Alteration of Copper Fluxes in Brain Aging: A Longitudinal Study in Rodent Using 64CuCl2-PET/CT

Brain aging is associated with changes of various metabolic pathways. Copper is required for brain development and function, but little is known about changes in copper metabolism during brain aging. The objective of this study was to investigate alteration of copper fluxes in the aging mouse brain with positron emission tomography/computed tomography using 64CuCl2 as a radiotracer (64CuCl2-PET/CT). A longitudinal study was conducted in C57BL/6 mice (n = 5) to measure age-dependent brain and whole-body changes of 64Cu radioactivity using PET/CT after oral administration of 64CuCl2 as a radiotracer. Cerebral 64Cu uptake at 13 months of age (0.17 ± 0.05 %ID/g) was higher than the cerebral 64Cu uptake at 5 months of age (0.11 ± 0.06 %ID/g, p < 0.001), followed by decrease to (0.14 ± 0.04 %ID/g, p = 0.02) at 26 months of age. In contrast, cerebral 18F-FDG uptake was highest at 5 months of age (7.8 ± 1.2 %ID/g) and decreased to similar values at 12 (5.2 ± 1.1 %ID/g, p < 0.001) and 22 (5.6 ± 1.1 %ID/g, p < 0.001) months of age. The findings demonstrated alteration of copper fluxes associated with brain aging and the time course of brain changes in copper fluxes differed from changes in brain glucose metabolism across time, suggesting independent underlying physiological processes. Hence, age-dependent changes of cerebral copper fluxes might represent a novel metabolic biomarker for assessment of human brain aging process with PET/CT using 64CuCl2 as a radiotracer.

Copper is a nutritional metal required for brain development and function [6][7][8][9][10]. On the other hand, accumulation of excess copper in brain tissues could be harmful. Copper deficiency due to malfunction of ATP7A copper transporter encoded by mutated ATP7A gene [11][12][13] causes neurological disorder in Menkes disease [14], while accumulation of excess copper due to malfunction of ATP7B copper transporter encoded by mutated ATP7B gene [15][16][17] causes neurological disorder in Wilson's disease (WD), or hepatolenticular neurodegeneration [18][19][20][21]. Significant progress has been made in understanding of cellular copper transport regulation by a delicate network of copper transporters and chaperons. However, systemic regulation of copper fluxes in brain aging and age-related neurodegenerative disorders remains poorly understood due to lack of a tool for longitudinal, real-time tracking of changes of brain copper fluxes with age.
Positron emission tomography (PET) is a useful tool for real-time tracking brain copper fluxes in vivo based on its high sensitivity and quantitative analysis capability [22]. Copper metabolism imbalance in Atp7b -/knockout mouse model of Wilson's disease was assessed with PET/CT using radioactive copper-64 chloride ( 64 CuCl2) as a tracer [23,24]. Recently, age-dependent changes of 64 Cu radioactivity were detected in the brains of Atp7b -/knockout mouse model of WD with PET/CT [25]. Using this new 64 CuCl2-PET/CT technique as a tool, this study aimed to conduct a longitudinal PET study to assess agedependent changes of copper fluxes in the brains of C57BL/6 mice. Moreover, a concurrent longitudinal PET study using 18 F-FDG) as a radiotracer [26] was performed to compare changes of copper fluxes with changes of glucose metabolism in mouse brain with aging. The findings of this pilot PET study for the first time demonstrated age-dependent changes of 64 Cu radioactivity in the brains of mice orally administered with 64 CuCl2 as a radiotracer, supporting potential use of cerebral 64 Cu uptake as a biomarker for noninvasive assessment of brain aging and age-related neurodegenerative disorders using 64 CuCl2-PET/CT as a tool.

Small animals and Radiopharmaceuticals
C57BL/6 mice were purchased from Taconic Biosciences (Hudson, NY) and housed in the animal housing facility, UT Southwestern Medical Center at Dallas, with free access to copper-adequate food (AIN-93M Purified Rodent Diet, Dyets Inc., Bethlehem, PA) and drinking water.
The tracer 64 CuCl2 was purchased from Washington University (St Louis, MO), which was produced via 64 Ni(p,n) 64 Cu using a biomedical cyclotron and supplied in the form of 64 CuCl2 in 0.1M HCl solution. The specific activity of 64 Cu was 6.9 ± 2.5 Ci/µmol. 18 F-FDG was purchased from PetNet Solutions (Dallas TX). All small animal experiments were conducted according to a protocol approved by the Institutional Animal Care and Use Committee, UT Southwestern Medical Center, Dallas, TX.

Study design
A longitudinal study was performed to assess agedependent changes of copper fluxes and glucose metabolism during normal aging process in a group of C57BL/6 mice (female, n = 5) by sequential 64 CuCl2 and 18 F-FDG PET/CT scans at 3 different time points (young, middle-aged, old) related to human age equivalents [ Fig.  1].  64 CuCl2-PET/CT imaging of C57BL/6 mice was performed using a method described previously [23][24][25]. Briefly, the mice were anesthetized by 3% isoflurane in 100% oxygen (3 L/min) at room temperature, using an isoflurane vaporizer (Summit Anesthesia Solutions, Salt Lake City, UT). The mice were positioned in a spreadsupine position on the imaging bed and subjected to inhalation of 2% isoflurane in 100% oxygen (3 L/min) during the PET/CT procedure. Static whole-body PET/CT imaging was then obtained for 15 minutes at 2 hour (h) post oral administration (PO) of 64 CuCl2 (2 μCi (74 kBq)/g body weight) diluted in a volume of 25 µL normal saline (0.9% sodium chloride) using a blunted oral feeding tube, followed by static whole-body scan for 15 minutes at 24 h PO. PET/CT images were reconstructed using the ordered subsets expectation maximization 3D algorithm (OSEM3D), and data was analyzed using the Inveon Research Workplace (IRW) software (Siemens) which allows fusion of CT and PET image volumes. For 18 F-FDG PET/CT imaging, C57BL/6 mice were subject to fasting for 10 to 12 hours prior to oral feeding of 18 F-FDG (2 μCi (74 kBq)/g body weight) with a blunted feeding tube using a method modified from those described previously [25]. Acquisition and reconstruction of 18 F-FDG PET/CT images were performed in analogy to the protocol used for 64 CuCl2-PET/CT imaging described above.  64 Cu uptake in the heart and muscle at old age higher than 64 Cu uptake at middle age. In contrast, 64 Cu radioactivity in the kidneys at old age was lower than renal 64 Cu radioactivity of C57BL/6 mice at young and middle age. Mean ± SD %ID/g: percentage of injected dose per gram tissue; PO: post oral administration of the tracer. 64 Cu and 18 F-FDG radioactivity in the brains of C57BL/6 mice were analyzed using the Inveon Research Workplace (IRW) software (Siemens). In order to measure 64 Cu and 18 F-FDG radioactivity in the different regions of the brains of C57BL/6 mice, 9 regions of interest (ROIs) were defined manually on the PET/CT images with reference to an MR imaging-based atlas of mouse brain anatomy [27] as described recently [25]. Regions included were: ROI 1 for olfactory bulb (OB), ROI 2 for frontal cortex (FC), ROI 3 for posterior cortex (PC), ROI 4 for hippocampus (HC), ROI 5 for basal ganglia (BG), ROI 6 for thalamus (T), ROI 7 for middle brain (MB), ROI 8 for brain stem (BS), ROI 9 for cerebellum (CB). The quantity of 64 Cu and 18 F-FDG radiotracer activity was obtained and recorded as a percentage of injected dose per gram tissue (%ID/g).

Quantification of 64 Cu and 18 F-FDG radioactivity in the brain of C57BL/6 mice
Additionally, the mean ± standard deviation (SD) of %ID/g obtained from various ROIs was calculated and recorded for further statistical analysis.

Statistical analysis
Initially, data was tested for normality using the Shapiro-Wilks test. Once normality was determined, parametric tests were performed in order to determine whether 64 Cu uptake (mean ± SD %ID/g) differs among the young, middle age, and old C57BL/6 mice. Specifically, we applied a (2 x 3 x 9) repeated-measures ANOVA, with the three within-subject's factors representing the scan time (2 or 24h PO), the age groups (young, middle age, and old) and the regions (ROI 1 -ROI 9). Following a significant overall test, pair-wise two-sample t-tests were conducted to determine significant differences among these groups. To correct for multiple comparisons, the adjusted Least Significant Difference (LSD) test (Sidak test) was performed for the post-hoc tests. Moreover, in order to determine whether the brain time course of 64 Cu uptake at 24h PO differs from the brain time course of glucose uptake, we applied a (2 x 3 x 9) repeated-measures ANOVA, where the three within-subjects' factors were the tracer ( 64 Cu, 18 F-FDG), the three timepoints and the nine regions. A P value of less than 0.05 was considered as statistically significant.

Biodistribution of 64 Cu in C57BL/6 mice orally administered with 64 CuCl2 at different ages
After oral administration of 64 CuCl2 as a radiotracer, highest uptake of 64 Cu was detected in the liver (1.73 ± 0.21 %ID/g) at 24h PO, followed by the kidneys (0.95 ± 0.13 %ID/g), with muscle showing lowest copper uptake (0.05 ± 0.05 %ID/g) in C57BL/6 mice at 5 months of age (Table 1, Fig. 2). Overall, 64 Cu uptake in organs was higher at 24h PO as compared to 2h PO (p = 0.01). There was decrease of 64 Cu uptake in the heart at middle age (0.29 ± 0.01%ID/g) compared to 64 Cu uptake in the heart at young age (0.43 ± 0.07 %ID/g), followed by increase to 0.56 ± 0.12 %ID/g at old age. There was also interval increase of 64 Cu uptake in the muscle from middle age (0.02 ± 0.02 %ID/g) to old age (0.12 ± 0.12 %ID/g) of C57BL/6 mice. In contrast, 64 Cu radioactivity in the kidneys at old age (0.48 ± 0.15 %ID/g) was lower than renal 64 Cu radioactivity of C57BL/6 mice at young (0.95 ± 0.13 %ID/g) and middle age (0.81 ± 0.07 %ID/g).

Age-dependent changes of regional 64 Cu radioactivity in the brains of C57BL/6 mice
We determined whole brain 64 Cu uptake for images acquired at both 2h (p = 0.04) and 24h (p = 0.03) PO by longitudinal PET of C57BL/6 mice at 5, 13, 26 months of age (Table 1, Fig. 3A). Because images acquired at 24h PO displayed a greater dynamic range, all further analyses were performed using images acquired at 24h PO. Moreover, quantitative assessment of regional brain 64 Cu tracer uptake showed significant differences between brain regions ( Table 2, and Fig. 3B). Highest 64 Cu uptake was determined in the olfactory bulb, followed by the brainstem and the basal ganglia, with cortical regions (frontal and parietal cortex) showing the lowest copper uptake ( Table 2 and Fig. 3B). Taken together, our data indicates a dynamic time course of copper flux in the brain, increasing by more than 40% of 64 Cu uptake from young to middle age, after which there is a decrease of 64 Cu uptake to about 15% above values determined at young age.

Biodistribution of 18 F-FDG in C57BL/6 mice orally administered with 18 F-FDG at different ages
Age-dependent changes of whole body biodistribution of 18 F-FDG in C57BL/6 mice were determined by a longitudinal PET/CT imaging after oral administration of 18 F-FDG radiotracer [ Table 3, and Figure 4]. Expected variation of muscular 18 F-FDG uptake was visualized in the C57BL/6 mice orally administered with 18 F-FDG radiotracer. Physiologic whole brain 18 F-FDG uptake at young age (7.92 ± 1.19 %ID/g) was significantly higher (p < 0.001) than uptake at both middle age (5.33 ± 1.63 %ID/g) and old age (5.63 ± 1.16 %ID/g), the latter two showing similar values (Table 3).

Relationship between age-dependent changes of 64 Cu and 18 F-FDG radioactivity in the brains of C57BL/6 mice
Age-dependent changes of regional brain 18 F-FDG uptake was analyzed by PET quantitative analysis (Table 4) and compared with regional brain 64 Cu uptake as described above (Table 2). In contrast to the observed low-level regional brain 64 Cu uptake, high regional 18 F-FDG uptake was determined in cortical, midbrain, and cerebellar regions (Table 4). Moreover, in addition to the large difference in overall tracer uptake, we determined differences in the temporal pattern between FDG and 64 Cu brain uptake. Whereas 64 Cu uptake increased from a low baseline value at young age to a maximum value at middle age and subsequently decrease to lower values at old age, 18 F-FDG uptake in brain was highest at young age and decreased to lower values that were similar at both middle age and old age (Table 4, and Figure 5). This comparative analysis clearly demonstrates differences with respect to physiological uptake mechanisms of 64 CuCl2 and 18 F-FDG during the brain aging process.

DISCUSSION
Brain aging process is associated with changes of various metabolic pathways. Development of radiotracers targeting brain metabolic changes is not only significant for a better understanding of physiology of healthy brain aging, but also significant for differentiation of healthy brain aging from pathological brain aging in AD and other neurodegenerative disorders in a content of brain aging. Copper is required for brain development and physiology although the role of copper in brain aging remains to be elucidated. Radionuclide 64 Cu is a positron emitting copper isotope and 64 CuCl2 was used as a radiotracer for noninvasive assessment of disturbance of cerebral copper fluxes in traumatic brain injury [28] and age-dependent changes of copper fluxes in Atp7b -/knockout mouse model of Wilson's disease [25]. Applying 64 CuCl2-PET/CT imaging in a mouse model of brain aging, we present data that show age-related differences in brain copper fluxes across the life span. Cerebral 64 Cu radioactivity detected in middle-aged C57BL/6 mice was higher than the cerebral 64 Cu radioactivity in young adult and middle-aged C57BL mice [ Figure 2, Table 1 and 2]. More importantly, our findings demonstrate differences in both the time course as well as in the regional pattern between brain 64 Cu and 18 F-FDG uptake, indicating different time course of brain copper and glucose metabolism in brain aging. As copper functions as a nutritional metal required for development and function of brain, independent non-invasive assessment of copper fluxes in vivo might provide an important tool for the study of copper metabolic changes during both normal and pathological brain aging. Aging is a major risk factor for pathogenesis of AD [29]. Age-dependent changes of cerebral 64 Cu radioactivity might be a useful biomarker for assessment of copper metabolic changes associated with AD with 64 CuCl2 -PET/CT. Recently, Torres et al. used radioactive glyoxalbis(N4-methyl-3thiosemicarbazonato)-64Cu(II) complex, 64Cu(II)-GTSM, as a radiopharmaceutical to examine intracranial copper transport in TASTPM transgenic AD mice [30]. Altered 64 Cu trafficking was detected in the brains of TASTPM transgenic AD mice, with increased 64 Cu concentration and faster brain 64 Cu clearance compared to 64 Cu trafficking in the brains of age-matched wild type control mice. Because 64 Cu(II)-GTSM penetrates blood brain barrier and delivers 64 Cu into brain tissue, 64 Cu(II)-GTSM might be limited for evaluation of systemic regulation of gastrointestinal absorption and brain uptake of copper. It will be desirable to compare cerebral copper trafficking in TASTPM transgenic AD mice administered with 64 Cu(II)-GTSM intravenously or 64 CuCl2 orally for changes of gastrointestinal absorption of copper and subsequent copper influx to brain in AD. Additional studies are necessary to address the following issues: (1) ROIs used for quantification of regional brain 64 Cu radioactivity were subjected to errors in view of small size of mouse brain and limited spatial resolution of PET/CT; (2) Correlation of age-dependent changes of cerebral 64 Cu radioactivity with changes of cognitive and behavioral activity of old C57BL/6 mice remains to be determined; (3) there may be difference of age-dependent changes of cerebral copper metabolism among different species; (4) Copper absorption may be affected by age and sex in humans [31]. However, our exploratory study only considered female C57BL/6 mice. Thus, further studies are required in male animals in order to determine whether age-dependent changes in copper flux differ between female and male C57BL/6 mice. Finally, the findings of this preclinical study of agedependent changes of brain copper fluxes using C57BL/6 mouse model of brain aging need to be validated in human subjects.
The molecular mechanisms of age-dependent changes of brain copper fluxes observed in this exploratory study remain to be elucidated. Different amount of 64 Cu uptake was detected in various regions of C57BL/6 mouse brain [ Table 2], likely related to copper fluxes controlled by functional activity of copper transporters and chaperons [9,32,33]. It remains to be determined whether high 64 Cu radioactivity in the thalamus/hypothalamus region of the C57BL/6 mice is correlated with high level of copper ions, because 64 Cu radioactivity measured by PET represents dynamic flow of copper and may or may not represent regional content of brain tissue copper ions. The increased 64 Cu uptake in the brains of middle-aged C57BL/6 mice compared with that in the brains of young adult C57BL/6 mice might reflect increased copper influx mediated by human copper transporter 1 (hCtr1) in response to physiological demand for higher copper ion concentration as a prerequisite for optimal functioning of copper-dependent enzymes. However, increased 64 Cu radioactivity in middle-aged C57BL/6 mice could be the result of either increased retention or reduced efflux of copper mediated (by Atp7a or Atp7b) copper transporters or disturbances in intracellular copper transport mediated by copper chaperons [9,32,33]. Brain 64 Cu uptake of old C57BL/6 mice was lower than that of middle-age C57BL/6 mice, possibly due to reduced demand for copper ions by copper-requiring enzymes in brain tissues or reduced copper influx mediated by hCtr1 or elevated copper efflux mediated by Atp7a or Atp7b. We believe that the creation of advanced imaging tools for longitudinal measurement of copper fluxes is highly relevant for the study of molecular mechanisms of age-dependent changes of brain copper fluxes regulated by copper transporters and chaperons, as well as the copper's role in the pathogenesis of neurodegenerative disorders in the context of brain aging.
In summary, age-dependent changes of cerebral 64 Cu radioactivity were detected in C57BL/6 mouse model of brain aging, showing increased 64 Cu radioactivity in the brains of middle-aged C57BL/6 mice compared with cerebral 64 Cu uptake in the young and old C57BL/6 mice. The findings support further investigation of age-dependent changes of copper fluxes in human brain aging and potential use of cerebral 64 Cu uptake as a biomarker for differentiating healthy brain aging from pathological brain aging in AD and other neurodegenerative disorders using 64 CuCl2-PET/CT as a tool.