A neurological mechanism for Fragile-X disease

Blogging on Peer-Reviewed Research

I’m busy preparing my lecture for genetics this morning, in which I’m going to be talking about some chromosomal disorders … and I noticed that this summary of Fragile-X syndrome that was on the old site hadn’t made it over here yet. A lot of the science stuff here actually gets used in my lectures, so they represent a kind of scattered online notes, so I figured I’d better put this one where I can find it.

I haven’t even finished grading the last of the developmental biology papers, and already my brain is swiveling towards the genetics literature, as I get in the right frame of mind to teach our core genetics course in the spring. And, lo, here is a new paper in PNAS that addresses details of a topic I bring up every time.

There are a surprising number of heritable diseases that share a couple of common traits: they are neurodegenerative, causing progressive loss of neural control, and they also exhibit a phenomenon called genetic anticipation—they tend to get worse, with earlier onset and more severe affects with each generation. Some of these diseases may be rather obscure, for instance
Haw-River Syndrome (AKA Dentatorubral-pallidoluysian atrophy),
Friedreich Ataxia,
Machado-Joseph Disease, or
X-linked Spinal and Bulbar Atrophy Disease (AKA Kennedy Disease), but others you’ve probably heard of, like
Myotonic dystrophy and
Huntington Disease. These are dreadful diseases that are variable in their pattern of appearance, and have terrible symptoms, like loss of motor control, chorea, seizures, dementia, and eventually, death.

Another tie that binds them together is their molecular cause: these are all trinucleotide repeat diseases. Our DNA has stretches in its sequence where the same nucleotides may be repeated a few times. The gene responsible for myotonic dystrophy, for instance, has a short stretch of cytosine-thymine-guanosine (CTG) repeated 5 times at the 3′ end of the transcript—CTGCTGCTGCTGCTG. People with a mild form of myotonic dystrophy may have 50 CTG repeats in the gene instead; in severe forms, they may have 1000.

The mechanism by which the sequences get excessively replicated is thought to be by mispairing. During the replication of the DNA, the individual strands unpair and re-pair, and with all of those similar repeated sequences, it’s easy for the two strands to get temporarily misaligned…and the replication enzymes err in making extra copies. I guess polymerases have the same problem reading the boring, repetitive bits that we do, and its easy to lose track of whether you’ve read CTG five times, or six times.

These excessive copies of codons tend to muck up the function of the gene. Huntington Disease, which is caused by an expansion of 19-21 copies of CAG in normal people into roughly 50 in affected individuals, affects an interaction domain. The CAG repeats form a repeated series of glutamine amino acids in the protein, which apparently bind to other proteins. Having many more glutamines makes the binding stronger, and the cell’s cytoplasm is afflicted with a gluey glop that leads to accumulating cell damage.

It’s easy to see how extra amino acids tagged onto a proteins can change its function. There is another trinucleotide repeat disease that has been less easy to understand, though. It’s called Fragile-X Mental Retardation (FMR), and it is the most common cause of heritable mental retardation, occurring in roughly 1 of every 1500 male births and 1 of every 2500 females.

FMR isn’t entirely a neurodegenerative disease. Its effects take place during development, and cause a characteristic suite of distinctive facial characters with moderate to severe mental retardation, although the degree of retardation may increase with age. Carriers of FMR, people who don’t express the overt symptoms but whose children are likely to inherit it, may also exhibit some late-onset difficulties such as tremor and ataxia.


Fragile X is a typical trinucleotide repeat disease. It is caused by an expansion of a CGG repeat in a specific region of the long arm of the X chromosome. Normally, we have 6-54 copies of CGG in this region; carriers have 50-200; and affected individuals have 200-1300 copies. There are enough copies to change the coiling and protein packing of the DNA in that region, which can give the chromosome a distinctive pinched appearance (as on the left), and in culture, cells with fragile-X are likely to break at that spot, hence the name.

One odd thing about Fragile-X, though, is suggested by the variable numbers of repeats in unaffected people: the repeats aren’t actually in a protein. They are in a stretch of DNA that is transcribed but not translated into a protein called FMRP, but the repeats are always neatly clipped off…so even people severely affected with Fragile-X Mental Retardation have a perfectly normal FMRP.

Part of this problem has been resolved. It turns out that all those CGGs seem to trigger a methylation response in the cell. Enzymes cruise along the chromosome, and attach methyl groups to the DNA backbone as a signal to the transcription enzymes to avoid this region. That means that although these individuals could make a functional FMRP, that gene has been inactivated, and they therefore have greatly reduced levels of FMRP. Of course, the next question is about what FMRP does in the cell, and that’s where the new paper by Weiler et al. comes in.

The brain is a very responsive kind of computer. It is constantly rewiring itself, modifying its pattern of connectivity in response to inputs. Much of this activity goes on at the synapse, or the regions where two neurons come into contact. What we generally think of when we talk about synaptic activity is the release of neurotransmitters from a presynaptic neuron, which bind to receptors on the postsynaptic neuron, which then cause ion channels to open or close and change the electrical properties of the postsynaptic membrane. Chemicals bind to receptors and cause a little blip of electrical activity, in other words. There is another kind of receptor, though, called a metabotropic receptor. Metabotropic receptors, when they receive a transmitter signal, do not directly cause an electrical change. Instead, they modify the metabolism of the cell. This can mean lots of things: they can change the rate of protein synthesis, they can cause phosphorylation of enzymes, they can enzymatically modify ion channels and thereby cause a change in electrical properties, they can even change the pattern of gene expression in the cell. The metabotropic receptor is a powerful thing that can dramatically change how a cell reacts to its environment.

Here’s where FMRP comes in. It seems to be part of the signal transduction mechanism that allows a transmitter signal received by a metabotropic receptor to be translated into a change in protein synthesis at the synapse. It helps to transport mRNA from the nucleus into the dendrites of a neuron, and there it can, when activated by synaptic signal, facilitate the translation of that RNA into protein. The somewhat complicated diagram below tells the story.

A putative role for FMRP in synaptic protein synthesis. (A) FMRP binds to its target mRNAs in the nucleus and helps export them to somatic cytoplasm. (B) FMRP and its target mRNA are packaged into transport assemblies and travel, likely by way of microtubules, down dendrites toward synapses. (C) FMRP–mRNA transport assemblies may take the form of nontranslating granules near synapses where they await some synaptic signal. (D) In response to activation of group 1 mGluRs (stimulated here with DHPG), phosphatidylinositol is cleaved into diacyl glycerol (DAG) and inositol triphosphate (IP3), initiating the release of intracellular calcium (Ca2+i) and activation of PKC. (E) PKC activation triggers an enzyme cascade that, by means of many intermediates (broken arrow), signals nontranslating granules to release FMRP-bound target mRNA for translation.

The circles in the gray, postsynaptic cell are copies of FMRP. It binds to mRNA in the nucleus and then carries it to various places.

In the closeup, we see some more details. At (D) is a metabotropic receptor, which has bound a glutamate transmitter molecule from the dark gray presynaptic cell at the bottom. The metabotropic receptor then activates some signalling pathways (the IP3 and DAG molecules, that then increase calcium levels and turn on PKC). These pathways then tell the FMRP+mRNA to wake up and start making new proteins.

Individuals with Fragile-X Mental Retardation have greatly reduced levels of FMRP, and therefore have lower levels of the specific transported mRNAs in their neurons. Furthermore, since they don’t have the FMRP-mRNA complexes present and ready to work, the metabotropic receptors have nothing to activate. The presynaptic cell is sending its signals, the metabotropic receptors are yelling at the cellular machinery to wake up and start working, but there’s little present to respond. Missing a piece of the response circuit means that the neurons have lost an important sensor that they use for synaptic remodeling.

You may be wondering if this offers the possibility of a cure for Fragile-X disease. The authors don’t speculate, but probably not, and if it provides any help at all, it’s a long way off. This is all work in mouse models, not people, and it’s an analysis of the cause of the problem. We don’t yet have a way to increase FMRP levels in people who have had the gene inactivated. If we could get more FMRP into the brains of people with this syndrome, maybe it would alleviate the progressive severity of the retardation, but it would be unlikely to correct the retardation itself. This is a developmental disorder; the brain has been deprived of some of its ability to respond to stimuli during its formation, and that is an error that can’t be retroactively corrected.

Weiler IJ, Spangler CC, Klintsova AY, Grossman AW, Kim SH, Bertaina-Anglade V, Khaliq H, de Vries FE, Lambers FAE, Hatia F, Base CK, Greenough WT (2004) Fragile X mental retardation protein is necessary for neurotransmitter-activated protein translation at synapses. PNAS 101 (50): 17504-17509.


  1. Darby says

    I’ve often wondered how CO and NO act as neurotransmitters in brain regions – are they interacting with metabotropic receptors?

  2. says

    It’s unfortunate that the hard (both senses of the word) science articles don’t get as many comments as the rest, but they’re much appreciated!

  3. says


    There just isn’t much that many of us are qualified to add, frankly. To be honest, I usually get rather lost on some of these articles!

  4. Nick Gotts says

    Thanks. Fascinating, and in the medium term, hopeful: what we understand, we can try to remedy.

    God not only moves in mysterious ways, he plays sadistic practical jokes.

  5. says

    #4 re: Isn’t the pinched chromosome on the right, not the left?

    Ah I spotted that as well but got distracted by the rest of the article. :^)

    #6 re: There just isn’t much that many of us are qualified to add, frankly. To be honest, I usually get rather lost on some of these articles!

    I like the getting lost part, though I call it “opportunity for learning.”

  6. imsd007 says

    > You may be wondering if this offers the possibility of a cure for Fragile-X disease.

    Instead of gene therapy, attempts to treat FXS may focus on pharmacologic reduction of mGluR signaling instead (at least in the near future).

    FMRP may negatively regulates local protein synthesis of transcripts in neuronal dendrites leading to their over-translation if FMRP is absent. This may lead to reduced synaptic strength due to AMPA receptor abnormalities. This “mGluR theory of FXS” postulates that an important consequence of FMRP reduction is excessive signaling through group 1 metabotropic receptors.

    A recent study represents an elegant test of this theory: Using a genetic approach the authors have shown that a genetic reduction of mGluR signaling can reverse several typical FXS symptoms in mice. This suggests that drugs targeting mGluR5 may be used as an effective treatment for FXS.

    Neuron. 2007 56:955-62.
    Correction of fragile X syndrome in mice.
    Dölen G, Osterweil E, Rao BS, Smith GB, Auerbach BD, Chattarji S, Bear MF


  7. John Vreeland says

    Very interesting. My aunt was killed by Friedreich’s Ataxia. This is the first news I have had that there is a test for it. Fortunately it’s recessive—and on chromosome 9, not X. I figure I have about a 25% chance of carrying it.

  8. Brachinus says

    If you taught a Developmental Biology course for non-majors, would it be known as HOX for Jocks?

  9. Knecht says

    CO and NO will bind to heme-containing proteins much like oxygen. Hemoglobin is the most well known example of a hemoprotein, but other proteins such as the cytochromes and soluble guanylate cylase also can bind gases. Soluble guanylate cyclase, in particular, has known cell signaling functions- its “ligand” is NO2, and in response to ligation, it catalyzes the production of cyclic GMP, a second messenger molecule (similar to cyclic AMP, the second messenger stabilized, most famously, by the action of caffeine) that has a diverse array of downstream effects (it activates, for one, protein kinases like PKC and PKG). I’m not too sure about CO and NO- they tend to be toxic due to their ability to outcompete oxygen and NO2 for binding to heme- but certainly NO2 has some described neuroactive effects that are mediated through soluble guanylate cyclase.

  10. David Marjanović, OM says

    I’ve often wondered how CO and NO act as neurotransmitters in brain regions

    Wait. Carbon monoxide? WTF?

    How did I miss that discovery!?!

  11. David Marjanović, OM says

    I’ve often wondered how CO and NO act as neurotransmitters in brain regions

    Wait. Carbon monoxide? WTF?

    How did I miss that discovery!?!

  12. caynazzo says

    Anyone know how a lack of FMRP brings about the “characteristic suite of distinctive facial characters” in heritable mental retardation?

  13. Confused says

    #1 – I’ve often wondered how CO and NO act as neurotransmitters in brain regions – are they interacting with metabotropic receptors?

    No. The primary role of membrane receptors, whether they’re ion channels, g-proteins (like most metabotropic receptors), receptor kinases or whatever, is to pass a signal across the membrane, and they all have some quite nifty ways of doing it (g-proteins tend to tug their transmembrane regions up and down causing a conformational change, kinases tend to dimerise, bringing together to halves of an enzyme to make one complete one).

    If your signalling molecule can readily diffuse across the membrane, like gases (NO and CO) can, there’s no point in all this messing about at the membrane – they just wade into the cytoplasm (sometimes right into the nucleus) and do their thing. Examples include Retinoic Acid (the receptor sits in the nucleus waiting to be activated), and endogenous cannabinoids (which iirc are in presynapses). Most of what I know about CO/NO is from blood vessels, where they activate an enzyme to make cGMP. I’d guess the brain is much the same.

    I guess polymerases have the same problem reading the boring, repetitive bits that we do, and its easy to lose track of whether you’ve read CTG five times, or six times.

    PZ, are you insinuating that molecules have intelligence? Careful, or the IDiots will quotemine you. ;)

    The metabotropic receptor is a powerful thing that can dramatically change how a cell reacts to its environment.

    One of my biggest criticisms of the blue brain project is that (as far as I can make out) it completely ignores metabotropic receptors.

    I could waffle on more, but I won’t.

  14. kgoldberg says

    Thank you for posting this article. It is interesting to me both as a nurse in a newborn ICU (where I have seen many of these diseases- especially myotonic dystrophy), and as a student interested in this area of biology. It doesn’t go unnoticed. Wish I lived in Minnesota. :)

  15. Sili says

    Funny. When I had to take biochem/cellbio &c I couldn’t get away fast enough. Bought Stryer, never opened it, and handed it over to the student book exchange after the exam (which was sans book anyway – wtf?) without ever checking if they managed to sell it and pay me.

    Now I wish I had it here …

  16. Mark says

    This is another good example of poor “design”. Why would an intelligent designer put such an important gene on the X chromosome where half the species get only one chance at a good copy? Maybe god IS a she-woman manhater! ;-)

  17. mlw says

    #4 & #9 re: I believe the “(as on the left)” refers to the location of the image relative to the text rather than which chromosome pictured within the image.

  18. Nick Sullivan says

    This is neat, I thought that cytoplasm soluble proteins just diffused down the axons, but instead some specific mRNA’s have FMRP tags which allow there to be a on site stockpile of mRNA’s ready for translation in to the required proteins. Neat.

  19. imsd007 says

    > This is neat, I thought that cytoplasm soluble proteins just diffused down the axons,…

    Dendrites in this case. Dendrites carry electrical stimulation received from other neural cells towards the cell body. Axons carry nerve impulses away from the cell body.

    Many proteins (and also whole organelles) are transported selectively into dendrites and axons in vesicles or transport granules.

    There is also axonal mRNA transport, as for example transport of beta-actin mRNA (via a protein called “zipcode-binding protein” ZBP1) to name just one.

  20. says

    Sigh. Not learned enough to opine effusively upon this article, so I shall have to resort to snark:

    1. Just as in every other aspect of life, too much repetition can be very bad. I’m looking at you, boring jobs!

    2. If I’d been willing to attribute “mind” to trinucleotide repeats, I could have had a hell of a lot of fun with the “mindless repeating of dogma” metaphor.

    3. And, in closing, I enjoy the fact that one could compare genetics to a game of “Telephone.” Anyone smarter than me on the biological details care to help me expand that into an “Evolutionary Biology for Dummies” sort of explanation for how errors in copying can lead to diversity (and hideous diseases)?

    Here endeth the snark. Over to you, brains!

  21. Confused says

    I thought that cytoplasm soluble proteins just diffused down the axons

    Even in motor neurons, where the axon can reach over a metre in length?