Welcome to a special year-end Brain Bits! Today I’ll discuss some of my favorite papers of the year.

*As always, “best of” = random papers I found interesting. Apologies to the authors of all the other awesome 2019 papers not mentioned here!

 

At long last, the sour taste receptor

I’ll start with a story near and dear to my heart: the long-awaited discovery of the sour taste receptor. As you may know, my own work focuses on how our brains perceive taste and decide how we should react. (Shameless plug: my latest paper was even about vinegar, a canonically sour food, and how fruit flies switch from eating it when they’re hungry to avoiding it when they’re full.)

Studies conducted over a decade ago identified the taste receptors on the tongue that detect other basic tastes such as sweet, savory, and bitter. But the receptor that detects sour taste has remained elusive, with several false leads over the years.

Then in 2018, a study from Emily Liman’s lab at USC discovered a new candidate receptor, Otop1, that enabled sour-sensing taste cells to be activated by acids (the type of chemical that creates sour taste). This year her group published a follow-up paper confirming that when Otop1 is mutated in mice, sour taste signals are absent from taste cells as well as the nerves that send taste signals to the brain.

 

A separate study from Charles Zuker’s lab at Columbia independently discovered Otop1 as the sour taste receptor and showed that sour signals are transmitted to dedicated sour-sensing cells in the brainstem. These studies finally reveal the molecular basis for sour taste, the taste that tells us super important things like when fruit is unripe and when milk’s gone bad.

 

Why we sleep?

We spend one-third of our lives sleeping, but why we (and all animals) even need to sleep in the first place has remained a biological mystery. A study from Gero Miesenbock’s lab at Oxford investigated this question in fruit flies, taking advantage of the fact that a small group of sleep-controlling neurons had previously been identified. These neurons are electrically active during sleep and silent during wakefulness, and changing their activity can force a fly to go to sleep or wake up.

The authors therefore asked what regulates the activity of these sleep-controlling neurons, reasoning that whatever triggers their activity might be what makes the fly need to sleep in the first place. Building on previous studies that identified the ion channels controlling the activity of these neurons, the authors found that these channels sense oxidative stress.

Oxidative stress is a fancy term for when potentially toxic chemicals containing oxygen (called reactive oxygen species, or ROS) build up and start damaging your cells. These chemicals are natural byproducts of metabolism so they are constantly being generated, but they need to be detoxified.

 

The authors show that reactive oxygen species build up during sleep deprivation and their buildup triggers sleep. Moreover, the paper elegantly elucidates the molecular and cellular mechanisms that link metabolism, oxidative stress, and the activity of sleep-controlling cells. The idea that sleep evolved as a way to control oxidative damage is a clear implication that now awaits testing in other animals.

 

Glia jump into the decision-making spotlight

How does our brain make decisions, such as deciding when to give up on something you’ve been vigorously working toward? A study from Misha Ahrens’ lab in at the Janelia Research Campus used zebrafish as a model to study this decision. The authors imaged the entire zebrafish brain while the fish was swimming in a virtual reality.

The fish keeps swimming as long as it receives visual feedback indicating that it’s making progress, but once that feedback is removed the fish eventually gives up. It’s like playing one of those racing video games like Mario Kart (umm, does this reveal my age?)—you keep racing because the screen shows you’re moving forward, but if the image were to stay stuck in one place matter how hard you pressed the button then it’d be stupid to keep playing.

Other studies have found neurons that integrate information over time to determine when to make a decision and what decision to make. These are pretty fancy neurons that usually reside in “higher” parts of the brain, such as parietal or frontal cortex. So I was very surprised to read that in the zebrafish study, the cells that integrate information to decide when to quit aren’t even neurons at all—they’re glia!

Glia are traditionally considered the “support cells” of the brain. They keep neurons happy and healthy, but they’re not thought to actually play a role in processing information. The authors found that calcium levels in a subset of glia called radial astrocytes accumulated every time the fish tried to swim forward but failed to make progress. Activating these astrocytes reduced swimming, whereas killing the astrocytes or reducing their calcium levels encouraged fish to keep swimming instead of giving up.

 

Through many more experiments, the authors show which neurons the radial astrocytes receive information from and who they send information to. But the highlight of this study is really the discovery that glia can perform computations that support decision-making, a pretty sophisticated behavior.

 

And more…

Here are a few more cool papers for you guys to check out:

“A deep learning framework for neuroscience”, a collaborative review by many big names in systems and theoretical neuroscience. This review suggests that the theories underlying deep learning and artificial neural networks should influence our experimental studies of the brain. While this article generated a lot of Twitter controversy, with some scientists skeptical or offended by a bunch of theorists telling experimentalists how to do things, I found many of the ideas to be interesting. For example, I liked the idea that we should be placing more emphasis on understanding the overall goals or “objective functions” of the brain rather than reading out whatever individual signals we can find, which may or may not be meaningful.

“Single-cell reconstruction of emerging population activity in an entire developing circuit” from Phillip Keller’s lab at the Janelia Research Campus. The authors developed new imaging methods to track the movement, activity, and identity of all neurons in the zebrafish spinal cord throughout their development. They analyzed how the complex activity patterns observed in mature animals emerge from smaller ensembles of active neurons.

Two papers that showed how small RNAs control transgenerational inheritance of behavior: “Neuronal small RNAs control behavior transgenerationally” and “Piwi/PRG-1 Argonaute and TGF-β mediate transgenerational learned pathogenic avoidance”. Transgenerational inheritance is the idea that our life experiences can be encoded in our DNA or RNA and passed on to our children, thus altering their physiology or behavior. This is a controversial notion that was long considered far-fetched and antithetical to Darwinian principles of evolution. These two studies demonstrate that it isn’t so far-fetched anymore and show the underlying mechanisms of how this works in the worm C. elegans.

“Discrete attractor dynamics underlies persistent activity in the frontal cortex” from Karel Svoboda’s lab at the Janelia Research Campus. Despite having a typo in the title (umm Nature I think you can afford a decent copyeditor!), this was a really cool study showing how short-term memory and decision-making in a simple task rely on attractor networks. Attractor networks are networks whose activity naturally converges onto discrete stable patterns, which can be used by the brain to represent specific memories or actions.

 

What were your favorite neuroscience stories of the year? Leave a comment below!