of stay. Cerebrospinal fluid diversion was increasingly required across bleed types (9.3% vs. 54.6% vs. 76.3% for perimesencephalic SAH vs. diffuse SAH vs. aneurysmal SAH, respectively, p < 0.001). Vasospasm and delayed cerebral ischemia were comparable between perimesencephalic SAH and diffuse SAH but significantly lower than aneurysmal SAH. Patients with diffuse SAH had intermediate functional outcomes, with significant rates of nonhome discharge (23.0%) and poor functional status on discharge (26.4%), significantly higher than patients with perimesencephalic SAH and lower than patients with aneurysmal SAH. Diffuse SAH similarly conferred an intermediate rate of good functional outcomes at 1-6 months post discharge (92.3% vs. 78.6% vs. 47.3% for perimesencephalic SAH vs. diffuse SAH vs. aneurysmal SAH, respectively; p < 0.016). CONCLUSIONS: We confirm the consensus data that perimesencephalic SAH is associated with a more benign clinical course but demonstrate that diffuse SAH confers an intermediate prognosis, more malignant than perimesencephalic SAH but not as morbid as aneurysmal SAH. These results highlight the significant morbidity associated with diffuse SAH and emphasize need for vigilance in the acute care of these patients. These patients will likely benefit from continued high-acuity observation and potential support to avert significant risk of morbidity and neurologic compromise.

Diagnostic Performance of Computed Tomography Angiography and Computed Tomography Perfusion Tissue Time-to-Maximum in Vasospasm Following Aneurysmal Subarachnoid Hemorrhage.

Background Vasospasm is a treatable cause of deterioration following aneurysmal subarachnoid hemorrhage. Cerebral computed tomography perfusion mean transit times have been proposed as a predictor of vasospasm but suffer from well-known technical limitations. We evaluated fully automated, thresholded time-to-maxima of the tissue residue function (Tmax) for determination of vasospasm following aneurysmal subarachnoid hemorrhage. Methods and Results Retrospective analysis of 540 arterial segments from 36 encounters in 31 consecutive patients with aneurysmal subarachnoid hemorrhage undergoing computed tomography angiography (CTA), computed tomography perfusion, and digital subtraction angiography (DSA) within 24 hours. Tmax at 4, 6, 8, and 10 s was generated using RAPID (iSchemaView Inc., Menlo Park, CA). Dual-reader CTA and computed tomography perfusion interpretations were compared for patients with and without vasospasm on DSA (DSA+ and DSA-). Logistic regression models were developed using CTA and Tmax as input predictors and DSA vasospasm as outcome in adjusted and unadjusted models. Imaging studies from all 31 subjects (mean age 47.3±11.1, 77% female, 65% with single aneurysm with mean size of 6.0±2.9 mm) were included. Vasospasm was identified in 42 segments on DSA and 59 segments on CTA, with significant associations across individual vessel segments (P<0.001). In adjusted analyses, DSA vasospasm was associated with CTA (odds ratio [OR], 2.43; 95% CI, 0.94-6.32; P=0.068) as well as territory-specific Tmax>6 seconds delays (OR, 3.57; 95% CI, 1.36-9.35; P=0.009). Sensitivity/specificity for DSA vasospasm was 31%/91% for CTA, 26%/89% for Tmax>6 seconds, and 12%/99% for CTA+Tmax>6 seconds. Conclusions CTA and Tmax offer high specificity for presence of vasospasm; their utility, even in combination, as screening tests is, however, limited by poor sensitivity.

Investigation of white matter and grey matter alteration in the monkey brain following ischemic stroke by using diffusion tensor imaging.

Background: Investigation of stroke lesion has mostly focused on grey matter (GM) in previous studies and white matter (WM) degeneration during acute stroke is understudied. In the present study, monkeys were utilized to investigate the alterations of GM and WM in the brain following ischemic occlusion using diffusion tensor imaging (DTI). Methods: Permanent middle cerebral artery occlusion (pMCAO) was induced in rhesus monkeys (n=6) with an interventional approach. Serial DTI was conducted on a clinical 3T in the hyperacute phase (2-6 hours), 48, and 96 hours post occlusion. Regions of interest in GM and WM of lesion areas were selected for data analysis. Results: Mean diffusivity (MD), radial diffusivity (RD), and axial Diffusivity (AD) in WM decreased substantially during hyperacute stroke, as similar as those seen in GM. No obvious fractional anasotropy (FA) changes were seen in GM and WM during hyper acute phase. until 48 hours post stroke when significant fiber losses were oberved also. Pseudo-normalization of MD, AD, and RD was seen at 96 hours. Pathological changes of WM and GM were observed in ischemic areas at 8, 48, and 96 hours post stroke. Relative changes of MD, AD and RD of WM were correlated negatively with infarction volumes at 6 hours post stroke. Conclusion: The present study revealed the microstructural changes in gray matter and white matter of monkey brains during acute stroke by using DTI. The preliminary results suggest axial and radial diffusivity (AD and RD) may be sensitive surrogate markers to assess specific microstructural changes in white matter during hyper-acute stroke.

Noise-trained deep neural networks effectively predict human vision and its neural responses to challenging images.

Deep neural networks (DNNs) for object classification have been argued to provide the most promising model of the visual system, accompanied by claims that they have attained or even surpassed human-level performance. Here, we evaluated whether DNNs provide a viable model of human vision when tested with challenging noisy images of objects, sometimes presented at the very limits of visibility. We show that popular state-of-the-art DNNs perform in a qualitatively different manner than humans-they are unusually susceptible to spatially uncorrelated white noise and less impaired by spatially correlated noise. We implemented a noise training procedure to determine whether noise-trained DNNs exhibit more robust responses that better match human behavioral and neural performance. We found that noise-trained DNNs provide a better qualitative match to human performance; moreover, they reliably predict human recognition thresholds on an image-by-image basis. Functional neuroimaging revealed that noise-trained DNNs provide a better correspondence to the pattern-specific neural representations found in both early visual areas and high-level object areas. A layer-specific analysis of the DNNs indicated that noise training led to broad-ranging modifications throughout the network, with greater benefits of noise robustness accruing in progressively higher layers. Our findings demonstrate that noise-trained DNNs provide a viable model to account for human behavioral and neural responses to objects in challenging noisy viewing conditions. Further, they suggest that robustness to noise may be acquired through a process of visual learning.

Convolutional neural networks trained with a developmental sequence of blurry to clear images reveal core differences between face and object processing.

Although convolutional neural networks (CNNs) provide a promising model for understanding human vision, most CNNs lack robustness to challenging viewing conditions, such as image blur, whereas human vision is much more reliable. Might robustness to blur be attributable to vision during infancy, given that acuity is initially poor but improves considerably over the first several months of life? Here, we evaluated the potential consequences of such early experiences by training CNN models on face and object recognition tasks while gradually reducing the amount of blur applied to the training images. For CNNs trained on blurry to clear faces, we observed sustained robustness to blur, consistent with a recent report by Vogelsang and colleagues (2018). By contrast, CNNs trained with blurry to clear objects failed to retain robustness to blur. Further analyses revealed that the spatial frequency tuning of the two CNNs was profoundly different. The blurry to clear face-trained network successfully retained a preference for low spatial frequencies, whereas the blurry to clear object-trained CNN exhibited a progressive shift toward higher spatial frequencies. Our findings provide novel computational evidence showing how face recognition, unlike object recognition, allows for more holistic processing. Moreover, our results suggest that blurry vision during infancy is insufficient to account for the robustness of adult vision to blurry objects.

Postmarket American Experience With Woven EndoBridge Device: Adjudicated Multicenter Case Series.

BACKGROUND: The Woven EndoBridge (WEB) device was granted premarket approval in the United States following results of the Woven EndoBridge Intrasaccular Therapy (WEB-IT) study. WEB-IT reported excellent adequate angiographic occlusion of treated aneurysms with a high safety profile. These results were achieved, however, in the context of a prospective study with strict inclusion criteria and rigorous training support. OBJECTIVE: To review early as-practiced clinical experience with the WEB device in the United States. METHODS: Retrospective review across 6 institutions identified 91 patients undergoing 92 treatment sessions for WEB device placement in treatment of 91 intracranial aneurysms. Details regarding demographics, aneurysm characteristics, treatment considerations, clinical outcomes, and aneurysm occlusion were obtained and analyzed in a multicenter database. Angiograms from the index procedure and follow-up studies were reviewed by a blinded and independent adjudicator. RESULTS: The middle cerebral, anterior communicating, and basilar artery complexes were the commonly treated locations. Eight patients presented with ruptured aneurysms. A mean of 1.2 devices were introduced per case. Technical failure without deployment of a WEB device occurred in 2% (2/92) of sessions. Complete aneurysm occlusion for patients with imaging follow-up was 49% (mean follow-up of 8 mo). Four aneurysms were retreated. 90% of patients had modified Rankin Scale ≤ 2 at last clinical follow-up with no mortalities. CONCLUSION: Immediate postmarket experience with the WEB device, newly introduced at American centers, confirms safe procedural use, but long-term efficacy remains unclear. Early challenges include accurate sizing and device selection.

Urokinase-type plasminogen activator promotes N-cadherin-mediated synaptic recovery in the ischemic brain.

Urokinase-type plasminogen activator (uPA) is a serine proteinase that catalyzes the generation of plasmin on the cell surface and activates cell signaling pathways that promote remodeling and repair. Neuronal cadherin (NCAD) is a transmembrane protein that in the mature brain mediates the formation of synaptic contacts in the II/III and V cortical layers. Our studies show that uPA is preferentially found in the II/III and V cortical laminae of the gyrencephalic cortex of the non-human primate. Furthermore, we found that in murine cerebral cortical neurons and induced pluripotent stem cell (iPSC)-derived neurons prepared from healthy human donors, most of this uPA is associated with pre-synaptic vesicles. Our in vivo experiments revealed that in both, the gyrencephalic cortex of the non-human primate and the lissecephalic murine brain, cerebral ischemia decreases the number of intact synaptic contacts and the expression of uPA and NCAD in a band of tissue surrounding the necrotic core. Additionally, our in vitro data show that uPA induces the synthesis of NCAD in cerebral cortical neurons, and in line with these observations, intravenous treatment with recombinant uPA three hours after the onset of cerebral ischemia induces NCAD-mediated repair of synaptic contacts in the area surrounding the necrotic core.