Lecture 31 - Flexibility and adaptability in neural systems continued

Bi-stability in a sensory neuron(the gastropyloric receptor GPR2)

• GPR2 is a stretch receptor in the stomatogastric system, like the crayfish MRO and the locust wing hinge stretch receptor
• monitors the activity of the gastric muscle gm9 and the pyloric muscle cpv3a
• GPR2 responds to passive stretch or active tension in these muscles
• exhibits two stable states: spiking (A in figure below) and bursting (B in figure below) mode

Spiking Mode...

In the spiking mode GPR2 responds to continuous stretch (DC) with an increase in spike rate.

• measured after initial adaptation
• approximately linear response
• upper firing rate of 4.5 Hz (measured at 0.52 mm stretch)

In the spiking mode GPR2 responds to oscillatory stretch (AC) with oscillations in spike rate.

• less adaptation than DC stretch
• upper firing rate of 15 Hz (measured at 0.28 mm stretch)

Bursting Mode...

In the bursting mode GPR2 responds to constant (DC) stretch with increase in burst rate.

• burst rate remains constant during stretch
• approximately linear response

However, in the bursting mode the GPR2 burst rate does not vary to oscillatory stretch.

How does GPR2 encode information?

• Slow movements (minutes to hours) are encoded by the bursting rate in the bursting mode - for example stretching when the gut is filled.
• Fast movements (seconds) are encoded by the spike rate in the spiking mode - for example the gastric mill rhythm
• May pass these two different kinds of information to separate targets by differential release of transmitters.

Escape swimming in Tritonia

Tritonia is a relatively attractive sea slug, and it is considered particularly attractive by some kinds of sea-star (starfish) which think it is a tasty snack. Not wanting to be eaten, for obvious reasons, the wily Tritonia has an escape behaviour triggered by (among other things) the odour of an approaching sea-star. The fixed action pattern consists of alternating bouts of dorsal and ventral flexion and contraction, which lift the slug into the water column. Once in the water column the currents (hopefully!) carry it away from the voracious echinoderm. The pattern of movement continues for some time after the stimulus has been removed (four to seven cycles of contraction and flexion).

The neural basis of Tritonia escape swimming

The escape rhythm is generated by three groups of interneurons on each side of the brain.

• 3 dorsal swim interneurons (DSI)
• 2 ventral swim interneurons (VSI)
• 1 C2 interneuron

These neurons are also used for other, smaller motions under different circumstances. When an escape response is triggered the neurons are strongly excited, but excitation decreases over time. The excitation comes from a single interneuron on each side of the brain, known as the dorsal ramp interneuron (DRI) Excitation of the DRI is necessary for swimming to take begin, and if this excitation is removed the bout of swimming will stop.

The neural circuit for swimming

• DRIs receive input from the sensory receptors
• Each DRI makes excitatory synapses with 3 DSIs
• DSIs excite C2 and inhibit VSIs
• VSIs, in turn, inhibit DSIs and C2
• C2 excites VSIs, and both excites and inhibits DSIs

Intrinsic neuromodulation in the swimming circuit

The excitation of DSIs by DRI has two functions

• DSIs play a role is generating the swim rhythm via synapses with C2 and VSIs
• DSIs sustain the rhythm by releasing serotonin

C2 function is enhanced by the serotonin released from the DSIs. DSIs fire rapid burst at beginning of swimming episode, releasing serotonin into the system. This neuromodulation probably 'sets' the circuit into escape mode by increasing the sensitivity of C2.

An overview of flexibility in neural systems

• use input to reconfigure network dynamics - crustacean stomatogastric ganglion (Simmons & Young - CH 7)
• utilise bi-stable or multi-stable states of individual neurons - R15 interneuron in Aplysia (Lechner et al., 1996)
• use input to reconfigure dynamics of networks of multi-stable neurons - neural network model of quadrupedal gaits (Canavier et al., 1997)
• use multi-stable sensory neurons to encode information on different time scales - the gastropyloric receptor GPR2 (Birmingham et al., 1999)
• use intrinsic neuromodulation to generate and sustain motor patterns - escape swimming in Tritonia (Simmons & Young - CH 7)