Can Alzheimer’s-in-a-dish models accelerate AD drug discovery?

Alzheimer’s disease (AD) transgenic mice have been used as a standard model for AD drug discovery and basic mechanistic studies. These mouse models overexpress amyloid β precursor protein (APP) or APP/presenilin (PS) with single or multiple familial AD (FAD) mutations, which lead to excess accumulation of amyloid β (Aβ), a well-known driver for AD pathogenesis. These mice exhibit several symbolic features of AD including Aβ-induced synaptic/memory deficits and Aβ aggregation (Aβ plaque), which have been proven very useful for testing candidate AD drugs before human trials. However, they have not been able to fully replicate key AD pathogenic cascades including clear tau tangle pathology and neurodegeneration, which might explain why many successful treatments in mice did not lead to similar success in humans.

The failure to fully recapitulate AD pathologies in mice challenges the reigning Aβ hypothesis, which predicts that accumulation of excess pathogenic Aβ leads to neurofibrillary tangles and eventual neurodegeneration. However, the shortcomings of current mouse models based on FAD mutations may possibly be due to fundamental species-specific differences. Indeed, adult mice do not express the six human tau isoforms essential for recapitulating tau tangle pathology, and furthermore, endogenous mouse tau proteins seem to interfere with the aggregation of human tau [1,2].

The problems that may rise from species-specific differences can be easily solved in human models. Induced pluripotent stem cell (iPSC) technology has provided a new model system to recapitulate AD pathogenesis in human neuronal environments. iPSC-derived human neurons have been generated from the fibroblasts of AD patients, including those with APP or PS FAD mutations [3-11]. As predicted, these FAD neurons showed significant increases in pathogenic Aβ species and interestingly, some of them showed accumulation of hyperphosphorylated tau, be an early signal of developing tau tangle pathology. However, these cellular models have been unsuccessful in demonstrating robust Aβ plaques, tau tangles, and neurodegeneration.

Recently, we developed a novel 3D human neural culture model that produces robust Aβ plaques and tau tangles [12,13]. As in AD mouse transgenic models, we overexpressed APP and PS1 with FAD mutations in immortalized human neural progenitor cells and differentiated them into neurons and astrocytes in 3D Matrigel culture conditions. We found that this 3D system dramatically accelerated the aggregation of pathogenic Aβ species (Aβ plaques) and more excitingly, induced accumulation of silver-positive, detergent-insoluble hyperphosphorylated tau protein, which clearly indicates the presence of tau tangles. Using this model, we were also able to show that blocking Aβ generation dramatically decreased hyperphosphorylated tau accumulation, which supports the current Aβ hypothesis.

These exciting new human cellular AD models offer faster, cheaper high-throughput screening platforms for novel AD drug discovery in comparison to current transgenic mouse models. While AD iPSC-derived models could be used for testing candidate drugs that intervene in early-stage AD pathogenesis, our 3D culture model may a good fit for discovering drugs that block later points in pathogenic cascades, such as accumulation/aggregation of Αβ and hyperphosphorylated tau. AD transgenic mouse models could subsequently be used to re-confirm drug targets identified in the cellular models. In addition to drug screening, human cellular AD models can be used to explore AD pathogenic mechanisms in human neuronal cells, which might provide additional targets for AD drug discovery.

The exciting new array of cellular models may be the keys we need to unlock the mysteries of AD pathogenesis, but we must strive to optimize our current models and continue to create new ones. Until now, the cellular AD models have not been able to precisely reconstitute the brain regions most affected in AD: the hippocampus and cortical layers. Recent progress in generation of organoid systems holds much promise [14]. Additionally, iPSC and our 3D culture model may have captured a significant part of AD pathogenesis, but we still have yet to show the clear neurodegeneration seen in brains of AD patients. Better understanding of the role microglial cells play in AD pathogenesis will be essential for the development of human cellular models that reconstitute robust neuronal death stemming from Aβ and tau pathologies. These integrated neuroinflammatory models will be crucial for movement towards better a mechanistic understanding of AD pathology, and development of effective therapeutics.

Currently, AD affects 5.3 million individuals in the US and that number is expected to increase dramatically over the next decade. We are cautiously optimistic that these new cellular AD models can accelerate discovery of new AD drugs and also enable dissection of molecular mechanisms underlying the pathogenic cascades of AD.

Doo Yeon Kim, PhD, Jenna Aronson, and Rudolph E. Tanzi, PhD

References
[1] Andorfer C, Kress Y, Espinoza M, de Silva R, Tucker KL, Barde YA, Duff K, Davies P (2003) Hyperphosphorylation and aggregation of tau in mice expressing normal human tau isoforms. J Neurochem 86, 582-590.
[2] Ando K, Leroy K, Héraud C, Yilmaz Z, Authelet M, Suain V, De Decker R, Brion JP (2011) Accelerated human mutant tau aggregation by knocking out murine tau in a transgenic mouse model. Am J Pathol 178, 803-816.
[3] Moore S, Evans LD, Andersson T, Portelius E, Smith J, Dias TB, Saurat N, McGlade A, Kirwan P, Blennow K, Hardy J, Zetterberg H, Livesey FJ (2015) APP metabolism regulates tau proteostasis in human cerebral cortex neurons. Cell Rep 11, 689-696.
[4] Muratore CR, Rice HC, Srikanth P, Callahan DG, Shin T, Benjamin LN, Walsh DM, Selkoe DJ, Young-Pearse TL (2014) The familial Alzheimer's disease APPV717I mutation alters APP processing and Tau expression in iPSC-derived neurons. Hum Mol Genet 23, 3523-3536.
[5] Sproul AA, Jacob S, Pre D, Kim SH, Nestor MW, Navarro-Sobrino M, Santa-Maria I, Zimmer M, Aubry S, Steele JW, Kahler DJ, Dranovsky A, Arancio O, Crary JF, Gandy S, Noggle SA (2014) Characterization and molecular profiling of PSEN1 familial Alzheimer's disease iPSC-derived neural progenitors. PLoS One 9, e84547.
[6] Duan L, Bhattacharyya BJ, Belmadani A, Pan L, Miller RJ, Kessler JA (2014) Stem cell derived basal forebrain cholinergic neurons from Alzheimer's disease patients are more susceptible to cell death. Mol Neurodegener 9, 3.
[7] Woodruff G, Young JE, Martinez FJ, Buen F, Gore A, Kinaga J, Li Z, Yuan SH, Zhang K, Goldstein LS (2013) The presenilin-1 ΔE9 mutation results in reduced γ-secretase activity, but not total loss of PS1 function, in isogenic human stem cells. Cell Rep 5, 974-985.
[8] Kondo T, Asai M, Tsukita K, Kutoku Y, Ohsawa Y, Sunada Y, Imamura K, Egawa N, Yahata N, Okita K, Takahashi K, Asaka I, Aoi T, Watanabe A, Watanabe K, Kadoya C, Nakano R, Watanabe D, Maruyama K, Hori O, Hibino S, Choshi T, Nakahata T, Hioki H, Kaneko T, Naitoh M, Yoshikawa K, Yamawaki S, Suzuki S, Hata R, Ueno S, Seki T, Kobayashi K, Toda T, Murakami K, Irie K, Klein WL, Mori H, Asada T, Takahashi R, Iwata N, Yamanaka S, Inoue H (2013) Modeling Alzheimer's disease with iPSCs reveals stress phenotypes associated with intracellular Aβ and differential drug responsiveness. Cell Stem Cell 12, 487-496.
[9] Xu X, Lei Y, Luo J, Wang J, Zhang S, Yang XJ, Sun M, Nuwaysir E, Fan G, Zhao J, Lei L, Zhong Z (2013) Prevention of β-amyloid induced toxicity in human iPS cell-derived neurons by inhibition of Cyclin-dependent kinases and associated cell cycle events. Stem Cell Res 10, 213-227.
[10] Israel MA, Yuan SH, Bardy C, Reyna SM, Mu Y, Herrera C, Hefferan MP, Van Gorp S, Nazor KL, Boscolo FS, Carson CT, Laurent LC, Marsala M, Gage FH, Remes AM, Koo EH, Goldstein LS (2012) Probing sporadic and familial Alzheimer's disease using induced pluripotent stem cells. Nature 482, 216-220.
[11] Yagi T, Ito D, Okada Y, Akamatsu W, Nihei Y, Yoshizaki T, Yamanaka S, Okano H, Suzuki N (2011) Modeling familial Alzheimer's disease with induced pluripotent stem cells. Hum Mol Genet 20, 4530-4539.
[12] Choi SH, Kim YH, Hebisch M, Sliwinski C, Lee S, D'Avanzo C, Chen H, Hooli B, Asselin C, Muffat J, Klee JB, Zhang C, Wainger BJ, Peitz M, Kovacs DM, Woolf CJ, Wagner SL, Tanzi RE, Kim DY (2014) A three-dimensional human neural cell culture model of Alzheimer's disease. Nature 515, 274-278.
[13] Kim YH, Choi SH, D'Avanzo C, Hebisch M, Sliwinski C, Bylykbashi E, Washicosky KJ, Klee JB, Brüstle O, Tanzi RE, Kim DY (2015) A 3D human neural cell culture system for modeling Alzheimer's disease. Nat Protoc 10, 985-1006.
[14] Lancaster MA, Renner M, Martin CA, Wenzel D, Bicknell LS, Hurles ME, Homfray T, Penninger JM, Jackson AP, Knoblich JA (2013) Cerebral organoids model human brain development and microcephaly. Nature 501, 373-379.

Last comment on 8 April 2016 by Christopher Navara

Comments

The blog by Kim and colleagues highlights the exciting potential of their novel Alzheimer's-in-a-dish model for drug discovery and uncovering novel mechanisms of AD pathogenesis. The authors rightly point out the limitations of the current transgenic mouse models of AD. Frustratingly, the AD field has seen time and time again that therapies that can cure Alzheimer's mouse models fail miserably when put to the test in humans. The embarrassing failure of the mouse models to recapitulate this important aspect of human AD has plagued the field now for nearly 20 years. Excitingly, the field is now poised for a big turnaround thanks to the authors’ 3-D Alzheimer's-in-a-dish model.

As Kim and colleagues point out, a key feature of their 3-D culture system is the fact that they use human-derived induced pluripotent stem cell neurons, which are able to recapitulate the neurofibrillary tangle pathology observed in actual AD brains, a feature that mouse models cannot do unless they overexpress human tau with frontotemporal dementia (FTD) mutations. This represents a truly seminal advance in the field and promises to rectify the previously mentioned deficiencies of the transgenic mouse models. Importantly, by recapitulating tau pathology without the need for FTD mutations, the Alzheimer's in a dish model also supports the current Abeta hypothesis of Alzheimer's disease pathogenesis.

It will be exciting to see in the future how this 3-D Alzheimer's in a dish model evolves. As the authors point out, the model promises to revolutionize AD drug discovery and allow investigators to rapidly and cost-effectively perform high throughput screening to identify novel Alzheimer’s disease drugs, thus opening the possibility of large scale drug discovery in academic groups, a purview that had previously been almost entirely limited to the pharmaceutical industry. Additionally, the authors predict that their 3-D model may also be useful for identifying drugs that block later points in the AD pathogenic cascade, such as the formation of tau pathology. Another exciting development would be the inclusion of other brain cell types in the 3-D model, such as microglia and astrocytes, both of which are important for the neuroinflammatory and cell death pathologies observed in human AD brain. As iPSC protocols become more sophisticated, allowing the differentiation of different neuronal cell types, the exciting possibility exists that different brain regions might be modeled in vitro in the 3-D system. Thus, beyond Alzheimer's disease, the authors’ 3-D system may be more widely applicable to modeling brain regions for the purpose of greater understanding of normal human brain function. Indeed, it may be that the sky is the limit for the potential applications of this exciting new technology.

We wholeheartedly agree with Kim et al. regarding the utility of induced pluripotent stem cells (iPSCs) as a “disease in a dish” model of Alzheimer’s disease (AD) and wish to highlight several advantages of this research tool that these authors did not mention. Neurons derived from iPSCs allow for the study of disease phenotypes within the complex genotype of human patients and unlike transgenic murine models provide a mechanism for the study of sporadic AD (SAD) [1-3] which represents the vast majority of AD patients. Further, transgenic mouse models fail to fully recapitulate human neurons, and therefore do not provide the most accurate model for testing candidate AD drugs. Pluripotent stem cell disease in a dish models do yield accurate human neurons—which can also recapitulate each patient’s unique genetic makeup. However, these neurons are not in their normal in vivo context. In addition to using transgenic mice to validate findings from the 3D culture model, the use of nonhuman primates for additional preclinical testing would provide a more accurate animal model for studies in vivo.

Both 2D and 3D models have utility to further our understanding of disease initiation and progression and are amenable to high-throughput screening for possible therapeutic agents. 2D cellular cultures provide the ability to examine subcellular changes occurring within the diseased cells [4] as well as probe for new markers of SAD [3]. 2D culture provides greater access for manipulations to test cell function at the cellular level including analysis of single cells [5] to probe the heterogeneity of neurons including live cell analysis. The 3D model discussed here can provide several characteristics that recreate the AD phenotype at the tissue level, which is very beneficial for testing AD drugs.

A singular advantage of disease in a dish modeling is the repeatability it affords to studies of human disease. While tissue samples from patients are in limited supply and precious. Induced pluripotent stem cells are likely immortal and thus provide a virtually endless supply of human neurons for study. This allows repeated measures within the same genetic background and further allows for the distribution of cell lines so that other researchers can repeat studies. To this end, the National Institute of Aging and the National Institute of Neurological Disorders and Stroke have an established cell repository of AD patient samples. This repository is a valuable resource for researchers; however, its current sample demographics do not represent the diversity of the AD patient population, and this is a need that must be addressed.

In summary, the disease in a dish cellular model approach offers great potential for studies of the pathogenesis of AD, potential new biomarkers of AD that might be used for diagnostics, and testing candidate AD drugs, and the 3D cellular model described here appears to provide the most accurate disease in a dish model

Doug Grow, John McCarrey and Christopher Navara
University of Texas at San Antonio, Department of Biology and the San Antonio Cellular Therapeutics Institute

References
[1] Israel MA, Yuan SH, Bardy C, Reyna SM, Mu Y, Herrera C, Hefferan MP, Van Gorp S, Nazor KL, Boscolo FS, Carson CT, Laurent LC, Marsala M, Gage FH, Remes AM, Koo EH, Goldstein LS (2012) Probing sporadic and familial Alzheimer's disease using induced pluripotent stem cells. Nature 482, 216-220.
[2] Kondo T, Asai M, Tsukita K, Kutoku Y, Ohsawa Y, Sunada Y, Imamura K, Egawa N, Yahata N, Okita K, Takahashi K, Asaka I, Aoi T, Watanabe A, Watanabe K, Kadoya C, Nakano R, Watanabe D, Maruyama K, Hori O, Hibino S, Choshi T, Nakahata T, Hioki H, Kaneko T, Naitoh M, Yoshikawa K, Yamawaki S, Suzuki S, Hata R, Ueno S, Seki T, Kobayashi K, Toda T, Murakami K, Irie K, Klein WL, Mori H, Asada T, Takahashi R, Iwata N, Yamanaka S, Inoue H (2013) Modeling Alzheimer's disease with iPSCs reveals stress phenotypes associated with intracellular Aβ and differential drug responsiveness. Cell Stem Cell 12, 487-496.
[3] Young JE, Boulanger-Weill J, Williams DA, Woodruff G, Buen F, Revilla AC, Herrera C, Israel MA, Yuan SH, Edland SD, Goldstein LS (2015) Elucidating molecular phenotypes caused by the SORL1 Alzheimer's disease genetic risk factor using human induced pluripotent stem cells. Cell Stem Cell 16, 373-385.
[4] Usenovic M, Niroomand S, Drolet RE, Yao L, Gaspar RC, Hatcher NG, Schachter J, Renger JJ, Parmentier-Batteur S (2015) Internalized tau oligomers cause neurodegeneration by inducing accumulation of pathogenic tau in human neurons derived from induced pluripotent stem cells. J Neurosci 35, 14234-14250.
[5] Liao MC, Muratore CR, Gierahn TM, Sullivan SE, Srikanth P, De Jager PL, Love JC, Young-Pearse TL (2016) Single-cell detection of secreted Aβ and sAPPα from human IPSC-derived neurons and astrocytes. J Neurosci 36, 1730-1846.

Pages