Date of Award


Document Type


Degree Name

Master of Science in Biology



First Reader/Committee Chair

Bournias-Vardiabasis, Nicole


Alzheimer’s disease (AD), identified over 100 years ago and intensively studied since the 1970s, has no effective treatments or mechanistic understanding of the underlying neurodegenerative process. Most investigators believe accumulation or aggregation of amyloid beta (Ab) proteins plays a causative role. Aβ peptides (~39-43 residues) are generated by proteolysis of the transmembrane protein APP. One reason we know so little about AD is an incomplete understanding of the cellular mechanisms responsible for Ab proteotoxicity. Human ES and iPSC models of AD are recent additions to many other models used to investigate these mechanisms. AD, however is a chronic progressive condition of old age and cultured neurons may not live long enough to model what goes wrong in neurons from AD patients. In my research, I used hESCs which directly express Ab peptides thus avoiding the time it takes to process APP. One App allele in H9 hESCs was previously edited using TALEN. A homologous recombination cassette coding directly for a secretory form of either Ab1-42 or Ab1-40 and containing a stop codon, was inserted into the first exon of App upstream of the normal translational start site. I used multiple independently isolated clones of edited cells with 3 genotypes: App/App (unedited), App/Aβ1-40 and App/Aβ1-42. Expression of Ab from edited alleles was confirmed by qRT-PCR using primers specific for the edit. I first sought to establish if editing changed any aspects of neuronal differentiation in culture. All 3 genotypes have similar embryoid body (EB) development, and similar numbers and sizes of neuronal clusters (NC) up to 34 days after EB dissociation and neural differentiation. Immunostaining of neuronal markers, NeuN and DCX (doublecortin), likewise revealed no difference among edited and unedited cells, suggesting that the edits do not affect the ability of my stem cells to differentiate into neurons. I next measured accumulation of aggregated Ab using an aggregate specific antibody, 7A1a. Data at 34-days post EB dissociation indicates NCs in the Aβ1-42 edited cells accumulate significantly more aggregates relative to either unedited or Ab1-40 edited lines, a result consistent with the increased ability for Ab1-42 to form aggregates. Aβ aggregates also appear to be concentrated around fragmented nuclei within neuronal clusters suggesting that intracellular accumulation may play a key role in proteotoxicity. Additionally, I observed a significant decrease in the number of synapsin1 puncta, a marker of synapses, another feature of AD. I documented a nearly 3-fold greater neuronal cell death in both the Aβ1-40 and Aβ1-42 neurons at 70 days after differentiation. RNA sequencing data also shows independently isolated clones group together and show differential expression of genes related to memory and neuronal cell death. The early presence of Aβaggregation and subsequent cell death is in line with the chronic and progressive nature of AD and this is the first known model to exhibit a neurodegenerative phenotype. These isogenic cell lines thus appear to be useful to screen for therapeutics that may prevent or slow Ab1-42 dependent neurodegeneration and a tool to investigate Ab-dependent mechanisms with relevance to AD.