May 9

May 9 Electrophysiology (i.e. optimization) appreciation post!

The 4th year of my PhD is winding down, so I want to use this post to look back on my progress as an electrophysiologist.

Frog eggs

The first glutamate receptor tracing I ever generated. The glutamate/glycine concentrations are crazy high, but it worked! So much optimization.

I started off using two-electrode voltage clamp (TEVC) in xenopus laevis oocytes (frog eggs) to study ionotropic glutamate (NMDA) and purinergic (P2X4) receptors, a technique that has its own pros and cons. In the early days, my cells suffered from contamination often, so I had to optimize the buffer that we store the cells in, and after that, getting my cells to express the genes I injected into them was a challenge. I finally figured out the best injection conditions and have been steadily building my project from there: incorporating other receptor subunits, mutating genes to see the effects, and incorporating drugs that affect the receptor’s ability to function.

Unhealthy oocytes. Membrane damage suggest storage issue; maybe the buffer was made incorrectly (pH, ionic strength, etc) or oocytes were stored at the wrong temperature.

Unhealthy oocytes contaminated with bacteria (black spots). There’s a great paper here on how to solve bacterial contamination issues in oocytes.

Healthy oocytes. I injected these cells yesterday, with both RNA and DNA (long story, I’ll explain in another post.)

Building stuff

I also rebuilt a few of our broken electrophysiology rigs and found that I really really like building things. The cool thing about electrophysiology is that you can set up a rig however you like, you just have to have a good understanding of physics (electricity and magnetism specifically.) One of my proudest contributions to lab has been replacing our old microscopes with electronic ones. They’re ridiculously cheap (<$300 on amazon!) compared to typical scientific microscopes, and make teaching students so much easier! You can take pictures of the frog eggs to keep track of the quality, and can even record video! The electrical noise they generate is minimal, and they look very cool. It also makes injecting the frog eggs way easier, since the old microscopes strain your eyes and require you to be hunched over a Petri dish for hours at a time.

Our injection station equipped with the digital microscope. Injecting cells is much easier and faster now.

Our rig with the new digital microscopes. Much easier to work with.

Brain slices

Going to the University of Illinois, Chicago to learn how to perform brain slice electrophysiology (extracellular recordings) was one of the hardest things I’ve had to do for my PhD. TEVC electrophysiology is hard for a few reasons: sometimes the health of the cell can affect your ability to get data, and/or sometimes the cell might not express your gene. With brain slice electrophysiology, you’re working with neurons (just like the name implies) which are even more sensitive than frog eggs. You must quickly and carefully sacrifice the animal, remove the brain, identify and cut the brain region you’re interested in studying, and then set up your experiment. If your technique is slow or sloppy, then it’ll very difficult (if not impossible) to get data.

Also, working with animals is tough. You can’t work with cells in a petri dish forever; at some point, you need to work on an animal model if you want your work to be translatable to human health. In Chicago, my technique was good; despite my “slow” brain prep time of 2 min and 25 seconds (a good time is <1min) I always got viable neurons. However, since Chicago, I’ve learned that I’m a much bigger proponent of the 3Rs of research than I thought, which stand for reduction, replacement, and refinement. You can read more on these principles here .

What’s next?

As I start the 5th year of my PhD, I’m still developing new tools and learning new skills. I set up the Roboocyte 2 automated TEVC electrophysiology rig, and I’m in the process of coding our experiments so that the robot can run our experiments overnight. I’m also optimizing the Robo-inject, which injects our frog eggs with DNA/RNA/proteins. In my initial runs, I’ve found that it can do injections 90% faster than our students, which makes it a very powerful tool. The issue I’m trying to solve right now is the injection parameters. When I teach my students to inject, I can see their technique and correct them as necessary. With the robot, I program how deep to pierce the cell, how fast to inject, and how much, and the robot follows this exactly. The problem is variability. The robot will always do what it’s told, but the cells will not always be the same shape/size/quality. I’m trying to figure out how deep I should inject DNA without hurting the cells but still get data.

The old OpusXpress TEVC system. An electrical problem caused it to break down, and the company no longer serviced the system.

The new Roboocyte 2 and Robo-inject. Cells are stored in 96 well plates and the system does the rest.

I’m also setting up the brain slice electrophysiology rig in our lab, which has way different equipment than what I learned to work with in Chicago. Despite these differences, the principals of physics still apply, and I’m making good progress in getting things up and running.

A mini-passion project I’d like to start in lab is getting our patch-clamp rig up and running. Patch clamp patches on to cells to record current/voltage changes in response to drugs/agonists, unlike our brain slice recordings, which only get close to the cell, or TEVC recordings, where electrodes actually pierce the frog eggs. I’ve made a lot of cool fluorescent receptors, and it’d be cool to validate them using our patch-clamp rig. The issue with our rig is that the computer’s motherboard burned out (to be fair, its >20+ years old) but we can’t just buy a new computer, since new computers don’t support the cable that controls the equipment (SCSI). We could buy new equipment that can work with a new computer, although that would cost >$10,000.

What I want to do is rebuild an old power mac G5 to interface with the old equipment. Parts for the G5 are inexpensive and easy enough to find online, and 3rd party software is available (and affordable) that still has technical support. This was the setup in Dr. John Woodward’s lab at the Medical University of Southern Carolina (MUSC) during my visit last year.