Diatom Colony Screen

A new lab video!



In this video I discuss designing a colony screen protocol for my transformed lines of diatoms. In one of my recent posts, I posted this picture diagram which oversimplified the process:

While the PCR steps are the same (starting in the top left panel with the diatoms being combined with the PCR mix), I discovered that diatoms grown in liquid culture (i.e. their natural state) needed extra care in preparing them for a colony screen.

Let's step back a bit and quickly talk about colony screens (in case you didn't read my above Tumblr post). A colony screen is a modified PCR reaction*, which is a cyclical amplification of a short sequence of DNA, exponentially copying the targeted DNA strand. (*Even though PCR stands for polymerase chain reaction, PCR is usually said out loud in conversation as PCR reaction.) Often the DNA source for a PCR reaction is a purified, such as plasmid DNA purified by a process to separate plasmid DNA from proteins and genomic DNA.

But colony screens use a colony of cells to supply the DNA for the PCR reaction. By initially lysing the cell by cooking it at a high temperature for a period of time, the cell's DNA is released into the PCR reaction mix and then amplified in the reaction. Colony screens are used to screen cells for the presence of DNA--that is if the targeted sequence of DNA is present, it will be amplified by the reaction. If a reaction gives a positive band on a gel (bottom right of the colony screen chart), then we know that the DNA from the cells in that particular reaction also have the DNA of interest.

This colony screen method should in theory work for other cells, like diatoms. My initial trial successfully amplified GFP from diatom colonies growing on agar plates, but it didn't work for diatom cultures growing in liquid media, a distinction I didn't make at the time.

I soon realized after a failed trial of screening  diatoms solely from liquid cultures that the residual salt water from their growth media was throwing off the delicate chemistry of the PCR reaction. This graphic paired with the above video discusses just this.

This DNA has a sequence and I have every intention to find out what it is! (and other stories)

In my quest to amplify a region of the nitrite reductase (NiR) gene that I can then ligate into a plasmid for in vivo testing, I needed to sequence my portion of DNA for three main reasons:

1. I needed to make sure that the restriction sites I added with gene specific primers were added to either end of the sequence. These restriction sites will allow me to cut out my amplified piece of DNA that is cloned into a cloning vector.
2. I need to verify that the sequence of DNA I've been cloning is indeed from the NiR gene, especially since the DNA sequence has been marginally longer on my agarose gels than anticipated.
3. Previous data acquired from a database told me I should be able to get a restriction enzyme to cut somewhere in the middle of the DNA sequence I had cloned, but the restriction enzyme wouldn't cut in the middle (but it did cut on either side of the DNA sequence that I cloned into a vector, as the vector had that restriction site on either side of the multiple cloning site). Essentially, that restriction site didn't exist in my DNA sequence like it should have.

So, couple weeks ago, I performed a PCR cycle sequencing reaction to start the process of sequencing. The purpose of this sequencing reaction is to amplify a region of DNA with DNA bases that have special tags. These tags are read by a sequencing machine yielding a series of base signals like below:

The color coded lines refer to a single base category (either A, T, G or C) and denote the signal strength at that spot in the sequence. Based on the predominant signal at a spot in the sequence, the software makes a judgment call as to which base is present at that location. Often it is quite clear to the machine and software that a particular letter of DNA is certainly an A at a specific location because there is a single strong signal peak. Other times there may be several signal peaks at a single location because of a lower quality sequencing reaction (e.g. there is too much or too little DNA to be read properly), which make it harder to determine which DNA letter actually exists in the DNA sequence.

In order for the DNA to be read by the sequencing machine, it needs to be single stranded. This means we make up separate PCR reactions with only one primer, which means that the one primer in a PCR reaction will attach to the DNA and copy the DNA until it falls off. To make it easier on myself, I followed a standard protocol that sequences a section of DNA that has been cloned into a plasmid vector. This way there is a buffer of vector plasmid sequence on either side of my DNA sequence of interest, where the primers that pair with the vector plasmid would bind.

I performed my PCR reaction and prepared it for the sequencing machine and was about to load my DNA onto the DNA sequencing plate (which goes into the machine), when I was notified that the machine was down and I couldn't sequence that day or possibly for the rest of the week. It took me a week or more for my samples to get read, but they finally did get read and my sequences were good enough to give me some answers.

I use a special computer program that will align the DNA sequences to one another. Running multiple samples of the same sequence will allow the program to compare answers and yield a single, best result sequence. Interestingly enough, while my four samples all matched each other, there was a small region (about 20 bases) that did not match up with the database sequence for my gene. This may have been why my restriction site wouldn't cut--because it wasn't there. One of my labmates is going to sequence the same gene soon, so I will be able to compare my results with hers and determine whether the predicted sequence in the database needs to be updated.

I can then take the sequence from the first program (the results) and pop it into another program, to quickly find all of the restriction sites that are present.

Here I can see that my BamHI and NotI restriction sites are indeed at either end of my sequence, with the EcoRI sites just outside of the Bam and Not sites. Because the EcoRI sites are in the vector which I cloned my sequence with, I know that the Bam and Not sites are on either side of my sequence like they should be.

At present I have ligated this section of DNA (the piece of the NiR gene) with the other portion of the NiR plasmid, and will transform some bacteria today. If the plasmid worked and was successfully cloned into my cells, they bacteria will be able to grow on the plate laced with ampicillin. Here's hoping!

It's Transformation Time.

Last time I checked in, my PCR reactions that were supposed to add restriction sites to either end of the nitrite reductase (NiR) terminator region appeared to work and work well. When I ran the PCR reactions on a gel, the amplified DNA bands were really strong and were the correct length.

This gave me the go ahead to continue the path in isolating & altering the NiR terminator in order to yield a sequence of DNA to fit the final NiR plasmid for my project's experiments.

Now I need to insert the NiR terminator into a vector plasmid, transform it into bacteria, and digest the NiR terminator back out of the plasmid to double check that the terminator I got on the gel from my PCR reaction (above) is the correct piece of DNA before I insert it into the NiR plasmid to complete the final NiR plasmid.

We use bacteria to amplify pieces of DNA because of their quick generation times. If you insert a plasmid into bacteria, they will duplicate the plasmid as if it were their own DNA as they grow and divide. A plasmid is a ring of DNA, which is essential for this to work, because bacteria will cut up and destroy any loose pieces of linear DNA. In order to get our NiR terminator to be duplicated by the bacteria, we insert it into a vector plasmid first. The vector plasmid is designed to accept small pieces of DNA from PCR reactions, lock in that piece of DNA within the plasmid. This plasmid can then be transformed into bacteria.

Bacterial transformation is really easy. Once the vector plasmid complete with our PCR DNA is ready, we add the plasmids to specially-altered E. coli cells, incubate the cells on ice for a short time (to lull them into a false sense of security), and then transfer them to a hot water bath (42°C) for thirty seconds. Thirty seconds is all we need for the bacteria cells to panic and scream "WHAT IS HAPPENING TO ME?" This prompts the bacteria, because they are stressed, to take up any DNA in their environment. Well good thing the only DNA in their environment is the plasmid we gave them! Through the heat shock, a large amount of bacteria should have taken up our plasmid. We then grow the bacteria over night while they recuperate, divide exponentially, and make copies of our plasmid. This process is summarized by the cartoon below:

Step two: Place science in a large bowl, add water, mix, and enjoy!

I just posted my second topic video on YouTube, where I bring a camera into lab to show everyone a sample of what I do in lab.

Last week I ran a gel to check on my DNA sequences yielded from gel purifying double digests. That is, I used enzymes to cut my plasmids, circular rings of DNA, which I then separated on an electrophoretic gel. I cut out my desired pieces of DNA, purified them, and then ran them on a second gel, which I show in the video.