Episode 23: Nathalie Rochefort, PhD

The following interview was conducted in-class, during the Spring 2024 session of Hidden Figures: Brain Science through Diversity, taught by Dr. Adema Ribic at the University of Virginia. What follows is an edited transcript of the interview, transcribed by Samantha Cecilia Anderson, Jocelyn Truc-Lam Pham, Vyshnavi Sistla, Olivia Paige Chapman, Salma Meraj, George Childress Quillen, Alex Liff and Elina Rastegar, who also drafted Dr. Rochefort’s biography. The final editing was by Dr. Adema Ribic.

Dr. Nathalie Rochefort received her Bachelor of Science in Biology and Epistemology at the Paris Diderot University and the Ecole Normale Superieure in Paris. Then, she obtained a PhD in Neuroscience in 2007, from the Sorbonne University and the Ruhr-Universität-Bochum in Germany. She completed her post-doctoral training at the Institute of Neuroscience at the Technical University of Munich in 2013. Shortly after, she joined the University of Edinburgh to establish her research group. Currently, Dr. Rochefort is a Professor of Neuroscience at the University of Edinburgh where her lab studies how brain neuronal networks process visual information in health and disease.

How did you get into neuroscience? Have you always been interested in pursuing science in your studies and as a career?

I went to study biology at University, but I had a strong focus on math and physics in high school. Neuroscience was a way to bridge biology, technology, and informatics. I had my main breakthrough during my Masters program, where I had many occasions to go to different labs. For example, I studied the locomotion of the locust. I also spent one month in the lab of Dr. Eric Kandel, where I studied mechanisms underlying LTP. These experiences made me begin to think about pursuing a career in research, specifically in neuroscience. However, even with these thoughts, it still took me quite a long time to realize that being a researcher could be my career and that I would be able to lead a lab.

What led you to pursue the degree in epistemology and how it correlates the neuroscience research that you’re doing currently?

I did very straight forward studies from high school through University without any break. After University, I wasn’t sure that I was ready to embark on a PhD, as I could not do it without a scholarship. I pursued epistemology because of the project required for my masters – you could either write a thesis or be involved with a company that was linked to your work. I worked for the French version of the Scientific American. I was a journalist during my Masters, which helped finance my masters program.

How did you like working as a journalist?

I enjoyed it a lot. I was able to contribute to a TV show about science, as well as contribute to a special edition about mathematicians. The magazine was focused on famous mathematicians in the 1930’s in France. I had the mission as a young trainee to go to visit these old mathematicians to collect pictures and anecdotes about their career where they revolutionized mathematics at that time. I questioned them about their work and recognized how much passion they had. After 3 or 4 interviews, I realized that I wanted to be them, not the person interviewing them. In other words, I wanted to produce science, not write about science or scientists.

How did this realization affect you in your studies and effort to finish your masters degree?

I still had to write an actual Master thesis, and I chose to write about the topic of color vision. After starting writing my thesis, I realized how much I enjoyed reading about neuroscience, which was quite a lonely activity in the library (writing and reading). I liked the idea of collective work after my experience working in labs and knew I wanted to choose a job that was collective.

Knowing you wanted to produce science, and work with a team in a lab setting, how did the rest of your education pan out?

After I received the Master for epistemology, I decided I wanted to do research. However, I still needed the money to pursue the PhD, so I knew that I needed to get a scholarship. At the time, the only chance that I had to get a scholarship was to do a second Master’s. At the end of this program, I would compete for the scholarship to fund my PhD. Upon this realization, I chose to embark on a second Master which got funded through a scholarship. Following that, I had to compete for the PhD scholarship, which I did not get. Even though I did not get that scholarship, a professor I had as an undergraduate told me about another European program that was looking for students. There were not as many applicants, so I applied, and got the fellowship. My PhD ended up being a collaborative project between labs in Germany and France.

What pivotal experiences, whether through coursework or research, shaped your decision to pursue visual neuroscience?

I was attracted to sensory neuroscience early on. It was one of the first things I had learned about in my undergraduate courses, specifically sensory neurophysiology. When I discovered that there was research in mammals, like the recordings in the brain of anesthetized cats, I felt very attracted to this type of research, which was pursued at my university. I didn’t plan to stay in the visual cortex. It just so happened that I moved from the cat visual cortex to the mouse visual cortex and kept going. For me, there’s an attraction to be as close as possible to the human brain. It was this attraction that led me to study the history of color vision for my masters in epistemology.

What aspects of your work at the Rochefort lab do you find most fulfilling, and how do you typically spend your time there?

One thing I like about this job is the diversity of the tasks. Every day is different and you can shape your job in an academic career to the things you like the most. There are researchers that like writing articles and others that like the technical aspect of the experiments. Other researchers like programming, teaching, mentoring, and giving talks. Leading a research group requires multitasking. Personally, I like to meet with people in my lab and discuss progress reports about our research. I love to see data and I am someone that is inspired by them. I always have one direction I’d like to go in, as a starting point, but I like to build the hypothesis upon the data and research. I teach a little bit, but I spend more time mentoring young researchers which I really enjoy.

What does your lab study now and which techniques do you use?

Our lab is interested in knowing how neuronal circuits underlie our visual perceptions and actions. And to understand that, we are using techniques that enable us to record the activity of large populations of neurons in awake and behaving animals. We are using the mouse primary visual cortex as a model system of a cortical circuit integrating sensory information, with non sensory contextual inputs. We are investigating  how our perceptions are influenced by our experiences and behavioral state.

How are you studying the integration of sensory and contextual information within the visual cortex?

You need a living and awake animal in which you can record populations of neurons while the animal is doing something. That’s why we’re using recordings in awake mice that we place in a virtual reality environment where we have visual stimuli with which they can interact with. The display of visual stimuli is synchronized with their movements, and we can teach them to find rewards in virtual reality and track, for example, the neurons before and after the learning of a task.

What are some of the more recent studies that came out from your lab?

We recently studied the effects of long-term food restriction on visual processing. We restricted the food given to the mice such that they lose 15% of their body weight for a period of 2-3 weeks. It’s a long period of food restriction for a mouse. We’re interested in understanding how the activity of the neurons adapted to the “low power regime”, and how this impacted the function of these neurons. For this, we used the visual cortex because it is a very convenient region to test a cortical function - you show a stimulus, you record activity, and you can get very clear measures of how precise and specific the response of the neurons is in relation to the different visual stimuli.

Does your lab study clinically-related areas, like atypical sensory perception in neurodevelopmental disorders?

My lab is part of the Center for Discovery Brain Sciences and of the Simons Initiative for the Developing Brain (SIDB). This institute (SIDB) funds research related to intellectual disabilities and autism spectrum disorders. One part of my lab is studying the non-stereotypical activities in the visual system of animal (mouse) models of intellectual disabilities and autistic spectrum disorders.

Is there a specific gene/protein/factor of interest that you are studying in these models?

We are focusing on the condition that is linked to single gene mutations in the gene, coding for the SynGAP protein which is a protein downstream of NMDA receptors. We found a deficit in visual discrimination in these SYNGAP mice, and we now want to know what are the mechanisms underlying this impairment.

What is it like to collaborate with and work for one of these projects in the Simons Initiative for the Developing Brain?

The big benefit of this institute is that we are working together with experts in other research fields. For example, we have the chance to work with psychiatrists who are directly in contact with patients who carry SynGAP mutations. They’re the ones we report our results to regularly, and they give us suggestions. For example, we’re currently testing a drug in our preclinical model (the mouse model) to support an application for a clinical trial.

Given the complexity of the neural circuits involved in visual processing and behavior, how do you identify specific neuronal populations?

Being able to identify the specific function of different neuronal populations, came both with the development of new techniques to probe neuronal function and the big advantage of genetic tools developed in mice. Why do people use mice? They’re small. They fit under a microscope easily. They breed well. It’s relatively cheap compared to other species. But mostly because you can use a lot of genetic tools. You can express specific markers in specific subpopulations of neurons: if you wanted to target a population of inhibitory neurons, you could genetically label that population with, for example, an activity marker (e.g. a calcium indicator)

After genetically tagging a subpopulation of neurons, we are then able to record and manipulate their activity. Optogenetics is a name for optical methods that allow you to manipulate the activity of neurons – either silence or activate them – by expressing a specific protein, such as channelrhodopsin, for example. You can express such light-sensitive protein in your neuronal subpopulation of interest and manipulate specifically the activity of these neurons.

What type of methods did you use/what were your research interests during your PhD?

During my PhD, I was studying the cat’s visual cortex by combining anatomical tracing of long-range callosal connections with optical imaging of intrinsic signals.. I had the Hubel and Weisel preparation where the cat was anesthetized and paralyzed such that there was no movement, no change in arousal, just stimulus-response association. The cat is placed in front of a computer screen where we show the stimuli. Then, we record the activity of neurons in the cat’s visual cortex and correlate the neuronal activity with the stimuli that we were showing on the screen. I was using optical imaging of intrinsic signals to determine the orientation selectivity of neurons connected by inter-hemispheric connections through the corpus callosum. In this method, the source of the signal is the change in light absorption from blood. The surface of the cat brain is illuminated with an orange light, that is differently absorbed by deoxygenated and oxygenated hemoglobin. The regions activated by a given stimulus will take more oxygen from the blood, leaving a dark blob where the neurons were active. Then, you have beautiful maps where you could see from the surface of the cat’s brain the domains of neurons that were activated for a visual stimulus of a given orientation. These domains correspond to orientation cortical columns. But the spatial resolution of this approach was poor ( about 50 micrometers), far from single cell resolution.

Knowing the size of a neuron, I can imagine this posed as a challenge. How did you navigate this resolution challenge? Where did you go from here?

 After studying the cat visual cortex, I moved on to study the mouse visual cortex. During this time, the way of accessing single-cell coding information was to combine the imaging of the big blobs of activity with an injection of anatomical tracer. As you can imagine, these were really demanding experiments. These blobs corresponded to orientation columns in which neurons would globally respond to an orientation of the stimulus. You would inject an anatomical tracer into one of the blobs, sacrifice the animal, cut the brain, reconstruct the axons. This became a problem for me during my PhD when I studied how the connections between the visual cortices of the left and the right hemisphere fuse the two visual fields to form one coherent field. We would inject the tracer in one hemisphere, reconstruct the axonal connection in the other hemisphere, and overlap these reconstructions with the functional maps to try to learn about single neuron coding. But, of course, we didn’t know whether the axon output would be functional or if they were connecting inhibitory neurons or excitatory neurons which would make a big difference. While I loved my work, I had a fundamental frustration in terms of knowledge because there were a lot of limitations due to the technique itself. The two-photon imaging method that I helped establish during my post-doc was a big shift, since it did allow the investigation of large population of neuronal activity at single cell resolution. .

Can you elaborate a little bit more about your contributions to two-photon calcium imaging?

Part of my post-doc was to develop two-photon calcium imaging with a big aim that we would be able to finally see individual neurons active in an in-vivo preparation during visual stimulation. This was done using anesthetized mice because at that time, head-fixed, awake recordings were not yet developed..

What are some of the advantages and limitations that you’ve found with any of the methods that you currently use in the lab and how have you learned to overcome those humps?

The limitations of calcium imaging are very clear. This is an indirect measure of neuronal activity. You’re using calcium, which is only indirectly related to the spiking activity. The intracellular calcium concentration is tightly linked to the spiking activity: when there is a spike, you have a huge entry of calcium and monitor this entry of calcium by using fluorescent indicators that will change their fluorescence depending on the change of intracellular calcium concentration. This is an indirect measure of spiking activity; it is not directly measuring a change in voltage (as in electrophysiological methods). Another limitation is that the relationship between the spiking activity and the calcium concentration may not be linear, and strongly depends on the calcium indicator. Depending on the research question, this relationship would have to be calibrated for different cell types, in different conditions (e.g. behavioral states) which is practically difficult.

How does imaging compare to electrophysiology?

Even the fast ones [calcium indicators] are working in the range of tens of milliseconds, usually hundreds of milliseconds, with decay times that are in the range of seconds. We are far from the millisecond resolution of firing that you would get with electrophysiology. However, with electrophysiology, you don’t actually see your neurons. It is quite hard to follow the exact same neurons across time. You are never sure, because you can’t see them (like with imaging).For some applications -for example, in my research on dendritic and spine imaging- the lack of temporal resolution of calcium indicators has been very frustrating. The way to go beyond that would be to use voltage-sensitive dyes, but this also comes with limitations.

For your research on mouse vision, have you faced other limitations and/or challenges?

Another limitation for me right now is that all the recent work in the labs is in head fixed mice. I am interested in knowing how the brain integrates different sources of information -sensory, motor, experience-related, metabolic, etc. In that case, you want to have the most physiological condition. Being head fixed under a microscope, one could argue that this is maybe not the most physiological condition. The animal cannot turn their head; orienting yourself in virtual reality is different from working in a room. Ideally in the next step, I would like to have the same resolution, the same control of inputs that I have with head fixed mice, but in freely moving animals.

You mentioned how you like collaborating with people in the lab and how you wanted to be the one that does the science, not the one that reports about it. Could we briefly discuss your early life and how this mentality came to be?

I went to a local state school as a child, and this was a school where the majority of children were first- or second-generation immigrants. It was called a priority school, which means that we had some funding for several activities to develop education. This funding came with very dedicated teachers who genuinely believed in their students, as well as provided many opportunities. Very early on, I sensed how it was important to be mentored.

Were there any specific events in your life that you could reference?

As a child, I wanted to be an astronaut – an astrophysicist. When I was eleven years old, children in Paris could write a letter explaining the career they wanted to do. I decided to write one of these letters and was selected by the school to spend a week at the space camp with astronauts, in the South of France, with many children, doing the training of the astronauts. I’m mentioning this because these are small initiatives, but make a huge difference for a child. I also realized this importance because in high school, I ended up with a much more sociologically homogeneous, more conventional high school. Despite the fact that I liked math, most of the girls, me included, were advised to do medicine and biology. At that time, I would have always told you it was my choice and not a social pressure. You realize these biases only in retrospect.

Do you enjoy mentoring today?

I feel a huge responsibility to give back. I have set up a “Women of Science” initiative, where I ask the guest speakers to spend an hour speaking about her career, as well as the barriers and the support she has received.

We see these barriers through gender, race, etc., but one that we don’t really talk about is language. I know that you’re a French native and fluent in French, so I was wondering if this has ever been an obstacle in your career?

It's a huge point, and it's something that’s completely ignored. To give you an idea, English was not taught so well in France. We could write okay, but we really couldn’t speak. Up until my masters, I had very few occasions to just present something in front of the class. Imagine during my PhD when someone told me I had to give a talk? I was paralyzed. I turned completely red and couldn’t speak, and on top of it in English! To read an article was very hard, and it took me much more time, than a native English speaker, to write and to speak as well. So, yes, it is a barrier, but getting good at this does come with practice.

What advice would you give to those that are currently experiencing these barriers today?

For all the people that are worried about freezing while giving public talks, just practice. You can train for that. If you really prepare your talk, then you will know your sentences. For me, the main trick is to prepare the first couple sentences so at least I know the first few. However, when it came to question time, I was so stressed and I still am. For many years, I simply couldn’t understand the content of the questions. Also, writing takes more time when it’s not your native language. That said, you learn.

Have you seen that the language barrier has improved from decades ago to now? Are the perceptions against people who speak different languages more accepting now?

People are more trained everywhere in the world. As a child, I was not exposed to English except one or two hours per week in secondary school with a recording tape. Now with the internet and huge flow of information, someone who wants to learn English has the opportunity to learn much better and faster. The awareness and tolerance for non-native English speakers is very variable. People who come from countries in which English is not the main language are usually much more open, speak slower, and are more aware. People who grew up in countries in which English is the official language and aren’t exposed to people speaking another language are maybe less aware. For example, when I'm talking to someone and there’s something that seems off in what this person said, one of my first thoughts is that it was lost in translation and is a misunderstanding. But sometimes people don’t think like that. Instead, they take the words for the words and don’t have the thought of “oh maybe that’s not what this person meant”.

How many languages do you speak?

Four. In order of how I learned them: French, English, Spanish, and German.

Could you tell us about your earlier work on spontaneous activity in the visual cortex?

The aim of the research was to study how the activity of neurons in the visual cortex changes before and after the mice open their eyes. Mice are born with their eyelids closed and open their eyes two weeks after their birth. This is the time when they start to explore the outside world. We wanted to know how the activity of visual cortex neurons would change when recorded before and after their eyes opened. This study was rejected three times: twice after two rounds of revision. The technique we used was not accepted by many in the community. There was no benchmark to determine if the data were consistent with previous findings. The paper was finally published in 2009 through the PNAS because it was sponsored by Dr. Bert Sakmann, a member of the National Academy of Science. What is spectacular is that this study has now been cited around 300 times, which is a large amount in the field of neuroscience. The research is still inspiring new projects. Considering its rough path to publication, it is a huge reward to see that the paper continues to inspire people even 15 years later.

What are some of the favorite findings from your lab?

Our lab’s first paper was on the behavioral state-dependent activity of different classes of inhibitory neurons in the mouse visual cortex. This was the first study demonstrating that the response of inhibitory neurons to locomotion is context-dependent, varying between light and dark conditions. This was important because it showed that the response of a neuron to a behavior variable in one context cannot be generalized to other contexts.

Can you discuss the significance of your recent work on the impact of food restriction on normal activity, and why it garnered significant public attention?

My recent work about the impact of food restriction on neuronal activity received a lot of attention. For me, this was the first experience where the public cared about my basic science. It was featured in the Quanta magazine. Since that publication, we received a lot of emails and attention from many different people that were interested in how intermittent fasting would or would not impair their vision, which was a bit of a misunderstanding of the article, since our study is based on long-term food restriction. It was still nice to be able to speak about this research and have these interactions with a broader audience.

Apart from the actual research, how do you feel about the personal and professional development of the individuals in your lab?

As a PI, one of the most exciting and rewarding experience is to see the development of the people in your lab. Some are very young when they join the lab (for example as undergraduate students) , and later on you see them attaining their PhDs, starting their own labs, or going into industry. I remember seeing one of my previous PhD student giving an amazing talk as a senior post-doc to a huge audience at a conference,: being part of this development is extremely rewarding..