The Last Post

Sadly, this will be my final blog post about my internship at Applied Biophysics. This year, Fridays have been my favorite day of the week - not in anticipation of the weekend - but because I got to spend every Friday morning learning about the nature of cell behavior and diving into the world of physics with some brilliant minds! (I mean, how many people can say that they were able to work in the same office as a Nobel Prize winner in physics?!) I have had the privilege to work with some extremely kind and intelligent men who have made this internship such and extraordinary experience. I could not have asked for a better mentor - Dr. Keese. His expertise and generosity has informed how I view research, the field of biophysics, and pursuing a career in the sciences. I have truly learned so much through this internship over the course of my senior year- from the physics behind electrode measurements, to culturing cells in a hood, to putting creative ideas into reality. I am so glad that I will be able to carry my newfound knowledge into college, and pursing research opportunities in the sciences.

Time Lapse Attempt #1

Today, Dr. Keese, Dr. Renken and I set up the equipment required for our time lapse imaging of experiment. In this experiment we will be examining how BSC-1 and MDCK cells interact with one another when co-cultred. As we have seen in the past, MDCK cells have the tendency to form like-cell islands amongst BSC-1 cells. We would like to see how these islands are formed over time. The basic mechanisms required for our time-lapse experiment are:

• Phase contrast microscope
• special ECIS loading dock with windows, water channel system, and gas inlet
• BSC-1 and MDCK cells
• CO2 pump and bubbler
• Nixon camera with microscope lens attachment
• Water pump 

Here is an image of the time lapse setup:

Most of the equipment used in this setup is to keep the cells healthy & happy. Cells like to stay in a warm, moist, environment- hence the water pump that sends 37 degree Celsius water through the channels of the dock they are encased in. The CO2 pump keeps the cells at ~ 5% CO2. The rest of the setup is pretty self explanatory: the camera is set up to take 1,000 frames over a 72 hour period, and the microscope is for magnifying the cells using phase contrast. Phase contrast is essentially a system of lenses that are arranged to allow a specific amount of light in, thus highlighting cells which are otherwise transparent bodies. Here is a picture that demonstrates how phase contrast highlights a cell:

Cells magnified with traditional bright field microscope (left) and phase contrast (right).

In Friday's experiment, we are just looking a BSC-1 cells using time lapse. This test run will help us work out any kinks such as adjusting the magnification, CO2 levels, etc. in order to be prepared for this Friday when we place a co-cultured sample of BSC-1 cells and MDCK cells into this contraption! I'm not sure if we are going to add ECIS measurements to this Friday's run, although that is possible for us to do. I'm hoping our experiment went well and I'll be able to post a time-lapse movie on my next blog post!

Hu Tu-80 Dilution Series and L-cell Islands

Today at Applied Biophysics I ran an experiment using a new type of cell known as Hu Tu-80. Despite its catchy sounding name, Hu Tu-80 cells are actually a bit sinister- they are human intestinal cancer cells. So far I have worked only with animal cells- from monkeys, dogs, and mice. This new human cell line looked similar to the animal cells I have worked with. Hu Tu-80 cells share some of the same properties as MDCK cells such as their cobblestone-shaped growth, but had more defined cell boundaries similar to that of BSC-1 cells.

This experiment is designed to examine the effects of BSA and Gelatin on varying concentrations of Hu Tu-80 cells. In this experiment, I used a 96W1E+ (which is an electrode array with 96 wells, and one electrode in each well). In the odd columns, I coated the wells with BSA (Bovine Serum Albumin), and the even columns were coated with Gelatin. BSA is used as a protective agent for cells from oxidative damage, and also serves to stabilize components of media such as fatty acids and pyridoxal. Gelatin is made from the collagen of animal parts and is primarily used to improve cell attachment. I let 200 micro liters of these two solutions sit in the wells for about 5 minutes, suctioned out the wells, and then added varying concentrations of Hu Tu-80 cell mixture to each of the wells. The Hu Tu-80 cell solution was diluted with a corresponding amount of regular medium. Here is the layout of my 96 array (the last two columns are empty):

Green columns are BSA coated, yellow columns are Gelatin coated.

L-cell and BSC-1 dilution series results:

Two weeks ago, I performed a dilution series experiment with L-cells and BSC-1 cells. I was testing these two cells types to see if they formed islands like the BSC-1 and MDCK cells did. The result: YES! As the concentration of L-cells increased, L-cell islands became more prevalent. Dr. Keese stained a row of my array to show how the number of islands increased. In the following pictures, you can see these stained wells:    ( *L-cells are the dark, circular cells* )

Pure BSC-1

Light inoculation - .0625 concentration of L cells

L-cells islands become more prevalent at .25 concentration

L-cell islands become as populous as BSC-1 cells

Pure L-cells

Next week, I will be working with Dr. Keese to begin setting up the time lapse equipment for our next experiment! We will be using the time lapse video to watch the formation of these islands in real time. Can't wait!!

L-Cell dilution series and hemocytometer usage

After a two-week vacation, it was refreshing to get back to my internship at Applied Biophysics. It was great to get back to the lab, creating a new co-culturing experiment and learning learning how to count cells using a Hemocytometer.

Before I delve into the details of my internship on Friday, I'd like to write a quick happy birthday to the amazing Dr. Ivar Giaever who celebrated his 85th birthday on Saturday! An esteemed physicist, Dr. Ivar Giaever is a professor emeritus at RPI, a professor-at-large at the University of Oslo, and president of Applied Biophysics. And did I mention he won the 1973 Nobel Prize in Physics for his studies in tunneling phenomena in superconductors? I love hearing stories about Dr. Giaever, from his stories of his travels to across the world, to his take on global climate change. Happy birthday!

Back to the science! I am now working on another dilution series that has the same format as my past 96w20idf experiments with MDCK and BSC-1 cells, but instead of MDCK, I am using L-cells (mouse endothelial cells). The object of this experiment is to compare the behaviors of our previous cell combinations of MDCK and BSC-1 (epithelial/epithelial) to a mixture of BSC-1 and L-Cells (epithelial/fibroblast). Here is the layout of this 96w20idf dilution series:

Cell Counting using Hemocytometer:

Once I finished setting up my dilution series experiment, I learned how to use a hemocytometer to manually count cells. The idea behind this device is to take a small, manageable amount sample of cell mixture, and use it to make an estimate of how many cells are present in a larger volume of that same mixture. Hemocytometers are commonly used to count blood cells (hence the name hemo (meaning blood) cyto (meaning cell) and meter (meaning count)). Here is what a hemocytometer looks like:

To use the hemocytometer, one pipettes a small drop of cells suspended in medium into the tiny v-shaped notch as seen in the image above. There is a cover slide that sits on top of the raised part of the hemocytometer, which the cell solution slips under and fills. Capillary action draws the liquid into the space between the cover slip and the glass surface, creating a uniform layer of liquid .1mm in depth. There is a microscopic, laser-etched grid on both sides of the hemocytometer, which looks like this:

Using the applicable objective, I magnified one of the 4x4 squares, and counted the number of cells within this square. With cells that are on the edge of the square, only count those that touch the top and left sides of the square (it's just common practice). To calculate the cell concentration per ml, use this simple formula:

I counted up 5 of the 4x4 squares for both a sample of L-cells and a sample of BSC-1 cells which were left over from my dilution series experiment. I counted up 32 L cells and 34 BSC-1 cells, two values that are relatively similar. This similarity is good because that means the cell solutions used in my dilution series will be relatively reliable. 

We did not perform this in my experiment, but one can use a hemocytometer to calculate the cell viability count by staining cells with Trypan Blue. Trypan Blue is a "vital stain" that colors dead cells  blue due to the incorporation of color into the proteins of dead cells, and leaves live cells colorless. 

March 14, 2014

I will be going on spring break for the next two weeks, as will Dr. Keese, therefore we did not start any new experiments on Friday. I prepared cells for freezing, cultured spare inoculations of BSC-1 cells, and analyzed my data from last week.

The L cells (mouse fibroblast cells) that Dr. Keese and I thawed last week are alive and healthy, and actually exhibiting some unexpected properties. Applied Biophysics acquired these cells from a lab in Korea because they were said to have unusual behavior when examined with ECIS. These cells supposedly caused the graph of impedance to ocilate. When I looked at these cells under the microscope, they looked very different than the BSC-1 and MDCK cells I am accustomed to working with. These L cells looked like they had grown in layers on top of each other. This behavior is abnormal, as cells are supposed to cease dividing and growing upon contact with other cells, forming a single confluent layer. This behavior is known as contact inhibition, and is characteristic of normal stoma tic cells grown in culture. Dr. Keese and I will look into the mystery of the L cells when we get back to break. I will be performing a dilution series experiment on these cells, coculturing them with BSC-1 cells. It will be interesting to see if the L cells and BSC-1 cells will form tiny "islands" like the BSC-1 and MDCK cells did.

Here are fluorescent stained images of the three cell types I have worked with in my internship thus far:
**(note the difference in cell junctions- these junctions are one of the main things we are looking at when analyzing these cells with ECIS.)

L-Cells (Mouse fibroblast cells)
Lay down collagen in the cell, giving the cell structure.
(Fluorescence Digital Image Gallery, FSU)

MDCK cells (Madin-Darby Canine Kidney cells)
Epithelial cells form tight junctions in confluent layer.
(MDCK Epithelial line, olympusmicro.com)

BSC-1 cells (African Green Monkey cells)
BSC-1 cells in the process of dividing
(Cell Division Mistakes, Laboratory News Network)

As for freezing the cells, I followed the same procedure outlined in last week's blog post. I suspended the cells, and placed them into a mixture of DMEM medium, 20% fetal bovine serum, and 10% DMSO (dimethyl sulfoxide). The DMSO will lower the freezing point of the medium, allowing the cells to freeze at a slower rate thereby reducing the chance of ice crystals forming which can rupture cell membranes. The cells are now ready to be put into small capsules to be cooled at 1°C per minute. Then, they are placed into a container of liquid nitrogen, which keeps them at a balmy -346°F and -320.44°F. I also resuspended a few flasks with a very sparse amount of cells, that way, they will reach confluence at a slower rate, and they will have enough medium to last them until my next visit in April.

My data from last week was in keeping with the results from our original co-culturing experiment. Because our discovery of these cell "islands" was replicated, we can now move on and study them under time-lapse, and test other cell types to see if they behave in a similar manner. Looking forward to putting together the time-lapse equipment for observation of these cell islands!

Figure 1: Micheal W. Davidson, FSU. (Oct 14, 2004).  Embryonic Swiss Mouse Fibroblast Cells (3T3). [Fluorescence digital image]. http://micro.magnet.fsu.edu/primer/techniques/fluorescence/gallery/cells/3t3/3t3cellslarge1.html

Figure 2: olympusmicro.com. (n.d.). Madin-Darby Canine Kidney Epithelial Cells (MDCK Line). [Fluorescence digital image]. http://www.olympusmicro.com/primer/techniques/fluorescence/gallery/cells/mdck/mdcksb10.html

Figure 3: Laboratory News Network. (Feb 2, 2011). Understanding how cell division mistakes lead to human disorders. [Fluorescence digital imaging]. http://labnewsnetwork.blogspot.com/2011/02/understanding-how-cell-division.html

Another busy day for me at Applied Biophysics! On Friday, I learned how to freeze and thaw cells, how to take images of stained cells under the microscope, and re-created the 96w20idf experiment that I had previously done with the help of my mentor on February the 7th. 


Thawing & Freezing Cells:
It is common lab practice to freeze cells that are not being used in order to preserve them for long periods of time. This coming Friday, I will be using a new kind of cell: L cells! These cells are mouse fibroblast cells, which are responsible for laying down collagen- giving structure to living tissues. These cells have been stored in a container of liquid nitrogen for years at quite the chilly temperature, -346°F and -320.44°F. Before cells are put into this frigid environment, complete medium is added along with a cryoprotective agent such as dimethylsulfoxide (DMSO). This agent is extremely permeable, and slips right past the phospholipid bilayer of the cell membrane, and into the cell. DMSO will lower the freezing point of the medium, allowing the cells to freeze at a slower rate thereby reducing the chance of ice crystals forming which can rupture cell membranes. To prevent cells from dying, it is necessary to freeze the cells at a controlled rate decreasing the temperature by 1°C per minute. 

Thawing cells is much easier than the freezing process. To thaw our L cells,  we located the vial within the liquid nitrogen chamber, and placed them into a water bath at 37°C. Once thawed, we pipetted the L cells into a T-25 flask and added fresh medium. Then, we placed the cells in the incubator and waited for them to attach to the surface of the flask and begin to form a confluent layer. At the end of my internship on Friday, Dr. Keese and I took a look at these cells, and saw that after about 3 hours, they still had not begun to attach to the surface of the flask. These cells might have been frozen improperly, causing them to die.  Hopefully this is not the case, but I will post an update on the status of my L cells on my next blog post.

Results of Cell-Staining Experiment:
Last week, I created a series of pure and 50/50 MDCK and BSC-1 cell dilutions. Dr. Keese stained these cells over a period of time to demonstrate the growth and behavior of my co-cultured cells. We found that the MDCK cells and BSC-1 cells grew into islands of like-cell types. Here is an example of an island of MDCK cells surrounded by BSC-1 cells: 

I took this picture with a camera that can be attached to the eye piece of a microscope. (Getting the cells in focus is much harder than it seems due to the fact that the microscope itself has to be in focus as does the zoom function on the camera.)

**Side note: After spring break, Dr. Keese said that we will put together a time lapse video of the growth of our co-cultured cells!!)

96w20idf Experiment:
In any experiment, it is necessary to be able to reproduce your findings in order to draw accurate conclusions about your data. I re-did the very complicated 96w20idf experiment from two weeks ago so that we can be sure our data is reliable. This week, I reproduced this experiment 100% on my own, and realized how far I've come throughout my internship experience at Applied Biophysics. I remember on day one when I was lost in the jargon of brilliant scientists talking about frequency scans, dilution series, and plasma etching, and had not the slightest idea of what Electric Cell-substrate Impedance Sensing was. On completing this co-culturing dilution series experiment, I realized how much I've learned, and how this internship has allowed me to apply what I've learned about biology to the real world. 

Co-Culture results

I'm very excited to share my results from my co-culturing experiment! Dr. Keese and I came across some very interesting results, especially in the to 50% MDCK/50%BSC-I mixtures. This is the first time anyone has performed a co-culturing experiment using ECIS, so Dr. Keese was just as excited as I was to see the outcomes of this experiment. We expected to see these two different types of cells to integrate with one another, attaching to the surface of the electrode in an evenly mixed layer. However, we found that these cells joined together in like-groups, looking like islands under the microscope. With this conclusion, it would not be logical to perform a wound healing assay on our next co-culturing experiment as we had originally planned. A wound-healing assay would kill off any cells on the electrode, and therefore would kill one type of cell- whichever island had attached to the electrode. 

With this new insight into how different cell types interact with one another, we need to redesign the next steps in our ongoing co-culturing experiment. 

First, it is critical to take a deeper look into how these islands are formed, and why. To do this, I filled 16 wells of a cell plate with varying concentrations of MDCK and BSC-I cells. I filled two wells with pure MDCK, two with pure BSC-I, and 12 with a 50/50 mixture of both cell types (1ml of solution per well). This cell plate does not contain any electrodes, it is purely for the purposes of examining the growth of cells under the microscope. Dr. Keese will be staining the cells incrementally to see how they attach to the surface, and at what rate they will form like-groups. I will be working with different cell types in the future to see if all cells form like-groups when grown together in the same environment.  I'm looking forward to posting pictures of these stained cells in next Friday's blog post!

Finally, Dr. Keese and I analyzed our results in ECIS. Although the number of MDCK cells seemed to have been was lower than the BSC1 cells (a possible source of error), we still had nicely mixed cell population data. Our results are best demonstrated in a frequency scan. Here is the frequency scan displaying the results from every well: 

And here is the frequency scan that I have manipulated to display the average of all pure MDCK wells, pure BSC-I, .5/.5 mixture, and cell-free solution (medium):

As you can see from these two frequency scans, MDCK cells show up better at lower frequencies, and then switch places with BSC-I cells at around 20 hours. At about this time, the .5/.5 mixture is seen as having a higher resistance than both the MDCK and BSC-I cells. What frequency tells us about our data: At higher frequencies, impedance is more affected by extent of cell-coverage, whereas at lower frequencies, we  are better able to see the changes in the spaces between or underneath the cells.  

A few weeks ago, Dr. Keese gave me a presentation on how ECIS works, and spoke about cell migration. I remember that in one slide of his presentation, there was a video of cells, and how they grew in response to a wound healing assay. I asked him if using this type of time-lapse video would be beneficial to seeing how our two types of cells are interacting with one other. We will have to wait and see what his answer is next week! Here is an example of the time-lapse documentation of cell migration I am referencing: