cytocentric visionaries sjeb01

Do You Know? Oxygen and the Sex of Your Cultured Cells

Sausan Jaber and Evan Bordt are co-first-authors of a recent publication, “Sex differences in the mitochondrial bioenergetics of astrocytes but not microglia at a physiologically relevant brain oxygen tension”, in Neurochemistry International [1]. To control atmospheric conditions for their cells, they used an Xvivo System made by BioSpherix for research.
Sausan Jaber is working on a PhD at the University of Maryland, Baltimore and Evan Bordt recently started a post-doc at Massachusetts General Hospital and Harvard Medical School. Here, Dr. Alicia Henn interviews these young researchers about their recent paper. The conversation was edited for length.

Alicia Henn: What made you want to look at these female/male differences in glial cells in the first place?
Sausan Jaber: Glial cells, especially astrocytes, play many important roles in the brain although they are often overlooked. Astrocytes provide support for neuronal survival and recent literature shows that they can be just as important for brain function as neurons. There are documented sex differences in astrocytes that have been observed prior to birth and hormonal changes occurring during development.
Evan Bordt: Glial cells are incredibly important support cells in addition to all of their other functions. For example, microglia and astrocytes both prune synapses to ensure proper neuronal functioning. There are alterations in both glial reactivity and mitochondrial function in brain developmental disorders, diseases, and brain traumas. There’s some great literature showing that female mitochondria are more resilient to brain injury than male mitochondria. There are also known sex differences in microglial cytokine production.

AH: Microglia function sort of like macrophages?
EB: The predominant view has been that they’re simply the macrophages of the central nervous system. However, there’s new literature suggesting that they are very much controlled by their environment and that they do act somewhat differently from macrophages.
There’s a growing consensus in the microglial field that they show large differences between in vivo function and in vitro function. We were wondering why that could be. Doing everything that we can to make the in vitro setting more physiologically relevant and then directly comparing that to an in vivo setting is our ultimate goal.


AH: What are the most important things about this paper?
SJ: The oxygen tension at which experiments are being run should always be considered as an experimental variable. There are often cases noted in the literature where there are significant differences between in vivo and in vitro models and there’s no explanation as to why. Oxygen as a variable should especially be considered in those situations.
Physiological oxygen is incredibly important, especially when looking at processes that directly involve oxygen, such as mitochondrial bioenergetics and reactive oxygen species generation. Tissue oxygenation is between 1 to 6% in the brain and that’s very far from 21%, or atmospheric oxygen.
EB: Since NIH changed the funding guidelines so that researchers have to include both sexes, the study of sex differences is becoming far more common. However, it still seems like in vitro research is lagging behind in vivo experimentation in this regard. When people think of cells they don’t often think of the entire organism’s sex.
With primary cultures or ex vivo preps you can easily control for the sex of the animal. Even in cell lines, which you definitely don’t think of as being sex specific, you can determine the sex. So, in addition to the obvious importance of controlling for oxygen, controlling for the sex of the cells is incredibly important to understanding their biology.

AH: You said something important there. It’s rare to think about the entire organism when working with cell lines?

EB: Yes and in addition, when we think of in vitro research we normally tend to think of the fact that we’re using cultures of neurons. However, we don’t necessarily think of what brain region the neurons came from or how those neurons are interacting with other brain cells or even other cells that happen to get into the brain.
It’s very important when dealing with a single cell preparation to remember that in the actual brain they’re going to be interacting with other cells and a different environment. That is what we’re trying to get at with this paper using different oxygen tensions. There are important variables to consider beyond just an isolated cell type.

AH: Do you think that’s part of an overall reductionist approach to science in general? The first thing we do is try to isolate just one cell type and ask one question about it.
SJ: You have to start somewhere. The thing is that there are two sides to that coin. Basic researchers look at the microscopic scale of things, and clinical researchers are doing the exact opposite. There has to be a middle ground where we start looking at not just one cell, but its environment.
EB: Sausan hit on something incredibly important there. The two of us, as basic researchers, start at the organellar, biochemical, or cellular level. Then the clinicians are coming from the human organismal level, and there’s just not enough interaction between basic science researchers and clinicians. Hopefully bench scientists experimenting in a more physiological environment, such as controlling the oxygen tension, can help start to bridge that gap.

AH: So why isn’t everybody working to control oxygen?
EB: The first main reason is likely a general lack of awareness that there would be an oxygen tension difference in vitro compared to what you experience in vivo. Even if people are aware of that, there’s a lack of knowledge about how you can actually perform your experiments at controlled oxygen.
One of our collaborators’ electrophysiological preps involves bubbling oxygen throughout a brain slice, but they don’t quantify the level of oxygen at all in those preps and it’s something that might change the cellular function. When I’ve talked to electrophysiologists, they mention that if you don’t perfuse slices with oxygen, then the tissues are going to die, but they are definitely worried about the levels of oxygen that they’re pumping through their system. So, we need a better ability to measure oxygen during experiments.


AH: In your paper, you wrote that you had to seed the cells at different numbers in order to get the comparable cell density at the time of analysis. The cells grew at different rates at different oxygen levels?
SJ: For the astrocytes yes, definitely. At low oxygen, they exhibited faster proliferation.
EB: And for the microglia as well. They grow a little bit slower at the lower oxygen levels. It’s funny. In terms of microglia in the brain, whenever there’s an insult or they’re trying to get rid of a pathogen, they will proliferate and move towards the injury site. We think of them as being less perturbed at a lower oxygen tension, so growing slower actually makes more sense with what they would be doing in vivo.
SJ: Truthfully, I couldn’t even say that they proliferate faster because we never checked to see if perhaps they just simply survived better.
EB: True.


AH: You got a different yield.
SJ: We’re not exactly sure why though we did correct our oxygen consumption measurements for protein.


AH: Have you compared unbroken vs. broken physiologic oxygen?
EB: We’ve done a little bit with it. Mainly we’ll culture cells at a lower oxygen and then place them in a Seahorse mitochondrial analyzer at 21% oxygen, or the reverse, and it seems that following acute switching the cells look rather different morphologically.


AH: They get stressed!
EB: Yes, exactly. Oxygen changes are an acute stressor.
SJ: There are papers out in the literature where researchers simply go from atmospheric to a physiological level between 1 and 3% O2 and they consider that a hypoxic stimulus. Just the stress of moving cells from one oxygen tension to the other was sufficient to see a response. We chose 3% oxygen for our experiments, which turned out to be 1.8% oxygen at the cellular level in astrocyte cultures. But the range in the brain is 1 to 6% oxygen, so people should look at the full range and not just one or an average. Even one or two percent likely can make a difference within the physiological range.


AH: How reliable are in vivo measurements that we use as guidelines to set our in vitro oxygen ranges?
EB: The initial experiments measuring oxygen tensions in brain were performed using microelectrodes. These demonstrated that the overall oxygen levels in vivo were below the atmospheric room air oxygen. But excitingly, in the past few years, two-photon phosphorescence lifetime microscopy has revealed a great deal of heterogeneity in the oxygen tension levels in different brain regions.
So the next steps that the field needs, in order to accurately model physiology in vitro, would be to actually determine the different oxygen tensions, not only in different brain regions, but possibly within different cells in those different brain regions.
I don’t think we would be surprised if different cells within the same brain region are exposed to different tissue oxygenation, possibly due to different cell types consuming oxygen at different rates. Once we get that optimal oxygen tension for whatever cells we’re studying then we need to move forward with that oxygen tension for this cell type from that particular region.

AH: Exciting times for closing that gap between in vitro and in vivo environments! Thank you for talking with me today. I’ll be looking for more of your papers soon.


1. Jaber, S.M., Bordt, E.A. et al., Sex differences in the mitochondrial bioenergetics of astrocytes but not microglia at a physiologically relevant brain oxygen tension. Neurochem Int, 2017.