Alzheimer’s disease: 3D culture system brings hope to drug discovery

Alzheimer’s disease (AD) is one of the most devastating human pathologies. AD is the leading cause of age-related dementia and currently afflicts more than 44 million persons worldwide (World Alzheimer Report 2014). It is characterized by a cognitive decline and memory loss and by the appearance of two pathological hallmarks: beta amyloid plaques and cytoskeletal pathology.

Our knowledge of the causes and mechanisms of Alzheimer’s disease are still limited, partly because no non-invasive and ethical methods currently exist to access brain tissue of human patients. To date, several mice models recapitulates diverse symptoms of AD but it does not exist a single model (in vivo or in vitro) showing all the pathophysiological changes occurring in the human disease (Aβ overproduction and accumulation in plaques and tau hyperphosphorylation and formation of neurofibrillar tangels). Here is where Choi and colleagues ¹ has brought some light to AD research. The authors have developed a new procedure to grow neurons in a 3D gel matrix that mimics all the pathological alterations induced in human patients.

Aβ overpdocution is produced by proteolysis of a bigger protein called Amyloid precursor protein or APP. APP can be cleved by several proteases but the ones related to AD are beta-secretases (BACE1) and gamma-secretases (i.e. Presenilin). So what the authors did first was to introduce two mutations in the APP gene (the Swedish mutation, that increase the levels of Aβ-40 and -42 and London, a mutation that increases the Aβ42/Aβ40 ratio by increasing Aβ42 levels, the most toxic APP fragment) together with a mutation in Presenilin1 via lentiviral vectors on human stem cells. These neuronal stem cells were then grown in a 3D gel matrix instead that in normal 2D cell cultures. A specific gel matrix with high levels of brain extracellular matrix proteins was chosen in order to mimic as much as possible the physiological situation. Under these conditions, neuronal stem cells differentiate more to neurons than in standard cell cultures (2D).

The surprisingly results obtained shown a completely representation of the human pathology regarding the two major hallmarks. First, under the hypothesis that in a gel matrix Aβ will not be diluted in the medium and will form plaques, the authors could find, after 6 weeks of culture the appearance of large amyloid deposits on the cultures expressing the APP mutations, visualized thanks to confocal microscopy. Importantly, the authors could observe also higher levels of phosphorylation of tau and Western blots revealed the classical accumulation of soluble and insoluble hyperphosphorylated tau. This is very important since the APP models are characterized by the absent of any tau pathology. The tau pathology was accompanied, like in humans, by an aberrant neuronal morphology, and after 10 weeks of cell culture, tau tangles were present in the neurons.

This nicely designed model is the first time that the scientific community is able to mimic the whole plethora of changes that occurs in humans. As mentioned above, we have available several mice models to assess the disease but normally they only develop a “small” part of the pathology. This new 3D cell culture system will undoubtedly allow the scientific community to better assess drugs in order to find therapeutical approaches that will delay or prevent AD. As an example, the authors also shown that treatment of the cultures with either beta- or gamma-secretase inhibitors not only reduced Aβ production but also tau phosphorylation. This is very interesting since is still not clear whether Aβ precedes tau phosphorylation or if the latest is produced as a consequence of Aβ overproduction. Amyloid hypothesis postulates that Aβ overproduction is the first molecular disarrangement produce in AD and that with time, precipitates the tau pathology. This 3D model demonstrates the validity of the Amyloid hypothesis opening an important therapeutical window to prevent the neurodegeneration in AD patients.

In general, the model proposed by Choi and colleagues open a new era on drug testing on AD since will allow a faster and cheaper method to screen for compounds with the ability not only to block Aβ overexpression, but also tau hyperphosphorylation.