Modelling recurrent discharge in the spinal α-motoneuron: reappraisal of the F wave.

Antidromic impulses travelling along a motor fibre can fail in invading the soma, due to their passing through neuronal regions characterised by non-uniform geometry and a reduced nerve conduction safety factor: the myelinated–unmyelinated junction, the axon–soma junction and profuse branching of dendrite terminals (Eccles, 1955). In particular, the axon–soma junction is clearly a non-uniform area due to the abrupt diameter change of the cable conductor, which affects both the membrane capacitive load and the intra-axial current downstream of the antidromic impulse progression. A larger soma diameter in fact increases the membrane area which needs to be charged before opening channels (i.e., capacitive load), thereby consequently reducing the amplitude of the action potential and rate of rise. However, the increase of diameter also reduces axial resistance, which, in turn, increases axial current at the expense of the transmembrane current (Goldstein and Rall, 1974; Joyner et al., 1980; López-Aguado et al., 2002). In this way, the antidromic axial currents at the axon–somatic junction could fail in propagating the impulse, or could slow down rise time of the somatic antidromic action potential, respectively causing either a block or a delay. Among the spinal a-motoneurons showing antidromic invasion of the soma, only a fraction is able to produce a re-excitation of the initial segment leading to an orthodromically conducted action potential, known as recurrent discharge. Recurrent discharge is the basis of a clinical neurophysiological test, the Fwave, mainly used to detect nerve conduction abnormalities in proximal segments of peripheral nerves (Kimura, 2001). In addition, the test is used in experimental paradigms so as to obtain information on spinal motoneuron excitability. In these cases, a direct relationship between motoneuron excitability and F-wave amplitude or frequency is often implicitly assumed, despite the fact that previous works clearly showed that such excitability influences the amplitude and persistence of the F wave based on complex and scarcely predictable mechanisms. In particular, it is well established that both increases and decreases in motoneuronal excitability can decrease the incidence of F waves (Gogan et al., 1984; Hultborn and Nielsen, 1995; Espiritu et al., 2003), even if this is not always recognised clinically. In order to show the reciprocal interplay of the axonal initial segment and the soma leading to recurrent discharge in detail, we developed a reduced conductance-based multi-compartmental computational model of a spinal a-motoneuron, by means of the NEURON v7.2 simulation environment (Carnevale and Hines, 2006).