JOURNAL TRANSCRIPT
Molecular Imaging of Cardiac Sympathetic Innervation by 11C-mHED and PET: From Man to Mouse? Marilyn P. Law1,2, Klaus Scha¨fers2, Klaus Kopka1, Stefan Wagner1, Otmar Schober1, and Michael Scha¨fers2 1Department
of Nuclear Medicine, University Hospital, University of Mu¨nster, Mu¨nster, Germany; and 2European Institute for Molecular Imaging (EIMI), University of Mu¨nster, Mu¨nster, Germany
Dysfunction of the sympathetic nervous system underlies many cardiac diseases and can be assessed by molecular imaging using PET in humans. Small-animal PET should enable noninvasive quantitation of the sympathetic nervous system in mouse models of human disease. For mice, however, the radioactivity needed to give acceptable image quality may be associated with a mass of unlabeled compound sufficient to block the binding of radioligand to its target. The present study assesses the feasibility of using [N-methyl-11C]metahydroxyephedrine (11C-mHED) to measure norepinephrine reuptake in humans, to determine cardiac innervation in mice. Methods: Anesthetized mice were placed in a small-animal PET scanner. 11C-mHED (containing 18% precursor metaraminol) was injected via a tail vein into each animal simultaneously. Fifteen minutes later, animals were injected with saline or metaraminol which competes with mHED for norepinephrine reuptake. 18F-FDG was injected at 60 min to identify heart regions. After reconstruction of the list-mode data, radioactivity in myocardial regions was computed using in-house software, and time–activity curves were plotted. Results: Hearts were clearly visualized after injection of 11C-mHED. Injection of metaraminol at doses less than 50 nmolkg21 had no effect, whereas doses greater than 100 nmolkg21 caused a dose-dependent loss of specifically bound radioactivity. Conclusion: 11C-mHED was successfully used to visualize and assess myocardial innervation in mice. Uptake of 11C-mHED is displaceable by the false transmitter metaraminol. The total molar dose of metaraminol and 11C-mHED must be considered in the analysis of PET data. Key Words: 11C-meta-hydroxyephedrine; norepinephrine reuptake; sympathetic innervation; cardiac imaging; smallanimal PET J Nucl Med 2010; 51:1269–1276 DOI: 10.2967/jnumed.110.074997
T
he activity of the sympathetic nervous system (SNS) is increased in heart failure because of an increased sympathetic drive and decreased activity and density of the neuronal norReceived Jan. 19, 2010; revision accepted Mar. 24, 2010. For correspondence or reprints contact: Marilyn P. Law, European Institute for Molecular Imaging, University of Mu¨nster, Mendelstrasse 11, D-48149 Mu¨nster, Germany. E-mail:
[email protected] COPYRIGHT ª 2010 by the Society of Nuclear Medicine, Inc.
epinephrine reuptake transporter (uptake 1). As a consequence, b-adrenoceptors are chronically activated, leading to a downregulation in cardiac b-adrenoceptor density (1). Several radioligands are available to investigate the cardiac SNS in patients using PET (2). The most widely used clinically are the norepinephrine mimetic 11C-meta-hydroxyephedrine (11C-mHED) for uptake1 and the nonselective b-adrenoceptor radioligand 11C-CGP 12177 for b-adrenoceptor density. Our group has observed global decreases in both uptake1 and b-adrenoceptor density in nonischemic arrhythmogenic cardiomyopathies (3–5), whereas enhanced reuptake occurred without a change in b-adrenoceptor density in Brugada syndrome (6). The development of small-animal PET scanners with both a high resolution and a high sensitivity offers the possibility of studying the progression of disease processes in rat and mouse models of human heart failure. In contrast to the wide use of 18F-FDG to assess myocardial metabolism (7), however, there are few published PET studies of myocardial SNS in rodents, although small-animal PET has been used in rats to assess myocardial sympathetic neuronal activity (8) and to evaluate radioligands for b1-adrenoceptors (9). The use of PET to image molecular targets such as transporters or receptors in small animals imposes challenges not apparent in studies using metabolic tracers such as 18F-FDG, which can be given at the high concentrations needed to achieve good images. Scanner design aims to optimize both resolution and sensitivity, but in dedicated animal scanners resolution is often pursued at the expense of sensitivity so that high doses of radioactivity or long acquisition times are required. With current radiopharmaceutical production methods, high doses of radioactivity are associated with a significant amount of unlabeled compound, which may compromise the specific binding of the radioligand (10). The quadHIDAC small-animal PET scanner (Oxford Positron Systems) (11) has a good spatial resolution and high sensitivity (12,13). Good-quality images of the mouse heart are achievable using less than 5 MBq of 18F-FDG— which, at a specific activity of 30 GBqmmol21, is equivalent to 0.17 nmol, giving 5.6 nmolkg21 for a 30-g mouse.
11C-MHED AND
PET: FROM MAN
TO
MOUSE? • Law et al.
1269
At this dose, the specific binding of a high-affinity radioligand, such as (S)-11C-CGP 12177 (in vivo KD ;1 nmolkg21 (14)), is not detectable, whereas specific uptake of radiotracers with moderate affinity, for example, 11C-mHED (in vivo KD ;100 nmolkg21 (15)), is evident. The aim of the present work was to test the feasibility of assessing myocardial uptake1 in mice using 11C-mHED with the high-resolution quadHIDAC small-animal PET scanner and to develop a quantitative measure of uptake1. With the current radiosynthesis of 11C-mHED, a small amount of the precursor metaraminol is also present in the dispensed 11C-mHED. Metaraminol is a norepinephrine mimetic with an in vivo Ki (;120 nmolkg21) in rat myocardium that is similar to that for mHED (15). Therefore, the effect of metaraminol on myocardial 11C-mHED radioactivity was investigated. MATERIALS AND METHODS Radiochemistry 11C-mHED was synthesized by direct N-methylation of metaraminol with no-carrier-added 11C-iodomethane (16). Specific radioactivities at the end of synthesis were 10–30 GBqmmol21. Radiochemical purities were 95% 6 5%. For 12 preparations, the mean concentrations of mHED and metaraminol were 9 and 2 mM, respectively. Pharmaceuticals Metaraminol bitartrate was purchased from Sigma Aldrich Chemie GmbH. It was dissolved in saline at concentrations of 50 nmolmL21 to 10 mmolmL21 for injection. Animals Studies were approved by the federal animal rights committee and were performed in accordance with institutional guidelines for health and care of experimental animals. Male C57Bl6 mice (25–35 g) were anesthetized by inhalation (isoflurane; 2%; oxygen, 0.5 Lmin21) for insertion of catheters into a lateral tail vein and, in some animals, the ventral tail artery. Animals were allowed to recover for 122 h under light restraint before being reanesthetized for PET. Ex vivo biodistribution studies were performed in conscious animals. PET PET was performed using a submillimeter-resolution (0.7 mm in full width at half maximum) dedicated small-animal scanner (32-module quadHIDAC), which uses wire-chamber detectors and offers uniform spatial resolution over a large cylindric field (diameter, 165 mm; axial length, 280 mm) (11–13). Two or 4 anesthetized mice (isoflurane, 2%; oxygen, 0.5 Lmin21 per mouse) with tail vein catheters were positioned in the scanner lying on their abdomens on a heating pad to maintain body temperature during the scan. Injections were performed via the tail vein catheters using injection loops made from fine-bore polythene tubing and flushed with saline by an infusion pump. To assess myocardial uptake, 11C-mHED (327 MBq in 100 mL per mouse; mHED, 2290 nmolkg21; metaraminol, 1–24 nmolkg21) was injected simultaneously into each mouse at 30 s after the start of data acquisition (total injection volume, including saline flush, 200 mL). To assess displacement of 11C-mHED, one mouse was given 50 mL of saline, and the other mouse (or the other
1270
THE JOURNAL
OF
3 mice in the 4-mouse groups) was given 50 mL of metaraminol (0.04240 mmolkg21) at 15 min after injection of 11C-mHED (total injection volume, 150 mL). List-mode data were acquired for 60 min. To confirm the location of the heart, 18F-FDG (4–8 MBq in 50 mL per mouse; total injection volume, 150 mL) was injected via the tail vein at 5210 min after completion of the 11C-mHED scan, and data were acquired for 30 min. A similar procedure was used to assess competition between 11C-mHED and metaraminol. Unlabeled metaraminol was added to dispensed 11C-mHED at concentrations to give doses of 0.121 mmolkg21. The prepared 11C-mHED (100 mL) was injected at 30 s after scan start, and list-mode data were acquired for 60 min. 18F-FDG was then injected to confirm the location of the heart. Data Analysis List-mode data were reconstructed into images with a voxel size of 0.4 · 0.4 · 0.4 mm3 in time frames of 10 s, 20 s, 1 min, 10 min, or 20 min for 11C-mHED and 15 min for 18F-FDG, using an iterative reconstruction algorithm (17). PET images were analyzed using in-house software programs in MATLAB (The MathWorks Co.) and C programming languages (13). To assess the total radioactivity in each animal, a cube encompassing the body, excluding the tail and paws, was drawn on the reconstructed 18F-FDG image (time frame, 15–30 min). The parameters defining this cube were saved and used to compute the whole-body radioactivities (counts per second [cps]mL21) for each time frame of the 11C-mHED scan. The parameters required to create images of each heart and compute time–activity curves were also defined using the 18FFDG scan (time frame, 15–30 min). The reconstructed volume was divided into 2 or 4 subimages, each showing 1 mouse. From the subimages, coronal images (64 · 64 · 64 pixels) encompassing the heart were made, and regions of interest were drawn manually for myocardium (left ventricular wall and septum) and blood (left ventricular chamber). The parameters were saved and used to create heart images (time frames, 5215 and 40260 min) and decay-corrected time–activity curves (cpsmL21 vs. mid-frame time, for each 10-s, 20-s, or 1-min time frame after injection) for the 11C-mHED scan. Ex Vivo Studies 11C-mHED (mHED, 2 or 47 nmolkg21; metaraminol, 1 or 3 nmolkg21) mixed with increasing concentrations of added metaraminol (0.1210 mmolg21) was injected as a bolus (100 mL) via the tail vein. Aliquots of each injectate were diluted in ethanol–saline and measured to determine the radioactivity injected into each mouse. Mice were sacrificed by intravenous injection of sodium pentobarbitone (Narcoren; Merial GmbH) at 200 mg(kg of body weight)21 at 30 min after injection. Blood was taken by cardiac puncture and tissues rapidly removed, blotted dry using filter paper, and weighed. Radioactivity was measured using an automated g-counter (Wallac Wizard 3; Perkin Elmer Life Sciences). To correct for differences in animal body weight and injected dose, results were expressed as an uptake index (15), defined as: Uptake index 5
Tissue radioactivity ðcpmÞ=tissue wet weight ðgÞ : Radioactivity injected ðcpmÞ=body weight ðgÞ
To assess clearance of 11C-mHED radioactivity from the blood, sequential blood samples were collected by dripping blood from a tail artery catheter into a multiwell plate (10-s collection times).
NUCLEAR MEDICINE • Vol. 51 • No. 8 • August 2010
Samples were weighed and radioactivity measured using the automated g-counter. Radioactivity was expressed as cpmg21. RESULTS PET Scans 11C-mHED
in combination with high-resolution smallanimal PET resulted in images of the SNS in mice that were comparable to those achieved in humans with respect to resolution and contrast. The heart was no longer visible after the injection of metaraminol, but its position was confirmed by 18F-FDG (Fig. 1). Myocardial Uptake of
11C-mHED
Myocardial images for the mice in Figure 1 are shown in Figures 2A22F. The late 11C-mHED images (Figs. 2B and 2E) indicated that the apex and part of the septum included spillover from the liver. Therefore, to construct time–activity curves for the myocardium, regions of interest were traced manually round the left ventricular wall and septum (omitting the apical region) on the 18F-FDG image and confirmed on the 11C-mHED images. Results for the left ventricular wall are shown in Figure 2G. The data for mouse 1 were fitted by a biexponential function of the form y 5 ae2bx 1 ce2dx. The data for mouse 2 were fitted by a biexponential function before injection of metaraminol and by a single exponential function of the form y 5 ae2bx 1 y0 after metaraminol administration. Myocardial radioactivity was detected immediately after injection and reached a maximum during the first minute. The early phase (,5 min) of rapid loss from the left ventricular wall (blood pool) was followed by a slow loss of myocardial radioactivity. The injection of unlabeled metaraminol, but not saline, caused a rapid loss of myocardial radioactivity. It also caused a rapid loss of activity from the ventricular space, an observation consistent with a high spillover from the myocardium (data not shown).
FIGURE 1. Distribution of radioactivity after simultaneous intravenous injection of 11C-mHED (3.6 MBq, 22.8 nmolkg21) into 2 mice. Fifteen minutes after 11C-mHED, mouse 2 received metaraminol (40 mmolkg21 intravenously) and mouse 1 saline. Both mice received 18F-FDG (5 MBq) after 11C-mHED scan. Arrows indicate hearts.
FIGURE 2. Uptake of radioactivity in mouse heart after intravenous injection of 11C-mHED. (A–F) Images of thoraces of the 2 mice in Figure 1 (scan 1): mouse 1 (A2C); mouse 2 (D2F). Both hearts (arrows) were visualized at 5215 min after 11C-mHED (A and D). Heart was still visible at 25–45 min after saline (B) but not after metaraminol (E) administration. 18F-FDG confirmed heart positions (C and F). (G) Time–activity curves for left ventricular wall (s, mouse 1; d, mouse 2). (H) Normalized time–activity curves for scan 1, with results for 2 other scans (scans 2 and 3). In each scan, one mouse was injected with saline at 15 min after 11C-mHED (s, scan 1; D, scan 2; ,, scan 3) and the other with metaraminol (d 5 40 mmolkg21, scan 1; : 5 208 nmolkg21, scan 2; ; 5 100 nmolkg21, scan 3). Broken line shows single exponential fit to control data (7 mice) and dotted lines 95% inclusion limits. Solid line shows fits to metaraminol displacement for each dose.
11C-MHED AND
PET: FROM MAN
TO
MOUSE? • Law et al.
1271
There was a discrepancy between blood time–activity curves derived from PET images and those assessed by continuous arterial blood sampling (Supplemental Fig. 1; supplemental materials are available online only at http:// jnm.snmjournals.org). In the absence of corrections for scatter and spillover (Supplemental Fig. 2), it is not possible to obtain an input function by drawing a region in the left ventricle. Displacement of
11C-mHED
by Metaraminol
Because there is no noninvasive method for obtaining an input function, myocardial radioactivity (cpsmL21myocardium) was normalized to the total radioactivity in the mouse (cpsmouse) for each time frame. cps mL1 % cpsmouse5100 cps mL1
myocardium
=ðcpsmouse Þ: Eq. 1
Figure 2H shows normalized myocardial uptake for the 2 mice in Figures 2A22G, with the results for 2 other pairs of mice. For each pair, one mouse was injected with saline (intravenously) at 15 min after 11C-mHED (control mouse) and the other mouse with metaraminol (test mouse). For clarity of illustration, data points for minute time frames are shown. Metaraminol at 40 mmolkg21 or 208 nmolkg21 caused rapid loss of myocardial 11C-mHED, but 100 nmolkg21 had no significant effect. Injecting metaraminol at high doses (.500 nmolkg21) after 11C-mHED caused a rapid loss of myocardial radioactivity, approximately 70% being displaced during the first 5 min. Therefore, to assess the effect of metaraminol, 10- and 20-s frames were reconstructed for the first 10 min after displacement. Normalized time–activity curves were computed, omitting the first 10-s frame, during which metaraminol was being injected, and using 10-s, 20-s, and subsequent minute time frames. Normalized time–activity curves for control mice were fitted by a single exponential function bx
y 5 ae
;
Eq. 2
where y is myocardial activity, and x is the time after injection of 11C-mHED. Normalized time–activity curves for test mice were fitted by a single exponential function (Eq. 2) before injection of metaraminol (5–15 min after 11C-mHED) and by a single exponential function with a nondisplaceable background B (Eq. 3) after metaraminol administration (15–50 min after 11C-mHED). y 5 aebx 1B:
Eq. 3
To compare individual control mice, the washout of radioactivity was expressed as the rate constant b in Equation 2 for 5–50 min after injection and area under the curve (AUC) for 5–50 min (AUC5–50 min). Values for slope b ranged from 0.004 to 0.017 (mean, 0.0122; SE, 0.0015; n 5 7), and there was no correlation between the nanomolar dose of norepinephrine mimetics (mHED 11C-mHED
1272
THE JOURNAL
OF
plus metaraminol) injected (,2–77 nmolkg21) and slope b. Values for the AUC5–50 min (mean, 480; SE, 19; n 5 7) were not related to injected doses below 35 nmolkg21, but 2 animals that received doses of 62 and 77 nmolkg21 showed low values (250 and 212, respectively). For comparison with mice that received displacement metaraminol, data for control mice before or after injection of saline (5–15 min or 15–45 min after 11C-mHED) were also fitted by a single exponential function (Eq. 2). Estimates of rate b and AUCs (AUC for 5–15 min, AUC5–50 min, and AUC for 15–50 min) were not significantly different from those for the fits to data for 5–50 min. The early washouts of myocardial radioactivity (5–15 min after 11C-mHED) for control (0.013 6 0.002 min21, n 5 9) or test (0.014 6 0.003 min21, n 5 12) mice were not significantly different (P 5 0.72, Student t test; P 5 0.65, paired t test). Relationships between rate b for 15–50 min after 11CmHED or AUC for 15–45 min (AUC15–45 min) and metaraminol dose are shown in Figure 3. Results for control mice are plotted against metaraminol (dispensed) injected with 11C-mHED (Figs. 3A and 3C), and those for test mice are plotted against metaraminol (displacement) injected at 15 min after 11C-mHED (Figs. 3B and 3D). The sum of mHED (2–69 nmolkg 21) and dispensed metaraminol (1–24 nmolkg21 ) ranged from 3 to 104 nmolkg21. In some cases, therefore, the amount of norepinephrine mimetic (mHED plus metaraminol) injected at time 0 was greater than the displacement dose of metaraminol given at 15 min (40 nmolkg 21 to 40 mmolkg21). For control mice, the rate of loss (0.0092 6 0.0005 min21) did not depend on the dose of mimetic injected (Fig. 3A), but displacement metaraminol at doses greater than 100 nmolkg21 caused rapid loss (Fig. 3B). The displacement data were fitted to an equation of the form y 5 ðratemax xÞ=ð Kdis 1xÞ;
Eq. 4
where ratemax is the maximum rate of radioactivity loss, x is the displacement metaraminol dose, and Kdis is the halfsaturation dose. Nonlinear least-squares estimates of the parameters calculated using all data points in Figure 3B were ratemax 5 0.33 6 0.4 min21 and Kdis 5 296 6 138 nmolkg21 (R 5 0.87). Omitting the 2 mice with the highest dose of mimetic at time 0 and the mouse with no quality control information did not significantly change the fit. The AUC15–45 min for control mice (Fig. 3C) indicates that metaraminol doses less than 25 nmolkg21 injected with mHED doses less than 60 nmolkg21 have no effect on myocardial uptake of radioactivity. Values for test mice, plotted against displacement metaraminol (Fig. 4D), howed a sharp decrease above approximately 50 nmolkg21. Mice in 1 experiment received exceptionally high doses of 11C-mHED (68–93 nmolkg21). The AUCs for these mice were significantly lower than for mice that
NUCLEAR MEDICINE • Vol. 51 • No. 8 • August 2010
FIGURE 3. Displacement of 11C-mHED radioactivity from myocardium by intravenous injection of metaraminol. 11CmHED injected at time 0 (d, 2–60 nmolkg21; s, 70–90 nmolkg21) was associated with metaraminol (dispensed) (1–24 nmolkg21). No quality control was available for 1 scan of 2 mice (:). (A and C) Data for control mice as function of dispensed metaraminol. (B and D) Data for test mice as function of metaraminol (displacement) injected 15 min after 11C-mHED. (B) Solid line shows nonlinear least-squares fit to all data points. (D) Solid line shows nonlinear least-squares fit to data points for which 11 C-mHED is #60 nmolkg 21 . Broken lines show 95% confidence limits.
received similar doses of metaraminol but lower doses of 11C-mHED and were omitted from further analysis. AUCs for mice that received displacement metaraminol were fitted by a 4-parameter logistic equation: B y 5 D1ðA DÞ= 11 x= 10log C ;
Eq. 5
where y is the AUC; x is the displacement metaraminol dose (nmolkg21); C is the metaraminol dose for half-maximal displacement; and A, B, and D are parameters giving a half-maximal dose of 124 nmolkg21. Effect of Injected Dose on Myocardial Uptake of 11C-mHED
Studies in rats have shown that myocardial uptake of is reduced if the combined dose of mHED and metaraminol is greater than approximately 20 nmolkg21 and the half-saturation dose is approximately 100 nmolkg21 11C-mHED
FIGURE 4. Dose–effect relationships for myocardial uptake of 11C-mHED radioactivity. (A and B) In vivo PET. Saline or increasing doses of metaraminol (100, 200, or 500 nmolkg21) were coinjected with fixed amounts of dispensed 11C-mHED (44 nmolkg21) and metaraminol (0.8 nmolkg21) (:). Control mice (from Fig. 3) were injected with 11C-mHED (d, 2–60 nmolkg21; s, 70–90 nmolkg21) and metaraminol at less than 24 nmolkg21. (A) Rates of loss of radioactivity. (B) AUCs. (C) Ex vivo dissection. Saline or increasing doses of metaraminol were coinjected with fixed amounts of dispensed 11C-mHED (2–47 nmolkg21) and metaraminol (1–3 nmolkg21) into conscious mice. Radioactivity at 30 min after injection expressed as uptake index is shown. Solid lines represent nonlinear least-squares fits to data points and broken lines 95% confidence limits.
(15). The present displacement studies suggest that these doses are higher in mice (.100 nmolkg21). To further investigate this effect, saline or increasing doses of metaraminol were coinjected with a fixed amount of dispensed
11C-MHED AND
PET: FROM MAN
TO
MOUSE? • Law et al.
1273
11C-mHED
and metaraminol. Time–activity curves were constructed using minute time frames, and the rates or AUCs (computed using Eq. 2) were plotted against the combined dose of 11C-mHED and metaraminol (Figs. 4A and 4B). Data for all control mice in Figure 3 are plotted for comparison. There was no correlation between rate of loss of 11C-mHED radioactivity and injected dose. The AUC, however, decreased with doses above 100 nmolkg21. Assuming 11C-mHED, mHED, and metaraminol have similar affinities and binding potentials (bmax/ Kd) for uptake1 transporters (15), the data were fitted to: AUC 5 B1ðbmax =KÞð1=½11C=KÞ;
Eq. 6
where C is the injected dose (mHED 1 metaraminol), B is the nondisplaceable background, bmax is the maximum uptake, and K is the half-saturation constant (15) ( halfsaturation dose). B was set at 87.09 (B in Eq. 3 for displacement by 10 and 40 mmolkg21 metaraminol multiplied by 45 min), giving a K value of 132 6 59 nmolkg21 (R 5 0.86). Ex vivo dissection studies gave comparable results (Fig. 4C). 11C-mHED mixed with increasing concentrations of added metaraminol was injected as a bolus via the tail vein into conscious mice. The data were fitted to: Uptake index 5 B1ðbmax =KÞð1=½11C=KÞ;
Eq. 7
where C is the injected dose (mHED 1 metaraminol), B is the nondisplaceable background, bmax is the maximum uptake, and K is the half-saturation constant. B was fixed to 0.14, the uptake index at 10 mmolkg21, giving a K value of 215 6 32 nmolkg1 (R 5 0.99). Dose–effect relationships are summarized in Table 1. DISCUSSION
The present study demonstrates that myocardial sympathetic innervation can be visualized in mice using the established tracer 11C-mHED with the high-resolution quadHIDAC small-animal PET scanner. Displaceable specific uptake can be demonstrated and pharmacokinetic parameters measured. This opens an exciting window of opportunity for future work on sympathetic innervation in mouse models of cardiovascular diseases. The unique advantage of the quadHIDAC scanner is that it offers a high sensitivity and high spatial resolution, which is constant over a large field of view (13). Consequently, several mice may be scanned at the same time, allowing the effects of pharmacologic interventions to be compared using the same radiochemical preparation. This is especially valuable for neuroreceptor systems in which specific uptake of a radioligand is reduced as the amount of unlabeled ligand increases (i.e., as specific activity decreases).
1274
THE JOURNAL
OF
Notwithstanding the limitations imposed by the lack of absolute quantitation in small-animal PET scanners, tracer modeling may give estimates of relevant physiologic parameters. Essentially two methods have been applied to assess myocardial innervation using 11C-mHED in patient studies, both of which require an arterial input function based on serial measurements of blood activity, either by blood sampling or by selecting volumes of interest in a heart chamber (4,18). Serial blood sampling in mice, however, necessitates catheterization of major blood vessels, which is not practical if serial scans are to be obtained to assess the development of disease or therapeutic efficacy. Determining the input function from PET images is an attractive alternative, but spillover from myocardium to the ventricles (blood pool) makes this approach inappropriate for mice. The only option in mice, therefore, was to use a semiquantitative index as previously used in humans (19). Time–activity curves were constructed using normalized myocardial radioactivity (cpsmL21% cpsmouse). Both washout of 11C-mHED radioactivity and AUCs were calculated. The PET scans of mice showed changes in myocardial radioactivity (Fig. 2) that were comparable to those after a bolus injection of 11C-mHED in rats, assessed by ex vivo dissection (15,20) or PET (8). An early increase in activity was followed by a phase of rapid decrease (from 20 s to 2 min) reflecting rapid loss from vascular space and then by a phase of slow loss from the myocardium. Metabolite analysis was not performed in this study. In rats, however, although radioactive metabolites were detected in plasma, only parent 11C-mHED was detected in the myocardium (15,20). The data for 5–50 min (Fig. 2) were fitted by a monoexponential function on the basis that loss of 11C-mHED from the nerve terminals is a process that follows first-order kinetics. The rate of loss ranged from 0.04 to 0.176 min21; there was no relationship between rate of loss and injected nanomolar dose. Values for areas under the time–activity curves (AUC5–50 min) ranged from 400 to 540 for injected doses less than 60 nmolkg21. A bolus injection of unlabeled metaraminol at 15 min after injection of 11C-mHED displaced myocardial radioactivity (Fig. 2), as has been observed ex vivo in rats (15). To assess the effect of metaraminol dose, the time–activity curves after injection of metaraminol were fitted by a monoexponential function with a background. The rate of loss increased from 0.1 to 1 mmolkg21 with metaraminol dose and was fitted by Michaelis–Menten kinetics, assuming that displacement by metaraminol is a pseudobimolecular reaction. An alternative measure of the effect of metaraminol was to calculate AUCs after injection of saline or metaraminol (AUC15–45 min, Fig. 3). This parameter showed a sharp decrease as metaraminol dose was increased above approximately 50 nmolkg21. The study in which metaraminol was coinjected with 11C-mHED (Fig. 4) indicated that the rate of loss of myo-
NUCLEAR MEDICINE • Vol. 51 • No. 8 • August 2010
TABLE 1. Myocardial Retention of
11C-mHED
Study Displacement by metaraminol (PET) Coinjection of metaraminol 1 Coinjection of metaraminol 1
11C-mHED 11C-mHED
(PET) (dissection)
cardial 11C-mHED did not depend on injected dose, whereas the AUC5–50 min showed a clear dependence due to a reduced initial uptake. Comparable results were obtained by ex vivo dissection studies (Fig. 4). The rate of displacement of 11C-mHED by metaraminol was the parameter least sensitive to metaraminol dose; the dose for half maximum was approximately 300 nmolkg21 (Table 1). The competition studies indicated that although the rate of loss of 11C-mHED between 5 and 50 min after injection did not depend on injected dose, the AUC decreased as the total dose of mimetics (11C-mHED 1 mHED 1 metaraminol) increased; the half-saturation constant K was 132 6 59 nmolkg21. A similar relationship between uptake index assessed by ex vivo dissection studies and total dose was observed. In this case, the halfsaturation constant K was higher (215 6 32 nmolkg21) but not significantly so. The data presented in Figures 3 and 4 indicate that myocardial uptake of 11C-mHED in mice decreases if the total injected dose of mHED and metaraminol is greater than approximately 50 nmolkg21. This dose, however, may be lower in mouse models of heart disease because the kinetics of uptake and retention of 11C-mHED depend on the number of neuronal noradrenaline transporters in the myocardium and on their intrinsic activity. A decrease in the number of transporters without a change in their activity will reduce myocardial uptake of 11C-mHED (AUC) but not the half-saturation constant K. A decrease in transporter activity, however, may reduce K. In addition, an increase in endogenous norepinephrine, as is observed in many heart diseases, may reduce the apparent K for injected mimetics. CONCLUSION
The reuptake of norepinephrine (uptake1) by myocardial sympathetic nerve endings can be successfully visualized in the mouse using 11C-mHED with the quadHIDAC small-animal PET scanner, opening exciting perspectives toward future preclinical studies in mouse models. A semiquantitative index can be used to compare uptake1 in individual mice and gain information about pharmacokinetic parameters. Myocardial uptake of 11C-mHED is dependent on the nanomolar dose of both mHED and metaraminol, the synthesis precursor; a total mimetic dose greater than approximately 50 nmolkg21 results in a decrease in myocardial uptake. Consequently, it is essential to ensure that the doses of 11C-mHED plus any meta-
Equation
Regression coefficient R
Dose for half-maximum effect (nmolkg21)
4 5 6 7
0.866 0.978 0.858 0.993
296 124 132 6 59 215 6 32
raminol used with PET to study uptake1 in mice are below approximately 50 nmolkg21. ACKNOWLEDGMENTS
We thank Christine Ba¨ tza, Irmgard Hoppe, Anne Kanzog, Sandra Schro¨ er, Daniel Burkert, and Sven Fatum for excellent technical assistance. This work was supported by the Deutsche Forschungsgemeinschaft, SFB 656 (projects A5, B2, B3, and Z5), by the Interdisciplinary Center for Clinical Research (IZKF core unit SmAP), Mu¨nster, Germany, and by the EU-FP6 project Diagnostic Molecular Imaging (DiMI) (WP 11.1), LSHB-CT-2005512146. REFERENCES 1. Brodde OE, Leinewever K. Autonomic receptor systems in the failing and aging human heart: similarities and differences. Eur J Pharmacol. 2004;500: 167–176. 2. Lautama¨ki R, Tipre D, Bengel FM. Cardiac sympathetic neuronal imaging using PET. Eur J Nucl Med Mol Imaging. 2007;34:S74–S85. 3. Scha¨fers M, Dutka D, Rhodes CG, et al. Myocardial presynaptic and postsynaptic autonomic dysfunction in hypertrophic cardiomyopathy. Circ Res. 1998;82:57–62. 4. Scha¨fers M, Lerch H, Wichter T, et al. Cardiac sympathetic innervation in patients with idiopathic right ventricular outflow tract tachycardia. J Am Coll Cardiol. 1998;32:181–186. 5. Wichter T, Scha¨fers M, Rhodes CG, et al. Abnormalities of cardiac sympathetic innervation in arrhythmogenic right ventricular cardiomyopathy: quantitative assessment of presynaptic norepinephrine reuptake and postsynaptic b-adrenergic receptor density with positron emission tomography. Circulation. 2000;101:1552–1558. 6. Kies P, Wichter T, Scha¨fers M, et al. Abnormal myocardial presynaptic norepinephrine recycling in patients with the Brugada syndrome. Circulation. 2004;110:3017–3022. 7. Schelbert HR, Inubushi M, Ross RS. PET imaging in small animals. J Nucl Cardiol. 2003;10:513–520. 8. Tipre DN, Fox JJ, Holt DP, et al. In vivo PET imaging of cardiac presynaptic sympathoneuronal mechanisms in the rat. J Nucl Med. 2008;49:1189–1195. 9. Law MP, Wagner S, Kopka K, Pike VW, Schober O, Scha¨fers M. Are [Omethyl-11C]derivatives of ICI 89,406 b1-adrenoceptor selective radioligands suitable for PET? Eur J Nucl Med Mol Imaging. 2008;35:174–185. 10. Hume SP, Gunn RN, Jones T. Pharmacological constraints associated with positron emission tomographic scanning of small laboratory animals. Eur J Nucl Med. 1988;25:173–176. 11. Jeavons AP, Chandler RA, Dettmar CAR. A 3D HIDAC-PET camera with submillimetre resolution for imaging small animals. IEEE Trans Nucl Sci. 1999;46:468–473. 12. Missimer J, Madi Z, Honer M, Keller C, Schubiger A, Ametamey SM. Performance evaluation of the 16-module quad-HIDAC small animal PET camera. Phys Med Biol. 2004;49:2069–2081. 13. Scha¨fers KP, Reader AJ, Kriens M, Knoess C, Schober O, Scha¨fers M. Performance Evaluation of the 32-module quadHIDAC small animal PET scanner. J Nucl Med. 2005;46:996–1004. 14. Law MP. Demonstration of the suitability of CGP 12177 for in vivo studies of b-adrenoceptors. Br J Pharmacol. 1993;109:1101–1109.
11C-MHED AND
PET: FROM MAN
TO
MOUSE? • Law et al.
1275
15. Law MP, Osman S, Davenport RJ, Cunningham VJ, Pike VW, Camici PG. Biodistribution and metabolism of [N-methyl-11C]-m-hydroxyephedrine in the rat. Nucl Med Biol. 1997;24:417–424. 16. Rosenspire KC, Haka MS, vanDort ME, et al. Synthesis and preliminary evaluation of carbon-11-meta-hydroxyepehdrine: a false transmitter agent for heart neuronal imaging. J Nucl Med. 1990;31:1328–1334. 17. Reader AJ, Ally S, Bakatselos F, et al. One-pass list-mode EM algorithm for high resolution 3D PET image reconstruction into large arrays. IEEE Trans Nucl Sci. 2002;49:693–699.
1276
THE JOURNAL
OF
18. Mu¨nch G, Nguyen N, Nekolla BS, et al. Evaluation of sympathetic nerve terminals with [11C]epinephrine and [11C]hydroxyephedrine and positron emission tomography. Circulation. 2000;101:516–523. 19. Malizia AL, Melicher JK, Rhodes CG, et al. Desipramine binding to noradrenaline reuptake sites in cardiac sympathetic neurones in man in vivo. Eur J Pharmacol. 2000;391:263–267. 20. Thackeray JT, Beanlands RS, Dasilva JN. Presence of specific 11C-metahydroxyephedrine retention in heart, lung, pancreas, and brown adipose tissue. J Nucl Med. 2007;48:1733–1740.
NUCLEAR MEDICINE • Vol. 51 • No. 8 • August 2010