Therapeutic Time Window And Effect Of Intracarotid Neural Stem Cells Transptantation For Neurological Sequelae After Experimental Intracerebral Hemorrhage

Background and Purpose:Intracerebral hemorrhage (ICH) is a major cause for morbidity. The damage induced by ICH involves a direct mechanical destruction of brain tissue by the hematoma and the following process of inflammation and brain edema. Currently, no medical therapy is available for ICH patients, with options limited to supportive care and invasive neurosurgical treatment. It is eager to search measures to treat those survivors with severe disabilities, so as to enhance their self-confidence to live, improves the quality of life, and reduce burden of both the family and the society. Due to the lack of ICH animal model simulating the spontaneous process of ICH accurately, those researches on the pathophysiological mechanism, restoration of neurological deficits and treatment strategies have been limited. In 1996, Del Bigio and colleague improved the ICH model established by Rosenberg. Intracerebral hemorrhage was successively induced in rats by stereotaxic, intrastriatal administration of bacterial collagenase and heparin, with fast speed of the formation of hematoma and similarity of hematoma volume. The method has been widely used to study brain edema, inflammation and ischemic penumbra following ICH. Neural stem cells are characterized by its self-review and pluripotentiality; in addition, neural stem cells has tropism for pathological changes in the CNS. Therefore, there is great interest in the possibility of repairing the nervous system by transplanting exogenously prepared NSCs that can replace those lost through damage or interact with cells in the host brain.However, concerns about where and when to transplant NSCs are still controversial. Most investigators have undertaken direct injections into the injured parenchyma at single or multiple sites. However, local intracerebral injection results in local brain damage, making its extrapolation to human beings unacceptable. Another problem is that NSCs directly implanted into the brains form cell clusters in the injected regions with a limited spread into the adjacent host brain tissues. In addition, exposing the cells to the hostile inflammatory environment of the lesion will lead to reduced survival of the grafted cells. Intraventricular transplantation is reported to permit a relatively widespread delivery to the brain, however, this technique, is also too invasive for clinical applicability, which makes its translation challenging. Moreover, injection into the lateral ventricle has to compromise the blood-brain and CSF-brain barriers. The percutaneous intracarotid injection, a minimally invasive procedure that can be easily performed at the bedside, has been used to obtain stronger efficacy against localized lesions in the brain. Therefore, we we tried to inject NSCs into the carotid artery of rat ICH models induced by injection of bacterial collagenase and heparin, so as to identify a more effective, less invasive strategies of NSCs delivery and ICH treatment. During the acute phase, the massive cell death, hemorrhage and the ongoing inflammatory response have neurotoxic effects, which will prevent surviving of the transplanted cells. Furthermore, severe inflammation leads to the production of an abundance of cytokines around the injured site, inducing those involved with astrocyte-differentiation such as IL-1, IL-6, TNF, and CNTF. Further, during the chronic phase of inflammation, the formation of cysts and development of glial scarring might gradually prevent NSCs from accessing the injured tissues. Therefore, we undertook the study, in which NSCs were transplanted respectively into rats at 2, 7, 14, 21, or 28 days after injury to determine optimal time windows of intraarterial NSCs transplantation. Methods:1. Brain tissues were obtained from the forebrain of embryonic day 14 Wistar rat embryos. After dissociated by adding trypsin (0.05% in 0.53 mM ethylene-diamine tetraacetic acid [EDTA]) and titurated by Pasteur pipette, cell suspensions were harvested into a 15-cm³ centrifuge tube. The cell suspension was centrifuged for 8 min at 800 rpm twice, and most of the supernatant was removed. After adding complete NSCs medium, a small sample was taken for counting, and then the cells were inoculated into preincubated NNSC medium at 1.0×10⁵cells/mL in 5 mL of medium. The T-flasks were placed in an incubator maintained at 37℃and 5% CO₂ in air.Enzymatic and mechanical methods were employed to break up neurospheres, respectively. Cell density was determined using a hemocytometer. Viability after dissociation was determined using the standard trypan blue exclusion test, so as to compare damage degree of NSCs dissociated by two methods.Growth curves of NSCs dissociated by enzymatic and mechanical methods were drawn, and doubling time was calculated, so as to study effects of different dissociating methods on cells proliferation.NSCs were cultured and passaged with different dissociating methods long term in vitro, and overall multiplication ratio was calculated, in order to explore the ability of neural stem cells to proliferate under conditions of long term dissociation with two methods.To examine the effect of long-term culture on NSCs neurogenesis, neurospheres of 4, 8, 12, 20 and 30 passages were plated on a laminin/poly-L-lysine-coated surface, and cultured in the absence of growth factors for 14 days. 2. Fifty Wistar rats were randomly divided into experimental group and control group. All rats were anesthetized with 10% Chloral Hydrate (0.3ml/100mg IP) and placed in a stereotactic frame. A microinfusion pump was used to infuse 0.7 uL saline containing 0.14U collagenase and 1.4U heparin into the caudate nucleus (3 mm lateral to midline, 0.2 mm posterior to bregma, depth 6 mm below the surface of the skull) of rats in the experimental group over 5 minutes, while Twenty-five sham-operated control rats received a similar 0.7μL infusion of saline without collagenase or heparin.Behavior was evaluated by an observer blinded to the identity of the rats beginning at 2 hours after collagenase/heparin injection and repeated daily.On 1, 3, 5, and 7 day after operation, respectively, five animals from each group were sacrificed to observe hematoma and edema in the brainsAt the end of the experiment, the last five rats in each group were killed for histological examination. Sections (6μm) were cut and stained with hematoxylin and eosin to observe pathological changes of the ICH brain.3. NSCs, expanded over 4-6 passages, were cultured with complete NSC medium supplemented with 5μmol/L 5-bromo-2-deoxyuridine (Brdu) three days before transplantation. The collected spheres were dissociated into single cells and resuspended at a density 1×10⁴ cells/μl. ICH model was induced by stereotaxic, intrastriatal administration of collagenase/heparin in forty-eight male Wistar rats, and forty animals were randomly divided into five groups to receive cell transplantation through intracarotid injection (4×10⁶ NSCs per animal) at 2, 7, 14, 21, or 27 day after injury respectively, while eight animals subjected to ICH only served as controls.Each rat was subjected to a series of behabioral tests to evaluate neurological function one day after ICH and every week after transplantation.Two months after collagenase/heparin injection, all rats were sacrificed, and brain sections (6μm) were cut. Single or double immunofluorescent staining was used to identify cells derived from grafted cells and determine the potential to differentiate along multiple lineages.To determine graft survival semiquantitatively, an average of 10 histology slides (200μm apart) of brain per experimental animal were evaluated. Two independent investigators counted the BrdU-positive cells and cells double labeled with MAP-2, GFAP for 3 sequential sections (200μm apart) per animal.The area of both hemispheres was measured on eight serial coronal sections per brain (200μm apart) stained with hematoxylin and eosin, and the area of the lesion size was averaged over these eight levels. Results:1. Primary and serial cultured neurospheres were Nestin positive, and were able to differentiate into MAP-2 positive neurons, GFAP positive astrocytes, and MBP positive oligodendrocytes, demonstrating self-reviewing and pluripotentiality of the cultured stem cells.In contrast with enzymatic dissociation, the method of mechanical dissociation, breaking up great neurospheres into small spheres with gradually reduced pipette, had advantages of raising cell viability (83.2±4.9% vs 73.6±4.3%), and shortening cell doubling time, which would increase the ability of cultured stem cells to proliferate.After NSCs were cultured long term in vitro, the number of neurons generated decreased with passaging, whereas the number of astrocytes increased.2. Rats were significantly impaired during the first day after collagenase injection, while sham-operated rats only showed slight behavioral deficit during the first day after sham injection. Significant differences were observed between two groups (p<0.0001).One day after operation, a round or oval hematoma was observed in the right caudate nucleus, with a narrow, irregular area of edema surrounded. Brain edema was significantly evident 3 days after ICH. In control group, brain edema was only observed in a limited area aroud the injection site during the first day. Water content of brain tissue in the experimental group was significantly higher than that in control group (p<0.001).Histological assessment showed a single roughly spherical hematoma, together with fragmented brain tissue, was present in the right injected caudate nucleus. Brain tissues, around the periphery of the hematoma, showed significant pathological changes, in which there were many neutrophils, microglias and degenerating cells. No obvious changes were observed in the control group.3. At 9 weeks of the experiment, ICH-induced neurological deficits were significantly ameliorated in animals receiving NSCs transplantation compared with that of the untreated controls (ANOVA p<0.05). Post hoc analysis revealed that animals receiving NSCs at 7 days after ICH had significantly lower scores at both 3 weeks posttransplantation and 9 weeks after ICH than animals treated at other time (p<0.05). Similarly, at the same time points, animals receiving NSCs at 2 or 14 days performed better than animals treated at later time, whereas therapeutic effect of cells transplanted at 14 days postinjury was superior to that of cells transplanted at 2 days postinjury (p<0.05). The therapeutic effects of NSCs given at 21 or 28 days postinjury were non-significant, and no significant difference was observed between them (p>0.05).Transplanted BrdU positive NSCs entered and survived in the brain. Instead of randomly dissipating into the whole brain tissue, the vast majority of the labeled cells, which seemed to be migrating selectively toward the lesions in line, were found distributed evenly in the perihematoma areas; also, a small number of cells were observed scattered in the contralateral hemisphere (p<0.05).Greatest surviving cell numbers were seen in animals treated at day 7 (p<0.001). In contrast with animals treated at 21 or 28 days after injury, there were also many more cells surviving in animals treated at 2 and 14 days after injury, whereas animals treated at 14 days still had significantly more surviving cells than animals treated at day 2.Some BrdU-positive cells were reactive for the astrocytic marker GFAP and for the neuronal markers MAP-2, but no double labeling for the oligodendrocytic marker MBP was observed.It seemed that the differentiation of the transplanted cells depended upon time of transplantation. In contrast to other groups, grafted cells of animals receiving NSCs injection at 2 days postinjury differentiated to a greater extent into astrocytes (84.5±7.6%), whereas to a less extent into neurons (8.6±1.3%). In 21 and 28-day group, a larger percent of neurons (35.4±3.1% and 37.2±4.1%, respectively) as well as a smaller percent of astrocytes (52.1±6.5% and 50.3±5.7%, respectively) derived from grafted cells were observed. However, when considering about absolute numbers of BrdU positive neurons, animals treated at 7 and 14 days postinjury gained the largest number of neurons, for that a great number of grafted cells survived (p<0.001).No significant reduction of volume of hemorrhagic damage was detected in rats receiving NSCs transplantation, compared with control animals subjected to ICH alone. Conclusions:1. In contrast with enzymatic dissociation, the method of mechanical dissociation, to break up great neurospheres into small spheres with gradually reduced pipette, helps minimizing damage to NSCs; maintain the integrity of cell-cell junction and receptors of NSCs, which results in enhancing proliferation of NSCs.After expanded long term in vitro, NSCs undgo decreased neurogenesis, due to the microenviroment and inherent factors,2. ICH model of Wistar rat induced by stereotaxic, intrastriatal administration of bacterial collagenase/heparin had significant and long-term behavioral disturbance, with little difference between individuals, and the hematoma size is similar, which is advantageous to evaluate the therapeutic effect for ICH.3. The percutaneous carotid artery injection is an efficient, minimally invasive delivery system of NSCs toward the site of ICH, due to the extensive damage in the blood-brain barrier and overexpression of cellular factors in the brain surrounding the hematoma, extensive tropism for pathology of NSCs, and diffusion gradients of intracarotid injection.The favorable timing for implanting the grafts would be approximately 7-14 days after the injury, during which new microenvironments might have been suitable for the grafted cells to survive and differentiate into more neurons and astrocytes.In vitro expanded NSCs administered through the carotid artery promoted robust functional improvement in rats with hemorrhagic brain injury, especially given at 7-14 day after ICH.