Friday, September 17, 2010

Ancient origins of the cerebral cortex

Just how special is the human brain? Compared to other mammals, the thing that stands out most is the size of the cerebral cortex – the thick sheet of cells on the outside of the brain, which is so expanded in humans that it has to be folded in on itself in order to fit inside the skull. The cortex is the seat of higher brain functions, the bit of the brain we see with, hear with, think with. In particular, one of its main functions is association – bringing sensory information together with information on internal states and motivation to enable flexible and context-dependent decisions to be taken, rather than simple reflexive actions in response to isolated stimuli. While undoubtedly vastly more developed in humans, a new study suggests the cerebral cortex may have much more ancient origins than previously suspected.

All mammals have a cortex and it generally increases in size over evolution. Mice and rats have a smooth cortex, while that of cats is somewhat expanded and folded. Monkeys and apes show progressively bigger cortices as they get evolutionarily closer to humans. Dolphins and elephants also show highly expanded and folded cortices, so we are not the only species to arrive at this arrangement.

Expansion is coupled with an increase in the complexity of the cortex, as defined by the number of distinct cortical areas. This is mostly due to the emergence of additional association areas, where information from different inputs is integrated, and, in humans, an increase in distinct areas in the frontal and prefrontal cortex – the seat of the most sophisticated executive functions, including decision-making and long-term planning.

But when in evolution did the cortex actually evolve? Does it have some ancient precursor or did it arise as a new invention at some point? There has been considerable debate for decades over whether birds and reptiles have a counterpart of the cortex. They do have some regions that occupy the dorsal parts of the brain and perform somewhat similar functions, but their organization is so different from that of the cortex in mammals (which is arranged into discrete layers, while these regions in birds and reptiles are arranged into clusters of cells) that it has been difficult to establish their relationship.

Whether particular brain structures in different species are related to each other (i.e., whether they diverged from a single structure in a common ancestor) is often a subject of debate and controversy. It can be difficult if not impossible to determine based solely on location, anatomical organization or functional similarity. This is because it is relatively easy for these parameters to change over the course of evolution – they can be affected by changes to one or two genes, which means there is plenty of variation in these phenotypes within the population – the raw material for evolution by natural selection.

If the final phenotypic end-point of any particular region is quite variable, the opposite is true of the genetic pathways that specify the identity of the region at earlier stages of development. These involve master regulatory genes, whose expression is turned on or off in various parts of the embryo in response to earlier pathways that specify positional information (head from tail, back from belly, etc.). So, there are genes that differentiate nervous system tissue from the rest of the embryo, that differentiate forebrain, midbrain and hindbrain and that differentiate later subdivisions, including the cerebral cortex.

These genes act as “transcription factors”, controlling the expression of sets of proteins which define the mature characteristics of any particular region. While it is relatively easy for one of these effector proteins to change over evolution – affecting some specific characteristic of the region – it is much harder for the master regulatory genes to change. This is because they do not work alone – each area is defined by the expression of a combination of such genes, which are often turned on or off in a specific sequence. These genes interact in a complicated network of feedforward and feedback loops to orchestrate this complicated sequence. The networks in which they operate are so interlocked and involved in so many different parts of the embryo that mutations to any one gene tend to have very drastic consequences and will be rapidly selected against. These early regulatory systems are thus far less variable and tend to be highly conserved across evolution. So much so, in fact, that in many cases the function of one of these genes in one species can be carried out perfectly well by a copy of the gene from even a very distantly related species.

It is thus possible to tell whether a brain region in one species is homologous to one in another species (which may look quite different in its mature characteristics) by looking at how those regions were specified. If they derive from regions of the embryo which are specified by the same sets of regulatory genes then one can infer they both evolved from the same region in a common ancestor, no matter how different they may look now.

Similar patterns of gene expression argue that the cortex of mammals and the “pallium” of birds and reptiles are indeed related to each other, but a new study from the lab of Detlev Arendt goes much further back in evolutionary time. They compared the pattern and sequence of genes involved in specifying the cerebral cortex in mammals and the mushroom bodies, a sensory-associative brain centre in a much simpler organism, an annelid worm, called Platynereis. While it had previously been suspected that there might be a relationship between these structures (particularly between the cortex and mushroom bodies in insects) it had been impossible to determine definitively due to technical difficulties in examining the expression patterns of more than one gene at a time. The researchers in Arendt’s lab solved this problem by developing a new image-registration technique so that many different gene expression patterns could be mapped onto a common template and compared.

They found the same set of genes is expressed in these regions, in the same temporal sequence, under the influence of the same patterning mechanisms (those that specify where different structures develop in the embryo). Even further, very similar profiles of gene expression were observed in specific types of neurons in the mushroom bodies and in the cerebral cortex. This similarity extend to mushroom bodies in the brain of the fruitfly Drosophila, which are well known to be involved in sensory-associative integration, as well as learning and memory.

The conclusion from all these data is that the common ancestor of annelids, insects and vertebrates (the common ancestor of protostomes and deuterostomes, for those keeping score) already possessed some brain structure, specified by this defined set of genes, which was involved in integrating sensory information and performing associative functions. The morphology and connectivity of this structure has diverged significantly in each lineage since then, but the underlying similarities in function remain.

The extension of this principle of conservation in the genetic mechanisms specifying various organs or cell-types – which has been observed in eyes, limbs, hearts, and practically everywhere else one looks – to the part of the brain that most defines our humanity reinforces the notion espoused by Darwin, that “the difference in mind between man and the higher animals, great as it is, certainly is one of degree and not of kind."

Tomer R, Denes AS, Tessmar-Raible K, & Arendt D (2010). Profiling by image registration reveals common origin of annelid mushroom bodies and vertebrate pallium. Cell, 142 (5), 800-9 PMID: 20813265

Wednesday, September 8, 2010

Wild-type humans

Wild-type is the term geneticists use to refer to non-mutants. It literally means organisms that are the same, genetically, as those in the wild, compared to ones that have been grown under coddled conditions in the lab for generations, going soft in the absence of natural selection, or that are specifically mutant at some gene or other. There are no wild-type humans.

Well, maybe there are a few, somewhere, but even they are not really non-mutants. We all carry millions of mutations in our genome – positions where the sequence in our genome differs from the typical sequence. Where everyone else has a “T”, you might have an “A”, for example. Most of these mutations have no consequence – they are simply neutral variation in DNA that has no discernible function. It turns out that most of the genome is not made of genes – the bits of DNA that code for proteins actually comprise only about 2-3% of the total sequence. Mutations that change the code for proteins are by far the most likely to cause disease or to result in an obvious phenotypic difference.

New DNA sequencing technologies have revealed how many mutations of that type each of us carries, on average. Lots: around 10,000 mutations that change the amino acid code of a protein. Those can be broken down based on frequency in the population. Some mutations are seen in many individuals in the population – this implies that they occurred long ago in some individual and have subsequently spread in the descendants of that individual. The inference is that such a mutation does not have a deleterious effect as it would have been selected against if it did. About 90% of protein-changing mutations fall into this common, ancient class. In fact, in many such cases it can be difficult to say which allele (which version of the sequence at a specific position) is “wild-type”.

Some of these common mutations are actually adaptive and may be much more common in some populations than others. These include mutations that affect skin colour, for example, reflecting adaptation to either high sunlight (requiring protective melanin) or lower sunlight (requiring less melanin to allow vitamin D production), as well as variants affecting diet, such as lactose tolerance, adaptation to environmental conditions, such as high altitude, or resistance to specific pathogens or parasites. So, what is wild-type in one population may be mutant in another.

The remaining 10% of mutations are either very rare or “private”, having only ever been observed in one individual. When searching for mutations responsible for genetic diseases, these are the variants that researchers go after. Of course, not all of these will have phenotypic effects. Many changes to the code of amino acids in a protein can be tolerated without compromising function. It is possible to estimate how many rare mutations each of us carries that are likely to affect protein function – this is between 100 and 200, quite a small number, really. As well as mutations that change one DNA base to another, these also include a different class – mutations which result in the deletion or duplication of a whole chunk of a chromosome (copy number variants).

This got me to idly musing about what would happen if you took someone’s DNA sequence and “corrected” all those mutations to the wild-type version. What would the result be? Those 200 or so rare mutations may generally be tolerated (they are clearly not lethal at least) but could still result in suboptimal performance of any number of biochemical, cellular or physiological processes in each one of us. They may also contribute to differences in morphology by subtly affecting processes of growth and development. As these mutations tend to reduce the function of the encoded protein, presumably it should be “better” to have the wild-type version. (For good measure, let’s imagine we can “correct” all the mutations predicted to affect protein function, even if they are slightly more common – say up to 5-10% frequency in the population, but not so common that we can’t say what the wild-type version is).

This was the premise of the excellent movie GATTACA. Apparently the book that inspired it was also good, but I haven’t read it because it didn’t have Uma Thurman in it. The movie did, Uma being somebody’s vision of what a wild-type human female would look like (and who would argue?). Her male counterpart, Jude Law, reinforces the impression that they would be, most importantly, ridiculously good-looking. Poor Ethan Hawke was cast as the guy born by traditional procreative methods, mutations and all.

Beauty is only skin deep, of course, and what really interests me is what would their brains look like? It takes a lot of genes to assemble a human brain and all of us carry mutations in many of those genes. Those differences affect how our brains are wired and influence many aspects of our personality, perception, cognition and behaviour (as pretty much all the posts on this blog describe). What would the brain of someone with each of those deleterious mutations corrected be like? Would they be a genius? Especially empathetic? A naturally coordinated athlete? Would they be left or right-handed? What would their personality be like? Is there a wild-type level of extroversion or neuroticism or open-mindedness?

For some of those traits the optimal level may be different from the maximal level. For brain size, for example, which is correlated with intelligence, there is a trade-off in, first, being able to make it out the birth canal and also in metabolic demand – big brains use a lot of energy. And for may personality traits it is difficult to define a single optimal point along the spectrum – there are many different strategies that may succeed better in different contexts. Being fearless and aggressive may attract the ladies, but could also get you killed young. So, our wild-type humans may have perfect vision and perfect teeth, but it’s much harder to define a perfect personality.

Another consideration is that natural selection has only ever acted on individuals with a genetic burden of mutations – we may thus in some way be adapted to that situation. Some mutations that decrease the function of one protein may be beneficial in the context of another mutation in a different protein. Perhaps putting all the perfect proteins together in one person would not actually generate an optimal system.

In the movie, the generation of these “genetically perfect” beings was accomplished by gradually selecting out all such mutations by screening embryos created by in vitro fertilization. The fatal flaw in this idea is that it considers the spectrum of mutations as static in the population, suggesting that once all the bad ones are weeded out, that will be that. This ignores the fact that the rate of new mutations is actually quite high. Each of us carries about 70 new mutations that are not inherited from our parents. Most of these arise during generation of sperm. The reason that mutations in sperm are more common than in eggs is that women are born with all their eggs already generated. The cells that generate sperm, in contrast, are constantly dividing throughout life. Each division increases the chance of incorporating an error. That is the reason why the rate of dominant Mendelian diseases – which are those caused by single mutations and which include many cases of common diseases such as schizophrenia and autism – increases with paternal age.

Of course, all of the discussion above is based on the premise that genetic effects on physical and psychological traits are predominant. This extreme form of genetic determinism was also espoused in GATTACA, to the point of predicting the cause and date of a person’s death! In reality, genetic factors have a large influence on many of these traits but by no means an exclusive one – intrinsic developmental variation, environmental effects and experience will all also contribute to varying extents. On the other hand, introducing mutations tends not only to change a phenotype but to increase the variance in the phenotype – as the system becomes more compromised, its output becomes more variable.

It would be interesting to ask, therefore, exactly how much variation in these traits would be left across our wild-type humans.

Ng, S., Turner, E., Robertson, P., Flygare, S., Bigham, A., Lee, C., Shaffer, T., Wong, M., Bhattacharjee, A., Eichler, E., Bamshad, M., Nickerson, D., & Shendure, J. (2009). Targeted capture and massively parallel sequencing of 12 human exomes Nature, 461 (7261), 272-276 DOI: 10.1038/nature08250

Roach, J., Glusman, G., Smit, A., Huff, C., Hubley, R., Shannon, P., Rowen, L., Pant, K., Goodman, N., Bamshad, M., Shendure, J., Drmanac, R., Jorde, L., Hood, L., & Galas, D. (2010). Analysis of Genetic Inheritance in a Family Quartet by Whole-Genome Sequencing Science, 328 (5978), 636-639 DOI: 10.1126/science.1186802