Oh wow, this is a lot longer than I expected. Anyway, on to the crazy science rambling!
First, let’s recall the story of Elle’s mutant SuperCoolGene and the Kid’s mutant SCG. Elle (and subsequently, I) put the mutation into pDest32 (a Y2H vector) using Gateway Cloning, and the Kid used the Quikchange Kit (thanks Agilent!) to make the mutation directly into pDest32-SCG.
Actually, now a slight retraction. The Kid was never able to get it to work, and thus HE DIDN’T ACTUALLY MAKE THAT MUTATION. A visiting scientist from another lab did (said scientist is currently collaborating with us while their lab is underfunded, but they are funded again and so they will return to their lab later this year once our collaboration is all done.) I am still going to refer to the Kid’s version of the mutant as his, since it was supposed to be his doing
So there is Elle’s mutant SLG, and the Kid’s mutant SLG, both in pDest32. Elle’s mutant binds to the Insulin-Related-Thing (IRT) and the other transcription factor (which Elle herself tested way back in the day), and the Kid’s mutant does not. Where I last left off, the Kid and I were stumped.
And that is why Bossman is around. Bossman is my boss, and when I am stumped, I ask him. That is hopefully one of the reasons he is paid (in addition to having to go to lots of boring meetings and having awesome ideas to write into grants). Bossman asks how the mutant compares to wildtype SCG.
OOPS. I did not do a Y2H of the wildtype versus the two mutants. The Kid mocks me for this, because he had asked why we didn’t do this before. It’s because I was lazy. I will admit that. This is a good life lesson to take away from this whole ordeal: don’t be lazy or else you may have to repeat things.
So the Kid and I set up plates comparing his mutation to wildtype to Elle’s mutation. I also prepare ANOTHER sequencing order of his and Elle’s mutants, however this time I intend to look at the entire inserted gene. What if there was a new mutation accidentally inserted that screws something up?
Several days later, we get our results. Once again, Elle’s mutant and the Kid’s mutant don’t match. However, neither match with wildtype either. Elle’s mutant is stronger than wildtype (this is a detail that I am constantly forgetting and have since confused other people regarding this detail. I blame my memory loss on THIS MUTATION IS NOT MY PROJECT and also don’t talk to me about science when I have food because I am focused on food and not science) and the Kid’s mutant is still crap.
Bossman’s next piece of advice is to do Western Blots to see if perhaps the yeast are making less of the Kid’s SCG protein compared to Elle’s SCG. I hunt down a protocol on how to isolate yeast protein from another professor, and the Kid and I plan to attempt Westerns in a few weeks once we have finished other projects. I have done Western Blots. I have learned Western Blots from three different people (from a different lab during a rotation, from the grad student who graduated before me and finally from that visiting scientist I mentioned earlier). Despite all their mentoring and help, I am a hopeless wreck at Western Blots. I can follow a protocol precisely, and somehow still end up with either a completely blank film, or a completely blacked out film. The only successful Western Blot I ever did was one where I was merely testing a tubulin loading control. I would love to practice more, but it doesn’t help that other people are hogging all the Western Blot equipment constantly. To summarize, I am not thrilled about doing Westerns, nor am I about to ask anyone else in the lab to do them for me (the Kid can go beg others for help though ヽ（´ー｀）┌)
And then the sequencing comes back. On Monday (Memorial Day), I decided my task for the day would be to examine it closely to see what exactly had gone wrong.
I did my standard sequencing analysis, which is paste everything into NCBI’s BLAST search where you can compare two sequences. If you do any kind of genomics or bioinformatics, then NCBI is your best friend, and so are BLAST searches. So I compared the 3 pieces of the Kid’s SCG mutant to the wildtype coding sequence. Looks okay, except there is one region where it looks like things are overlapping funny? I initially predict it may be due to a stretch of Ns in the sequencing results.
DNA is made up of 4 nucleotides (or bases): A, T, C, G. For my intents and purposes, that is all that matters. A Sanger sequencer can detect these bases and will call them as such, but when it can’t, it will assign it as N. This is where you have to look NOT just at the text your lovely sequencing service has sent you, but at the DNA chromatogram file. A DNA chromatogram is just the visual representation of your sequence, usually annotates with the sequence and color coded so a red peak will be T and a black peak is G, etc. I use DNA chromatograms to compare the best peak to what the nucleotide in the sequence ought to be, and usually, the Ns are the correct nucleotide but the sequencer wasn’t able to confidently call it. Generally, the first 100 base pairs and the last several hundred base pairs of a sequencing reaction will look like absolute trash (long stretches of Ns) so you pick your primers carefully so the parts you want will look nice.
All of the Ns in the Kid’s SCG mutant end up being the right nucleotide. Thus far, everything looks just fine. But it can’t be fine? And what was that funky overlap?
ExPASY is another website with amazing tools for figuring out biology. I rarely use it myself, except for it’s translate function. Recall the (very simplified) central dogma of biology:
DNA > RNA > Protein
Three DNA bases code for a single amino acid (the building blocks of proteins), however a single amino acid can be coded by several combinations of three DNA bases. There are also start and stop codons. Translating my DNA sequence into an amino acid sequence by hand would be a huge pain. Instead, Translate will generate the translation for you in all three frames, in both 5>3 and 3>5 directions.
I ran the coding sequence for wildtype SCG through translate and get a beautiful amino acid sequence; methionine to the several motifs we study to a stop at the end. When I run the Kid’s SCG mutant, it should be identical except for a single amino acid base change. Instead, it is identical until about 5 amino acids PAST the mutation, where suddenly there are a few different ones, and then a stop codon. Wait, WHAT?!
Translate, in all it’s beautiful usefulness, as a function where it will show your amino acids lined up with your nucleotide sequence. I take advantage of this to see exactly where in the sequencing that things have gone south. I try to line things up in a word file manually to try and figure out why things are suddenly not matching.
Remember earlier how when I BLASTed my DNA (man they picked a great name for that program…) there was a weird overlap? The overlap occurs slightly upstream of the mutation, but thing go to hell slightly downstream. What did the primers look like? Being Memorial Day, the kid is not around, so I look at Elle’s primers, since they should be about the same. The upstream overlap matches the primer exactly, which makes sense because the way primers are designed is to match the region of DNA you want to change, except for the pieces you intend on changing (note: Most special cloning kits, such as Gateway, Quikchange, Gibson, Q5, etc, will tell you HOW they want you to design their primers, and if you’re lucky, they’ll point you to their online primer designer).
While staring at this bit of sequence, I notice that right after the primer sequence is the same sequence, except the mutation is not present. Say the original amino acid is glutamine (nucleotides GAA) and the mutation is glycine (nucleotides GGA). The primer sequence has GGA and is still in line with everything, and then right after it is the original sequence containing GAA, and this extra primer sequence has thrown things out of frame, introducing a stop codon downstream.
Congratulations, we now have a new mutant SCG which is now missing it’s entire C-terminal domain and has a mutation too. Mystery solved! (*-`ω´- )人 I am so proud of myself! But how did this happen?!? I don’t know. I haven’t used the Quikchange kit. If I was
Sadly, no one else was as enthusiastic about this as I was. The Kid proceeded to blame the visiting scientist, who was quite baffled and wasn’t sure how this made any sense. The sequencing data shows this region is present twice and should introduce a stop, and when actually tested, it DOESN’T WORK so it kinda makes sense. I just have no idea how this came to be. °˖✧∩(◎ヮ◎∩)✧˖° However, I no longer have to worry about this, and I also don’t need to do Western Blots (relating to this anyway….), so I am happy with this outcome.
I also checked Elle’s sequencing data. The first two pieces came up as mostly Ns, but the last piece matched exactly with the end of the wildtype SCG amino acid sequence (I checked her mutant in ExPasy too), so I think it’s safe to say her mutant is good. Which is great, considering her mutation is the one IN ALL OF OUR FLY STOCKS. ⊙︿⊙ And we do NOT need another crisis where people have worked on a mutation for years only to find out later that it was wrong.
And so I can now move on to something that isn’t this stupid mutation.