As a PhD student Anne Marie CraigIn my lab at the University of Illinois at Urbana-Champaign, I studied what directs two glutamate receptors to their location for function in hippocampal neurons. We thought the cytoplasmic ends of the proteins might be the culprit, so we swapped the C-termini of these proteins to create chimeras. To make it easier to detect both the chimera and the two original proteins in different cell cultures, we inserted a three amino acid tag sequence at the N-terminus of all proteins that could be detected with an antibody.
Julia Darby, who now tutors in school districts in Iowa and Illinois, was a graduate student studying glutamate receptors at the University of Illinois at Urbana-Champaign when, after defending her dissertation, she learned that some of her research methods might not be as effective as she thought.
Julia Darby
When we expressed different protein constructs in neuronal cells in culture, we demonstrated that these C-terminal tails determine the localization of glutamate receptors, and we obtained beautiful images showing that one receptor is found in the axon and a second receptor localized along the axon in the dendrites. Neuron When we published the data, we chose them to be the cover of the issue.
I followed this success story with my PhD and moved to Washington University in St. Louis for a postdoc position. About a year later, a postdoc in Craig’s lab followed up on my project, studying the native receptors. Their data showed that my receptors, which appeared uniformly along axons, were actually clustered in defined locations on axons and synapses. It turned out that the tags I had introduced were disrupting this activity, causing the proteins to not interact as they should.
When I found out about this discrepancy, my first thought was that I was glad I had already completed my PhD: although the alignment we observed was wrong, our findings about the role of the C-terminus were correct, so I did not retract the paper.
This incident reminds me that in science we raise many questions, but how we answer those questions is more important than the results themselves.
This interview has been edited for length and clarity.
Share your best and worst fails with us for a chance to be featured in the “Epic Fail” column.
