Tag Archives: intro

The SciFund Experiment: Yeast Adaptation to D2O

It has taken me quite a while, but I’m finally at the point where I can begin the experiments that my SciFund quest funded. To refresh your memory, here is my project proposal:

Water is arguably the most important molecule in the universe. It’s a simple molecule that is composed of two hydrogen atoms and one oxygen. But did you know there are different types of water molecules? Every element has alternate forms known as isotopes, and hydrogen is no different. A common isotope of hydrogen, known as deuterium (D) which is twice the mass of hydrogren, can bond with oxygen to make heavy water (D2O).

Immediately after heavy water was first purified from naturally occurring water in the early 1930s, it was discovered that most organisms cannot survive in pure heavy water. It was also shown that increased (but not toxic) levels of heavy water significantly affect many systems in these same organisms, like fertility, metabolism, temperature regulation, and many more, all of which are essential for healthy organism function. Interestingly, on a cellular level the increased mass of heavy water may affect chemical processes. Not many studies have been performed in this area, unfortunately, because many experimenters ignore the effects of water even though it is by far the most abundant molecule in these experiments.

Because of the presence of deuterium in naturally occurring water, life may have evolved essential uses for deuterium. I plan to study the effects of heavy water on E. coli and S. cerevisiae (baker’s yeast). I will be growing cultures of these microbes in water with varying amounts of heavy water (from 0% to 99.9%) and comparing characteristics between the cultures, looking for effects in growth, development, appearance, and other physical differences.

But that isn’t specific enough to describe what I actually intend to do with the money. The full explanation of the experiment is way cooler!

Initially I’m going to continue the experiments from April and May that compared the growth of yeast grown in DI water, DDW, and D2O. Each time I did a time trial experiment I would start from scratch to show that yeast grows much slower in D2O. This time I’m going to carry over the yeast each day. Over time I hope to show that the yeast is adapting to growing in D2O and begins to develop at a similar rate to how it develops in DI water and DDW.

Once that happens I’ll be doing two experiments.

First I’ll be looking for phenotype differences between yeast grown in H2O vs yeast grown in D2O. I’ll be looking for shape, size, motility, etc of the yeast using the lab’s light microscopes.

In the second experiment, I’ll flip the script on the yeast. Once I get the yeast to adapt to D2O, I would like to determine if H2O is as harmful to D2O adapted yeast as D2O is to H2O yeast (aka natural yeast). I’ll do this by growing the D2O adapted yeast in DDW/DI water and recording the growth. No matter what happens with this experiment, the results will be interesting.

3-piece ligation: BpALS, EpBR, 5′-bio/int-bio adapter

A few days ago I did a 3-piece ligation that didn’t turn out so well. Today’s ligation is more like the traditional ligation that I was supposed to do with the exception that I’m using the pALS anchor instead of the pRL anchor.

I mentioned earlier, that BpALS (pALS anchor digested with BstXI) is better overall because it is longer and makes the optical tweezer operations a little easier and better.

pBR322 can be digested with EarI or SapI for use in our construct for DNA unzipping. I haven’t done any studies that show which method produces better results in terms of final yield, but I’m pretty convinced SapI will because you only have to perform one gel extraction. If you digest pBR with EarI you get two linear fragments. And according to separate analysis by Koch and myself the overhang that we need for the unzipping construct ligation is the longer piece (in the gel image linked above it is the brighter top band).

The benefit to following this method is you can perform a three piece ligation (essentially ligating the entire construct together in one shot) because the EpBR piece (pBR322 digested with EarI) can only ligate to the adapter, and the BpALS/BpRL piece can only ligate to the other end of the adapter (both adapters that I’ve been using work the same way).

And so today I’m doing the proper three-piece ligation reaction. The reaction setup and method can be seen in the table below:

pALS PCR digestion with BstXI

The pALS anchor is special in that it has two purposes. The first is that it is immediately usable in DNA stretching experiments because it has a dig molecule (to stick to glass) on one end and a biotin (to stick to microspheres) on the other. The second is that I can digest it with BstXI restriction enzyme and use it as the anchor segment for unzipping DNA.

And when it all works well, pALS is much more useful than the pRL574 anchor (1.1kb). It’s extra length makes it easier to calibrate the optical tweezers for unzipping and we can get higher forces in the optical trap by using bigger beads. (Note: The tweezers are the entire device, and the trap is the focal point of the laser in the microscope. So the trap is a subset of the tweezers.)

With the huge success of the pALS PCR yesterday, I’m going to digest some of it and then ligate this piece to EpBR and both adapters. But first here is my digestion reaction:

Annealing the adapter oligos

I purchased some oligos and now I need to create the adapter duplex from those oligos. This requires an annealing reaction which is super simple to do. Basically you just mix your oligos together, heat them to 95C and then slowly cool the mixture. Nature takes care of all the leg work. I’ve done this reaction 3 different ways:

  1. Heat a cup of water to boiling in a microwave, then remove the water, place your annealing mix in the water (make sure the top is floating), and allow the water to cool on a lab top. Basically allow the water to cool to room temperature (RT).
  2. Put your mix into a heating block and heat that to 95C (or close to boiling). Once the mix has been heated, remove the block from the heating unit and place on a lab top to cool to RT.
  3. Put your annealing mix in a PCR machine which can control the temperature very specifically. Create a program that will: (1) heat the mix to 95C for about 5 minutes, (2) slowly lower the temperature to about RT or 4C or whatever cool temp you want, (3) hold at that low temp until you are ready to remove it and move on.

For today’s reaction I will be using option 3. My protocol for the experiment is below. Unfortunately there is no easy way to verify the annealing reaction is successful. Well that’s not 100% true. You can run an SDS-PAGE gel which has the ability to resolve very short DNA sequences, but I don’t really have the equipment for that right now. I do have a high resolution gel that I’ve been wanting to try. Hmmmm…

And here is a link to what my annealing buffer is: Annealing Buffer Recipe.

Shotgun DNA Mapping: Microspheres

Today I need to buy microspheres. In the first experiments we used spheres with a diameter of 0.5um (or 500nm), and over time we eventually switched to using 1.0um beads. The reason is because: 1) big beads are easier to see if they are tethered or not, 2) you get better tweezer forces with the larger beads, and 3) the big beads clump less. The only drawback to using the larger beads is that there is a lot of repulsion between the beads and our glass surface so our DNA tethers need to be longer. I invented the 4kb pALS anchor to solve this very problem.

Here are some pictures of the different sized beads in the tethering environment:

And for completeness here are some old notebook entries regarding those pictures:

And here is a video that shows the tethering results of the DNA experiments listed above:

We want beads that are coated in streptavidin (or some form of avidin) because this molecule creates a very strong bond with biotin which is attached to our DNA for stretching/unzipping experiments.Most commonly, you can order beads with streptavidin, but some companies offer alternatives like avidin or neutravidin. In my experience neither works any better or worse than streptavidin. And in the case of these experiments the bond either holds or it doesn’t.

But believe it or not, it is hard to quantify the effectiveness of the beads. As you can see in the pictures above, both bead concentration and DNA concentration can affect tethering efficiency. And I have suspicions that the sonication process (what we do to prevent the beads from clumping) may affect the streptavidin in some way: in my head the vibration shakes off the molecules from the beads.

With all that said, there are places that I trust buying beads from. In the past I’ve purchased beads from Bangs Labs, Invitrogen, and Poly Sciences. I’ve never noticed that any bead from any company seems to work better than the rest. Because of this I think I’ll order a new stock of beads from Bangs because they pretty much only make beads (so they should do it the best). Note: I just remembered that Bangs, and Poly Sciences may be the same company and it turns out they are affiliated in some way. So I suppose there really is no difference between the two.

Update: I’m placing my order with Bangs Labs. I’m ordering 0.53um beads and 1.04um beads, both coated in streptavidin, and neither are fluorescent.

Shotgun DNA Mapping: Unzippable DNA

Unzipping the complete construct…

We’ve discussed the anchor DNA and the adapter duplex, but we wouldn’t be able to measure unzipping forces and shotgun DNA mapping would fail if we didn’t have any DNA to unzip. As I’ve mentioned several times in this series of posts, the entire construct is assembled through a reaction known as ligation.

For the purposes of this experiment, the reaction works as follows: an enzyme known as DNA ligase looks for compatible ends of DNA and attaches them together. In our construct those ends are the overhangs that I referred to in the other posts. And the construct is designed so that the anchor can only attach to one end of the adapter and the unzipping segment can only attach to the other end of the adapter.

Now technically we can use any piece of DNA to unzip. The catch is that we need to use a plasmid to get the overhang that we need. As I said earlier, one side of the adapter can ligate to the anchor. The other end’s overhang is created by a cut from the enzyme EarI and is specific to the plasmid pBR322 and any other plasmids that have the same multiple cloning site. For instance, for shotgun clones (which will be explained much later) we use pBluescript II, and the enzyme SapI cuts the plasmid with the exact same overhang as EarI does in pBR322.

Because of the proximity of the SapI site to the multiple cloning site, we can stick any piece of DNA into the plasmid for cloning. Then we can cut the plasmid with the unzipping insert with SapI and then ligate this long piece to our unzipping construct.

In calibration experiments, we use pBR322 to test to make sure we have unzipping. And eventually we will move up to use the pBluescript clones that I made a few years ago. Although I have a feeling I’ll be doing that all over again.

Shotgun DNA Mapping: The Unzipping Adapter

Ignoring the circle, the adapter duplex (the middle piece, red) will be the topic of today’s discussion.

The ligation reaction that I keep referring to requires three pieces of DNA. They get fused together all in one shot, that is slightly complicated. The most crucial of which is the adapter duplex, because without it the anchor and the unzipping DNA would not attach and the reaction would yield nothing. And because of how important the adapter is, this has been the source of my troubles for the past 4 years. But before I go into that, let me tell you about the duplex.

It’s called an adapter duplex because it is actually two single stranded pieces of DNA. We call them the top and bottom strand. They are short DNA sequences manufactured from biotech companies. In the past we’ve used Alpha DNA, but I’m thinking of trying someone new. How short are the strands? The bottom strand has about 35 bases and the top is only a few bases longer. Compare that to the anchor sequence which is either 1100 bp (base pairs) or 4400 bp or the unzipping sequence which can be as long or as short as we want (but typically around 3000bp for calibration sequences).

Once our single stranded sequences arrive via mail (we call these short sequence oligonucleotides, or oligos for short), we need to bind the top and bottom strands together in a process called annealing. Most molecular biological reactions involve some kind of enzyme to help the reaction, but annealing is quite a natural process. DNA naturally wants the bases to bind to complementary bases (A-T, G-C) and even in single stranded form, the DNA will self anneal, that is bind to itself. So to get our top and bottom strands to stick together we just put them together in the same tube, heat it up to near boiling temperatures, and slowly bring the temp down so that the top and bottom strands find each other and bind. Once it’s cooled, the adapter duplex is formed and will stay that way unless heated to very high temperatures (near boiling).

There are three key features of the adapter duplex: (1) a biotin molecule, (2) a gap in the DNA backbone, and (3) two non-palindromic overhangs. The overhangs are designed to bind with a very specific sequence. One side can only bind with the overhang I mentioned in the anchor DNA, the other side can only bind with the overhang contained in the unzipping DNA. Right now that particular sequence is very specific to cutting plasmid pBR322 with the enzyme SapI (and any other plasmids that share similar properties).

The biotin is necessary for unzipping. The biotin has a high affinity for streptavidin which coats the microspheres we use for optical tweezing. Typically the biotin in our bottom adapter strand is near the start, but not at the start of the sequence. In more recent iterations, we moved it to the 5′ end completely or added a poly-A overhang with several biotin there. The reason for this is because we’ve been having issues actually unzipping, which I’ll explain in another post. The hope was that by moving the biotin we would get better tethering efficiency and better unzipping. We ended up not getting unzipping results and the tethering efficiency studies were inconclusive.

See wikipedia, DNA article

The bottom strand has both the biotin and the gap (key feature 2), which actually plays a role in the unzipping. Since the tweezers will pull on this side, the gap was designed to aid in the unzipping. Basically the gap was the weakest point in the complete DNA chain and since the microsphere is so close to it the DNA would begin to unzip from this location. The gap is actually a missing phosphorus (the yellow in the image to the right), which prevents the anchor and the bottom adapter strand from connecting to each other.  In later iterations we completely removed the first base to make the gap wider, and the poly-A tail I mentioned was also used to prevent there from being any attachment.

Ultimately I never got unzipping to work. Oddly enough, I ran experiments that verified the ligation reaction worked, but could never get the completed structure to unzip. That’s what this new set of experiments is going to attempt. But before I get to that, I need to tell you about the unzipping DNA portion!