Ten years ago, neuroimaging was largely restricted to determining the localization of pathological lesions in the human brain. Due to the rapid advances in magnetic resonance imaging (MRI) and related technologies, methods have now been developed to determine the precise functional importance of brain lesions, how cognitive operations are carried out within the brain and why they fail.
MRI is an example of the technological development that is no longer restricted to the crude location of a brain lesion. The recently introduced analytical approach enables the size of the image structures in the MRI to be determined, thereby enabling the investigator to track small changes in the structure over time. This permits the abnormal pattern of development in grey or white matter to be detected. For example, using such techniques it has been possible to track the distribution of grey and white matter during the onset of schizophrenia in children, and to show that the condition is characterized by an abnormal time course in the reduction in grey matter in several brain regions. Such changes suggest that there are abnormalities in synaptic pruning which normally takes place during the early stages of neurodevelopment. A further extension of the MRI technique, the diffusion-tensor technique (whereby the diffusibility of water molecules is rendered visible by the preferential orientation of their movement in neurons), has enabled the researcher to track major fibre bundles in white matter.
Another important change has been the shift from positron emission tomography (PET) to MRI-based techniques for the indirect measurement of neuronal activity. Nevertheless, PET and single photon emission computed tomography (SPECT) remain the only viable techniques for studying ligand binding in the human brain and the increasing resolution of PET is constantly improving as new detectors are developed. Recently, new radiotracers have enabled signal transduction mechanisms and gene expression to be evaluated in the human brain.
Specific molecules can now be determined in the human brain by means of magnetic resonance spectroscopy (MRS). Such methods have enabled the glutamate-GABA system to be assessed during neuronal activation and have shown how this pathway is defective in depression and altered by pharmacological treatments. Unlike most of the techniques described here, MRS is also applicable to analysing changes in energy metabolism and the glutamate/GABA pathway in the rat brain.
Undoubtedly, one of the most important advances has been in functional magnetic resonance imaging (fMRI), which enables the researcher to distinguish between the encoding and retrieval phases of memory. An additional application of the fMRI technique in the study of mood and emotion has revealed that the systems involved are widely distributed throughout the brain. However, despite these improvements it will never be possible using fMRI to approach the accuracy of event-related potential (ERP) methods which can quantify events occurring over milliseconds. ERP methods have now been combined with fMRI so that it is possible, for example, to measure signals that occur in visual tasks when they arrive at the cortex, how the signals are modulated by attention and how they are decoded into semantic information. Once these intricate processes have been elucidated in normal behaviour, it should be possible to apply them to specific psychiatric disorders such as schizophrenia where there appears to be a major deficit in the processing of visual stimuli.
Finally, it has now become possible to use imaging methods to study the functional interaction between different brain regions. This has been made possible by the development of effective connectivity mapping which is based on moment-to-moment relationships between fMRI signals in different brain regions to create equations which enable the contribution of the activity of one brain region with another to be quantified. Another approach has been to combine fMRI with transcranial magnetic stimulation (TMS). In this technique, areas of the brain are directly stimulated by magnetic currents and the resulting changes in brain regions quantified by fMRI.
To date, these techniques have been applied almost entirely to man. Over the next decade it will be equally important to further refine them so they may become applicable to the brains of experimental animals, particularly rodents. So far it has only been possible to determine gross structural changes in the rat, for example, using MRI.
Some examples of the application of new imaging methods to psychopharmacology
(a) Application of MRS
It is well known that GABA is the major inhibitory neurotransmitter in the mammalian cerebral cortex and that the functional activity of the GABAergic system is disrupted in several neurological and psychiatric disorders. Furthermore, several classes of psychotropic drugs are known to specifically affect the GABAergic system. Thus MRS should be an ideal method for assessing changes in the glutamate-GABA system in both man and animals.
The initial studies using MRS were performed in rats following the administration of the antiepileptic drug, and GABA transaminase inhibitor, vigabatrin and subsequently in epileptic patients being treated with the drug. In the MRS studies on epileptic patients, it was shown that vigabatrin failed to raise the GABA concentration at a dose that exceeded 3g/day (an antiepileptic dose). This led to a detailed analysis of the mechanism regulating the GABA concentration in the brain. The enzyme synthesizing GABA from glutamate, glutamate decarboxylase (GAD), exists in two major isoforms (GAD 67 and 65) in the brain and these are products of separate genes, differentially distributed (GAD 67 in the cytoplasm and GAD 65 in synaptic terminals) and have different kinetic properties. Using MRS, it was found that GAD 67 activity was reduced in response to the elevated GABA concentration and that, in rats, vigabatrin selectively inhibited the GAD 67 isoform. Other studies using the MRS technique have shown that several antiepileptic drugs with no known action on the GABAergic system (for example, GABApentin, topiramate and lamotrigine) also increase the concentration of GABA in vivo. MRS studies also showed that the regulation of GABA metabolism was closely integrated with GABAergic function. In epileptic patients, these studies showed that the GABA concentration was decreased and that it was the cytosolic GABA concentration which was important in the suppression of seizures by antiepileptic drugs. Chronic vigabatrin administration was shown to reduce the seizure frequency in parallel with the rise in cytosolic GABA.
In addition to epilepsy, reduced GABA has been recorded in patients with unipolar depression, following alcohol withdrawal and in hepatic encephalopathy. The finding that the concentration of GABA is reduced in depression is unexpected as there is no evidence that the disorder is associated with an increased cortical excitability. One possibility is that the reduction in GABA is a reflection of a decreased availability in its excitatory amino acid precursor glutamate.
(b) Application of TMS
A fundamental problem with conventional functional imaging has been an inability to probe the causal relationship between regional brain activity and behaviour. For example, if a brain region utilizes more glucose or oxygen while the subject performs a behavioural task, it is only possible to conclude that the change in regional activity correlates with the behaviour; a causal relationship between the metabolic and behavioural changes can only be inferred. By combining TMS with fMRI it is now possible to directly test how information flows within the brain.
With TMS, a brief but powerful electric current is passed through a small coil held against the scalp of a conscious patient. This generates a powerful local magnetic field which passes unimpeded through the skull and induces a weaker, less focused electric current within the brain. Due to the non-invasive nature of this method, the important physiological effects of TMS are likely to be a consequence of the density of the electric current and the electric field which is induced in the cortex. It is believed that the induced electrical fields cause neuronal depolarization which changes the neuro-transmitter release mechanisms.
TMS has now been combined with glucose utilization studies and fMRI.
Repetitive TMS, unlike electroconvulsive therapy (ECT), uses sub-convulsive stimuli to treat depression. Compared to ECT, TMS has a potential to target specific brain regions and to stimulate brain areas thought to be primarily involved in depression while sparing areas like the hippocampus, thereby reducing the probability of cognitive side effects. However, the therapeutic efficacy of TMS as a treatment for depression is, unlike ECT, modest. Most TMS studies use high-frequency, fast stimulation (> 10 Hz) over the left dorsolateral prefrontal cortex, an area which has been shown to be hypofunctional in PET and electroencephalogram (EEG) studies of depressed patients. Most ''open'' and double-blind studies have confirmed that TMS has a modest antidepressant response in non-psychotically depressed patients. No seizures or cognitive side effects have so far been reported following fast TMS, pain at the treatment site being the only recorded problem.
Hopefully the combination of TMS with fMRI will enable the more precise location of the regional dysfunction in depression to be located and thereby enable the neuronal pathways concerned to be identified. To date, the early studies of TMS with fMRI have shown that the effects of TMS occur in brain regions distant from the site of stimulation, including the caudate, orbitofrontal cortex and the cerebellum.
Was this article helpful?