Limited predictive power of DAT mRNA mappings for DAT protein expressions

Dopamine transporter (DAT) is a transmembrane protein that translocates dopamine from the extracellular space into the presynaptic neurons. We aimed to investigate the predictive power of DAT mRNA for DAT protein expression measured from positron emission tomography (PET). Thirty-ve healthy subjects were imaged with 18 F-FP-CIT PET. Binding potential (BP ND ) from ventral striatum, caudate nucleus, putamen, middle frontal, orbitofrontal, cingulate, parietal, temporal cortices were measured. DAT gene expression investigated in this study was obtained from the freely available Allen Human Brain Atlas (www.brain-map.org) derived from 6 healthy donors.


Abstract
Background Dopamine transporter (DAT) is a transmembrane protein that translocates dopamine from the extracellular space into the presynaptic neurons. We aimed to investigate the predictive power of DAT mRNA for DAT protein expression measured from positron emission tomography (PET).

Methods
Thirty-ve healthy subjects were imaged with 18 F-FP-CIT PET. Binding potential (BP ND ) from ventral striatum, caudate nucleus, putamen, middle frontal, orbitofrontal, cingulate, parietal, temporal cortices were measured. DAT gene expression investigated in this study was obtained from the freely available Allen Human Brain Atlas (www.brainmap.org) derived from 6 healthy donors.

Results
PET-derived BP ND s from 35 subjects, and DAT mRNA expressions of 6 probes from 6 subjects were included in this study. The auto-correlation of PET-derived BP ND s for DAT was intermediate (mean 2 = 0.6588) with 2 ranging from 0.0811 to 1. However, the auto-correlation of mRNA expression was weak across the probes with mean 2 from 0.0889 to 0.2263. Cross-correlations between PET-derived BP ND s, and mRNA expression were weak with mean 2 ranging from 0.0011 to 0.2220 across the probes.

Conclusion
we observed weak associations between DAT mRNA expressions, and DAT protein expressions in human brains. Therefore, there might be a limited predictive power of DAT mRNA mappings for DAT protein expressions in humans.

Background
Dopamine is a neurotransmitter that plays a major role in reward-motivated behavior, and motor control (1). Brain dopamine neurotransmission is regulated by dopamine transporter (DAT), a transmembrane protein that actively translocates dopamine from the extracellular space into the presynaptic neurons in dopaminergic system (2). Dysfunction of DAT has been known to be linked to neuropsychiatric disorders, such as attention-de cit/hyperactivity disorder (3), biopolar disorder (4), and alcoholism (5). In addition, DAT is a major target for various pharmacologically active drugs (2).
Allen Human Brain Atlas is a freely available multi-modal atlas of gene expression and anatomy of all brain regions (6). Six postmortem brains were analyzed for microarray-based expression, in situ hybridization gene expression, and also magnetic resonance imaging-based brain mapping (6). Previously, the predictive power of brain mRNA mappings of Allen Human Brain Atlas was reported after comparison with positron emission tomography (PET)derived protein expressions including serotonin receptor (7-9), serotonin transporter (8,9), opioid receptor (7), monoamine oxidase A (MAO-A) (9, 10). Therefore, we measured striatal DAT availability using dynamic positron emission tomography (PET) scans with 18 F-FP-CIT, a high-a nity radioligand for DAT from a study by Pak et al (11) to measure DAT protein expression.
We hypothesized that DAT mRNA expression can predict DAT protein expression measured from PET scans. Therefore, we investigated the predictive power of DAT mRNA expression with 1) auto-correlation, and 2) crosscorrelation with PET-derived DAT protein expression.

PET data
Thirty-ve healthy, male subjects without brain injury, neuropsychological disorders were included in this study. The majority of the participants in this study were included in a previous study of striatal DAT changes after glucose loading (11). This study was approved by the institutional review board of Pusan National University Hospital. An intravenous bolus injection of 18 F-FP-CIT was administered. The emission data were acquired over 90 mins with 50 frames of progressively increasing durations (15 s × 8 frames, 30 s × 16 frames, 60 s × 10 frames, 240 s × 10 frames, and 300 s × 6 frames) using the Siemens Biograph 40 Truepoint PET/CT (Siemens Healthcare, Knoxville, Tennessee, USA). The dynamic PET data were collected in the 3-dimensional mode, with 148 slices with image sizes of 256 × 256 and pixel sizes of 1.3364 × 1.3364 mm 2 . These were reconstructed by ltered back projection using a Gaussian lter. For a volume-of-interest (VOI)-based analysis, an averaged image (0-10 min after injection) was created from dynamic PET frames and spatially normalized to 15 O-Water PET template in statistical parametric mapping 5 (Wellcome Trust Centre for Neuroimaging, United Kingdom). To extract time-activity curves (TACs) of VOIs from full dynamic PET scans, Oxford-GSK-Imanova striatal atlas (https://fsl.fmrib.ox.ac.uk/fsl) for ventral striatum (VST), caudate nucleus, putamen, and Automated Anatomical Labeling atlas (12) for middle frontal, orbitofrontal, anterior/middle/posterior cingulate, parietal, temporal cortices were applied. DAT availability, expressed in terms of binding potential (BP ND ), were measured by analyzing TACs with the simpli ed reference tissue method (13) with the cerebellum as a reference. Image analysis was done using pmod version 3.6 (PMOD Technologies LLC, Zurich, Switzerland). mRNA data Gene expression information investigated in this study was obtained from the freely available Allen Human Brain Atlas (www.brain-map.org). The details of the procedures regarding Allen Human Brain Atlas are reported previously by Hawrylycz et al (14). We downloaded mRNA values of the dopamine transporter to test for an association with the radiopharmaceutical 18 F-FP-CIT. The dataset of Allen Human Brain Atlas is derived from 6 healthy donors. Allen Human Brain Atlas contains gene expression pro les of DAT throughout the brain with 6 probes. We downloaded DAT mRNA expression data in log2-values for each sample. For VOI-based analysis, mRNA samples were averaged within VST, caudate nucleus, putamen, middle frontal, orbitofrontal, anterior/middle/posterior cingulate, parietal, temporal cortices.

Statistical analysis
Normality was assessed using the D'Agostino & Pearson normality test. Spearman correlation analysis was done for both 1) auto-correlation of BP ND s of 18 F-FP-CIT PET, and DAT mRNA expressions, and 2) cross-correlation between BP ND s of 18 F-FP-CIT PET, and DAT mRNA expressions to ensure validity and consistency. In addition, with mean BP ND s of 18 F-FP-CIT PET, and mean DAT mRNA expressions, Spearman correlation analysis was done for each VOIs.

Results
PET-derived BP ND s from 35 subjects with mean age of 24.4 years (range 20 ~ 31 years), and DAT mRNA expressions of 6 probes from 6 subjects were included in this study. As the limited number of genomic samples was included, normal distribution of DAT mRNA expression could not be assumed. PET-derived BP ND s from VST, caudate nucleus, putamen showed substantially higher levels than those from middle frontal, orbitofrontal, anterior/middle/posterior cingulate, parietal, temporal cortices. However, mRNA expression levels from probe 1 (A_24_P397435), 2 (CUST_942_PI417557136), 3 (CUST_941_PI417557136), 4 (CUST_588_PI416408490), 5 (CUST_927_PI417557136) were systemically lower than those from probe 6 (CUST_15799_PI416261804), which could be due to a lack of sensitivity of these probes (Fig. 1).

Discussion
In this study, we examined the association between in-vivo protein expression of DAT from PET, and DAT mRNA expression from Allen Human Brain Atlas. Among 6 probes for DAT mRNA from Allen Human Brain Atlas database, 5 probes showed much lower expression levels of DAT mRNA than probe 6, probably due to a lack of sensitivity.
Although mRNA expression of probe 6 showed the highest expression levels in each VOI, the auto-correlation between subjects (inter-subject correlation) was weak with mean 2 of 0.2263, and cross-correlation with PET-derived BP ND s was also weak with mean 2 of 0.2220. However, mean mRNA expression of probe 6 showed the signi cant correlation with mean PET-derived BP ND .
Dopamine is a neurotransmitter that involves in reward-motivated behavior, and motor control (1). Brain dopamine neurotransmission is regulated by DAT, which drives reuptake of extracellular dopamine into presynaptic neurons (2).
Dysfunction of DAT has been known to be linked to neuropsychiatric disorders, such as attention-de cit/hyperactivity disorder (3), biopolar disorder (4), and alcoholism (5). In addition, DAT is a major target for various pharmacologically active drugs (2). From genomic maps such as Allen Human Brain Atlas, gene expression can be visualized across whole brain regions yielding insights into the relationship between structure and function of human brain (15), leading to help brain investigation, and drug development (7). Therefore, we investigated the association of DAT between protein expression measured by PET, and mRNA expression from Allen Human Brain Atlas. However, in this study, we observed relatively weak associations between DAT protein expressions, and DAT mRNA expressions regardless of probes. In addition, the correlation between mean PET-derived BP ND s, and mean DAT mRNA expression levels were marginally signi cant in 1 of 6 probes. Therefore, maps of the human mRNA transcription architecture from DNA microarray analysis of DAT might have limited predictive value, after comparison with PET-derived protein expression in this study. Previously, Rizzo et al. explained the possible mechanisms of the limited predictive role of mRNA mapping for protein expression (7). First, posttranscriptional mechanisms of splicing, or translational modi cations might in uence protein expression for each cell type (7,16). In addition, mRNA expression is analyzed in the cytoplasm, while DAT protein is predominantly expressed presynaptically (9), which might affect the limited role of mRNA mapping for DAT.
There are several limitations in this study. First, the sample size of Allen Human Brain Atlas database is small with 6 donors. Second, although mRNA data of 6 probes were included across the regions, mRNA expression from 5 probes were systemically low, probably due to a lack of sensitivity of probes. In addition, the auto-correlation of mRNA expression between individuals was weak, probably due to interindividual variations. Also, DAT of both mRNA expressions, and PET-derived BP ND s might be affected by the functional status of dopaminergic system, dopamine levels in the brain. Lastly, substantia nigra and ventral tegmental area might have a relevant role in the description of mRNA for DAT protein expression.

Conclusion
We observed weak associations between DAT mRNA expressions, and DAT protein expressions in human brains.
Therefore, there might be a limited predictive power of DAT mRNA mappings for DAT protein expressions in humans.

Declarations
Ethics approval and consent to participate All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Declaration of Helsinki and its later amendments or comparable ethical standards. Ethical permission for the study procedures was obtained from the Institutional Review Boards at Pusan National University Hospital. Subject consent has been obtained by Pusan National University Hospital. This article does not contain any studies with animals performed by any of the authors.

Consent for publication
Not applicable.