[Activation of axo-axonic synapses by serial impulses in afferent C fibres: the gate-control theory disproved by Manfred Zimmermann.].

The gate-control theory of pain, as originally proposed by Melzack and Wall [8], is nothing but a hypothesis concerning the spinal processing of non-noxious and noxious afferent information. Its basic tenant is that the P cells (projecting neurons) convey noxious information to supraspinal pain systems only after a critical threshold of excitation has been passed, and that access to the P cells is controlled by the SG cells (cells of the substantia gelatinosa Rolandi) or, in other words, the SG cells act as the gate. Since the primary afferent fibres have monosynaptic connections with the P cells the gate can only operate-and this is the critical point of the whole hypothesis-via presynaptic inhibition exerted by axoaxonic contacts on these afferents (Fig. 1). The SG cells are excited by thick (low-threshold) afferent fibres, whereas input from fine (noxious) afferents has inhibitory effects. Low-threshold afferent input, therefore, produces little activation of P cells, since the collateral activation of the SG cells leads to vigorous activation of the presynaptic inhibitory gate (Fig. 1 a). But as soon as noxious input via fine afferent units inhibits the SG cells the gate will be opened and the P cells activated (Fig. 1b). Presynaptic inhibition of primary afferent fibres is accompanied by primary afferent depolarization (PAD), which can be recorded from dorsal root filaments as a negative potential change, the dorsal root potential or DRP (Fig. 4). Removal of tonic presynaptic inhibition should involve a primary afferent hyperpolarization (PAH), and this should evoke a positive DRP. Mendell and Wall (1964) claimed to have recorded such positive DRP following preferential or exclusive activation of fine afferent units (group III [A] and IV [C] fibres). This exclusive and therefore crucial experimental support of the gate-control theory has not been confirmed in other laboratories. In particular, as shown in this paper, a series of very elegant and convincing experiments by Manfred Zimmermann gave unequivocal evidence that afferent input via fine afferents produces only PAD, under a variety of experimental conditions. A number of other experimental approaches have also so far failed to demonstrate any PAH following afferent input via fine somatic afferent fibres. Thus, the gatecontrol theory has been disproved by these experiments. As a consequence, Melzack and Wall [9] have now modified their hypothesis considerably. Its present formulation is not much more than a very general statement to the effect that all kinds of afferent input, including noxious input, is under the modulating influence of various mechanisms operating both at the spinal level and from supraspinal (descending) structures. No implications for therapy can be derived directly from such an undefined, in many ways trivial, assumption.