Why Do Some People Need Less Sleep? The Genetics of Short Sleepers

Most of us treat five or six hours of sleep as a debt we’ll eventually have to pay back. But there’s a small group of people for whom that’s just a normal night. They wake up rested, don’t nap, don’t crave a lie-in on weekends, and have been doing this since childhood. Researchers call them natural short sleepers, and over the past fifteen years a handful of labs have tracked down the actual genes responsible. What they found says less about sleep hygiene and more about how the brain decides when it’s had enough rest.

What makes someone a “natural” short sleeper

The key word is natural. Natural short sleepers (NSS) typically sleep four to six hours a night without the daytime sleepiness, cognitive fog, or health consequences that come with ordinary sleep deprivation [1]. This is different from someone who simply forces themselves through a short night and then feels terrible. It’s also different from familial advanced sleep phase, where people fall asleep and wake up earlier than usual but still get a normal total amount of sleep [1]. NSS is specifically about needing less total sleep, not shifted sleep.

Because the trait is rare and tends to run in families, researchers have spent the last decade and a half recruiting these families, sequencing their genomes, and then testing the candidate mutations in mice (and occasionally fruit flies) to see if the same mutation actually shortens sleep in an animal model. So far five genes have held up to that scrutiny: DEC2, ADRB1, NPSR1, GRM1, and most recently SIK3.

DEC2: the first short-sleep gene

DEC2 was the first one found, back in 2009, when researchers sequencing DNA from a small family noticed two relatives who both had a mutation in this gene and both slept around 6.25 hours a night compared to about 8 hours for their non-carrier relatives [1]. The mutation swaps a proline for an arginine at position 385 (or 384, depending on the numbering convention used) in a protein that normally acts as a brake on the circadian clock machinery.

To confirm this wasn’t a coincidence, the team built transgenic mice carrying the same mutation. The mice stayed awake for a longer stretch of each day and slept less overall, particularly during their normal rest period, and their sleep was more fragmented, with more, shorter bouts of both non-REM and REM sleep [1]. They even engineered fruit flies with the equivalent mutation, and the flies slept less too, which is a strong hint that whatever DEC2 is doing has been conserved across a huge stretch of evolutionary time.

Later work traced the mechanism further. DEC2 normally suppresses a gene called Hcrt, which encodes the wake-promoting peptide orexin. The mutant version of DEC2 is worse at this job, so orexin levels creep up, and that extra orexin appears to be a major reason these mice (and by extension, these people) spend more time awake [4]. Giving the mutant mice an orexin receptor blocker partially restored normal sleep, which is about as close to a smoking gun as you get in this field.

ADRB1: turning up a wake-promoting switch in the brainstem

Ten years after DEC2, the same broader research group (based at UCSF) identified a second gene: ADRB1, which encodes the beta-1 adrenergic receptor. This is the receptor that noradrenaline and adrenaline act on, and it’s a well known player in the sympathetic nervous system, but its role in the brain was much less understood [2].

The family in this study carried a mutation, A187V, sitting in one of the receptor’s transmembrane domains. Carriers slept about 5.7 hours a night on average versus 7.9 hours for non-carriers [2]. When the researchers engineered the same mutation into mice, the mutant mice lost roughly 55 minutes of sleep per day, almost entirely from the dark (active) phase, which is when mice are normally awake and moving around anyway [2].

What’s interesting here is where this receptor is doing its work. It’s highly expressed in a brainstem region called the dorsal pons, and using calcium imaging the researchers showed that neurons expressing ADRB1 there are active specifically during wakefulness and REM sleep, going quiet during non-REM sleep [2]. When they used light to artificially switch these neurons on during non-REM sleep, the mice woke up almost immediately. In the mutant mice, this same population of neurons was more active overall, which lines up neatly with the shorter, more wakeful nights [2].

The mutation itself makes the receptor protein less stable, so there’s actually less of it around. But the effect isn’t uniform: the receptor’s inhibitory signaling is more sensitive to that drop in protein levels than its excitatory signaling is. The net result is a population of wake-promoting neurons that’s harder to inhibit, so it stays active more of the time [2].

NPSR1 and GRM1: two more brakes released

Two other genes rounded out the picture before the most recent discovery. NPSR1 encodes a receptor for neuropeptide S, a signaling molecule that activates histamine and orexin-producing neurons involved in wakefulness [4]. A gain-of-function mutation, Y206H, was found in a family whose carriers slept just 4.3 and 5.5 hours a night, and the equivalent mutation in mice hyperactivated a thalamic region involved in switching the brain between sleep and wake states [4].

GRM1, meanwhile, encodes a metabotropic glutamate receptor (mGluR1) involved in synaptic signaling. Two separate families turned up two different loss-of-function mutations in this gene, both of which reduced activation of a signaling pathway (ERK) previously shown to lengthen sleep in mouse models. Less ERK activation, shorter sleep [4].

What’s notable across DEC2, ADRB1, and NPSR1 in particular is that none of them seem to work by reducing how much sleep the body needs in a homeostatic sense. Instead they all seem to strengthen wake-promoting circuits. That’s a meaningfully different idea: these people may not need less sleep pressure to build up before feeling sleepy, they may just have a stronger competing signal keeping them awake, one that doesn’t come with the usual costs of overriding sleep pressure [4].

SIK3: the newest addition, and a look under the hood

The most recent gene to join this list is SIK3, reported in 2025. A woman in her seventies who self-reported sleeping around 3 hours a night (actigraphy put it closer to 6.3) was found to carry a mutation, N783Y, in this kinase gene [3]. Unlike the other four genes, SIK3 had already been heavily studied in mice through unrelated genetic screens, where different mutations were known to push sleep need up or down depending on whether they increased or decreased the kinase’s activity. This made SIK3 a strong candidate the moment it turned up in a human short sleeper.

Lab tests confirmed the N783Y mutation reduces the kinase’s activity, and mice engineered with the same mutation slept about 32 minutes less per day, with less rebound sleep after being kept awake, and elevated delta power indicating higher sleep pressure that the mutant mice were apparently more able to tolerate [3]. The researchers went a step further than most of these studies and looked at what SIK3 is actually doing at the molecular level: profiling nearly 12,000 phosphorylation sites in the mutant mouse brains, they found that the mutation causes widespread loss of phosphorylation, concentrated heavily at synapses, and specifically dials down the activity of protein kinase A while ramping up MAPK signaling [3]. That’s a level of mechanistic detail the earlier short-sleep genes still don’t fully have, and it points toward a kinase signaling network at synapses as the place where sleep need actually gets computed.

Takeaway

Across all five genes, a pattern is starting to emerge. These aren’t people whose bodies quietly need less rest in some abstract sense. They’re people whose brains have a stronger push toward wakefulness, whether that’s from extra orexin, an overactive wake-promoting neuron population in the brainstem, or altered kinase signaling at synapses, and that push doesn’t seem to come at the cost of the usual sleep debt. Mouse models carrying these mutations generally don’t show the memory problems or exaggerated rebound sleep you’d expect from ordinary sleep deprivation.

None of this is a blueprint for skipping sleep. These mutations are each individually rare, found in a handful of families worldwide, and the animal work, while thorough, still can’t fully answer whether there are subtler long-term costs in humans. But as a natural experiment, this small set of families has done more to reveal how sleep is actually regulated at the molecular level than decades of studying sleep-deprived rodents, precisely because these people show what the brain looks like when it needs less sleep and nothing else seems to go wrong.

Common Questions

Is being a natural short sleeper the same as just getting used to less sleep?

No. Natural short sleepers don't train themselves into it and they don't build up sleep debt the way someone chronically sleep deprived does. The trait shows up from childhood, runs in families, and carriers report feeling fully rested on 4 to 6 hours without needing to catch up on weekends.

If I only sleep 5 hours a night, do I have one of these mutations?

Almost certainly not. These mutations are individually very rare, and most people who sleep 5 hours are simply sleep deprived, not naturally short sleepers. The giveaway is how you feel: natural short sleepers don't feel tired, don't nap, and don't want more sleep even when it's offered.

Do any of these mutations make you sleep-deprived without you knowing it?

The animal studies say no. Mice carrying these mutations show normal or even reduced sleep debt after sleep deprivation, and the human carriers score normally on cognitive tests. That said, researchers are upfront that long-term, larger-scale health effects in humans haven't been fully ruled out.

Could these findings lead to a pill that lets anyone sleep less?

That's the long-term hope of several of these labs, since genes like DEC2, ADRB1, and SIK3 are considered plausible drug targets. But none of this is close to a product. Right now the mutations are studied because they're a rare natural experiment in sleep regulation, not because there's a therapy ready to go.

References