Introduction:
The serotonin (5-HT) system is highly diverse with 7 families of receptors (5-HT1 to 5-HT7), including 14 subtypes, and a transporter. Serotonin is implicated in a myriad of brain functions and dysfunction of this system is linked
to many disorders [1,2]. We here aimed to generate a high resolution atlas of the cerebral 5-HT receptors and transporter distribution in a healthy population, to be offered to the scientific community. The Center for Integrated
Molecular and Brain Imaging (Cimbi) has been extensively studied the 5-HT system and has accumulated a rich database of healthy subjects including high-resolution structural MRI and image positron emission tomography
(PET) images targeting the receptors 5-HT1A, 5-HT1B, 5-HT2A (agonist and antagonist) and 5-HT4, and the transporter 5-HTT. These receptors and the transporter represent the major components of the 5-HT system that can
be studied with PET neuroimaging. We utilized this data to create a high-resolution surface and volume-based in vivo atlas of these receptors and transporter in healthy subjects.
Methods:
High resolution PET images were acquired using a Siemens ECAT HRRT scanner using the following tracers: [11C]CUMI-101 (5-HT1A, n=8), [11C]AZ10419369 (5-HT1B, n=22), [18F]Altanserin (5-HT2A, n=18), [11C]Cimbi-36
(5-HT2A, n=30), [11C]SB-207145 (5-HT4, n=51), and [11C]DASB (5-HTT, n=106). Corresponding structural MRI were also available for all subjects. Dynamic PET images were reconstructed using a 3D-OSEM-PSF algorithm
[3]. The data was analyzed with FreeSurfer [4] (FS, 5.3) using a surface and a volume stream. The individual cortical surfaces were reconstructed using the structural MRI corrected for gradient non-linearities. Partial volume
correction of the TACs was performed using a region-based voxelwise method [5]. TACs were transformed to the common volume space (MNI152) using Combined Volume-Surface registration [6] and to a common surface
space (fsaverage) using FS. The data was surface or volume smoothed by 5 mm full width half maximum. Kinetic modeling of the nondisplaceable binding potential (BPND) was performed using the FS PET pipeline [7] with a
Multilinear Reference Tissue Model 2 [8] for all tracers, using a high-binding region for estimating k2', with the exception of [18F]Altanserin for which steady-state quantification (BPP) was used. The individual BP maps were
then averaged for each tracer to create the final maps.
Results:
Surface and volume-based average BPND and BPP maps were created for all tracers (Figures 1-4). The availability of structural MRI enabled us to perform accurate normalization in both surface and volume, a quality previously
absent from similar work [9]. These maps highlight key features characterizing the 5-HT system and interesting patterns can be observed across tracers. For example, occipital cortex (pericalcarine) has high density for 5-HT1B,
5-HT2A and 5-HTT while striatum has high density for both 5-HT4 and 5-HTT, but not for other tracers.