You can't teach a middle-aged ganglion new tricks

Repeated imaging of the same individual neuron for over a year in mice allows the authors of a new study in this issue to show that presynaptic axon terminals become progressively more stable as the animals age, changing little after 6 to 12 months.

A typical neuron has 10,000 neighbors, as compared to only tens or at most hundreds for other cells of the body. It matters, of course, not only how many neighbors you have but also which of your neighbors you talk to and who talks back to you. Furthermore, the complexity of the fully developed mammalian nervous system is many orders of magnitude larger than what could be possibly genetically determined. It is all the more important, therefore, to understand the rules governing neural (self-) organization, particularly during development, repair and adult plasticity.

We have known how neurons look, down to the finest axonal and dendritic branches, since Golgi discovered the eponymous staining method1 and highly corrected microscope optics were perfected. More recently, electron microscopy refined this picture and allowed, in particular, the unambiguous identification of synaptic connections. All these histological methods, however well they resolve fine details, work only on fixed tissue, thus completely missing information on neural activity and, even more important in the present context, being unable to detect the dynamics of individual morphological features. To assay such dynamics requires repeated imaging of the very same neuron in the very same animal, which to date only optical microscopy can deliver. But imaging in intact tissue or intact animals still faces substantial obstacles.

In a study reported in this issue, Gan and colleagues have overcome these obstacles and pushed long-term in vivo imaging to a new level2. Using the rodent parasympathetic submandibular ganglion preparation that the Lichtman and Purves labs have been perfecting over the past decades3,4, Gan et al.2 now show that at least some synaptic connections become increasingly morphologically stable as the nervous system matures, changing hardly at all after middle age (in a mouse 6–12 months). The reported observations were only possible by combining a number of cutting-edge technologies: microsurgical preparation, allowing the same area to be imaged several times during a period of more than a year, confocal microscopy and transgenic expression of green fluorescence protein (GFP), allowing stable fluorescent labeling of cells. Also, the submandibular ganglion preparation used by Gan et al.2 may be particularly conducive to such studies as the neurons are either completely isolated or present in small clusters on the salivary ducts, making relocation and imaging easier.

This work follows on the heels of two studies5,6 that examined neuronal morphology over time in the cerebral cortex and, incidentally, used very similar transgenic animals7. Interestingly, these earlier studies came to contradictory conclusions concerning the stability of dendritic spines. One found that in the mouse barrel cortex, although dendritic structure was stable over weeks, spines appeared and disappeared5. But according to the other study6, which used similar GFP-expressing mice, spines on pyramidal neurons in layer 5 of primary visual cortex show remarkable plasticity during the critical period early in development, but become much more stable in the adult. The current study by Gan et al.2 now adds complementary results on presynaptic morphological stability and thus provides further information about how stable the adult nervous system really is.

There are, however, a number of issues that still need to be resolved, including the physiological relevance of morphological stability or plasticity. Could a ganglionic synapse, which seems to serve as a relay, with reliable action potential initiation in the postsynaptic cell following every presynaptic spike, ever be the site of and hence model for 'memory'? Are morphological changes correlated to memory formation at all, as has been suggested, for example, by observations of spine genesis during synaptic stimulation8,9? Then there is, of course, the question of what the biochemical mechanisms are that provide stability and malleability, respectively, to synaptic morphology. The system used by Gan et al.2 may be well suited to study this very question, as the effects of transgenic and pharmacological manipulations on long-term morphological stability can probability be assessed more clearly and quantitatively in the submandibular ganglion than, say, in cortex.

Figure 1

Diagram of experimental methods.


  1. 1

    Golgi, C. Gazzetta Medica Italiana, Lombardia 33, 244–246 (1873).

  2. 2

    Gan, W.B., Kwon, E., Feng, G., Sanes, J.R. & Lichtman, J.W. Nat. Neurosci. 6, 956–960 (2003).

    CAS  Article  Google Scholar 

  3. 3

    Lichtman, J.W. J. Physiol. (Lond.) 273, 155–177 (1977).

    CAS  Article  Google Scholar 

  4. 4

    Purves, D. & Lichtman, J.W. J. Neurosci. 7, 1492–1497 (1987).

    CAS  Article  Google Scholar 

  5. 5

    Trachtenberg, J.T. et al. Nature 420, 788–794 (2002).

    CAS  Article  Google Scholar 

  6. 6

    Grutzendler, J., Kasthuri, N. & Gan, W.B. Nature 420, 812–816 (2002).

    CAS  Article  Google Scholar 

  7. 7

    Feng, G. et al. Neuron 28, 41–51 (2000).

    CAS  Article  Google Scholar 

  8. 8

    Engert, F. & Bonhoeffer, T. Nature 399, 66–70 (1999).

    CAS  Article  Google Scholar 

  9. 9

    Maletic-Savatic, M., Malinow, R. & Svoboda, K. Science 283, 1923–1927 (1999).

    CAS  Article  Google Scholar 

Download references

Author information



Rights and permissions

Reprints and Permissions

About this article

Cite this article

Denk, W. You can't teach a middle-aged ganglion new tricks. Nat Neurosci 6, 908–909 (2003).

Download citation


Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing