Humans are two developmental stages away from monkeys

Less than 50 of our 20,000 genes are unique to humans. What separates us from monkeys?

Anyone with access to a television can probably think of at least one movie comparing people to primates. There's MVP: Most Valuable Primate, or The Planet of the Apes. The trope in science-fiction books is even more popular. As these popular books and movies suggest, we seem to be forever asking the question—just how are we different? 

At the most fundamental level, we're nearly identical to apes. Humans share 99 percent of our genetic information with chimpanzees, our closest relatives, and 93 percent with rhesus macaques, one of the most common monkey species. Out of our roughly 20,000 genes, less than 50 are exclusive to humans. From such tremendous similarity, how do our substantial differences arise? 

The PsychENCODE Consortium—a U.S. government-funded initiative, involving 116 researchers at 15 institutes—is trying to finally come up with an answer. The collective recently published their findings as a series of 11 papers in a special issue of Science. One of these studies, spearheaded by Yale neuroscientist Nenad Sestan, radically redefined our understanding of how our brains become human.

Cover of Science magazine for the PsychENCODE issue

Science, AAAS

Sestan’s team studied both humans and rhesus monkeys as they grew up, from infancy into early adulthood. The researchers followed how the human and monkey brains changed over time, using a technique known as single cell RNA sequencing. This allowed them to isolate individual cells from the brains at different ages, profiling the amount of every gene in each cell. This means that the scientists could track changes, uncovering differences that are masked when millions of cells are combined, as they are in traditional genetic tests. 

The amount of a gene in each cell can matter just as much, if not more, as the identities of the genes themselves. Consider baking cookies: Two recipes can have all the same ingredients, but if one calls for twice as much salt or baking powder, the final results will be dramatically different. If what separates people from primates is a result of how many genes of a certain type are in certain groups of cells—well, this would be the first study that could possibly detect these differences.

The researchers quickly noticed that for the most part, humans and monkeys don't actually diverge much as they're growing up. Humans go through every stage of macaque development, and most aspects of brain growth happen at the same time across the two species. In both humans and monkeys, brain cells—including neurons, cells that send messages to and from the brain, and glia, the cells that support neurons by providing nutrients and protection—develop and diversify in the same order, and at about the same time. As a result, our brains don't just grow in parallel—during certain periods, they actually become more similar to each other as they develop. 

The next logical question, is what, exactly, causes humans to deviate so radically from monkeys? The Yale researchers found a few notable differences: Though humans advance through all stages of macaque development, humans also have two extra stages of brain development, ones found in no other animal. The first comes at the very beginning of life, when fetuses experience a short period of great change. Interestingly, during this phase fetuses show developments in their prefrontal cortex, which is  crucial to complex thought and decision making, the most distinctive parts of human thought, and an area of the brain that's comparatively. After this period, our brains  rapidly become more similar to those of monkeys—until the next period of divergence.

Sestan’s team found this second period comes during late childhood, around the age of ten. Humans have a prolonged childhood, meaning we take longer to mature, and also keep some childlike traits, like having large eyes and heads, permanently. But until now, scientists didn't understand why we take so long to develop.  Only around the end of elementary school do the brain cells from our two species begin to form connections and transmit information differently. As a result, the researchers speculate that this phase could be responsible for how humans learn, think, and remember differently than monkeys.

Surprisingly, in all stages of development, humans and monkeys shared most cell types. This means if a neuron is present in humans, it is likely present in monkeys too. Our two species' genetic differences are generally found in a few small groups of cells, mostly subgroups of neurons. But even in these, the genetic differences across species remained relatively small—in total, the researchers found out of 20,000 genes, only about 50 diverged significantly. Despite the small number, some of these genes play major roles in mental function; many have been linked to disorders such as schizophrenia or bipolar disorder. For example, the gene GRIA1 controls learning and memory, causing schizophrenia by distorting the underlying basis of thought.

Baby rhesus macaque eating flowers

Prem Jejurkar on Pexels.com

Even though this work represents a major step forward in our understanding of what makes humans unique, this study only examined the genes found in a cell. Human and monkey brains may also differ in many other ways—like how brain cells connect to one another, or how these cells  communicate. Further research along these lines will be necessary to fully understand the relationship between our two species. 

Regardless, Sestan’s study illustrates how seemingly minor changes can yield major effects. Only two short periods of development meaningfully separate the brains of humans and monkeys, where only a few cell types receive the brunt of these changes. A minute number of genes diverge in just a few cells, and suddenly, we acquire the ability to read, to write, to speak—to be human. 

After the Planet of the Apes came out, there was some speculation on what hypothetically could've made these apes become so close to humans, and what such a scenario would require in real life. We now finally have an answer: not much at all.

Peer Commentary

Feedback and follow-up from other members of our community

Fascinating story James. I just found the lack of distinction between gene and transcript very confusing, I had to keep reminding myself you meant RNA levels and not multiple copies of the gene. Also when you say “in total, the researchers found out of 20,000 genes, only about 50 diverged significantly” are you referring to the 55 genes they found differentially expressed in the single-nucleus RNAseq data? Because [as far as I know] they only detected about 450-2500 genes in each of those samples anyways, so the 20,000 number is incorrect.

Also, what do you think of this story about researchers claiming to have transferred a single human gene into macaques and found an increase in intelligence? Credible? And does that mean that they’re just fifty odd genes away from engineering human-like intelligence in monkeys? 

James R. Howe VI

I do scRNA-seq myself, so I’m very cognizant of the gene/transcript distinction and detection limit. I also agree with both of your points! I found those two things a bit too technical to give a lot of space to (each would probably take at least a paragraph to fully flesh out), so I consciously elided those to ensure it stayed understandable to a general audience. I kind of think that if a person has those things in mind when reading the piece, they are likely already aware of the paper or probably have a sufficient knowledge base to read the original article in Science themselves, and these details do not really affect the overall thrust or conclusions of what I’m saying. 

Oh, I feel the same way as you on the RNA stuff. I think single cell omics studies are one of the most scarily complex things in biology at first glance, and there are so many technical considerations to keep in mind that there just isn’t time to hit in a piece, such as where to draw the line on splitting vs clustering cell types and differences in measures of similarity.

That monkey paper you linked is… interesting, to say the least. I’m looking through it a bit now, and it makes me feel kind of weird. From what they’ve reported, I see that some monkeys have 6 or 9 copies of the human gene knocked in, which probably skews things massively, and the behavioral assay is kind of weird and limited, which makes me wonder as to the generalizability and significance there. I would also be very skeptical if someone thought they could insert 50 genes and just expect monkeys to start talking. One of the big takeaways of the piece is that a lot of these things are related to expression levels, timing, and feedback loops that vary across individual regions and individual types of cells within these regions, so even small changes can have massive effects. We can even see that in relation to Neandertals, which have nearly identical genomes (more or less, obviously some things differ) but still were far less intelligent than modern humans. If we could exert such a precise degree of control synthetically, I think it could be possible, but I also feel like we would have pretty much solved biology at the stage when that’s possible.

This also isn’t even touching the ethical issues here, which I think would be super interesting to write about if every science writer hasn’t already beaten me to the punch!

This is a really cool article and a very interesting finding! RNA seq is such a powerful tool to assess such general processes like development. You mention that the fetus shows development in the pre-frontal cortex during the first divergent developmental change-does their RNA seq data reflect this idea, specifically in this area? I imagine there were differences across the whole brain, but I was wondering if they discuss which, if any, specific brain areas actually drive these two developmental stages. I imagine that identifying these would have important implications in studying neurodevelopmental diseases.

James R. Howe VI

Yes, they do! There are two cool figures in the original paper that shows the degree of difference for each region side-by-side and exactly what the differences they detected were. I would put it below but I have no idea if you’re allowed to reproduce non-OA journal figures without weird copyright stuff happening, so just check out the linked paper if you’re interested!