Earliest Tactile Sensory Neurons in Vertebrate Evolution
by David D. Olmsted (Copyright - 1998, 2006. Free to use for personal and
educational purposes)
Last Revised September 1, 2006
Tactile Sensory Neurons of the Tadpole Body
Figure 1
Sensory Rohon-Beard fibers from the tadpole of the clawed toad Xenopus (Roberts and Hayes - 1977)
 |
The evolutionarily
earliest tactile sensory neurons in vertebrates seem to be the Rohon-Beard cells
(named after the people who first described them in fish, Rohon (1885) and
Beard (1896). These Rohon-Beard cells are found in adult lampreys which represent
the evolutionarily earliest class of vertebrate, the jawless fish. In amphibians
and fish they are present only in the embryonic and young larval (tadpole)
stages and originate in the spinal cord. In adults they are displaced by a more
specific sensory system originating from cells in the dorsal root ganglions
located just outside and along the the segments of the spinal cord. So the
Rohon-Beard cells in fish and amphibians are an example of "ontogeny follows
phylogeny" in which more recent evolutionary advances are grafted on
top of older structures in the developing embryo.
The sensory fibers of the Rohon-Beard neurons run towards the spine from the underside of the fish or tadpole. These unmylinated
input fibers have a diameter of 0.2 to 1.2 micrometers and run just under
the skin forming a loose mesh as shown in figure 1. They are sensitive to a light
touch as produced by a light stroking with a hair. These fibers have no specialized
endings so that any deformation along their length tends to produce a signal.
While
the Rohon-Beard fibers signal touch in these animals other skin disturbances which
might be interpreted as pain are signaled by a general electrical skin response.
The skin itself becomes depolarized and propagates a signal which is received elsewhere
by the nervous system (exactly where is not known). Further development of
this property allows some fish (such as Apteronotus albifrons) to sense electric
field disturbances and other fish such as electric eels to give a shock when touched.
Evidence that this “pain” detecting system is separate from the touch system is
shown in figure 2. It shows that the animal can still respond to strong pokes (the
x's) even when the Rohon-Beard cells are destroyed which cover the same area thus
preventing any stroking response.
Figure 2
Responses to Pokes are Separate from Responses to Strokes (Roberts and Smyth - 1974)
 |
The motor responses are always the same when the skin and the Rohon-Beard neurons
are activated suggesting that they produce reflexes and not motivation
releases. The Rohon-Beard cells produce a spinal cord flexure while a skin impulse
due to a poke, a shock to the gill area, or a pinch produces a half second
burst of tail swimming activity (Roberts and Smyth - 1974)
The number of Rohon-Beard
cells in stage 35/36 embryonic tadpoles averages 190 covering a total area of 15.7
square mm. This works out to a mean field size of 0.083 square mm per cell.
Yet the measured mean receptive field size is 0.16 square mm. Consequently, the
receptive fields almost completely overlap having an overlap ratio of 1.93
(0.16/0.083). A ratio of 2 would indicate complete overlapping. The receptive field
sizes range from 0.04 to 0.25 square mm. (Roberts and Hayes - 1977). The action
potential responses to stroking are shown below in figure 3. Notice that the
greater is the intensity
of the stimulus the longer is the pulse showing that in this case the frequency
of the neuron's action potentials do not correspond to stimulus intensity as is
assumed by conventional neural networks. Since
these are reflexive actions the longer the pulse length the longer the motor action
command and the greater the spinal flexure. If the frequency of the action potentials
were also modulated then the rate of the flexure would also be modulated but this
type of modulation either is not possible with this neural system or it is not required
of the animals environment. since the purpose of this touch response is probably just
to allow the animal to squeeze around obsticals.
Figure 3
Responses of the Rohon-Beard Cells to Stroking as Recorded from the Spinal Cord (Roberts and Hayes - 1977)
 |
The Tactile Sensors of the Tadpole Head
Figure 4
Responses of Neurons in the Head of an Embryonic Tadpole (Roberts - 1980)
 |
While the tactile sensors of the body trigger reflexive avoidance actions
most of the tactile sensors of the head trigger reflexive stop actions since they
sense when the animal runs into something. These stop action commands not only end
swimming actions but also prevent the normal triggering of (motivation released?)
swimming by dimming light (Roberts - 1980).
The neural responses of
the stopping tactile neurons (called movement detectors by Roberts) are shown in
figure 4. Their mean receptive field size is 0.017 square mm. Repeated stimulations
habituate these sensors reducing the average number of action potential spikes
(the length of the pulse) by 3/4 unless resting periods of 2 to 3 minutes
between stimuli occur. Notice that like the Rohon-Beard cells the responses
modulate the length of the pulse more than the action potential frequency inside
the pulse such that the frequency remains consistently highly valued at 120 to 140
Hz (the A example has a different scale than the rest as explained at the bottom
of the caption). This again suggests reflexive behavior as opposed to motivation
releasing behavior so stopping reflexes can consistently override any released
motivation actions. (a good experiment would be to compare the responses when
reflex stopping actions are triggered at the same time as the reflexive escape
actions).
A more sensitive and specialized version of a stopping neuron is found
in the cement gland located near the "chin" area of the tadpole. This gland
secretes mucus over small hairs making it very sensitive to pressure.
Yet these
stopping neurons are not the only types of neurons found in the head.
Types that trigger swimming actions, called rapid-transient detectors, are also found. They produce
very short pulses (usually only one to four action potentials) and habituate
very rapidly to all tactile events except vibrations. A vibration frequency of 50
Hz. results in an action potential for each cycle for 0.3 seconds. Vibrations like
this in water would tend to signal the presence of moving creatures which for most
tadpoles means danger.. The greater the frequency of the vibration the more likely
it is due to background noise. They have a mean receptive field size of 0.015 square
mm.
References
Beard, J. (1896) The History of a Transient Nervous Apparatus in
Certain Ichthyopsida. Zool Jb. 9, 319-426
Roberts, Alan and Smyth, Deborah (1974)
The Development of a Dual touch Sensory System in Embryos of the Amphibian
Xenopus laevis, Journal of Comparative Physiology, 88:31-42
Roberts, Alan
and Hayes, B.P. (1977) The Anatomy and function of "Free" Nerve Endings in
an Amphibian skin Sensory System, Proc. R. Soc. Lond. B. 196: 415-429
Roberts,
Alan (1980) The Function and Role of Two types of Mechanoreceptive "Free"
Nerve Endings in the Head Skin of Ambhibian Embryos, Journal of Comparative
Physiology, 135:341-348
Rohon, V. (1885) Zur Histogenese des Ruckenmarkes der Forelle.
S.B bayer. Akad. Wiss. Math. (Phys. Kl. Jahrg. (1884) 14, 39-57