In-vivo Astrocyte-to-neuron Reprogramming For Treating Alzheimer's Disease Mouse Model

Part ? Characteristics of NeuroD1-mediated in-vivo reprogramming in different systemsBackground: By overexpressing specific transcription factors,in-vivo reprogramming technique makes use of endogenous astrocytes as source material and convert them into functional neurons in situ.These newborn neurons make up for the lost ones,repair neural circuits and finally improve neural function.With the development of this technique,it will show great application potential in the treatment for nerve injury and neurodegenerative diseases in the future.For expressing specific transcription factors in astrocytes to regenerate neurons,variety of virus vectors and transgenic mouse model systems are currently being used.Our previous work has shown that Neurogenic Differentiation 1(NeuroD1)can induce in-vivo reprogramming robustly,enabling the conversion from astrocytes to functional neurons.Moreover,the subiculum(SUB)serves as a hub for neural communication between the hippocampus and the cortex.In the 5xFAD mouse model,the SUB is the first site where A? deposits and neurons are lost.Therefore,the SUB is an ideal region for studying the treatment of AD mouse using in-vivo reprogramming.In this part,with NeuroD1 as the key reprogramming transcription factor andSUB as the area for in-vivo reprogramming,the features of reprogramming under different overexpressing systems were systemically investigated and compared.The results will help us to develop an optimal strategy for efficient in-vivo neuronal regeneration and for clinical treatment of Alzheimer's disease in the future.Purpose: The present part aims to compare four systems generated from combinations of adenoassociated virus(AAV)vectors and transgenic mice,including the AAV-FLEx + AAV-GFAP::Cre system,AAV-GFAP::Neur D1 system,AAV-FLEx + GFAP::Cre transgenic mouse system and AAV-FLEx + Aldh1l1::Cre/ER2 transgenic mouse system.For elucidating the theoretical principles of reprogramming and developing an effective expressing system for NeuroD1-driven astrocyte-toneuron(At N)conversion,the differences of the four systems were compared.The first two gene therapy systems are purely based on AAV and thus are applicable for non-transgenic animals,which means they are promising for clinical use.The latter two systems are combinations of AAV and transgenic animals and thus are suitable for lineage tracing in exploring the mechanism of in-vivo astrocyte reprogramming.Methods: In each of the four systems,we performed stereotactic injection of virus vectors into the dorsal subiculum(dSUB)to induce NeuroD1-mediated in-vivo At N reprogramming.Immunochemistry was used to analyze the NeuroD1 overexpression,reprogramming efficiency,astrocyte specificity and subtypes of astrocyte-converted neurons.Electrophysiology was used for functional analysis of converted neurons.Results: Among all the four systems,AAV-FLEx+Aldh1l1::Cre/ER2 transgenic mouse system had the highest specificity(99.3%)for astrocytes.However,the NeuroD1 expression was too low to induce At N reprogramming when using this system.Notably,in this system,abnormally increased expression of Oligodendrocyte Transcription Factor 2(Olig2)was observed in infected astrocytes in the NeuroD1 group but not in the control group.The AAV-FLEx + Cre-FLEx system and AAV-FLEx + GFAP::Cre mouse system overexpressed NeuroD1 robustly and both generated primarily glutamatergic neurons.NeuroD1 expression was rather low in the AAV-GFAP::NeuroD1 system and the neurons regenerated were mostly GABAergic.The reprogramming efficiency of the AAVGFAP::NeuroD1 system,AAV-FLEx + Cre-FLEx system and AAV-FLEx + GFAP::Cre transgenic mouse system was 43.9%,70.0%,and 92.0%,respectively.All the systems showed high astrocyte specificity(>90%)except the AAV-FLEx + Cre-FLEx system(64.4%).Conclusion: Due to the different features of the four reprogramming systems,decisions should be made appropriately according to the needs of basic research and clinical application.In addition,the abnormal expression of Olig2 in astrocytes induced in the AAV-FLEx + Aldh1l1::Cre/ER2 system might be an obstacle for NeuroD1-driven At N reprogramming.The mechanism remains elusive and further in-depth study is warranted.Part ? Treating Alzheimer's disease mouse with NeuroD1-mediated in-vivo reprogrammingBackground: Alzheimer's disease(AD)is a neurodegenerative disease with insidious onset and slow progression.Cognitive impairment is its major manifestation.The disease mechanism remains unclear.However,pharmacological interventions for AD can only yield limited effectiveness and almost all clinical trials for new medication have failed in recent years.For now,an estimated 10 million Chinese are living with AD.As the population age rapidly,this number is expected to reach 20 million in the next ten years,which will impose a great financial burden for families and the whole society.Multiple lines of evidence have suggested that gradual loss of neurons and synapse plays an important role in the progressive cognitive decline in AD.Meanwhile,animal studies and postmortem studies have identified severe neuronal loss in the cortex and hippocampus of AD brain,both of which are key areas for learning and memory.Accordingly,neuronal replacement therapy based on in-vivo reprogramming might be promising for treating AD.Purpose: As evidenced by Part I,NeuroD1-driven At N reprogramming based on the AAV-FLEx + AAV-GFAP::Cre system can generate glutamatergic neurons in the dSUB of mice.The subtype is the same with that of the lost ones in the hippocampus of 5xFAD mice as previously reported.Therefore,using the AAV-FLEx + AAV-GFAP::Cre system,the present Part of study aims to investigate whether NeuroD1-mediated At N conversion can replenish lost neurons in the dSUB,increase synaptic connections and improve spatial memory of AD mice.Methods: AAV-FLEx + AAV-GFAP::Cre virus vectors were injected into the dSUB of AD mice to induce NeuroD1-mediated At N reprogramming.Immunochemistry was used to quantify dSUB volume change as well as neuron number and synaptic change.Finally,the Morris water maze was applied to measure spatial memory.Results: Immunofluorescence staining results showed that neuron number increased in the dSUB and the atrophy of dSUB was ameliorated after NeuroD1-mediated reprogramming.Furthermore,the dendrites and glutamatergic synapse were also elevated.However,the time spent in target quadrant between the control group and the NeuroD1 group was not significantly different,indicating limited improvement in spatial memory.Conclusion: NeuroD1-mediated At N conversion can alleviate pathological impairment in the dSUB of AD mice whereas the spatial memory was not improved significantly,which indicates that this technique only enables local neuronal repair in AD model.More extensive neuronal regeneration might be necessary for cognitive function improvement.Part ? Olig2 distribution pattern in the adult mouse CNSBackground: Both Olig2 and NeuroD1 belong to b HLH transcription factor family.However,they play different roles in neural development.Briefly,NeuroD1 promotes neurogenesis while Olig2 is essential for the differentiation of motor neurons in spinal cord and oligodendrocytes.In adult mouse brain,Olig2 increases gene expression related to oligodendroglia and restrains endogenous neuronal regeneration.It has also been proved that the promoter and enhancer regions of Olig2 are among those most specific targets of NeuroD1 binding.Therefore,these two transcription factors might interact with each other.As shown in Part I,a 28-day overexpression of NeuroD1 at a low level is insufficient for in-vivo At N reprogramming.Meanwhile,unexpected expression of Olig2 was identified in astrocytes.Hence,we speculated that high expression of Olig2 in astrocytes might inhibit NeuroD1-driven At N reprogramming.Moreover,in Part II,we regenerated neurons in situ in the SUB of AD mouse brain,which ameliorated the atrophy of SUB and improved synaptic connections.However,the spatial memory wasn't improved significantly.Considering that AD pathology occurs in a lot more regions besides SUB,more extensive or even global in-vivo reprogramming is critical for treating AD mouse.Furthermore,considering that Olig2 might suppress NeuroD1-driven reprogramming,the baseline expression pattern of Olig2 in astrocytes in different regions should be clarified before we proceed to study large-scale in-vivo reprogramming.It will also provide us with preliminary data for region-specific in-vivo reprogramming.Purpose: Using immunohistochemistry and Laser Scanning Confocal Microscopy(LSCM) techniques,the present study aims to systemically study Olig2 distribution pattern in the adult mouse CNS and the baseline expression of Olig2 in astrocytes in different CNS regions.Methods: In thirteen selected regions of brain and spinal cord of healthy adult mice,immunofluorescence staining and confocal imaging were employed for colocalization analysis between Olig2 and specific markers for oligodendrocyte progenitor cells(OPCs),oligodendrocyte,neurons,microglia and astrocytes respectively.Olig2 expression level in different cell types was compared based on immunofluorescence intensity.Results: Olig2 was ubiquitously expressed in OPCs and oligodendrocytes,but not in neurons or microglia in the adult mouse CNS.Olig2 was highly expressed in most astrocytes in a series of CNS regions(the olfactory bulbs,midbrain,thalamus,medulla and spinal cord)but rarely expressed in those in the cortex,hippocampus and striatum.Moreover,the expression level of Olig2 in those astrocytes is comparable to that in oligodendrocytes.Conclusion: Olig2~+ astrocytes are enriched in a series of regions in healthy adult mouse CNS.Further research is warranted to answer how Olig2 may affect normal astrocyte function,whether Olig2 inhibits NeuroD1-driven in-vivo reprogramming and whether Olig2 alters the subtypes of converted neurons.These preliminary results in this part will guide our future work in in-vivo astrocyte reprogramming in different CNS regions.