Co-Culturing: Take 2!

Today at Applied Biophysics I worked with last week's data I from my co-culturing experiment. Sadly, the data did not turn out as expected. The experiment started out looking promising, however, the cells in every well died out around 15 hours. Our hypothesis as to why this happened? Using an array just after it has been etched is tricky. Arrays that have been recently etched tend to allow more liquid to wick up the sides of the wells. This means that the amount of medium in each well is thinly distributed, and the cells are unable to grow and divide in an environment with insufficient nutrients. Also, this causes the liquid to evaporate faster, thus leaving the remaining medium highly concentrated with saline. Cells cannot survive in such a hypertonic environment- they will shrivel and die due to osmosis.

I treated each of the wells with cysteine, which usually helps with the problem of liquid wicking up the sides of wells, but my array might have been too fresh (last week, I took my array straight from the plasma etcher to the hood). So, this week I repeated our experiment. All conditions remained the same, except I used an array that had been etched a few weeks ago. That should have fixed our problem! Hopefully next week's data goes smoothly!

Friday was a busy day for me at Applied Biophysics. I learned how to plasma etch an array, interpreted the data from my 96 array experiment last week, and began my newest project: co-culturing. 

Plasma etching:

At Applied Biophysics, we use plasma etching to sterilize electrode arrays. On Friday, Dr. Keese put me in charge of the etching machine! First, I placed my array (an 8W10E+) into the chamber that will be pumped down to create a vacuum. We then pumped oxygen into this low-pressure chamber. After a few repetitions of pumping down the chamber and adding gas, we were able to start the plasma. Plasma is known as the fourth state of matter, because it is not a solid, liquid or gas. It is instead a mixture of atoms, ions, and free radicals. We sent an electrical charge through the chamber to create the plasma. The plasma ions become activated by this electrical field, and begin to vibrate and glow a white to purple color.  The vibrating ions essentially “scrub” the arrays in the chamber, leaving them sterile. Here is a picture of the plasma we created inside the chamber (and a super nerdy picture of me in front of the etcher):

Data from last week:

Last week Dr. Keese and I created a 96 array dilution series, with half MDCK-II cells and half BSC-I cells. Here is our graph using the ECIS software: 

Kind of crazy looking! Dr. Keese added a wound to the cells at 48 hours, which can be seen in the dip and recovery of the green and dark blue lines. The rainbow series of dots in the left hand column is the layout of the array wells. Columns 1-6 are MDCK-II cells, and 7-12 are BSC-I cells. Each column has same type and about the same concentration of cells, therefore we can group each column (take the average of all wells per column), making a much more manageable graph:

As you can see, the lines that are a shade of blue (representing the MDCK-II cells) take much more time to attach to the surface of the wells and form a confluent layer compared to the BSC-I cells. 

Co-culturing:

On Friday, I began my own study on the subject of co-culturing. Co-culturing cells means growing two different cell types together in one culture (or in my case, well). By studying cells in this way, we are able to see how cells interact with one another, in this case, how BSC-I cells will affect MDCK. Will one cell type overtake the other? Will the cells grow in isolated groups? Will they together form an evenly distributed confluent layer? To answer these questions, I created a dilution series, (explained in last week's post), diluting MDCK cells with BSC-1. I filled four tubes with 1ml of MDCK cells, and then added 1ml of BSC-I to the second tube. I then mixed the contents of that tube, took 1ml of the mixture, and dispensed it into the next tube. The overall array had 2 wells of pure MDCK cells, 2 wells of a 50/50 mixture, 2 of .25/.75, and 2 of .125/.875. I am looking forward to seeing these results next week! 

96W1E

It's been so long since I've been at my internship! It was a good thing I didn't forget what I had learned over winter break, because we started a much more complex experiment today! Today I got to work with a 96 well electrode known as a 96W1E. This wellpad works the same way an 8 well electrode would, but looks quite a bit different! Here is an image of my 96W1E, all filled up and ready to be placed into the ECIS station:

In today's experiment, we used two cell types: MDCK and BSC-1 cells in a dilution series. Dr. Keese demonstrated how to work with a 96 electrode, and proper pipetting techniques using an 8 tip pipette. He first pipetted 150 µl of cysteine into the first six columns of the wells. Cysteine is an amino acid that reacts with the gold of the electrodes, allowing proteins cells to attach and spread easily. While we waited for the cystine to react with the gold, a process that takes at least 5 minutes, he began to treat the MDCK cells with EDTA and trypsin to get the confluent layer of cells to be free-floating. Setting those cells aside, we began to prepare our dilution series. The main idea behind a dilution series is to examine how cells grow in environments of varying sparsity. We created cell mixtures in six test tubes with the following cell compositions: 1x, 1/2x, 1/4x, 1/8x, 1/16x and medium (0x). We got these mixtures by filling each test tube with 3mL of medium (except the 1x tube). The 1x tube was 6mL of a suspended MDCK cell mixture. We took 3mL from this tube and dispensed it into the second tube- 1/2x. We mixed the cells and the medium using a titration technique (aspirating and dispensing the mixture over and over again to make sure the cells are uniformly distributed). This step was repeated for each of the subsequent tubes, therefore making each tube half as concentrated as the preceding tube. Dr. Keese then pipetted the mixtures of varying concentration into 6 of the 12 columns. I repeated these steps on my own with the last 6 columns, using BSC-1 cells. Here is an image that shows the layout of the 96 wells:

Finally, we were ready to place our prepared 96W1E into its station! This station looked different than those that I have used with 8 well electrodes. Instead of teeth that attached to one end of the electrode, this station uses what Dr. Keese called, a "bed of nails" to transfer signals from the electrodes to the ECIS machine. Each of the gold "nails" would connect to a single electrode, relaying the electrical signal to the more advanced ECIS machine, the Zθ. This machine is different than the Z (which I am used to working with) because it uses a "complex impedance spectrum", meaning it can read cell behavior in Z, R, and C (impedance, resistance and capacitance). 

Here is a sneak peak of the first 15 minutes of my graph. Looking forward to seeing my results next week!

On Friday, I met with my mentor for the last time in 2013. We went over the data collected from my wound healing assay experiment last week. For the most part, my data was as expected. However, there were two wells that did not follow the general trend of my results. Possible error in these two wells could have been that I did not have the same amount of cells in these wells as the other six. Therefore, the graphs of these were significantly lower than the average. However, the cells in these wells caught up to the others after about 15 hours.

I also passaged my BSC-1 cells and MDCK cells. With practice, I am learning how to better passage these cells. The MDCK cells are the hardest to passage because they have tight junctions and are therefore difficult to separate from one another and the bottom of the flask. On Friday, I only used about one drop of cell culture in each of my new flasks because I will not be working with the cells for a few weeks. We do not want the cells to multiply and grow too quickly because they will use up all of the nutrients in the medium. Dr. Keese will be babysitting my cells while I am away, making sure they have enough medium and are healthy and pathogen-free!

Dr. Keese taught me how to manipulate the ECIS software to better understand my data. My favorite function saws the 3D graph option! He showed me how to put my graphs into three dimensions, and how to interpret these results. The 3D graphs plotted time (hours), frequency (hz) and impedance (Z). I would have loved to attach the graph of my results, but I had a bit of trouble getting the data onto my computer. I will try to get these images uploaded onto my blog for my next post. Sadly, I will have to wait until January 10th until my next visit to Applied Biophysics!

On Friday, I passaged my BSC-1 cells to make two new flasks of cells. I will be using these cells in a co-culture experient in which I will combine BSC-1 and MDCK cells together and examine their behavior with an ECIS run. I also passaged MDCK cells today. These cells behave very differently than BSC-1 cells becasue they form tight junctions and are tough to suspend. In order to get these cells off of the bottom of the flask, I needed to add three rinses of EDTA, which is a solution that chelates the cells (takes the magnesium and calcium ions off of the cells). I then added trypsin, which also helps suspend the cells. I then took my MDCK cells, added medium, and put 400 µL of this cell-medium mixture into seven wells of an 8W1E electrode. The eigth well is just filled with medium, no cells.
I then plugged my electrode array into the ECIS port, and started to run my experiment! In this experiment, I will be performing a wound healing assay. In a wound healing assay, the electrode sends out a current of electricity that will kill the cells in the vicinity of the electrode. In this experiment, I sent a 1400 uA current through four wells for 20 seconds.This process will cause the cell membranes to become porous, and medium will leak into the cell, causing it to die. Why would I want to kill of perfectly healthy cells? Because then we can see how the other cells will move in to heal the wound. We had to wait for the cells to settle to the bottom of the electrode and grow into a confluent layer before we wounded the cells. I set the wound time for 9 hours into the experiment, and I will see what my results look like this Friday!

I was able to connect some of my knowledge about genetic modification that I had learned in AP Biology to this wounding process. ECIS allows one to either wound their cells, completely killing the cells around the electrode, or electroporate them. Electroporating cells introduces a high frequency current that damages cell wall (not killing them), causing them to become more porous. The increased porosity of the cells allows substances (such as large or non polar molecules) that could not usually pass through the selective cell membrane to enter. Electroporation allows a scientist to introduce genetic information into the cells, often by a virus, changing the cells' DNA.

On Friday I observed Dr. Keese while he performed an experiment using an array with filters. By growing cells on a porated filter, we area able to better simulate an in vivo environment, and to measure the barrier function between cell junctions. There are many different types of cell junctions,  each with a different strength. We used MDCK II (canine kidney cells) in our experiment which form tight junctions. These cells are very hard to separate, which makes it difficult for us to passage them. We added 3 rinses of a solution called EDTA which takes away the magnesium and calcium ions of the cells. Then, we added trypsin, which is an enzyme that essentially eats away at the cell walls, weakening their junctions. Once the cells became rounded and were no longer attached to one another or the surface they were grown on, we were ready to put them into our array!

The electrode array we used today has a very deep well which allows for our filters and a suspended electrode to sit in the well. Here is what our electrode array looks like:

Here is a diagram of a single well in our electrode:

As you can see, medium sits above and below the cells, therefore creating an environment that would be similar to that of the body. 

So what, exactly are we measuring in this type of experiment? Using the two electrodes, we can pass a current through the layer of cells growing on top of the filter, therefore seeing how tight the cells junctions are. This measurement is called barrier function, which is much higher for MDCK II cells than the BSC-1 cells I have been working with. The higher measure of barrier function, the tighter cell junctions are. For example, endothelial cells that comprise the blood-brain barrier would have an extremely high barrier function because they must have a low permeability in order to keep out microorganisms and other large or hydrophilic molecules. 

I also worked on passaging my BSC-1 cells. Here some candid shots of me working with my cells:
(Photo credits to my awesome mentor, Dr. Keese!)

Holding a flask of BSC-1 cells

Working in the hood

I learned a lot about the physics behind Applied Biophysics' ECIS instrument this week. (As more of a biology person, I need a lot of visuals to understand the physics side of things, so get ready for a lot of pictures and diagrams!) Dr. Keese and Dr. Renken gave me an in-depth presentation on how cells located in each well of the electrode array are subjected to varying frequencies to measure their behavior. There are two types of current- AC (alternating current) and DC (direct current). The ECIS instrument sends an AC current through the electrode, allowing us to get a constant reading. Here is a diagram I made to explain the main idea of using AC current:

For our purposes, the capacitor represents the cells that are sitting on the electrode. They act just like a capacitor because cells are mostly saltwater, and do not let current flow through.

In my experiment from last week, I used the most basic ECIS instrument, the Z. It sent out super-low frequencies of 1000 hz, all the way up to 96000 hz. At low frequencies, we were able to see how tightly joined our cells were. This is because current always wants to follow the path of least resistance, and will therefore travel around and underneath the cells. At higher frequencies, the current will pass right through the cell, and therefore allows us to measure how many cells are in the vicinity of the electrode. It is important to note that cells don't sit flat on a surface. Instead, they have little anchors that connect the cell to the surface it is growing and spreading out on. Here is a diagram that illustrates this concept:

With all this new information, I was able to understand the basics of my own ECIS graphs that were produced from the experiment I ran last week. My cells responded just as they were supposed to, with the exception of my control well which was filled with only medium. Apparently, I somehow got a few cells into this well and began to grow. (Opps!) 

Here is a graph of my data at a medium frequency of 16000 Hz. This frequency gives you a good overview of how my cells were growing, but it doesn't measure specific things like cell density or cell junctions.