Increased Vesicular Monoamine Transporter 2 (VMAT2) and Dopamine Transporter (DAT) Expression in Adolescent Brain Development: A Longitudinal Micro-PET/CT Study in Rodent

Background: Brain development and maturation in adolescence is a complex process with active changes of metabolic and neurotransmission pathways. Positron emission tomography (PET) is a useful imaging modality for tracking metabolic and functional changes in adolescent brain. In this study, changes of glucose metabolism, expression of vesicular monoamine transporter 2 and dopamine transporter during adolescent brain development in rats were investigated with PET/CT. Methods: A longitudinal PET/CT study of age-dependent changes of VMAT2, DAT and glucose metabolism in adolescent brain was conducted in a group of Wistar rats (n = 6) post sequential intravenous injection of 18F-PF-(+)-DTBZ, 11C-CFT, and 18F-FDG, respectively. PET acquisition was performed at 2, 4, 9, and 12 months of age. Radiotracer uptake in different brain regions, including the striatum, cerebellum, and hippocampus, were quantified and recorded as Standardized uptake value (SUV) and striatal specific uptake ratio (SUVR: SUV in brain regions/SUV in cerebellum). Results: Variable uptake of 18F-PF-(+)-DTBZ and 11C-CFT were detected, with highest level uptake in the striatum and accumbens. There was significant age-dependent increase of 18F-PF-(+)-DTBZ and 11C-CFT uptake in the striatum from 2 months of age (SUV: 1.36 ± 0.22, 1.37 ± 0.39, respectively), to 4 months (SUV: 2.22 ± 0.29, 2.04 ± 0.33), 9 months (1.98 ± 0.34, 2.09 ± 0.18), 12 months (SUV: 1.93 ± 0.19, 2.00 ± 0.17) of age, SUV of 18F-FDG also increased from 2 months of age to older ages (SUV in the striatum: 3.71 ± 0.78 at 2 month, 5.28 ± 0.81, 5.14 ± 0.73, 4.94 ± 0.50 at 4, 9, 12 month, respectively). Conclusion: Age-dependent increases of striatal of 18F-FDG, 18F-PF-(+)-DTBZ, and 11C-CFT uptake were detected in rats from 2 to 4 month of age, demonstrating striatal development presents over the first 4 months of age. Four months of age can be considered a safe threshold to launch brain disease studies for exclusion of confusion of continuing tissue development. These findings support further investigation of age-dependent changes in expression of DAT, VMAT2, and glucose metabolism for their potential use as a new imaging biomarker for study of brain development and functional maturation.

development presents over the first 4 months of age. Four months of age can be considered a safe threshold to launch brain disease studies for exclusion of confusion of continuing tissue development. These findings support further investigation of agedependent changes in expression of DAT, VMAT2, and glucose metabolism for their potential use as a new imaging biomarker for study of brain development and functional maturation.

INTRODUCTION
Brain development is a dynamic and complex process, including the change of brain volume and morphology, metabolism, expression of receptors and transporters, and function (Chugani and Phelps, 1986;Chugani, 1998;Khundrakpam et al., 2016). A number of investigations indicated brain regional growth with changes of cortical thickness during brain development and maturation using MR (Sowell et al., 2004;Aljabar et al., 2008;Mengler et al., 2014). However, metabolic and functional changes associated with brain development and maturation remain largely unknown or poorly understood, particularly in adolescent population. Because of vulnerability caused by unmature status of brain development, the adolescent population is at a high risk for development of various mental health problems, such as uncontrolled violence, depression, and drug abuse. It is important to explore metabolic changes and changes of receptors or neurotransmission pathways during brain development and maturation process in adolescent population.
Positron emission tomography (PET) is a particularly useful functional imaging tool for tracking metabolic changes or changes of neuro-receptor expression during brain development using a radiotracer. PET imaging with [ 18 F]fluoro-2-deoxy-D-glucose ( 18 F-FDG) is widely used to detect alterations of cerebral glucose metabolism in brain development, aging, and various brain disorders (Mosconi et al., 2008;Smailagic et al., 2015). Some study present the glucose metabolism change in brain maturation using 18 F-FDG/PET (Choi et al., 2015). Using radioligands binding to brain receptors and transporters, such as dopamine transporter (DAT) and vesicle monoamine transporter 2 (VMAT2), PET is also a useful tool for studying of changes of the expression and function of brain receptors and transporters during brain development and maturation (Alexander et al., 2017).
It is well-known that there are changes of glucose metabolism that have some inherent connections with the development of brain transporters and 18 F-FDG PET may be used to evaluate changes of glucose metabolism (Chugani and Phelps, 1986;Goyal and Raichle, 2018). There are continued efforts to explore changes of other metabolic pathways in brain development such as those related to synapse activity and biometals. DAT are protein molecules located in the dopaminergic nerve terminals, which mediate neuronal uptake of dopamine from extracellular space into extravesicular cytoplasmic compartments. [ 11 C] 2-β-carbomethoxy-3-β-(4-fluorophenyl)tropane ( 11 C-CFT) is a radioligand with high affinity binding to DAT (Brownell et al., 1996). VMAT2 are responsible for packaging and transporting neurotransmitters, such as dopamine, into synaptic vesicles (Eiden and Weihe, 2011). [ 18 F]9-fluoropropyl-(+)dihydrotetrabenazine ( 18 F-FP-(+)-DTBZ) can bind to VMAT2 with high affinity. Both DAT and VMAT2 regulate the synaptic concentration of neurotransmitters in the brain (Hall et al., 2014). Expression level of DAT and VMAT2 reflect the density of dopamine terminals in brain regions, which may change during brain development and pathogenesis of brain disorders such as Parkinson's disease. It was demonstrated that 11 C-CFT and 18 F-FP-(+)-DTBZ could be used for quantification of the density of DAT and VMAT2 in brain (Hsiao et al., 2014;Wood, 2014).
Rat adulthood is defined as full sexual maturity in many studies, and the time window is typically described by age and body weight, which lack clear and unambiguous boundaries. Mengler et al. (2014) found clear indications of ongoing developmental changes of cortical thickness, myelination in the rat brain, which point to the time window of maturity within 3 months of age. In this study, to better characterize the brain development of glucose metabolism, DAT and VMAT2, three radiotracers were utilized to observe the development of rat brains in maturation. We recorded the PET signals of these three radiotracers in Wistar rats from 2 to 12 months of age in a longitudinal study. with this multi-tracer approach, we undertook to chart the interrelations of DAT and VMAT2 in vivo during the early life of rats, and find a time window of brain maturation.

Animals and Radiopharmaceuticals
Wistar rats were purchased from Shanghai Laboratory Animal Center (Shanghai, China) and housed in the animal housing facility, with free access to food and drinking water. 18 F-PF-(+)-DTBZ, 11 C-CFT, and 18 F-FDG were prepared in the radiochemistry facility of PET center, Huashan Hospital, Fudan University for clinical use under requirement of GMP. All small animal experiments were conducted according to a protocol approved by Huashan Hospital, Shanghai, China.

Study Design
A longitudinal, sequential micro-PET/CT study was performed to assess age-dependent changes of VMAT2, DAT, and glucose metabolism in a group of Wistar rats (male, n = 6) from 2 FIGURE 1 | Schematic representation of longitudinal study of change of VMAT2, DAT, and glucose metabolism in rat brains with PET/CT at different ages.
to 12 months of age (equivalent to 12 to 30 years in humans) using 18 F-PF-(+)-DTBZ, 11 C-CFT, and 18 F-FDG, respectively, (Figure 1). Small animal imaging was conducted at 2, 4, 9 months and approximately 12 months of age. Rat ages were defined as the human age equivalents outlined below according to literature (Quinn, 2005).

Micro-PET/CT Imaging
Micro-PET/CT imaging of Wistar rats was conducted by a method described previously using Inveon microPET-CT scanner (Siemens Inc., United States) (Zilliox et al., 2016). Briefly, the rats were anesthetized using 2-3% isoflurane in medical oxygen (1 L/min) at room temperature with an isoflurane vaporizer (Molecular Imaging Products Company, United States). The rats were positioned in a spread-supine position on the imaging bed and subjected to inhalation of the anesthetic during the PET/CT procedure. Static PET/CT imaging was obtained for 10 min at 60 min post intravenous administration of 18 F-FDG (∼10 µCi/g body weight), at 60 min post intravenous administration of 18 F-PF-(+)-DTBZ (∼10 µCi/g body weight), and 45 min post administration of 11 C-CFT (∼20 µCi/g body weight). The imaging time is based on previous dynamic study we performed. PET/CT images were reconstructed using the ordered subsets expectation maximization 3D algorithm (OSEM3D), and data was reviewed using the Inveon Research Workplace (IRW) software (Siemens) and processed by PMOD software (version 3.4, PMOD Technologies Ltd., Zurich, Switzerland).

Quantification of Radiotracer Activity in the Brain Regions of Rats
18 F-FDG PET images were automatically fused using the PMOD FDG rat brain template (Schiffer et al., 2006) while 18 F-PF-(+)-DTBZ and 11 C-CFT PET images were manually fused with the T2-MRI template. 58 brain regions of interest (ROIs) were collected on the PET/CT images with reference to the MR imaging-based atlas (Schiffer atlas) (Schiffer et al., 2006). The regions analyzed included: cerebellum, accumbens (left and right), amygdala (left and right), striatum (left and right), hippocampus posterior (left and right), hypothalamus (left and right), olfactory (left and right), midbrain (left and right), thalamus (left and right), and cortex etc., All collected brain regions with a sample of 18 F-FDG are present in Figure 2. The quantity of 18 F-PF-(+)-DTBZ, 11 C-CFT, and 18 F-FDG activity FIGURE 2 | The ROIs applied to the PET images using MR imaging-based atlas by PMOD (top) and corresponding example for FDG (bottom). The main brain regions included: cerebellum (marigold), accumbens (blue), amygdala (yellow), striatum (red), hippocampus posterior (green), midbrain (purple), thalamus (lime), and cortex etc.
was obtained and recorded as a standardized uptake value (SUV). SUV ratio (SUVR) was calculated by SUV uptake in the specific brain regions compared to that of the cerebellum.

Statistical Analysis
In order to determine whether the SUV and SUVR of [ 18 F]FDG, [ 11 C]CFT, [ 18 F]DTBZ in these rats differ among different ages, we applied two-way analysis of variance (ANOVA), followed by Tukey multiple comparison post-test, with the between-subjects factor representing the different age groups and the withinsubjects factor representing the brain regions. Statistical analyses was performed in Prism (version 7.0, GraphPad Software, San Diego, CA, United States). A P-value of less than 0.05 was considered statistically significant.

Comparison of 18 F-PF-(+)-DTBZ, 11 C-CFT, and 18 F-FDG Uptake in Rat Brains
Bodyweight of these rats increased from 222.0 ± 14.6 to 424.5 ± 7.2 from 2 to 12 months of age, and kept in a relatively stable range between 371.7 ± 32.9 and 424.5 ± 7.2 from 4 to 12 months of age (Supplementary Figure S1). 18 F-PF-(+)-DTBZ and 11 C-CFT displayed similar distribution patterns in rats after intravenous administration. Both radiotracers mainly accumulated in the striatum and accumbens, yet uptake of 18 F-PF-(+)-DTBZ could also be found in the skull (Figure 3). The pattern of 18 F-FDG uptake in rat brains was different from these two tracers, as the cortex, striatum and accumbens had the highest FDG uptake, followed by the cerebellum, while the pituitary gland demonstrated the lowest uptake (Figure 3 and Supplementary Tables S1-S6). Additionally, age-dependent increase of whole brain radiotracer uptake were observed, FIGURE 3 | Biodistribution of radiotracer activity in rat brains by PET/CT imaging at different ages, post intravenous administration of 18 F-PF-(+)-DTBZ, 11 C-CFT and 18 F-FDG, respectively. There was increase of brain uptake of these 3 radiotracers increased from 2 months to 4, 9, 12 months of age by PET quantification.
Frontiers in Neuroscience | www.frontiersin.org radioactivity uptake of these 3 radiotracers were higher in the rat brains at 4, 9, 12 months of age than that at 2 months of age.

Age-Dependent Changes of 11 C-CFT Uptake in Rat Brains
A decrease in DAT expression was observed in human and small animal brain aging (Hall et al., 2014). In order to determine whether there are age-dependent changes of DAT expression in brain development and maturation, a longitudinal PET/CT study was performed to measure 11 C-CFT in the brains of rats at 2, 4, 9, and 12 months of age. 11 C-CFT uptake (SUV) reached a slightly descending plateau at 4 months of age in the striatum and accumbens (Figure 4). The striatum displayed a significantly lower 11 C-CFT uptake at 2 months of age (SUV: 1.37 ± 0.39) compared to 4 months (SUV: 2.04 ± 0.33, P < 0.0001), 8 months (SUV: 2.09 ± 0.18, P < 0.0001 ) and 12 months (SUV: 2.00 ± 0.17, P < 0.0001). The same trend was found in the accumbens, as increased SUV was observed from 2 months (1.01 ± 0.32) to 4 months (1.38 ± 0.22, P = 0.0021), 9 months (1.36 ± 0.24, P = 0.0033), and 12 months (1.19 ± 0.25, P = 0.3705). SUV uptake in nearly all other brain regions did not demonstrate significant change, aside from the thalamus from 2 to 4 months (P = 0.02) (Supplementary Table S5).

Age-Dependent Changes of Glucose Metabolism in Brains of Rats
In order to study the role of glucose metabolism during brain growth and the relationship among glucose metabolism and DAT and VMAT2 expression, we also examined glucose changes in the brain using 18 F-FDG (Figure 6). A similar increased age-dependent FDG SUV uptake pattern was discovered from 2 months to 4, 9, and 12 months of age, as seen in both 11 C-CFT and 18 F-PF-(+)-DTBZ. However, high accumulations were found in the cortex, striatum, and accumbens, while the lowest uptake was observed in the pituitary gland. SUV in the striatum at 2 months of age increased significantly from 3.71 ± 0.78 to 5.28 ± 0.81 (P < 0.0001) at 4 months of age, and then decreased gradually from 4 to 9 and 12 months of age. Almost all of the brain regions demonstrated a significantly lower FDG SUV uptake at 2 months compared to older ages, except in the pituitary gland, which possessed the lowest uptake and displayed no significant change of uptake at different ages. SUVR uptake decreased significantly from 2 months to older ages in most brain regions, including the striatum, accumbens and cortex etc., this decrease was not observed in the thalamus, pons, and auditory cortex (Supplementary Tables S1, S2).

Relationship Between Age-Dependent Changes of 18 F-FDG, 18 F-PF-(+)-DTBZ, and 11 C-CFT Uptake in Brains
In order to determine the relationship of glucose metabolism, DAT, and VMAT2 during brain development, we compared the age-dependent change of striatal uptake of these three radiotracers. There existed a similar pattern of age-dependent change in regional brain SUV uptake among these three radiotracers from 2 to 12 month of age, but no correlation of SUVR uptake in these brain regions. As shown in Figure 7, 18 F-FDG, 18 F-PF-(+)-DTBZ, and 11 C-CFT SUV in the striatum displayed similar patterns by increasing from 2 to 4 months and then reaching a slightly descending plateau until 12 months of age. Interestingly, SUV of 18 F-PF-(+)-DTBZ, and 11 C-CFT were similar at different individual ages (P > 0.05 at 2, 4, 9, and 12 months). However, SUVR demonstrated different patterns in striatum for these three radiotracers. The striatum 11 C-CFT SUVR was higher than 18 F-PF-(+)-DTBZ SUVR at all ages.

DISCUSSION
Investigation of age-dependent changes of various metabolic pathways or expression or function of receptors or transporters is critical for a better understanding regulation of brain development and maturation. Many studies revealed brain function alterations in brain aging, such as fluctuations in DAT, glucose levels, and copper metabolism (Kawamura et al., 2003;Peng et al., 2018). However, few studies were dedicated to study FIGURE 4 | Alterations of SUV and SUVR of 11 C-CFT in brain regions from 2 to 12 months of age. SUV (left) increased from 2 to 4 months of age in the striatum and accumbens, along with other regions. SUVR (right) did not fluctuate in most brain regions, aside from varying patterns in the striatum and accumbens. All data were presented by mean ± SD.
FIGURE 5 | SUV and SUVR of 18 F-PF-(+)-DTBZ in brain regions of rats at different ages. Increased SUV (left) and decreased SUVR (right) 18 F-PF-(+)-DTBZ uptake was observed from 2 months of age onward. All data were presented by mean ± SD. of morphological and functional changes in brain development and maturation at adolescent population, partly due to technical limitation and other factors such as access to adolescent human subject population. In vivo imaging techniques, such as MRI and PET, are useful tools for monitoring disease progression along with changes in biological pathways and metabolism. The definition of brain maturity is divided into many types, such as social, cognitive, and neurotransmitter system etc., Previous studies using MRI found brain structures, especially the rat brain cortex, are not fully developed at 3 months of age. Therefore, they recommend careful examination of the brain structures before longitudinal examinations (Mengler et al., 2014). However, the function examination of brain development is rarely considered, particularly the examination of expression of dopamine-related receptors and transporters in brain maturation.
Dopamine levels are considered to represent the density of dopamine terminals. DAT and VMAT2 can reflect dopamine levels, and they are biomarkers brain disease and used to study brain aging (Colebrooke et al., 2006;Troiano et al., 2010;Oh et al., 2012). DAT and VMAT2 in the striatum were found to be decreased in the brains of aging rats and in human brains using 11 C-CFT and 18 F-PF-(+)-DTBZ as radiotracers (Kawamura et al., 2003;Hall et al., 2014). More importantly, DAT and VMAT2 were biomarkers for Parkinson's Disease (PD). 11 C-CFT and 18 F-PF-(+)-DTBZ were applied to diagnose PD and evaluate the severity of PD (Nurmi et al., 2000;Hsiao et al., 2014). The increased expression of dopamine terminals can reflect the process of brain maturation (O'Donnell, 2010). Age-dependent changes of the DAT and VMAT2 biomarkers could be useful in the assessment FIGURE 6 | Age-dependent changes of SUV and SUVR uptake of 18 F-FDG in brain regions of rats at different ages by a longitudinal PET study after intravenous administration. Increased SUV (left) and decreased SUVR (right) of 18 F-FDG was observed from 2 months of age to older ages. All data were presented by mean ± SD.
FIGURE 7 | Correlation of 18 F-FDG, 18 F-PF-(+)-DTBZ and 11 C-CFT uptake in the striatum of rats. A similar trend of SUV (left) at different ages for these 3 radiotracers, age-dependent increase uptake was observed from 2 months of age to an older age. And a coincidence of SUV uptake pattern for 18 F-PF-(+)-DTBZ and 11 C-CFT, but no correlation was observed for SUVR uptake (right) of these radiotracers. All data were presented by mean ± SD. of density changes of dopamine terminals associated with maturation.
In this study, we found age-dependent increases in FDG uptake in rat brains from 2 to 4 months of age, and a slightly nonsignificant decline of glucose metabolism was observed from 4 to 12 month. This finding is different from decreased glucose levels in aging brains described in previous reports (Chetelat et al., 2013;Ewers et al., 2014;Nugent et al., 2014). The increased glucose metabolism may be caused by the increase in glucose demand in brain maturation. In addition to the increase in glucose metabolism, expression of DAT and VMAT2 was also elevated from 2 to 4 months of age within the striatum. This increased density of dopamine terminals in the striatum and suggests an increase in dopamine level or concentration. Mechanisms of concurrent increase of FDG uptake and elevated expression of DAT and VMAT2 are not clear.
Aside from the measurement of SUV, we also calculated the SUVR of these three radiotracers. Calculation of SUVR is considered standard practice in clinical 18 F-FDG PET brain imaging. Interestingly, the pattern of SUVR is different from the pattern of SUV among these three radiotracers. SUV of 18 F-FDG increased from 2 to 4 months of age in most brain regions, while SUVR declined slightly, which is caused by global increased glucose uptake in rat brains and faster increase of FDG uptake in the cerebellum. For SUVR of 11 C-CFT and 18 F-PF-(+)-DTBZ, no significant changes were observed in these brain regions, except in the stratum (2 m vs. 12 m, P = 0.04; 4 m vs. 12 m, P = 0.002) for 11 C-CFT. This may be induced by the global brain increase of 11 C-CFT uptake. The age-dependent higher and increased SUVR of 11 C-CFT in the striatum compared with a decline in SUVR of 18 F-PF-(+)-DTBZ showed the potential of 11 C-CFT in the diagnosis of brain disease in the striatum, such as PD (Yagi et al., 2010).
Owing to the high concentrations of dopamine in the striatal regions, several studies of dopamine in aging have been focused on this region. Different from those studies in aged animals, we found a correlation in the SUV uptake curves of 11 C-CFT and 18 F-PF-(+)-DTBZ in the striatum (Figure 7).
Both radiotracers could reflect the expression of dopamine levels. Our results indicated the positive correlation of absolute quantification (SUV) of DAT and VMAT2 in the striatum in maturation, which demonstrates growth of dopamine terminals in the striatum, and the numbers of DAT and VMAT2 may also be similar within the striatum. 18 F-FDG possessed a higher SUV uptake and similar SUV pattern compared with the other two radiotracers within the stratum, indicating the significance of glucose metabolism in the development of DAT and VMAT2 in this region.
The age of small animals is a key impact factor in the study of brain disease, yet many brain studies are based on the assumption of matured brains. When the animals are too young with undeveloped brain-blood barrier (BBB), radiotracers will not need to penetrate the BBB to induce high brain uptake (Patel and Gibson, 2008). Additionally, immature brains will introduce deviation in longitudinal studies. Therefore, mature mice or rats are essential in the evaluation of new radiotracers targeting the brain and longitudinal studies examining brain disease and brain functions. However, about 300 g of body weight or 2 to 3 months of age have often been applied to define the time window of maturity in rats.
In previous studies, volume change, cortical thickness, and cell density in the striatum were found to be stabilized in 3 months of age, but myelination in the striata continued up to 3 months of age and in the cortex up to 6 months (Mengler et al., 2014). Our study confirms that brain synaptic development is not stabilized within the first 4 months, as the densities of DAT and VMAT2 continue grow at 4 months of age, which coincided with previous study. Therefore, 4 months of age may be considered as the window time for brain maturation of dopamine. Longitudinal study for brain, especially PET imaging study, should not start before 4 months of age in rats. And future studies need to better define the end of the developmental phase before a related study can begin.
There were several limitations in this study. Brain size could be influent factor for the measurement of brain uptake. In previous study, brain size did not change significantly after 2 month of age in Wistar rats (Mengler et al., 2014). Brain uptake pattern of these rats at 2 and 12 month matched PMOD template (Supplementary Figure S2), which means brain uptake from PMOD can be used for comparison. That conclusion supported us to start our study from 2 month of age. However, Lack of MRI study for brain size still limit this study. Perfusion of the brain uptake as a consequence of brain development and differential response to anesthetization is another limitation. Different type of anesthetization brought different response to brain uptake. Brain uptake of 18 F-FDG decreased in condition of isoflurane and ketamine/xylazine anesthetization, no change of striatal 18 F-PF-(+)-DTBZ uptake was observed in isoflurane but ketamine increased 18 F-PF-(+)-DTBZ striatal uptake (Toyama et al., 2004;Chen et al., 2016). All of these results indicated anesthetization is one of the confounding factor to brain uptake. Especially at different age of rats, differential response to anesthetization may cause different perfusion pattern, relative proportion of the tracer in the brain may alter. This is also one of the main limitation in this study. All of these limitations desired further study to investigate the infections.

CONCLUSION
Age-dependent increase of striatal 18 F-FDG, 18 F-PF-(+)-DTBZ and 11 C-CFT radiotracer uptake was detected in Wistar rats from 2 to 4 months of age, reaching plateau of uptake in matured rats at 4 month of age. The findings suggest that 4 months of age may represent a desirable time point for conducting neurological study of brain disease to avoid confounding effects of ongoing changes of glucose metabolism and neuroreceptor expression on neurological studies involving use of PET and radiopharmaceuticals. The findings support further investigation of age-dependent changes of DAT and VMAT2 expression during brain development and maturation with PET using 18 F-PF-(+)-DTBZ and 11 C-CFT radiotracer, respectively.

AUTHOR CONTRIBUTIONS
DJ and FX designed and performed the experiments. XL and FX produced the radiotracers. DJ, ZL, FH, CZ, YG, and FX contributed to conception and designed the study. DJ, FX, and FH contributed to analysis and interpretation of data. FX wrote the draft of this manuscript. DJ, NR, FP, and FX contributed to critical review and revision of the manuscript for this article.