May 30, 2012
I'm researching carboxysomes in the Department of Energy Joint Genome Institute (JGI) in Walnut Creek. Carboxysomes are quasi-icosahedral structures made completely out of protein and contain ribulose-1,5-bisphosophate carboxylase oxygenase (RuBisCO), an enzyme in the Calvin cycle. These structures are present in autotrophic bacteria and participate in its carbon concentrating mechanism (CCM). If we could understand carboxysomes well enough, then there is potential to apply the design or concept into other mechanisms. I'll specifically be researching about alpha-carboxysome in the cyanobacteriumProchlorococcus marinus.
I spent most of the first day acquainting myself to the environment and brushing up on the necessary background information that I learned from my lower division biology classes and labs. The postdoc who supervises me gave me a project that investigates a protein which we do not yet know much of, except for the information we can derive from its sequence of amino acids. We have a hypothesis that it interacts with another protein in the carboxysome, so right now I’m testing it.
During my first week, the postdoc taught me many methods and protocols that lead up to how I’m supposed to prove such an interaction. I grew Escherichia coli cells, learned how to transform them, induce protein expression, perform a pull-down assay, run SDS-PAGE (short for sodium dodecyl sulfate-polyacrylamide gel electrophoresis), and now I’m learning how to perform and analyze Western blot.
One thing that I found very interesting is that experiments (in biology, at least) sometimes have extremely long wait hours (one hour, three hours, sometimes overnight). Some of my experiments take a whole afternoon, or even two days to complete. Due to these long waiting times, I actually have a lot of down time, which I use to make notes and study the background of what I’m doing. Also, because the experiments take so much time, I now understand why the undergraduate lab classes don’t have better, more interesting experiments - we are in student labs only once per week for four hours at a time, which makes it impossible to complete anything worthwhile, especially because some of that time is lost to figuring out what to do. On top of that, the selection of experiments is usually broad with no specific focus because labs have to be easily reproduced and restocked. With a tight budget and limited time in lab, there is no way that they could give students experience in something like running an SDS-PAGE.
Working full-time in a research lab is quite the experience already. The protocols that I have learned are so much more complex and interesting than any lab experience that I had before, and I am able to do these because I am working in the lab full-time.
June 19, 2012
This week is my fifth week at JGI, which means the program is almost half over for me. My post-doc has taught me the most interesting protocols. While working on my project, I have learned how to grow E. coli on plates and in LB media, transform bacteria, induce protein expression, purify protein, design primers, run a few machines (sonicator, nanodrop), make and run a SDS-PAGE, and perform a Western Blot. In addition, I have been helping my post-doc run some of her experiments, so I have performed DNA ligation, more transformation, DNA purification, and more.
My project is involved with finding out how one of the shell proteins (I’ll call it X here) of alpha-carboxysome interacts with another, more known shell protein (I’ll call it Y). Much of protein X is unknown. We currently do not know the 3D structure of X because attempts of crystallizing this protein has not succeeded yet; we could only guess from its sequence of amino acids, which has already been determined. We have good reason to suspect that X interacts with Y, because (1) X is part of “ghost” carboxysome (an empty shell after freeze-thaw treatment of purified carboxysomes); (2) tagged carboxysome can be purified by affinity column when fusing C-terminal SPA-tag onto X; (3) X and Y precipitated out after mixing (not due to an electrostatic interaction). These two experiments were done before I came to the lab.
Right now, I’m trying to detect the two proteins interacting; so I’ve combined the two proteins into one tube then tagged Y to see if there is X along with it. Once I get this far, then I should be able to analyze the protein. I first tried to grow two separate cultures of E. coli that express the two proteins separately, thinking that if I incubate the two proteins together for a long enough time (i.e. overnight), they would be able to form the theoretical complex that I want. I couldn’t get it to work.
One of the things I recently figured out was that the column that I was trying to purify the protein with did not work as well as I wanted. After purification, my SDS-PAGE results showed very faint bands, if not nothing. The resin form of the column was available to me, so I tried using that one. Even though the resin required many repeated wash steps (to filter out uninteresting junk), it gave me very nice bands. Now I’m testing whether my experiment will work with the resin.
If the resin does not yield the results I want, I have designed and am going to test a plasmid that would make E. coli express the two proteins together. Perhaps they have to be assembled together right after they are made.
July 16, 2012
I’m currently trying to make a plasmid that allows E. coli to coexpress X and Y together, a very long process. Basically, I amplify the two genes and the plasmid and then use DNA ligase to combine the genes and plasmid together.
First, I find a vector that had two multiple cloning sites (MCS) so that I can insert the two proteins. The plasmid vector pCOLADuet-1 has the necessary requirement. I then amplify these vectors by simply transferring them into E. coli, growing a massive quantity of cells and then harvesting the plasmids.
The genes for the proteins are located in other vectors, so I create primers to amplify the sequence for my proteins. A primer is a DNA sequence that includes three parts: the complementary sequence to the gene (so the primer can attach to the plasmid), a restriction site (so restriction enzymes can cut out my sequence later), and a tail (a bunch of random base pairs at the end to stabilize the primer and provide easier attachment of the restriction enzyme and necessary for restriction enzyme to function correctly). There are many things to keep in mind when making cloning primers. I make two primers, a forward and reverse primer and ensure that both primers have relatively similar melting temperatures. That way they can both attach during the annealing stage. The reverse primer has to be reverse-complemented. While picking restriction enzymes, I find ones that cut pCOLADuet-1 at the MCS and not my gene of interest.
After designing the primers, I send it out to get made. Upon receiving my primers, I use polymerase chain reaction (PCR) to amplify my target genes from the vectors. After PCR, I insert my genes into another vector, which is put into competent E. coli cells, and grow cultures to amplify the gene even more. I then pick a few colonies and isolate my genes for sequencing. When I get the sequence back, I analyze them and try to find my genes. If the sequences are right, I’ll sequentially digest one of the plasmids with two restriction enzymes. During this time, I also double digest pCOLADuet-1 to open it up, plus treat it with calf intestinal alkaline phosphatase (CIP) to preventing self-ligation in the next step (only the phosphate groups are removed, sticky ends are still there). After I purify those two, I used DNA ligase to stick them together. (If I don’t treat pCOLADuet-1 with CIP, then it could stick back to itself, and I could end up with my vector without insert) Then I transform more E. coli, so that I have one of the genes in pCOLADuet-1. Then, I repeat the digestion-ligation procedure to put the second gene into pCOLODuet-1.
With this vector, I will try to use His-tag to capture (his-tag cannot detect interaction, it can only detect the presence of his-CsoS1) the interaction between my two proteins (complex).
The most time-consuming steps are the ones where I have to grow cultures for plasmid prep. I usually have to put them into the incubator overnight. Then, I spend the following morning extracting the plasmid. The most unfortunate part of the process is that when attempting to make the vector, the steps are usually all-or-nothing. I can tell if I succeeded only at the last step, when I have sequenced my supposed vector with gene insert. This is because at each processing step, I lose a lot of desired material: I could start with a 150 mL culture of bacteria and at the end of several steps, have only several hundred nanograms of digested DNA. If this does not succeed, I lose a week’s work.
Such is the life of a researcher.