Application of neurosonography to experimental physiology.

When Horsley and Clark invented the stereotaxic technique they revolutionized experimental neurobiology. For the first time it became possible to repeatably place experimental or surgical probes at precise locations within the skull. Unfortunately, variations in the position and size of neuroanatomical structures within the cranium have always limited the efficiency of this technology. Recent advances in diagnostic medical ultrasonography, however, allow for the real-time visualization of anatomical structures, in some cases with resolutions of up to 150 microm. We report here that commercially available ultrasonographs can be used in the laboratory to generate real-time in vivo images of brain structures in both anesthetized and awake-behaving animals. We found that ultrasonic imaging is compatible with many types of experimental probes including single neuron recording electrodes, microinjection pipettes, and electrodes for producing electrolytic lesions. Ultrasonic imaging can be used to place, monitor and visualize these probes in vivo. In our hands, commercially available ultrasonic probes designed for pediatric use allowed us to visualize anatomical structures with sub-millimeter resolution in primate brains. Finally, ultrasonic imaging allowed us to reduce the risk of accidentally damaging major blood vessels, greatly reducing the incidence of stroke as an unintended complication of an experimental neurosurgical procedure. Diagnostic ultrasound holds the promise of reducing the uncertainty associated with stereotaxic surgery, an improvement which would significantly improve the efficiency of many neurobiological investigations, reducing the number of animal subjects employed in this research. While this demonstration focuses on sonographic imaging in non-human primates, similar advances should also be possible for studies in other species, including rodents.

Stimulus probability directs spatial attention: an enhancement of sensitivity in humans and monkeys.

We examined whether improvements in sensory processing, defined as changes in sensitivity, could be elicited in a simple luminance discrimination task without eliciting concomitant changes in decision processing. To this end we developed a task, for use in both humans and monkeys, in which prior knowledge about where a discriminative stimulus was likely to appear (1) offered no decisional advantage in solving our task and (2) could be parametrically varied to yield a psychometric function. We found that if we parametrically varied the quality of prior knowledge, by increasing the probability, and thus the certainty, that a discriminative stimulus would appear at a particular location under these conditions, luminance discrimination improved for both human and monkey subjects. This improvement was correlated with an enhancement in sensory processing, but not with any systematic changes in decisional processing, as assessed by signal detection theory. These results suggest that (1) sensory processing and decision processing can be separated by task design and (2) systematic changes in prior knowledge about where a stimulus may appear can lead to systematic changes in sensitivity; providing a psychometric function for the influence of prior knowledge on perceptual sensitivity. Importantly, these results were obtained from both human and monkey subjects. Similar task designs could be used in physiological studies attempting to generate linking hypotheses between psychometric and neurometric functions, ultimately allowing changes in perceptual sensitivity to be linked to changes in an underlying neural substrate.

Disinhibition of the superior colliculus restores orienting to visual stimuli in the hemianopic field of the cat.

Following unilateral removal of all known visual cortical areas, a cat is rendered hemianopic in the contralateral visual field. Visual orientation can be restored to the blind hemifield by transection of the commissure of the superior colliculus or by destruction of the superior colliculus (SC) or the substantia nigra pars reticulata (SNpr) contralateral to the cortical lesion. It is hypothesized that a mechanism mediating recovery is disinhibition of the SC ipsilateral to the cortical lesion. The ipsilateral nigrotectal projection exerts a robust inhibitory tone onto cells in the SC. However, ibotenic acid destruction of SNpr neurons, which should decrease inhibition onto the SC, does not result in recovery. The failure of ipsilateral SNpr lesions to produce recovery puts into question the validity of SC disinhibition as a mechanism of recovery. We directly tested the disinhibition hypothesis by reversibly disinhibiting the SC ipsilateral to a visual cortical lesion with a gamma-aminobutyric acid (GABA)A antagonist, bicuculline methiodide. In accordance with the hypothesis, transient disinhibition of the SC restored visual orienting for several hours in three of eight animals. Recovery was not a volume or pH effect and was distinct from the release of irrepressible motor effects (i.e., approach and avoidance behaviors) seen within the first hour after injection. Thus, in the absence of all visual cortical areas unilaterally, disinhibition of the SC can transiently restore the ability of the cat to orient to visual stimuli in the previously "blind" hemifield.

Ibotenic acid lesions of the substantia nigra pars reticulata ipsilateral to a visual cortical lesion fail to restore visual orienting responses in the cat.

Unilateral removal of all known visual cortical areas in the cat renders the animal hemianopic in the contralateral visual field as measured by visual perimetry and other behavioral tests. We have shown that visual orientation behavior can be restored to the previously blind hemifield by destruction of a critical zone in the substantia nigra pars reticulata contralateral to a cortical lesion (Wallace et al., J. Comp. Neurol. 296:222-252, 1990). The model proposed to explain this recovery postulates that damage to the crossed nigrotectal projection disinhibits the superior colliculus ipsilateral to the cortical lesion and this leads to recovery. If disinhibition can account for recovery, then destruction of the uncrossed nigrotectal projection, which is known to exert a tonic inhibition on the superior colliculus, should also result in recovery. We made unilateral visual cortical ablations and ipsilateral ibotenic acid lesions of the substantia nigra pars reticulata. Visual orienting behavior was assessed in animals for a period of 4 to 31 weeks. Contrary to the prediction of the model, we failed to observe a recovery of visual orienting behavior in the blind hemifield in any of 23 animals.

Unilateral nigrostriatal lesions induce a bilateral increase in glutamate decarboxylase messenger RNA in the reticular thalamic nucleus.

The reticular thalamic nucleus consists of densely packed neurons containing the neurotransmitter GABA. It surrounds the lateral border of the thalamus, has extensive reciprocal connections with thalamocortical neurons, and is thought to be involved in attentional processes. The reticular thalamic nucleus also receives direct and indirect inputs from the basal ganglia, suggesting that it may be involved in relaying motor information to the thalamus and cortex. We examined the possibility that decreased dopaminergic transmission in the basal ganglia indirectly affects the reticular thalamic nucleus. Rats received unilateral 6-hydroxydopamine lesions of the substantia nigra pars compacta and were killed two or three weeks after the lesion. Sections of the reticular thalamic nucleus were processed for in situ hybridization histochemistry at the single cell level with RNA probes for both isoforms of glutamate decarboxylase (M(r) 65,000: glutamate decarboxylase 65 and M(r) 67,000: glutamate decarboxylase 67), the rate limiting enzyme of GABA synthesis. Unilateral nigrostriatal dopaminergic lesions induced a topographically specific, bilateral increase in glutamate decarboxylase 67 messenger RNA in neurons of the lateral and ventral reticular thalamic nucleus. A much smaller increase in glutamate decarboxylase 65 messenger RNA was observed which was significant only ipsilateral to the lesion. Short- (seven day) and long-term (eight month) treatments with the antipsychotic drug haloperidol, in regimens that preferentially block D2 dopamine receptors, induced catalepsy and orofacial dyskinesia, respectively, but did not alter glutamate decarboxylase 67 messenger RNA levels in the reticular thalamic nucleus. Thus, loss of dopaminergic terminals, but not blockade of D2 dopamine receptors, induced the effects observed in the reticular thalamic nucleus. The results reveal a novel bilateral effect of unilateral dopamine depletion. In view of the role of the reticular thalamic nucleus in tremor and attentional processes, which are altered in Parkinson's disease, this effect may contribute to the clinical manifestations of nigrostriatal dopamine depletion.

Ibotenic acid lesions of the superior colliculus produce longer lasting deficits in visual orienting behavior than aspiration lesions in the cat.

We compared the effects of unilateral surgical aspiration and ibotenic acid produced lesions of the superior colliculus (SC) on visual orienting behavior in 20 cats. Four animals with aspiration lesions initially showed an hemianopia in the contralateral hemifield which recovered fully in 4.5 weeks or less. These lesions also destroyed axons in the commissure of the superior colliculus (CS). In 9 animals we produced complete loss of cells in one SC, with preservation of axons in the CSC, by injections of ibotenic acid. In these animals the contralateral hemianopia persisted for an average of 16.6 weeks, but may have persisted longer had we not intervened by either sacrificing the animal or ablating the visual cortex contralateral to the SC lesion. The cortical lesion produced an immediate hemianopia in the contralateral hemifield and a recovery in the previously hemianopic ('collicular') hemifield. In the remaining 7 animals with attempted ibotenic acid lesions, 5 had incomplete lesions and 2 others sustained major damage to the SC as well as the CSC. These 7 animals recovered visual orienting on an average of 3.0 weeks postoperatively. We conclude that unilateral loss of collicular cell function and the presence of fibers coursing through the commissure of the superior colliculus are both necessary for the prolonged deficit in visual orienting behavior. We suggest that competition between the two hemifields may play a role in the hemianopia caused by collicular manipulations and that the cholinergic pathway from the pedunculopontine nucleus to the contralateral SC via the CSC may be involved.

Subthalamic nucleus lesions: widespread effects on changes in gene expression induced by nigrostriatal dopamine depletion in rats.

Lesions of the subthalamic nucleus block behavioral effects of nigrostriatal dopamine depletion in rats and primates, but the contribution of this region to the molecular effects of dopaminergic lesions is unknown. The effects of subthalamic nucleus lesions alone or in combination with a 6-hydroxydopamine-induced lesion of the substantia nigra were examined in adult rats. Unilateral subthalamic nucleus lesions caused ipsiversive rotation after peripheral administration of apomorphine and a small decrease in glutamic acid decarboxylase (GAD) mRNA in the ipsilateral globus pallidus (external pallidum). Confirming previous results, nigrostriatal dopaminergic lesions caused contraversive rotation after apomorphine injection, and increased enkephalin mRNA in the striatum, GAD mRNA in the globus pallidus, and somatostatin mRNA in the entopeduncular nucleus (internal pallidum) ipsilateral to the lesion. In addition, the lesion decreased substance P mRNA in the ipsilateral striatum compared to the contralateral side, and GAD mRNA in the contralateral entopeduncular nucleus. These effects were abolished in rats with lesions of the subthalamic nucleus and substantia nigra on the same side. Thus, the subthalamic lesions prevented changes in gene expression induced by dopamine depletion, not only in regions receiving a direct input from the subthalamic nucleus (ipsilateral pallidum), but also in regions which do not (striatum and contralateral pallidum). This suggests that polysynaptic pathways regulated by the subthalamic nucleus contribute to the effects of dopaminergic lesions in many regions of the basal ganglia. This pivotal role of the subthalamic nucleus may account for the beneficial effects of subthalamic nucleus lesions on motor symptoms resulting from dopamine depletion.