You are currently browsing the category archive for the ‘Epigenetics’ category.

Brains are made up of many different types of cells that come about through common stem cells called neuroblast progenitors. These progenitors can produce the number and diversity of cell types necessary to form a central nervous system. As neuroblasts divide to make more cells, they can potentially specify different types of cells with each round of division, and depending upon the stage of development. 

Lineage development from larva to adult. Cartoon diagram highlighting a single neural lineage in the larval brain (left) and in the adult brain (right). Note the decreased contribution of primary neurons in the adult brain compared to the larval brain. GMC ganglion mother cell. Spindler and Hartenstein, Dev Genes Evol. 2010 220: 1–10. doi: 10.1007/s00427-010-0323-7

In the fruit fly, most of the cells in the adult central nervous system have been mapped back to their embyonic neuroblast progenitor, of which there are about 100 or so. As such, the fruit fly brain is a popular model for understanding neural circuitry and development. One of the better-studied neuroblast lineages, referred to as NB7-1, gives rise to seven progenitor cells, the first five of which give rise to lineages of motorneurons that will wire the nervous system to the muscles. The remaining two neuroblasts give rise to interneurons; the cells that connect neurons within the nervous system.

Though not innervated by NB7-1 motorneurons, these are  developing flight muscles of the fruit fly (visualised in GREEN), which are remodeled from larval muscle templates by the fusion of adult muscle precursors (visualised in BLUE). The motor neurons (visualised in RED) of the larval muscles are re-specified to innervate the flight muscles. These muscles and motor neurons are a unique example of how adult structures are refashioned from embryonic counterparts during metamorphosis in animals. Imaged by: S. Roy (Institute of Molecular and Cell Biology, Singapore; Cover image, EMBO J, 08.08.2007)

Each of the five motorneuron-specifying neuroblasts gives rise to a different lineage of cells. The lineages specified in the first two neuroblasts, called U1 and U2, are governed by a gene called hunchback (hb), which is normally only active in the first two progenitors. The hb gene codes for a protein that regulates the activity of other genes called a transcription factorTranscription factors work by recognising DNA sequences in a gene’s promoter; the bit immediately preceding the coding sequence to a gene. Once bound to the promoter, they summon the molecular machinery (or are sometimes already pre-assembled to it) necessary to decode that gene into a template to make a protein. In some cases, like with hb, transcription factors (the proteins) bind to their own gene promoters. This is usually done to ensure that a lot of that particular transcription factor is produced, as it may be necessary to activate a great number of downstream genes. In fact, some transcription factors are responsible for turning on a whole cascades of gene activity; in some cases altering the identity of a lineage of cells.

In fact, if hb is experimentally turned on in any of the first five progenitors, it will turn on the same U1/U2 motorneuron identity in these cells. Curiously though, hb’s affect is limited to those cells as the remaining two (interneuron-specifying) progenitors and their resulting neurons do not respond to the hb signal. When a sub-group of cells is capable of responding to an inductive signal like this, it is said to be competent.

Chris Doe’s group at the Howard Hughes Medical Institute utilises this predictable progression of neuronal specification to understand how competence is regulated in cells. As the identity of each cell has been mapped from the embryo through larval stages and on to the adult, NB7-1 gives us a system to understand how cellular identity is fundamentally achieved at the genetic level. So the next time someone flips off fly research within earshot, enlighten them.

So if hb‘s job is to turn on the genes involved in the U1/U2 motorneuron program, why can the first five cells respond, but not the other two? Remember that the DNA is very highly packaged and organised within the cell. There are a number of ways that the cell can organise a specific region of DNA on a chromosome in order to ensure it is ready for activity when necessary. One of the mechanisms involves specifically positioning different regions of chromosomes in different parts of the nucleus. This filing system is not entirely different than that used by most first year university students in determining if clothes are still wearable (I’ll speak for the men here):  the pile of clothes on the floor in the middle of the room (though subject to a smell-test) are fair game. The clothes on the edges of the room; still on their shelves and hung up in the closet, like that sweater your lovely Aunt [so-and-so] knitted you for Christmas last year, are out of the way and generally not (read: never, in the case of the sweater) used. In a similar way, by tethering regions of a chromosome to the edge of the nucleus, to a structure called the nuclear lamina, they’re effectively inaccessible, no matter what transcription factor proteins are available to decode those regions.

Chris Doe’s group recently published findings that show that in the first two progenitors, the hb gene is physically positioned in the interior of the nucleus, a region that promotes its activity. In the 3rd, 4th and 5th progenitors, the gene is silent but remains poised in the interior (ie. it failed the smell-test; it remains on the dorm floor) and responsive to hb-induction. After the 5th division, hb gets repositioned to the edge of the nucleus (right next to the sweater) where it is silenced. This positioning is inherited by any of the future cells in that neuroblast’s lineage, ensuring that they hb protein is no longer produced in these cells. Experimentally interfering with this repositioning enables more cells to respond to the hb signal, resulting in the production in a greater number of U1/U2 motorneurons. This adds to the growing body of examples of genes that are regulated in a developmentally-relevant and heritable manner by tethering to the nuclear lamina. This also provides a mechanistic frame work to understand developmental competence at the molecular level in cells. 


We should have this whole biological-ball-of-wax figured out in about 100 years, or so (that bit is generally always implied in these sorts of assertions). At least that was a brief point that Professor Adrian Bird made at the start of his lecture at the Royal Society before accepting the RSGSK Prize last Tuesday night. It’s an intriguing prospect, but one that largely fell out of the scope of his talk. That said though, if biologists do find themselves out of work a century from now, the blame can be pointed squarely at the likes of Professor Bird. His research has contributed a significant amount to our understanding of one of the mechanisms involved in silencing the activity of genes across the genome.  In the process, he worked out the cause to a severe autism-spectrum disorder affecting one in 10,000 female births and developed a mouse model to help understand it, and potentially reverse it.

Twelve years ago the first draft sequence of the human genome was published and we are only just scratching the surface of understanding all of the information encoded within it. The genome’s publication and the tools born of the genomic era changed the way we approached molecular biology; we turned to more targeted approaches. However it quickly became apparent that there was a lot more information there and a lot fewer genes than we anticipated. The 21,000 (or so) genes that code for proteins proved to be just as numerous as animals we used to stare down our supposedly evolved noses at (read: flatworms).  These, while central as the blueprints to the various parts that make up the cell, are only 1% of the story.

The other 99% is composed of swathes of genetic information strewn throughout the genome that carry instructions on how and when to use the 1%.  It’s a shame economics couldn’t take a page from this arrangement, right? Consortium projects like ENCODE have significantly furthered our understanding how some rest of this genomic DNA works, but large gaps remain in our understanding of how all of this information is organised and utilised inside cells.If you could take all of the DNA that is inside a single human cell, strip it of its protein packaging and arrange the bits from the 23-pairs of chromosomes end-to-end, you would have something in the neighbourhood of two metres in length. All of that fits into a cell nucleus that has a diameter in the order of 10s of microns (that’s one-one hundred thousandth of a metre). Needless to say, one incredible folding act needs to take place to get all of it in there. DNA wraps around a spindle of proteins called histones, forming the repeating unit of DNA architecture that packages those two-ish meters of double helix into chromatin. This nucleosomal structure coils back on itself almost the same way a phone cord will if you start to twist it, packing all of that chromatin in the nucleus rather tightly. The big question is then, how does the cell access the right gene, at the right time, because the process of turning on genes (aka. transcription) requires that they are physically read, and therefore accessible amongst all of that coiling.

Part of this is explained by a field called Epigenetics that really took off in the post-genome era. Its origins go back at least as far as the 1940s-50s, when scientists like Barbara McClintock were discovering that the story of how genes worked didn’t stop at the DNA sequence itself.  She went on to win a Nobel Prize in 1983 for her work on mobile DNA elements that regulate gene expression, even though at the time her work was regarded by some as an eccentricity of Maize genetics. But that’s a whole other post!

Colour variation in maize is regulated by mobile DNA elements in their genome called transposons.

The part of the DNA sequence that immediately precedes that of a gene is called the promoter.  Think of this as a staging ground for the molecular machines that read the genes. In short, the promoter promotes gene activity, thus some or all of the following information can be found there:

  • The When – During what developmental stage should gene-X be active; embryo? child? adult?
  • The Where – In what ‘type’ of cell should gene-X be active; muscle? bone? brain?
  • and The How Much – Is a lot or only a little of gene-product-X needed?

This is nowhere near the whole story at the sequence level, but for now let’s assume it is, as I’m about to add a layer of complexity to it. The prefix epi- in epigenetics comes from Greek, meaning over or on top of as it represents the genetic regulation that supersedes that which is imparted by the DNA sequence, itself. Put another way, epigenetics is like a very elaborate filing system for genetic information in a cell.

Think of it this way, if all cells contain essentially the same set of genetic blueprints, then there have to be ways to mark the genes pertinent to a neuron’s function, for example, as active, while leaving the others who could be either irrelevant or potentially detrimental to neuronal function marked as silent. Epigenetics can be thought of as the collective set of molecular sign posts that index an epigenome, ensuring that only those genes are available to that cell for use when and where they are promoted to do so.

There are principally two ways that these  marks work.  One set involves the marking the histones, the proteins that package DNA into chromatin. These are chemically modified in a number of ways, facilitating a rich code of information that marks genes, where their promoters start, where their coding sequences start/end and in what degree of  readiness for transcription genes should adopt for that particular cell type.  How this code is read by the cell and how the cell organises this information in a functional way inside cells are still very actively researched topics.

The other type involves the direct chemical modification of DNA, called DNA methylation, which is generally associated with gene silencing.  It was this, and the mechanisms that the cell uses to detect this modification that Professor Bird’s group works on.  Certain DNA sequences can be chemically modified by the addition of methyl groups, and these tend to accumulate near the start of gene promoters. As is often the case with biology, when molecules get chemically modified, there’s usually another molecule that is able to read  or recognise this modification.  This, in essence, is how biological information flows in the cell.  One molecule is modified, another binds to it, and that binding effects and outcome.  In this case, methylating the DNA in promoters suddenly makes promoters a target for a protein called MeCP2, which leads to a cascade of events that results in the silencing of that gene. Mutations in MeCP2, as it turns out, are the major cause of Rett Syndrome.  During the lecture, professor Bird showed encouraging results that Rett Syndrome symptoms could be reversed with gene therapy treatment in the animal models within four weeks of treatment (and yes, that’s the same mouse).

June 2019
« Jul    


%d bloggers like this: