Improved Visualization and Quantitative Analysis of Drug Effects Using Micropatterned Cells

The overall goal of the following procedure is to show how normalization of cell architecture and position achieved on specific adhesive micro patterns allows easy automation of image acquisition and straightforward image processing with sensitive and robust quantification of drug effects, which is unattainable on conventional glass cover slip supports. This is achieved by seeding heli cells on SI two chips that bear L-shaped SI three labeled micro patterns. The cells adopt the triangular shape building a major stress fiber spanning the two ends of the L.The pattern cells are next incubated with lebos statin or left untreated, then fixed and stained to visualize the acton cytoskeleton and nuclei.

Images of the micropattern cells are then acquired and processed using the cell ref macro to generate a stack of images for analysis and the hypotenuse macro to quantify phenotypes. Ultimately, results can be obtained that show a significant effect of low doses of BLEs statin on micropattern HELOC cells, which is undetectable in standard cell culture conditions. Hi, welcome.

We're here today to walk you through our experimental protocol explaining how to use adhesive micro patterns for cell analysis as well as showing you experimental data illustrating the easy quantification of very low doses of drugs on these adhesive micro patterns. So when first seeing these micro patterns, you will immediately understand how it can solve variability issues in image capture during cell analysis. Because by having a really array of cells, much like a micro array, you can easily move the microscope stage from cell to cell to be able to capture images in a stepwise fashion.

But that of course is not the only advantage of micro patterns it offers. It offers much more improvements because in fact, on certain typical micro patterns such as the crossbo or the Y, we are in fact are normalizing internal cell architecture, meaning that organelles, the mitotic spindle, the cytoskeleton and different protein networks are all in the same locations from cell to cell or in the same orientation from cell to cell, which makes it much easier to predict and quantify the effects of drugs on these particular patterns. Now this may seem to be quite counterintuitive from what is normally done in the classical Petri dish or microplate.

Well, because you often see of course, cells moving around and adopting different morphologies. Whereas on the adhesive micro patterns, all cells will start to look alike because they are responding to this adhesive micro pattern and organizing their architecture in the same way. This may seem in fact a bit artificial, but when you come to look at it closely, isn't the Petri dish even more artificial?

Because in tissues, cells are constrained by the neighboring cells as well as by the extracellular matrix. So very much they're receiving a lot of spatial information, which is orienting the cell architecture. And this on onsite to ad of micro patterns is what we are doing.

We are restoring this spatial information to cells and all cells are responding to this and organizing in the same way. And now because cells all lookalike, we can introduce the concept of the reference cell, meaning that we can average all images and obtain a single image, which is really representative what the cell looks like in a given condition. Then you can add a drug and look again at that reference cell and see how it has modified the morphology or the organization of the cell in response to that drug.

Place two situ two chips individually into two independent wells of a six well plate, ensuring that the situ two logo is readable. The situ two chips used for this experiment have micro patterns stained with FibroGen SI three and should be kept in the dark before use. Collect hela cells that are approximately 80%confluent by trypsin and check that they're properly individualized under a microscope centrifuge at 300 G for four minutes and resus suspend the cells gently count and dilute the cells to a concentration of 15, 000 cells per milliliter.

In incomplete D-M-E-M-F 12 culture media dispense 60, 000 cells per well. The plate should be moved as little as possible to avoid introducing rotating movements in the medium, which will tend to concentrate the cells at the center. Allow the cells to sediment for 10 minutes under the hood, then move them to the cell incubator.

Within 10 to 20 minutes in the cell incubator, cells will start to adhere after 10 minutes, check regularly under the microscope when cells have attached. Change the cell medium and remove cell excess by repeating the following procedure four times. Keeping the plate flat gently aspirate the medium from the side of the well.

Be careful to keep enough media to avoid the chip from drying out. Add four milliliters of PBS and aspirate starting at the center of the chip. Then moving to the side of the well to aspirate most of the PBS as before, check under the microscope for floating cells.

If a large amount of floating cells remain, repeat the washing procedure. If very few cells are present, aspirate off part of the PBS and replace it with two milliliters of fresh medium aspirate and add four milliliters of fresh medium twice. Incubate the cells for three hours in a cell incubator to allow the cells to fully spread.

Using this method, between 10 and 30%of the micro patterns will be occupied by a single cell, whereas other patterns will be either empty or occupied by multiple cells. To treat the cells with BLEs statin dilute the 17 millimolar drug stock solution to a final concentration of five micromolar with 0.1%DMSO in four milliliters of medium. For the control prepare medium with 0.1%DMSO only aspirate the cell medium and replace it quickly with the BLEs statin or the control solutions to avoid the chip from drying out.

Incubate the cells for one hour at 37 degrees Celsius. During this incubation, prepare the cytoskeleton buffer. Add one gram of sucrose to two milliliters of cb.

This helps preserve the internal cell structures. Mix one milliliter of CB sucrose with four milliliters of 5%PFA. Add two milliliters of PFA in CB sucrose solution.

In two wells of a fresh six well plate. To fix the cells use forceps to pick up the CY two chip and place it in the six well plate containing two milliliters of PFA in CB sucrose solution. Incubate the cells for 10 minutes at room temperature, remove the PFA and wash the cells once with PBS then wash the cells with two milliliters of 100 millimolar ammonium chloride for 10 minutes.

To quench the residual PFA cross-linking activity, permeable the cells by adding two milliliters of PBS containing 0.1%tritton X 100 for three minutes, washed twice with PBS stain actin filaments with two milliliters of ZI conjugated fain at one to 2000 in PBS with 1.5%BSA incubate for one hour at room temperature. Then wash the cells once with two milliliters of PBS. Next, sustain the nuclei.

Add two milliliters of one milligram per milliliter hawked in PBS and incubate for three minutes at room temperature. Wash twice with two milliliters of PBS. Mount the side two chips on a standard slide using 25 to 30 microliters of mounting solutions such as al.

To acquire images in three wavelengths, we use Metamorph imaging software and an NIK icon eclipse T microscope. Equipped with the CCD Hema Matsu camera and an intens lite mercury fiber illuminator. First select 20 x magnification.

Next in the menu bar, open multidimensional acquisition. To set up the different parameters of the experiment. First indicate where to save your data.

Set the number of wavelengths to three, select the SI three wavelength and position the cover slip slide so that 12 micro patterns are centered in the camera field. The number of micro patterns visualized per field will vary depending on the camera and objective setups. Set the exposure time for each wavelength and auto focus.

For PS three, create a journal running multi-dimensional acquisition and save it as MDA jnl. Open the scan stage function to acquire 24 images each containing 12 micro patterns. Enter six columns and four rows with minus 400 micron x step size, and 300 micron y step size in set journal to execute, upload the journal MDA jnl press scan to start the automated acquisition of the 24 images in the three wavelengths.

In this video, L micro patterns are used to trigger the formation of a single major actin stress fiber that upholds the unattached part of the cell and simplify the visualization and measurement of BLEs statin's impact on cells compared to cells seated on plain fibronectin. For automated image processing and analysis, we outline two custom made macros written for the open source program image J from a stack of raw images, the first macro extracts individual cells and creates a reference cell. The second macro measures collapse of the cell membrane.

The first step is the segmentation of individual cells from raw images. The fluorescently labeled micro pattern images are used to delit regions centered on the micro patterns which are cropped on images for each wavelength and reassembled into stacks for each wavelength. To select micro patterns with single cells, the macro detects the number of nuclei per image and eliminates in all stacks those containing either more than one nucleus or no nuclei.

Finally, the image J plugin multi-stack reg is used to realign all the collected images and all the channels based on the micropattern position. This step generates a line stacks of individual images that can now be used for single cell analysis or to create a reference cell. In image J, the reference cell is constructed by making a projection of the filtered and aligned images from a stack.

Several projection methods can be applied, but usually a median projection is chosen to eliminate outliers of the image stack. The reference cell is a unique and useful tool for representation and image analysis only obtained with micropattern cells. To facilitate its examination, a lookup table or LUT is applied to convert the mono color image into a coded frequency map.

To characterize the BLEs statin effect, the rigidity and collapse of the cell were analyzed using a second image J macro. Briefly, threshold images of the L micro pattern are used to define a theoretical hypotenuse of the cell triangle shape. Next thresholding of the acton stain images is performed to define cell shape.

The difference in area between the theoretical hypotenuse and the actual concave cell membrane is then measured cells presenting an area underneath the theoretical hypotenuse are considered to be collapsed. The impact of five micromolar BLEs statin was characterized by comparing the number of collapsed cells in treated and control conditions. Typical images of L micropattern cells treated with BLEs statin or left untreated are shown cells were fixed and stained micro patterns.

Actin and nuclei are shown in red, green, and blue respectively. Note the impact of BLEs statin on cell shape compared to control conditions. Representative cells after incubation with BLEs statin or controls on plain fibronectin are shown here.

Cells were fixed and stained actin and nuclei are in green and blue respectively. Note the similarity between treated and controlled conditions. Typical images of reference cells obtained from cells treated with BLEs statin or left untreated are shown.

Note the potential of this approach for easy observation of the phenotype under study. Here is an example of an analysis of the effects of BLEs statin on cells using the macro hypotenuse. While 25%of control cells were collapsed, this increased threefold 75%in the five micromolar BLEs statin treated cells reflecting the weakened actinin contractile network.

After watching this, you should now have a good understanding how adhesive micro patterns solve several bottlenecks in high content analysis. First, the use of adhesive micro patterns made image capture processing analysis much more straightforward. Second, micro patterns allow detection of very subtle effects of drugs at very low doses because you're reducing cell to cell variability by normalizing their morphology and their behavior.

And finally, compared to the thousands of cells that are usually necessary to get a good result in cell analysis, you only need a few tens of cells to get a robust and significant result using our reference cell.