How to Enter  Winning Entries  Past Prize Coverage  Contact Information  Eppendorf & Science Prize Press Releases

The Brain That Nature Built

Maxwell G. Heiman

Buckminster Fuller marvelled at how order emerges from chaos: "To how many places does nature carry out pi," he asked, "when she makes each successive bubble in the white-cresting surf of each successive wave?" (1). This wonderment lies at the heart of many great scientific questions, including how an embryo calculates the shape of every neuron in the most amazing self-organizing structure in the world—the human brain.

The past 15 years have revealed stunning views of the brain's construction. First, genetic studies of the simple nervous system of the nematode C. elegans identified a secreted protein, netrin, needed to guide axons to their destinations (2). Soon after, landmark experiments with rat neurons showed that the tip of an axon computes its position in a netrin gradient (3), and since then, numerous studies have elucidated how this calculation is performed (4). We have learned that gradients of secreted factors such as netrin, Slit, Wnt, and others define the compass points along which the great thoroughfares of the nervous system are built.

As a neuron travels these paths, it accomplishes an astounding feat. Each neuron sees the same landscape, yet adopts a unique shape. To me, the most fascinating question in biology is how this vast diversity of architecture is encoded and how it is read out during development.

Rather than chasing these questions through the dense urban cityscape of the human brain, I turned instead to a tiny village of a nervous system. C. elegans has only 302 neurons and, remarkably, each develops in precisely the same way in every individual (5). As a postdoctoral fellow in the laboratory of Shai Shaham, I wanted to understand how the embryo calculates the shapes of these neurons so reliably.

We decided to focus on a group of 12 sensory neurons that share a simple shape. From their orb-shaped cell bodies, each extends two thin projections: a loop-shaped axon, through which it communicates with other neurons, and a single thread of a dendrite that extends to the animal's nose (6). Through these sensory dendrites the worm collects most of its information about the world (7). We asked a profoundly simple question: How does it know how long to make them?

To pluck out the genes that control dendrite length, we cast a wide net. Using a classical genetic strategy, we randomly mutated genes and looked for individuals whose sensory dendrites were the wrong length. We expected these mutations would define the genes that determine dendrite length.

While looking through thousands of nematodes, each with its neurons lit by a fluorescent protein, I found one that was exactly what we hoped for. The neurons had normal-looking cell bodies and normal-looking axons, but instead of the elegant threadlike dendrites extending to the nose, there were only stubs (see the figure) (8). Within a few days, I had a handful of mutants with the same problem.

To understand why these dendrites failed to form properly, we watched them as they grew—or failed to grow. C. elegans embryos are small and transparent, and most cells get their shapes in a defined window of about 2 hours (9). We used a recently developed imaging trick (10) to watch, for the first time, the growth of single dendrites. What we saw surprised us. In normal embryos, a sensory neuron was born near the nose and the cell body migrated away, yet the tip of its dendrite remained firmly anchored at the nose (8). This anchoring caused the dendrite to stretch to exactly the length needed to reach the nose. In our mutants, this anchoring failed, the dendrite tip was dragged along with the cell body, and the resulting dendrite did not reach the nose (8).

Our mutations, we knew, had disrupted the dendrite anchor. What did the mutated genes normally do? We found that each mutant disrupted either of two genes encoding extracellular proteins resembling sperm-egg proteins: one that makes the zona pellucida (ZP), the gelatinous coat surrounding the egg, and another that is present in sperm and binds the ZP (8). Neither looks anything like the proteins that usually tether neurons to their targets—with one exception. A similar pair of proteins is expressed in the human ear and forms a gelatinous matrix that anchors the tips of sensory cells used for hearing (11). There were hints of a similar system, too, in fruit flies (12).

Here, then, was a possible explanation for how these proteins worked: Perhaps, as in the ear, they were secreted and formed a matrix anchoring the tips of sensory neurons, resisting the force of cell migration much as their counterparts resist the motion caused by sound waves. This part of our hearing system might even have evolved from such a dendrite anchor.

To test this idea, we examined these proteins further. They are expressed during dendrite formation(one by the neurons, the other by neighboring cells) and act only at this time (8). They are present at dendrite tips, suggesting a direct role in anchoring (8). And, most importantly, both proteins are secreted and one of them forms multimers: the signature of a matrix (8).

We had found an amazingly simple solution to our question. The embryo does not calculate the length of a dendrite any more than a wave calculates the diameter of a bubble. In each case, structure emerges spontaneously from the properties of the system. In the embryo, the fixed pattern of cell division deposits sensory neurons at the nose where they immediately drop anchor, ensuring that this connection persists no matter where the cell body migrates.

This simplicity might have pleased Buckminster Fuller, who sought to uncover the "elegant and exquisitely exact mathematical coordinate system [nature uses to] formulate and mass-produce all the botanical and zoological phenomena" (1). One dimension of this system lies in gradients of secreted cues such as netrin. Another is probably a network of adhesion points specific to each neuronal class, like the anchor that connects sensory neurons to the nose. Together, they form the grid on which nature constructs a brain as orderly as the nematode's or as wild and restless as our own.

References and Notes

1. R. B. Fuller, E. J. Apple, Synergetics 2: Explorations in the Geometry of Thinking (Macmillan, New York, 1979).
2. N. Ishii, W. G. Wadsworth, B. D. Stern, J. G. Culotti, E. M. Hedgecock, Neuron 9, 873  (1992).
3. T. Serafini et al., Cell 78, 409 (1994).
4. L. A. Lowery, D. Van Vactor, Nat. Rev. Mol. Cell Biol. 10, 332  (2009).
5. J. G. White, E. Southgate, J. H. Thomas, S. Brenner, Philos. Trans. R. Soc. London Ser. B 314, 1(1986).
6. S. Ward, N. Thomson, J. G. White, S. Brenner, J. Comp. Neurol. 160, 313 (1975).
7. C. I. Bargmann, WormBook 2006, 1 (2006).
8. M. G. Heiman, S. Shaham, Cell 137, 344  (2009).
9. J. E. Sulston, E. Schierenberg, J. G. White, J. N. Thomson, Dev. Biol. 100, 64  (1983).
10. R. Ando, H. Hama, M. Yamamoto-Hino, H. Mizuno, A. Miyawaki, Proc. Natl. Acad. Sci. U.S.A. 99, 12651  (2002).
11. C. Petit, J. Levilliers, J. P. Hardelin, Annu. Rev. Genet .35, 589 (2001).
12. Y. D. Chung, J. Zhu, Y. Han, M. J. Kernan, Neuron 29, 415  (2001).
13. I thank S. Shaham for guidance and C. K. Patil for the Buckminster Fuller references.

To Advertise   |   Find Products