User:Andy Maloney/Notebook/Lab Notebook of Andy Maloney/2009/04/16/Kinesin & Microtubules

Disclaimer

These are my notes and thoughts on this paper. Do not take anything I say as being correct or incorrect until you have read the paper first. If I do say something wrong, please feel free to correct me.

This was a tough paper but well written. I had to read it 4 times before I was able to understand what was going on and write this.

Paper

Direct observation of kinesin stepping by optical trapping interferometry

Review

• Step size - distance a kinesin molecule will move forward between dwell states.
• Sliding distance - distance a kinesin molecule will move forward for every ATP hydrolysed.
• People like using kinesin and microtubules because:
  • Kinesin
    • You can vary the speed at which kinesin walks along microtubules.
    • Remains attached to microtubules as it walks.
    • You can make recombinant kinesin.
  • Microtubules
    • Microtubules can be visualized in regular DIC microscopy.
    • They are more rigid than actin.
    • Note: Koch believes that we can make recombinant microtubules. This is a kick ass idea. (Steve Koch 00:47, 17 April 2009 (EDT):Hey! Didn't I say it would be great if we could make recombinant tubulin, but that actually it's pretty difficult? :) )
• They used a dual beam interferometer trap. Note: Oh so cool and yet such a simple idea. I think I should look into trying to do this since we will have DIC capabilities.
• To determine the step sizes of kinesin, they did 3 experiments.

1. Moderate levels of ATP in solution (10 µm) and low loads.
2. Saturating levels of ATP in solution (500 µM) and and high loads.
3. Low levels of ATP in solution (~1-2 µM) and low loads.

• • They noted that in the first experiment, they were not able to directly measure the stepping nature of kinesin. But, they were able to see the steps in experiments 2 and 3. To reconcile the stepping nature in experiment 1, they did a lot of fancy statistical analysis that I have no clue as how to do. Question: Will I need to know how to do these statistical methods?
• They briefly describe their optical trapping interferometer. They reference where the idea came from...Watt Webb. Note: I'm going to have to reteach myself DIC microscopy. I'll make sure to put a page up that describes it.
• They talk a bit about noise in their system. I'm not sure why they do but I can glean that since they did, it's a big deal.
• They actually used the TMC vibration table I suggested us getting below. Perhaps we should get it as well.
• Question: I still don't know what a power spectrum means. What power are they talking about?
• They talk about normalizing this power spectrum. Question: Is this important to do?
• They purified their own kinesin with squid optic lobes. Eww, but I guess necessary.
• Their buffer consisted of:

80 mM PIPES

1 mM MgCl₂

1 mM EGTA

pH 6.9

• They coated/incubated their beads with:

Buffer

50 mM KCl

0.5 mM dithiothreitol

50 µg/mL casein

~ 50 µg/mL kinesin for some amount of beads

• Their solutions for experiments consisted of:

Buffer + Beads

Either

1. 0.5 µg/mL phosphocreatine kinase + 1-500 µM ATP + 2 mM phosphocreatine
2. 2 mM AMP-PNP

• Their microtubules were stabilized with Taxol and put on coverslips that had been treated with a silane. Question: Will we need to use a silane coated coverslips? I thought the purpose of using casein was to coat the exposed glass after you stuck microtubules on it. After introducing the microtubules, they flushed the coverslips with casein and Buffer.
• They specifically chose microtubules that were aligned with the Wolaton shear axis. Note: I think this step is very important.
• They state that through Poisson statistics, the probability for a kinesin to be attached right next to another kinesin on a bead was about 2% if you generously consider the head of a kinesin stretching out 100 nm. Question: Do kinesin stretch out that far?
• This is really cool. They noticed that as the bead moved to the edge of the trap (i.e. higher loads) thermal energy displacements of the bead decreased.
  • Note: What does that mean? Well, that means that the bead stopped jostling around so much, enough so that they were able to see stepping motions. So, what does that mean for kinesin? Well, they effectively argue that kinesin is stretched. This stretching causes a nonlinear effect to the kinesin if you were to model it as a spring. Linear springs don't care about thermal motion and ideally, you can stretch them infinitely without any problems. This means that the properties of the spring, do not depend on how far you stretch it. However, nonlinear springs do care about how far you stretch them and thus you get a dependence on thermal fluctuations to how far you stretch the kinesin.
• They continue to say that the nonlinearity may be because of the bead-kinesin system.
  • Note: I like how they start this argument. They say that the kinesin molecule is an entropic spring. I agree totally with this statement. However, they say that attaching a bead to the end of a kinesin actually gives the motor more degrees of freedom. I would disagree with this statement because a kinesin molecule is free to move in any way it wants to without the bead. In all actuality, I would argue that the bead reduces the area of momentum space a kinesin molecule can go. Mainly because the "head" of a kinesin molecule can in principle touch the microtubules when no bead is attached to it. When there is a bead attached to its head, it cannot touch the microtubule. This effectively reduces the space the kinesin molecule can visit. Reducing the space an entropic body can visit means that you reduce the entropy of the system. Reducing the system's entropy means that there is an entropic force resisting this change. What that entropic force is eludes me but, it is possible that it contributes to stretching the kinesin molecule in the trap.
    • Uh oh. Did I just say that the forces we measure in an optical trap are in fact lower than what they should be due to entropy?
• They talk a lot about pairwise distance distribution functions. I have no idea what they mean.
• They say pairwise distance distribution functions are the same thing as the autocorrelation function?

Take home

The Block lab kicks ass. They also were able to show that kinesin moves in discrete 8 nm steps.

• I think I need to understand their PDF analysis.
• I should understand their tweezer interferometer.

Steve Koch 00:59, 17 April 2009 (EDT): I think you'll easily be able to understand both of these. PDF was (or maybe is) popular for analyzing stepping of motors when step size was unknown and / or noisy. I understood it back then, so I don't think it's too tricky. What I remember being more tricky (and very cool) was the variance analysis (randomness parameter) that Mark Schnitzer did in block lab. It was pretty sweet, but I don't recall seeing it much recently. It reveals the number of rate limiting steps, even if you can't see the substeps. The more rate limiting steps, the more clocklike it becomes. As for using the Wollaston as an interferometer, I'm not ever sure if anyone does that anymore. It turns out you can get very good precision with the QPD back focal plane method. It can be an annoyance being a strictly 1-D method. I had an idea back in the day of using AODs (or better yet EODs) to make an interferometric technique by deflecting one beam around the other beam (instead of using a wollaston prism to split the beams). But I don't think you get any benefit out of it, relative to using a QPD. Although there could be a cool scanning DIC microscopy technique you could do with EODs and lasers. Definitely I'm off on a tangent now.

Optical tables

Below are comparables to what we have now in the lab. I think if we want more vibration control, then we may want to go with the TMC tables and legs.

• Thorlabs
  • Active vibration control legs.
  • Breadboard.
  • Total = $3976.30
• TMC
  • Same legs as we have now.
  • Breadboard.
  • This might be cool to have for isolating the microscope. Has anyone used these before? I know they have them at the cancer research center.
    • Steve Koch: It reminds me of the "minus k" isolation platforms, which rely on a passive mechanical system. Alex Corwin at Sandia had one that he really liked, but I don't have other verification. I think I convinced Adelman to get one that she ended up not liking, not sure. The main disadvantage of the minus k (not sure if it's similar to the one you linked to) is you have to tweak it for the exact mass. Alex dealt with this by buying a 200 pound weight set so he could balance any changes without having to re-tune it.
  • Total = $4900.00
• Newport
  • Newport doesn't have any comparable leg system.
  • Breadboard
  • Apparently Newport likes to hide their products. I was able to find a shelving system that will fit the table we have now here.
• New Focus
  • Complete system, funky shapes.
  • Total > $4000
• Vere
  • Vere makes shelves to size. The shelves have 1/4-20 threaded holes on them. (Koch is smiling now for some reason.)

---

Andy Maloney 01:58, 17 April 2009 (EDT):

I just found a new company, Kinetic Systems, that has reps here in NM. Apparently they will send people out to our lab, assess what kind of vibrations we have and recommend their products dependent on what kind of isolation we want. Seems like it's way overkill but I'm not sure.

• Koch Question: Could you give me your input in regards to the optical table? Lead time for tables and legs are like 3 - 4 weeks and I'd like to order something this week so we can get it around the same time the wet lab components come in.
  • Steve Koch 02:07, 17 April 2009 (EDT):Sorry, I think we still need to wait. I'm still hoping we can use the other lab (which has tables). We can probably get another little table or some other "tide me over" option until we know what we're doing. Make sense?
    • Andy Maloney 02:15, 17 April 2009 (EDT): Yes. I believe I'm interpreting your answer as: Don't order anything. We will borrow a table from someone if we can't use the other tables in the other lab.