High-Content-Screen Microscopy Enables Large-Scale Multiplex Image Analysis

Jinhe Li and Min Hu, Abbott Laboratories

Quantitative analysis of cellular images is critical for investigating neuronal function and disease mechanisms in modern neuroscience drug discovery research. In the study of Alzheimer’s disease, neurotoxicity of β-amyloid peptide 1-42 (Aβ_1-42) has been a primary focus of research for the past decade.

Although β-amyloid has long been known to play a central role in Alzheimer’s disease, the underlying mechanisms still remain unclear. For example, although β-amyloid-peptide-mediated cell death has been observed in various cell-based models — in transgenic mice overexpressing the mutant forms of the amyloid-β precursor protein that demonstrate cognitive deficit — there has not been profound cell death. Instead, these mice showed decreased dendritic spine density that seems to correlate with impaired long-term potentiation in the hippocampus as well as with behavioral deficits.

In Alzheimer’s disease patients, Aβ_1-42 appears to disrupt synaptic function during early onset, and cholinergic neurons either are spared or affected only in late stages of the disease. It is thus emerging that Aβ_1-42-induced memory deficits may involve subtler neuronal changes that lead to synaptic deficits before neurodegeneration occurs. One way to assess synaptic properties of neurons is to measure neurite outgrowth in vitro with quantitative image analysis performed by microscopy.

We studied Aβ_1-42 toxicity by measuring neurite outgrowth in primary rat cortical cell cultures. Neurite outgrowth typically is measured by conventional microscopy, but this technique alone cannot perform high-throughput analysis of subtle changes. Thus, we employed high-content screening using automated fluorescence microscopy, which can acquire and analyze a large number of cell images.

Figure 1. This representative image assembly shows part of a 96-well plate containing rat cortical cell cultures. Cells were labeled in a way that indicated total cell number (blue), glia (red) and neurite outgrowth (green). Different color intensities in various wells (corresponding to groups of four images) were from different levels of neurite labeling, indicating various effects of certain treatments on neurite outgrowth.

Images were automatically acquired from several fields of view in each compartment of a 96-well plate (Figure 1). Each field of view contained hundreds of cells. The number of cells (blue) and glial number (red) as well as neurite outgrowth (green) were analyzed in each image (Figure 2). Multiple measurements were performed using the same cell population in an array of samples, which generated large data sets under various conditions. This avoids subjectiveness in the selection of images and reduces variability between successive measurements, typically encountered with conventional microscopy. In other words, neurite outgrowth measurements can be based on a larger amount of quantitative data. These procedures ensured that a reliable and sensitive measurement of markers relevant to neuronal function was performed.

Figure 2. These images are representive of an individual field of view. Untreated rat cortical cells show normal neurite outgrowth (green) (A). Cells treated with Aβ_1-42show reduced neurite outgrowth (green) (B).

Our studies using this approach demonstrated that neurite outgrowth could be significantly reduced by the Aβ_1-42 peptide without substantially reducing neuronal cell numbers, as detailed in the March issue of Brain Research. The Aβ_1-42-induced reduction of neurite outgrowth was not observed with a scrambled peptide but was attenuated by memantine (the receptor antagonist for an N-methyl-D-aspartate) and by α7 nicotinic acetylcholine receptor-selective ligands.

Our studies suggest that reduction of neurite outgrowth may serve as a model representing Aβ_1-42-mediated synaptic failure in Alzheimer’s disease, which, in combination with high-content screening, provides a high-throughput cell-based assay that can be used to evaluate potential neuroprotective compounds in vitro.

Meet the authors

Jinhe Li and Min Hu are research investigators at Neuroscience Research, Global Pharmaceutical Research & Development at Abbott Laboratories in Abbott Park, Ill.; e-mail: jinhe.li@abbott.com.