Success at last!

If you recall, on my last post I had said that none of my experiments had really worked. Well, as of Friday the 13th, all of that changed...

In my final week in Dr. Dolan's lab, which also coincided with finals week (which wasn't a big deal since I only had one take home final to work on), I decided that it would be worthwhile to repeat my knockdown experiment one last time, except with a few changes: 1) shorter time point of 6 hours vs. 24 hours, 2) added the siRNA to the nucleofection 96-well plate before adding the cells+NFS mixture, 3) allowed the cells to rest for 10 minutes at room temperature in the hood before adding pre-warmed media after the nucleofection, and 4) pooled four wells per sample instead of two.

On top of changes to the protocol, my technique was solid on the RNA isolation, RT-PCR, and real-time PCR. My RNA yield was the best it'd been all quarter, and my real-time results were not only beautiful but more importantly were believable because the standard curves (for the standards and samples) had a slope of about -3.4 which is much closer to the ideal -3.3 that's expected for totally pure cDNA.

So what was the final result? At the highest concentrations tested (40 and 60 pmol), I saw an almost 80% decrease in CYP1B1 RNA for both cell lines after 6 hours of transfection. These results are very exciting since it shows that transfection is possible with these LCLs (otherwise considered a very difficult system in which to transfect siRNA) and that the vector system we've chosen (96-well, the particular program and NFS) are workable, if not necessarily optimal, choices.

The next step is to carefully note any and all the changes to the protocol I made (some of which were really just reverting back to what Amaxa, the supplier of the transfection stuff, recommended) and then explain those to others in the lab. Woohoo!

Real-time PCR

So I've taken you from cells to RNA to DNA, now it's time to find out whether my siRNA knockdown experiment actually worked. What real-time PCR does is implied in the name; it provides information about the quantity of a particular gene (or genes) at the time the assay is performed. The PCR part comes from the fact that while during the reverse-transcription step we use random primers to amplify all the RNA in our sample, this time we want to only amplify particular genes so we use special primers for them. What makes these primers special? Let's say they give the reaction a special glow.

A company, Applied Biosystems, has developed a system (or perfected it), in which a special probe with a colored dye and a quencher (a compound that masks the dye) is affixed to one end of the primer (a small stretch of DNA that will anneal to one end of a gene of interest) such that when the DNA polymerase enzyme that adds DNA nucleotides together runs into it, the probe is cleaved and the dye and its quencher are separated such that when the sample is excited, light is emitted which is read by a detector. This cleavage, excitation, and light emission occurs during each cycle, so the more of the target DNA there is, the more light will be emitted and thus read by the machine. Make sense? Multiple dyes for multiple genes can be added to each sample. In our case, we want to know the levels for our gene of interest, CYP1B1, and a housekeeping "standard" gene (HKG), huB2M, that lets us know the quality of cDNA we're testing and to control for any extra high overall gene expression in our cells (which is a frequent issue with tumor cells, though our LCLs are from normal, healthy individuals).

Once the samples are read, fold-change is assessed by dividing the amount of CYP1B1 by the amount of the HKG, and then dividing our concentrations by the 0 pmol (control) sample. If things work as they should, we expect to see a dose-dependent decrease in CYP1B1 in cells receiving the siRNA such that the higher the concentration of siRNA added, the lower the resulting gene expression.

My experiments, unfortunately, did not work. I got crazy fold-change numbers such that for one cell line I first saw a knockdown of about 40% and then two weeks later saw a 1×10^7 (that's 10 million fold) increase in expression! Crazy. Naturally, I repeated the experiment and will get my results tomorrow. Later!