Happy Friday! I apologize for leaving you hanging about the picogreen DNA assay results, but your wait is over:)
Monday I shared with you some of the background behind the picogreen DNA assay and how to read the results. So here’s what happened Tuesday:
I needed to run four different sets of samples in this experiment. I would be using light to measure how much DNA was present in a sample. I had my frozen day 1 and day 7 scaffolds of mouse and human bone marrow cells, a control of bacteriophage DNA that would show the buffer I used free the DNA worked properly, a dilution series of the same type of bacteriophage DNA to create the line of best fit, and a set of controls of buffer without DNA to make sure there was not any biological contaminates. That’s a lot right? Dr. Simon shared with me that much of the work he does involves test controls to make sure all components of an experiment are working properly.
Before I go on, I wanted to share an image of how a bacteriophage works: When I first learned about these amazing viruses I could not believe how cool they were in their appearance and delivery. Basically it is a virus that injects its DNA into bacteria so that the bacteria does the work of replicating the virus and assembling new phages. It looks and acts like something straight out a science fiction movie right?! That’s one of the many things I love about science, learning about this cool stuff! But, back to the assay…
My first task was to make the serial dilution. I would be diluting a concentration by a factor of 3 at each step with a total of 10 steps. I started with known concentration of 1000 ng/mL or 0.000001 g/mL (nano= 10-9). I placed 150 uL DNA sample into well 1. The I took 50 uL from well 1 and placed it into well 2 which already had 100 uL of buffer in the well. I then mixed up the contents in the well then transferred 50 uL from well 2 to well 3. I continued this process 7 mores times. Each time I was careful to mix the contents before transferring the sample. Why? What would happen to the concentrations if I did not mix it? What might happen to the concentration if I had an air bubble in the tip of the pipette? Needless to say, this was a slow process.
Once the dilution series was complete, it was time to transfer 100uL of the DNA in lysis solution to this new plate
In order to make sure I knew where each sample would go I made a small mark on the first well on each row: I also marked the top of the plate with the location of each of my samples: and recorded this information in my lab book as well. Each time I transferred a sample I made sure the mix up the sample in the well and change the tip after moved the sample, Why? What was the point?
Next up I transferred the lysed bacteriophage DNA as one control and transferred a sample of plain buffer to 3 wells to run as another control. Now it was time to add the fluorescent tag o the DNA, the picogreen. This is a light-sensitive ingredient so I covered the vial with foil to reduce the amount of light entering the sample: and transferred 100 uL of picogreen in buffer to each of the wells: .
The it was time to run the samples in the fluorescence imaging machine: I loaded the plate onto the arm and then closed the drawer: The machine is attached to a computer with specific software used to capture the data as the plate is “read.” There is a mercury lightbulb similar to the one used to take the fluorescence images with the microscope: In addition there are numerous mirrors that beam the light through the sample. Picogreen has a specific excitation wavelength emits a certain wavelength (see 6-28-12-more-fluorescence-imaging-and-seeding-human-bone-marrow-stem-cells for more info on how this works) How bright this light is emitted is called the fluorescence intensity and depends on how much of the tag is present. This tag binds to the DNA in our sample so the more tag, the more DNA!
It took about a minute to run the plate and here’s what the results looked like: We needed to export these results to an Excel spreadsheet in order to analyze them and determine if the experiment worked as it was supposed to and how much DNA was present on each scaffold. That’s what I did today.
I took the raw data created the standard curve with the line of best fit: and found the formula for the line of best fit in the y-intercept form. (See I told you math was important!!!!!) I knew what the fluorescence intensity reading for each of my wells (y value), and using this formula, I could figure out what the concentration of DNA was in each well (x value). For example: the fluorescence intensity in well 1 was 11778. I plugged this y value into the formula: y = 1.8476x + 4.8972 where y is the value on the y axis, 1.8476 is the slope of the line (how far the line rises (change in y) as it runs (change is x)) and 4.8972 is where the line of best fit crosses the y-axis. If I lost you, check out Brainpop: Slope and Intercept (sligoms, brainpop). So when I substituted y in the formula with 1178= 1.8476x + 4.8972 I needed to solve for x. In order to do that I needed to subtract 4.8972 from both sides of the equation and I got 1173.1028 = 1.8476x Now I needed to divide both sides of the equation with 1.8476 to isolate x and I got 634. That meant I had 634 ng/mL of DNA in well 1.
But I still needed to know how much DNA was present without the volume since the cells were grown on the scaffold not in solution: . So I needed to figure out how to remove the mL units. Since I had worked with a sample of 200uL (0.2 uL) in each well, I could multiply the amount of DNA by 0.2 mL and that would cancel out the mL units. But I also needed to remember that my original lysed solution contained 200 uL and I only test 100 uL so I needed to multiply the amount of DNA by 2 to find the total amount of DNA on the scaffold. I used the formula DNA (ng) = 0.2 * d * 2 where d= the amount of DNA in solution. SO for well 1 I found DNA (ng) = 0.2 * 634 * 2 = 253 ng. I did this for all of my samples and created the following data tables: In analyzing this data I found a few things. One, the mouse cells proliferated (reproduced) more easily than human cells. There were about twice as many cells on day 7 as on day 1. And something went very wrong with one of my scaffolds. Can you tell which one? There was very little DNA on scaffold 3 in the day 7 from the hBMSC-8001 cells. Dr. Simon and I spent a long time discussing this piece of data and what to do with it. There are a number of statistical analysis tools scientists use to analyze their data and determine the chances of randomly getting their results. If one of the data points very different from the rest of the set, it is known as an outlier and we can do a statistical analysis to find the probability of getting that result again. If it is extremely unlikely to be reproducible ( how often I could repeat this test and get the same results) scientists can choose to remove the data point from their set. I was very hesitant about doing this. I wanted to make sure that I was being honest in my representation of the results as well as make sure I used this mistake as a learning experience ( try to think what went wrong so I don’t do it again). After much debating I agreed to create two representations of the data, one using scaffold 3 and one without it. Here’s the graphs of the results without the outlier. What do you notice about the average amount of DNA on day 7 of the hBMSC cells in the graph below compared to the amount in the graph above? You have to be careful to check the values of the y axis but you should see that the average amount of DNA in the graph above is close to 90 ng and the average amount of DNA in the graph below is close to 275 ng. One first glance, the graphs appear very similar, it is only with close study, you see the big difference in values. This is another way scientists may skew or misrepresent their results in what is known as showing a bias.
After taking a look at this data and the photographs taken in this experiment shown in this powerpoint 2012 Trial 1 hBMSC and mC3T3-E1, Dr. Simon and I agreed that I would perform one more experiment. I will repeat the same experiment I have done so far with one adjustment. We noticed that there were many more hBMSC cells that grew in the corners of the scaffolds than the mC3T3-E1 cells: To determine if this was true we are going to add a third set of scaffolds to the experiment where we will grow both human and mouse osteoblasts for 14 days. At that time I will fix and stain and try to compare the amount of growth in the corner of the scaffolds. If this is true, it may lead into another experiment for Dr. Simon and his team. That’s the way research works. For every experiment, you analyze the data, compare and discuss the results, try to determine a theory as to why you saw the results you did and in many times you end up with more questions than you started with and this leads to your next project.
So up next? Growing mC3T3-E1 cells in culture on Monday!