When Heike Schuler joined the laboratory of Rosemary Bagot, PhD, she was interested in studying the neurobiology of depression. More specifically, she wanted to use mouse models to understand the predisposition of developing depression or anxiety-like symptoms at the gene expression and transcription level.
“But very early on, I hit a wall when we included female mice in our samples. Ultimately, the rodent models of depression didn’t work equally well for male and female mice,” says Schuler.
Historically and up until recent years, most of the animal models of neuropsychiatric conditions were performed exclusively in male mice. This was largely based on the misconception that female rodents would be too difficult to study due to their hormonal cycles. This led Heike to take an unexpected trajectory with her PhD research to work on developing protocols that allowed her to study both male and female typical phenotypes. She just finished her doctorate degree, and we were very glad that she made the time to sit down with us and talk about the importance of her findings and her work before taking the next step in her career.
How did your protocol differ with what was previously used in models of depression?
One of the widely used protocols for studying depression is a chronic stress model, which is based on territorial aggression in rodents. There's a lot of difficulty implementing that in female mice because they simply do not express this behaviour.
For my PhD, I studied social and exploratory interactions at baseline and after stress to then try and understand how both male and female mice reacted to chronic stress protocols. What we found was that male mice tend to display social avoidance, and are in general less likely to explore the test environment. In contrast, female mice seem to become hyperactive: they tend to run around more, travel longer distances and move faster than mice who have not been subjected to stress.
Were you only using chronic stress in your behavioural tasks?
I also wanted to look at the processing of emotional, or valenced, stimuli. People with depression tend to have a negativity biases, where they perceive something neutral or positive as more negative than people without depression. In other cases, we might see anhedonia, which is the inability to perceive something as pleasurable. For my PhD, I wanted to look at the difference in valence discrimination in male and female mice.
Very rarely in life is everything around us is either negative or positive; we wanted to include different experiences for the mice during the task to reflect some variations in experience. We incorporated three distinct tones, each presented before a negative (shock), a reward (chocolate milk) and a neutral (no outcome) stimuli. This brought more complexity to the environment and the associations that the animal had to learn to provide a model that's a bit closer to reality.
Why was it important to study stimuli valence in male and female mice?
When we used fear conditioning alone, we saw that female mice displayed generalized freezing, and didn’t move in the environment. On the other hand, male mice froze to the tone associated with the shock; however in the absence of the shock predicting stimuli, they were still exploring more than the females. In the past, this has been interpreted as the females inability to discriminate between a negative cue and a neutral cue as well as the males. Using this new approach where the environment itself did not carry a single valence, we saw less bias in the behavior between male and female mice, and could prove that females are equally able to discriminate between a negative and a neutral cue. We were able to demonstrate this by showing that at baseline, they don’t explore as much as male mice in a new environment.
Do you think we’re in a better position now to find these differences in behavioural trends between the sexes?
Decades ago when behavioural neuroscience experiments had started, researchers relied on recording and observation alone. These approaches leave a lot of room for debate. Researchers were trying to put themselves in the position of the mouse to try and understand the behaviour. There is a widely used example of the forced swim test: researchers placed a mouse in a pool of water in which there's no way for it to escape. Eventually, the mice stopped trying to swim and floated. This was interpreted as learned helplessness, until someone proposed that the behaviour may be adaptive. In the last decade, there’s been enormous stride with respect to the technology available to us as scientists. In our lab, we now analyze the behavioural readouts using machine learning paradigms. This enables us to take a less biased approach, and to capture the whole spectrum of behaviours displayed by the animals without trying to anthropomorphize.
What do you think is the next step to build on your research?
There are a lot of possibilities to look into sexually dimorphic brain mechanisms that are contributing to the behaviour to better understand the effect of stress in female mice. In the absence of strong stress phenotypes, researchers used to either assume there was nothing to see, or just exclude females from their neurobiological essays. Now that we have a model that can include broader phenotypically relevant behaviour, we can better understand what's going on in female mice. We think there's going to be different neurobiological mechanisms for males and female at the underlying molecular level. This still needs to be studied, and there's going to be interesting work coming out of the Bagot lab in the future.
What’s your next step in your career?
I want to pivot from academic research towards working in the industry. During my PhD, I really enjoyed the bioinformatics aspect of my research, working with large data sets, and building statistical and machine learning paradigms. Prior to McGill, I completed my Masters in neuroscience with a bioinformatics track at Utrecht University in the Netherlands. Through my trajectory, I was very lucky to have a strong statistics background. During my time at McGill, I learned to code and implement bioinformatics to my data and research. This all merged as a good foundation for my PhD. Ideally, I would like to pursue opportunities as a biostatistician or bioinformatician.
What advice do you have for an aspiring student that wants to do a PhD in neuroscience?
It requires a lot of commitment and self-initiation to do a PhD. My advice is to think about what you're really interested in, and what about doing research and science brings you joy. It's much more fun if you're asking the questions that you're truly interested in.
