Intelligence is our most complex characteristic. Some would even say it defines us, setting us apart from other primates. And now, a new study – published this week by Hennady P. Shulha and colleagues in PLOS Biology – brings us a step closer to explaining the genetic and evolutionary basis of human intelligence.
To understand the significance of this research, we need to take a step back and consider some basic genetics.
The genome project
The human genome was sequenced a decade ago, but we still don’t have much idea about how it works. That is, we understand how genes encode proteins, and that mutations in genes sometimes make defective proteins that cause disease.
But, for the most part, we can’t track complex human characteristics back to any genetic explanation.
Most researchers would say that the 20,000-odd protein-coding genes encoded in the human genome are necessary but not sufficient to build the complexity of the organism they describe.
There are roughly 250 different types of cells that make up a human body, and the set of genes that is “expressed” (i.e. when a gene’s information is used in the synthesis of a functional gene product) in each cell type is different.
So there is clearly a complex mechanism controlling which genes are expressed where and when. Modern biologists increasingly believe this is where the real meat of the human genome project is.
Understanding how genes are controlled, how they control each other, and how turning genes on and off in different places and at different times allows not just the building of a human from a single cell, but also the development of a human brain that can recognise itself, understand language, write plays and build robots to explore other planets.
And this is where epigenetics comes in.
We know there are short DNA sequences located close to genes that affect how those genes are expressed. But DNA, it seems, isn’t just a long list of instructions that are read off, but rather is itself an active participant in the reading and interpretation of those instructions.
Parts of chromosomes (thread-like structures in the nuclei of cells, containing a DNA molecule) loop around to contact each other so often a region that controls a gene is some distance away.
Sometimes regions of the genome are wrapped up so the genes can’t be accessed and sometimes the DNA itself is chemically altered to make it more, or less, readable.
To make matters even more complicated, these different states of the DNA can often be inherited through cell division and even passed through generations.
As mentioned, the catch-all term used to describe these types of genomic activities is “epigenetic” – characteristics of the genome that affect and influence the expression of genes in time and space, so the same sequence of DNA can do different things in different contexts.
One common epigenetic “marker” is the chemical state of proteins which make up the superstructure around which DNA is wrapped.
These proteins, called histones, have various states which can be altered by adding and removing small methyl (CH3) groups. In turn, the methylation state of the histones around which a particular region of DNA is wrapped affects whether or not that region of DNA is expressed.
There are enzymes that add and remove the methylation groups, so one way of controlling lots of genomic regions in concert might be to increase an enzyme that changes the methylation state of particular histones.
You can see that this isn’t far off the idea of a master switch.
So different cells have the same genome sequence but different epigenetic interactions. If we want to find out how complex human characteristics are controlled at the molecular level, we need to look at the networks of epigenetic interactions that control genes, and try to track down those interactions that make certain cells behave in certain ways.
Which brings us back to the new research published by Shulha and colleagues in PLOS Biology.
(The prefrontal cortex is an area of the brain associated with evolution of the primate brain, and, by inference, intelligence.)
By comparing the locations of a particular type of methylated histone called H3K4me3 (which is often found at the start site of regulated genes) between human, chimp and macaque neurons, they found several hundred regions in which humans appear to use the process of histone methylation to control areas which chimps and macaques don’t.
Why is this interesting? Because these may be the genomic areas that are involved in human brain functions, including intelligence.
Of course, humans, chimps and macaques are different in lots of ways (not just in intelligence). To address this, the authors showed that some of the regulatory regions they had identified were only found in human neurons, not in the cells surrounding neurons.
The regions that are different between humans and monkeys might just relate to general differences between the species – e.g. regions that control tail growth – rather than ones responsible for brain functions.
The fact the regulatory regions were only found in neurons and not in non-neuron human cells probably means they are specific to neurons. And therefore, these regulatory regions are more likely to relate to brain differences rather than just species differences.
The basis of intelligence?
So we have a set of highly regulated DNA regions that are only found in prefrontal cortex neurons and are not found in less intelligent but closely related species. Furthermore, some of those regions are close to genes that are known to be involved in brain-related diseases.
And the regions all share a common control mechanism, leading us to think there may be some way of tweaking them as a group.
Sure, we’re still a long way off understanding the link between the genome and human physiology. But new tools which allow us to query epigenetic interactions are bringing us closer to being able to describe how different types of cells look and behave and why that might be the case.
In this way, the new research by Shulha and colleagues provides hints as to the genetic networks that control the development of higher-level functions such as human intelligence.
Andrew Lonie is a computational biologist and Head of the Life Sciences Computation Centre at the Victorian Life Sciences Computation Initiative hosted at the University of Melbourne, where he is also Co-ordinator of the Masters of Science in Bioinformatics.