Short sleep and moving beyond DEC2
Some additional considerations for hacking sleep if DEC2 doesn't work out
I am not a short sleeper. I used to think I was, but it turns out feeling physically awful and fully caffeine dependent (3-4 large cups minimum daily) because I slept 4-6 hours every night from 16-23 is a bad consistent homeostatic state to be in. Maybe you’re one of the blessed few short sleepers, but my guess is you unfortunately aren’t, but if you’re on Twitter you almost certainly have seen the buzz about short sleepers and all the hype associated, conjuring personal visions of all that added efficiency, and I hear you – it’s enticing to want that for yourself. Even today at 28 I go through cycles of sleep deprivation (I’m in one right now, actually) which has me thinking about sleep, so here are some thoughts below, mostly regarding DEC2 and alternative routes to “shorter” sleep. I want to note that these are really more like musings on what I think are promising avenues for hacking sleep in a way that doesn’t rely on all-inning on DEC2. I welcome dissent, though, so feel free to DM or @ me on Twitter, or email me if you disagree.
But first, a word from our sponsor, husband to former First Lady Jacqueline Kennedy, Aristotle Onassis.
What happens when we sleep?
I personally think the foundation is important to cover before jumping into the more complicated aspects that might be helpful for sleep-optimizers, so it’s worth covering some basics about the mechanics of sleep. At its core, sleep is a process of consolidation and repair from the day’s metabolic processes, the details of which are specific to the stages that divide up our sleep into cycles. Our brain relies on a fluid that runs in a channel up and down our spinal cord and into our brains called cerebrospinal fluid (CSF) for most of what drives the restorative process of sleep. Well, the toxin removal parts, anyway.
If you haven’t had the privilege of seeing a skull sawed open with a live brain sitting inside it, this probably seems confusing. Why even have CSF at all? What benefit does this thing provide that makes it special from blood or whatever else our body makes? Essentially, our brains don’t sit directly on top of the skull, but are instead suspended in CSF, which provides basic needs of physics – nutrient delivery through diffusion, heat and homeostatic regulation, physical protection from rattling around, and, as I alluded to above, waste clearance. CSF makes up the fluid that fills what’s called the glymphatics system (this is real).
Now let’s return to how CSF and sleep stages fit together. The stages, put very simply, are light sleep, deep sleep and REM sleep. During deep, or slow-wave, sleep, the glymphatic system becomes more active. Brain cells actually shrink slightly during this phase, creating wider interstitial spaces that allow CSF to flow more efficiently and flush out accumulated metabolic waste products in a process not too unlike rinsing scraps off a dinner plate. This includes proteins like beta-amyloid and tau, which have been implicated in neurodegenerative diseases when they form plaques and tangles.
The stages of sleep are:
Light Sleep (NREM Stages 1 & 2)
This is the entry point to sleep, where your brain begins to slow down but can still be easily roused. During Stage 1, you might experience hypnic jerks – those sudden muscle twitches that sometimes wake you up (happens to me a lot when I drink coffee too late in the day). Stage 2 introduces sleep spindles and K-complexes, which are distinctive brain wave patterns visible on an EEG (I don’t really know how to read these that well but you can find guides online if you’re curious). These features help suppress processing of external stimuli, allowing you to stay asleep despite environmental noise. It’s not quite as important as the next stage, but it still makes up about half of our time in sleep, so it’s worth noting.
Deep Sleep (NREM Stages 3 & 4, now commonly combined)
This is where the magic happens for physical restoration. Your brain generates slow delta waves, heart rate and breathing slow dramatically, and muscle tone decreases. Beyond the increased glymphatic activity I mentioned earlier, deep sleep is when your body releases growth hormone, stimulating tissue repair and cellular regeneration. The immune system gets a boost, and glucose metabolism is regulated. In terms of timing, most deep sleep occurs in the first half of the night, which is why going to bed earlier than normal rather than sleeping past our typical alarm tends to feel more restorative.
REM (Rapid Eye Movement) Sleep
While your body is essentially paralyzed (a condition called atonia to prevent you from acting out your dreams), your brain becomes highly active – nearly as active as when you're awake. This is the primary domain of dreaming. REM sleep plays a crucial role in emotional processing and creative problem-solving. It helps consolidate procedural memories (how to do things) and may facilitate the integration of new information with existing knowledge. For whatever reason, the glymphatic system is actually less active during REM than during deep sleep, suggesting these stages complement each other in function. I’m not a neurologist or neuroscientist though, so I don’t know exactly what the deal is there.
Zooming out on the sleep cycle
These stages aren't experienced just once per night but cycle throughout. A typical cycle progresses from light sleep to deep sleep and then to REM, lasting about 90-110 minutes. As the night progresses, the proportion of deep sleep decreases while REM periods lengthen, explaining why you're more likely to remember dreams from just before waking.
The architecture of these sleep stages provides a framework for understanding how DEC2 mutation and other potential sleep-optimizing mechanisms might work. Now, let’s talk about that.
How does DEC2 fit into this exactly?
OK, so you’ve heard of DEC2 (or BHLHE41) mutations, but what does this mutation actually do? Yes, it shortens sleep, but as we’ve (kind of) laid out, there’s actual physics involved in sleep. Fluid comes in, it power-washes our brain and clears out debris. There’s memory consolidation. There’s, uh, other stuff (again, sorry neuroscientists…I’m sure there’s hidden functions you need to kill 500 mice in a lab before they tell you about, but my tally is not so high).
DEC2 is a transcriptional repressor, meaning it decreases our expression of particular genes which, in this case, are involved in our circadian rhythm (i.e. the global system that tells us whether we feel alert or tired). Some people have a mutation in this gene where they can get away with <6 hours of sleep. Lots of reporting has been done there in preclinical models, but I’ll note that studies on short sleepers in the adult human population via the UK Biobank are not so conclusive.
“Here, we take advantage of up to 191,929 individuals from four population-based studies, including the UK Biobank, to estimate the effects of these variants on sleep duration and timing using self-reported and accelerometer-based sleep estimates coupled with sequencing data. Our analysis revealed no association between variants previously reported and extreme sleep conditions. … Our results indicate that previously reported variants are not causal for extreme sleep conditions in the general population” (PLoS Genetics).
Theoretically, DEC2 mutations drive changes in sleep needs by making sleep more efficient rather than eliminating some need for sleep. The molecular specifics are suggested to be related to altered sleep pressure, which is what ultimately drives our feeling of being tired. During sleep itself, DEC2 mutations increase the capacity for cells to deal with stress. So, people with these mutations get less sleep, but the body is able to handle that better via a few functions. Increased mitochondrial function (make ATP better), improved stress resistance (oxidative, endoplasmic reticulum etc) and orexin upregulation to make DEC2 mutants feel more wakeful despite overall decreased levels of sleep.
If we put it all together, what we can appreciate is that DEC2 mutants attack the sleep problem from two directions – staving off feelings of fatigue accrued through the day, and making better use fo the deep sleep hours they do get. Though it’s hard to disentangle the dominant driver of the sought-after phenotype of short sleep, I lean towards the side of improved efficiency being the key aspect of sleep. Decomposing sleep tells us that the mechanical features of metabolite clearance and also stress remediation both need to happen for sleep to be successful. Even though DEC2 might reduce the sleep drive that accumulates over the course of the day, this alone would likely lead to long-term consequences of DEC2 mutants. After all, toxins that build up in the brain, especially tau and amyloid, are linked to neurodegenerative diseases. Yet we know that at least in mice, DEC2 mutants do not experience the deleterious effects of short sleep. Taking this to the logical endpoint, it appears DEC2 mutants feel less tired because they are able to adequately clear accumulated central nervous system waste during shortened sleep.
The search for short sleep probably should start with improving efficiency of mechanics of deep sleep.
Short sleep might be inducible by other means
I’m just thinking out loud here, but what this all suggests to me is that short sleep can be achieved by different means, not just DEC2. Decomposing the processes of sleep – the mechanical but also mole cular – reveals some outlets to relieve heightened metabolic stress associated with less deep, slow-wave sleep.
Some ideas, in no particular order:
Alternative Short Sleep Mutations: The beta-adrenergic receptor variant ADRB1 seems promising. Someone should look into that one!
Glymphatic Power-washing efficiency: Improving the efficiency of waste clearance during sleep seems necessary to make us need less sleep. Research into things like aquaporin-4 channels, which facilitate CSF flow through brain tissue, shows that their distribution and function affects clearance efficiency. Targeted interventions that enhance these channels or increase their density should improve waste removal, thereby decreasing our need to be in deep sleep. Genes to consider: CLDN1, MFSD2A and HDAC2.
Slow-Wave optimization: Technologies that enhance slow-wave activity, like transcranial direct current stimulation (tDCS) or acoustic stimulation synchronized to brain waves, have shown promising results in deepening slow-wave sleep. I view this as another way of making sleep more restful.
Memory Consolidation Accelerators: Sleep spindles and hippocampal ripples coordinate memory transfer during sleep. Compounds or techniques that enhance these specific oscillations could potentially speed up memory processing, allowing for shorter overall sleep duration. Some options that come to mind are Gadd45y (DNA repair gene), c-Rel (another repair and inflammation one for the hippocampus) and AP-1.
I don’t know if DEC2 will work out. It seems like some folks are working on trying to replicate academic studies, and I hope for my sake it does prove true. Regardless, I think hacking sleep, whether by molecular means or otherwise, requires revisiting the fundamentals of sleep. A lot of talk has been had about DEC2 so I thought I would just help the conversation by zooming out a little and open things up to other possibilities, lest we put all of our eggs in one basket.
For now I’ll simply refrain from taking up napping as a hobby in anticipation of good things to come.
Thank you to Stephen Malina for feedback on this post.