Gonna make you fall in love with…Spines

Share on FacebookShare on Google+Email this to someoneTweet about this on Twitter

Never ever let a n* ride if you think he’s gonna slide pop’em in the spine…fo’ the money. – Bone Thugs N Harmony

The Genius is dead. Genius Re-post:

Your standard neuron in the cortex is called a pyramidal neuron. It generally has on big apical dendrite sticking out the top and a few basilar dendrites radiating around the bottom. Here’s a glowing one:

This one comes from a somewhat troublesome paper from the lab of Karel Svoboda who as far as I can tell is a guru of new imaging techniques who happens to have an interest in the details of dendrites as well. The reason why the paper was a little vexing is because the results seemed to run against a large literature showing that the major identifiable morphological problem with neurons in Fragile X Syndrome brains has to do with dendritic spines. Let me back up. Your standard neuron in the cortex has dendrites and an axon. The axon is the output structure that goes and pokes at another neuron and can fire action potentials that release neurotransmitters from its tip to send signals. The dendrites are the parts that receive signals from axons. Where an axon and a dendrite meet is called a synapse. Dendrites are studded with these little knobby-doos called spines. Spines are where most excitatory synapses in the cortex are made.

Spines go through a maturation process whereby they initially come out as long wormy things called filopodia and then kind of settle back into a variety of shapes often resembling the power-up mushrooms in Super Mario Bros. or like a large kind of extended comma. The fat part at the end is referred to as the spine head and the thin part connecting to the dendritic shaft is the spine shaft.

Spine shapes change in the lifespan of the spine, but also change over the lifespan of the organism. And along side the changes in shape come changes in the number of spines per dendrite (the spine density). Since people had looked at adult human Fragile X brains and seen abnormalities with regard to spine morphology and density it seemed clear that something must be occurring during development.

Nimchinsky et al (2001) wanted to see what happens in development so they got some normal mice and some mice lacking fragile X mental retardation protein (FMRP) that are a pretty good model for what’s happening in humans with the disorder. When the mice were one, two, or four weeks old they injected a particular part of the cortex with a fancy shmancy virus that lights up a good portion of neurons (anyone who reads science crap knows about GFP the magical jellyfish protein, that’s how they’re doing this). This allowed them to use laser microscopy to do detailed quantitative analysis on the spine lengths and density. The troublesome bit is that while they found the expected differences between FraX mice (the fragile X model) and wild-type (read: normal) mice at one week (shown below), the differences disappeared by 4 weeks. How can this be?!

They offered up a number of considerations including that they were looking at a different chunk of cortex than other people usually do and that perhaps there were some limitations to their visualization technique. The most important limitation, as it turns out, is that they can’t use their technique past about 6 weeks of age because the virus they used is too bad at infecting neurons after that. A more recent paper from the Greenough group in Illinois, who were some of the major reporters of these fragile X spine abnormalities in the first place, arrived at a resolution to this conflict.

Galvez and Greenough (2005) took basically the same tack as the Svoboda group except they chose a couple of different ages. They used 25-day old mice to map onto the 4-week time point previously reported and they took another sample at age 73-76 days old. For those of you who have trouble dividing by 7, that’s about 10-11 weeks. The reason for doing this is that it is understood that a major developmental process called pruning might be at play here. Initially developing brains produce huge amounts of connections during a period of widespread synaptogenesis. They don’t need all of these connections. So they have to be ‘pruned’ back. There are lots of tree metaphors when talking about dendrites that can make it very pleasant for a neuroscientist to contemplate a tree in the park on a summer afternoon. This apparently happens in mice some time after one month of age.

The Greenough paper used a more rudimentary staining procedure so they could look at spines in the older brains. They managed to replicate for the most part the finding that FraX mice and wild-type mice don’t differ at 4 weeks with regard to spine shape and density. But at the much later time point they differ pretty radically as illustrated here:

It appears that the major malfunction with Fragile X spines then isn’t that they can’t grow out right or anything like that. Its that they can’t be eliminated after the fact. Notice how much more bare the adult wild-type dendrite (Figure 1C from the paper) looks compared to the FraX dendrite (Fig 1D). This is awfully nice to see for two reasons. One, because it clears up an apparent discrepancy in the literature and it turns out that everyone was right which ought to make all the labs involved feel marvelous. It’s also nice because it indicates that the real neurological problems in Fragile X development start later than expected in development. I’m not quite clear on when this massive pruning event is supposed to happen in human development, but it opens a window whereby if we ever get the means to replace this protein we might ameliorate some of the effects of the syndrome.


Nimchinsky EA, Oberlander AM, Svoboda K (2001). Abnormal development of dendritic spines in FMR1 knock-out mice. J. Neurosci. 21:5139-5146.

Galvez R, Greenough WT (2005). Sequence of abnormal dendritic spine development in primary somatosensory cortex of a mouse model of the fragile X mental retardation syndrome. Am. J. Med. Genet. 135A:155-160.


  1. This has got to be a dumb question, but how can there be more dendrites than axons on a neuron? If every dendrite connects to an axon and vice versa, this would imply equality. Or is it that each axon sends signals to many dendrites in different cells, but each dendrite connects to only one axon? That’s probably it, but I never see it explained. Why not?

  2. Re: Interconnection between axons and dendrites 
    View the dendrites of a neuron as branches of a bush and the axon as the root system. (Some neurons have bushy axon roots; others have one long carrot-like root.) An axon ?root? may pass near the dendrites of thousands of different neurons. Or the axon may pass close to many dendritic branches of the same neuron. Picture a long vine growing through the branches of several neighboring bushes. When an axon and a dendrite branch are sufficiently close a synapse can form. So an axon ?root? could have many connections to the same dendritic branch, could connect to many branches of the same neuron, or could connect to the branches of many different neurons. In a similar manner other neuron axon ?roots? are passing through the same neuron bushes. So a single neuron dendrite branch could pass near the axon ?roots? of many different neurons. 
    Initially, there are a very large number of weak synaptic connections between the axons and dendrites. As the brain matures, a few of the synapses are strengthened while most weaken and disappear, as shown in the images Coffee Mug provided. In a mature human brain, there is an average of 1,000 to 10,000 connections per neuron. 
    The interconnection pattern is different for different neural tissue. Some axons connect nerve centers to distant sensory or motor neurons. Some axons connect different neural layers.

  3. This was a good read. Does this mean if we could find some kind of dendritic spine growth inhibitor (and implement it) FraX would go away? And what is it that the excess spines do to the person exactly? Firing some conflicting impulses?

  4. Shan: ?And what is it that the excess spines do to the person exactly?? 
    At this stage, we only have speculation.  
    I believe that the total connectivity strength of a neuron must be kept within a certain range. Too little or too much and the neuron dies. (A living cell must maintain homeostasis.) As some neurons are strengthened during learning, others should weaken. Thus as a brain matures we should see a few strong synapses rather than many weak synapses. 
    My guess is that the problem isn?t necessarily excess spines. It could be a failure to strengthen synapses which results in the retention of weak synapses. Or it could be a failure to further weaken the weak synapses and so the total connectivity strength isn?t regulated properly. That could lead to inappropriate cell death. 
    (Hopefully Coffee Mug or someone else will correct my mistakes and offer their own conjectures.)

  5. That makes sense, perhaps at an early age our brains are unable to precisely control the impulses released by the neurons, overdoing it could probably damage critical neurons at their most sensitive phase of development.

  6. Opps… 
    “As some neurons are strengthened during learning, others should weaken.” 
    Should read, “As some synapses are strengthened during learning, others should weaken.”

  7. yeah. i don’t think it’s clear what the functional outcome of immature-looking spines is. in a sense, we know what the final outcome is because we know what the symptoms of fragile X syndrome are. on the other hand, the mental retardation could be caused by some other neurological malfunction and the spine morphology is more like a side-effect.  
    if spines can’t be matured you might expect it to be difficult to ever make a secure, permanent synapse. this should in principle make it difficult to learn and maintain information, but only certain types of info are affected, i think. i wonder if the same morphology phenotype is also observed in the cerebellum for instance. searching… i guess so: 
    The fragile X-cerebellum connection. 
    * Huber KM. 
    Center for Basic Neuroscience, Department of Physiology, University of Texas Southwestern Medical Center, Dallas, TX 75390, USA. kimberly.huber@utsouthwestern.edu 
    Fragile X syndrome (FXS) is an inherited form of mental retardation that results from the loss of function of the fragile X mental retardation protein (FMRP). A recent report demonstrated alterations in the structure and plasticity of synapses on cerebellar Purkinje cells in Fmr1 knockout mice, which are a model of FXS. These synaptic alterations are associated with deficits in the cerebellar learning both in the mice and humans with FXS. This work forges an important link between the FMR1 gene, altered synaptic plasticity in the cerebellum and mental retardation. 
    so i guess one can predict global delay in establishing permanency of connection. my question then is, how do the available mechanisms left over provide any permanency? how do fragile X patients learn how to walk or eat? 
    there must be some compensatory mechanism that is doing the job, but not quite..