PET and MRI – better together
Since their inception, imaging studies – whether MRI to investigate anatomical structure in detail or PET to visualize the workings of the body’s tissues – have yielded amazing amounts of information in our search to better understand the brain. Now able to be used simultaneously in purpose built machines, the investigation of drug effects in the living human brain becomes even more informative. Here Gitte Moos Knudsen, Denmark, describes some recent findings using these techniques.
Cimbi – an atlas of the brain
The Cimbi database is a multimodality neuroimaging database covering healthy volunteers and patients. Of particular relevance to neurobiologists interested in the serotonergic transmitter system, it contains data gathered from normal subjects concerning 5-HT1A, 5-HT1B, 5-HT2A, and 5-HT4 and the 5-HT transporter (5-HTT).The Cimbi database currently comprises a total of 1,200 PET and structural and functional MRI scans.
Use of both high-resolution PET scanning and structural MRI, together with a novel quantification methodology to correct for partial volume effects, has facilitated the creation of a population-based atlas that details the 5-HT receptors and 5-HTT.
Comparison of this atlas with the ALLAN postmorten mRNA expression atlas shows good correlation in the siting of the various 5-HT receptors, though not with 5-HTT. This may be because, while the transporter acts presynaptically, it is likely to be produced elsewhere.
Due to be published shortly, scientists will soon have access to a valuable resource - a whole brain, in vivo atlas of 5-HT targets created by multimodal imaging that will further facilitate study and manipulation of the serotonergic network
5-HT concentrations by proxy
Through use of a multimodal approach, 5HT4 receptor binding has been shown to be a suitable proxy or in vivo biomarker for brain serotonin concentrations.
Before and after a three-week intervention with either placebo or the selective serotonin reuptake inhibitor (SSRI) fluoxetine, healthy volunteers underwent fMRI and PET (using a 5-HT4 receptor radioligand). There were no significant between-group differences in emotional face processing on fMRI until individual SSRI intervention-associated changes in central 5-HT levels, as measured with the 5-HT4 receptor radioligand, were taken into account. It then appeared that the greater the increase in central 5-HT levels (as assessed using 5-HT4 as proxy), the lower the threat-related amygdala reactivity. This suggests that individual changes in brain 5-HT levels are linked to threat-related amygdala reactivity.
Simultaneous fMRI-PET – best of both worlds
Simultaneous fMRI-PET allows both assessment of pharmacological interventions and capture of regional differences. For example, occupancy of the 5-HT1B receptor has been studied when a specific receptor partial agonist, an agonist or an antagonist were given in animal models. By combining the two imaging approaches it was possible to tease out any connections that were not apparent at the beginning of the study.
Administration of the partial agonist, as expected, brought about a decline in the 5-HT1B receptor binding potential due to an increase in the occupancy of the receptor binding sites. However, fMRI showed the partial agonist had elicited an unexpected biphasic change in cerebral blood volume.
Comparing the effects of administering the agonist to the partial agonist revealed that both had similar extracranial tissue effects. However, only the partial agonist generated the intracranial biphasic cerebral blood volume response. Might the haemodynamic effect caused by the partial agonist be blocking 5-HT signaling in some way? Use of an antagonist suggested this was not the case as it too promoted changes in the cerebral blood volume, albeit to a different extent when compared to those of the agonist.
Correlation of the PET and the MRI signal changes noted with each scenario suggests that the partial agonist has a lower capacity to cross the blood brain barrier. While the agonist and antagonist exhibit similar occupancy of the receptor, their effects on the cerebral blood volume are very different.
Future possibilities
Exactly how pharmacologically-induced changes in receptor occupancy relate to regional blood flow changes remains to be better understood. Clearly, however, simultaneous use of PET and MRI has much to offer in the future assessment of drug blood brain barrier penetrance, target occupancy, drug function and downstream effects.
Further information on Cimbi can be obtained at www.cimbi.dk
