Evaluation of Dopamine Receptor Integrity after Sevoflurane Exposure in Neonatal Rat Brain Using Positron Emission Tomography

Aims: The volatile general anesthetic sevofl urane is commonly used across all ages in the clinic. Sevofl urane-induced neurotoxic effects on the developing dopaminergic system are still unclear. The aim of this study was to evaluate the integrity of the D2/D3 receptor in developing rat brain utilizing molecular imaging techniques. Method: Positron Emission Tomography (PET) coupled with Computerized Tomography (CT) approaches were used to evaluate the effects of developmental sevofl urane exposure on D2/D3 receptors in rat pups. The PET radiotracer, [18F]-fallypride, was used to examine whether dopamine receptors were affected by prolonged sevofl urane-exposure during the brain growth spurt. Six postnatal day (PND) 7 rats in the experimental group were exposed to sevofl urane at the clinically-relevant concentration of 2.5% (v/v) for 8 hours, and an equal number of control animals were exposed to room air for 8 hours. MicroPET/ CT scans were sequentially collected on PNDs 14, 21, 28, and 35. Results: No signifi cant differences in the retention of [18F]-fallypride in striatum were detected between sevofl urane-exposed and control animals at any of the scan times. D2/D3 receptors are not affected by prolonged sevofl urane-induced general anesthesia during development. Conclusions: Sevofl urane-induced neurotoxicity in the developing rodent brain was not associated with apparent derangements in dopamine D2/D3 receptors. Research Article Evaluation of Dopamine Receptor Integrity after Sevofl urane Exposure in Neonatal Rat Brain Using Positron Emission Tomography Qi Yin1, Merle G. Paule1, Tucker A. Patterson2, Fang Liu1, Charles M. Fogle1, Scott M. Apana3, Marc Berridge3, William Slikker Jr2, Cheng Wang1 and Xuan Zhang1* 1Division of Neurotoxicology, National Center for Toxicological Research (NCTR), U.S. Food and Drug Administration (FDA), Jefferson, AR, USA 2Offi ce of the Director, National Center for Toxicological Research (NCTR), U.S. Food and Drug Administration (FDA), Jefferson, AR, USA 33D Imaging, LLC, Little Rock, AR, USA Received: 29 November, 2017 Accepted: 16 January, 2018 Published: 18 January, 2018 *Corresponding author: Xuan Zhang, MD, Ph D, Division of Neurotoxicology, National Center for Toxicological Research (NCTR), U.S. Food and Drug Administration (FDA), Jefferson, Arkansas, USA, E-mail:

There are substantial data indicating that prolonged exposure to sevofl urane can induce neurotoxicity in both rodents and nonhuman primates developing brains [13,14]. Signifi cant behavioral research has also shown that exposure of neonatal mammalian animals to sevofl urane for a long period and/or multiple times will impair performance in the Morris water maze (MWM), a novel-object recognition test (NOR), social behaviors, fear conditioning and recognition memory at all ages [15][16][17]. Since there are many different kinds of neurons in the developing brain which may be affected by sevofl urane, the role dopaminergic neurons play in this complicated procedure need to be addressed. The function and expression levels of the dopamine D2/D3 receptor in the developing rodent brain exposed to a neurotoxic regimen of sevofl urane, has not been fully investigated.
Like glutamate and gamma aminobutyric acid, dopamine is an important neurotransmitter in the brain. It acts on dopamine receptors, which have fi ve different subtypes (from D1 to D5), to regulate brain reward and pleasure, and many  The D2-like receptor subfamily includes D2, D3, and D4 subtype members, and they usually couple to Gi/Go proteins, which may decrease the production of cAMP and the activation of protein kinase A (PKA) in the cytoplasm [18,19]. D2 receptors are highly expressed in the caudate, putamen (basal ganglia), and nucleus accumbens. D3 receptors have a more limited pattern of anatomical distribution such as in the nucleus accumbens [20,21]. The caudate, putamen and ventral striatum are three main nuclei of the striatum. The nucleus accumbens was included in the ventral striatum area [22], so the present research focused on evaluating the integrity of D2  [23][24][25]. As a minimally invasive, dynamic and leading approach, PET imaging can acquire quantitative images from living animal organs such as the brain [26,27]. Since PET imaging data can be collected from the same subject at multiple time points, it provides the possibility of continuously examining physiological and pharmacological processes of interesting targets in vivo.
The present study was designed to determine whether exposure to sevofl urane during a sensitive period of brain development in the rat alters specifi c aspects of dopaminergic systems, especially for D2/D3 receptors, which can be effi ciently measured using a PET radiotracer.

Radiotracer preparation
prepared by 3D Imaging LLC (Little Rock, USA) following published procedures [28,29]. GraphPad Prism version 6 (GraphPad Software, La Jolla, CA. USA) was used for data analyses and to produce graphs.

MicroPET/CT image acquisition
All assays were analyzed using unpaired student t-tests. All analyses were two-tailed and considered to be statistically different at p-values less than 0.05.

Results
In order to improve the location accuracy of microPET imaging, a CT scan was scheduled immediately after the 90min microPET scan. A representative picture is shown in fi gure 1. The image of a rat brain is presented in three different kinds of anatomical panels including the horizontal, coronal, and sagittal plane. CT images that were used as anatomical references are presented in the top row. PET images are presented in the middle row. The merged images of PET and CT were automatically given by the IRW software co-registration function. The combination of both structure and functional images were presented in the bottom row. The striatum is indicated by the green cross circle. As mentioned, D2 and D3 receptors are predominantly distributed in the striatum, so the strongest [ 18 F]-fallypride retention area coincides with the striatum. In order to compare the radioactivity intensity of each subject a three-dimensional region of ROIs with the same volume was manually and randomly drawn in the left striatum. Since 18 F is a sensitive bone-seeking isotope and much of the [ 18 F]-fallypride is retained in bone during the scan procedure [31], high radioactive areas outside the brain were primarily observed in bones and teeth. Meanwhile, ROIs in the left frontal cortex were analyzed and compared. No apparent differences were found (data not shown).
On PND 14 (one week following exposure), all animals were weighed and injected intravenously with [ 18 F]-fallypride. The decay-corrected amount of injected [ 18 F]-fallypride was calculated by subtracting the activity of the empty syringe from the activity of the full syringe. The radioactivity concentrations of ROIs were measured by the IRW. SUVs were calculated by average concentration of radioactivity in the ROIs multiplied by body weight and then divided by injected dose [32,33]. No statistically signifi cant differences of [ 18 F]-fallypride SUVs in the striatum were detected between exposure and control groups on PND 14. The uptake of radioactivity in control ( Figure  2A) and sevofl urane-exposed rats ( Figure 2B) were similar.
Serial microPET scans were performed on PND 21, 28, and 35 in order to monitor the dynamic development of the brain after sevofl urane exposure following the same procedure as described on PND 14. The SUVs of each time point were analyzed and no changes in D2/D3 receptor expression levels were found in the striatum of the sevofl urane-exposed and control rats ( Figure 2C). The results indicate that an 8-hour sevofl urane exposure does not impair D2/D3 dopaminergic receptor function in the striatum.

Discussion
MicroPET/CT is a minimally invasive molecular imaging multimodality that was broadly used in both preclinical and clinical research [34]. To determine the expression levels,  [39].
Also, the high affi nity is not affected by the density of the D2/ D3 receptor and the concentration of endogenous dopamine in different brain areas [39]. Second, the high-affi nity gives [ [44][45][46][47]. Also, [ 18 F]-fallypride can easily cross the blood brain barrier and reach maximal uptake activity within  30 min after a single bolus IV injection. An equilibrium state may last at least two hours before gradual clearance. Therefore, the relatively slow washout and long half-life (110 minutes) of [ 18 F]-fallypride are two key factors which support our study designs [48]. Thus, [ 18 F]-fallypride was selected in this study to evaluate the potential effect of prolonged sevofl urane exposure on D2 and D3 dopamine receptor expression levels. In present the study both control and sevofl urane-exposed animals were initially scanned 30 min after intravenous injection of [ 18 F]fallypride and the data recorded for 90 min in order to collect as many signals as possible. Our microPET data demonstrated that the accumulation of the [ 18 F]-fallypride signal was heavily expressed in the striatum as evidenced by horizontal, coronal and sagittal examination (Figure 1), in both control and sevofl urane-exposed animals.
It should be mentioned that currently only a few studies have investigated the roles of monoaminergic neurotransmission in sevofl urane anesthesia [10]. In general, dopamine and dopamine receptors help control brain reward and pleasure, and their dysregulation and/or defi ciency can result in neuronal degeneration and cognitive-disorders including Parkinson's disease [49]. Also, numerous behavioral studies have disclosed that long-term sevofl urane exposure in developing brains may impair cognition, memory and learning ability later in life [15][16][17]50,51]. It has been reported that altered dopamine neurotransmission, especially dopamine receptor expression levels, may mediate the capacity of inhaled anesthetics to provide immobility and decrease or increase MAC (the minimum alveolar concentration of an inhaled anesthetic required to suppress movement in response to a noxious stimulus in 50% of test subjects) [52,53]. Therefore, it is originally proposed that any kind of D2/D3 receptor dysregulation or dysfunction, after prolonged anesthetic exposure, might subsequently result in increased neuronal damage in various brain areas including striatum. Since imaging tools provide a means to analyze the mechanism underlying behavior dynamically, the subjects (experimental animals) could receive "follow up" scans repeatedly. Thus, minimally invasive microPET/ CT imaging was applied in the present study to detect subtle variations in vivo and to measure the molecular level variation of the dopaminergic system following 1, 2, 3, and 4 weeks of sevofl urane exposure. Particularly, striatal dopamine receptor levels were monitored to determine whether the dynamic alteration of D2 and D3 receptors could be associated with the sevofl urane-induced neuro-apoptosis, which was previously demonstrated [14,54]. Our imaging data demonstrate that no statistically signifi cant changes in D2/D3 receptors were detected in striatum on PND 14 or older ages after an 8-hour sevofl urane exposure in the developing rodent brain. Our previous publication demonstrated that the frontal cortex was one of the most vulnerable areas of anesthetic-drug induced neural damage [55], so we also evaluated the D2/D3 receptor expression levels in this area. No obvious changes were found in frontal cortex at multiple time points (data not shown).
These data, plus data from studies by others [56], indicate that dopamine receptors do not mediate the immobility or altered MAC produced by inhaled anesthetics, including sevofl urane, and that D2/D3 receptors in striatum have no contribution to sevofl urane-induced behavioral disturbance and sevofl uraneinduced neurotoxicity [14,54].
Commonly used anesthetics may induce neuro-apoptosis in immature animal brains. Such mechanisms have been widely categorized and associated defi cits of cognitive behavior are usually caused by brain and/or neuronal damage in areas that affect learning and memory. According to our PET and immunochemical data, D2/D3 neurons in the striatum may be less sensitive or completely not related to sevofl urane-induced neuronal damage and behavioral defi cits. Accumulating evidence indicates that anesthetic-induced neuronal damage could be closely associated with receptor subtype activation and the maturity of animals at the time of exposure [57,58].
Therefore, neuro-apoptosis induced by prolonged sevofl urane could be mediated via all mechanisms whereby sevofl urane acts; including sevofl urane-induced depression of glutamate release [59], GABA receptor excitation [60], increased intracellular calcium infl ux [61], anesthetics-induced decrease in neurotrophic factors [62] and anesthetic-induced activation of inositol 1,4,5,-trisphosphate (IP3) receptors [63]. Thus, our immediate priority for future projects is to search and identify specifi c radiotracers to target potential alteration of glutamate neurotransmission and GABA excitation after gaseous anesthetic exposure. Also, the effects of anesthetics on mitochondria could be another factor as why immature neurons are subject to anesthetic-induced apoptosis.
The limitation of molecular imaging studies is that a variety of different brain cell types cannot be characterized using microPET/CT scanning. There are many different kinds of cells in the central nervous system, including neurons, astrocytes, oligodendrocytes and microglia. An obvious question for further studies is to categorize, quantify and determine which types of brain cells can be affected under the situation of sevofl urane exposure. Neural cell and apoptosis-related biomarkers may be useful to solve the limitation [64].
In summary, microPET/CT provides a minimally invasive way to interrogate brain status affected by general anesthesia during development. The current results demonstrate that dopamine neurotransmission, especially in the D2/D3 receptor subtypes, is not affected by prolonged sevofl urane-induced general anesthesia. D2/D3 receptors in striatum do not contribute to sevofl urane-induced neurotoxicity.

Disclaimer
This document has been reviewed in accordance with United States Food and Drug Administration (FDA) policy and approved for publication. Approval dose not signify that the contents necessarily refl ect the position or opinions of the FDA.
The fi ndings and conclusions in this report are those of the authors and do not necessarily represent the views of the FDA.