Improve sensitivity, speed, and assay quality for complex biological assays

Oksana Sirenko, PhD | Sr. Research Scientist | Molecular Devices
Jayne Hesley | Sr. Applications Scientist | Molecular Devices
Matt Chan | Optical Engineer | Molecular Devices

With the increasing use of highly-complex cell-based 2D and 3D assays in biologic research, there is a pressing demand for improvements to the capabilities of automated high-content imaging. Here, we demonstrate improvement in assay sensitivity, quality, and speed of acquisition with the ImageXpress® Confocal HT.ai High-Content Imaging System. Using a high-power laser light source, the system significantly increases light throughput to the sample, which results in brighter images, increased sensitivity, and increased assay throughput. The impact is especially important for the assays where sensitivity and imaging time are the limiting factors. To demonstrate the practical impact of laser light sources, we present results from several complex biological assay systems: a GPCR activation assay, cancer spheroids, and lung organoids.

Methods

The Instrument

The ImageXpress Confocal HT.ai system is equipped with a multi-line laser light source with matched filters, spanning 405 nm to 730 nm excitation. This light source provides drastically increased illumination power to the sample in comparison to the previous-generation LED light source.

In the examples presented below, comparisons were done between the ImageXpress Confocal HT.ai system and the previous generation ImageXpress Micro Confocal system.

Cell assays

The Transfluor® Assay was performed as previously described using cell line expressing GFP-tagged betaarrestin that associates with the receptor of interest upon activation¹. Cells were stimulated with isoproterenol causing dose-dependent appearance of aggregated internalized receptors (pits), that were visualized with a 20X Plan Apo objective (60’ spinning disc) in the FITC channel.

3D spheroids were formed from an HTC116 colon cancer cell line (ATCC) using U-bottom low attachment (Corning) microplates as previously described² (Sirenko, 2015). Spheroids were treated for 48 hours with anti-cancer drugs, fixed and stained with DRAQ5, HCS CellMask Orange or AF555 Phalloidin, and Whole Cell Green or AF488 Phalloidin (Thermo Fisher Scientific). Spheroids were imaged in 3D (Z-series with 5 μm steps) using a 20X Plan Apo objective. Cell analysis was performed on maximum projection images in MetaXpress® High-Content Image Acquisition and Analysis Software.

3D organoids were formed from primary human lung epithelial cells (ScienCells) in Matrigel (Corning) domes, prepared using a Stem Cell Technologies kit and protocols³. Mature lung organoids (after six weeks of development) were fixed and stained with Hoechst, MitoTracker Orange, and AF488 Phalloidin. Organoids were imaged at 10X magnification in 3D (Z-stack covering a range of 150-250 μm deep) and the images were analyzed using MetaXpress 3D image analysis tools.

Results

GPCR activation assay, Transfluor

G-protein coupled receptors are the largest class of pharmaceutical targets that play a major role in biological screening assays. The Transfluor Assay quantitates internalization of GFP-tagged beta-arrestin that associates with the activated receptor of interest. This internalization results in the appearance of small fluorescent pits that are counted and characterized by high-content imaging. Pits are relatively dim and require ~1 sec exposure with an LED light source. The laser light source of the ImageXpress Confocal HT.ai system reduced exposure time by 3-4X, and correspondingly reduced imaging time by 33% (1.5X gain in assay speed). In addition, assay Z’ values also increased by approximately 20%.

Figure 1. A. Small dots (pits) in green formed upon receptor activation with isoproterenol. Nuclei were stained with Hoechst (blue). Images were acquired using the same exposure times for LED and laser light sources (optimized for lasers). B. Optimized exposure times and duration of imaging were substantially reduced using the laser light source. C. The assay quality (Z’ value) was compared for images taken with laser and LED systems.

3D cancer spheroids

In the second assay, we compared a standard imaging system ImageXpress Confocal with LED excitation to one with laser excitation. HCT-116 spheroids were grown for four days in round bottom plates, with some wells receiving anti-cancer compound treatment with 5 μM of either Cytochalasin D or Nocodazole for the last two days. Spheroids were fixed and stained with either DRAQ5 (nuclei), Whole Cell Green, and Alexa Fluor 555 Phalloidin (actin); or DRAQ5, HCS CellMask Orange (whole cell), and Alexa Fluor 488 Phalloidin (actin). Z-series images were acquired with a 20X Plan Apo objective to a spheroid depth of 150 μm (31 steps at 5 μm step size). Image analysis was performed on the maximum projection images to count nuclei and determine spheroid area. The images were taken using an exposure that yielded a 14-bit image whether acquired using LED or Laser excitation. By independently optimizing exposure times for each light source, 3D spheroid acquisition was performed in approximately half the time using lasers vs LED, depending upon number of channels collected and fluorophores used.

Figure 2. A. 2D projection image of DRAQ5 and Phalloidin-AF488 overlay of an untreated spheroid (right half) and nuclear segmentation mask (left half). B. Exposure times optimized to acquire images of equivalent pixel intensities. Different exposures were required for different staining protocols. C. Speed of acquiring a 96-well spheroid plate in the ImageXpress Confocal HT.ai system at 20X using LED vs laser light sources.

Figure 3. A. Confocal image of organoid culture in Matrigel (maximum projection), 10X. B. Images were taken using same exposure times for LED and laser light sources. Cell count and average intensity of objects (cells with intact mitochondria) shown for representative images. C. Exposure times were matched (14 bit intensity range) for both laser and LED. Optimized exposure times and duration of imaging were substantially reduced by using lasers.

3D lung organoids

The lung organoid cultures were started from primary lung epithelial cells, and then grown in Matrigel domes using reagents and protocol from Stem Cell Technologies. Briefly, cells were first expanded in 2D, then mixed with growth factor reduced Matrigel and seeded into Matrigel domes in 24-well plates, or other plate formats. Organoids are a very useful tool for disease modeling and assessment of compound effects. Automated imaging and analysis of organoids are important methods for quantitative assessment of phenotypic changes in organoids, and for increasing the throughput of organoid experiments. Confocal Imaging and 3D image analysis are especially useful for capturing the complexity of 3D biological systems.

We have evaluated the impact of lasers on imaging of 3D organoid samples. Organoids comprised spherical objects with complex cavities and vesical structures that were imaged using 20-30 Z-planes through the Matrigel with 10X-40X magnification. Organoids were treated with compounds known to damage lung tissues, and were stained with markers visualizing mitochondria, cytoskeleton, and cell junctions. 3D volumetric analysis was performed to count and characterize organoids and cells inside the organoids. We were able to observe increased pixel intensity, resolution, and sharpness of objects that resulted in higher quality analysis. Importantly, the speed of imaging was increased 2.3X (51-57% time reduction) due to an 8X decrease in total exposure time.

Implications of laser light source for assay speed and throughput

To test the impact of lasers on assay speed and throughput of high-content screening assays we have compared exposure times and acquisition speed for 10 independent assays performed by five scientists. The data below illustrate the observed decrease in exposure times and increased speed (time reduction) for these 10 assays.

Conclusion

The illumination power of the laser light source significantly increases image intensity and assay sensitivity, which is especially important for dim samples. The imaging system with lasers substantially decreased exposure times, which resulted in increased imaging speed and assay throughput. 3D imaging especially benefited from the laser light source. We characterized several biological assays and demonstrated higher image intensities and improved image quality, which resulted in increased assay sensitivity and imaging speed.

1. Transfluor, MolDev https://www.moleculardevices.com/en/assets/app-note/dd/img/gpcr-activation-using-metaxpress-acuityxpresssoftware- and-transfluor-assay-system#gref
2. Sirenko et.al., 2015, Assay and drug development technologies https://www.liebertpub.com/doi/pdfplus/10.1089/adt.2015.655
3. Stem Cell Technologies, https://www.stemcell.com/pneumacult-airway-organoid-kit.html