Vasopressin in the locus coeruleus and dorsal pontine tegmentum affects posture and vestibulospinal reflexes.

Vasopressin (VP) acts on both the locus coeruleus (LC) neurons and the neighbouring dorsal pontine reticular formation (PRF) neurons by exciting them. Experiments performed in precollicular decerebrate cats have shown that microinjection of 0.25 x 10(-11) micrograms VP into the LC complex of one side increased the extensor rigidity of the ipsilateral limbs, while rigidity of the contralateral limbs remained unmodified or slightly decreased. The amplitude of modulation and thus the response gain of both the ipsilateral and the contralateral forelimb extensor triceps brachii to sinusoidal roll tilt of the animal (at 0.15 Hz, +/- 10 degrees), leading to stimulation of labyrinth receptors, decreased significantly, while there was only a slight decrease in phase lead of the responses. These effects occurred 5-10 min after the injection, were fully developed within 30 min and disappeared in about 2 h. VP activation of presumed noradrenergic LC neurons had a facilitatory influence on ipsilateral limb extensor motoneurons, either directly through the coeruleospinal (CS) pathway, or indirectly by inhibiting the dorsal PRF and the related medullary inhibitory reticulospinal (RS) neurons. Moreover, because the facilitatory CS neurons fire out-of-phase with respect to the excitatory VS neurons, we postulated that the higher the firing rate of the CS neurons in the animal at rest, the greater the disfacilitation affecting the limb extensor motoneurons during side-down animal tilt. These motoneurons would then respond less efficiently to the excitatory VS volleys elicited for the same direction of animal orientation, leading to a reduced gain of the EMG responses of the forelimb extensors to labyrinth stimulation. In contrast to these findings, unilateral injections of the same dose of VP immediately ventral to the LC, i.e., in the peri-LC alpha and the surrounding dorsal PRF, where presumed cholinergic neurons are located, decreased extensor rigidity in the ipsilateral limbs while that of the contralateral limbs either decreased or increased. The same injection also produced either a moderate or a marked increase in gain of the multiunit EMG response of the ipsilateral triceps brachii to animal tilt. In the first instance the response gain of the contralateral triceps brachii to animal tilt increased slightly, while the corresponding response pattern remained unmodified, as shown for the ipsilateral responses (increased EMG activity during ipsilateral tilt and decreased activity during contralateral tilt). In the second instance, however, the response gain of the contralateral triceps brachii showed only slight changes, while the pattern of response was reversed. These effects occurred 5-20 min after the injection, developed fully within 20-60 min and disappeared in 2-3 h. We postulated that VP increased the discharge of the dorsal PRF neurons and the related medullary inhibitory RS neurons of the injected side, leading to reduced postural activity of the ipsilateral limbs. However, because these inhibitory RS neurons fire out-of-phase with respect to the excitatory VS neurons, it appeared that the higher the firing rate of the RS neurons in the animal at rest, the greater the disinhibition affecting the limb extensor motoneurons during ipsilateral tilt. These motoneurons would then respond more efficiently to the same excitatory VS volleys elicited by given parameters of stimulation, leading to an increased gain of the EMG responses. The contralateral effects could be attributed to crossed excitation by dorsal PRF neurons of one side, either of medullary inhibitory RS neurons or of excitatory CS neurons of the opposite side, respectively. We conclude that VP controls posture and gain of the VS reflex by acting on LC neurons as well as on dorsal PRF and the related medullary inhibitory RS neurons.