Of all the fluids produced by the human body, the cerebrospinal fluid (CSF) may be the least offensive. A clear, unassuming liquid, CSF circulates throughout the ventricular system of the brain and the central canal of the spinal cord, cushioning the nervous system from injury and rinsing away metabolic waste. The composition of CSF is very similar to blood, but lacks proteins. If you were to taste CSF, it would be salty and a little bit sweet, like a mix of Gatorade and seawater. And unlike blood (or feces), a cup of CSF would probably not make you feel afraid or disgusted.
But despite its prosaic reputation, CSF has some interesting quirks. One of them involves its relationship to the nervous system. Nearly a century ago, two neuroanatomists, W. Kolmer and E. Agdhur, discovered a population of neurons that innervate the central canal of the spinal cord, and make direct contact with the CSF. These CSF-contacting neurons, or CSF-cNs, appear to be ubiquitous among vertebrates, as Kolmer and Agdhur found them in over 200 species.
CSF-cNs are peculiar-looking neurons. Each cell body possesses a bushy protuberance, which extends into the central canal. This protuberance includes a primary cilium, which is characteristic of other sensory neurons, such as olfactory receptors and auditory hair cells. Thus, it has been proposed that CSF-cNs may be sensory neurons that monitor the chemical or mechanical properties of the CSF. There is also evidence that CSF-cNs can affect locomotion by inhibiting premotor interneurons in the spinal cord. However, the specific sensory properties of CSF-cNs have remained elusive.
Now, a recent study from Claire Wyart's lab at the Institut du Cerveau et de la Moelle épinière, Paris, has explored the mechanosensory function of CSF-cNs in larval zebrafish. Using genetic tools for labeling CSF-cNs, Urs Lucas Böhm, Andrew Prendergast, and colleagues began by showing that the cilium that extends into the central canal is free to bend, which could allow it to detect CSF flow. They then used 2-photon calcium imaging to show that these neurons respond to active contraction of tail muscles, as well as passive mechanical bending of the tail. These mechanosensory responses were abolished in mutant fish that lacked an ion channel specific to CSF-cNs (called PKD2L1), indicating that the CSF-cNs may be directly mechanosensitive.
Böhm, Prendergast and the team then used fish with the same PKD2L1 mutation, as well as an independent genetic method, to examine how CSF-cNs contribute to zebrafish swimming behavior. Although fish lacking feedback from the mutated CSF-cNs were able to swim, they exhibited a marked reduction in their tailbeat frequency. Thus, CSF-cNs appear to provide mechanosensory feedback to motor circuits in the spinal cord, leading to an increase in tailbeat frequency during swimming.
Although these data provide the first evidence that zebrafish CSF-cNs respond to mechanical stimulation generated by CSF motion, many questions remain. For example, what are the dynamics and distribution of mechanical force generated by CSF movement during locomotion? Which aspects of CSF flow do the CSF-cNs detect and encode? The molecular mechanisms by which CSF-cNs sense mechanical signals are also not clear. PKD2L1, the ion channel required for mechanosensory responses in CSF-cNs, is highly sensitive to extracellular pH. In fact, a separate study by Elham Jalalvand and colleagues has recently suggested that a primary function of CSF-cNs is to monitor the pH of CSF and decrease locomotion when it drifts outside the appropriate physiological range. Given all of these complex and provocative links to the central nervous system, CSF may soon shed its reputation as the most boring of the bodily fluids.