Community Research and Development Information Service - CORDIS

FP7

TARGETBRAIN Report Summary

Project ID: 279017
Funded under: FP7-HEALTH
Country: Sweden

Final Report Summary - TARGETBRAIN (Targeting Brain Inflammation For Improved Functional Recovery in Acute Neurodegenerative Disorders)

Executive Summary:
A summary description of project context and objectives

The “Call” for the pour project was focused on the cross talk between the brain and the immune system under pathological conditions and on the implications for therapy of neurodegenerative diseases. We decided to focus on an acute neurodegenerative disorder, ischemic stroke. In this disease, following a sudden insult, neurons rapidly die, but additional cellular loss can occur hours and days thereafter with apparently limited repair. This disease is desperate for a cure, and can serve as a far-reaching model for general neurodegenerative conditions.

Our main objectives, which were detailed in the work plans, were therefore as follows:
1. To study the temporal and spatial relationship of different types of CNS-resident and blood-borne infiltrating immune cells in stroke-induced brain damage and functional recovery.
2. To elucidate the role of inflammatory cells, astrocytes and their secreted molecules in the sequential control of the local milieu needed for supporting cell survival and regenerative processes, including formation of new neurons from endogenous and transplanted NSCs.
3. To investigate the ability of transplanted NSCs to modulate the inflammatory response and to affect the characteristics of the lesion and the course of recovery after stroke.
4. To develop an immune-based therapeutic protocol to promote functional recovery in stroke patients based on enhancement of cellular plasticity.

TargetBraIn, aimed to understand the role of neuroinflammation in neurodegenerative diseases. This was achieved by basic studies of brain inflammatory processes, thereby developing new therapeutic strategies to obtain improved functional recovery in stroke patients. The ultimate goal is to modulate the inflammatory responses in the stroke-damaged brain and to enhance cellular and functional plasticity. In accordance with the call, we aimed to unravel the role of neuroinflammation in the neurodegeneration caused by stroke, and to develop methodologies for targeting its different components for future development of neurodegenerative disease modifying therapeutics.

It is well established that brain inflammation contributes to the pathogenesis of acute and chronic neurodegenerative diseases, including stroke, but application of anti-inflammatory treatments in humans has been unsuccessful. This could reflect the heterogeneity of the cells that are involved in the inflammatory reaction and the diverse molecules that they produce, the different origin of the cells (resident versus infiltrating from the blood), the length of time that elapsed since the injury, and the mechanism and route of entry of the infiltrating cells, which determine the cue molecules that they encounter on their way to the damaged area. Another issue that was overlooked for decades is the role of the systemic immune system in regulating the local inflammation, which alters with time after injury.

Our project aimed to provide three major advances compared to the current state-of-the-art: Firstly, to gain an in-depth understanding of the role of inflammation for neuroprotection, cell renewal and functional plasticity following an acute neurodegenerative event; secondly, to create the basis for a novel immunomodulatory therapeutic strategy for improved recovery after stroke, based on the possibility of fostering regenerative/repair mechanisms sustained by the synergistic activity of monocyte-derived macrophages, CNS-resident microglia, and exogenous or endogenous NSCs; and thirdly, to enhance our basic understanding of the role of immune-mediated processes in the damage occurring in neurodegenerative pathologies. Since this proposal was focused on the recovery phase, our findings could have a major impact on basic understanding of other neurodegenerative diseases, including chronic ones, and thereby on identifying novel therapeutic approaches to such diseases. Importantly, thus far, the therapeutic window for stroke was considered to be very narrow, primarily reflecting the emergence of the threatening self-compounds (ions, neurotransmitters, free radicals, etc). The immune resolution clearly leaves a wider therapeutic window.

As the working hypothesis we proposed that following ischemic stroke, the CNS-resident microglia and astrocytes are activated and locally harnessed to remove dead cells and cell debris and to buffer any neurotransmitter that exceeds its physiological levels. The astrocytes also demarcate the lesion site. Subsequently, circulating immune cells infiltrate from the blood; the activity of these cells, as well as cytokines and growth factors secreted as a consequence of their activation, are needed to support the survival of spared cells and to enhance regenerative processes. The infiltrating immune cells are needed for resolving/terminating the local response of the microglia, by acquiring a phenotype of alternatively activated cells; if the local response is not terminated on time, its effect becomes detrimental. Moreover, we proposed that under these conditions, NSCs not only differentiate to mature neurons but also contribute to immuno-modulation, modifying disease pathology and supporting functional recovery. The key remaining questions were concerning the spatial-temporal and functional relationships among these players, as well as the interactions with additional resident and infiltrating cells.

To address issues raised in our hypothesis we designed 6 Work Packages (WPs) covering of scientific and pre-clinical work and 2 more WPs for management and dissemination of the knowledge, respectively.

WP1: Involvement of blood monocytes in recovery after stroke
In this WP, we have aimed to study whether and how blood-borne monocytes contribute to CNS plasticity and recovery following lesion. We explored, when, and how circulating immune cells including blood-borne monocytes, infiltrate the brain following stroke and studied the mechanism by which infiltrating cells contribute to increased plasticity and regeneration. We determined how ablation of monocytes effects the local brain tissue environment and showed phenotype of monocyte-derived macrophages at the site of ischemia. We also demonstrated that the role of choroid plexus for the infiltration of monocytes to the lesioned CNS and studied whether monocyte delivery through CSF can promote functional recovery.

WP2: Role of microglia in cell genesis, plasticity and functional recovery after stroke.
The main objective of this WP was to explore how resident immune cells, microglia contributes to the production of new neurons in response to stroke and behavioral recovery. To explore the contribution of activated microglia to recovery after stroke and contribution to post-stroke neurogenesis. This WP was also designed to study the influence of activated microglia on properties of grafted NSCs, resident and NSC-generated nascent neurons and their functional integration in ischemically damaged brain.
WP3: Evaluation of ion channels in immune cells as potential targets for modulation of stroke-induced inflammation
In WP3 we planned to study ion channel expression in microglia and monocyte-derived macrophages of brain slices in animals exposed to stroke and explore the potential role of ion channels in regulating behavior of microglia- and monocyte-derived macrophages in stroke models.

WP4: Role of activated astrocytes in cell genesis and plasticity and functional recovery after stroke.
The goal of WP4 was to study the role of activated astrocytes for functional recovery after stroke and to evaluate effect of activated astrocytes on post-stroke neurogenesis. We also aimed to determine the mechanisms by which reactive gliosis inhibits functional recovery after stroke and how pharmacological attenuation of astrocyte activation can affect functional recovery after stroke.

WP5: Functional role of new cells for recovery after stroke and modulation of inflammation.
The main objective of this WP was to evaluate contribution of endogenous neurogenesis to recovery after stroke and to study the role of NSC transplantation for neuroregeneration after stroke. Moreover, we aimed to develop methods for non-invasive monitoring of the fate of grafted and endogenous NSCs when modulating the inflammatory response after stroke.

WP6: Development of pre-clinical protocol for novel regenerative therapy in stroke patients based advancing upstream knowledge, clinical relevance, commercial potential on immunomodulation
Overall objective of this WP was to translate the findings in other WPs efficiently to the clinical protocol. The specific objectives were to assess the pre-clinical safety pharmacology. and to develop the human cellular product.

WP7: Management of the consortium
The WP7 served to ensure that the consortium was managed and functioned properly and that all deliverables and reports (scientific and financial) were delivered as planned. Within this WP7 we arranged Steering Group and Consortium meetings, and discussion meetings between individual or group of partners.

WP8: Dissemination of knowledge, and intellectual property management
The overall objective was development of consortium Webpage and other dissemination tools to ensure proper dissemination of the results obtained by TargetBraIn. In addition, this WP also dealt with exploitation and IP issues.

The main Achievements and Discoveries of TargetBraIn by topics are as follows:

Role of monocyts for stroke rcovery and choroid plexus for monocyte infiltarion in CNS (advancing upstream knowledge)
We...
• ...demonstrated that bone marrow-derived monocytes contribute to the post-stroke long-term recovery most likely through switching ischemia-lesioned brain environment from pro-inflammatory to anti-inflammatory (Watananit et al., 2016, J Neuroscience).
• ...identified monocytes as prime mediators of tissue destruction and responsiveness to pro-inflammatory mediators (e.g. GM-CSF) (Croxford et al., 2015, Immunity).
• ...showed that entry of monocyte-derived macrophages to the CNS territory following injury does not necessitate breaching of the blood-brain-barrier; entry of monocytes is orchestrated through the choroid plexus epithelium within the blood-cerebrospinal fluid barrier (Shechter et al., 2013. Immunity; Kunis et al., 2013. Brain 136).
• ...demonstrated brain’s choroid plexus epithelium is immunologically suppressed in brain aging and neurodegenerative conditions (Baruch et al., 2014. Science, 346: 89-93. 6. Baruch et al., 2015. Nature Communication)
• ...discovered that systemic immune suppression can modify neurodegenerative diseases, and this could be is a new therapeutic target (Baruch et al., 2016. Nature Meicine).
• ...uncovered of a novel strategy to influence the inflammatory cells driving them improved neuronal survival and improved post-stroke functional recovery by brain-selective mRNA, miR-124 (Hamzei Taj et al., 2016, J Neuroimmune Pharmacol; Hamzei Taj et al., 2016, Biomaterials).

Contribution of microglia in neuroinflammation and stroke pathophysiology (advancing upstream knowledge)
We...
• ...characterized the genetic profile of microglia and tissue invading myeloid cells in stroke and tissue inflammation (Buttgereit et al., 2016, Nature Immunology).
• ...demonstrated that microglia clustering into germinal niches of the developing cortex is largely dependent on basal progenitors expressing Cxcl12 and that the mutual bidirectional relationship between these two cell types is vital to protect the developing cerebral cortex from unexpected damage impairing the proper proliferation and cortical positioning of neural precursors (Arno et al., 2015, Nature Communication).
• ...observed that reduction or complete ablation of microglial cells in the acute phase of cerebral ischemia does not impact on the early stroke outcome in terms of disability and lesion volume.
• ...discovered IL-34 being a unique brain-derived growth factor for microglia (Greter et al., 2012, Immunity).

Involvement of astrocytes in brain inflammation and post-stroke plasticity (advancing upstream knowledge)
We...
• ...demonstrated the role of astrocyte nanofilaments (intermediate filaments) in Notch-mediated control of neuronal differentiation of neural stem cells, and identification of astrocyte subpopulations with respect to their capacity to send/receive Notch signals (Wilhelmsson et al., 2012, Stem Cells; Lebkuechner et al., 2015, J. Neurochemistry)
• ...showed that the pro-inflammatory complement peptide C3a promotes survival of astrocytes after ischemic stress (Shinjyo et al., 2016, Mol Neurobiol).
• ...discovered that complement peptide C3a stimulates neural plasticity after ischemic stroke and intranasal treatment with C3a leads to faster and sustained functional recovery after ischemic stroke(Stokowska et al., 2016, Brain).
• ...demonstrated that overexpression of C3a in astrocytes reduces brain tissue loss due to hypoxic-ischemic injury to the immature brain and intranasal treatment with C3a ameliorates cognitive impairment after hypoxic-ischemic brain injury (Järlestedt et al., 2013, FASEB J; Moran et al., 2017, Exp Neurology).

Ion channels as specific targets for the immune cells in CNS (advancing upstream knowledge)
We...
• ...identified microglial ion channels regulating cytokine and chemokine release in an in situ stroke model (Charolidi et al., 2015, PloS One) and channels priming of reactive oxygen species production in neuroinflammatory condition mice (Spencer et al., 2016, PloS One).
• ...described ion channels expressed by microglia in the brain of adult, aged and ischemia-exposed mice (Schilling et al., 2015, Glia) and ion channels regulating proliferation and polarization of macrophages (Schilling et al., 2014, J Cell Sci).
• ...described specific differences in the expression of ion channels between acutely isolated monocytes and monocyte-derived macrophages infiltrating stroke-damaged brain.

Imaging of neuroiflammation and immune cell interaction with post-stroke neurogenesis (advancing upstream knowledge)
We
• ...defined using in vivo imaging methods optimal stem cell graft survival for cell replacement therapy after stroke including lack of effect on survival by proximity to the ischemic target zone and demonstrated that negative inflammatory response to cell survival is predominantly due to damage by the injection process (Boehm-Sturm et al., 2014, Biomaterials; Tennstaedt et al., 2015, PLoS One).
• ...demonstrated by using cell specific imaging reporters under in vivo conditions the time profile of various phases of early neuronal differentiation all the way to fully mature neurons exhibiting synaptogenesis, the prerequisite phase for tissue integration (Tennstaedt et al., 2015, Biomaterials).
• ...showed increased neurogenic response in aged animals subjected to stroke compensates for age-dependent decrease of basal neurogenesis (Adamczak et al., 2017, Neurobiol Dis).
• ...provided the first evidence that adult neural stem and progenitor cells located within the subventricular zone exert an ‘innate’ homeostatic regulatory role by protecting striatal neurons from glutamate-mediated excitotoxicity in stroke (Butti et al., 2012, Brain).

Transplantation of stem and reprogrammed cells in stroke-injured brain and interaction with neuroinflammation (advancing upstream knowledge)
We...
• ...uncovered a novel strategy to influence the inflammatory cells driving them improved neuronal survival and improved post-stroke functional recovery by brain-selective mRNA, miR-124 (Hamzei Taj et al., 2016, J Neuroimmune Pharmacol; Hamzei Taj et al., 2016, Biomaterials).
• ...demonstrated that functional neurons with genetic and morphological properties of cortical pyramidal projection neurons could be generated from reprogrammed human somatic cells and they can integrate in host rodent brain network responding to sensory stimulation through functional synapses (Tornero et al., 2013, Brain; Tornero et al., 2017, Brain).
• ...showed that neurons generated from human fibroblast-derived iPS cells could exhibit their recovery-promoting action through modulation of inflammatory environment (Tatarashvili et al., 2014, Restroarative Neurology & Neuroscience).
• ...showed that neural precursor cell transplantation, by modulating the excitatory-inhibitory balance and stroke microenvironment, is a promising therapy to ameliorate disability, to promote tissue recovery and plasticity processes after stroke (Bacigaluppi et al., 2016, J Neuroscience; Morini et al., 2015, Stem Cell Res. Ther).

Pre-clinical and translational data on use of immunecells as a target for CNS functional recovery (clinical relevance, commercial potential)
We...
• ...obtained proof-of-principle that CSF delivery of monocytes can improve impaired motor and sensory functions in rats subjected to cortical stroke.
• ...identified a new drug candidate (GEMST) acts against the inflammatory response by reducing or the number of stroke-activated CD45+ and CD11b+ cells at 21 days after GEMST treatment.
• ...identified and characterized the human sub-population of blood-derived monocytes that upon injection into the CSF effectively facilitates recovery following acute damage to the CNS.
• ...developed semi-automated and in a close system for manufacturing of the human monocytes.
• ... defined the set of release criteria of monocytes including purity, sterility and expression of specific surface markers, in compliance with the regulatory requirements for freshly administered cellular products.
• ...developed tests for assurance of biological activity and potency of isolated human monocytes.

Project Context and Objectives:
WP1: Involvement of blood monocytes in recovery after stroke
D 1.1.1. The report of establishing the “proof of concept” that circulating immature monocytes can improve the outcome of stroke

In order to explore the potential role of monocyte-derived macrophages (MDMs) for spontaneous functional recovery after stroke, we depleted circulating monocytes during the first week after the stroke induced by occlusion of middle cerebral artery, i. e., at a time when maximum monocyte infiltration takes place base on our previous data. We selectively depleted Ly6C+/CCR2+ monocytes from peripheral blood using the anti-CCR2 antibody MC-21 (Mack et al., 2001; Shechter et al., 2009). The CCR2 receptor is the binding site for the CCL2 ligand (also known as monocyte chemoattractant protein-1, MCP-1), which mediates monocyte recruitment from the bone marrow to the circulation. Also, the CCR2+ cells are the ones recruited to injured tissues outside (Bose and Cho, 2013; Peters and Charo, 2001; Peters et al., 2001; Wetzler et al., 2000) as well as to the CNS (Prinz and Priller, 2010; Saederup et al., 2010; Shechter et al., 2009; Yan et al., 2007). Animals were subjected to stroke and injected with MC-21 antibody the same day as well as 1, 2 and 3 days after the insult. Sham-treated and stroke-subjected mice were injected with vehicle and served as controls. Blood samples collected at 4 days after stroke (1 day after the last injection of MC-21) revealed nearly complete loss of circulating monocytes in all MC-21-treated animals (Fig. 1). At 10 days, the CCR2+ monocytes in peripheral blood had started to rebound and at 14 days had returned to normal level. Thus, the MC-21 antibody that we administered efficiently depleted circulating monocytes during the first week after stroke, as previously observed in other models of CNS insults (Shechter et al., 2009).

Fig.1. MC-21 antibody efficiently depletes circulating monocytes. Numbers of circulating CCR2+ monocytes in sham-operated, and in stroke-subjected mice, injected either with vehicle or with MC-21 antibody. Data are means ± SEM; *, p < 0.05, one-way ANOVA.

Based on these findings, we carried out behavioral tests to assess how depletion of monocytes during the first week after stroke would affect the long-term functional recovery. All animals were subjected to corridor (1 week before and 1, 3, 7, and 11 weeks after stroke) and staircase tests (1 week before and 1, 3, and 7 weeks after stroke). Sham-treated animals showed normal behavior in corridor and staircase tests. We observed impairments in the corridor test on the contralateral side from 1 to 7 weeks after stroke in both vehicle- and MC-21-injected animals (Fig. 2A).

Fig. 2. Depletion of circulating CCR2+ monocytes impairs long-term spontaneous behavioral recovery after stroke. Comparison between sham-treated and vehicle-injected (“sham”, n =10), stroke-subjected and vehicle-injected (“vehicle”, n=9), and stroke-subjected and MC-21-injected (“MC-21”, n=10) in performance in corridor (A) and staircase (B, C) tests. Performance in the corridor test was calculated by dividing the number of contralateral retrievals with total number of retrievals from both sides. Performance in staircase test was calculated as the number of retrieved or eaten pellets on the impaired side divided by the total number of pellets on both sides, and expressed as percentage of performance at baseline. Repeated measures analysis of variance (ANOVA); *, p < 0.05.

Interestingly, at 11 weeks, we found spontaneous behavioral recovery, the test performance reaching control level, in vehicle-injected mice. In contrast, in MC-21-treated animals, the impairment was maintained at the same level as at earlier time points (Fig. 2A). In the staircase test, both vehicle- and MC-21-injected mice showed similar impairment in number of retrieved pellets on the contralateral side at 1 week after stroke. At 3 weeks, the performance of the vehicle-injected mice did not differ from that of the sham-operated animals, whereas MC-21-treated mice still showed impairment. At 7 weeks, the number of retrieved pellets was similar in all three groups. These findings suggest a delayed recovery in this task in the monocyte-depleted group. Similarly, when behavior in the staircase test was assessed by number of eaten pellets, we found that the vehicle-injected group had recovered at 7 weeks, whereas the MC-21-injected animals remained impaired (Fig. 2C). To rule out the possibility that the worsening of long-term behavioral recovery in MC-21-injected mice was an outcome of more extensive ischemic injury, we analyzed the lesion site by immunohistochemistry.
These data clearly demonstrated that depletion of circulating monocytes during the first week after stroke and the resulting decrease of MDMs at the site of the injury caused impaired recovery of sensorimotor deficits in the chronic phase following MCAO. These data represent proof of principle that we circulating monocytes after infiltration in the brain lesioned by stroke participate in plasticity and regeneration of the brain and can improve functional recovery after stroke.

D 1.2.1. The report on the role of infiltrating blood-borne monocytes in neuroprotection following ischemic stroke.

We have explored whether the bone-marrow-derived monocytes are neuroprotective in the brain after the induction of stroke. Animals (mice) were subjected to focal cerebral ischemia by transient occlusion of the middle cerebral artery (MCA). The stroke model was induced in anaesthetized c57/B mice by insertion of a surgical filament into the right carotid artery until the origin of MCAO. The mice were re- anaesthetized after 45 min of occlusion and the filament was removed. Animals were sacrificed at 18 weeks post-injury and immunocytochemistry with neuron-specific antibody NeuN was used to detect the injury site and characterize the extent of damage.
In order to evaluate the contribution of infiltrating monocytes on the extent of the ischemic lesion the level of the circulating monocytes was reduced by injected one group of mice (n=10) intraperitoneally with MC-21 (anti-CCR2 antibody) starting on the day of the injury, and throughout the first week of recovery (day 0, 2, 4, and 6 post-injury). This treatment has been previously shown to reduce extremely efficiently the number of circulating monocyte with the capacity to infiltrate into nervous system (D 1.1.1). The animals were sacrificed at 18 weeks after the stroke and brains were processed for histochemical staining and evaluation of the extent of lesion (Fig.1). Stereological assessment of brain sections and statistical analysis revealed that neither pattern nor the volume of the brain tissue loss is affected by the depletion of monocyte-derived macrophages. This in turn leads to the suggestion that the effect of endogenous monocytes on functional recovery after stroke observed in other deliverables (D1.1.1) is most likely mediated by the contribution of monocytes to the brain plasticity rather than neuroprotection against ischemic lesion.

Fig.1. Depletion of circulating CCR2+ monocytes does not affect the pattern and extent of neuronal loss after stroke
A, Location and pattern of ischemic injury, mainly confined to lateral and dorsolateral parts of striatum, shown by NeuN-staining in brain sections from stroke-subjected mice, treated with vehicle or MC-21, at 18 weeks after insult. Inserts are enlargements from respective coronal sections. Scale bar= 1 mm. B, Mean volume of ischemic lesion treated with vehicle (n=9) or MC-21 (n=10), at 18 weeks after insult. Data are means±SEM; *, p<0,05, unpaired t-test.

In another series of experiments we investigated whether intraventricular injection of bone marrow-derived CD115+ monocytes primed towards M2 phenotype could affect the extent of ischemic lesion in mice subjected to stroke. Mice subjected to cortical lesion were randomly allocated to cell transplantation group or PBS injection control group. In vitro derived M2 macrophages were injected 24 hours after cortical stroke in mice. At 7 days after the injection, mice brains were examined for ischemic damage. Cell transplanted group and PBS injected control group showed no difference in infarct volume (Fig.2).

Fig.2. Transplantation of M2-primed monocytes does not affect the extent of neuronal loss after stroke. Mean volume of ischemic lesion treated with vehicle (n=5) or M2 primed monocytes (n=6), at 1 week after insult. Data are means±SEM; unpaired t-test.

We have shown that the same pattern of monocyte transplantation improves behavioural recovery after stroke. This data suggests that recovery-promoting effect of M2 macrophage transplantation affects brain plasticity has no effect on ischemic damage.

D 1.3.1. Report on the influence of monocyte-derived macrophages on microglial profile and astrocytic scar formation.

We aimed to understand interaction between monocyte-derived macrophages (MDMs) and resident microglia. First, we generated chimeric animals through whole body irradiation with10Gy with head protection and transplantation of bone marrow cells from CX3CR1-EGFP donors. We analyzed the ratio between MDM and microglia in chimeric animals at days 3 and 7 after stroke. At 3 days, when the number of MDMs was at its maximum, they represented 61.6% of the whole CD45+/CD11b+ cell population, whereas microglia constituted 24.6% (Fig. 1A). Conversely, at 7 days after stroke microglia had become the dominant population, representing 66.9% and MDMs only 12.6%. The phenotypic characterization of microglia based on flow cytometry analysis of Ly6C and CX3CR1 expression revealed (Fig. 1B) that at both time points, these cells had predominantly pro-inflammatory activity (65.2% and 88.2% at 3 and 7 days, respectively).
Fig.1. A. Estimation of the percentage of microglia and MDMs in injured hemisphere of mice subjected to stroke and sacrificed 3 and 7 days thereafter. B, Estimation of the percentage of microglia with pro- and anti-inflammatory phenotype in injured hemisphere of mice subjected to stroke and sacrificed 3 and 7 days thereafter. Data are means±SEM; unpaired t-test. Data are means±SEM; unpaired t-test.

In order to explore how the number of infiltrating monocytes influence the number of resident microglia microglia, we administered 4 million monocytes from CD45.2 mice to stroke-subjected CD45.1 mice. Microglia and infiltrating, endogenously recruited or grafted MDMs were then distinguished based on levels of CD45, being low and high, respectively (Sedgwick et al., 1991; Shechter et al., 2013). In the contralateral, intact hemisphere (Fig. 2A), the MDMs (CD45.1high) constituted about 3 % of all CD11b+ cells but there were virtually no grafted CD45.2+ MDMs.

Fig. 2. Flow cytometry analysis of brain hemispheres (contra- (A) and ipsilateral (B) to the lesion) from CD45.1 mice subjected to MCAO and injected intravenously with 4 million monocytes from CD45.2 mice on the day after the insult and killed 2 days later. Note the presence of high numbers of grafted CD45.2high/ CD11bhigh and endogenous CD45.1high/ CD11bhigh monocytes ipsilateral to the ischemic lesion. The CD45.1low/ CD11bhigh cells are microglia.

The vast majority (85.1%) of CD11b+ cells were resident CD45.1low microglia. In the ipsilateral hemisphere, we detected a small population (0.69% of all CD45high cells) of CD45.2high grafted MDMs (Fig. 2B). However, they were extremely few as compared to the infiltrated, endogenously recruited CD 45.1high MDMs, which represented about half of all CD11b+ cells. Taken together, our data show that within 2 days after transplantation the grafted monocytes home to the stroke-injured hemisphere, where they represent only a very small fraction of CD11b+ microglia and are much fewer than endogenously derived monocytes.

In separate series of experiments, we studied how the ablation of MDM in early phases after stroke affect the number of microglia/macrophages in stroke lesioned brain. Animals were subjected to stroke and injected with MC-21 antibody the same day as well as 1, 2 and 3 days after the insult. Sham-treated and stroke-subjected mice were injected with vehicle and served as controls. Animals were sacrificed at 1 week after stroke and brain sections were processed for immunocytochemistry with antibodies for resident (non-activated) microglia/macrophages and activated macroglia/macrophages (ED1). No differences in the striatum between MC21 and sham-treated mice in the number of Iba1+ microglia or (Fig. 3A), Iba1+/ED1+ activated microglia/macrophages (Fig. 3B). We also did not detect any differences in the and in the % Iba1+/ED1+ over all Iba1+ cells (Fig. 3C).
Fig.3. Number of Iba1+ (A), ED1+/Iba1+, and % ED1+/Iba1+ of Iba1+ microglia/ macrophages in injured brain tissue of stroke-subjected, MC-21- and vehicle-injected rats. Data are means±SEM; unpaired t-test.

In both groups there is a significant increase of the number of microglia and activated microglia/ macrophages in the ipsilateral side compare to the contralateral side.

Next, we examined how depletion of MDMs effect the number of activated astrocytes. Since it is impossible to count the number of individual GFAP+ astrocytes due to cytoplasmic staining pattern and overlapping of individual cell, GFAP immunoreactivity was assessed in the astrocytes in regions which covered lesioned striatim and surrounding area including astrocytic scar and the corresponding areas of vehicle-injected animals. To quantify GFAP immunoreactivity, images of different regions of interest (ROIs) in two representative sections from each brain doublestained for GFAP and NeuN (to visualize lesioned area) were acquired. The area of GFAP-immunoreactivity ROI of MC-21-injected animals and corresponding vehicle-injected regions was determined by image analysis using Cellsens Dimension 2010 software (Olympus). In each section, areas of immunoreactivity were identified using defined representative ranges of threshold for specific signal. Using these defined parameters, the images of each ROI were analyzed by software, which calculated the total area covered by pixels/specific immunopositive signal. The values corresponding to total fluorescence areas were averaged and expressed as the mean GFAP+ area per animal. Quantification of obtained data showed significant increase in GFAP coverage in the ipsilateral compared to contralateral side in both MC-21 and saline treated groups, but values do not differ based on treatment (Fig.4.).

Fig.4. Area of GFAP immunoreactivity in the region of interest (n=4, in 2 representative sections) in stroke-damaged mouse brain after MC-21 (n=5) or saline (n=4). Student’s unpaired t test, *, p < .05; mean±SEM.

We have also performed measurement of angiogenesis using immunocytochemistry for the assessment of the area covered by CD31+ blood vessels the sham and MC-21 treated stroke-subjected animals. However, microscopic and morphometric analysis of the data did not reveal any differences in angiogenesis or astrocytosis caused by the depletion of monocyte-derived macrophages during first week after stroke (data not shown).

There exists a significantly large population of glial cells in the mammalian CNS that can be identified by the expression of the NG2 proteoglycan. Cells that express NG2 (NG2 cells) are found in the developing and mature CNS and are distinct from neurons, astrocytes, microglia, and mature oligodendrocytes. They are often referred to as oligodendrocyte progenitor cells because of their ability to differentiate into oligodendrocytes in culture. However, the observation that a large number of NG2 cells persist uniformly and ubiquitously in the adult CNS and display a differentiated morphology is not entirely consistent with the notion that NG2 cells are all oligodendrocyte progenitor cells. Therefore, we also the number of NG2+ cells in animals subjected to stroke ands injected with MC-21 antibody or saline. Our quantification revealed that that ablation of infiltrating monocytes leads to is significant reduction of NG2+ cells in the striatum ipsilateral to lesion as compared to saline injected control animals (Fig.5).

Fig.5. Number of NG2 immunoreactive cells in the region of interest (n=4, in 2 representative sections) in stroke-damaged mouse brain after MC-21 (n=5) or saline (n=4). Student’s unpaired t test, *, p < .05; mean±SEM.

In conclusion, our experimental data indicate that infiltration of bone marrow-derived monocytes in stroke-lesioned brain does not influence the number of resting or activated microglia/macrophages or the degree of the activation of astrocytes. However, monocytes influence the number of NG2+ cells which could participate in cellular plasticity and regeneration process spontaneously occurring in stroke-damaged brain. These findings are novel and instrumental in explaining the role of monocytes in the functional recovery after stroke-induced ischemic lesion of the brain.

In our originally submitted grant proposal, based on our previous studies using the spinal cord injury model (Shechter et al., PloS Medicine, 2009), we aimed to test in an ischemic stroke model, whether monocyte-derived macrophages contribute to repair and if so, whether their contribution is dependent on expression of IL-10 and TGFβ. It soon became apparent that the rate limiting step in recovery is the number of recruited cells, and therefore we shifted our focus towards the mechanism that is responsible for the limited recruitment, rather than the activity of these cells following entry. To this end, we used the well calibrated model of spinal cord injury, in which we have an established model to evaluate the contribution of recruited cells to repair. The advantage with the spinal cord injury model is that it can anatomically disassociate the site of primary injury and the site of entry of monocytes. Retrospectively, it turned out that this was a good choice, since we discovered that the entry of recruited immune cells is through the brain-cerebrospinal fluid barrier (Shechter et al, Immunity, 2013; Kunis et al., Brain, 2013).

D 1.4.1. Report on the influence of monocyte-derived macrophages on cell genesis

The role of myeloid cells in the functional recovery process following stroke is poorly understood. In hemorrhagic stroke, monocyte derived macrophages (MDMs) are needed to promote vascular healing. Yet, the involvement of MDMs in the repair processes beyond the acute phase, at the remodeling stage, and during long-term functional restoration is unknown. The objectives of the present study to explore the contribution of infiltrating MDMs to post-stroke neurogenesis.

We selectively depleted Ly6C+/CCR2+ monocytes from peripheral blood using the anti-CCR2 antibody MC-21. The CCR2 receptor is the binding site for the CCL2 ligand (also known as monocyte chemoattractant protein-1, MCP-1), which mediates monocyte recruitment from the bone marrow to the circulation. Also, the CCR2+ cells are the ones recruited to injured tissues outside as well as to the CNS. Production of anti-CCR2 monoclonal antibody MC-21 was performed as described previously. MC-21 was used to selectively deplete CD115+/CD11b+/Ly6C+ monocytes from the blood.
We explored whether monocytes regulate neural stem/progenitor cell (NSPC) proliferation in the SVZ after stroke. To study the influence of the MDMs in the post-stroke neurogenesis, C57BL6 mice were subjected to 30 min. MCAO. Half of the animals were injected with MC-21 antibody intraperitoneally starting immediately after MCAO and on the first 3 days of recovery (d0, d1, d2 and d3 post- injury). The second half of the mice were control animals and received only vehicle injection. BrdU was administered into animals at day 4, 5, and 6 after stroke to be able to trace for cell proliferation at that certain time point. Animals were sacrificed at days 7 and 14.
To reveal changes in cell cycle duration or cell division of NSPC in SVZ we analyzed BrdU incorporation in replicated DNA and DCX expression. We have previously shown that stroke caused by MCAO induces a transient increase of NSPC proliferation in SVZ (Arvidsson et al., 2002; Thored et al., 2006). To determine whether MDM influences NSPC proliferation in SVZ under pathological conditions, we compared NSPC proliferation and early survival of newly formed cells in SVZ of MDM depleted and control mice at the peak of the increased proliferation, 7 days after stroke.
We observed significant increase in BrDU+/DCX+ cells in the ipsilateral hemisphere of MC-21 treated group compared to saline group in the SVZ and in the striatum close to the SVZ (500 μm far from the SVZ)(Fig.1).

Fig.1. Number of DCX+/BrdU+ neuroblasts in the subventricular zone (left) and striatum (right) of animals subjected to stroke and injected either by saline or MC-21. Mean±SEM.

We did not detect any difference of the number of DCX+/BrdU+ cells in the olfactory bulbs (data not shown). However, we observed significant reduced number of BrDU+/DCX+ cells in the hippocampus in the ipsilateral hemisphere of MC-21 treated group compared to saline group (Fig.2).

Fig.2. Number of DCX+/BrdU+ neuroblasts in the hippocampus of animals subjected to stroke and injected either by saline or MC-21. Mean±SEM.

There were no differences in SVZ proliferation as judged by the number of ki67+ cells (data not shown).

D 1.5.1. Report on the phenotype/activity of monocyte-derived macrophage at the site of ischemia

To explore in detail the extent and dynamics of spontaneous monocyte infiltration to the stroke-injured brain, we euthanized mice at different time points after stroke, and analyzed by flow cytometry the myeloid cell composition in both hemispheres (Fig. 1A).
We found that the numbers of spontaneously infiltrating MDMs in the hemisphere ipsilateral to the insult reached a peak (more than 60-fold increase as compared to the hemisphere on contralateral side or in intact or sham-treated animals) at 3 days after stroke and then declined rapidly, still being higher at 7 days compared to contralateral side but reaching control values at 14 days (Fig. 1B). The numbers of MDMs in sham-operated and intact mice at 1 day after treatment (Fig. 2B) were similar to those following stroke in the ipsilateral hemisphere on the day of the injury, and in the contralateral hemisphere at all tested time points (Fig. 1A and B). We assumed that the majority of these cells were residual blood monocytes in brain capillaries. We stained the brains of sham-operated animals with CD31 as a marker for vessels, IB4 for activated macrophages, and Iba1 as microglia/macrophage marker (Fig. 1C).
In support of our contention, the IB4+ monocytes were clearly located within the blood vessel lumen while the Iba1+ microglia, exhibiting ramified resting phenotype, were found in the brain parenchyma.

Fig. 2. Spontaneous infiltration of circulating monocytes to sites of lesion peaks at 3 days after stroke. A, Examples of flow cytometry analysis of brain hemispheres (contra- and ipsilateral to lesion) of mice subjected to MCAO, identifying MDMs and microglia as CD45high/CD11bhigh and CD45low/CD11bhigh, respectively. B, Time course of numbers of MDMs based on flow cytometry analysis in hemispheres contra- and ipsilateral to MCAO or sham treatment and in control hemisphere. Data are means±SEM; *, p<0,05, paired t-test between contralateral and ipsilateral sides for each group. C, Fluorescence microscopic images of mouse brain showing CD31+ vessel, IB4+ activated monocytes, Iba1+ microglia, Hoechst+ nuclei in the striatum of sham-treated animal, and merged image. SSC - side scatter; FSC - forward scatter. Scale bar = 20 μm

In order to explore potential mechanisms underlying the recovery-promoting effect of MDMs, we first wanted to distinguish MDMs from activated microglia and determine their phenotype. Therefore, we generated chimeric mice by subjecting wild-type mice to bone marrow transplantation from CX3CR1-GFP mice. Notably, all mice were head-protected during the irradiation that precedes the bone marrow transplantation (Shechter et al., 2009). In CX3CR1-GFP mice, CX3CR1+ monocytes as well as microglia are GFP+ (Jung et al., 2000). However, in the chimeric animals only bone marrow-derived monocytes are GFP+, which allows identification of MDMs both with immunocytochemistry and flow cytometry (Mildner et al., 2007).
Immunocytochemistry of brain sections from chimeric mice showed that at 7 days after stroke the majority of GFP+ MDMs were distributed within the ischemically injured tissue with some of them being localized in close proximity to the lesion border (Fig. 2A-C). This was in line with our earlier experiments showing that already at 3 days after stroke, intravenously transplanted monocytes infiltrate the stroke-injured tissue. The lesion border was clearly delineated by GFAP+ activated astrocytes, most of which were found outside the lesion core (Fig. 2B, C). The vast majority of the MDMs showed IB4 immunoreactivity (Fig. 2D). In brain tissue from contralateral hemisphere or from sham-operated animals, virtually no GFP+ bone marrow-derived monocytes were detected.

Fig. 2. CX3CR1-GFP+ monocyte-derived macrophages infiltrate lesion site of chimeric mice subjected to ischemic stroke. A, B, Fluorescence microscopic images of mouse brain coronal sections showing distribution of GFP+ MDMs and GFAP+ astrocytes within and outside the ischemically injured tissue, respectively. Arrowheads depict lesion border. C, Enlargement of inset depicted in (B). Arrows point to individual GFP+ MDMs. D, Fluorescence microscopic image showing double-immunostaining of GFP+ MDMs (green) with activation marker IB4 (red). Majority of MDMs are immunopositive for IB4 (arrows). Scale bar = 250 μm for A and B, and 250 μm for C and D.

The flow cytometry analysis of tissue from the stroke-subjected mice confirmed that blood-born monocytes had efficiently infiltrated only the injured hemisphere (Fig. 3A). Two major sub-populations of MDMs have previously been described: the pro-inflammatory (Ly6Chigh) population and the alternatively-activated anti-inflammatory population (Ly6Clow) (Gordon and Taylor, 2005). We separated infiltrating MDMs and resident microglia based on CD45 and CD11b expression, and then we further separated the MDM population based on expression of CXCR1 and Ly6C. We observed that the two sub-populations of MDMs in the brain after stroke, namely Ly6Chigh/CX3CR1low and Ly6Clow/CX3CR1int, underwent dramatic changes in the ipsilateral hemisphere: at the early time point (day 3 after injury), the relative percentage of the Ly6Chigh/CX3CR1low MDM subpopulation was high, but was then significantly reduced by day 7. In contrast, the relative percentage of the Ly6Clow/CX3CR1int population remained relatively unchanged (Fig. 3A). We further analyzed the phenotype of MDMs in intact and stroke-injured hemispheres at different time points after the insult (Fig. 3B). In intact animals, more than 70% of monocytes were Ly6Clow/CX3CR1int and only 5 % were Ly6Chigh/CX3CR1low (Fig. 3B). Most of the monocytes in the intact brain are those that have not been washed out from capillaries during the perfusion of animals prior to brain excision (see Fig. 2C). At day 1 after stroke, approximately 40% of the MDMs were Ly6Chigh/CX3CR1low, reaching the peak at day 3 (53%). Their percentage then gradually decreased to 18.3 % at day 21. At day 3 after stroke, the percentage of Ly6Clow/CX3CR1int MDMs was 19 %, and their percentage increased at day 7 (34.2%) and then remained stable at days 14 and 21 (40.1 and 38.7%, respectively). Taken together, our data show that the relative ratio between the two sub-populations of MDMs was shifted towards the pro-inflammatory phenotype at day 3 and towards the anti-inflammatory phenotype thereafter (Fig. 3B).

Fig. 3. Monocyte-derived macrophages switch from pro- to anti-inflammatory phenotype during first weeks after stroke. A, Flow cytometry analysis of brain hemisphere ipsilateral to lesion in mice subjected to stroke and sacrificed 3 and 7 days thereafter. CD45/CD11b immunoreactivity is used to distinguish MDMs and microglia, and CX3CR1/Ly6C to define pro- and anti-inflammatory phenotype of MDMs. B, Changes as a function of time in percentage of MDMs with pro- and anti-inflammatory phenotype defined by flow cytometry analysis in the ischemically injured brains; intact hemispheres were used as controls. C, Estimation of the percentage of CD204+, CD206+ and Declin+ cells within MDM.

We then asked whether the increase of the number of MDMs with Ly6Clow/CX3CR1int phenotype from day 3 to day 7 after stroke was accompanied by corresponding changes in anti-inflammatory characteristics. The Ly6Chigh/CX3CR1low and Ly6Clow/CX3CR1int populations were analysed by flow cytometry for the expression of characteristic markers of anti-inflammatory activity, CD206, Dectin-1 and CD204. The percentage of cells expressing CD206 and Dectin-1 was increased within the Ly6Clow/ CX3CR1int but not within the Ly6Chigh/CX3CR1low sub-population (Fig. 3C). The percentage of cells immunoreactive for CD204+, known as macrophage scavenger receptor 1, did not change in either population (Fig. 3C).

At the lesion site, MDMs with two distinct phenotypes were found, first pro-inflammatory and subsequently anti-inflammatory. The transition of MDMs from pro-inflammatory to anti-inflammatory bias during the first three weeks post-ischemia leads to modulation of the inflammatory tissue environment and is associated with improved functional outcome. This new insight could have important therapeutic implications by raising the possibility that inadequate recruitment of MDMs after stroke underlies the incomplete functional recovery seen in both animals and patients. Future studies will show whether this physiological repair mechanism can be potentiated by increasing the homing of macrophages, derived from endogenous or grafted monocytes in the peripheral blood, to the ischemically injured brain.

D 1.6.1. Characterization of monocyte dynamics in the ischemic brain using imaging data

The goal was to image the temporal profile of monocyte infiltration into the ischemic brain after systemic injection of labeled monocytes. Bioluminescence imaging was chosen for vitality monitoring and MRI for high resolution 3D tracking in the mouse brain. For this purpose, the monocytes were isolated from bone marrow of syngeneic mice. Monocytes seeded in culture were incubated with iron-oxide nanoparticles overnight to efficiently incorporate the SPIO particles for MRI detection. The same process was repeated with monocytes isolated from β-actin-luc mice, constitutively expressing the firefly luciferase for bioluminescence imaging (BLI). SPIO uptake efficiency was determined by titration of concentration in culture and 168 um Fe/ml was found to be optimal for cell tolerance of labeling and detection threshold. No difference between WT and β-actin-luc monocytes was observed. In vitro detection limits were 2 500 cells with BLI and 5 000 cells with MRI. In vivo implantation into the striatum of mouse brains allowed observation of 1 500 cells by MRI and permitted the longtime monitoring over approximately 2 weeks. BLI signal of such implantations showed a substantial loss of cell viability during the first two weeks, reducing the available cell number to 20-30% after two weeks.

For the monitoring of monocytes after stroke, stroke was first induced with the intraluminal filament occlusion technique, occluding the middle cerebral artery for 30 min, followed by reperfusion after withdrawal of the filament. Successful MCA occlusion and characterization of lesion extent and severity was assessed by MRI on the day of occlusion. One day later, 2.5 Million monocytes were injected through the tail vein. BLI was performed to detect cells with high sensitivity and to assess cell vitality, while high resolution MRI was performed to localize cells in specific brain regions. These experiments demonstrated by BLI screening that most cells get trapped in the lungs, thus being lost for arrival in the target are of the ischemic brain territory. The small fraction of monocytes reaching the target zone were detected by MRI in the ischemic territory 24 hours after iv injection. Small dark spots in the hyperintense T2-weighted MRI images showed with high sensitivity the small numbers of cells, spread across the whole stroke territory. However, further 24 hours later, these hypointense spots were no longer detectable.
In conclusion, the imaging based approach is highly sensitive but the rapid dynamics of the monocytes makes it difficult to monitor their wash-in and distribution across the brain using a reasonable temporal resolution of repetitive imaging experimentation. The experimental condition will be substantially improved with injection of increased number of cells for sensitive, longer observation periods in the target zone of the ischemic territory.

D 1.7.1. Report on treatment of ischemic stroke based on passive transfer of bone marrow-derived monocytes to the CSF

Inflammation following ischemic stroke can have both detrimental and beneficial effects. Microglia and blood-borne macrophages can exacerbate ischemic damage by releasing ROS and various pro-inflammatory cytokines (Amantea et al., 2009; Block et al., 2007). On the other hand, recent work found that blood-borne macrophages are indispensable for post-stroke recovery (Wattananit et al., 2016). Macrophages can exit in two states: pro-inflammatory M1 macrophages and anti-inflammatory M2 macrophages (Mantovani et al., 2013). It has been shown that choroid plexus serves as a gateway for M2 macrophages infiltration into CNS mediated by VCAM1 as well as NT5E, and these M2 macrophages promote post-injury recovery in SCI model (Shechter et al., 2013). Whether choroid plexus act in a similar way after stroke needs to be examined. Infiltration of monocytes through choroid plexus after stroke might be occurring in parallel to BBB. If this is the case, for therapeutic purposes, injection of monocytes in CSF might be alternative and more safe way of cell delivery as compared to systemic injection. Therefore, the primary study was carried out to investigate possible infiltration of monocytes through choroid plexus after stroke.
First, we investigated whether choroid plexus is activated after stroke. We used mice cortical stroke model, where the lesion site is physically far away from choroid plexus in the ventricles. We used six probes that were shown to be up-regulated in choroid plexus after spinal cord injury as indicators for choroid plexus activation: Vcam1, Madcam1, Ifnγ, Nt5e, Cx3cl1, and Ccl2 (Fig.1). At 6 hours after stroke, none of the examined factors showed significant change in combined lateral and ¾ ventricle choroid plexus samples (data not shown).
Fig. 1. Expression of genes related to choroid plexus activation.

In choroid plexus samples, right ventricle choroid plexus close to the lesion showed up-regulation of Nt5e. At 24 hours after stroke, Vcam1, Madcam1 and Cx3cl1 showed up-regulation, indicating activation of choroid plexus possibly for leukocyte infiltration. Up-regulation of Cx3cl1 remained at 3 days after stroke. At 7 days after stroke, Nt5e was upregulated. At 2 weeks after stroke, Cx3cl1 and Nt5e showed downregulation. Taken together, these data showed that though physically remote from cortical lesion area, choroid plexus senses certain factors triggered by cortical lesion and up-regulate various inflammatory factors. Our data clearly indicate that this response occurs mainly at the first 3 days after stroke.
We next explored whether activation of choroid plexus could lead to increased macrophages infiltration in choroid plexus. Consistent with up-regulation of Vcam1, Madcam1 and Cx3cl1 in choroid plexus at 24 hours and Cx3cl1 at 3 days after stroke, the number of CD45+Ly6C+ macrophage in choroid plexus increased at 3 days after stroke (Fig. 2). The percentage of CD45+Ly6C+ macrophages in total CD45+ leukocytes was also increased (Fig. 2). At 7 days after stroke, the infiltration of CD45+Ly6C+ macrophage in choroid plexus returned to baseline in naïve condition.
Fig. 2. Number of leukocytes and macrophages in the choroid plexus of naïve and stroke-subjected animals.

In choroid plexus, about 10% of all macrophages expressed M2 marker CD206 at 3 days after stroke. Taken together, these data indicated that recruitment of macrophages in choroid plexus increased at 3 days after stroke, and some of these macrophages were skewed towards M2 macrophages.

Fig. 3. Images of Iba1+ cells expressing M2 marker CD206

To examine whether macrophages in the choroid plexus could migrate into CSF, we collected CSF from cisterna magna where CSF just flows out from the ventricular system. We then did a correlation between the number of CD45+ leukocytes in CSF and the percentage of CD45+ leukocytes in choroid plexus (Naïve n=2, Stroke n=4). We found a positive correlation between the two CD45+ leukocytes populations. In CSF, 50% of all CD45+ leukocytes are CD45+CD11b+ macrophages. We also did a correlation between the number of CD45+CD11b+ macrophages in CSF and the percentage of CD45+CD11b+ macrophages in choroid plexus. However, no significant correlation was found (data not shown), possibly because of lower macrophages number in CSF and choroid plexus compared with leukocytes. We then asked whether monocytes in CSF could migrate into the lesion site after stroke, we isolated Cx3cr1GFP monocytes from bone marrow using CD115 antibody conjugated with magnetic beads. We then directly injected the freshly isolated monocytes into ipsilateral ventricle 6 hours after stroke. At 3 days after the injection, GFP+ monocytes were found in the lesion sites. These data indicated that resting monocytes could sense chemokines released from the lesion site and then migrate towards them. Taken together, these data indicated that macrophages in choroid plexus could migrate into CSF, and then to the lesion site after stroke.

Fig. 4. Infiltration of GFP+ monocytes in the stroke-damaged brain tissue after intraventricular injection.

Next, we investigated if we could take advantage of the CSF-lesion site migration pathway to enhance anti-inflammatory M2 macrophages infiltration into the lesion site. Freshly isolated Cx3cr1GFP monocytes from bone marrow were cultured with IL4 and IL13 to prime them towards the anti-inflammatory M2 fate. 5 days after the culture, many M2 markers, such as YM1, ARG1, IGF1 and TGFβ1 were upregulated, and M1 marker TNFα was downregulated, indicating that monocytes have obtained M2 phenotype. These M2 macrophages were then injected into ipsilateral cerebroventricle 24 hours after stroke. 3 days later, many GFP+ macrophages were found in the lesion sites. The majority of the macrophages expressed M2 marker CD206, suggesting that they maintained their M2 fate at this timepoint. Taken together, these data showed that intraventricular injection of anti-inflammatory M2 macrophages could significantly increase the M2 macrophages number in the lesion site.

Fig. 5. Activation of monocytes in vitro after treatment with IL4 and IL13 towards M2 phenotype.

Finally, we examined if enhancement of M2 macrophages infiltration via CSF-lesion route would influence stroke damage and recovery. Mice subjected to cortical lesion were randomly allocated to cell transplantation group or PBS injection control group. In vitro derived M2 macrophages were injected 24 hours after cortical stroke in mice. 7 days after the injection, mice brains were examined for ischemic damage. Cell transplanted group and PBS injected control group showed no difference in infarct volume. This data suggests that M2 macrophages transplantation has no effect on ischemic damage.

We then investigated the effect of M2 macrophages transplantation on post-stroke motor function recovery using open field test and corridor test. At 20 days after injection, while clockwise rotation involving mainly ipsilateral muscles showed no difference in cell transplantation and PBS injection groups, mice with M2 macrophages showed increased anticlockwise rotation in open field test compared with PBS injected ones. Also, in corridor test, mice with cell transplantation showed increased preference of using contralateral limbs. Taken together, these data showed that M2 macrophages transplantation increased post-stroke recovery in motor function.

Fig. 6. Effect of CSF injection of monocytes on the behavioural recovery after stroke. Mean±SEM. unpaired t-test.

Our data clearly indicate that in response to stroke, there is activation of carotid plexus and injection of monocytes through CSF of lateral ventricle leads to their infiltration in stroke-lesioned brain. Moreover, this treatment leads to improvement of stroke-impaired sensory and motor behaviour. These data pave the way towards further exploration of the possibility to deliver autologous monocytes through the CSF for the treatment of stoke patients in order to promote improvement of spontaneous recovery.

WP2: Role of microglia in cell genesis, plasticity and functional recovery after stroke

D 2.1.1. Contribution of activated microglia to recovery after stroke
In contrast to most other resident tissue macrophages, microglia derive exclusively from primitive myeloid precursors that accumulate in the fetal yolk sac prior to embryonic day E8.5. Subsequently these precursors migrate to the brain through blood vessels around E9.5 and differentiate into mature microglial cells. In addition to their distinct origin another unique feature distinguishing microglia from most other myeloid populations is their radiation- resistance and ability to maintain themselves locally and independently of circulating bone marrow (BM)-derived precursors in the steady state.
The contribution of myeloid cells and the factors that regulate their function remain largely unknown in the ischemic stroke. The development of microglia, the resident macrophages of the CNS, is highly dependent on Csf-1r signaling but independent of Csf-1. In addition to Csf-1, IL-34 is an alternative ligand for Csf-1 receptor secreted by neurons and mice lacking IL-34 have a 50% decrease in microglial cells. Thus the contribution of microglia to the outcome of stroke has been studied using the IL34LacZ/LacZ transgenic mouse line generated by Partner 5. This transgenic mouse is, as said above, characterized by a reduction of microglia of about 50% in the forebrain, and in particular in the cortex, striatum and hippocampus. Importantly this transgenic mouse has no overt alterations in peripheral immunity since IL34 is prevalently expressed in the brain and in the skin.

We have first studied by arterial AngioMR the vascularization of the brain in IL34LacZ/LacZ mice and compared it to IL34LacZ/+ and to IL34+/+ (wildtype) mice. No overt alterations in the main cerebral arteries were found between the three groups. We thus proceeded to induce transient 45 minute long proximal occlusions of the left middle cerebral artery in wild- type, IL34LacZ/+ and IL34LacZ/LacZ mice. For 7 days post-ischemia (7dpi) mice were studied for behavior (mNSS test and cylinder test) and analysis of the inflammatory infiltration, ischemic lesion volume and edema was performed at 7 dpi. By studying about n>20 mice per group we found that IL34LacZ/LacZ mice had a slightly improved survival compared to IL34LacZ/+ and WT mice. Nonetheless ischemic lesion volume and edema was comparable between the three groups (p>0.05). Also behavioral test showed no overt differences between the three groups.
In collaboration with Melanie Greter and Iva Lelios (Partner 5) we studied by flow cytometry myeloid and lymphoid cells infiltrating the brain of wild-type and IL34LacZ/LacZ mice at 2 and 7 days after stroke. We found that early after stroke, at 2 dpi, IL34LacZ/LacZ had less microglial cells (31% compared to wild-type mice), less monocytes (49%) and less moDC (55%) in the ischemic hemisphere. These differences were then lost at 7 dpi when IL34LacZ/LacZ and wild- type mice had again similar numbers of myeloid cells. These early differences observed at 2 dpi and not anymore present at 7 dpi were also observed on histology by counting IBA-1 positive cells in the ischemic and non-ischemic hemisphere of IL34LacZ/LacZ and control mice.
Since the studied IL34LacZ/LacZ transgenic mice have a 50% of reduction of microglia in the forebrain we imported from Partner 5 the newly available CX3CR1-CreERT2 that has been crossed to the iDTR mouse line to ablate more efficiently, upon diphtheria toxin administration, brain microglia (47, 48). However experiments suggested that upon treatment of Cx3Cr1- CreER:iDTR mice with diphtheria toxin, microglia are ablated to varying degrees (30-90%) and do repopulate the CNS within 3-5 days after treatment.
We thus studied ischemic stroke by administering the newly developed inhibitor of the CSFR1 known as pexidartinib (PLX3397) that became only available in late 2016. Indeed as described recently 3 weeks of PLX3397 administration was efficacious in ablating about 99% of microglia cells in the brain. It is noteworthy that this treatment scheme in part also reduces circulating blood monocytes as well as resident macrophages in the brain. Interestingly ablation of microglia with PLX3397 did not affect acute stroke outcome in terms of behavior, survival and ischemic lesion volume outcome up to 7 days post stroke.
In order to determine whether Csf-1r signaling on microglia plays a role in microglia homeostasis and function in steady state, we generated Cx3Cr1- CreER :Csf1r-fl/fl mice. We observed that deletion of Csf-1r on microglia did not lead to a significant microglia number reduction but increased microglia proliferation. Furthermore, we observed that in bone marrow chimeras where Csf-1r was deleted only on microglia cells but not on other myeloid populations (WT x Cx3Cr1-CreER :Csf1r-fl/fl), monocyte- derived cells from the periphery did not contribute to the pool of parenchymal microglia. In the view of examining the contribution of Csf-1r signaling for the infiltration, function and polarization of monocyte-derived DCs in the course of stroke, bone marrow chimeric mice where Csf-1r is deleted only in the periphery but not on microglia cells (Csf1r-/- -WT) could be generated.
In order to better understand the role of microglia under physiological and pathological conditions and to target microglia in vivo our goal was to identify a marker that is specifically expressed by these cells but not by other members of the mononuclear phagocyte system. For this we compared the transcriptome of different macrophage populations by Affimetrix gene chip array and selected the transcription factor Sall1 as a promising microglia-specific gene candidate.
We could show that the transcription factor Sall1 is exclusively expressed by microglia within the hematopoietic system and that no other cell population in the CNS of adult mice expresses Sall1 even upon CNS inflammation and cell infiltration into the brain. By utilizing the tamoxifen-inducible Sall1Cre-ER system we were able to specifically target and modulate microglia in vivo.
We could show that the disruption of M-CSFR signaling, which has previously been shown to be essential for microglia development, leads to a significant decrease in microglia numbers during steady state. In addition we observed that the few remaining cells in this model start to proliferate in order to repopulate the microglia niche. In contrast to that TGF-βR signaling, also recently shown to be important for microglia development, is not crucial for their maintenance. However, in homeostatic conditions signaling via TGF-βR appears to maintain a resting phenotype of microglia since its deletion leads to an activation of these cells.

D 2.1.2: Report on contribution of different inflammatory compartments in repair after stroke

Before proceeding to the deletion of specific inflammatory cell subsets in ischemic stroke we have characterized at 0, 2, 5, 7 and 14 days post ischemia the myeloid inflammatory infiltrate in the brain as well as in the spleen by flow cytometry –in collaboration with Melanie Greter and Iva Lelios (Partner 5). We used multiple markers to identify microglial (CD45low CD11b+ F4/80+ Ly6C-), monocytic (CD11b+ F4/80+ CD45-hi SiglecF Ly6C+) and monocyte derived dendritic cells (moDC, CD11b+ F4/80+ CD45-hi SiglecF Ly6C+ MHCI+) in the brain. We observed that microglial cells increase in the ischemic hemisphere steadily up to 7 dpi (14’280± 2497 cells/ischemic hemisphere) and were still elevated at 14 dpi (12’022± 1696 cells/ischemic hemisphere). Monocytes as well as moDC had an early peak at 2 dpi (2’321± 700 cells/ischemic hemisphere and cells/ischemic hemisphere respectively) but were then again reduced at almost baseline levels at 5 dpi (297± 233 cells/ischemic hemisphere) although showing again at later time-points a slight increase (597± 299 cells/ischemic hemisphere at 14 dpi). Interestingly lymphoid cells (CD11b- cells) being only slightly increased at 2 (1150± 451 cells/ischemic hemisphere), 5 and 7 dpi showed a marked increase at the time-point of 14 dpi (4493± 1730 cells/ischemic hemisphere).
We further detailed the inflammatory infiltrate using the CX3CR1-CreERT2-YFP mouse line that has been recently imported from Partner 5 that can selectively tag and trace microglial cells in the brain. This mouse line permitted us to specifically quantify and study microglial cells in the brain after stroke. Indeed we confirmed that using multiple markers (CD45lowCD11b+F4/80+Ly6C-Ly6G-) we can identify with a very good confidence microglial cells in the brain with superimposable results as by using the CX3CR1-CreERT2-YFP reporter mouse.
Since we observed that microglia was the prevalent immune cell population in the brain in the post-acute phase after stroke and that there were complex technical issues to solve to ablate efficiently microglial cells we concentrated out efforts in the study of microglia in stroke and did not further proceed to ablate other inflammatory cell populations after stroke except for neutrophils (see D2.8.1).

D 2.2.1. Report on the influence of activated microglia on stroke- induced neurogenesis.
Ischemic stroke triggers neurogenesis from neural stem/progenitor cells (NSPCs) in the subventricular zone (SVZ) and migration of newly formed neuroblasts towards the damaged striatum where they differentiate to mature neurons. Whether it is the injury per se or the associated inflammation that gives rise to this endogenous neurogenic response is unknown. Here we showed that inflammation without corresponding neuronal loss caused by intrastriatal lipopolysaccharide (LPS) injection leads to striatal neurogenesis in rats comparable to that after a 30 min middle cerebral artery occlusion, as characterized by striatal DCX+ neuroblast recruitment and mature NeuN+/BrdU+ neuron formation. Using global gene expression analysis, changes in several factors that could potentially regulate striatal neurogenesis were identified in microglia sorted from SVZ and striatum of LPS-injected and stroke-subjected rats. Among the upregulated factors, one chemokine, CXCL13, was found to promote neuroblast migration from neonatal mouse SVZ explants in vitro. However, neuroblast migration to the striatum was not affected in constitutive CXCL13 receptor CXCR5-/- mice subjected to stroke. Infarct volume and pro-inflammatory M1 microglia/macrophage density were increased in CXCR5-/- mice, suggesting that microglia-derived CXCL13, acting through CXCR5, might be involved in neuroprotection following stroke. Our findings raise the possibility that the inflammation accompanying an ischemic insult is the major inducer of striatal neurogenesis after stroke. Although previous studies have established that inflammation can influence different steps of neurogenesis, i.e., cell proliferation, migration, differentiation, survival and functional synaptic integration, our study provides the first experimental evidence that inflammation per se triggers a neurogenic response in the mammalian brain. These findings support the notion that the cross-talk between immune cells and NSPCs and their progeny is of crucial importance for efficient neuroregenerative responses following disease and damage in the adult brain. Better understanding of the molecular mechanisms involved in this dialog could provide novel therapeutic targets to promote more efficient repair of the damaged and degenerated brain. These data are published in Neurobiology of Disease (Chapman et al., 2013).

D 2.2.2. Report on the fate and properties of SVZ-derived neuroblasts and new neurons in stroke-subjected transgenic mice following ablation of microglia
We have studied neurogenesis in healthy and in stroke conditions in the IL34LacZ/LacZ transgenic mouse line. As said above this mouse model has a 50% reduction of microglia in healthy conditions, while showing similar microglia cell numbers in the post-acute phase after stroke. Analyzing doublecortin (DCX)+ cells in the subventricular zone (SVZ) and in the lesion border zone, and analyzing the proliferation of DCX in the SVZ (BrdU+ cells) at 7 days post ischemia we did not observe significant differences between IL34LacZ/LacZ and WT mice. Also at gene-expression level, analyzing the expression of Nestin, DCX and DLX-2 in the striatum and SVZ of IL34LacZ/LacZ and WT mice at 3 days after stroke no difference in expression levels of these markers could be noticed. Due to the limited and time-constrained reduction of microglia in the IL34LacZ/LacZ mice after stroke the experiments for studying in this transgenic model the survival and engraftment of transplanted progenitors and iPS cells in stroke-subjected transgenic mice (D 2.3.1. and D 2.4.2.) were not performed. To overcome the transient and limited microglia reduction of the IL34LacZ/LacZ we focused our attention on the ablation of microglia with the CSF1R inhibitor PLX3397 and in the study of stroke outcome and of neurogenesis. Indeed we observed that iPLX3397 treated animals have a relative increase of doublecortin positive cells (type A cells) within the SVZ takes place while transient amplifying cells as labeled by BrdU are unchanged.

D 2.3.1. Report on fate and properties of grafted neural progenitors and iPS cells in stroke-subjected transgenic mice following ablation of microglia

We have previously shown that the best survival of grafted human NSPCs in the stroke- damaged brain requires optimum numbers of cells to be transplanted in the early post- stroke phase, before the inflammatory response has been established (Darsalia, 2011, JCBFM). The inflammatory response in the brain is mainly represented by macrophages derived from activated local microglia and infiltrating monocytes from blood stream.

Since the Il34LacZ/LacZ transgenic mouse only had a 50% reduction of microglial cells that even turned to be less upon ischemic stroke induction and since the use of PLX3397 while ablating microglial cells in the brain in part reduced also circulating monocytes, we decided to study the role of monocyte-derived macrophages for the intracerebral transplantation studies. Moreover, whether MDMs affect reprogrammed human-derived cells when transplanted in stroke-damaged brain was unknown. To clarify this issue important from translational perspectives, animals were subjected to stroke (distal MCAO) and depletion of circulating and brain-infiltrating monocytes was carried out by MC-21 injections. Vehicle-injected animals served as control. BrdU was injected for 3 days to label diving cells and at 1 week after stroke, when the peak of inflammation occurs (Darsalia, 2011, JCBFM), animals received intracortical transplants of cortically fated iPS cells as described previously (Tornero, 2012, Brain). At 2 and 8 weeks after transplantation, animals were sacrificed. We assessed proliferation, survival, differentiation and migration of the grafted iPS cells using immunohistochemistry and unbiased morphometric quantification (stereology).

Quantification of the number of grafted cells by means of immunocytochemistry with human-specific antibody HuNu revealed that depletion of monocyte-derived macrophages did not have any cear effect on survival of grafted human skin-derived lt-NES cells at 2 and 8 weeks after transplantation in stroke-lesioned rat brains, respectively (Fig. 1).

Fig.1. Effect of depletion of monocyte-derived macrophages on number HuNu+ grafted lt- NES cells.

The analysis of proliferation of grafted cells by means of mitotic marker Ki67 revealed that proliferation was not affected (Fig.2). Proliferation rate varied between 8 and 12 of all grafted cells and there were no differences between experimental and control as well as different time point groups

Fig.2. Effect of depletion of monocyte-derived macrophages on proliferation of grafted GFP+ lt-NES cells.

We also carried out evaluation of neurogenesis from grafted cells by counting of BrdU+/DCX+ cells which did not reveal any difference between groups subjected to saline and MC-21 injections, respectively (Fig.3).

Fig.3. Effect of depletion of monocyte-derived macrophages on number grafted GFP+ lt- NES cell graft-derive DCX+ neuroblasts

Although quantification of newly born neuroblasts from grafted cells did not show any differences we still quantified the number of mature neurons by counting number of Fox3+/GFP+ neurons. There was clear increase in the number of mature neurons from 2 weeks after transplantation to 8 weeks illustrating maturation of grafted lt-NES cells in both groups, control and MC-21 injected (Fig.4). However, there were no differences in the number of grafted Fox3+ neurons between saline and MC21 injected groups.

Fig.4. Effect of depletion of monocyte-derived macrophages on number GFP+ grafted lt- NES cell-derived neurons.

Next, we examined inflammatory environment by quantifying the number of activated microglia/macrophages in the lesioned and contralateral striatum. To depict this population we used double-immunocytochemistry and stained for Iba1 and ED1. We carried out counting in SVZ, and the lesioned whole striatum (Fig.5). In the striatum of both groups, control and MC-21 injected, counting of cells in the ipsilateral to lesion side revealed increase of the number of activated microglia/macrophages as compared to contralateral intact side. However, only in MC21 group we observed significant decrease of Iba1 immunoreactivity from 3 weeks after stroke to 9 week timepoint.

Fig.5. Effect of depletion of monocyte-derived macrophages on number Iba+ and ED1+ microglia/ macrophages near the lesion.

To further explore the inflammatory environment we quantified number of Iba1/ED1+ double-positive cells inside the graft of lt-NES cells and around the core (Fig.14). While there was no difference between saline and MC-21groups at 3 weeks, at 9 weeks we detected increased inflammation only in MC-21 group but not in control one. However, this increase was attributed to the number of only activated macrophages (Iba1+/ED1+) because the number of single positive Iba1+ cells was not changed in either group or time point (Fig.6).

Fig.6. Effect of depletion of monocyte-derived macrophages on number Iba+ and ED1+ microglia/ macrophages inside the graft.

As a part of inflammatory environment we evaluated the activation of astrocytes in the lesioned striatum. Measuring the area covered by GFAP immunoreactivity (see above) did not reveal any difference between control and experimental groups (Fig. 7). However, evaluation of the effect of depletion of monocyte-derived macrophages on the degree of astrocyte activation near and inside the graft of lt-NES cells revealed that MC-21 antibody and blockade of monocyte infiltration lead to decrease of glial scar inside and close proximity to the transplant.

Fig.7. Effect of depletion of monocyte-derived macrophages on coverage of GFAP+ cells in the lesioned striatum (upper graph) and near and inside the graft (lower graph).

These data indicate that monocyte-derived macrophages mot like do not affect the survival and differentiation as well as proliferation of grafted stem cells. However, it is affecting the degree of inflammation inside the graft and in close proximaty. How this in turn affects functionality of the neurons generated from the grafted lt-NES cells remains to be elucidated.

D 2.4.1. Report on electrophysiological properties of SVZ-derived new neurons in stroke-subjected transgenic mice following ablation of microglia

As previously reported we have studied the NestinCreERT2-YFP mouse, in which it is possible to specifically trace neural stem cells by injection of tamoxifen followed by 15 days of washout before the induction of stroke. We observed that after stroke NPC from the SVZ migrate towards the ischemic lesion and express GFAP (20.8± 14.2, 26.8±7.4 and 55.2±4.2, percentage of YFP+GFAP+ over YFP+ cells at 6, 10 and 30 dpi respectively), doublecortin (11.1± 2.7, 20.7±12.3 and 20.6±9.7, percentage of YFP+DCX+ over YFP+ cells at 6, 10 and 30 dpi respectively), NeuN (0, 0 and 1.4±1.0, percentage of YFP+NeuN+ over YFP+ cells at 6, 10 and 30 dpi respectively) and olig2 (22.1, 11.9±5.5 and 9.6±5.7, percentage of YFP+Olig2+ over YFP+ cells at 6, 10 and 30 dpi respectively). From this data as well as from recent literature (Faiz et al., 2015) it emerges that NPC derived from the SVZ to the ischemic lesion differentiate mainly in astrocytes while only very few differentiate in neurons at the time-points analyzed (up to 30 days post stroke). We therefore did not proceed in studying electrophysiological properties of SVZ- derived new neurons after stroke both for the rarity of finding these cells as well as for the difficulty up to the recent usage of PLX3397 in ablating microglial cells. However even in PLX3397 treated animals the number of YFP+NeuN+ at 30 days post stroke was too low to enable us to proceed to the study of electrophysiogical properties of SVZ- derived new neurons.

D 2.4.2. Report on electrophysiological properties of grafted progenitors and iPS cell-derived neurons in stroke-subjected transgenic mice following ablation of microglia

We studied the electrophysiological properties of cells generated from iPSC-derived lt-NES cells at 5 months after transplantation into stroke-lesioned striatum or cerebral cortex of Nude rats by performing whole-cell patch-clamp recordings in acute brain slice preparations (Fig. 1A-G).

Fig. 1. Human-induced pluripotent stem cell-derived long-term expandable neuroepithelial-like stem (lt-NES) cells have developed electrophysiological properties of functional neurons at 4.5–5.5 months after grafting into intact rat striatum and cortex. Whole-cell configuration of GFP-expressing lt-NES cells (A–C). Photomicrograph of a grafted in the striatum (D) and cortex (F) and recorded cells (in the boxed area) labelled with biocytin and GFP. Enlargement of boxed areas in (D) and (F) showing (arrows) grafted GFP+ cell filled with biocytin after recording (E, G). Patch-clamp recordings from the cells grafted in the striatum (H–L) and the cortex (M–S). Representative traces of membrane potential responses (H, M, striatum and cortex, respectively) to step injection of hyperpolarizing and depolarizing current (10 pA steps) showing action potentials that are blocked by TTX (right). Representative traces of whole-cell Na+ and K+ currents (I, N), blocked using TTX and TEA (I), respectively, elicited by voltage steps from -70 mV to +40 mV in 10 mV increments. Representative traces of spontaneous (J, O) and excitatory (K, Q) postsynaptic currents recorded in the presence of PTX, in voltage-clamp configuration at -70 mV. Absence of postsynaptic currents after addition of PTX and glutamate receptor antagonists, NBQX and D-AP5 (L) or only NBQX (R). Representative traces of miniature (P) postsynaptic currents recorded in the presence of TTX, in voltage-clamp configuration at -70 mV. Postsynaptic AMPA-receptor-mediated currents (S) are evoked by paired-pulse electrical stimulation delivered from a stainless-steel electrode placed approximately 300 lm away from the transplant in the cortex (three representative traces superimposed.

Grafted cells were classified as neuronal or glial based on resting membrane potential and input resistance (Table). A proportion of the cells was functionally mature neurons with a resting membrane potential of approximately -50 mV and -58 mV in the striatum and cortex, respectively, the majority generated action potentials in response to depolarizing current injection (Fig. 1H, M). In voltage-clamp configuration, depolarizing voltage steps induced characteristic Na. and K. whole-cell currents, which were sensitive to the voltage-gated Na. channel blocker, TTX, and the voltage-gated K. channel blocker, TEA, respectively (Fig. 1I-N). Spontaneous postsynaptic currents were frequently observed in grafted cells in the striatum (Fig. 1J-O). Spontaneous excitatory postsynaptic currents (sEPSCs) were recorded in the presence of the GABAA receptor antagonist, PTX (Fig. 1K-Q), and addition of glutamate receptor antagonists NBQX and D-AP5 blocked all postsynaptic currents (Fig. 1L), indicating that the cells had functional excitatory synapses. Interestingly, spontaneous inhibitory postsynaptic currents were not observed (n =5 cells).

Table Striatum
(n = 2 rats, 13 cells) Cortex
(n = 2 rats, 12 cells)
Neuronal
(n = 6 cells) Glial
(n = 7 cells)
Resting membrane potential (mV) -51.4 ± 4.2 -81.6 ± 2.1 -57.8 ± 5.3
Input resistance (MΩ) 1478 ± 253 172 ± 40 656 ± 175
Membrane capacitance (pF) 8.7 ± 1.5 11.3 ± 2.5 11.8 ± 1.5
Cells exhibiting action potentials 5/6 0/7 7/12
Action potential threshold (mV) -32.5 ± 1.3 - -34.9 ± 1.3
Action potential amplitude (mV) 41.6 ± 5.0 - 47.0 ± 3.7
Action potential duration (ms) 1.4 ± 0.2 - 1.3 ± 0.1
After hyperpolarization amplitude (mV) 11.6 ± 2.7 - 7.1 ± 0.7
After hyperpolarization duration (ms) 58.4 ± 25.0 - 20.3 ± 7.2

In addition, we detected miniature postsynaptic currents that were action potential independent (not blocked by TTX) (Fig. 1P). In two of 10 grafted cells, excitatory, AMPA receptor- mediated currents could be evoked by stimulating a cortical region remote from the transplant. This finding suggests that transplanted neurons generated from iPSC-derived lt-NES cells receive synaptic input from host neurons and functionally integrate into host brain neural circuitries.
In order to explore the cross-talk between grafted iPSC-derived lt-NES cells and microglia/ macrophages we examined the number of resting and activated macrophages and macrophages with different level of activation based on their morphology in vehicle-injected and cell-transplanted animals subjected to stroke (Fig.2).

Fig. 2. (A - F) Photomicrographs showing similar distribution and density of Iba1+, ED1+, and Iba1+/ED1+ microglia/macrophages in injured cortex of stroke-subjected, vehicle-injected (A-C) and cell-grafted (D-F) aged rats. (G-H) Photomicrographs showing examples of ramified (G), intermediate (H), amoeboid (I) and round (J) microglia/macrophage morphologies. (K) Microglia/ macrophages with less activated (ramified + intermediate) and more activated (amoeboid + round) phenotypes expressed as percentage of total number of Iba1+ cells in the peri-infarct cortex 8 weeks after dMCAO in vehicle-injected and cell-grafted animals. Means ±SEM, *, P < 0.05 unpaired t-test. Scale bar = 50 µm (A-F), and 12 µm (G-J).
No differences were detected between vehicle-injected and cell-transplanted animals with regard to the total number of microglia/macrophages (Iba1+ cells; 967.9±71.1 and 762.5±87.6, respectively) or percentage of activated microglia/ macrophages (Iba1+/ED1+ cells; 87.9±3.6 and 83.4±2.1, respectively) in stroke-injured cortex (Fig. 2A- F). However, the morphology of microglia/macrophages differed between the two groups (Fig. 2 G-J). The majority of microglia/macrophages in vehicle-injected animals exhibited a more activated morphological phenotype, being ameboid or round (Fig. 2K). In contrast, considerably fewer microglia/macrophages showed this phenotype in cell-grafted rats, most of them being ramified or intermediate.

D 2.5.1. Report on mechanisms of neuron-to-microglia signaling in control animals, and animals subjected to stroke

Microglia are observed in the early developing forebrain and interact with neurons by regulation of neurogenesis through only partially known mechanisms. In the developing cerebral cortex, microglial cells cluster in the ventricular/subventricular zone (VZ/SVZ), a region containing in particular Cxcl12-expressing basal progenitors (BPs).

We have investigated how ablation of neuronal BP as well as genetic loss of Cxcl12 of BP reduces microglia recruitment into the SVZ. Ectopic Cxcl12 expression or pharmacological blockage of CxcR4 further supported that Cxcl12/CxcR4 signaling is involved in microglial recruitment during cortical development. Furthermore, we found that cell death in the developing forebrain triggers microglial proliferation and that this is mediated by the release of macrophage migration inhibitory factor (MIF). Indeed is has been consistently shown that MIF can modulate myeloid cell proliferation by binding to the CD74 receptor.

Interestingly we found that also microglia is important for neurogenesis. In fact when we depleted microglia by using mice lacking receptor for colony-stimulating factor-1 (Csf-1R) we found a reduced number of BPs into the cerebral cortex. This suggests that the microglia and neural stem cells are mutually important for the development of the brain.
We then moved to study in adult mice whether an altered number of microglial cells would also affect formation of new neurons in the SVZ and in the hippocampus. We thus studied neurogenesis in the IL34LacZ/LacZ that has a 50% reduction of microglia in the forebrain. We observed that the in vitro growth rate of neural precursor cells (NPCs) derived from IL34LacZ/LacZ was comparable to wild-type NPCs. By counting however neuroblasts (doublecortin, DCX+ cells) in the SVZ and hippocampus of IL34LacZ/LacZ mice we found an increase of these cells in particular in the hippocampus. Previous studies by Nandi et al. have shown that IL34 can act as repressor of NPC proliferation. Further in vivo and in vitro studies are thus currently ongoing to better understand and characterize the role of IL34 on NPCs in vivo as well as after stroke.

D 2.5.2. Report on functional consequences of neuron-to-microglia signaling on microglial functional state and behaviour in control and stroke-subjected animals

Communication between neurons and microglial cells has been found to be important for controlling microglial behaviour. Partner 3 has demonstrated (Arnò et al. Nat Comm 2014) that the mutual relationship between microglia and developing neural progenitors is fundamental for the proper cortical development. Byin vivo functional approaches, we have observed that microglia recruitment into the cortical ventricular zone (VZ) and subventricular zone (SVZ) is dependent on Cxcl12-expressing basal progenitor cells (BP) in the developing brain. We found that primary microglia cell cultures and cortical microglia express Cxcl12 receptors. This is a necessary feature for basal progenitors (BP) expressing Cxcl12 to attract microglia into the VZ/SVZ. Accordingly, the downregulation of Cxcl12 in GFAPCre/Tbr2flox/flox mice led to a 25% reduction of microglia into the cortex. In line with these results, the ectopic expression of either Tbr2 or Cxcl12 substantially increased the number of microglia within treated cortical areas. Furthermore, the injection of AMD3100 - a blocker of Cxcl12/ CxcR4 interaction - as well as the constitutive inactivation of CxcR4 and the conditional inactivation of Cxcl12 (GFAPCreCxcl12flox/flox) caused a significant reduction of cortical microglia cells.

To further sustain the finding that basal progenitor cells of the brain are involved in the recruitment of microglia at the VZ/SVZ, we examined transgenic mice in which the proper generation of BP within the developing cerebral cortex is impaired. In mice carrying a constitutive active form of β-Catenin in radial glia (RG) the reduction of BP affected Cxcl12 expression in the SVZ and in consequence the number of microglia was reduced to about 60%. Furthermore in Pax6Sey/Sey mice, carrying a null allele for the transcription factor Pax6, Tbr2+ cells were greatly reduced, Cxcl12 expression was suppressed into the SVZ and a significant reduction of microglia was observed. In mice lacking Emx2 - a transcription factor promoting RG proliferation in the occipital cortex—the number of Tbr2+ cells was reduced to 30% and the expression of Cxcl12 was greatly impaired. In the same mice, the reduction of Cxcl12 was paralleled by a 30% reduction of microglia in the occipital cortex. On the other hand, the inhibition of Notch signaling in radial glia increased both basal progenitor cells and the expression of Cxcl12 in the developing cortex, a finding accompanied by a threefold increase in the microglia number. Interestingly also microglia influences the developing cerebral cortex. We used Csf-1R null mice that are microglia-free and measured basal progenitors. We found that Tbr2+ basal progenitors were slightly, but significantly, reduced in Csf-1R null mice at E13.5 and at E17.5. This observation suggests that in the developing brain microglia can certainly contribute to removing progenitor cells but its primary role seems to support neurogenesis. In conclusion, we demonstrated that microglia clustering into germinal niches of the developing cortex is largely dependent on basal progenitors expressing Cxcl12 and the mutual bidirectional relationship between these two cell types is vital to protect the developing cerebral cortex from unexpected damage impairing the proper proliferation and cortical positioning of neural precursors.

With this study on the crosstalk between microglia and neurons during cortical development we can conclude that neuronogenesis needs the interaction with microglia and that interference in the signaling between these two cells might alter regenerative programs of the cortex.

D 2.6.1. Report on the role of ROS, and metabolites of the IDO and NO pathways on recovery after stroke
Cerebral ischemia initiates a cascade of detrimental events characterized by rapid activation of resident microglia and astrocytes, and the production of pro-inflammatory mediators resulting of the activation of NOS and the IDO pathways.
The first phase is called excitotoxicity and corresponds to a high liberation of glutamate and ROS: elevated intracellular calcium ion concentration mediated by increased levels of catecholamines (CA) (Fink et al., 2002); activation of nNOS and iNOS to generate NO and consequently ROS after NADPH oxidase activation and cytotoxic substances; the increase of glutamate, taurine, catecholamines, NO and NO2 metabolites. The latter metabolites induce the formation of protein-adducts, thus leading the cell to apoptosis. After focal ischemia, the primary neuronal death appears rapidly in the core area and it is followed by the secondary death in the ischemic penumbra, which evolves from the delayed activation of multiple cellular death pathways. The core anoxic ischemic depolarization induces a release of neurotransmitters such as glutamate. Once released, glutamate generates a phenomenon of peri-infarct depolarization, which increases energy consumption and promotes Ca++ influx into the cells (Guillemin et al., 2001).

The second phase is characterized by an inflammatory activation and apoptosis. This phase is characterized by: the activation of cytokines, caspases and both IDO and NO pathways. IDO-1 is up-regulated by several inflammatory molecules including interferon gamma (IFN-γ) (Shimizu Tet al., 1978; Werner et al., 1989). Infiltrating macrophages, activated microglia, and neurons express the full range of kynurenine (KP) enzymes, whereas astrocytes and oligodendrocytes lack the crucial enzymes kynurenine 3-monooxygenase (KMO) and IDO-1, respectively (Guillemin et al., 2000; Lim et al., 2016). Thanks to this enzymatic variability we can discriminate later using immunohistochemical techniques which are the IDO metabolites implicated on inflammation processes.
A. We have firstly developed antibodies directed against the different metabolites, in order to evaluate the IDO and NO pathway metabolites. Thanks to these antibodies we have then conducted immunocytochemical techniques.
Thus, monoclonal antibodies were raised directed against NO, NO2 and IDO metabolites. Immunochemical characteristics of these antibodies were resumed as very high affinity and specificity comparing with close competitors. These data are very important for later studies since for example 3-hydroxyanthranilic acid (3HAA) is a free-radical generator (Goldstein et al., 2000), the excitotoxin quinolinic is a N-methyl-D-aspartate (NMDA) receptor agonist (Stone et al., 1981) and kynurenic acid (KYNA) is a NMDA antagonist (Perkins et al., 1982).
Thus, IDO-1 up-regulation as well as accelerated and sustained degradation of tryptophan represent key indicators of inflammation. Indeed, inflammation and resulting immune activation lead to KP activation and the concomitant increased production of the excitotoxin QUIN (Moffett et al., 1997).

B. Thus, the upregulation of IDO, NO and NO2 metabolites were evaluated after conducting transient middle cerebral artery occlusion (tMCAO, a stroke model) on 4 experimental groups of rats, treated with:
In order to know if these molecules (IDO, NO, ROS) play an important role, we have carried out immunocytochemical techniques using monoclonal antibodies directed against conjugated IDO metabolites, NO-metabolites. These immunocytochemical studies have been combined with other evaluation of markers (anti-Iba1; anti-GFAP; anti-CD45; anti-CD11b and anti-Doublecortin). After the mapping of the brain only two IDO metabolites were visualized in the ischemic region (Striatum/Cortex). Thus, it seems that these molecules must play an important role during the different phases of the ischemia. It is also possible that the other metabolites that were not found exert a role as well in other moment of the process. Both molecules, matches perfectly on distribution found with that observed after using anti-Iba1; anti-GFAP; anti-CD45; anti-CD11b. In the model of stroke we have found that 3HAA and KYNA were found in astrocytes (Mangas et al., 2016b; Eur J Histochem; Mangas et al., 2017 “in press”; Annals of Anatomy). Furthermore, astrocytes are KMO negative, it means that the presence of 3HAA is only explainable by an uptake system since this molecule is synthetized in microglia (KMO positive) (Guillemin et al., 2000; Lim et al., 2016). More, other aim of this study was to evaluate the role played by the NO pathway in stroke. But we did not observe immunoreactivity in the infarcted region using monoclonal antibodies directed against conjugated NO-tryptophan (NO-W) and conjugated NO2-Tyrosine (NO-Tyr). Finally, we have recently published the presence of NO modified tryptophan in the rat septum (Mangas et al., 2016a, Eur J Histochem), but this NO modified small molecule was not present in the ischemic region of the stroke model. Thus, it is possible that the absence of immunoreactivity for the NO pathway was mainly due: 1) other markers not studied here could be involved such as conjugated NO-Cysteine, 2) there is not an excess of NO and/or 3) molecules of the NO pathway have short half-lives and they cannot be detected after two days by immunohistochemistry techniques, except for the conjugated NO-W in the septum where this molecule appears is endogenous (Mangas et al., 2016a, Eur J Histochem).
In all the cases, we have demonstrated the specificity of the signal with the following histological controls: 1-elimination of the first or secondary antibodies, 2-preabsorption of the antibody with the targeted molecule (antibody pre-incubated with conjugated small molecule). In all cases, the results showed the extinction of the signal.

Conclusions
After application of immunocytochemical techniques, we have reached the following conclusions:
1) Striatum/Cortex damage induced by the transient middle cerebral artery occlusion (tMCAO) after 3, 5 and 21 days was demonstrated by the markers (anti-Iba1; anti-GFAP; anti-CD45; anti-CD11b; see Deliverable 2.7.1) and this signal was coincident with the signals found with two IDO metabolites (KYNA and 3HAA).
2) The presence of these molecules in the ischemic region demonstrate that these IDO-metabolites are playing an important role in the pathogenesis of Ischemia, and these IDO metabolites (KYNA and 3HAA) are early overexpressed in stroke..
3) NO and NO2 metabolites were not visualized at this stage (3, 5, 21 days post I/R), probably because these molecules are not present anymore, at this stage, in the ischemic region. At this point, we have considered that maybe using the tools applied here it was not possible to detect NO and NO2 conjugated metabolites (which are ephemeral due to the own properties of the molecules being quickly combined). But this hypothesis was quickly discarded because we have observed immunoreactive neurons in other brain regions.
4) We have visualized for the first time using antibodies developed by Gemac neuronal populations that have never been described until now: 1) NO-W. The signal was found in tMCAO animals independently of the composition of the treatment and it was also found in animals that were not operated (negative controls). This means that the visualized signal in the rat brain for the NO-W has an endogenous origin. 2) KYNA and 3HAA were found in astrocytes exclusively found in the damaged region.

D 2.7.1. Report on the impact of GEMST on the early and late phase of inflammatory response elicited by experimental ischemic stroke

The aim of these studies was to demonstrate the proof of concept of GEMST on the experimental ischemic stroke model.
GEMST was conceived in order to treat the excitotoxicity environmental conditions of brain tissue after an acute ischemic/reperfusion (I/R) process using a known model of stroke (transient middle cerebral artery occlusion (tMCAO)).
In vitro model
Before conducting experimental stroke on animals, we have done in vitro studies on glutamate intoxicated neurons (in vitro experimental model), this experiment was realized with subcontractor (Bioalternatives, France). In the in vitro stroke model, neurons were intoxicated with glutamate and exposed to several concentrations of different poly-L. Lysine compounds (PLL) belonging to the formulation of GEMST. Unfortunately, these studies did not show much more information since when high levels of glutamate (excitotoxicity) were added to the neurons, PLL compounds did not show any efficacy in this in vitro model. In order to evaluate the proof of concept of GEMST, we used the tMCAO animal model.
tMCAO mice model with WP2 leader and partner 3 of TargetBrain project
First of all, it was conducted a transient tMCAO mice model in which animals were treated with a single dose of a PLL-amino acid. These experiments were conducted in collaboration with Martino's group. The first results obtained encouraged us to focus our efforts in this research line since a single dose of a PLL compound (0.65 mg/kg) decreased significantly in this model the edema, but there wasn’t a favorable impact on the ischemic region. The technical difficulties found in the mice model were almost exclusively due to the size of the animals.
tMCAO rat model
At this point and in order to evaluate in early phases the efficacy of the GEMST components, a tMCAO rat model was conducted. Three different compositions of GEMST have been tested versus the vehicle: Composition 1 (PLL linked to different vitamins), Composition 2 (PLL linked to different amino acids) and Composition 3 (whole formula: PLL linked to different vitamins and amino acids). Two subcutaneous injections were carried out (one hour after I/R and 25 hours after I/R), this experiment was realized with subcontractor (Etyca, Spain). This experiment did not show significant differences between the different groups:
1) Composition 1 (7.5 mg/ subcutaneous injection).
2) Composition 2 (7.5mg/ subcutaneous injection).
3) Composition 3 (7.5 mg/ subcutaneous injection).
4) Vehicle group (NaCl) (subcutaneous injection)].
This experiment did not show much more information since animals were treated only for 24 hours. Thus, the inflammatory response at early phase was not decreased. After sharing and discussing the results with the coordinator and partner 3, it was concluded that experiments must be conducted by using treatments for at least two weeks in order to see the inflammatory response and to avoid any possible behavioral interference.
Thus, a second large rat experiment at late phase was done using the three formulations in order to study the inflammatory response during the late phase. Experiments were done in collaboration with the Institute of Neurosciences of Castilla y Leòn, Spain. Thus, PLL-Compounds were tested using four different groups (Composition 1, Composition 2, Composition 3, Vehicle group); survival animals to I/R procedures by group was n=5 (21 days after the I/R procedure).
The first animal experiment conducted with subcontracting (Etyca) at early phase was not comparable to the study of the late phase since tissue was treated with 2,3,5-tripheniltetraazolium chloride (TTC) and this technique does not allow to apply immunocytochemical techniques and compare with late phase study. Thus, a second early phase experiment was conducted treating the animals for 3 and 5 days with the Composition 3 (better composition) versus controls. Thus, animals were treated from two days before suffering the tMCAO surgical procedure to the day of the perfusion (3 or 5 days after the surgical procedure). Furthermore, this new study include a deep immunocytochemical study that allow to compare histological results in the different phases with key markers.

Immunocytochemical studies
In order to avoid possible interference by endogenous peroxidase, free-floating sections were treated with a mixture of methanol and H2O2 (2:1) for 30 min. Then, sections were thoroughly washed with an orbital tube shaker (OTS) in 0.15 M phosphate-buffered saline (PBS, pH 7.2) (2 x 15 minutes at room temperature “Rt”) and pre-incubated in PBS containing 1% of normal horse serum and 0.3% of Triton X-100 (mix solution) for 30 min (OTS, Rt). The sections were then incubated overnight at 4º C in the same buffer containing anti-Iba-1 (1/1,500, Abcam), anti-GFAP (1/400, Abcam; 1/100, Dako), anti-CD45 (1/50-1/100, Abcam), anti-CD11b (1/50-1/100, Abcam) or anti-Doublecortin (1/300; Santa Cruz Biotechnology) in the mix solution. Following this, the sections were washed in PBS (2 x 15 min) and then incubated for 60 min with biotinylated anti-rabbit immunogammaglobulin (Vector), diluted 1/200 in the mix solution (OTS, Rt). Following rinses with PBS (2 x 15 minutes, OTS, Rt), the sections were incubated for 60 min with a 1/100-diluted avidin-biotin-peroxidase complex (Vectastain) in the mix solution (OTS, Rt). Finally, after rinses of the sections with PBS (2 x 15 minutes, OTS, Rt) and Tris-HCl buffer (pH 7.6, 10 min, Rt), tissue-bound peroxidase was developed with H2O2, using 3, 3’ diaminobenzidine as chromogen (10 min, Rt), as previously described (Mangas et al., 2007, 2009, 2012). The sections were then thoroughly rinsed with PBS and coverslipped with PBS/Glycerol (1/1).

Results
It was observed an increase in the signal of the animals belonging to the positive control group (C+, vehicle) and only in the ipsilateral side of the brain (damaged region due to the occlusion) (see Table1), but not in treated animals with the following markers:

• Anti-Iba-1. Ionized calcium-binding adapter molecule 1 (IBA1) is a microglia/macrophage-specific calcium-binding protein. It is specifically expressed in macrophages/microglia and is upregulated during the activation of these cells. Iba1 expression is up-regulated in microglia following nerve injury, central nervous system ischemia, and several other brain diseases.
• Anti-GFAP. Glial fibrillary acidic protein (GFAP). GFAP is a monomeric intermediate filament protein in astrocytes and its protein level in the serum reaches a maximum between day 2 and 4 after acute ischemic stroke.
• Anti-CD45. CD45 (lymphocyte common antigen) is a receptor-linked protein tyrosine phosphatase that is expressed on all leukocytes, and which plays a crucial role in the function of these cells; CD45 antibody identifies the rat leukocyte common antigen expressed in all leukocytes and, at lower levels, in resting microglia.
• Anti-CD11b. CD11b is expressed on the surface of many leukocytes including monocytes, neutrophils, natural killer cells, granulocytes and macrophages, CD11b regulates leukocyte adhesion and migration to mediate the inflammatory response. During microglial activation, the expression of CD11b, microglia secrete microglial surface markers, CD11b is the most important one with biological significance.
• Anti-Doublecortin. Doublecortin is a reliable and specific marker that reflects level of adult neurogenesis and its modulation

Table 1. tMCAO rat model, immunocytochemical results visualized after three weeks with the different markers used.

Marker

Group

Iba-1

Ipsi / Contra

GFAP

Ipsi / Contra

CD45

Ipsi / Contra

CD11b

Ipsi / Contra

C18

Ipsi / Contra

Composition 1
(n=5)

++ +*

+
+*
-*

+

-*

-

-*

-

+*

+

Composition 2
(n=5)


++

+

+
-*

+

+
-*

-

-*

-

+

+

Composition 3
(n=5)


+*


+

-*

+

-

-

-

-

+

+

Vehicle (NaCl)
(n=5)

+++
++

+
+++
++
+

+
+++
++
-*

-
+++
+
-


-
++
+

+

Table 1 legend: +++: high density; ++: moderate density; +: low density; +*: indicate that punctually we found a moderate density but in general is a low density; -: absence of immunoreactivity; -*: in general absence of immunoreactivity, and punctually a low density. Contra: Contralateral side of the brain; Ipsi: Ipsilateral side of the brain.

An increase in the immunoreactivity was observed in the sections belonging to the positive control group animals (C+, vehicle) and in those belonging to groups 3 and 5 days, but not in non-operated animals. In all cases, only the ipsilateral side of the brain (damaged region due to the occlusion procedure) showed the overexpression of the following markers: anti-Iba-1, anti-GFAP, anti-CD45, anti-CD11b, anti-3HAA and anti-KYNA (Mangas et al., 2016, 2017).

Conclusions
Once the mapping of the animals was finished, we reached the following conclusions:

1) It has been demonstrated that in the focused region of the lesion (striatum), there's an increase in the signal of the animals belonging to the positive control group (C+) and only in the ipsilateral side of the brain (damaged region due to the occlusion) (see Table1), but not in treated animals with the following markers: Anti-Iba-1, Anti-GFAP, Anti-CD45, Anti-CD11b and Anti-Doublecortin.

2) We unknown to date the mechanisms occurring but compositions 1, 2 and 3 of GEMST exert an effect on the inflammation compared with the positive control group (treated with NaCl), since only in some sections of some animals belonging to groups treated with Composition 1 and 2, a few leukocyte infiltrates were observed at 21 days after stroke. At 3 -5 days it seems that GEMST exert only a minor effect.

3) Any animal belonging to the group treated with composition 3 showed ipsilateral leukocyte infiltrates; this means that Composition 3 is the most efficacious composition formula of GEMST.

4) Elimination of the antibodies (primary and secondary) confirmed the specificity of the reaction. Furthermore, preabsorption of Iba-1 with anti Iba-1 confirmed also the specificity of the reaction.

In sum, accordingly to all the data obtained, it seems that GEMST acts against the inflammatory response reducing or abolishing deleterious processes induced by the I/R procedure and caused by excitotoxicity (microglia, astrocytes, monocytes, NK, macrophages). This is in agreement with our postulates because PLL-Compounds of GEMST were included in the formulation in order to treat specifically the deleterious mechanisms induced by glutamate and radical oxygen species. The data are also in agreement with previous results obtained in other models of Multiple Sclerosis (Mangas et al., 2006, 2008, 2010), Amyotrophic Lateral Sclerosis (Nicaise et al., 2008) and in clinical trials (Geffard et al., 2010, 2011).

GEMST could be suggested as a preventive therapy and post-stroke treatment but not as a shock immediate drug after stroke. Many patients suffering an ischemia use to have more than one episode, so a possible therapy window is still open if PLL-compounds acts on late phase (see Deliverable 2.7.2).
Moreover, these data will be described in a paper that will be submitted to a scientific journal in February-April 2017, after the corresponding patent applications.

D 2.7.2. Report on the optimal dose, administration regimen, and route of administration of GEMST in experimental stroke

As previously shown in D2.7.1 GEMST is a new drug candidate homologue of previous compounds developed by GEMAC for the treatment of other pathologies like GEMSP (for Multiple Sclerosis treatment) and GEMALS (for Amyotrophic Lateral Sclerosis treatment), among others. In humans, these new drug candidates were conceived for a sublingual treatment. They are directly absorbed by the mucosa and delivered by the blood flow.
GEMST was conceived in order to treat the excitotoxicity environmental conditions of the brain tissue after an acute ischemic/reperfusion (I/R) process using a known model of stroke (transient middle cerebral artery occlusion (tMCAO)).
Dose Range
Previous dose range conducted in EAE (Mangas et al., 2008, 2010) and ALS (Nicaise et al., 2008) models shown that higher dose used was the most efficacious (7.5mg/day). Our large experience with previous Poly-L-Lysine (PLL) compounds showed that the optimal dose in animal models is similar to the dose used in humans. Maybe, this is due to the metabolic differences between species and because it is rapidly incorporated into the metabolism. It should be noted that the efficacious dose is only 7.5 mg/day (for the whole compounds). 95% of the PLL compounds weight of GEMST correspond to the PLL carrier which is not immunogenic and it is easily degraded without any adverse events as it will be explained below. Furthermore, it should be noted that the same quantities of small molecules and PLL but without linkage have not effect in animal models (Mangas et al., 2008). This occurs because the free components of the drug are rapidly incorporated into the metabolism and degraded (Mangas et al., 2006, 2010).
Therefore, the dose tested in humans (in the case of GEMSP and GEMALS) was always the best concentration, once a dose range was carried out (Mangas et al., 2006, 2010; Nicaise et al., 2008). Thus, we have used the same concentration than that used for previous compounds (7.5mg/day). Composition 1 (PLL linked to different vitamins), Composition 2 (PLL linked to different amino acids) and Composition 3 (whole formula: PLL linked to different vitamins and amino acids). We have focused our efforts on the different possible formulations which include 3 different compositions plus the vehicle control:
• Composition 1 (7.5 mg/ subcutaneous injection day).
• Composition 2 (7.5mg/ subcutaneous injection day).
• Composition 3 (7.5 mg/ subcutaneous injection day).
The use of different compositions demonstrated that the efficacy of GEMST Composition 3 was the best in order to treat the excitotoxicity occurring after the I/R process (see Figure 1)(see as well deliverable 2.7.1). As noted above the concentration of the different compositions were the same in weight, it means that Composition 1 or 2 versus Composition 3 have a lower efficacy and this means that in order to stop the deleterious processes of the model, Compositions 1 and 2 have less efficacy than GEMST Composition 3.
Figure 1. I/R occurring processes. Taken from the article of Tatro JB 2006 (Endocrinology 147 (3):1122–1125)

Route of Administration
In humans, the candidate must be kept under the tongue for around 3-5 minutes in order to be absorbed and hence this is not possible to be carried out in animals. Thus, we take in consideration the use of pumps in order to deliver the candidate (compositions 1-3) into the animal, but this is not possible because the different components of the GEMST must be kept at 4° C until their administration. Thus, the use of pumps is prejudicial for its conservation during long term treatments as it has been conducted in the tMCAO model for three weeks. In the case of the acute phase (24-48 hours after I/R) (Compositions 1-3) a single daily dose was administered, so a procedure more complicated than a simple subcutaneous injection was not taken into account due to undesirable side effects.
Therefore, the route of administration used with other homologues of GEMST (GEMSP, GEMALS,...) was considered the best in order to be used in our animal model. Consequently, a single subcutaneous injection (six days a week) was considered the best way to deliver the GEMST in the tMCAO animal model. This route has been the best way in other animal models and its efficacy has been demonstrated (Mangas et al., 2006, 2008, 2010; Nicaise et al., 2008).
Administration Regimen
Why have we administered a single daily injection of GEMST six days a week? All the information that we have collected during the past twenty years showed that our candidates remain at least 24 hours in the animals (unpublished and confidential data) even when small quantities of the compounds were administered. This is due to the fact that small molecules conjugated to PLL have a longer half-life. Furthermore, no side effects were described in previous studies even when higher doses were administered intravenously (study conducted by subcontracting, EVIC, DL50=0). In humans, an open clinical trial for Multiple Sclerosis was conducted. No side effects were observed in any patient after six months of treatment with GEMSP. Moreover, any biochemical, hematological nor hepatic parameter was altered (Geffard et al., 2011; Mangas et al., 2010).
We have conducted an acute and chronic toxicity with GEMST and as previously demonstrated with subcontracting DL50=0.
1) Acute Toxicity
After a single subcutaneous dose (five times chronic dose), the healthy rats were sacrificed 3 days later and organs were examined macroscopically. No alteration was observed on heart, lungs, liver, kidney, spleen, stomach or gut.
2) Chronic toxicity
A diary subcutaneous injection (7.5 mg) of Compositions 1-3 and vehicle were administered in healthy rats for a period of 38 days (> 3 human years). At the end, animals were sacrificed and organs were examined macroscopically. No alteration was observed on heart, lungs, liver, kidney, spleen, stomach or gut.
Weight evolution of animals was invariable and it was not linked to the PLL compounds administered on different healthy rats groups in comparison with vehicle group.

3) Immunotoxicity
Even if PLL is a non-immunogenic carrier, we wanted to control the immune response of the animals. Thus, antibodies levels (directed against conjugated GEMST compounds) were controlled using an ELISA Tests. No immunotoxicity was revealed since IgG and IgM were unaltered.

Conclusions
1) The best dose is that previously reported in other models with similar compounds. The efficacy of that dose in the tMCAO model developed here and in other models has been demonstrated: 7.5mg/day.
2) The route of administration is a subcutaneous injection.
3) According to our experience, a single daily dose (six days a week) is enough for an optimal treatment.
4) Furthermore, the toxicity results confirmed previous studies which shown DL50 = 0; no alteration of weight; no macroscopic alteration of heart, lungs, liver, kidney, spleen, stomach or gut was observed and finally no immunotoxicity was found.

Moreover, these data will be described in a paper that will be submitted to a scientific journal in February-March 2017, after the corresponding patent applications.

D 2.8.1. Report on the impact of innate lymphocytes invading the CNS in experimental stroke

We studied at 2, 5, 7 and 14 days after stroke the infiltration of lymphocytes within the brain. We observed a slow decrease over time of infiltrating lymphocytes (CD11b-) that peaked at 14 dpi (436±161, 1150±451; 1183±480; 1213±333 and 2827± 551 cells per ischemic hemisphere at 0, 2, 5, 7 and 14 dpi respectively). Subpopulations of CD4+, CD8+, NK and B cells were similarly represented at 2 and 7 days post ischemia. Interestingly by performing the analysis of the inflammatory infiltrate in the brain post stroke in young (2 months of age) and aged (>17 months of age) C57Bl/6 mice at 2 days post ischemia we noticed that aged mice, while suffering from increased disability and mortality after experimental stroke also had a significantly increased number of neutrophils infiltrating the ischemic brain. Since from literature it is known that in human stroke an increased peripheral blood neutrophil to lymphocyte ratio correlates with poor stroke outcome in terms of disability and stroke lesion volume (Buck et al.; Mangold et al.) we investigated further the role of the increased neutrophil to lymphocyte ratio in experimental stroke. Indeed we found that aged mice had in physiological conditions an increased neutrophil to lymphocyte ratio in the blood and had after stroke a significant increase of neutrophil infiltration. To further investigate the role in stroke outcome we adopted a neutrophil ablation strategy. Using the anti-neutrophil antibody Ly6G (1A8) at various dosages we did not observe a significant depletion of circulating neutrophils but rather an opsonization of neutrophils. Interestingly anti-neutrophil treatment ameliorated stroke related no-reflow phenomenon in the brain but was followed by a significant increase of hemorrhagic transformation of the ischemic brain lesion determining an overall lack of advantage in terms of disability and mortality compared to isotype treated aged mice.

WP 3: Evaluation of ion channels in immune cells as potential targets for modulation of stroke-induced inflammation
D 3.1.1. List of ion channels expressed in resting microglia, pro-inflammatory and anti-inflammatory microglia in brain slices of normal and stroke-subjected animals

In accordance to the Technical Annex, we have mainly focused on two major research topics, namely (1) we have identified the functional importance of ion channels in macrophages of distinct functional state, and (2) we have identified ion channels expressed by microglia in situ, i.e. in brain slices.
In a first set of experiments, we investigated bone marrow-derived macrophages, which have been pre-treated with lipopolysaccharide/interferon-gamma (as a model of pro-inflammatory M1 macrophages), with interleukin-4 (as a model of anti- inflammatory M2a macrophages) or with interleukin-10 (as a model of anti-inflammatory M2c macrophages). To date, we have revealed major differences between these macrophage populations in the expression of Kir2.1 inward rectifier K+ channels, Kv1.3 voltage-activated K+ channels, KCa3.1 Ca2+-activated K+ channels, G protein-activated K+ channels, Hv1 voltage-activated H+ channels, TRPM7 non- selective cation channels and swelling-activated Cl- channels. In addition, we have found that KCa3.1 Ca2+-activated K+ channels appear to be expressed solely by macrophages, but not by microglia in brain tissue. Thus, KCa3.1 channels may be considered as potential marker for invading macrophages after stroke. In contrast, KCa1.1 Ca2+-activated K+ channels, which are expressed by activated microglial cells in brain tissue, were undetectable in macrophages, and, thus, could be considered as potential marker for activated microglia after stroke. Furthermore, we have tested the specificity of certain ion channel inhibitors, which will be applied in functional tests in the future. In addition to ion channels in control M0 macrophages, in classically activated M1 macrophages and in alternatively activated M2 macrophages (see above), we have now identified the functional importance of ion channels in these macrophage populations in vitro. In particular, we have focussed on Kv1.3, TRPM2 and TRPM7 channels, because we have found in our studies last year that Kv1.3 and TRPM2 channels are upregulated in M1 macrophages, whereas TRPM7 channels are upregulated in M2 macrophages. In order to investigate the role of these ion channels, specific ion channel inhibitors have been identified and subsequently used in all functional studies. Margatoxin was used to block Kv1.3 channels, N-(p- amylcinnamoyl) anthranilic acid (ACA) was used to block TRPM2 channels, N-[(1R)- 1,2,3,4-Tetrahydro-1-naphthalenyl]-1H-Benzimidazol-2-amine hydrochloride (NS8593) and FTY720 were used to block TRPM7 channels.

In addition to ion channels in control M0 macrophages, in classically activated M1 macrophages and in alternatively activated M2 macrophages (see above), we have now identified the functional importance of ion channels in these macrophage populations in vitro. In particular, we have focussed on Kv1.3, TRPM2 and TRPM7 channels, because we have found in our studies last year that Kv1.3 and TRPM2 channels are upregulated in M1 macrophages, whereas TRPM7 channels are upregulated in M2 macrophages. In order to investigate the role of these ion channels, specific ion channel inhibitors have been identified and subsequently used in all functional studies. Margatoxin was used to block Kv1.3 channels, N-(p- amylcinnamoyl) anthranilic acid (ACA) was used to block TRPM2 channels, N-[(1R)- 1,2,3,4-Tetrahydro-1-naphthalenyl]-1H-Benzimidazol-2-amine hydrochloride (NS8593) and FTY720 were used to block TRPM7 channels.
In M1 macrophages, blockade of Kv1.3 and TRPM2 channels inhibited the production of reactive oxygen species as detected by fluorescence imaging using the dye DCFDA. Furthermore, the production of pro-inflammatory cytokines, e.g. TNF- alpha, and chemokines, e.g. CXCL2, by M1 macrophages were inhibited by Kv1.3 and TRPM2 channel inhibitors. In contrast, production of anti-inflammatory substances, e.g. IGF-1 or TGF-β, by M2 macrophages was unaffected by inhibitors of Kv1.3 or TRPM2 channels. Inhibitors of TRPM7 channels did not affect production of reactive oxygen species, pro-inflammatory cytokines or chemokines by M1 macrophages. TRPM7 channels were found to be required for IL-4-induced proliferation of M2 macrophages. Furthermore, specific TRPM7 channel inhibitors affected polarization of M2 macrophages. For example, TRPM7 channel blockade caused reduction of arginase activity and inhibited downregulation of TNF-alpha in M2 macrophages.
In addition, using the patch clamp technique we have recorded ion currents from microglial cells in cortex and striatum of adult mice. Microglial cells were visually identified using transgenic IBA-1/EGFP mice, which have been provided by partner 1. Using specific voltage protocols and/or specific ion channel agonists, the following ion channel types have been identified: K+ channels, namely Kir2.1, Kv1.3, KCa1.1, KCa3.1 and G-protein-activated K+ channels; H+ channels, namely Hv1; Cl- channels, namely swelling-activated Cl- channels; non-selective cation/TRP channels, namely TRPM2, TRPM4 and TRPM7 channels. Comparison of microglial ion channels in cortex and striatum revealed significant differences in the expression levels of KCa1.1 channels, i.e. the current density of microglial KCa1.1 Ca2+- activated K+ channels was significantly larger in striatum than that in cortex.
Intriguingly, KCa3.1 Ca2+-activated K+ channels, which are a hallmark of peripheral tissue macrophages, were not expressed in microglia of cortical or striatal slices of adult mice.
Technology used to obtain the above described data are presented in 2 published papers (below) where the support of EC is acknowledged:
Schilling T. & Eder C. Fluorescence imaging of intracellular Ca2+, Na+ and H+ in cultured microglia. Methods Mol. Biol., 1041: 147-161, 2013.
Schilling T. & Eder C. Patch clamp protocols to study ion channel activity in microglia. Methods Mol. Biol., 1041:163-182, 2013.

D 3.1.2. List of ion channels expressed in monocyte-derived macrophages in brain slices at different time points following invasion of monocytes into stroke-damaged brain.
We have performed in situ patch clamp recordings on microglia in mixed striatum/neocortex slices from healthy mice to investigate neuron-to-microglia signalling. We have found that a small percentage of microglial cells in the cortex were able to detect neuronal activity.
We have carried out experiments to identify functional ion channels in brain slices from adult (2-3 months old) and aged (18-24 months) mice, and have discovered several differences between microglia of both age groups, which are relevant to our future studies on stroke-exposed mice. In detail, we have identified voltage- and Ca2+-activated K+ channels as well as TRPM channels in microglia of striatum and neocortex from adult and aged mice. Differences between microglia of adult and aged mice were found in the cells’ passive membrane properties, i.e. in the mean input resistance and in the mean resting membrane potential of the cells. Furthermore, differences in the expression of voltage- activated, but not Ca2+-activated, potassium channels as well as in the expression of TRPM7, but not TRPM2, channels were found between microglia of adult and aged mice. Mean current densities of Kir2.1 inward rectifier K+ channels, Kv1.3 outward rectifier K+ channels and TRPM7 channels were significantly larger in microglia of aged mice compared to those of microglia in adult mice.

Furthermore, we have tested whether specific ion channel inhibitors identified in our recent functional studies on bone marrow-derived macrophages could also block the production of cytokines by microglial cells in brain slices. Inhibition of Kv1.3 channels with margatoxin or charybdotoxin in microglia reduced the production of the pro-inflammatory cytokine IL-6 and of the chemokines CCL2 and CXCL2. Upon inhibition of TRPM2 channels with N-(p- amylcinnamoyl) anthranilic acid (ACA), cytokine and chemokine production by microglial cells in brain tissue was also inhibited.

D 3.2.1. Report on the role of ion channels in ROS production by pro-inflammatory microglia and macrophages in ischemia-lesioned brain
We have investigated the role of ion channels in pro-inflammatory activities of microglia and macrophages with the aim to identify therapeutic targets in brain pathology. We focused on the role of ion channels in priming and activation of reactive oxygen species (ROS) production (Spencer et al., 2016).
Production of large amounts of reactive oxygen species (ROS) and subsequent oxidative stress play a pivotal role acute neurodegenerative disorder such as stroke. While activated microglial cells are the major source of ROS production in brain pathology ROS can have beneficial roles via regulation of cellular signalling mechanisms. Furthermore, intracellularly produced ROS can contribute to microglial neurotoxicity by enhancing the production of pro-inflammatory substances.
Activation of microglia with an appropriate stimulus, such as ATP, induces NADPH oxidase activity leading to the production of a certain amount of ROS. Pre-exposure of microglia to various agents, which do not cause ROS production themselves, can lead to significant enhancement of ROS production upon subsequent stimulation, e.g., with ATP. This so-called microglial priming represents one of the mechanisms leading to excessive ROS production and subsequent neuronal damage in brain pathology. It is now well recognized that brain aging, stroke and other neurodegenerative diseases lead to the formation of primed microglia, while the proinflammatory cytokine IFNγ has been identified as a microglial priming factor. In the recently published study (Spencer et al., 2016), we investigated IFNγ -induced priming of microglial ROS production.
In this study, we elucidated mechanisms underlying priming of microglial ROS production. We demonstrate that IFNγ -induced priming simultaneously stimulates three mechanisms, namely (i) upregulation of NADPH oxidase subunit NOX2, (ii) upregulation of NO production and (iii) reduction of intracellular GSH levels. All three mechanisms were found to be dependent on p38 MAPK activity, whereas p42/p44 ERK was not involved. Furthermore, by testing a variety of ion channel inhibitors, we identified Kir2.1 inward rectifier K+ channels as crucial regulators of IFNγ -induced priming of microglial ROS production. Blockade of Kir2.1 channels inhibited effects of IFNγ on GSH levels and NO production. However, unlike p38 MAPK inhibition, Kir2.1 channel inhibition did not affect the upregulation of NOX2. These data suggest that IFNγ-induced p38 MAPK and Kir2.1 inward rectifier K+ channel activity are not directly linked. We further elucidated underlying mechanisms by which Kir2.1 K+ channels inhibit priming of microglial ROS production, namely, (i) by inhibiting nitric oxide (NO) synthase-mediated enhanced NO production and (ii) by blocking the reduction of the intracellular antioxidant glutathione. Our data lead to the conclusion that microglia priming leading to excessive ROS production upon secondary stimulation represents a major risk factor for the development of neurodegenerative diseases and delayed degeneration following stroke. A better understanding of mechanisms underlying priming of microglial ROS production may lead to the development of strategies aiming at the reduction of microglia-induced neurotoxicity. We suggest that inhibition of Kir2.1 channels provides a potential therapeutic strategy to reduce microglial priming and subsequent enhanced oxidative stress in brain pathology including stroke.
The data from this deliverable are published in PloS One (Spencer NG, Schilling T, Miralles F, Eder C. Mechanisms underlying interferon-γ-induced priming of microglial reactive oxygen species production. PLoS One 11(9): e0162497, 2016).

D 3.2.2. Report on the functional role of ion channel in release of cytokines, chemokines, matrix metalloproteinases and other substances from activated microglia/macrophages in ischemia-lesioned brain.
In collaborative effort carried out by Partners 1 and 6, Dr. Schilling performed study in Lund on mice subjected to stroke with objective to characterize monocyte-derived macrophages, which are spontaneously recruited to the stroke-injured brain.
Twenty-eight mice (8 wild-type and 20 β-actin-GFP+ C57BL/C male mice) were subjected to 30 min MCAO. To be able to distinct between the monocyte-derived macrophages (MDM) and the resident activated microglia in the stroke-affected brain we transplanted homologous monocytes isolated from β-actin-GFP+ C57BL/C mice into wild-type mice one day after MCAO.
Whole-cell patch-clamp recordings was performed from GFP+ MDMs in acute brain slices at three and seven days post stroke to characterize MDM electrophysiologically. We measured the ion channel expression in monocytes directly after isolation, 2 days after transplantation (3 days after stroke) and 6 days after transplantation (7 days after stroke). The tested ion channels were the voltage activated potassium channels Kv1.3 and Kir2.1, the calcium-activated potassium channels KCa3.1 and KCa1.1, the calcium-activated non- selective cation channel TRPM4, the ADPribose-activated non-selective cation channel TRPM2 and the non-selective cation channels TRPM7, TRPA1, TRPV4, TRPV2, TRPV1 and TRPM3.
All channels, which we were able to investigate, seem to change their expression levels, when monocytes invade the stroke area, or differ from the expression in microglial cells in situ in healthy animals. Most interestingly, this is illustrated by the expression of calcium-activated potassium channels. Freshly isolated GFP+ monocytes had a moderate expression of the KCa3.1 channel, which increased 3d after stroke and seem to disappear 7d after stroke. In contrast, the large conductance channel KCa1.1 was not expressed in GFP+ monocytes at any stage.
This fits well with our previous data: The intermediate conductance channel KCa3.1 is expressed in cultured macrophages, but not in microglial cells in situ. Microglial cells in situ expressed instead the large conductance channel KCa1.1.
GFP+ monocytes had directly after monocyte isolation a low expression of the voltage-activated potassium channel Kv1.3, which increased 3d after stroke. This upregulation might be transient, but due to the low number of cells measured after 7d, this is difficult to interpret. From our previous experiments and published literature, we know that both macrophages and microglial cells express this channel type in abundance, when exposed to a pro- inflammatory environment. So a transient (or even sustained) upregulation of Kv1.3 channels in GFP+ monocytes after stroke is in agreement with previous findings.
The voltage-activated potassium channel Kir2.1 was expressed at low levels in freshly isolated GFP+ monocytes, but I was not able to detect this channel type in GFP+ cells in stroke animals.
This result came unexpected. Murine microglial cells in situ as well as macrophages in culture expressed in previous experiments this channel type quite consistently. At the moment it is not clear, why GFP+ monocytes didn’t express the Kir2.1 channel after stroke.
The non-selective cation channel TRPM2 was expressed at a very low level in GFP+ monocytes at the day of isolation. 3 and 7 days after stroke induction the invading GFP+ monocytes seem to upregulate this channel type.
TRPM2 has been described in the literature as being upregulated under pro- inflammatory conditions by macrophages and other immune cells. In previous experiments, TRPM2 was also upregulated in cultured macrophages exposed to pro- inflammatory stimuli and down-regulated by anti-inflammatory conditions. An upregulation of TRPM2 by GFP+ monocytes in a post-ischemic environment would correspond well to previous observations.
The calcium-activated transient receptor potential channel TRPM4 was not detectable in freshly isolated GFP+ monocytes. It also was not detectable 3d after stroke, but TRPM4 seems to be expressed in GFP+ cells 7d after stroke induction.
In our previous experiments, cultured macrophages upregulated TRPM4 channel activity, when treated with anti-inflammatory cytokines. Nothing is known about the TRPM4 channel expression of microglial cells in situ. It is a compelling thought that TRPM4 might be upregulated, when GFP+ monocytes switch their phenotype from pro- to anti-inflammatory.
Additionally, we did not detect TRPA1, TRPV4, TRPV2, TRPV1 and TRPM3 activity in freshly isolated bone marrow derived monocytes and found a low expression level for TRPM7 in these cells. However, TRPM2 and TRPM4 channels were expressed at higher levels when compared to those of isolated resting monocytes and of monocytes in mice 3 days after stroke. These data are summarized in table (see below) and under now preparation for publication. (Schilling and Kokaia, in preparation).

Ion channel activity in GFP+ cells
(n cells in patch clamp experiments) Freshly isolated GFP+ monocytes 3 days after stroke 7 days after stroke
Outward rectifying voltage activated potassium channel Kv1.3 low
(n=8) upregulated (n=3) downregulated (n=2)
Inward rectifying voltage activated potassium channel Kir2.1 low
(n=8) not detected (n=3) not detected (n=2)
Intermediate conductance calcium- activated potassium channel IKCa Moderate
(n=6) upregulated (n=3) not detected (n=2)
Large conductance calcium- activated potassium channel BK not detected (n=6) not detected (n=3) not detected (n=2)
ADPribose activated non-selective cation channel TRPM2 Low
(n=7) upregulated (n=3) upregulated (n=1)
Calcium activated non-selective cation channel TRPM4 not detected (n=8) not detected (n=3) upregulated (n=3)
Non-selective cation channel TRPM7 Low
(n=4) not tested not tested
Non-selective cation channel TRPA1 not detected (n=6) not detected (n=1) not tested
Non-selective cation channel TRPV4 not detected (n=5) not detected (n=1) not tested
Non-selective cation channel TRPV2 not detected (n=4) not tested not tested
Non-selective cation channel TRPV1 not detected (n=3) not tested not tested
Non-selective cation channel TRPM3 not detected (n=4) not tested not tested

D3.3.1. Report on the functional role of monocyte ion channels in regulating monocyte infiltration into the damaged brain, transformation of monocytes to macrophages, neurogenesis, interactions of monocytes with neurons, microglia, and astrocytes.

Over studies revealed that monocytes do not have specific ion channels which are activated in response to stroke which are not also present on microglia and might not get activated as well. In addition, there was lack of mouse model which would allow to distinguish histologically between monocyte-derived macrophages and activated microglia. Therefore, within this deliverable, we decided to explore the role of monocytes on the overall number of activated microglia/macrophages and also study inflamasomes. Increasing evidence support that multiprotein complexes known as inflammasomes contribute significantly to ischemic brain injury following stroke (de Rivero Vaccari et al., 2014; Trendelenburg, 2014). The NOD (nucleotide-binding oligomerization domain)-like receptor (NLR) pyrin-domain-containing 3 (NLRP3) inflammasomes are abundantly expressed in the brain and immune cells. Recent findings demonstrate the participation of NLRP3 in ischemic stroke (Fann et al., 2013; Yang et al., 2014; Tong et al., 2015; Zhu et al., 2016). These findings indicate that NLRP3 inflammasomes may play an important role in detecting cellular damage and mediating inflammatory responses to aseptic tissue injury during ischemic stroke. Thus, the pharmacological targeting of the NLRP3-mediated inflammatory response may help with the design of a new approach to develop therapeutic strategies for preventing the deterioration of cerebral function and for treating stroke. Hence, there is substantial interest in the discovery of potentially therapeutic inflammasome inhibitors for treatment of stroke. One such compound, MCC950, has been identified as a potent NLRP3-selective inhibitor, but is not yet available for clinical use.

Fenamate NSAIDs have been shown to inhibit IL-1β secretion from macrophages, although the significance of COX inhibition remains unclear. Using immortalized mouse bone marrow-derived macrophages (iBMDMs) we show that the fenamate class of NSAIDs inhibit the NLRP3 inflammasome via reversible blockade of volume-regulated anion channels in the plasma membrane. Moreover, we demonstrated that this treatment inhibit cognitive impairments in neurodegenerative model in rodents, which is important component of stroke-induced cognitive problems.

Thus, our data clearly indicate that offering a safe and rapidly translatable option to treat NLRP3-related inflammatory diseases. These data (Daniels et al., 2016) are now published in Nature Communication, 2016, 7:12504.

We have previously shown that depletion of brain from monocyte-derived macrophages could affect the brain environment and make it more pro-inflammatory which could explain the beneficial role of monocyte-derived macrophages for the functional recovery after stroke (Wattananit et al., 2016). Therefore, we decided to explore the inflammatory environment by quantifying the number of activated microglia/macrophages. To depict this population we used double-immunocytochemistry and stained for Iba1 and ED1. We carried out counting in SVZ, and the lesioned whole striatum (Fig. 5). In the striatum of both groups, control and MC-21 injected, counting of cells in the ipsilateral to lesion side revealed very strong and significant increase of the number of activated microglia/ macrophages as compared to contralateral intact side. However, in the SVZ which is the main source for newly formed neuroblasts, only MC-21 injected group showed clear increase although also in the control group we detected tendency of the increase but with very high variability. Since at 1 week time point, the number of circulating monocytes and also the number of infiltrating monocytes in the brain is substantially decreased (Wattananit et al., 2016), most likely the increase of activated macrophages in the SVZ is accounted to the number of the increased activated microglia.

Fig.5. Effect of depletion of monocyte-derived macrophages on number of activated Iba1+/Ed1+ macrophages in the striatum, and SVZ.

Post-stroke environment is also defined by the number of activated astrocytes. It is well- known that stroke-induced lesion leads to activation of astrocytes and formation of glial scar. Since major marker of the activated astrocytes, GFAP, is labels the whole elaborated body and extension of the astrocytes, it is virtually impossible to identify and count individual glial cell in the lesioned tissue. Therefore, to assess the astrocytic reaction, we measured the are of the brain tissue covered by GFAP immunoreactivity. This quantification revealed that in both groups, there was clear increase of the area covered by astrocytes but there was no difference between saline and MC-21 injected groups, respectively (Fig.6).

Fig.6. Effect of depletion of monocyte-derived macrophages on number on area covered by GFAP immunoreactivity.
We next examined how the depletion of infiltrating monocyte-derived macrophages affected the brain tissue milieu. For this purpose, animals subjected to stroke and MC-21 or saline treatment were killed at 1 and 2 weeks after the insult, and the ipsilateral hemispheres were processed for quantitative PCR analysis of genes with pro-inflammatory or anti- inflammatory function. At 1 week after stroke, when circulating monocytes were completely depleted, we did not detect any dramatic change in the expression of inflammatory cytokines between saline and MC-21 injected groups. However, at 2 weeks time point we observed significantly decreased expression of CXCL2 in the MC-21 injected group and it was also significantly decreased as compared to 1week group of the same treatment (Fig. 7). Similarly, IGF 1 expression was decreased in MC21 group at 2 weeks as compared to MC21 group at 1 week. Other tested markers were not altered.

some genes did not changed at all (no differences comparing ipsi and contra hemispheres, MC21 and saline treatments and the 2 time points): CCL2, Ncan, Acan, MMP13, Brevican, Vescican, CXCR4, TNFaR1a, GDNF, GDNFRa1, GDNFRa2, checked only in the ipsilateral hemisphere; BDNF, PPARG, CCL22, IL13, checked in both hemispheres.
- some genes showed a consistent increase in MC21 and saline treated groups in the ipsilateral hemisphere compared to the contralateral hemisphere: IGF1, Iba1, IL6, IL12a.
- some genes showed significant changes ipsi/contra only in saline or MC21 treated mice: Arg1, NOS2, Ym1, Cxcl13, IL10, IL1β at 1wk significantly unregulated in ipsi only in saline group; TNFa both 1 and 2 wk significantly unregulated in ipsi only in saline group; TGFβ1&TGFβ2 at 1wk significantly unregulated in ipsi only in MC21 group .
- some genes significantly changed between 1 and 2 weeks: VEGF significantly increased from 1 to 2 weeks in the ipsi hemisphere both in saline and MC21 groups. Arg1, CXCL12 significantly decreased from 1 to 2 weeks in the ipsi hemisphere of MC21 group; GFAP, IL12a significantly increased from 1 to 2 weeks in the ipsi hemisphere of MC21 group.
- some genes significantly changed between MC-21 and saline treated mice: significant reduction in the ipsi hemisphere of MC21 compared to saline treated mice in Ym1, Cxcl13 at 1 week time point and in TGFβ 1&TGFβ2, Nos2, Arg1, VCAM, CD163, TNFαR1b, CXCL12 at 2 weeks time point.
- some genes changed both during time and because of the treatment: CXCL12, Arg1.

Our findings reveal a critical role of the monocyte-derived macrophages infiltrating to the injured brain early after the insult might contribute to different steps of neurogenic response from SVZ. Ths effect might be mediated by transition from pro-inflammatory to anti-inflammatory bias during the initial phase after stoke and modulation of the inflammatory tissue environment. Whether these changes in neurogenic response contribute to previously observed post-stroke recovery-promoting effect of monocyte- derived macrophages will require further investigations. However, Moreover, future studies will show whether this physiological repair mechanism can be potentiated by increasing the homing of macrophages, derived from endogenous or grafted monocytes in the peripheral blood, to the ischemically injured brain.

WP 4: Role of activated astrocytes in cell genesis and plasticity and functional recovery after stroke

D 4.1.1. Report on the effect of GFAP and vimentin ablation in astrocytes on neuronal response to oxygen and glucose deprivation
Oxygen and glucose deprivation (OGD) induces intermediate filament reorganization
Under normal culture conditions, the intermediate filament bundles in astrocytes were composed of vimentin, nestin and GFAP in varying proportions (Fig.1a). OGD induced intermediate filament reorganization. We observed cells with condensed bundles of intermediate filaments and many cells showed a diffused non-filamentous pattern when visualized with antibodies against GFAP, vimentin and nestin. This phenomenon was reversible and during reperfusion, the intermediate filament bundles became again clearly visible and more abundant (Fig. 1a). These changes were present in cells with regular nuclear morphology, indicating that they were not a consequence of cell death. Prior to OGD, the majority of astrocytes showed vimentin positive bundles of intermediate filaments, while GFAP containing bundles of intermediate filaments were undetectable in 25% of astrocytes and 23% of astrocytes were highly GFAP positive (Fig. 1b). At the end of OGD, the fraction of GFAP negative astrocytes increased (p<0.01). GFAP immunoreactivity was recovered during reperfusion with the number of GFAP positive astrocytes at 2 hr of reperfusion returning to their basal level. The fraction of astrocytes with high GFAP immunoreactivity decreased after OGD (p<0.01) and increased again during reperfusion (p<0.001, Fig. 1b).

Figure 1. After exposure to 18 hr OGD, astrocyte cultures were fixed immediately (OGD+0) or maintained under normal culture conditions for 2 hr (OGD+2h). Astrocyte cultures maintained under normal culture conditions served as a control (Ctrl). (a) Intermediate filaments were visualized by antibodies against GFAP (blue), vimentin (red) and nestin (green). In some astrocytes, the immunostaining showed condensed filament bundles (arrows), whereas in other astrocytes the staining was diffused (arrowhead). Scale bar, 50 µm. (b) Fraction of cells without visible GFAP filament bundles and those highly positive for GFAP in astrocyte cultures maintained under the three conditions. Mean ± SEM, p<0.05,
GFAP-/-Vim-/- astrocytes are more sensitive to OGD
Under normal culture conditions, GFAP-/-Vim-/- astrocytes were smaller than wild-type (p<0.001, Fig. 2a), as demonstrated before (Lepekhin et al. 2001). Both wild-type and GFAP-/-Vim-/- astrocytes showed a decrease in cell size during OGD (to 57%, p<0.001 and 72% p<0.05 of original cell size, respectively; Fig. 2a). At 2 hr of reperfusion, their cell size remained smaller. Under normal culture conditions, astrocytes formed a monolayer of cells in close contact with each other. Bright membrane staining, indicative of extensive areas of cell-cell contact, was lost during OGD in both GFAP-/-Vim-/- and wild-type astrocytes (Fig. 2b). During reperfusion, adjacent cells re-established cell-cell contact in some areas whereas larger gaps remained in other areas (Fig. 2b). Western blot analysis showed substantial reduction in the amount of connexin 43, a component of gap-junctions, after OGD, and this reduction was more pronounced in GFAP-/-Vim-/- astrocytes (p<0.05; Fig. 2c).

Figure 2. After exposure to 18 hr OGD, astrocyte cultures were fixed immediately (OGD+0) or maintained under normal culture conditions for 2 hr (OGD+2h). Astrocyte cultures maintained under normal culture conditions served as a control (Ctrl). Cell membranes were labeled with a green-fluorescent lypophilic dye. (a) Cell area of individual astrocytes. (b) Representative images; lypophilic dye (green), DAPI-stained cell nuclei (blue); arrows, areas of bright membrane staining; arrowhead, a cell free space. Scale bar, 50 µm. (c) OGD-induced decrease in the amount of connexin 43 as determined by Western blot analysis. Mean ± SEM, * p<0.05, ** p<0.01, *** p<0.001

The extent of cell death measured as LDH release was small during OGD and increased during the early reperfusion. Immediately after OGD, and at 2 hr of reperfusion, the extent of cell death was higher in GFAP-/-Vim-/- than wild-type astrocytes (p<0.05; Fig. 3a). When cell metabolic activity was assessed by the MTT assay, we observed a decrease in cell metabolic activity/viability at 2 hr of reperfusion and this was more pronounced in GFAP-/-Vim-/- than wild-type astrocytes (p<0.01, Fig. 3b). At 7 hr of reperfusion, GFAP-/-Vim-/- astrocytes seemed to have recovered to a similar extent as wild-type astrocytes (Fig. 3b). Thus, GFAP-/-Vim-/- astrocytes are more sensitive to OGD and early reperfusion phase, but later on show a similar recovery as wild-type astrocytes.

Figure 3. Astrocyte cultures were exposed to 18 hr OGD and different length of reperfusion. (a) Cell death expressed as the percentage of LDH released (b) Metabolic activity/viability of cells as determined with a MTT assay after 18 hr OGD and 2 or 7 hr of reperfusion. Controls (Ctrl) were maintained under normal culture conditions for 7 hr. Values are normalized to the controls prior to reperfusion. Mean ± SEM, * p<0.05, *** p<0.001

GFAP-/-Vim-/- astrocytes are less neuroprotective
Neurons are more sensitive to OGD than astrocytes and hence even OGD of short duration induces substantial neuronal death and neurite fragmentation (Almeida et al. 2002). It is known that astrocytes are protective to cocultured neurons under OGD (Jones et al. 2011). In order to evaluate the effect of intermediate filament ablation on astrocyte-mediated neuronal survival, we cultured cortical neurons obtained from wild-type mice on top of confluent astrocyte monolayers obtained from wild-type or GFAP-/-Vim-/- mice. The number of neurons in control conditions did not differ in cocultures with wild-type or GFAP-/-Vim-/- astrocytes (Fig. 4). After 2 hr of OGD followed by 2 hr of reperfusion, the neuronal death was more pronounced when neurons were cocultured with GFAP-/-Vim-/- compared to wild-type astrocytes (54% vs. 23%, respectively, p<0.01, Fig. 4). This indicates that GFAP-/-Vim-/- astrocytes are less neuroprotective.

Figure 4. Cocultures of wild-type neurons on wild-type or GFAP-/-Vim-/- astrocytes were exposed to OGD for 2 hr followed by 2 hr of reperfusion. (a) Neurons were visualized by antibodies against β–III-tubulin (green), nuclei were labeled with DAPI (blue). Scale bar, 50 μm. (b) Number of live neurons defined as β-III-tubulin positive cells extending neurites with an intact nuclear morphology. Mean ± SEM, ** p<0.01

ROS elimination is impaired in GFAP-/-Vim-/- astrocytes
Oxidative stress is one of the causes of cell death during ischemia-reperfusion (Sugawara and Chan 2003). We observed an increase in intracellular ROS in astrocytes after 20 hr of OGD followed by 2 hr of reperfusion and this increase was more pronounced in GFAP-/-Vim-/- astrocytes (p<0.001, Fig. 5a). To assess the ability of astrocytes to eliminate ROS, we measured the accumulation of ROS in astrocytes after the addition of exogenous hydrogen peroxide. After 30 min of exposure to 100 μM hydrogen peroxide, the accumulation of ROS was higher in GFAP-/-Vim-/- astrocytes compared to wild-type astrocytes (p<0.001, Fig. 5b). These data indicate that astrocyte intermediate filaments play a role in ROS elimination after OGD or hydrogen peroxide exposure.

Figure 5. (a) Intracellular ROS in astrocyte cultures exposed to 20 hr OGD followed by 2 hr reperfusion (OGD+2h). Astrocyte cultures maintained under normal culture conditions served as a control (Ctrl). (b) Intracellular ROS in control astrocyte cultures (Ctrl) and cultures treated for 30 min with 100 μm hydrogen peroxide (H2O2). Mean ± SEM, ** p<0.01, *** p<0.001

In summary, we present evidence that GFAP and vimentin are important for astrocyte survival and their ability to cope with oxidative stress in an in vitro model of ischemic stroke. We further show that the exposure to exogenous oxidant (hydrogen peroxide) leads to a higher ROS accumulation in GFAP-/-Vim-/- astrocytes, implying their decreased ability to eliminate ROS. Astrocyte intermediate filament system modulation could be a novel and interesting approach in designing the treatment for ischemic stroke and/or neurodegenerative diseases.

D 4.1.2. Report on the effect of complement-derived peptides C3a, C5a and C3adesArg on astrocyte response in an in vitro stroke model

First, we evaluated the effect of ischemia on the expression of C3a receptor (C3aR) expression. We found that in astrocytes exposed to chemical ischemia for 2h, C3aR mRNA levels were increased compared to non-ischemic control astrocytes. After 4 h of recovery, C3aR mRNA levels were reduced in comparison to ischemic astrocytes but still remained elevated compared to non-ischemic control astrocytes. Western blot analysis showed that C3aR protein levels were higher in astrocytes subjected to 2h ischemia followed by 24 h recovery as compared to non-ischemic control astrocytes. These results demonstrate that ischemia up-regulates C3aR mRNA and protein levels in cultured cortical astrocytes, and indicate that ischemic stress increases the responsiveness of astrocytes to C3a.
To study the effect of C3a on astrocyte activation, we therefore performed immunocytochemistry and Western blotting using antibodies against GFAP. Immunostaining of cells with antibodies against GFAP demonstrated considerably higher GFAP signal intensity in astrocytes subjected to mild chemical ischemia and allowed to recover for 24h as compared to non-ischemic control astrocytes. Western blot quantification by densitometry confirmed that chemical ischemia increased GFAP protein levels in astrocytes. In contrast, GFAP levels in astrocytes treated with C3a during the recovery period did not differ from non-ischemic control astrocytes. These findings suggest that C3a attenuates astrocyte activation in response to ischemia.
To investigate whether the reduced stress response in C3atreated astrocytes is associated with effects on their survival, we examined ischemia-induced astrocyte death in the presence or absence of C3a. As only minimal cell death was induced by mild chemical ischemia (less than 2% TUNEL positive cells), we subjected astrocytes to severe chemical ischemia for 5h, which kills about one quarter (24%) of astrocytes. C3a treatment during severe chemical ischemia reduced the fraction of dead cells. Next, we measured caspase-3 activation by Western blot analysis and found that the signal intensity ratio between cleaved and pro-caspase-3 was lower when astrocytes were subjected to ischemia in the presence of C3a, suggesting that C3a inhibits ischemia-induced apoptosis of these cells. The protective effects of C3a were confirmed in another in vitro model of severe ischemia, namely astrocytes subjected to OGD.
To further confirm that the protective effects of C3a are directly mediated by caspase activation and to determine the role of ERK signaling therein, we used Z-VAD-FMK (pan-caspase inhibitor) and U0126 (inhibitor of ERK phosphorylation). We found that the inhibitors increased astrocyte survival after ischemia to the same extent as C3a. The combination of C3a with any of the inhibitors did not lead to a more pronounced effect on astrocyte survival compared to C3a or inhibitor alone. Jointly, these results show that the protective effects of C3a astrocytes are mediated by ERK signaling and caspase activation. To determine whether the effects of C3a on astrocyte survival after ischemia are dependent on the intermediate filament (IF) proteins GFAP and vimentin, we used GFAP-/-vim-/- astrocytes. We found that C3a increased astrocyte survival after ischemia regardless of the expression of GFAP and vimentin. These results demonstrate that the protective effects of C3a on astrocytes are independent of IF.
We have also assessed the potential involvement of C3adesArg and C5a signaling in astrocytes. Ligand binding assay showed no specific binding for C5a/C3adesArg to murine astrocytes in culture and immunocytochemical analysis showed no positive signals for the C5a or C3adesArg receptor in these cells, indicating the absence of C5a or C3adesArg- dependent regulation. The expression levels of C5a receptor (C5aR) and second C5a-like receptor (C5L2) by murine astrocytes, both under control and ischemic condition, are below the detection limit of RT-qPCR analysis. Jointly, these results show that murine astrocytes express C5aR and C5L2 at very low levels which are not increased in response to ischemia. Thus, further investigation into the effects of C5a and C3adesArg on astrocyte response to ischemia is not warranted.

In summary, these results provide evidence that the complement-derived peptide C3a modulates the response of astrocytes to ischemic stress such that it reduces the reactive phenotype of these cells and renders them more resistant to ischemia-induced cell death. These findings reveal new important links between complement signaling and brain ischemia, highlight new complement functions on astrocytes and show that targeting C3aR with agonists is a new strategy for protecting the brain tissue from ischemia-induced damage.

D 4.2.1. Report on the effect of astrocyte intermediate filaments in intracellular vesicle trafficking and its role in Notch signalling

Notch signaling suppresses neuronal differentiation through cell-cell contact (58). Here, we investigated the possible involvement of Notch signaling in the GFAP–/–Vim–/– astrocyte-mediated neuronal differentiation of wild-type neurospheres and neural stem/progenitor cells. We co-cultured P2 wild-type and GFAP–/–Vim–/– astrocytes with Notch reporter cells and found less Notch signaling between GFAP–/– Vim–/– astrocytes and reporter cells than between wild-type astrocytes and reporter cells.
Next, we assessed the expression level of Jagged1, the principal Notch ligand, in GFAP–/–Vim–/– and wild-type astrocytes by quantitative real-time polymerase chain reaction (qRT-PCR) in three independent experiments. Expression of Jagged1 was down regulated by 40% in GFAP–/–Vim–/– astrocytes. Western blot analysis confirmed that GFAP–/–Vim–/– astrocytes contained less Jagged1 protein. Fluorescence activated cell sorting (FACS) analyses of Jagged1pos astrocytes showed comparable amount of cell membrane bound Jagged1 on wild-type and GFAP–/–Vim–/– astrocytes (46.9 ± 2.4 and 40.9 ± 4.1 mean fluorescence intensity, respectively). Thus, although the total amount of Jagged1 in GFAP–/–Vim–/– astrocytes is reduced, the membrane-associated fraction is not altered.
Both general endocytosis and Jagged1-specific endocytosis is decreased in GFAP–/–Vim–/– astrocytes

Notch ligand and receptor availability is known to be regulated by endocytosis and membrane trafficking and we have previously shown that intermediate filaments are important for astrocyte vesicle trafficking dynamics and IFN-gamma induced mobility of MHC class II compartment. Thus, we investigated both the general endocytosis as well as the endocytosis of Jagged1, which is important for eliciting a Notch signal. In GFAP–/–Vim–/– astrocytes, we observed a general reduction in endocytosis as shown by FACS analysis of the uptake of dextran-coated beads and a decrease in the Notch ligand-mediated internalization of the Notch extracellular domain. Also, the number of Jagged1pos vesicles was reduced in GFAP–/–Vim–/– astrocytes, suggesting that reduced endocytosis of Jagged1 in intermediate filament-deficient astrocytes might be the cause of decreased Notch signalling from GFAP–/–Vim–/– astrocytes to neural stem/progenitor cells.
To determine the efficiency of Notch signaling from GFAP–/–Vim–/– astrocytes specifically to neural stem cells, we transfected adult mouse neural stem cells with a Notch reporter. We found that the Notch signaling activity in neural stem cells co-cultured with GFAP–/–Vim–/– compared to wild-type P2 astrocytes was reduced by 78%. Next, we assessed Notch signaling in P4 neurospheres, which contain both astrocytes and stem/progenitor cells, by quantifying the expression of the Notch intracellular domain (NICD) using an antibody specific for cleaved/active Notch. After 5 days of differentiation, GFAP–/–Vim–/– neurospheres showed a 19% decrease in NICD+ cells compared to wild-type controls (62 ± 2.4% versus 76 ± 0.1% NICDpos cells/well).
To determine whether adding Jagged1 to the system would abrogate the effect of GFAP–/–Vim–/ - astrocytes on neuronal differentiation, we allowed P4 neurosphere cells to differentiate in the presence of immobilized recombinant Jagged1-Fc or a control protein Fc. Under control conditions, neuronal differentiation was greater in GFAP–/–Vim–/– neurosphere cells. In the presence of Jagged1, however, neuronal differentiation of GFAP–/– Vim–/– neurospheres decreased to the level comparable to wild-type neurospheres.
To investigate if this decrease was specific to Notch-mediated signaling, we added DAPT, a gamma- secretase inhibitor that prevents cleavage and activation of the Notch receptor, to GFAP–/–Vim–/– neurosphere cells differentiating in the presence of Jagged1. The addition of DAPT abrogated the Jagged1-mediated decrease in neuronal differentiation of GFAP–/–Vim–/– neurospheres.
To determine the Notch signaling competence of individual astrocytes, we used real-time quantitative PCR (RT-qPCR) to analyze the expression of selected genes in individual mouse primary astrocytes. Individual astrocytes were subdivided into four groups based on their expression of the two selected ligands, Jag1 and Dlk2, and the receptor Notch1 mRNA. Thirty five percent of the astrocytes did not express any of the ligands or the receptor Notch1 and were therefore considered as Notch signaling incompetent (group A). Astrocytes which expressed one or both ligands were classified as competent to send Notch signals (group B, 27% of all cells). The cells expressing Notch1 mRNA were classified as competent to receive Notch signals (group C, 53% of all cells). Astrocytes that expressed both the receptor and at least one of the ligands were regarded as being competent to both send and receive Notch signals (group D, 15% of all cells). Thus, individual primary astrocytes differ with regard to their competence to send and receive Notch signals.
Next, we assessed the effect of GFAP and vimentin deficiency (GFAP-/-Vim-/-) on the Notch signaling competence of individual astrocytes. The fraction of Notch signaling incompetent (group A, 43% of all cells), Notch signal receiving (group C, 50% of all cells) and Notch signal sending as well as receiving (group D, 10% of all cells) competent astrocytes did not differ between wild-type and GFAP-/- Vim-/- single cell preparations. Fewer GFAP-/-Vim-/- astrocytes showed Notch signal sending competence (group B): 16% of all GFAP-/-Vim-/- cells versus 27% of wild-type cells.
We conclude that individual primary mouse astrocytes differ with regard to their Notch signaling competence. Whereas the majority of astrocytes are competent to receive Notch signals, only a minority of astrocytes are competent to send Notch signals. By using GFAP-/-Vim-/- cells we show that intermediate filament proteins regulate Notch signaling in individual astrocytes.

In summary, we show that endocytosis of Notch ligand Jagged1 by astrocytes plays an essential role in Notch signaling from astrocytes to neural stem/progenitor cells and this is dependent on normal function of the intermediate filament proteins GFAP and vimentin in these cells.

D 4.2.2. Report on the effects of attenuated astrocyte activation on stroke-induced neurogenesis

Astrocytes are key homeostatic regulators in the CNS with functions ranging from the regulation of blood flow and neurotransmitter recycling to defense against oxidative stress. In response to any CNS insult, astrocytes become activated. The response of astrocytes to ischemia is particularly important in the acute phase, when it limits the loss of neurons, but at a later stage activated astrocytes may inhibit the neural plasticity processes necessary for full recovery of function. We have previously reported that the complement peptide C3a promotes astrocyte survival after ischemic stress. Here we report on our findings on the effects of C3a on astrocyte activation and post-stroke neural plasticity including neurogenesis.
We evaluated the effect of C3a treatment on astrocyte intermediate filament protein expression after ischemia. As astrocytes become activated in response to stress or injury, they up-regulate the expression of intermediate filament proteins, in particular GFAP. To study the effect of C3a on astrocyte activation, we therefore assessed GFAP expression in these cells. Immunostaining with antibodies against GFAP demonstrated considerably higher GFAP signal intensity in astrocytes subjected to mild chemical ischemia and allowed to recover for 24 h as compared to non-ischemic control astrocytes. Western blot quantification by densitometry showed that chemical ischemia increased GFAP protein levels in astrocytes. In contrast, GFAP levels in astrocytes treated with C3a during the recovery period did not differ from non- ischemic control astrocytes. To determine whether the effects of C3a on astrocyte
survival after ischemia are dependent on the intermediate filament proteins GFAP and vimentin, we used GFAP-/-Vim-/- astrocytes which lack the intermediate filament system. In agreement with our previous findings, the absence of the astrocyte intermediate filament system decreased the astrocyte survival in response to the in vitro ischemia, with or without reperfusion. However, the C3a treatment increased astrocyte survival after ischemia regardless of the expression of GFAP and vimentin.
Jointly, these results show that C3a attenuates astrocyte activation in response to ischemia but the protective effects of C3a are independent of the astrocyte intermediate filament system.
As we have previously reported that C3aR signaling stimulates basal neurogenesis, we quantified newly born neurons (BrdU+/NeuN+) in the peri-infarct region in C3aR-/-, C3a-GFAP mice (i.e. overexpressing C3a in the brain) and their respective WT control mice. C3aR-/- mice had 25% fewer BrdU+/NeuN+ cells than C3aR+/+ mice 21 days after focal ischemia, whereas there were nearly 30% more BrdU+/NeuN+ cells in the peri-infarct region of C3a-GFAP mice compared with their WT littermates. The BrdU+/NeuN+ cells were distributed loosely within the cortex up to approximately 800 µm medially and laterally from the injury site. This observation points to the BrdU+/NeuN+ cells originating from a resident cortical stem cell-like cells rather than from SVZ-derived neural progenitor cells diverted from the rostral migratory stream. These results indicate that C3a signaling through its canonical receptor C3aR stimulates ischemia-induced cortical neurogenesis.

D 4.3.1. Report on the functional recovery after stroke in GFAP-/-Vim-/- mice

In order to investigate the effect of reactive astrocytes on neurological recovery after stroke, we compared the behavioral outcome during the recovery process between wild-type (WT) mice and GFAP-/- Vim-/- mice. For this purpose, we performed Rose Bengal induced photothrombosis to the forelimb motor cortex. The ischemic infarct was localized to the directly illuminated cortical tissue. Unlike increased infarct volume observed in GFAP-/-Vim-/- mice compared to that in WT mice 7 days after distal middle cerebral artery transection in a previous study, there was no difference in the lesion volume between WT and GFAP-/-Vim-/- mice 28 days after cortical photothrombosis.
In addition, the modest cortical lesion in the right forelimb area led to severe, however, comparable behavioral deficits of the left forepaw in both WT and GFAP-/-Vim-/- mice 3 days after stroke measured with foot-fault test and single pellet reaching test. Gradual recovery was observed in both WT and GFAP-/-Vim-/- mice; however, compared with WT mice, the motor performance of the left forepaw in GFAP-/-Vim-/- mice was reduced from day 14 to day 28 after stroke.

We further investigated contralesional CST axonal plasticity by injecting BDA into the forelimb area of the contralesional cortex, to anterogradely label the CST axons. In WT mice subjected to unilateral right photothrombosis in the forelimb cortical area, in the denervated left side of the cervical cord, BDA- labeled contralesional CST axons that crossed the midline of the spinal cord, and extended toward the ventral horn, were evident, while in GFAP-/-Vim-/- mice, BDA-labeled CST axons were rarely observed in the left denervated spinal cord. We measure the BDA-labeled axonal length on 30 consecutive cervical sections in each animal. Quantitative data showed that the total length of BDA-labeled CST axons was significantly reduced in GFAP-/-Vim-/- mice.

Reactive astrocyte generating CSPGs are important inhibitory components of neurite outgrowth in glial scarring following CNS injury. To investigate whether reduced CST axonal plasticity in GFAP-
/-Vim-/- mice is associated with CSPG, we examined the distribution of CSPG in both ipsilesional and contralesional cerebral hemispheres with immunohistochemistry using monoclonal CSPG
antibody CS56, which is specific for the glycosaminoglycan (GAG) portion of native CSPG with
epitope recognizing CS octasaccharide containing A-D tetrasaccharide sequence. The CSPG positive staining was present in the corpus callosum and striatum of GFAP-/-Vim-/- mice. In the contralesional cerebral hemisphere, CSPG expression increased as early as day 3 after stroke in both WT and GFAP-/-Vim-/- mice, while at day 14 after stroke, CSPG in the contralateral hemisphere decreased in WT mice, but further increased in GFAP-/-Vim-/- mice. Quantitative data showed that CSPG expression in the contralesional hemisphere at day 14 after stroke was higher in GFAP-/-Vim-/- mice than in WT mice. Interestingly, in the ipsilesional hemisphere, increased CSPG expression was observed in the ischemic lesion boundary zone (arrows) in WT mice; however, in GFAP-/-Vim-/- mice, CSPG expression was increased in the cortical area outer lesion boundary zone (arrowheads), but not in the lesion boundary zone. Quantitative measurements of CSPG positive areas in the ischemic lesion boundary zone and the cortical area outer lesion boundary zone showed differences between WT and GFAP-/-Vim-/- mice.

We cannot exclude the possibility, that phototrombotic stroke leads to a delayed, rather than reduced, CST axonal remodeling and neurological recovery in mice deficient for GFAP and vimentin, since four weeks post stroke these mice continued to improve their motor functions. In support of this possibility, we recently reported delayed but complete axonal regeneration after sciatic nerve lesion in GFAP-/-Vim-/- mice that implied altered response dynamics but a comparable outcome after peripheral nervous system lesion in mice deficient in the intermediate filament system in astrocytes and Schwann cells. Therefore, a study is ongoing to address the effect of attenuated reactive
gliosis on the functional recovery at later stages after stroke.

In summary, our current data suggest that attenuated reactive gliosis either delays or impairs neurological recovery by reducing CST axonal remodeling in the denervated spinal cord. This indicates that modulation of reactive gliosis post stroke may represent a therapeutic target for neurorestorative strategies.

D 4.3.2. Report on molecular mechanisms linking reactive gliosis and functional recovery after stroke

Here, we report on our findings on the molecular mechanisms linking reactive gliosis and functional recovery in neurodegenerative condition. We reasoned that for this task, chronic neurodegeneration affecting the entire motor cortex as well as the spinal cord would be more informative than stroke, as chronic and more global neurodegeneration enables more robust and unbiased assessment of the effects of pharmacological intervention on astrocyte activation such as quantification of protein and gene expression. Therefore, we chose a mouse model of ALS which has many similarities with stroke on molecular and cellular levels as well as allows accurate and reproducible long-term quantification of the clinical symptoms of the disease. Our findings represent an important step towards modulation of astrocyte activity in stroke. This report (D4.3.2) is closely connected with the report D4.4.3 on the effects of potential drugs on astrocyte activation after stroke, which uses a stroke experimental model.
ALS is a neurodegenerative disorder affecting lower motor neurons in the spinal cord and brain stem with no efficient treatment available. Astrocyte activation and reactive gliosis are prominent neuropathological hallmarks of ALS, as they are in stroke, and astrocytes have emerged as potentially key players in ALS pathogenesis. Neurotoxic effects of reactive astrocytes were implicated as an important part of the pathogenesis in both sporadic and familiar ALS. Our aim is to dissect the role of reactive astrocytes in stroke as well as ALS and identify potential therapeutic interventions.
Mice carrying a mutation in Vps54, a protein involved in vesicle recycling, develop ALS-like symptomatology and serve as one of the ALS models. To assess the role of reactive gliosis in ALS, we crossed these mice with GFAP–/–Vim–/– mice, which have genetically attenuated reactive gliosis due to the lack of astrocyte intermediate filament (nanofilament) proteins GFAP and vimentin.
Attenuation of reactive gliosis slowed down ALS progression. At the time when control ALS mice reached the end stage of the disease, many ALS mice with attenuated reactive gliosis exhibited a relatively mild motor deficit.
Astrocytes from ALS GFAP–/–Vim–/– mice compared to ALS astrocytes showed already at postnatal day 1.5 (i.e. long before the disease onset) a 3-fold up-regulation of C3 mRNA (the most up-regulated gene as seen by Affymetrics). This is compatible with a scenario that C3 has a neuroprotective role, and is supported also by results linking higher complement activation with slower ALS progression in mice.
In conclusion, we demonstrate that genetic attenuation of reactive gliosis (GFAP–/–Vim–/– background) in ALS mice that carry mutated Vps54 slows down ALS progression. Thus, astrocyte activation and reactive gliosis in ALS seem to be maladaptive and therefore a target for pharmacological modulation.
These results further clarify the role of reactive gliosis in neurodegenerative disorders. By using suitable chronic neurodegeneration model, we were able to determine if the conclusions inferred from the stroke model can be applied more generally, i.e. in neurodegenerative context across different neurological diseases. We also want to point out that the goals set for the D4.3.2. were already included in D4.3.1. In D4.3.1, we described the finding that genetic attenuation of reactive gliosis after stroke leads to a lower expression of chondroitin sulphate proteoglycans. In addition, the results presented in D4.3.2 connect to findings reported in D4.4.3 on the effects of potential drugs on astrocyte activation after ischemia.

D 4.3.3. Report on the effects of reactive gliosis on microglia activation after stroke

Microglia perform many functions in healthy and diseased brain, ranging from maintenance of homeostasis and regulation of neural plasticity to trophic support and neuroprotection. Although experimental evidence suggests that microglia are not a homogenous cell population, the knowledge about their functional diversity and its molecular basis is limited. We used single-cell gene expression profiling of freshly isolated cells from uninjured mouse hippocampus and hippocampus after partial deafferentation to assess the heterogeneity of hippocampal microglia/monocytes and determine their response to injury. Here we report that in the absence of injury, Cx3cr1 and the astrocyte marker GFAP were expressed in two non-overlapping populations of cells. Injury led to the co-expression of these markers both in the injured and contralesional hippocampus. Cells co-expressing astrocyte and microglia markers were also detected in sections from human brain affected by Alzheimer’s disease, Lewy body dementia and ischemic stroke. Our findings indicate that ischemic stroke and other types of brain injury lead to the appearance of cells that share molecular characteristics of both microglia and astrocytes, two cell types with different embryonic origin. This report shows results from two experimental model systems (hippocampal de-afferentation and activation of glial cells in vitro) and three disease situations in humans, i.e. ischemic stroke, Alzheimer's disease and Lewy body dementia. The experimental models that were used here were crucial for the identification of the dual identity subpopulation of glial cells, and the experiments on human tissues have implied that this cell subpopulation is present also in stroke and other neurological diseases. The specific functions of these cells remain to be identified.

The proportion of Ccr2 positive cells was not changed after injury whereas the mean expression levels of Ccr2 mRNA were reduced in cells isolated from the injured and contralesional hippocampus compared to individual cells isolated from the control hippocampus. Regardless of injury, more than 50% of Cx3cr1 positive cells also expressed Ccr2. Principal component analysis (PCA) showed that the Cx3cr1+Ccr2+ cells clustered together with the Cx3cr1 single positive cells, indicating that they are microglia/ macrophages. Thus, our results identify two subpopulations of Cx3cr1 positive cells, namely Cx3cr1+Ccr2+ cells and Cx3cr1+Ccr2– cells; these subpopulations do not show any differences in the expression levels of any of the other markers assessed. The expression levels of Cx3cr1 were relatively stable, we detected only a single cell that could be classified as Cx3cr1lowCcr2+, although the expression levels of Cx3cr1 mRNA in the Cx3cr1+Ccr2+ cells were almost two-fold higher compared to the Cx3cr1+Ccr2– cell population. In PCA, the majority of Ccr2+ cells appeared outside the microglia/ macrophage cluster, suggesting that the Ccr2 expressing cells comprise at least two distinct subpopulations.

Regardless of injury, a large fraction of Ccr2+ cells co-expressed vimentin (Vim) mRNA and over 25% of those cells also expressed GFAP mRNA. All cells positive for von Willebrand factor (Vwf) expressed also Vim mRNA. In the control and contralateral hippocampus, at least 50% of Vwf + cells expressed Ccr2 mRNA, the fraction of Vwf +Ccr2+ cells was lower among cells isolated from the hippocampus ipsilateral to injury.

Jointly, these results suggest that Ccr2 mRNA expression is not limited to the monocyte/macrophage population and can also be found in astrocytes and endothelial cells. Based on PCA results, the Vim+Ccr2+Vwf+ cells appear to constitute a homogenous population (all found in single cell cluster) as compared to cells positive for Vim, Ccr2 and GFAP mRNA and cells positive for Vim and Ccr2 mRNA, which show a much broader distribution pattern.

Our data show that whereas the expression of Cx3cr1 and GFAP define two non-overlapping populations of cells in the control hippocampus, injury leads to the co-expression of these markers in some cells both in the injured and contralesional hippocampus. Almost 80% of Cx3cr1+ microglia/macrophages expressed Slc1a3 in the control hippocampus and 44-75 % of microglia/macrophages expressed Slc1a3 after injury. The majority of cells expressing Slc1a3 were negative for Cx3cr1 in the control hippocampus but the fraction of cells co-expressing Slc1a3 and Cx3cr1 increased after injury from 12% in uninjured control to 27% (p<0.05), and 28 % in injured and contralesional hippocampus, respectively. Thus, Slc1a3 is expressed by microglia/macrophages and GFAP positive astrocytes isolated from the control, injured as well as contralesional hippocampus. Further, Ccr2 mRNA was co-expressed with GFAP (and Slc1a3) regardless of injury, supporting the above conclusion that Ccr2 mRNA expression is not limited to the monocyte population. In the heat maps with cluster dendrograms based on the expression profiles, cells co-expressing Cx3cr1 and GFAP are found solely in the microglia cluster.

To investigate whether CX3CR1 and GFAP co-expression can be detected in cultured cells and to avoid and possible cross-reactivity of antibodies used for marker visualization, we prepared primary astrocyte and microglia cultures from CX3CR1-GFP mice. We found that in both cell cultures, less than 5% of GFP expressing cells were positive for GFAP under control conditions (serum-free). Addition of LPS but not serum or oxygen-glucose deprivation led to a dramatic increase in the fraction of GFP+ cells that co-expressed GFAP in astrocyte as well as microglial cell cultures. As both primary cell cultures contain microglia as well as astrocytes (microglia constituted 0.8±0.30% of cells in the astrocyte cultures; astrocytes constituted 13.5±1.50% of cells in the microglia cultures) and microglia are prone to cell fusion, we addressed the possibility of fusion between the GFP+ microglia and GFAP+ astrocytes. We found that in astrocyte cultures treated with LPS, 56±0.2% of the GFP+GFAP+ cells contained two nuclei. These findings show that in primary cell cultures that contain astrocytes and microglia, LPS increases the fraction of cells expressing both microglial and astrocyte markers and at least some of these dual identity cells are the result of cell fusion.

Having established that cells co-express markers of microglia and astrocytes in injured mouse brains and in LPS-induced microglial cultures, we wanted to determine whether such cells are present also in human brain. We used tissue microarrays containing temporal cortex samples from Alzheimer’s disease (10 subjects), Lewy body dementia (10 subjects) and 10 control subjects as well as tissue microarray containing cortical samples from 14 ischemic stroke subjects on which microglia/monocytes and astrocytes were visualized by immunostaining with antibodies against AIF1 and against the astrocyte marker S100β, respectively. We detected cells that expressed both markers in samples from one subject affected by Alzheimer’s disease, one subject affected by Lewy body dementia as well as in grey matter samples from two subjects affected by ischemic stroke. The cells showed morphological features of astrocytes . We did not find such cells in the samples from control subjects. These results show that cells co-expressing microglia and astrocyte markers are present in the diseased human brain including brain affected by ischemic stroke.

These results further clarify the role of activated astrocytes and microglia in neurodegenerative disorders. As in D4.3.2 report, we show that the findings obtained in stroke have broader implications, i.e. that the cells with dual identity (exhibiting molecular features of both astrocytes and microglia) can be found not only in stroke but also in other neurodegenerative disorders such as Lewy body dementia or any other age-related cognitive retardations.

D 4.4.4: Report on the effects of inhibitors of astrocyte activation on functional recovery after stroke
Astrocytes are key homeostatic regulators in the CNS with functions ranging from the regulation of blood flow and neurotransmitter recycling to defense against oxidative stress. In response to any CNS insult, astrocytes become activated. As astrocytes become activated in response to stress or injury, they up-regulate the expression of intermediate filament proteins, in particular GFAP. The response of astrocytes to ischemia is particularly important in the acute phase, when it limits the loss of neurons, but at a later stage activated astrocytes may inhibit the neural plasticity processes necessary for full recovery of function. We have previously reported that the complement peptide C3a attenuates astrocyte activation in response to ischemia in vitro (previously reported in D4.2.2), and signalling through the C3a receptor stimulates peri-infarct neurogenesis in mice after experimental ischemic stroke. Here we report on the effects of intranasal treatment with C3a on functional recovery and astrocyte activation after ischemic stroke.

We evaluated the effect of C3a treatment on astrocyte intermediate filament protein expression after ischemia. Using fluorescently labelled C3a and epifluorescent imaging in live animals, we first confirmed that C3a can be delivered to the mouse brain through intranasal administration and can be subsequently detected in the brain tissue for at least 3 hours. Next, mice were subjected to photothrombotic stroke in the motor cortex, leading to substantial impairment of the forepaw function, and treated daily with C3a or vehicle (PBS) between days 7 and 21 post-stroke. Motor function was assessed once a week using a cylinder test and grid-walk test. Intranasal C3a treatment had no effect on infarct volume. In the grid-walking task, C3a-treated mice showed a tendency toward reduced number of right paw foot faults on days 14 and 21 compared with day 7, whereas no trend toward significant improvement was observed in PBS-treated mice. At all time points after stroke, both groups showed significant impairment with respect to the baseline performance. In the cylinder test, C3a-treated mice showed continuous improvement between days 7 and 21 such that on day 21 their frequency of right paw use for body support did not differ from baseline performance. The PBS-treated mice showed sustained impairment until the end of the testing period. These results show that intranasal C3a treatment can promote the recovery of forepaw function after ischemic stroke.

To assess the effect of C3a treatment on astrocyte activation in the peri-infarct region, GFAP expression in the peri-infarct region was determined by immunostaing with antibodies against GFAP and high content image analysis. In both groups of mice, we observed considerably higher GFAP positive area in the peri-infarct region as compared to the corresponding region in the contralesional hemisphere. However, the GFAP positive area in the peri-infarct motor cortex of C3a treated mice was significantly lower compared to vehicle treated mice.

Because neither of the treatment groups showed full recovery in terms of forepaw motor function as assessed by the grid walking task by 21 days post-stroke, we next asked whether longer intranasal C3a exposure could provide greater benefit for functional recovery and whether functional improvement would be sustained after cessation of the treatment. Starting on day 7 after motor cortex stroke induction, mice were treated with C3a or PBS for 3 weeks and behavioral performance was assessed until day 56 post-stroke. In the grid-walking task, both groups displayed a substantial degree of recovery over the two-month period. However, the extent and time course of functional recovery were markedly different between the groups. C3a-treated mice showed significantly fewer right foot faults compared with PBS-treated mice at days 14 and 56 post-stroke. C3a-treated mice also had a significant reduction in foot faults within the first week of the treatment, while control mice did not show significant improvement until day 28 post-stroke. The functional improvement of C3a-treated mice continued after the conclusion of the treatment period and by the final day of testing their performance did not differ from pre-stroke baseline levels. Performance of PBS- treated mice plateaued at day 42 post-stroke and did not reach baseline levels by day 56 post- stroke.

A similar positive effect of C3a treatment on post-stroke functional recovery was observed in the cylinder task. The average scores on the last day of testing showed only a trend toward a difference between groups, although the C3a-treated mice displayed a sustained functional improvement compared with day 7 at days 28 and 56 post-stroke while changes in performance of PBS- treated mice were inconsistent and not statistically significant. Also, paired analysis of individual mice showed that C3a-treated animals readily increased their affected paw usage between the treatment initiation and 4 weeks after the completion of the treatment period, while in PBS-treated mice, overall right paw impairment did not change only 3/9 of mice improved. Taken together, these data indicate that intranasal treatment with C3a supports faster and more complete motor function recovery, which is sustained beyond the treatment period.

WP 5: Functional role of new cells for recovery after stroke and modulation of inflammation

D 5.1.1. Report on tracing and dating of SVZ cells in the double transgenic stroke mice upon TAM administration

The functional role of neurogenesis after stroke is still unclear. We generated a NestfloxGFPfloxTK-IRES-LacZ transgenic mouse line using lentiviral vector technology. In this mouse, three stop codons were introduced at the end of the GFP gene, so that the TK gene can be translated only after a removal of the GFP segment. We here studied the NestfloxGFPfloxTK- IRES-LacZ mouse to determine, based on the expression of the Nestin-GFP transgene, the best suited protocol to administer Gancyclovir (GCV) to ablate neural precursor cells (NPC) in the subventricular zone (SVZ).

NestfloxGFPfloxTK-IRES-LacZ mice were subjected to 45 minutes long middle cerebral artery occlusion (MCAO). The expression of the Nestin-GFP at various time points after stroke in pathology was evaluated. In particular we were interested in observing whether Nestin can be re- expressed also by astrocytes that are known to partially recapitulate, in pathological conditions, phenotypic characteristics of NPCs (i.e. Nestin expression). Early after stroke, we observed the GFP signal mainly expressed at the lesion border including in some reactive astrocytes. The expression of the GFP by astrocytes increased over time up to 16 days post ischemia.

These results, together with the putative toxicity of GCV when administered during a pathological process, prompted us to choose a GCV protocol where the GCV was given for 4 weeks before the pathological process to the mice in order to exclusively ablate the NPCs. This is because, the GCV administration after the onset of stroke might induce the incidental ablation not only of NPCs in the SVZ but also of the proliferating astrocytes within the lesion border. This protocol has the advantage that GCV neither interferes with the Nestin re-expression by astrocytes, nor affects other inflammatory cells that massively proliferate after ischemia. Using this protocol, we have been able to study the effect of ablating NPCs in the acute phase of stroke but not the effect of NPC ablation in the chronic phase of stroke. To reach this latter goal, we thus have to ablate permanently the SVZ before stroke (with a modified GCV administration protocol). This was theoretically possible by crossing AspMCreERT mice with the NestfloxGFPfloxTK-IRES-LacZ mice and this is what we originally proposed. However, while studying in detail the AspMCreERT mouse we realize that AspM was expressed in adulthood only in a subset of NPCs. Thus, crossing the two mouse lines would have not allowed us to eliminate the NPC compartment. To reach our goal a contingency plan has been elaborated and we have now two options: crossing the NestfloxGFPfloxTK-IRES- LacZ with a DLX2 Cre mouse (to reduce the possibility that cells other than NPCs express the Nestin after stroke onset) or tracing the SVZ NPCs using the Nestin-Cre/ERT2 mouse that we have recently imported in the lab.

D 5.1.2. Report on the functional effect of NPC ablation in acute and chronic experimental stroke

We aimed at investigating the specific role of NPCs after acute stroke by using a custom made transgenic mouse – expressing the herpes simplex virus thymidine kinase (TK) under control of the 2nd intron enhancer of Nestin (2ndI-Nestin-Tk) – in which NPCs of the SVZ can be selectively traced and, then, specifically killed by subcutaneous (s.c.) or intracerebroventricular (icv) injection of ganciclovir (GCV). In particular s.c or icv GCV administration for 2 or 4 weeks in NestinTK mice, generated from line 7457, specifically and transiently ablates endogenous proliferating aNPCs, residing within the SVZ, without affecting hippocampal neurogenesis. Most importantly GCV administration in these mice did not induce microglia activation, gliosis and physiological alterations (no alterations of cell blood counts, serum electrolytes, arterial pressure) although it induced a reduction in weight gain.
Transient, 45 minutes long, middle cerebral artery occlusion (MCAO) was induced in 2ndI-Nestin-Tk and control mice after 28 days of s.c. GCV or saline treatment. Neurological disability, ischemic volume, endogenous neurogenesis, inflammatory mediators, glutamatergic striatal synaptic currents as well blood brain barrier integrity were assessed.
We found that post-ischemic functional deficits, measured using the modified neurological severity score were greater in the GCV-treated 2ndI-Nestin-Tk than in saline treated 2ndI-Nestin-Tk or saline or GCV-treated wild type mice (thereafter referred as control mice). GCV treatment of 2ndI-Nestin-Tk determined an increased infarct lesion volume and an increased number of hemorrhages compared with that observed after saline treatment of 2ndI-Nestin-Tk or saline or GCV treatment of WT mice. Survival of GCV treated 2ndI-Nestin-Tk mice at seven days was found to be significantly inferior to the other treatment groups. No difference regarding brain edema was found, but reduced blood brain barrier permeability to medium and high molecular weight compounds was noted. Furthermore, an increase of spontaneous excitatory post-synaptic currents in striatal neurons whose morphology was markedly deranged was measured in MCAO 2ndI-Nestin-Tk mice compared to controls; in vitro studies revealed that NPCs regulate striatal glutamatergic tone – via the release of the endocannabinoid anandamide acting, at both pre- and post-synaptic level, through the type-1 cannabinoid (CB1) receptor – mainly in response to inflammatory conditions. The increased mortality to MCAO was, thus, in vivo reverted in 2ndI-Nestin-Tk mice by intrastriatal administration of CB1 receptor agonists.
Our results indicate that ablation of neural precursor cells in acute stroke results in increased lesion volume, worsening of neurological disability, and decreased overall survival thus supporting a protective mechanisms exerted by endogenous neural progenitor cells. So far one of the protective mechanism exerted by SVZ NPCs has been elucidated being the capacity of such cells to protect neurons from glutamate-mediated excitotoxicity via the release of endogenous cannabinoids. Additional mechanisms seem, however, also implicated. In particular, our preliminary data suggest that SVZ NPCs might exert their protective action by regulating BBB permeability, which might be pathologically deranged in stroke when SVZ NPCs are altered due to the ischemic insults.
We have reported above the effect of neural stem cells ablation in the acute phase after stroke. Due to the difficulties in using the NestinTK mouse to ablate NPCs in the chronic phase after stroke, since Nestin is also re-expressed by astrocytes (see D5.1.1), we studied the NestinCreER-T2- YFP mouse where it is possible to specifically trace neural stem cells by administering tamoxifen before the induction of stroke. We observed that after stroke NPC from the SVZ migrate towards the ischemic lesion mainly differentiating into astrocytes (YFP+/GFAP+) at 10 and 30 dpi. Some SVZ NPCs also express doublecortin (YFP+/DCX+) but mainly migrate along the rostral migratory stream and only few doublecortin positive cells were seen close to the ischemic lesion at 10 and 30 dpi. Interestingly many YFP+ cells were observed to localize close to blood vessels in the peri-ischemic area retaining a mainly undifferentiated phenotype (GFAP-/DCX-). We tried to ablate NPCs using the NestinCreER-T2 crossed with the iDTR mouse recently imported from Partner 5, so to be able to kill selectively upon diphtheria toxin administration the NPCs of the SVZ. However we obtained no efficacious reduction of NPCs in basal conditions using the NestinCreER-T2-iDTR mouse so that we did not further proceed in inducing stroke in these animals since ablation of NPCs was not reliable.
We have however studied the effect of stroke in aged mice, since these mice suffer from a physiologically reduction of neurogenesis. Interestingly we observed that aged mice, despite have a similar ischemic lesion volume and edema at 3 dpi when compared to young mice. However aged mice have an increased neuronal loss in the peri-ischemic area, as well as suffer from significantly increased disability and mortality compared to young mice. Since we observed that neurogenesis is still reduced at 3 dpi, whether the reduced neurogenesis observed in aged mice could causally be relevant for the observed mortality and disability remains an issue to be investigated. In conclusion we observed that in the acute phase of stroke endogenous NPCs of the SVZ are protective by dampening glutamate toxicity in the striatum.

D 5.1.3. Summary on the blood-brain and inflammatory infiltrate alterations induced by NPC ablation

The evolution of ischemic stroke is not only depending on tissue damage induced by glutamatergic excitotoxicity but also, at least in part, by the progressive disruption of the Blood- Brain Barrier (BBB). There are two reasons why a relationship between SVZ NPC and BBB can be inferred in ischemic stroke. On one hand, there are several studies indicating that BBB dysfunction is one of the causative detrimental events in the sub acute phase of stroke; on the other hand, NPCs reside in close contact with vessels forming the BBB within the periventricular area.

We first analyzed the morphological features of the BBB in the NestinTK GCV treated transgenic mice that have ablated NPCs (Butti et al.). We found that the ablation of NPCs in the SVZ lead to the following structural changes: (i) an increase of the pericyte coverage (possibly due to cell hyperplasia and not hypertrophy) of the BBB forming vessel; (ii) an increase of extracellular deposition of collagen IV; (iii) a thinned BBB endothelial basal lamina; and, (iv) a decreases expression of tight junctions on the BBB as exemplified by the decrease of Zo-1 as well as a decrease of other junction proteins (Claud5). Contrariwise we did not find any alteration of the astrocytic end feet coverage of the BBB vessels. All these structural changes were paralleled - thus suggesting a causative role - by an alteration of the local function of the BBB that showed an increased permeability to low and medium sized solutes in the absence of modifications of microvessel density and length. While our results clearly indicate that NPCs actively participate in the maintenance of a semipermeable BBB (that is unique to the SVZ) rather than passively exploiting a pre- existent leaky vascular niche (Shen et al.), it still remains to be investigated how exactly pericyte hyperplasia dysregulated endothelial function and even more importantly how NPCs interact both with endothelial and pericytes in maintaining an immature local BBB.

We next tested whether the structural alterations present at the level of the periventricular BBB in NPC ablated mice might also affect ischemic stroke evolution and outcome. Contrariwise to what we had observed in physiological conditions, ischemic stroke in NPC ablated mice induced a reduced extravasation of medium and high molecular solutes compared to non-ablated control mice. How these alterations might relate to the worsening of the stroke outcome is not entirely clear. However, we can speculate that pericyte hyperplasia - induced by NPC ablation - might increase the no-reflow phenomena thus determining a reduced blood flow within the peri-ischemic areas. Besides, the reduced blood reflow might explain why the extravasation of endogenous or exogenous traces is reduced in the ischemic tissue.

D 5.2.1. Report on the morphological alterations of the pyramidal tract occurring in stroke after NPC transplantation: comparison of histology and MRI findings.

We have shown that mice undergoing middle cerebral artery occlusion (MCAO) and systemic NPC transplantation have a reduced ischemic lesion volume as demonstrated both by MRI imaging as well as by histological analysis. Importantly the NPC treated mice recover from stroke due to the ability of the transplanted cells to specifically reach the lesioned areas and buffer the excess of glutamate that occurs in response to the tissue damage. This mode of action, in turn, promotes robust neuronal plasticity in both the ipsilateral and contralateral hemisphere as documented by the local increase of dendritic arborization and of the axonal sprouting.
Using patch clamp technique on striatal sections from NPC-treated vs. control mice, we could study in collaboration with Dr. Diego Centonze in Rome, from a functional point of view, the ipsi- and contra- lateral neuroprotection exerted by the transplanted NPCs. We, in fact, found that ipsilesionally located transplanted NPCs induce an early inhibition of the ischemic hemisphere, indicated by the increased
GABA currents and lowered glutamate release, and at the same time, an increased excitatory tone in the contralateral hemisphere.

We subsequently dissected the molecular mechanisms sustaining the NPC-mediated protective function. We found that the buffering activity exerted by NPCs was possible due to (i) the expression on the NPC membrane of two glutamate ion transporters (i.e. GLT1, GLAST) and (ii) the up regulation of the GLT1 on the membrane of peri-lesional endogenous astrocytes. To verify whether GLT1 up-regulation is essential for the increased post-ischemic plasticity induced by NPC transplantation, we have inhibited - after NPC transplantation - the functionality of GLT1 by implanting in situ (at the level of the lesioned striatum) a cannula connected to a miniosmotic pump secreting dihydrokainic acid (DHK), a selective GLT1- transporter blocker. We observed that NPC transplanted MCAO mice treated with DHK did not show any clinical amelioration over 60 days post- treatment when compared to MCAO mice only NPC transplanted. Moreover, when we performed neurophysiological analysis, the NPC-induced post-synaptic enhancement of NMDA- dependent excitatory transmission was absent in the contralesional striatum of NPC transplanted MCAO mice treated with DHK. In fact, NPC transplantation failed to increase decay time and half width in MCAO NPC transplanted mice treated with DHK.

Finally, we found that transplanted NPCs do up-regulate GLT1 on the surface of endogenous astrocytes via VEGF production. VEGF was identified among several other growth factors expressed by NPCs via the selective pharmacological block of its downstream intracellular pathway. We thus found that systemically delivered NPCs might improve stroke outcome by promoting ipsilesional neuroprotection and, most importantly, long-ranged plasticity processes by regulating glutamate transporters.

Using resting state fMRI, alterations of the functional connectivity of the sensorimotor networks were monitored for twelve weeks following stroke. Within the first week, the interhemispheric connections were weakened and the intrahemispheric networks of both hemispheres were affected with no substantial recovery during the three months. Intracortical implantation of human NSCs two days after stroke completely stabilized the functional connectivity networks. Vitality of the NSC graft was monitored using bioluminescence imaging. When at four weeks the vitality decreased substantially, the functional networks deteriorated and approached the disturbed situation of the untreated mice. In conclusion, early stem cell implantation positively stabilizes functional connectivity networks but requires persistent viability of large grafts.

D 5.2.2. Validation of MRI methods to detect NPC in vivo upon transplantation in mouse stroke model

In collaboration with Gianvito Martino (P3) we had already reported last year that we are working on the functional connectivity network changes under conditions of neurogenesis ablation. These experiments are performed on transgenic animals (provided from Milano) where Nestin positive cells in the subventricular zone express TK. Functional connectivity network is analyzed in a group of animals under normal, healthy conditions. Then, six weeks after treatment with ganciclovir, when ablation is maximal, the measurement was repeated, and finally, a third recording took place six weeks after ganciclovir application was stopped. Ablation results in substantial decrease of interhemispheric connectivity between motor cortices and between visual cortices, while the correlation coefficient between the somatosensory cortices remains stable. Also the striatal connectivity is reduced by ablation.

Interestingly, after recovery of neurogenesis, the connectivity between striata and between visual cortices normalizes completely, while the interhemispheric connection strength between motor and somatosensory cortices increases even beyond the normal, pre-ablation value. The connection between motor and visual cortex is the only intra-hemispheric connection that is lost during the recovery period, despite its prior preservation during the ablation period. Presently, control experiments with non- transgenic littermates are prepared to exclude a direct influence by the ganciclovir itself.

These observations clearly indicate the long ranging influence of the SVZ neurogenesis on stable functional connectivity, particularly across the hemispheres. The effect goes far beyond the earlier described electrophysiological hyperexcitability exclusively of the striatum, noted after ablation.

We have also evaluated the possibility to detected transplanted monocytes/macrophages. Transgenic, labeled monocytes and macrophages, respectively, have implanted into the striatum of mouse brain to study their in vivo detectability and vitality after implantation. There is a substantial loss of cell vitality over the first two weeks, both for monocytes and macrophages, indicated by the decreasing bioluminescence signal. During the whole observation period, cell clusters were distinctly visible by 3D MRI.

D 5.2.3. Validation of MRI methods to assess tissue plasticity upon in vivo NPC transplantation in experimental mouse stroke.

In connection to the functional network alteration under pathophysiological conditions, we have established a high-resolution diffusion MRI protocol and applied it to structural connectivity studies of fiber tracking under stroke conditions.

On the ischemic hemisphere, severe alterations are observed. Within the first week after stroke, the fiber connection between the thalamus and the somatosensory cortex is completely lost. Instead, only completely unorganized fiber situation is detected. However, in the following weeks, there is substantial dynamics in the affected hemisphere pointing towards spontaneous regeneration processes. At week four after stroke induction, connectivity between thalamus and somatosensory cortex appears to be re- established. However, the fibers circumvent the ischemic core and reach the cortex rather tangentially instead of the original radial orientation. Presently, experiments are undergoing, combining structural and functional connectivity studies during 12 weeks after stroke induction to assess the effect of the structural derangement on the functional deficit, functional improvement or plasticity. Moreover, part of this ongoing study includes cortical implantation of human neural progenitor cells to investigate the therapeutic effect of stem cells modulation of the functional deficit and improvement.

We have established detailed protocols to assess viability and cell behavior after labeling monocytes or macrophages in culture with iron oxide nanoparticles, MRI contrast agents. To investigate the viability of cells after injection into mouse brain, we have isolated monocytes from transgenic animals expressing luciferase under beta-actin control. Iron uptake, tested using photometric iron analysis, was equal for WT and beta-actin derived monocytes and macrophages.

Macrophages, derived from the beta-actin luc transgenic animals, were characterized using flow cytometry. Comparison with the same cells, labeled earlier with iron oxide nanoparticles demonstrated that the labeling process has no serious effect on the functional characterization of the macrophages.

In conclusion, both monocytes and macrophages take up sufficient amount of iron so that they can be labeled with MRI contrast agents for sensitive in vivo detectability. This labeling procedure has no apparent strong effect on the functional characterization of the cells, as described by FACS analysis.

D 5.3.1. Quantitative bioluminescence signal discrimination from endogenous and grafted NSCs and from monocytes, first in vitro, then in vivo

Under point 1) the quantitative bioluminescence imaging (BLI) of endogenous and grafted NSCs has already been demonstrated, both in vitro and in vivo. The corresponding BLI of luciferase expressing monocytes was reported under 2). Here we present our approach to discriminate BLI signals from two different cell populations. We have a series of mutated luciferases available that span the whole wavelength spectrum from the green end to the near infrared end (luciferase rainbow). Transduction of one cell population with a luciferase with the maximum signal in the green while transducing a second population with a different luciferase emitting in the red spectrum range then requires spectral decomposition for the identification and quantification of the respective cell population’s BLI signal contribution. For a proof-of-concept study we have transfected HEK-293T cells with either red or green emitting luciferases. Such cell populations were mixed in stoichiometric ratios. Decomposition by unmixing of the emission spectra permitted the quantitative contribution of either population.

In the next step, these cell populations were transplanted into the cortex or into the striatum of nude mice. The two different locations were chosen to analyze the effect of higher signal absorption by cell location deep in the brain. We demonstrated that the stoichiometric mixture can be analyzed with very high accuracy. For the striatal location of the graft, the error bars are a bit larger but accuracy of quantification remains at a high qualitative level. Thus, these results convincingly demonstrate that with the approach of rainbow luciferases two different cell populations can be discriminated using quantitative BLI. At present this strategy is repeated to demonstrate the distinction of NSCs from already neuronally differentiated NSCs in one graft location. With the same approach, monocytes isolated from beta-actin-luc mice (see 2) above) can be separated from NSCs that carry a red-shifted luciferase.

D 5.3.2. Characterization of grafted and endogenous NSC dynamics in ischemic brain with/without modulation of inflammatory activity

Interleukin-34 (IL-34), an alternative ligand for Csf-1 receptor, is produced by keratinocytes in the epidermis and by neurons in the brain. Mice lacking IL-34 display a marked reduction of Langerhans cells in the epidermis and a decrease of brain microglia Partner 3 has analysed whether healthy IL34LacZ/LacZ mouse present an initial deficit in neurogenesis. Inhibitor of DNA Binding 1(Id1) antigen is highly present in neural stem cell and has been shown to be necessary for self-renewal, a characteristic of neural stem cell (also called B cells). No difference was highlighted in the SVZ of IL34LacZ/LacZ mice compared to control using this marker. We next evaluated fast cycling C-type cells also called transit- amplifying cells. We observed a small reduction of transit amplifying cells both in the SVZ and the dentate gyrus of the hippocampus. Finally we studied doublecortin (DCX) positive neural precursor cells also known as A-type cells or neuroblasts. Healthy brains were immunostained for DCX, no significant difference in the cell number was observed in the SVZ between groups. Analysis for DCX gene expression did not reveal major difference either. Further investigations of neurogenesis in the IL34LacZ/LacZ mice with reduced microglia were done in vitro. Following established procedures, self-renewal was assessed by continuous in vitro propagation and multipotentiality by differentiation induced by addition of serum and growth factor deprivation. Growth rate of NPC derived from IL34LacZ/LacZ or CSF1RKO did not differ from controls and multipotentiality of NPCs of IL34LacZ/LacZ was not diverse to control NPCs as confirmed by a similar presence of GFAP+, beta Tublin III+ and olig4+ cells. We also quantified stem cells and progenitors from primary adult IL34LacZ/LacZ mouse SVZ by performing a neural colony forming cell assay. At 17days in vitro (div), no difference among the B cells (colonies >1mm) was found. A 30% reduction was observed among cells having a limited self-renewal capacity (C cells) in NPCs of IL34LacZ/LacZ, recalling the small BrdU reduction found in vivo. IL34LacZ/LacZ NPCs were also analysed for the gene expression of marker such as EGFR, GLAST, DLX2, Dcx but no difference appeared.

In previous (Bacigaluppi et al. Brain 2009) and recent works (Bacigaluppi et al. J Neuroscience 2016) Partner 3 observed that intravenous transplantation of adult neural precursor cells (aNPCs) three days after experimental ischemic stroke results in selective accumulation and persistence over time of transplanted cells within the ischemic-, peri- ischemic brain tissue. Interestingly the transplanted cells retain over time and up to 60 days post- transplantation a rather undifferentiated phenotype localising close to blood vessels and in vicinity of inflammatory cells, in so- called ‘ectopic neurogenic niches’ (Martino et al. Nat Rev Neurosci 2016). Transplanted aNPCs retain an undifferentiated phenotype and secrete a wide variety of growth factors an in particular they seem secrete vascular endothelial growth factor (VEGF) that influences peri-ischemic reactive astrogliosis.

Interestingly enough the cross-talk between immune cells and NPCs and their progeny seems to determine the mechanism of action as well as the fate and functional integration of grafted NSPCs. To further investigate the effect of the tissue on the fate of transplanted NPCs we took advantage of a mouse model of demyelination with reduced inflammation by injecting the demyelinating agent lysolecithin in the adult spinal cord. (Mozafari et al. JCI 2015). Interestingly, coronal and longitudinal sections showed that locally transplanted mouse iPS- derived NPCs (miPS-NPCs) and also mouse epithelial NPC (mE-NPCs) extensively remyelinated the demyelinated spinal cord funiculus, as confirmed by the wide expression of myelin basic protein (MBP) in the adult MBP-deficient Shi/Shi Rag2–/– mice. miPS-NPC– and mE-NPC–derived oligodendrocytes harbored a typical morphology of bona fide myelin- forming oligodendrocytes with several T-shape ending processes. We also quantified the extent of remyelination based on MBP expression on coronal sections in Shi/Shi Rag2–/– mice. Based on MBP+ area, miPS-NPCs were able to produce MPB+ myelin to the same extent of mE-NPCs with 6,200 ± 1,582 µm and 6,200 ± 1,240 µm for miPS-NPC and mE- NPC, respectively, indicating that migration and remyelination were tightly correlated. Furthermore, GFP+ cells did not express protein zero (P0), a major protein of peripheral myelin, indicating their entire commitment to CNS myelin-forming cells. Grafted miPS-NPC generated compact myelin around host axons and restored nodes of Ranvier and conduction velocity as efficiently as CNS-derived mE-NPC.

WP 6: Development of preclinical protocol for novel regenerative therapy in stroke patients based on immunomodulation

D 6.1.1. Report on cellular distribution in the CNS

The distribution of autologous cells injected to the cerebrospinal fluid (CSF) of injured animals was examined by using β-actin-GFP mice as the monocyte donors. Three days following spinal injury, cells were injected into the CSF via two routes, ICV and LP. At 7 dpi the mice were sacrificed and the brain and 3 segments from the spinal cord (lesion site, rostral and caudal to the lesion site) were taken separately for analysis. We found GFP positive cells in high numbers in both the brain and lesion site segment in the spinal cord but only neglected number of cells in the rostral or caudal segments of the spinal cord. We concluded that autologous monocytes injected into the CSF will migrate preferentially into site of lesions, hence the cells home to the brain that was injured by the needle penetration during the course of the ICV injection.

Additional groups of animals were injected via LP, which is located caudally to the lesion site in the spinal cord. As expected, the LP-injected cells homed mainly to the lesion site and accumulated in similar amounts as was previously seen using the ICV injection. Concomitantly, cells were also found in the caudal segment of the spinal cord but in fewer amounts. Only a very limited number of cells were found in the rostral segment or in the brain suggesting that the CSF is an appropriate compartment for monocytes transplantation strategy.
Homing of the human monocytes sub-populations injected to the CSF of animals following spinal cord injury was investigated in the nude mice model. Nude mice were subjected to spinal cord injury and on day 3 post SCI they were treated with human monocytes separated into either CD16- or CD16+ subpopulations. The human monocytes were identified in the mice tissue using antibodies against human HLA- DR and antibodies against human CD45, both markers for human monocytes and leukocytes respectively. Human cells were detected only in the brain (that was injured by the procedure of ICV injection of cells) in naïve (no SCI) mice. In the SCI mice, human cells were detected in the site of SCI in animals injected with CD16- monocyte subpopulation, yet, the CD16+ monocytes were found in much less numbers at the site of injury. This may indicate that the CD16+ monocytes have much less migratory capability in comparison to the CD16- subpopulation.
In addition, the cellular distribution and profile of monocytes infiltrating the brain was assessed in animals subjected to the stroke. We first assessed whether monocytes home to sites of injury in the stroke-affected brain. To be able to trace the monocytes and identify their homing site, we passively transferred homologous monocytes isolated from the bone marrow of β-actin-GFP+ C57BlC mice into syngeneic wild-type mice that do not express GFP. This allows distinction between infiltrating monocytes and resident activated microglia. Two groups of animals were subjected to MCAO and on the next day injected through the tail vein with 4 million GFP+/CD115+ monocytes or vehicle, respectively. The purity of the CD115+ population was 94-98% as defined by flow cytometry. Immunocytochemistry revealed that grafted GFP+ monocytes were located exclusively in the ipsilateral hemisphere, the overall majority within the ischemically injured striatum (Fig. 1B, C and I). About 6000 GFP+ cells per brain were found (Fig. 1C). The injured striatum was filled with cells immunoreactive for the microglia marker, ionized calcium-binding adapter molecule 1 (Iba1) (Fig. 1D). Most likely, these cells were resident microglia because freshly recruited monocytes are not Iba1+ and, moreover, not a single GFP+ cell showed Iba1 immunoreactivity (Fig. 1E-G).

Figure 1. Transplanted and endogenous monocytes are recruited to injured brain tissue after stroke. B-G, Fluorescence microscopic images of mouse brain coronal sections showing the ischemic lesion in the striatum visualized by NeuN staining (B), distribution of grafted GFP+ monocytes within the lesion (C and F), expression of Iba1 (D and E) by cells within the injured striatum. (E)-(G) Confocal images showing GFP+ grafted monocytes in the lesioned striatum (F) not expressing Iba1 (E) with merged image in (G). H-J, Fluorescence microscopic images of mouse brain coronal sections showing extensive GFAP staining mostly outside the lesion (H), distribution of grafted GFP+ monocytes within the lesion (I), expression of IB4 (J) by cells within the injured striatum. Scale bar = 420 μm for B-D and H-J; Scale bar = 50 μm for E-G and K-M.

In order to detect distribution of endogenously (bone marrow) – derived monocytes in the parenchima of stroke-lesioned brain, we generated chimeric mice by subjecting wild-type mice to bone marrow transplantation from CX3CR1-GFP mice. In these chimeric animals only bone marrow-derived monocytes are GFP+, which allows identification of monocyte-derived macrophages (MDMs) both with immunocytochemistry and flow cytometry (Mildner et al., 2007). Immunocytochemistry of brain sections from chimeric mice showed that at 7 days after stroke the majority of GFP+ MDMs were distributed within the ischemically injured tissue with some of them being localized in close proximity to the lesion border (Fig. 2 A and B; from D 1.5.1 as above).

In conclusion, both intravenously injected and bone-marrow-derived monocyte after infiltration in stroke-damaged brain preferentially if not exclusively invade lesioned tissue and reside in activated form within few days after injection (transplantation) or brain insult (spontaneous).

D 6.1. Report on efficacy of monocyte therapy in male and female species, and aged animals

Ischemic stroke occurs more often in aged humans. It is therefore of major clinical interest to explore whether in the aged brain immune system retains the capacity to contribute to post-stroke recovery. Females experience poorer recovery after ischemic stroke compared to males, even after controlling for age and stroke severity. Therefore, it is important to validate and compare the effect immune system on post-stroke recovery in female and aged mice. To achieve this goal we carried out a pilot study with behavioral experiments in which we polled out aged animals (12-16 months old) and also some female mice to compare how depletion of bone marrow-derived monocytes from the blood circulation will affect the recovery after stroke.
In order to explore the potential role of MDMs for spontaneous functional recovery after stroke in male and female species and aged animals, we depleted circulating monocytes by means of MC-21 antibody during the first week after the insult, i. e., at a time when maximum monocyte infiltration takes place as described in D1.1.1 (Fig.1). We carried out behavioral tests to assess how depletion of monocytes during the first week after stroke would affect the long-term functional recovery.
Animals which were more aged as compared to normal young counterparts commonly used for the experiments were subjected to corridor (1 week before and 3, and 11 weeks after stroke) and staircase tests (1 week before and 1, 3, and 7 weeks after stroke). Sham-treated animals showed normal behavior in corridor and staircase tests. We observed impairments in the corridor test on the contralateral side at 3 weeks after stroke in both vehicle- and MC-21-injected animals (Fig. 1).

Fig. 1. Comparison between sham-treated and vehicle-injected (“sham”), stroke-subjected and vehicle-injected (“vehicle”), and stroke-subjected and MC-21-injected (“MC-21”) in performance in corridor test. Performance was calculated by dividing the number of contralateral retrievals with total number of retrievals from both sides.

Interestingly, already at 3 weeks animals injected with MC-21 had more severe impairment as compared to vehicle injected controls. At 11 weeks, we found spontaneous behavioral recovery, the test performance reaching control level, in vehicle-injected mice. In contrast, in MC-21-treated animals, the impairment was maintained at the same level as at earlier time points (Fig. 1).
In order to explore the contribution of sex to the observed effect of monocyte-derived macrophage effect on behavioural recovery, we subjected female mice to the monocyte depletion and compared to the female control, vehicle injected animals. In the staircase test, both vehicle- and MC-21-injected female mice showed similar impairment in number of retrieved pellets on the contralateral side at 1 week after stroke. At 7 weeks, the performance of the vehicle- injected mice did not differ from that of the sham-operated animals, whereas MC-21-treated mice still showed impairment. These data are closely replicating the behavioural impaired effect of monocyte depletion which was observed in male mice.

Fig. 2. Comparison between sham-treated and vehicle-injected (“sham”), stroke-subjected and vehicle-injected (“vehicle”), and stroke-subjected and MC-21-injected (“MC-21”) in performance in staircase tests. Performance was calculated as the number of eaten pellets on the impaired side.

These findings suggest that depletion of monocytes delay recovery in spontaneous recovery as observed in our previous study (Watanannit et al., 2016 J. Neuroscience) in aged animals similar to that in young mice. Moreover, the data obtained in female mice was very similar to that in male animals indicating that the sex of species does not play any role in the effect of infiltrating monocytes for post-stroke recovery. Therefore, we concluded that the immune cells act universally and special consideration to age or gender most like should not be taken into account in respect when considering the stimulation of monocyte infiltration in stroke therapy.

D 6.2.1. Protocols for production of human monocytes

We have developed method for the production of autologous monocytes from patient’s own blood. Although, this method work very well with healthy patients and ones with chronic neurodegenerative diseases, for the acute generation of the monocytes from the stroke patient depending on the condition of the patient it requires further development of feasibility.

This method includes the procedure for labeling cells with fluorochrome-labeled antibodies against the surface markers of monocytes, lymphocytes, granulocytes, red blood cells and thrombocytes. In addition, this method describes the procedure to determine the cell viability of the product cells as well as determine their cell cycle stage.

The method is based on the specificity of antibodies to their respective antigens on the surface of cells. CD14 is a human monocyte/macrophage marker. CD235a (i.e. Glycophorin A) is a specific human erythrocyte marker. CD42b is a specific human thrombocyte marker. CD45 is a specific human white blood cells marker which is not expressed on erythrocytes or thrombocytes. CD66b is a specific human granulocyte marker. CD3, CD4, CD8, CD19, CD25, CD56 and δγTCR are surface markers for specific subsets of lymphocytes. CD11b, CD11c, CD16, CD33, CD34, CD86, CD163, CCR2, CX3CR1 and HLA-DR are surface proteins which their combined expression or absence of expression patterns allows for identification of specific monocyte subsets. Fluorochrome-conjugated antibodies for all of these markers are commercially available. When excited by a flow cytometer laser system, the specific wavelength of fluorescence emitted by the different fluorochromes allow for detection of specific cell populations in the product. Propidium Iodide is a dye that binds DNA in proportion to the DNA amount present in the cell which allows for identification of the cell cycle stage of the cell. 7-AAD is a DNA dye that allows for exclusion of nonviable cells in flow cytometric analysis.

The developed manufacturing process which consist of several steps. During Step 1 (Collecting Blood Leukocytes) there is collection of 15x109 white blood cells via an apheresis procedure. During Step 2 (Monocyte Enrichment RBC and Platelet Elimination) there is elutriation of cells by passing the apheresis product obtained in Step 1 through the ELUTRA machine. Next is Step 3 (Labeling CD16+ cells with magnetic beads) which includes injection of anti-CD16 antibodies conjugated to magnetic beads into the apheresis product bag. During the following Step 4 (Depletion of CD16+ Cells), the bags are placed on Life’s CTS magnet and the cells are transferred into the clean bag. Next Step 5 (Preparation of Cells for Injection) includes washing and concentration procedure for the isolated product. The Step 6 (Quality Control Tests) is very important and includes tests for safety, purity and activity. The final Step 7 (Injection to the Patient) is preceded by packaging of the final product.

The system developed by TargetBraIn ensures very efficient and high quality and safety collection of circulating monocytes from human blood and will definitely qualify for clinical testing in forthcoming clinical trials based on the use of monocytes and modulators of the neuroinflammation for improved functional recovery in patients with in acute and chronic neurodenerative diseases.

D 6.2.2. Set the release criteria for the human product

A set of release criteria was developed which include purity, sterility and expression of specific surface markers measured by flow cytometry methods, as listed below:
• Survival – survival of the cells in the final product was set to be ≥98% as determined by 7-AAD staining, a DNA dye that incorporate into nonviable cells.
• Purity - purity of monocytes out of all live cells was set to be ≥80% as defined using CD14 immunostaining as the human monocyte/macrophage marker.
• Impurity – Characterizing the other cellular components in the final product using the following set of markers; CD235a (i.e. Glycophorin A) is a specific human erythrocyte marker, CD42b is a specific human thrombocyte marker, CD45 is a specific human white blood cells marker which is not expressed on erythrocytes or thrombocytes, CD66b is a specific human granulocyte marker. Using this set-up we calculate the percentage of each of these cells and set the criteria to ≤2% for erythrocytes and thrombocytes.
• Monocyte sub-population – The distribution of monocyte sub population as determined by expression of CD14 and CD16 was set to a frequency of ≤5% for CD14+CD16+ cells out of total CD14 + cells.
• Sterility – two immediate sterility tests were performed to validate the cell product sterility. LAL test of cell suspension set to ≤1.66 Endotoxin unit per mL. Gram stains of cell suspension set to show no evidence of bacteria.

In consistency with the deliverables D 6.2.1 and D 6.2.2, this set of release criteria was drawn based of the experimental runs that we performed using the final manufacturing procedure as described in our SOPs and in compliance with the regulatory requirements for freshly administered cellular products. The SOPs is available to the consortia members upon request.

D 6.3.1 Summary of the analytical characterization
Preclinical studies for the GEMST development. Summary of the analytical characterization.

GEMST was conceived in order to treat the excitotoxicity environmental conditions of the brain
tissue after an acute ischemic/reperfusion (I/R) process using a known model of stroke (transient middle cerebral artery occlusion (MCAO)). GEMST is a new drug candidate homologue of previous compounds developed by GEMAC for the treatment of other drug candidates like GEMSP (for Multiple Sclerosis treatment) and GEMALS (for Amyotrophic Lateral Sclerosis treatment), among others.
For preclinical studies, a major phase is the preparation of PLL compounds with the verification of their qualities by appropriate analytical methods (see Figure 1).
Figure 1. Summary of preparation process of PLL compounds with synthesis, purification until final assays and report.

In order to establish an analysis certificate of the Poly-L-lysine (PLL) compounds, the quality control and assurance were set up following the French Good Manufacturing Practices (GMP) (traceability, procedure, acceptance of batch, samples bank).
Using a Fourier Transformed InfraRed (FTIR) spectroscopy, GEMAC has characterized the PLL compounds and raw materials according to EMA and FDA requirements. By FTIR, the qualitative analysis allowed to check the linkage of small molecules to PLL and to identify the specific peaks of these molecules.
Because the identification of each batch of PLL compound was done by comparing measured spectra with reference spectra, Gemac has created a spectral library in addition to those pre-existing molecules saved in other libraries. The qualitative evaluations were based on the identification of a known substance. A particular functional group shows IR absorption bands with a characteristic spectral range: this is called group vibration, such as stretching, contracting and bending. The position and intensity of the absorption bands are both extremely specific in the case of a pure substance. This enables the IR spectrum, similar to fingerprint region, to be used as a highly characteristic feature for the identification of molecules.
The quantification process has been finished for raw material (RM) linked to PLL by different linking agents (Figure 2) and purified by Tangential Flow Filtration (TFF) (Tables 1 & 2). Finally, PLL compounds were lyophilized for storage and, using the FTIR-ATR, each batch was analysed (see Figure 1 and Table 1).

Figure 2. Small molecules linked to PLL

At Gemac, the concentration of both free small molecules and PLL compounds were determined by UV/Visible (Perkin-Elmer). In addition, PLL compounds must be tested by other methods such as Gas-Liquid Chromatography and Mass Spectrometry was done in collaboration with CESAMO’s analytical platform at the University of Bordeaux. In order to control impurities during the synthesis of PLL compounds, residual solvents were controlled by an external subcontractor (Eurofins).

Table 1. Since the acquisition of FTIR in 2012, the number of analysis carried out for several PLL compounds and raw materials (RM) are shown. These assays include TFF tests and tests with IR cell accessory.
PLL Compound Assay numbers RM numbers Stability numbers
Total by category 546 68 120
TOTAL 734

Since 2012 to date, Gemac has performed a total IR number of : 3,386
- 2012-2013: 937
- 2013-2014: 1,008
- 2014-2015: 729
- 2015-2016: 712
-
Stability studies
Gemac has subdivided each batch in multiple samples in order to conduct analyses over the time. Thus, thanks to the stored samples, the assessment of stability study of PLL compounds (in the lyophilized form) has been conducted firstly for a period of 6 months, over 2 years. In conclusion, all PLL compounds were stable for at least 2 years (see Table 2).

Table 2. Summarized data regarding the changes in the synthesis process, purification and stability of PLL compounds.
PLL Compound Synthesis Purification 2 years
Stability by FTIR
PLL-Compounds
Scale up OK

TFF OK

Stables

The process applied by changing the linker and the methodological procedure, allowed to optimize the synthesis processes.

Secondary structure of PLL compounds
Since 2011, the Institute of Molecular Science (ISM, Bordeaux, France) of the University of Sciences at Bordeaux has analyzed the PLL compounds by spectroscopy IR coupled to a vibrational circular dichroism (VCD). This spectrum revealed the secondary structure of PLL compounds, allowing to compare it with derivatives spectra. All PLL compounds were VCD spectra, slightly modified in comparison with PLL (random coil). Currently, this partnership is continuing.
The IR spectral data of high polymers are usually interpreted as vibrations in a structural repeated unit. The most sensitive spectral region of the polypeptide secondary structural components is the amide I band (1700-1600 cm-1), which is mainly due to the C=O stretch vibrations of the peptide linkage. The frequencies of the amide I band components were found to be closely correlated to each secondary structural element of the polypeptides. PLL, and some PLL-compounds show a random coil as secondary structure. Finally, it should be noted that the linkage of small molecules to PLL did not modify the PLL secondary structure.
The collaboration with the ISM are still ongoing. The results found will be submitted to a scientific journal in the coming months. The paper will be entitled “New therapeutic polypeptide compounds investigated by FTIR and VCD spectroscopy.”

WP7: Management of the consortium

The overall objective of this WP was to manage consortium and ensure that it functions properly, all deliverables are delivered, annual scientific and financial reports and audit certificates are produced. The management of the project always made sure that during the reporting period all partners and team members receive all necessary information and support, and that annual scientific and financial reports were produced on time and with high quality.

The overall responsibility for the research program was on the Coordinator and Steering Group. They were assisted in the management of the project by Project Administrator Katarina Turesson. In addition, Research Service Office at Lund University was rendering help with preparation of contract, annual and final reports and any issues related to legal questions and interaction with EC. The daily decisions for running the project and interaction with EC, as well as all work with all formal documents and reports. Co-coordinator was in charge of overseeing the progress of the Tasks, arrangement of scientific data discussions and brainstorming (whenever needed or requested by Partners), suggestions for modifications of the projects of the proposal whenever it is required based on the new published data and unpredicted data obtained by the consortium.
Regular meetings of all consortia members will take place every 12 months. The meeting were organized in Jerusalem (2011), Bergamo (2012), Tbilisi (2013), Lund (2014), Gothenburg (2015), Milan (2016). These meetings updated the scientists involved in the project by presenting the data from ongoing and finished scientific tasks and based on active discussions, were making decisions. In the light of experimental data, it was always discussed and decided how to continue with different projects and whether or not it will be necessary to modify planned work.
In addition to annual meeting, the steering group and individual PIs arranged more than 10 ad-hoc meetings or meetings during different conferences and congresses. In addition, Consortium has arranged several Steering group meetings in Copenhagen, (November 24, 2011, attended by P1- Kokaia, P2 – Schwartz, P3-Martino), Garmisch-Partenkirchen (December 15, 2011, attended by P1- Kokaia, P3-Martino, P4-Hoehn during the European Stroke Science Workshop), Zurich (February 17, 2012: attended by P1- Kokaia, P2 Schwartz, P3-Martino, P4-Hoehn; P5-Betcher), Boston (November 5, 2012, during International Society of Neuroimmunology (ISNI) Congress, attended by P1- Kokaia, P2 – Schwartz, P3-Martino, P5- Betcher; Prof.Tomas Olsson – SAB member). Several individual meetings of PIs were also arranged (e.g., visit of P4- Hoehn and P6-Eder visit to Lund on June 11 and 5, 2012 respectively, meeting of P1-Kokaia with P2-Schwartz on July 2012 at Rome, Visit of a Ph.D. student of Schwartz - P2 at the laboratory of P5-Betcher for training on February/March2012 ). Meeting was arranged in in Prague during a symposium at Featured Regional Meeting of Federation of European Neuroscience Societies, September 11-14, 2013 (attended by P1- Kokaia, P4 – Hoehn, P7-Pekny) and in Milan during the symposium “The Neurobiology of the Immune System” within the 15th International Congress of Immunology August, 22-27, 2013 (P1- Kokaia, P2 – Schwartz, P3-Martino). Consortium has arranged several individual PI planning meetings Aspenäs, Sweden, April 2014 (Schwartz, Pekny), Washington DC, USA, Society for Neuroscience meeting, Nov 2014 (Schwartz, Pekny and the researchers from the two laboratories), Gothenburg, Sweden, Dec 2014 (Kokaia, Pekny), during Science Festival in Bergamo (Kokaia, Martino, Lindvall), October 2-5, 2014, PI meeting in Lund (Kokaia, Hoehn) November, 2014. In addition, PI Milos Pekny organized the COST Action NanoNet Training school and International Research Conference Astrocyte intermediate filaments (nanofilaments) and astrocyte function in health and disease, Aspenäs, Sweden, 2014 (PI Michal Schwartz as one of the speakers). He also co-organized of a symposium Brain plasticity and regeneration – from animal models to clinical rehabilitation, 8th World Congress for Neuro Rehabilitation, Istanbul, Turkey, 2014. Consortium has also arranged several individual PI planning meetings: Tbilisi, Georgia, June 2015 during Immunology conference (Kokaia, Martino, Hoehn); Sydney, Australia, Aug 2015, a planning meeting (Pekny and Becher), in conjunction with the Society for Neurochemistry Meeting Understanding the Function of Glia in the Healthy and Diseased CNS; Quebec City, Canada, Sept 2015, a planning meeting (Pekny and Schwartz), in conjunction with 11th Andre-Delambre ALS Symposium, Laval University; Chicago, USA, a planning meeting (Pekny and Schwartz), in conjunction with the Society for Neuroscience Meeting. Partners 1, 2 and 3 (Kokaia, Schwartz and Martino) had a brainstorm and planning meeting at Copenhagen Airport (Hotel Hilton) on March 9, 2015. In addition, Consortium Steering Group members had several video and telephone conferences based on the needs of the running the consortium and discussing data and plans. Partner 4 (Hohn) organized Kick-off meeting of the Molecular Neuro-Imaging Study Group. 10th European Molecular Imaging Meeting (EMIM), Tübingen, Germany, March 18-20, 2015 in which Partner 1 (Kokaia) gave a talk and participated in discussions. For collaboration with UZH Partner 5 (B. Becher and M. Greter) Partner 3 traveled Zurich (Switzerland) on the 10-11 September 2015 and on the 24-27 November 2015 to discuss data and analyze the inflammatory infiltrate in ischemic brain at various time-points after stroke with particular regards to myeloid cells and specifically to microglia.

The management of the consortium, in collaboration with LU lawyers has created the consortium agreement which was signed by all Partners.
As suggested by the review panel of the TargetBraIn proposal (D 7.1.1), the Consortium management has created the International Scientific Advisory Board (SAB) consisting of world leaders in respective field of activity. The SAB consist of following members:
Tomas Olsson - Professor of Neurology, Department of Clinical Neurosciences,
Neuroimmunology Unit, Center for Molecular Medicine, Karolinska Hospital, Stockolm, Sweden - neuro and clinical immunology.
Costantino Iadecola - Cotzias Distinguished Professor of Neurology and Neuroscience Chief of Division of Neurobiology, Weill Cornell Medical College, New York, USA – stroke and neuropathology and immunology.
Oliver Brüstle - Professor at Institute of Reconstructive Neurobiology, Director of LIFE & BRAIN Center, University of Bonn, Germany – stem cells, neurodegenerative diseases and neurotransplantation.
The members of SAB have been present at several annual meetings. Prof. Olson participate in Bergamo meeting (2012), Profs. Olsson, Iadecola and Brustle in Tbilisi meeting (2013), Profs. Olsson and Brustle in Gothenburg (2014), and Prof. Iadecola in Milano final meeting (2016). The have produced 3 reports which were presented during the annual reports to the commission and served as guidelines for the consortium (the reports are attached).
Consortium appointed Professor Olle Lindvall as Chief Clinical Advisor world-class specialist in cell therapy who was in charge of providing consortium with clinical advices together with other active clinicians and members of SAB (Prof. Olsson, prominent clinician with neuroimmunology background and Prof. Iadecola, clinical neurologist specialized in stroke). Prof. Lindvall actively participated in all consortium meetings and discussions.

Project Results:
Stroke is the second most common cause of death worldwide (Lopez et al., 2006). It also represents the greatest cause of disability with about 72 million disability-adjusted life years. The extension of the average human life-span has resulted in a constant increase in the incidence of stroke among the expanding elderly population, creating a serious hazard to the quality of life and increased health care costs. According to the WHO estimates the number of stroke events in EU countries is likely to increase from 1.1 million per year in 2000 to more than 1.5 million per year in 2025 solely because of the demographic changes (Truelsen et al., 2006). From this perspective, all efforts at bringing the latest advances of science and technology a step closer to the development of therapies against stroke should be vigorously promoted. Given the relatively long-term prospects of these types of efforts, both basic and applied initiatives are urgently needed. The overriding goal of our project was to promote development of a novel therapeutic strategy leading to improved functional recovery in stroke patients. It directly addresses the Community social objective of improving the quality of life and health in stroke patients by shortening the recovery phase and minimizing the motor and possible cognitive impairments.
We have explored the possibility to promote functional recovery after stroke targeting immune mechanisms and utilizing the brain’s own capacity for self-repair. Our data show feasibility of improving post-stroke recovery based on manipulation of immune system and altering neuroinflammation. This approach avoids the major ethical problems related to use of human ES or fetal stem cells for therapy. After our earlier discovery (Arvidsson et al., 2002) that new neurons are formed from the adult brain’s own NSCs after stroke, several biotech companies in the US, Canada and Europe have been established solely for developing therapeutic strategies for neurodegenerative diseases based on adult neurogenesis. We hope that data from our work will also promote the development of new approaches to the stroke treatment.
There is a considerable activity and investment in the US in research for development of stem cell therapy for stroke and other neurodegenerative diseases. Several clinical trials are ongoing to evaluate the safety and effectiveness of stem cell transplantation in patients. The Chief Clinical Advisor and active member of Partner 1 research team Prof. Lindvall is one of the pioneers of cell therapy and the establishment of proof-of-principle that cell replacement strategies can work in the human brain with neurodegeneration is closely related to his clinical work. Although such proof-of-principle for cell therapy in stroke patients is still lacking, our studies showed an enormous potential for NSCs. It is now becoming clear, through the pioneering studies of the Partners of the consortium, that immune cells can benefit repair in neurodegenerative disorders. Partners of the consortium have also demonstrated that NSCs can act beyond cell replacement and participate in local and systemic immunomodulation. Based on our data it is possible to develop novel therapeutic approaches for this condition as well as for other acute and chronic neurodegenerative diseases. Since Partners of this proposal are very prominent in their respective fields, our research had a major impact on advances in this area. As a proof practical value of the knowledge generated by the consortium, we can name 2 examples:
1. The coordinator of the consortium has been invited for the planning of Phase III clinical trial by ReNeuron which is the only clinical trial which is using neural stem cells for the stroke patients. The knowledge and experience generated by the consortium in respect to the involvement of neuroinflammation in cell therapy for stroke was shared by extensively discussed during the meeting on March 4, 2016 at Amsterdam Airport Schiphol with Clinical Trial Management.
2. The knowledge generated by consortium was shared by coordinator with the special meeting by company EVER Neuro Pharma GmbH, which is producing drug cerebrolysin used as a supplement for the post stroke recovery. The meeting was arranged on November 10, 2014 in Salzburg (Austria).
The details of the meetings can not be shared because of the Confidentiality Agreement but these facts indicate potential impact of the research conducted by the consortium on development of new therapies for stroke and other neurodegenerative diseases.
We propose an integrated and comprehensive, goal-directed study at state-of-the-art level. It will contribute to increase competitiveness of Europe in fields of high strategic importance such as basic and clinical neuroscience, stem cell research, health care, economic welfare and quality of life.
An additional societal benefit addressed by our proposal concerns opportunities for education and training. Several visits of the students between the Partner labs was arranged to learn different technics and methods and standardize research and exchange the experience. For example,
- Marina Dobrivojevic and Anuka Minassian (from Partner 4 group; Cologne, Germany) visited Prof. Zaal Kokaia (Partner 1), Lund, Sweden with purpose to get training on surgical technique of distal middle cerebral artery occlusion as cortical stroke model;
- Ruimin Ge from Partner 1 (Lund, Sweden) visited Partne2 (Revovat, Irael) to learn choroid plexus isolation and intracystrnal injection of monocytes;
- Ruimin Ge and Cecilia Laterza from Partner 1 visited Partner 7 (Gothenburg, Sweden) to perform stroke and intracebral transplantation experiments and train local students in these procedures.
This type of activities ensured the spreading and dissemination of valuable tools and know-how across different countries of Europe.
Participation of consortium members in more than 250 scientific meeting and arrangement of several targeting meetings (see below) as well as publications by the consortium members scientific articles in top international journals (see below) promoted strengthening of European competitiveness in neuroscience, neuroimmunology and stem cell research, which are fields of high strategic importance. Our problem-solving approach enlarged the knowledge of the general principles of on one hand neuroinflammation and on the other hand, stem cell biology and their application for cell therapy.
The work in this proposal was carried out by the coordinated effort of 9 different centres in 7 European countries. This collaboration brought together academic and industrial European laboratories to realize a common goal. We believe our objectives were best tackled by a combination of contributions from different fields, disciplines and techniques, and which was supported by composition of our partnership. Each centre concentrated on specific tasks in which it has acquired a distinctive, well-established competence. The expertise which we collected jointly could not have been obtained within anyone of the countries participating in this network. The collaboration of these centres resulted in the spread of knowledge, technologies and expertise for the benefit of all the participants.

As a part of dissemination activity, the consortium created interactive webpage www.targetbrain.eu. The publicly accessible part of the website provides information about the TargetBraIn objectives and goals, and address general issues related to the brain immune system and inflammation. The website acknowledge European Commission's FP7 support and display the EU flag and FP7 logo. The page also has internal part, protected by the password, and containing all information regarding the consortium including the contracts and list of deliverables and milestones, consortium agreement, the personal and professional data of all involved scientific and technical staff, experimental protocols, manuals, list of available reagents etc., which facilitated the interaction between the partner teams and disseminate knowledge within the consortium.

The major form to disseminate results to the scientific community is the publications in peer-reviewed scientific journals. We aimed to publish data generated by the consortium in high-impact journals easily accessible for a wide range of scientists and physicians. Overall, consortium published 165 scientific peer-reviewed articles. Among them, publication appeared in such top journals as Science (4), Nature Group (Total: 13. Nature: - Immunology (1), -Medicine (2), - Neuroscience (3), -Communications (4), -Review Neuroscience (1), -Review Immunology (1), -Review Neurology (1)),
Journal of Neuroscience (6), Journal of Clinical Investigation (2), Immunity (4), Cell Stem Cell (1), EMBO J (3), Trends Immunology (3), Brain (5) J Experimental Medicine (2), Cell Reports (1), Lancet Neurology (1) and Stem Cells (3). It should be emphasized that at the very beginning of the consortium activity the Editorial Board of Nature Neuroscience approached some members of the consortium with the proposal to write an overview article for the special issue about the cross-talk between stem cells and immune cells during the neurodegenerative diseases. We have published this article and also produced cover page for the issue. Importantly, the authors of this paper are only members of the TargetBraIn consortium and EU support and affiliation of the authors to TargetBraIn is the only acknowledgement of the whole paper (see: Kokaia et al., 2012, Nature Neuroscience, vol.15, No.8, p. 10781087). (102 Citations).

In addition, all members of the consortium presented data from our consortium research at almost 300 European and other international conferences. We have primarily target conferences with a focus on neuroimmunology, stem cell research, neurodegenerative diseases, and regenerative medicine
The presentations at the conferences and congresses always acknowledged the affiliation to the TargetBraIn consortium and financial support from EC. We would like to mention several of them which were organized and/or dominated by consortium members.

Several high-profile publications had press releases and obtained very high international media coverage. For example, Science publication by Baruch (Partner 2) has received the enormous public attention and the discussion about this paper by Dr. Meydani and Dr. Steindler has been posted at Youtube (https://www.youtube.com/watch?v=xaAPEGfz25Y). This paper was Editor’s Choice, Science Translational Medicine. “Old-Age Interfer(on)ing” by Anne Schaefer (Vol. 6, Issue 252) and described in “Perspective” at Science. “Good barriers make good neighbors” by Richard M. Ransohoff (Vol. 346, Issue 6205). It also received “Research highlight” at Nature Immunology “Interferons in the brain” by Laurie A Dempsey (Vol .15, Issue 909). Another Science paper published by Partner 1 was also covered widely by media and appeared also on Fox News (http://www.foxnews.com/health/2014/10/10/scientists-find-cells-self-repair-brain-after-stroke/). The press release was covered by more than 30 internet news sites. “The Telegraph” dedicated article (http://www.telegraph.co.uk/science/2016/04/15/stroke-treatment-to-fight-inflammation-could-harm-recovery/) to the collaborative publication of Partners 1 and 2 dedicated to the role of immune system in recovery after stroke published in Journal of Neuroscience, 2016.

- The meeting, entitled “Neurodegenerative diseases, stem cells and inflammation – new prospects for therapy” was arranged by the consortium in conjunction with kick-off meeting in Jerusalem (December 1, 2011) and was arranged at Weizmann Institute (Rehovot, Israel). The program of this symposium was fully composed of lectures by consortium members and was attended by great number of researchers from different fields.
- The meeting entitled “Cellular plasticity after stroke - important for recovery?” was organized as a symposium at European Stroke Science Workshop by European Stroke Organization on December 15-17, 2011, at Garmisch-Partenkirchen, Germany. Four out of 5 presentations were presented by Consortium members and the symposium was chaired also by consortium representative Prof. Olle Lindvall.
- The meeting, which was arranged and chaired by TargetBraIn coordinator Prof. Kokaia, was symposium “Inflammation in neurodegenerative disorders” within the International Society of Neuroimmunology (ISNI) Congress in Boston (November 5, 2012). These 2 out of 3 lectures were presented by TargetBraIn PIs, Kokaia and Schwartz.
- The meeting, entitled “Inflammation, neurodegeneration and stem cells” was arranged by the consortium in conjunction with annual meeting in Tbilisi (November 23, 2013) and was arranged at Georgian National Museum (Tbilisi, Georgia). The program of this symposium was fully composed of lectures by consortium and SAB members and was attended by great number of researchers and physicians from different fields.
- The meeting was entitled “Neural Stem Cells and Regeneration Following Stroke” and was organized as a symposium at Featured Regional Meeting of Federation of European Neuroscience Societies, September 11-14, 2013, Prague, Czech Republic. Three out of 4 presentations were presented by Consortium members (Kokaia, Hoehn and Pekny; program enclosed) and the symposium was chaired also by consortium coordinator Prof. Zaal Kokaia.
- The meeting, which was arranged and chaired by TargetBraIn Partner Prof. Martino, was symposium “The Neurobiology of the Immune System” within the 15th International Congress of Immunology August, 22-27, 2013 (Milan, Italy). These 4 out of 6 lectures were presented by TargetBraIn Partners, (Kokaia, Schwartz and Martino) and TargetBraIn SAB member Prof. Tomas Olsson.
- Partner 7 (Milos Pekny) co-organized of the 9th European Conference on Intermediate filaments (nanofilaments) in health and disease, Stockholm, 2015 and
- Partner 1 (Zaal Kokaia) Organized symposium “New Frontiers in Neuroimmunology” (Chair of symposium). Symposium: Tbilisi, Georgia, June 24-35, 2015 with lecture participation of Partners 3 and 4 (Gianvito Martino and Mathias Hoehn).
- Partner 4 (Hohn) organized Kick-off meeting of the Molecular Neuro-Imaging Study Group. 10th European Molecular Imaging Meeting (EMIM), Tübingen, Germany, March 18-20, 2015 in which Partner 1 (Kokaia) gave a talk and participated in discussions.

Two presentations by young scientists at the conferences received prizes. Hipponion Stroke National 3rd Prize was received by to Marco Bacigaluppi for the presentation entitled: Neural stem cell transplantation promotes post-ischemic recovery through the glutamate transporter GLT1. Tamar Memanishvili from the Kokaia lab, received first prize for the best poster at the Drug Discovery & Therapy World Congress 2014, June 16-19 Boston, MA, USA. Title of the poster: Biodegradable Amino Acid-Based Polymeric Microparticles for Improved Functional Recovery in Stem Cell Therapy After Stroke.

Exploitation of foreground knowledge and intellectual property (IP) by individual partners not directly involved in their generation was be restricted to internal use. The objectives and the topics of this proposal were of great importance for the ongoing current and future research of most of the partners. The members of the consortium are committed to the future scientific exploitation of the results obtained from the consortium for the further development of a solid and competitive European knowledge-base in the field of neuroinflammation, stroke, stem cell biology and cell therapy. These plans include: 1. Further studies on mechanisms of brain inflammation in stroke and interaction of NSCs with blood- and brain-derived immune cells; 2. Development of new methods for modulation of neuroinflammation for affecting the functional recovery after stroke and fate of endogenous and transplanted NSCs; 3. Further development of pre-clinical protocols for improvement of post-stroke recovery by modulation of neuroinflammation. 4. Development of molecular imaging strategies and tools for a detailed monitoring of the therapeutic effects of these neuroinflammation-modulated cell strategies.

During he course of the project several patents were developed and most of them are currently either under investigation of innovation bodies or pending approval. We foresee that in coming years the data obtained form the project will also serve as a bases for several more patents. The partners have no previous agreements, which may impose limitations on a subsequent exploitation of the results. The handling of the IP properties were and will be carried out as described in Consortium Agreement.

Commitment to the field of research, solid and original data obtained during the course of consortium activity and the high-quality standard and track record of partners in the consortium constitute strong reassurances of the credibility and capability of the consortium for effective future scientific exploitation of the results obtained from the work.

Potential Impact:
www.targetbrain.eu

List of Websites:
The logo of the consortium and the Brochure are provided.

Related information

Reported by

LUNDS UNIVERSITET
Sweden

Subjects

Life Sciences
Follow us on: RSS Facebook Twitter YouTube Managed by the EU Publications Office Top