What I Have Been Up To!
So in this post I want to talk about a new research paper that two students of mine published recently. This paper is important to me for a couple of reasons: 1) These two students designed and executed this experiment on their own, with me serving as little more than a cheerleader-and as such these two students are legitimately co-first authors. 2) These students’ work extended the utility of one of my earlier experiments. And 3) The data in this manuscript provide more comprehensive data regarding the interaction of Fmrp and Arc proteins for learning and memory processes.
The paper that I am writing about in this post is “Reduced activity-dependent protein levels in a mouse model of the fragile X premutation” by von Leden, Curley, et al. It is in Neurobiology of Learning and Memory. If you do not have access, please email me and I will send you an author’s copy by pdf.
What Did We Study and Why?
In this manuscript, Ramona von Leden and Lindsey Curley (the students that did 90% of the work-Romana was a post graduate research associate and Lindsey was an undergraduate student) sought out to answer what seemed a simple question at the time: Does the CGG KI mouse model of the fragile X premutation show similar learning-related increases in Arc protein as seen in wildtype mice, or do they show dysregulation as seen in the Fmr1 KO mouse model of fragile X syndrome?.
The reason this is an important question is that the CGG KI mice show a 20% reduction in Fmrp levels when compared with wildtype littermates, whereas the Fmr1 KO mice lack Fmrp entirely. Looking at how Fmrp and Arc interact in mice with intermediate levels of Fmrp (rather than 100% of wildtype or 0% in Fmr1 KO mice) may provide critical information that allows researchers to more adequately model Fmrp-Arc interactions.
The second question was whether differences in Fmrp or Arc protein levels would explain the impaired performance seen in the CGG KI mouse on skilled forelimb reaching tasks, as previously demonstrated by another student of mine Link.
Our Innovative Approach
The earlier experiment done by Amanda Diep I mentioned above demonstrated that CGG KI mice show a CGG repeat length dependent impairment in skilled forelimb reaching acquisition. They also show impaired asymptotic levels of performance. Ongoing research in the lab that has since been published demonstrated clearly that there were CGG repeat dependent reductions in Fmrp levels in these mice Link. So Ramona and Lindsey set out to test if these two sets of data were related.
More to the point, Ramona came up with the general experiment, including the regions of interest to sample (somatosensory and primary motor cortex) and that Fmrp should be assayed by Western Blot-a method she was adept at and I have only ever done myself once (and poorly at that).
At this point, this was clearly her experiment, not mine. I sat down with Lindsey and asked if she wanted to be involved with the project, and if she could identify a control protein that we could assay in parallel with Fmrp. She did me one better, she presented me a number of John Guzowski’s papers looking at Arc immediate early gene expression as a marker of learning and asked if we could look at Arc protein, using c-Fos protein as a general activity marker using Western Blot. I told her to speak with Ramona, set up the experiments, and go have fun.
They apparently did have fun.
They designed the task, chose the dependent measures, and notified me once they had finished. They chose to look at Fmrp, Arc, and c-Fos levels in the somatosensory and motor cortices contralateral to the paw used to reach. They chose to use the ipsilateral cortex as a within-animal control. They also chose to use mice placed in the apparatus with reward pellets, but no reaching requirement as a control (in case the reward had something to do with expression levels), as well as untrained home cage control mice. They also chose Western Blotting with Gapdh as a loading control, as the lab typically used Actin, but since Actin interacts with Arc it was deemed a poor control for this experiment. They also used Ponceau Red on a number of blots to verify within-experiment consistency.
One day I received an email from Ramona asking if I would run the skilled forelimb reaching task and once the mice learned, to run one last trial and bring the mice to them for tissue sampling 1 hour later. Like any good mentor, I did what my students asked me to do.
Long story short, we replicated Amanda Diep’s earlier results, with the CGG KI mice taking 2 extra days to learn the skilled forelimb reaching than wildtype littermates. We also saw poorer performance at the asymptotic phase of the task. What was intriguing was that we saw a relatively strong relationship between Fmrp and Arc expression levels and performance in both wildtype and CGG KI mice. What made this so interesting was that it has previously been shown that Arc and Fmrp levels are associated with learning and memory performance and synaptic activity. What made this exciting to me was that c-Fos levels (an important control as c-Fos mRNA does not interact with Fmrp) did not show any relationship with performance.
Based upon Ramona and Lindsey’s data, it appears that the Fmrp and Arc effects we observed may be associated with learning, whereas c-Fos is associated with cellular activity, replicating a decade of previous research. Given that the CGG KI mice show impaired performance for skilled reaching, and have reduced Fmrp and Arc protein expression levels, it is not a stretch to assume that there may be something going on at the level of Fmrp and Arc interacting. However, that was beyond the scope of our experiment and remains unanswered.
Take Home Message
TL;DR, Arc and Fmrp interact in the cell, as has been demonstrated in wildtype mice. However, the Fmrp-Arc dynamics have to date been modeled using only the endpoints of the relationship (i.e., 100% Fmrp in wildtype mice vs. 0% Fmrp in the Fmr1 KO mouse). The work of my students has suggested that when one looks at intermediate Fmrp levels, this relationship is not as clearly linear as previously assumed. Further research is necessary to fully characterize this relationship.
Also, the most important TL;DR: undergraduate research assistants ROCK! Often in my experience they can design experiments better than I can-so I have learned to let them!