Our brains are split into two halves, a left hemisphere and a right hemisphere. While the left brain specializes in languages, the right brain specializes in faces. But the two halves don’t exist as two separate entities. Instead, both halves or hemispheres are connected at several points. These connections are important to transfer and coordinate information between hemispheres. But how do the correct neurons “know” if and where they should cross? The textbook model so far has described a positive chemical signal called Netrin that diffuses in the developing brain and guides crossing neurons. But new approaches to this question recently suggested a very different answer.
Our cognition and behavior, such as math-savviness or creativity, depend on highly complex neuronal circuits in the nervous system. In the adult human brain, over 100 billion neurons form on average 7,000 connections with other cells, including those in the other brain hemisphere. The precise wiring of neuronal circuits requires that each axon— a neuron’s long projection that conducts electrical signals— can find and identify its correct target. Four categories of guidance cues are thought to guide axons on their journey. These signals act at either a long range or a short range. Additionally, chemicals acting at each range can be either attractive or repulsive. These four possible combinations of properties belong to four major protein families: the netrins, ephrins, semaphorins, and slits.
“Like a lighthouse that guides ships from afar, netrin1 from the floorplate would orient commissural axons from a distance.”
Netrin is the textbook model of chemoattractant. The first member of this protein family was identified more than 25 years ago as Unc-6 in the worm Caenorhabditis elegans. This round worm has been extensively studied by neuroscientists because of its simple nervous system, which consists of only 302 neurons. In Unc-6 mutants, neurons do not extend correctly along the dorsal-ventral axis, the imaginary line from the worm’s back to its belly. The corresponding vertebrate gene, or ortholog of Unc-6, netrin1 (derived from the Sanskrit word netr, or “one who guides”), was identified in a painstaking search in chicken and mice for factors that can influence the guidance of axons in the dorsal spinal cord.
In the spinal cord as well as in the hindbrain, netrin1 is produced in high levels at the so-called floorplate at the midline. The midline is the line of symmetry that divides our body and brain into left and right. In the brain and spinal cord, the midline is a staging post. Commissural axons, which are axons that connect the left and right sides of brain and spinal cord, grow towards the floor plate. They then cross the midline to reach the opposite side of the nervous system. In mice in which the function of netrin1 is strongly reduced, the commissural axons fail to cross the midline, and several brain commissures do not form. Experiments performed on cells in a dish suggested that a gradient of netrin1 guides axons. Altogether, these observations caused researchers to presume that in the vertebrate nervous system, netrin1 forms a ventral to dorsal gradient that is sensed by growing axons. Like a lighthouse that guides ships from afar, netrin1 from the floorplate would orient commissural axons from a distance.
But there was a fly in the ointment. First, there was no direct proof for a ventral to dorsal gradient of soluble netrin1 in the nervous system of mouse or chick. Work in the fruitfly Drosophila melanogaster also showed that while many commissural axons fail to cross the midline in a netrin mutant fly, this failure is overcome when netrin is replaced in a form that cannot diffuse. This suggests that in Drosophila, a diffusion of netrin from the midline is not required for the guidance of commissural axons across the midline.
“If we imagine the nervous system as a a slightly squished donut, the ventricular zone is the region adjacent to the donut hole.”
Recently, two studies suggested that, in fact, the same is the case in the spinal cord and hindbrain of mice. Writing independently in two highly respected journals, groups led by Samantha Butler at the University of California, Los Angeles, and by Alain Chedotal and Patrick Mehlen, at Sorbonne University and the University of Lyon, respectively, used new genetic tools in the mouse to finally shed a light on whether or not netrin1 really acts as a “lighthouse.” They found that netrin1 was not only produced in the floorplate, but also in the so-called ventricular zone. If we imagine the nervous system as a a slightly squished donut, the ventricular zone is the region adjacent to the donut hole. Neural progenitors in the ventricular zone are cells that produce new neurons. While the progenitors’ cell bodies line the “donut hole”, they also have long processes that extend to the opposite side of the CNS – the donut’s outside or pial surface. Both groups suggest that the neural progenitors produce netrin1 and then transport it along their processes to the pial surface, where netrin1 protein is also found.
The researchers then used new genetic tools, including mouse lines that express Cre-recombinase. This tool allows researchers to remove DNA sequences in specific cells or tissues. By combining this with conditional mutations of netrin1, the researchers removed netrin1 from one specific area in the mouse brain, while leaving it intact in other areas. When they removed netrin1 from the floorplate, the researchers saw no obvious defects in the axons growing to and across the midline. So, both in the spinal cord and the hindbrain, the groups showed that netrin1 from the floorplate (the assumed source of a ventral-to-dorsal “lighthouse” gradient) is not required to allow axons to grow to and across the midline. In fact, it is neither necessary nor sufficient. When netrin1 is removed from neural progenitors in the ventricular zone, many commissural axons fail to extend to the midline, even though netrin1 is still produced at the floorplate. Instead, netrin1 from the ventricular zone plays a critical role in guiding axons.
But if netrin1 is not a lighthouse, how does it guide commissural axons? Both groups find that while netrin1 protein is made in the ventricular zone, the protein itself is not actually found there. Instead, the neural progenitors are proposed to transport netrin1 protein along their radial processes to the pial surface. Commissural axons were observed to grow immediately alongside this netrin1-positive surface. After the axons grow along this surface, netrin1 protein also accumulates on axons. The new observations suggest a new model: neural progenitors and their processes lay down the netrin1-positive surface as growth substrate, which acts like a sticky track. Commissural axons then grow next to this growth substrate that leads them to the midline. It is, however, unclear how exactly the sticky Netrin track can orient the commissural axons.
Applying new methods to an old question showed that a sticky Velcro track of Netrin, rather than a beacon from a lighthouse, appears to establish the connections between our left and our right cerebral hemispheres (or, at least, those of our mouse cousins).
Artwork by Sean Noah, schematic diagram by Sophie Fessl.
Varadarajan SG, Kong JH, Phan KD, Kao TJ, Panaitof SC, Cardin J, Eltzschig H, Kania A, Novitch BG, Butler SJ. Netrin1 Produced by Neural Progenitors, Not Floor Plate Cells, Is Required for Axon Guidance in the Spinal Cord. Neuron. 2017 Apr 20. PubMed PMID: 28434801.
Dominici C, Moreno-Bravo JA, Puiggros SR, Rappeneau Q, Rama N, Vieugue P, Bernet A, Mehlen P, Chedotal A. Floor-plate-derived netrin-1 is dispensable for commissural axon guidance. Nature. 2017 Apr 26. PubMed PMID: 28445456.