Please visit our new site!

Lab Rats


When that experiment refuses to work

Veronica Akle 26 May 2011

Secrets: zebrafish eggs have a story to tell

Now that I look back at it, I realize that it was a very silly mistake, but everything looks different in retrospect

When an experiment refuses to work, how stubborn does a researcher have to be to continue working on it?

The answer is simple: one has to be very persistent, but not blindly so. Today, after many months of trying the same experiment, something almost unbelievable occurred: I got a decent result. It is not that I had obtained negative results up until that time – in other words, that the experiment gave me a clean answer of ‘no’. This would have been fine. Instead, the problem was that I was getting results that did not lead anywhere at all.

Being a graduate student, I am very used to experiments not working. But as time has progressed, I am finally realizing that I have to graduate sometime soon, and I’ve grown impatient and somehow bitter about things not working in the lab. I've changed aims in my project numerous times. At first, from simple curiosity, I engaged in multiple, fun experiments; but later, I changed gears according to experiments’ willingness to work.

However, there is one part of my project that must work, as my supervisor says, “no matter what it takes”. So, since there was no escape and the pressure of a graduation date was looming, I sat down to design the plan for the next experiment...again.

My experiment in general has consisted of injecting specialized chemicals (called morpholinos) into one-cell stage zebrafish embryos to block the expression of my protein of interest – in this way, I hoped to try to work out what it was doing normally. In other words, I would monitor the developing fish and try to establish a function for the protein based on the phenotype it produces when the organism doesn’t have it.

After trying numerous morpholino targets, multiple combinations of them, and even new injectors and needles, I was close to declaring that the protein has no relevance during development whatsoever. Nevertheless, I’ve known all along that this cannot be true, as we could see that the protein was normally present in essential structures leading to the formation of the nervous system.

So what was wrong? I took a long break (of one entire week!) and decided to think the experiment through all over again.

Now that I look back at it, I realize that it was a very silly mistake, but everything looks different in retrospect. The only thing I had not done was to observe the animals overnight after injection. I had checked on them late before going home, and I had observed them early the following morning, but the time in between was a mystery. As a result, I could never understand what had happened, and simply assumed failure and started the next experiment with a different set of conditions the time.

Today, I am at the lab and it is 4 am. I have checked my animals every half-hour under the microscope since 9 am when they got fertilized and we injected them. At 8 hours post-fertilization, I start seeing a difference with the controls; and by 14 hours post-fertilization, I know that the development of the animals is compromised. I continue to check all 800 eggs and take a few pictures of them. Most have stalled development and look like they might be about to die; however, I notice that the controls are starting to succumb too, but in a different way. After 20-hours post fertilization, all the embryos are dying: the controls are surrendering to a fungus that is taking over their bodies, and the experimental animals are not developing because of the effect of the morpholino that blocked an important signaling molecule. For these two very different reasons, I now know why all the eggs were always gone in the past when I’d come in in the morning.

I can tell the story with a smile in my face tonight, but I have suffered painstakingly over the past six months. I plan to enjoy the moment and not think (today) about all the controls I now have to run to make sure that my results are true. Also, I won’t punish myself (today) for not checking the eggs more carefully before. But what I will do is to celebrate being a graduate student, pat myself on the back and then go to sleep happily after a very productive long experiment.