Characterization of the four isomers of 123I-CMICE-013: A potential SPECT myocardial perfusion imaging agent

Bioorganic & Medicinal Chemistry 22 (2014) 2033–2044

Contents lists available at ScienceDirect

Bioorganic & Medicinal Chemistry journal homepage: www.elsevier.com/locate/bmc

Characterization of the four isomers of 123I-CMICE-013: A potential SPECT myocardial perfusion imaging agent Lihui Wei a,b,c,⇑, , Corinne Bensimon a, , Xuxu Yan a, Julia Lockwood b,c, Wei Gan b,c, R. Glenn Wells b,c, Yin Duan a,c, Pasan Fernando a,b,c,d, Bram Gottlieb a, Wayne Mullett a, Terrence D. Ruddy b,c a

Nordion Inc., 447 March Road, Ottawa, ON K2K 1X8, Canada Division of Cardiology, University of Ottawa Heart Institute, 40 Ruskin Street, Ottawa, ON K1Y 4W7, Canada Canadian Molecular Imaging Center of Excellence (CMICE), Nordion Lab/University of Ottawa Heart Institute, 40 Ruskin Street, Ottawa, ON K1Y 4W7, Canada d Department of Cellular and Molecular Medicine, Faculty of Medicine, University of Ottawa, Ottawa, ON K1H 8M5, Canada b c

a r t i c l e

i n f o

Article history: Received 12 January 2014 Revised 12 February 2014 Accepted 24 February 2014 Available online 4 March 2014 Keywords: 123 I-CMICE-013 127 I-CMICE-013 Myocardial perfusion imaging SPECT

a b s t r a c t Myocardial perfusion imaging (MPI) with single photon emission computed tomography (SPECT) is widely used in the assessment of coronary artery disease (CAD). We have developed 123I-CMICE-013 based on rotenone, a mitochondrial complex I (MC-1) inhibitor, as a promising new MPI agent. Our synthesis results in a mixture of four species of 123I-CMICE-013 A, B, C, D. In this study, we separated the four species and evaluated their biodistribution and imaging properties. The cold analogs 127I-CMICE-013 A, B, C, D were isolated and characterized and their chemical structures proposed. Methods: 123I-CMICE-013 was synthesized by radiolabeling rotenone with Na123I in trifluoroacetic acid (TFA) with iodogen as the oxidizing agent at 60 °C for 45 min, and the four species were separated by RP-HPLC. The cold analogs 127 I-CMICE-013 A, B, C and D were isolated with a similar procedure and characterized by NMR and mass spectrometry. Biodistribution and microSPECT imaging studies were carried out on normal rats. Results: We propose the mechanism of the rotenone iodination and the structures of the four species. First, I+ forms an intermediate three-membered ring with 60 and 70 carbons. Second, the lone electron pair of the water molecule attacks the 60 or 70 -carbon, following by the formation of 60 -OH, and 70 -I bonds as in major products C and D, or 60 -I and 70 -OH bonds as in minor products A and B. The weaker 60 -I bond in the intermediate prompts the nucleophilic attachment of water at the favorable 60 -carbon to generate C and D. MicroSPECT images of 123I-CMICE-013 A, B, C, D in rats showed clear visualization of myocardium and little interference from lung and liver. The imaging time activity curves and biodistribution data showed complex profiles for the four isomers, which is not expected from the structure activity relationship theory. Conclusion: 123/127I-CMICE-013 A and B are constitutional isomers with C and D, while A and C are diastereomers of B and D, respectively. Overall, the biological characteristics of the four species are not correlated perfectly with their molecular structures. Ó 2014 Elsevier Ltd. All rights reserved.

1. Introduction Stress myocardial perfusion imaging (MPI) has been increasingly utilized in the assessment of coronary artery disease (CAD).1–3 The most commonly used MPI modality is single photon emission computed tomography (SPECT),4 in conjunction with one of three SPECT radiotracers, 99mTc-sestamibi, 99mTc-tetrofosmin, and 201Tl.5–7 Important information has been obtained with these agents for evaluation of myocardial perfusion and viability in patients. ⇑ Corresponding author. Tel.: +1 613 798 5555x16500; fax: +1 613 595 4599.  

E-mail address: [email protected] (L. Wei). Lihui Wei and Corinne Bensimon contributed equally to this study.

http://dx.doi.org/10.1016/j.bmc.2014.02.052 0968-0896/Ó 2014 Elsevier Ltd. All rights reserved.

However, these radiotracers have certain suboptimal pharmacokinetic characteristics. 201Tl SPECT MPI has poor image quality in obese patients, redistribution to non-target tissues over time, and fair radiodosimetry.2,5 A shortcoming with both 99mTc MPI agents and 201Tl is the disproportional low myocardial uptake under stress conditions with myocardial hyperaemia (the ‘roll-off’ phenomenon).2,8–10 201 Tl is a potassium analog that accumulates in myocardial cells by uptake mediated by Na+/K+ adenosine triphosphatase.5,11 99mTcsestamibi and 99mTc-tetrofosmin, lipophilic and cationic complexes, accumulate in cardiac cells based on the mitochondrial membrane potential.12–15 Most recently, a new class of radiotracers that specifically target complex I of the mitochondrial electron

2034

L. Wei et al. / Bioorg. Med. Chem. 22 (2014) 2033–2044

transport chain (mitochondrial complex I: MC-I) has demonstrated superior properties compared to current MPI radiotracers.16–30 Mitochondria are abundant in high energy consuming tissues such as in the heart, where they comprise 20–30% of the myocardial intracellular volume.31 Three well-known potent MC-I inhibitors are rotenone,16–19,22,32 fenazaquin,21 and pyridaben,20,23–30 and have been radiolabeled with 99mTc, 125I/123I and 18F. The 18F pyridaben analog (18F-BMS-747158-02) is the most extensively studied tracer with promising results in preclinical studies using small and large animal models and on-going clinical studies.20,23–27 18 F-BMS-747158-02 is a positron emission tomography (PET) radiotracer and takes advantage of the higher sensitivity for tracer detection, better spatial resolution, and accurate attenuation correction of PET. However, the clinical use of cardiac PET is limited due to the limited availability of PET cameras for nuclear cardiology. The majority of PET and PET/CT scanners in North America are used for oncology applications. Recently, we developed a promising new radiotracer for SPECT MPI, 123I-CMICE-013, a 123I labelled rotenone analog. Our research indicated that 123I-CMICE-013 exhibits high myocardial uptake, very good target to background ratios, and favourable biodistribution characteristics.33 Rotenone has been previously radiolabeled with 125I and 123I through a tin precursor, resulting in 70 -iodinication on a 60 ,70 -double bond, and E and Z-isomers.16,17,22 The biodistribution profiles of the two isomers are significantly different.22 Our previous study showed that 123I-CMICE-013 contains four species, A, B, C, and D.33 Due to the different iodination mechanism, it is expected that the type of isomerisation is different than the reported 70 -125I/123Irotenone. Because the retention times of the four isomers of 123 I-CMICE-013 on HPLC are very close to each other, it is challenging to completely separate them on large scale. The radiotracer used for pre-clinical animal studies included all four species, 123 I-CMICE-013 A, B, C, D, with D as the major product.33 It is crucial to explore the chemical structures of the four species, compare their biological properties, and understand the contribution of each compound to the mixed product in our pre-clinical studies. In this study, we report the synthesis, isolation and chemical, biological characterization of 123I-CMICE-013 A, B, C, D, and their cold analogs. 2. Materials and methods 2.1. Materials All chemicals and solvents were analytical grade and were used without further purification. Rotenone was purchased from Sigma Aldrich. 123I was provided by Nordion Inc. Dionex GP50 gradient pump, Water’s Fraction Collector III, and Rheodyne six-port sample injection valve (7725i) were used for the HPLC purification of 123 I-CMICE-013 compounds. Purification of 127I-CMICE-013 compounds and analytical HPLC were performed on a Water0 s system with 1525 Pump, 2998 Photodiode Array Detector, and 717+ Autosampler with 200 lL injection loop. The UV detection wavelength was set to 290 nm. The injection volume was 1 lL for radioactive compounds, and 10 lL for cold compounds. Perkin Elmer’s 150TR Flow Scintillation Analyzer was used as radiomatic detector. Capintec dose calibrator (CRC-25R) was used to measure the radioactivity. 1D, 2D Nuclear Magnetic Resonance (NMR) was measured on Varian Inova NMR spectrometers. Nuclear Overhauser Effect Spectroscopy (NOESY) was measured with a 400 msec mixing time. High Resolution Mass Spectrometry (HRMS) was measured on Micromass GCT (GC-EI Time of Flight Mass Spectrometer) under positive mode.

2.2. Preparation of 123I-CMICE-013 A, B, C, D for microSPECT imaging and biodistribution study 463 MBq (12.5 mCi) of Na123I solution in 0.1 M NaOH (42–62 lL) was added to a 1.5 mL BioRad vial. Subsequently, 170 lL of rotenone solution (2.5 mg/mL) in trifluoroacetic acid (TFA) and 30 lL of iodogen solution (0.75 mg/mL) in TFA were added to the vial. The mixture was heated on a thermomixer (Eppendorf) at 60 °C, 600 rpm for 45 min. Four reaction vials with total activity of 50 mCi were set up. After cooling at room temperature for 5 min, the solutions from four reaction vials were combined, and the reaction mixture was purified by reverse-phase HPLC using a preparative Luna C18(2), 5 lm, 100 Å, 250  21.2 mm column (Phenomenex, CA, USA) at ambient temperature with ethanol: water = 48:52 (v/v) as mobile phase, and a Dionex pump with a flow rate of 5 mL/min. The four components of 123I-CMICE-013 A, B, C and D were collected at retention times of 40-43, 45-48, 57-59, and 60-62 mins, respectively, and subjected to analytical HPLC using a Luna C18(2), 5 lm, 100 Å, 250  4.6 mm column (Phenomenex, CA, USA) under ambient temperature with ethanol: water = 48:52 (v/v) as mobile phase, and a flow rate of 1 mL/min. Retention times of 123I-CMICE-013 A, B, C, D on analytical HPLC were 11.5, 12.9, 15.7, and 16.6 min, respectively. The purities of 123 I-CMICE-013 A, B, C, D were 100%, 80-85%, 95% and 96-97%, respectively. 123I-CMICE-013 B was subjected to second purification using the preparative Luna C18(2) column, after which the purity was increased to 100%. The four components were heated at 40 °C under a constant supply of nitrogen to evaporate ethanol. Charcoal filter was used as a vent and also to absorb free 123I. The products were reconstructed in 5% ethanol (v/v) in 10 mM NaOAc pH 6.5 to a volume of 0.8–1 mL, and analyzed by HPLC again before injected to animals. The pH of the final formulation was around 5. The radiochemical purity, as determined by analytical HPLC, was 100%, 100%, 95% and 96% for 123I-CMICE-013 A, B, C and D, respectively. The recovery yields after radiolabeling and HPLC purification for 123I-CMICE-013 A, B, C, D were approximately 6%, 2%, 6% and 12%, respectively. The total yield for the four radioiodinated species was 26%. The activity of the tracer was measured on the day of tracer production by dose calibrator. The tracer was then stored at 80 °C for 3–5 days to allow the activity to decay. 100–150 lL of product was injected into HPLC to obtain a UV peak at 290 nm wavelength with the same retention time as the radiomatic peak. The UV peak area was converted to weight (lg) using a calibration curve generated from a set of rotenone standards. The specific activity (mCi/lg) was calculated and converted to mCi/ lmol and GBq/lmol. Using this method, the specific activities of the four tracers were estimated to be 14.8–22.2 GBq/lmol (400–600 mCi/lmol). 2.3. Preparation of 127I-CMICE-013 A, B, C and D: the cold analogues of 123I-CMICE-013 A, B, C and D Rotenone (42.5 mg) in TFA (17 mL) was mixed with NaI (81 mg) in NaOH solution (0.1 M, 5 mL). Iodogen™ (45 mg) in TFA (3 mL) was added to the mixture at room temperature. The reaction mixture was stirred at 60 °C for 45 min and concentrated under reduced pressure. Water (20 mL) was poured in and the raw product was extracted with CH2Cl2 (20 mL  3), dried over anhydrous Na2SO4, filtered, and concentrated under reduced pressure to yield a green oil. This residue was dissolved in CH2Cl2 (1 mL) and subjected to HPLC purification using a preparative Luna C18(2), 5 lm, 100 Å, 250  21.2 mm column. The sample was eluted at a flow rate of 6 mL/min, using ethanol/water = 48:52 (v/v) as mobile phase. The detector was set to 290 nm. The

L. Wei et al. / Bioorg. Med. Chem. 22 (2014) 2033–2044

products at retention time 35–45 min were collected and evaporated to dryness to yield a mixture of 127I-CMICE-013 A and 127I-CMICE-013 B as a white solid (1.0 mg, 1.7%). In order to separate A and B, the mixture was further purified by HPLC using the same HPLC conditions. The desired products were collected and evaporated to dryness to yield 127I-CMICE-013 A (0.4 mg, 0.7% yield) and 127I-CMICE-013 B (0.1 mg, 0.2% yield) as a white solid (see Scheme 1 for molecular structure of A and B). Multiple batches were prepared in order to collect enough amount for NMR and mass spectroscopy analysis. The chemical purity was estimated from NMR spectrum. 127 I-CMICE-013 A: purity: 95%. 1H NMR (300 MHz, CDCl3) d 7.83 (d, J = 8.6 Hz, 1H), 6.74 (s, 1H), 6.49 (d, J = 8.6 Hz, 1H), 6.43 (s, 1H), 5.02–4.89 (m, 1H), 4.62 (dd, J = 12.1, 3.1 Hz, 1H), 4.28–4.10 (m, 2H), 3.92 (d, J = 11.8 Hz, 1H), 3.84 (d, J = 4.0 Hz, 2H), 3.79 (s, 3H), 3.74 (s, 3H), 3.31 (dd, J = 14.3, 7.9 Hz, 1H), 3.13 (dd, J = 15.9, 8.2 Hz, 1H), 1.91 (s, 3H); 13C NMR (300 MHz, CDCl3) d 188.97, 166.40, 157.79, 149.46, 147.38, 143.85, 130.15, 113.65, 112.54, 110.15, 104.89, 104.66, 100.89, 88.44, 72.24, 71.29, 66.28, 59.80, 56.32, 55.89, 44.61, 32.56, 26.61. HRMS for C23H23IO7: calcd 538.0488, found 538.0476. 127 I-CMICE-013 B: purity: 90%. 1H NMR (500 MHz, CDCl3) d 7.82 (d, J = 7.7 Hz, 1H), 6.74 (s, 1H), 6.53 (d, 1H), 6.46 (s, 1H), 5.09 (t, J = 8.7 Hz, 1H), 4.96 (s, 1H), 4.70–4.55 (m, 1H), 4.19 (d, J = 12.1 Hz, 1H), 3.88–3.67 (m, 9H), 3.53–3.34 (m, 1H), 3.29–3.10 (m, 1H), 1.86 (s, 3H); 13C NMR (500 MHz, CDCl3) d 188.91, 166.87, 157.71, 149.49, 147.36, 143.84, 130.04, 113.60, 112.26, 110.20, 104.72, 104.55, 100.90, 88.09, 72.25, 71.43, 66.17, 58.68, 56.29, 55.83, 44.58, 31.65, 23.52. HRMS for C23H23IO7: calcd 538.0488, found 538.0473. The products at retention time 51–63 min were also collected and evaporated to dryness to yield a mixture of 127I-CMICE-013 C and 127I-CMICE-013 D as a white solid (5.1 mg, 8.6%). The mixture was further purified by HPLC using the same conditions as above to separate 127I-CMICE-013 C and 127I-CMICE-013 D. The products were collected and evaporated to dryness to yield 127I-CMICE013 C (0.1 mg, 0.2% yield) and 127I-CMICE-013 D (3 mg, 5.1% yield) as white solid (see Scheme 1 for molecular structure of C and D). 127 I-CMICE-013 C: purity: 94%. 1H NMR (300 MHz, CDCl3) d 7.85 (d, J = 8.6 Hz, 1H), 6.76 (s, 1H), 6.52 (d, J = 8.6 Hz, 1H), 6.48 (s, 1H), 5.02 (t, J = 9.2 Hz, 1H), 4.98–4.87 (m, 1H), 4.64 (dd, J = 12.1, 3.0 Hz, 1H), 4.20 (d, J = 12.1 Hz, 1H), 3.87 (d, J = 3.9 Hz, 1H), 3.84 (s, 3H), 3.79 (s, 3H), 3.50 (d, J = 9.0 Hz, 1H), 3.43

2035

(d, J = 9.0 Hz, 1H), 3.19 (d, J = 9.2 Hz, 2H), 1.37 (s, 3H). 13C NMR (300 MHz, CDCl3) d 189.05, 166.65, 157.90, 149.51, 147.39, 143.88, 129.98, 113.68, 113.35, 110.21, 104.89, 104.62, 100.92, 87.73, 72.25, 72.21, 66.25, 56.33, 55.88, 44.63, 27.74, 23.04, 15.57. HRMS for C23H23IO7: calcd 538.0488, found 538.0501. 127 I-CMICE-013 D: purity: 94%. NMR and HRMS data has been reported in our previous work.33 2.4. Micro-SPECT imaging studies in rats All animal experiments were conducted in compliance with the guidelines of the Canadian Council on Animal Care (CCAC) and with approval from the Animal Care Committee (ACC) at the University of Ottawa. Male Sprague Dawley rats (Charles River Laboratories, MA, USA) weighing approximately 250–450 g were used for this experiment. Rats were sedated under light anesthetic with 1–2% isoflurane delivered through a nose cone. A small volume catheter was placed in the tail vein for tracer injection. 14.8–55.5 MBq (0.4–1.5 mCi), 14.8–22.2 MBq (0.4–0.6 mCi), 22.2–55.5 MBq (0.6–1.5 mCi) and 62.9–92.5 MBq (1.7–2.5 mCi) of 123I-CMICE-013 A (n = 3), B (n = 4), C (n = 4), D (n = 4), respectively, were administered. The volume of the injected tracer was 0.8 to 1 mL for each rat. The rats were placed in the supine position on the heated scanner bed and scanned for twelve 10 min scans over a 2 h period on a Bioscan NanoSPECT/CT in vivo preclinical scanner (Washington DC). Image acquisition began immediately following tracer injection. The scanner is a 4-head scanner with 9  2.5 mm diameter pinhole collimation on each head. The data were acquired with a spiral scan covering head to mid-abdomen of the rat. The linearity of the microSPECT camera was calibrated with a set of phantoms with different amounts of activities. Physiologic parameters were monitored and recorded including respiratory rate, heart rate, ECG and body temperature during the scanning process (Model 1025T Small Animal Monitoring System, SA Instruments, Stony Brook, NY). A CT scan was acquired prior to the SPECT acquisition for localization of tracer uptake. The resolution is 100 lm with optical magnification of 1.3. Other CT parameters are: tube voltage: 45 kV; tube current: 177 lA; time per projection: 500 ms; number of projection: 360; total acquisition time: 6 min. Images were reconstructed using HiSPECT software provided by the camera manufacturer. The images acquired at different time points after tracer injection were analyzed with MATLAB-based software (Mathwork, version R2007b, Boston, MA). Manual volumes of

Scheme 1. Proposed iodination mechanism and structures of

123

I-CMICE-013 A, B, C, and D.

2036

L. Wei et al. / Bioorg. Med. Chem. 22 (2014) 2033–2044

interest (VOI) were drawn around the heart and liver areas of focal uptake in the summed images. The total counts in the VOI at each time point were decay corrected and recorded. The converting factor from measured counts to activity (Bq) was pre-determined by measuring a set of I-123 calibration sources with known activities. The percentage injected dose (%ID) was calculated by dividing the activity at the VOI region by the injection dose. The volume of the organ was estimated as the number of voxels in the VOI surrounding the organ. Each voxel has a size of 0.6  0.6  0.6 mm, which equals to 2.16  104 mL. The activity density was then calculated as %ID/mL, and time activity curves were generated.

vs. 127I-CMICE-013 C, D. The two mixtures were further purified by a second HPLC step to yield the four species. Several batches of products were prepared and combined in order to collect enough compounds for NMR and/or mass spectrometry characterization. Figure 2 shows that the UV traces of the HPLC chromatograms of the four species match the radiomatic traces in Figure 1b. Note that Figure 2 shows the chromatograms of one fraction from HPLC purification for each compound. The purities of the combined compounds for NMR and mass spectrometry analysis were a little lower for 127I-CMICE-013 A, B, which can be observed from the impurities in NMR spectra (see Figs. 3 and 4 and Figures S8–S12 in the Supplemental Information).

2.5. Biodistribution studies 3.3. Characterization of Following the imaging at 2 h post injection, the animals were sacrificed. The heart, liver, kidney, muscle, femur, spleen, blood, brain, intestine, lung, stomach, urine/bladder, testes and thyroid were extracted, weighed and analyzed for total gamma counts (Wizard 3 2480 Automatic Gamma Counter, Perkin Elmer). The tissue uptake was decay corrected, and the percentage of injected dose per gram (%ID/g) was then calculated. 2.6. Statistical methods All of the data are presented as mean ± SD. Statistical analysis was performed using GraphPad PRISM (San Diego, CA). Repeated measures two-way (two-factor) ANOVA with Bonferroni post test was performed on the time activity curves (Fig. 6) of the imaging data. The compound and time point are the two factors. To determine the difference of the biodistribution of the four compounds (groups) and 123I-CMICE-013 in our previous study33 (Fig. 7), a one-ANOVA with a Bonferroni post hoc multiple comparison test was used. For comparing imaging and biodistribution data (Fig. 8), a two-tailed unpaired Student t-test was used. Differences at the 95% confidence level (p <0.05) were considered significant. 3. Results 3.1. Preparation of 123I-CMICE-013 A, B, C, D for microSPECT imaging and biodistribution study The four radioiodinated rotenone derivatives 123I-CMICE-013 A, B, C, D were synthesized following similar procedure as 123 I-CMICE-013.33 Briefly, rotenone was mixed with Na123I and iodogen in trifluoroacetic acid (TFA). After heated at 60 °C for 45 min, the reaction mixture (see Fig. 1a) for the HPLC chromatogram) was subjected to HPLC purification. A preparative Luna C18(2) column was used with ethanol/water = 48:52 (v/v) as mobile phase and a flow rate of 5 mL/min. Analytical HPLC chromatograms show the purities of the components were higher than 95%, except for 123I-CMICE-013 B, with purity between 80% to 85%. A second round of HPLC purification with the same preparative column was performed to achieve 100% purity for B. Fig. 1b shows the typical analytical HPLC chromatograms and the purities of the final products 123I-CMICE-013 A, B, C, D used for microSPECT imaging and biodistribution studies. 3.2. Preparation of

127

I-CMICE-013 A, B, C, D

The cold analogs 127I-CMICE-013 A, B, C, D were prepared following similar procedure as 123I-CMICE-013 A, B, C and D. Based on the amount of rotenone used, the cold reaction is a 100-fold scale up reaction. In order to separate the four species, the reaction mixture was purified by two rounds of HPLC using preparative Luna C18 column. The first HPLC purification separated 127I-CMICE-013 A, B

127

I-CMICE-013 D

The purity of the 127I-CMICE-013 D compound was 94%, with 6% of I-CMICE-013 C. In our previous work,33 we proposed the iodination position and the structure of 127I-CMICE-013 D as shown in Scheme 1. Here we intend to provide more thorough explanation to confirm our hypothesis. The published 1H NMR spectral data of rotenone34 showed the two protons at 70 -carbon as two singlets at 5.01 and 4.88 ppm, respectively. In 127I-CMICE013 D33 (see Fig. S1 in Ref. 33), the signals of the 70 -proton are shifted upfield to 3.46 and 3.36 ppm, and also changed to two sets of doublets, with coupling constant 10.5 Hz, typical for geminal protons. 13C NMR of rotenone showed 60 and 70 -carbons as 142.83 and 112.40 ppm, respectively. In 127I-CMICE-013 D33, chemical shifts of 60 and 70 -carbons are more upfield to 71.71 and 17.46 ppm33 (see Fig. S2 in Ref. 33), respectively. The NMR spectra changes from rotenone to 127I-CMICE-013 D clearly suggest that the 60 , 70 -double bond is reduced to a single bond, and rotenone is iodinated at 70 -carbon. The mass spectrometry shows the molecular weight of 127I-CMICE-013 D is 538,33 which is 17 higher than rotenone plus iodine, indicating a hydroxy group is added to the 60 -carbon. To gather further evidence of the proposed structure of 127 I-CMICE-013 D, we found the closest compound, a 60 -OH, 70 -Cl derivative of rotenone, compound 21 in previous studies.35,36 Both 1 H and 13C NMR spectra of the two compounds are very similar. For example, compound 21 shows the 70 -proton at 3.60–3.76 ppm, 60 -carbon at 73 ppm. The difference of the 70 -carbon in compound 21 (50.1 ppm) and 127I-CMICE-013 D (17.5 ppm) is due to the lower electron withdraw power of iodine versus chlorine. 127

3.4. Characterization of 127

127

I-CMICE-013 C

I-CMICE-013 C has the same molecular weight (538 daltons) as I-CMICE-013 D as shown in mass spectrometry. The retention times of the two compounds are very close to each other, which makes the separation quite challenging. With two rounds of HPLC purification using preparative column, we were able to separate 127I-CMICE-013 C with D. The NMR spectra of C shows similar pattern as D (compare Figs. S1-S5 in Supplemental Information and Figs. S1–S5 in Ref. 33). The 70 and 40 proton signals of C are slightly different than D, suggesting the different environment around the chiral center 60 -carbon for C and D. The similarity of HPLC retention time and the NMR spectra and the same molecular weights indicate that 127I-CMICE-013 C is a diastereomer of D (see Scheme 1). The published 60 -OH, 70 -Cl derivative of rotenone (compound 21 in Refs. 35,36) also had two isomers with different stereochemistry at 60 -carbon. In fact, the two isomers of compound 21 were not separated, as they appear as a single peak in C18 HPLC, but the 1H and 13C NMR spectra indicated a mixture of the two isomers. 127

2037

L. Wei et al. / Bioorg. Med. Chem. 22 (2014) 2033–2044

(a)

C

A

D

Reacon mixture

B

(b)

A: 100 %

B: 100 %

C: 95 %

D: 96 %

Figure 1. HPLC chromatograms (radiomatic trace) of: (a) the reaction mixture; and (b) biodistribution studies).

3.5. NOE NMR measurement of CMICE-013 D

127

I-CMICE-013 C and

127

I-

In order to gather evidence of the stereochemistry at 60 -carbon for the diastereomeric pairs 127I-CMICE-013 C and D, we attempted the Nuclear Overhauser Effect NMR measurement. The NOE coupling pattern of 127I-CMICE-013 C and D is different (see Figs. S6 and S7). The stereo-orientation of 80 -CH3 and 60 -OH is different around the 60 -chiral center in 127I-CMICE-013 C and D, resulting in different proton coupling interaction as shown in Table 1. In 127 I-CMICE-013 C, 80 -proton shows strongest coupling with 40 -proton, while in D, 80 -proton has stronger interaction with 70 -proton. The assignment of the correct diastereomers for 127I-CMICE-013 C and D is very challenging. In order to confirm the absolute stereochemical structure, the NOE coupling needs to correlate with the bond distances calculated from ab initio quantum chemistry method. X-ray crystallography is another method to obtain the absolute structure.

123

I-CMICE-013 A, B, C, D (final purified products used for microSPECT and

3.6. Characterization of

127

I-CMICE-013 A and

127

I-CMICE-013 B

The high resolution mass spectrometry (HRMS) of 127I-CMICE-013 A and B shows their molecular weights are 538 daltons, same as 127 I-CMICE-013 C and D, suggesting that A and B are constitutional isomers of C and D. The HPLC retention time (RT) difference of A and B is similar to the RT difference of C and D (see Figs. 1 and 2). Based on the hypothesis of C and D, we propose that A and B are also diastereomers (see Scheme 1). The peak assignment for the 1H and 13C NMR of 127I-CMICE-013 A (Figs. 3 and 4) is based on 2D COSY, HMQC and HMBC NMR spectra (see Figs. S8–S10). Compared to the two sets of doublet of 70 -protons at 3.46 and 3.36 ppm for 127I-CMICE-013 D and 3.50 and 3.43 ppm for C, the 70 -protons for 127I-CMICE-013 A undergo a down-field shift to around 3.9 ppm. Due to the overlap of 70 , 12a, OCH3 proton and the solvent (ethanol) peaks, the doublet–doublet feature of the 70 -proton is not as clearly indicated as in 127I-CMICE013 D. The 70 -carbon of 127I-CMICE-013 A is significantly shifted

2038

L. Wei et al. / Bioorg. Med. Chem. 22 (2014) 2033–2044

A: 100%

B: 99%

C: 94%

D: 94%

Figure 2. HPLC chromatograms (UV trace) of

127

Figure 3. 1H NMR of

127

down-field to 71.3 ppm from 17.5 ppm in 127I-CMICE-013 D and 15.4 ppm in C. In contrast, the 60 -carbon undergoes up-field shift from 71.7 ppm in 127I-CMICE-013 D and 72.2 ppm in C to

I-CMICE-013 A, B, C, D (HPLC fractions).

I-CMICE-013 A.

59.8 ppm in A. The 50 -proton is also shifted up-field from 4.9 ppm in 127I-CMICE-013 D and 5.0 ppm in C to 4.2 ppm in A. The differences of the 50 , 70 -proton and 60 , 70 -carbon peaks suggest

L. Wei et al. / Bioorg. Med. Chem. 22 (2014) 2033–2044

Figure 4.

13

C NMR of

the exchange of the iodine and hydroxy groups between 60 and 70 carbon. Therefore, we propose 127I-CMICE-013 A is a 60 -iodo-70 -hydroxy derivative of rotenone. Although the NMR spectra of 127I-CMICE-013 B (Figs. S11 and S12) contain some impurity peaks, we were able to assign the product peaks. Both 1H and 13C NMR spectra of 127I-CMICE-013 B are very similar to 127I-CMICE-013 A, indicating they are diastereomers. The only noticeable difference is the 50 -proton peak, which shows at 4.2 ppm for 127I-CMICE-013 A and 5.1 ppm for B. This may be caused by the different stereo orientation of the hydroxy and methyl group at 60 -carbon. 3.7. MicroSPECT imaging studies with

123

I-CMICE-013 A, B, C, D

Figure 5 shows the CT co-registered SPECT images from the summation of the data from the rats acquired at 0 to 60 min; 30–90 min, and 60–120 min post-injection of the four tracers 123 I-CMICE-013 A, B, C, D. The radiolabeling and recovery yields were different for the four compounds, resulting in different injected activities and intensity of the imaging signals. The injected dose of 123I-CMICE-013 B was lower than the other three compounds, due to the lower yield of B caused by two rounds of HPLC purification. All four compounds showed significant uptake in the myocardium, and low lung uptake. The contrast between heart and liver was different for the four tracers, with 123I-CMICE-013 A and D showed higher heart to liver ratio than B and C. For all compounds, the activities in the background regions did not affect the visibility of the heart. The images in Figure 5 show the one hour summed accumulation of the tracers at three intervals, which qualitatively indicate

2039

127

I-CMICE-013 A.

the tissue uptake and the clearance trend during two hours. To further analyze the imaging data, the images were processed with a MATLAB-based software, the decayed corrected counts were calculated and the percentage injected dose per volume of organ (%ID/mL) was estimated for heart and liver. These are the only two organs that can be clearly seen in the field of view of the images. The time activity curves (TAC) were generated for the four tracers as shown in Figure 6. Two-way ANOVA showed that the heart uptake of 123I-CMICE013 B was significantly lower than A, C and D at all time points, while A, C, D were not statistically different (P >0.05) (Fig. 6). The difference between 123I-CMICE-013 B and A is significant with P values less than 0.05. The differences between 123I-CMICE-013 B and C, B and D were more significant at earlier time points than later time points. For example, as shown in Figure 6 (i) for comparison between B and D, the P values were less than 0.0001 from 0 to 67 min, less than 0.001 from 80 to 93 min, less than 0.01 at 107 min, and less than 0.05 at 120 min. The liver uptake of the four tracers was not statistically different, except the uptake between B and C. As shown in Figure 6 (ii), the uptake of B was significantly lower than C (from 13 to 40 min, P <0.0001, P <0.001 at 53 min, P <0.01 at 67 min, and P <0.5 at 80 min). Although the microSPECT images showed different heart to liver contrast, the heart to liver ratios in the time activity curves did not indicate statistically significant differences among the four tracers. 3.8. Biodistribution Studies The biodistribution of 123I-CMICE-013 A (n = 3), B (n = 4), C (n = 4) and D (n = 4) in Sprague Dawley rats at 2 h post injection

2040

L. Wei et al. / Bioorg. Med. Chem. 22 (2014) 2033–2044

(i) 0 – 60 min

(ii) 30 – 90 min

(iii) 60 – 120 min

A 1.34 mCi

B 0.44 mCi

C 1.50 mCi

D 2.09 mCi

Figure 5. SPECT/CT images of 123I-CMICE-013 A, B, C and D in 4 male Sprague Dawley rats (for each rat, in a clockwise direction: maximum intensity projection, coronal, transversal and sagittal views; column (i): 0–60 min sum; column (ii): 30–90 min sum; column (iii): 60–120 min sum).

(p.i.) is shown in Figure 7 and summarized in Table 2. The myocardial uptake of 123I-CMICE-013 B (0.54 ± 0.12 %ID/g) was significantly lower than the other three tracers (vs 1.52 ± 0.38 %ID/g for A, P <0.01; vs 1.39 ± 0.46 %ID/g for C, P <0.05; vs 1.65 ± 0.7 %ID/g for D, P <0.01), while 123I-CMICE-013 A, C, and D show similar heart uptake. This is consistent with the imaging data (see Fig. 6 (i) at 120 min time point). As shown in Figure 7 and Table 2, the intestine uptake of 123 I-CMICE-013 A and B was significantly higher than C and D. while the two pairs of diastereomers (A vs B; C vs D) showed

similar uptake. The liver, blood and kidney uptake of B was relatively lower than C. The accumulation of A and B in the thyroid was significantly lower than D. Another noticeable difference is that the stomach uptake of A was significantly lower than D. Other than the listed statistical difference in Table 2, the tracer concentrations in other organs were not significantly different for the four compounds. The estimated values from microSPECT imaging study were compared with biodistribution results for the heart and liver uptake and heart to liver ratio at 2 h p.i, as shown in Figure 8. A

2041

L. Wei et al. / Bioorg. Med. Chem. 22 (2014) 2033–2044

Figure 6. Comparison of time activity curves for heart (i), liver uptake (ii) and heart to liver ratio (iii) for 123I-CMICE-013 A (n = 3), B (n = 4), C (n = 4) and D (n = 4). Two-way ANOVA: for heart uptake, comparison between B and D; for liver uptake, comparison between B and C. (⁄P <0.05, ⁄⁄P <0.01, ⁄⁄⁄P <0.001, ⁄⁄⁄⁄P <0.0001).

Figure 7. Biodistribution of

123

I-CMICE-013 A (n = 3), B (n = 4), C (n = 4), D (n = 4) and

two-tailed Student t-test indicates that the imaging results matched well with the biodistribution data, with no statistically significant differences. (e.g. the heart to liver ratio of 123I-CMICE013 A: 3.57 ± 1.47 %ID/mL from imaging vs. 6.63 ± 2.30 %ID/g from biodistribution, P >0.1; the heart to liver ratio of B: 1.89 ± 0.91 from imaging versus 2.81 ± 1.45 from biodistribution, P >0.1).

123

I-CMICE-01333 (n = 6) at 2 h p.i.

4. Discussion Our research has been focused on the SPECT MPI tracers based on rotenone. Rotenone, widely used as an insecticide, acaricide, and miticide,37,38 is a natural product and has been demonstrated as a potent inhibitor of MC-1.39,40 We are particularly interested in 123I

2042

L. Wei et al. / Bioorg. Med. Chem. 22 (2014) 2033–2044

Figure 8. Comparison of imaging and biodistribution results for heart, liver uptake and heart to liver ratio for 123I-CMICE-013 A (n = 3), B (n = 4), C (n = 4), D (n = 4). Imaging data is presented in %ID/mL, biodistribution data is presented in %ID/g.

Table 1 NOE coupling (%) measured from NOESY-NMR at 400 ms mixing time 127

127

0.7 1.3 1.0

0.6 0.8 1.0

I-CMICE-013 C

0

0

CH3 (8 )—CH(ring) (5 ) CH3 (80 )—CH2(ring) (40 ) CH3 (80 )—CH2I (70 )

I-CMICE-013 D

labelled MPI agents, taking advantage of the wide availability of SPECT cameras. 123I has a 13.2 hour half life, which allows for cross-country shipment, eliminating the need to invest in costly on-site cyclotrons. The gamma energy of 123I (159 keV) is close to 99m Tc, which is optimal for SPECT imaging. Radiometal labelled pharmaceuticals usually need the conjugation of bulky chelates and in the case of small molecules, this could significantly alter the pharmacokinetic properties of the host compound. The advantage of 123I labelled radiopharmaceuticals is that iodine can be directly attached onto the carbon atom without adding the bulky moiety, resulting in less impact on the pharmacological characteristics. Rotenone has been labelled with 125I and 123I by the route of oxidative destannylation reaction using tributyltin as the precursor, which involved multi-step reactions with low overall yield.16,17,22 Our group has developed a simple, one step procedure that allows the direct labelling of commercial available rotenone with 125I and 123I. Our signature compound 123I-CMICE-013 is a promising SPECT MPI agent with good myocardial uptake, low background activity, and favourable biodistribution properties.33

Our previous study showed that 123I-CMICE-013 contains four species, A, B, C and D, with D as the main product. In this study, we synthesized and isolated the cold analogs of the four compounds 127 I-CMICE-013 A, B, C, and D, characterized them by NMR and mass spectrometry, and proposed their structures. In addition, we isolated the radiolabeled tracers 123I-CMICE-013 A, B, C, and D, and compared their in vivo imaging and biodistribution profile in normal rats. Based on the extensive 1D, 2D NMR and mass spectrometry characterization of 127I-CMICE-013 A, B, C and D, we proposed the structures of the four isomers. As indicated in our previous study, the iodination reaction in trifluoroacetic acid (TFA) with iodogen as the oxidation agent reduces the 60 , 70 -double bond to a single bond.33 In 127I-CMICE-013 A and B, which are diastereomers to each other, iodine is attached at the 60 -carbon position, and a hydroxy group is added onto the 70 -carbon. For the other diastereomer pair 127I-CMICE-013 C and D, the iodine and hydroxy groups are attached on the 70 -carbon and 60 -carbon, respectively. We proposed the mechanism of the rotenone iodination reaction as shown in Scheme 1. (1) NaI is oxidized to I+ by iodogen. (2) I+ forms an intermediate three-membered ring with 60 and 70 -carbon (an iodonium ion). (3) The lone electron pair of the water molecule attacks the 60 or 70 -carbon, following by the formation of 60 -OH, and 70 -I bonds as in major products 127I-CMICE-013 C and D, or 60 -I and 70 -OH bonds as in minor products 127I-CMICE-013 A and B. The weaker 60 -I bond in the intermediate prompts the nucleophilic attachment of water at the favorable 60 position to generate 127 I-CMICE-013 C and D.

L. Wei et al. / Bioorg. Med. Chem. 22 (2014) 2033–2044 Table 2 Biodistribution of

123

2043

I-CMICE-013 A, B, C, and D in Sprague–Dawley rats at 2 h p.i. and one-way ANOVA

Tissue

A (n = 3)

B (n = 4)

C (n = 4)

D (n = 4)

One-way ANOVA

Urine Liver Femur Muscle Spleen Blood Brain Intestine Kidney Heart Lung Stomach Thyroid Testis Heart/Liver Heart/Blood Heart/Lung

0.30 ± 0.19 0.27 ± 0.18 0.13 ± 0.05 0.31 ±0.14 0.09 ± 0.01 0.08 ± 0.01 0.04 ± 0.01 2.45 ± 0.92 0.34 ± 0.06 1.52 ± 0.38 0.33 ± 0.08 0.97 ± 0.68 0.51 ± 0.17 0.06 ± 0.01 6.63 ± 2.30 17.96 ± 2.10 4.96 ± 2.49

0.83 ± 0.62 0.22 ± 0.08 0.12 ± 0.04 0.25 ± 0.10 0.09 ± 0.02 0.10 ± 0.03 0.03 ± 0.00 2.06 ± 0.55 0.24 ± 0.03 0.54 ± 0.12 0.38 ± 0.28 2.27 ± 0.85 0.84 ± 0.29 0.08 ± 0.01 2.81 ± 1.45 5.89 ± 2.30 1.90 ± 1.05

0.62 ± 0.38 0.73 ± 0.34 0.32 ± 0.16 0.49 ± 0.26 0.26 ± 0.15 0.35 ± 0.15 0.06 ± 0.03 0.59 ± 0.26 0.64 ± 0.31 1.39 ± 0.46 0.46 ± 0.18 2.05 ± 0.50 1.08 ± 0.15 0.20 ± 0.11 1.99 ± 0.53 4.11 ± 0.64 3.05 ± 0.45

1.07 ± 0.62 0.49 ± 0.11 0.17 ± 0.07 0.31 ± 0.14 0.16 ± 0.02 0.23 ± 0.06 0.06 ± 0.01 0.45 ± 0.11 0.44 ± 0.07 1.65 ± 0.17 0.33 ± 0.09 2.68 ± 0.62 1.61 ± 0.45 0.13 ± 0.03 3.47 ± 0.65 7.62 ± 2.12 5.20 ± 1.44

NS B versus C: P <0.05; other: NS NS NS NS A versus C: P <0.01; B versus C: P <0.01; Other: NS NS A versus C: P <0.01; A versus D: P <0.01; B versus C: P <0.05; B versus D: P <0.01; Other: NS B versus C: P <0.05; Other: NS A versus B: P <0.01; B versus C: P <0.05; B versus D P <0.01; other: NS NS A versus D: P <0.05; Other: NS A versus D: P <0.01; B versus D: P <0.05; other: NS NS

Sprague–Dawley rats (250–450 g) were injected intravenously with 14.8–92.5 MBq (0.4–2.5 mCi) of 123I-CMICE-013 A, B, C or D. The organ uptake values are reported as percentage injected dose per gram (%ID/g) and presented as mean ± SD. One-way ANOVA was performed with a Bonferroni post hoc test. NS: not significant.

Previously reported 125I-iodorotenone was produced through tin-precursors at 70 -position, resulting in 70 -iodination on a 60 , 70 -double bond.16,22 Two isomers were isolated, E and Z isomers, with the 70 -iodine cis and trans to the 60 -methyl group, respectively. The in vivo biodistribution studies of the two isomers in rats showed significant different heart uptake and heart to background ratios.22 The heart uptake of the E-isomer is approximately two-fold higher than the Z-isomer. The heart to blood ratios of the E isomer is about 7-times higher than Z-isomer, indicating that the E-isomer has higher extraction properties in the heart tissue than the Z-isomer. Ref. 22 mentioned ‘The absolute heart uptake and heart to blood ratios for the E compounds relative to the Z compounds were not expected and could not be predicted from this small structural change.’ Compared to 125I-iodorotenone,16,22 the structure of 123 I-CMICE-013 is quite different due to the different labelling approach and mechanism, resulting in different types of isomers. Four isomers were produced. 123I-CMICE-013 A and B are constitutional isomers of C and D, and they are diastereomers of each other. The microSPECT imaging and biodistribution studies indicate that the biological characteristics of the four isomers are more complicated than expected; the characteristics are different in some aspects, but similar in others. Similar to what was found with the Z and E isomers of 125I-iodorotenone,22 structure activity relationship (SAR) theory does not explain the biological profiles of the four isomers in all cases. Based on SAR, the biological properties of the two pairs of diasteromers (A vs B; C vs D) should be similar to each other, while the constitutional isomers (A/B vs C/D) should be different. Our data demonstrated that different organs/tissues show different results. For instance, the intestine, liver and blood uptake of the two pairs A versus B, C versus D is close to each other, while A and B show different uptake than C and D, which fits the SAR. 123I-CMICE-013 B shows significantly lower heart uptake than A, while the heart uptake of A is similar to C and D. Also the stomach and urine uptake of B is more close to C and D, which can not be explained by SAR. For some organs, such as lung, femur, muscle, spleen, brain, the uptake of all four tracers is not significantly different. We also compared the biodistribution properties of the four pure compounds 123I-CMICE-013 A, B, C, D with the mixture 123 I-CMICE-013 in our previous study,33 as shown in Figure 7. One-way ANOVA analysis indicates no significant differences between 123I-CMICE-013 and C and D. This is expected due to the

fact that in the mixture 123I-CMICE-013, D (74%) and C (14%) are major products (see Figure 1 (ii) in Ref. 33). With regards to the targeted organ in this study, 123I-CMICE013 B stands out from the other three isomers, with the heart uptake approximately three times lower. This indicates that for the application of myocardial perfusion imaging agent, it is beneficial to minimize the amount of 123I-CMICE-013 B in the product. 123 I-CMICE-013 A shows good biodistribution profile, except the high intestine uptake. From overall evaluation of the imaging and biodistribution data, 123I-CMICE-013 D has better properties, although the differences between C and D are not statistically significant. The ideal radiotracer would be 100% 123I-CMICE-013 D, however, a mixture of C and D is acceptable. This is feasible for manufacturing, due to the fact that C and D are major products of the radiolabeling reaction. 5. Conclusion As the next step in the development of a promising SPECT MPI agent 123I-CMICE-013, we isolated, characterized and proposed the structures of the cold analogs of the four species 127I-CMICE-013 A, B, C and D. 127I-CMICE-013 A, B and C, D are two pairs of diastereomers, while A and B are constitutional isomers of C and D. For 127 I-CMICE-013 A and B, the iodination occurs at 60 -carbon, with a hydroxy added to 70 -carbon. For C and D, the positions of iodine and hydroxy attachment switch to 70 -carbon and 60 -carbon, respectively. The four radiolabeled tracers 123I-CMICE-013 A, B, C and D were also isolated and characterized by microSPECT imaging and biodistrubution studies. All four tracers demonstrate good myocardial uptake and little background interference from lung and liver. The biodistribution profiles of the four isomers show both similarities and differences, depending on the organs and tissues, and do not have perfect correlation with their molecular structures. Acknowledgements This project was supported by grant #RMIPJ 389641-09 from Canadian Institute of Health Research (CIHR) and Natural Sciences and Engineering Research Council of Canada (NSERC). Dr. X.Y. was supported through NSERC Industrial R&D Fellowship. Dr. W.G. is supported by the Mitacs Elevate Postdoctoral Fellowship Program. We are very grateful for the technical support from the University

2044

L. Wei et al. / Bioorg. Med. Chem. 22 (2014) 2033–2044

of Ottawa animal care technicians. We sincerely thank Dr. Glenn A. Facey from the University of Ottawa Chemistry Department, who provided the NOE NMR measurement and thorough discussion for the stereochemistry of the disatereomers. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.bmc.2014.02.052. References and notes 1. Foot, D. K.; Lewis, R. P.; Pearson, T. A.; Beller, G. A. J. Am. Coll. Cardiol. 2000, 35, 66B. 2. Beller, G. A.; Bergmann, S. R. J. Nucl. Cardiol. 2004, 11, 71. 3. Clark, A. N.; Beller, G. A. Q. J. Nucl. Med. Mol. Imaging 2005, 49, 43. 4. Russell, R. R., 3rd; Zaret, B. L. Curr. Probl. Cardiol. 2006, 31, 557. 5. Gibbons, R. J. Heart 2000, 83, 355. 6. Kapur, A.; Latus, K. A.; Davies, G.; Dhawan, R. T.; Eastick, S.; Jarritt, P. H.; Roussakis, G.; Young, M. C.; Anagnostopoulos, C.; Bomanji, J.; Costa, D. C.; Pennell, D. J.; Prvulovich, E. M.; Ell, P. J.; Underwood, S. R. Eur. J. Nucl. Med. Mol. Imaging 2002, 29, 1608. 7. Slart, R. H.; Bax, J. J.; van Veldhuisen, D. J.; van der Wall, E. E.; Dierckx, R. A.; Jager, P. L. Int. J. Cardiovasc. Imaging 2006, 22, 63. 8. Glover, D. K.; Ruiz, M.; Edwards, N. C.; Cunnungham, M.; Simanis, J. P.; Smith, W. H.; Watson, D. D.; Beller, G. A. Circulation 1995, 91, 813. 9. Glover, D. K.; Okada, R. D. Circulation 1990, 81, 628. 10. Glover, D. K.; Ruiz, M.; Yang, J. Y.; Smith, W. H.; Watson, D. D.; Beller, G. A. Circulation 1997, 96, 2332. 11. Iida, H.; Hayashi, T.; Eberl, S.; Saji, H. J. Nucl. Med. 2003, 44, 40. 12. Carvalho, P. A.; Chiu, M. L.; Kronauge, J. F.; Kawamura, M.; Jones, A. G.; Holman, B. L.; Piwnica-Worms, D. J. Nucl. Med. 1992, 33, 1516. 13. Arbab, A. S.; Koizumi, K.; Toyama, K.; Arai, T.; Araki, T. J. Nucl. Med. 1998, 39, 266. 14. Piwnica-Worms, D.; Kronauge, J. F.; LeFurgey, A.; Backus, M.; Hockett, D.; Ingram, P.; Liberman, M.; Holman, B. L.; Jones, A. G.; Davison, A. Magn. Reson. Imaging 1994, 12, 641. 15. Piwnica-Worms, D.; Kronauge, J. F.; Delmon, L.; Holman, B. L.; Marsh, J. D.; Jones, A. G. J. Nucl. Med. 1990, 31, 464. 16. VanBrocklin, H. F.; Hanrahan, S. M.; Enas, J. D.; Nandanan, E.; O’Neil, J. P. Nucl. Med. Biol. 2007, 34, 109. 17. Broisat, A.; Ruiz, M.; Goodman, N. C.; Hanrahan, S. M.; Reutter, B. W.; Brennan, K. M.; Janabi, M.; Schaefer, S.; Watson, D. D.; Beller, G. A.; Vanbrocklin, H. F.; Glover, D. K. Circ. Cardiovasc. Imaging 2011, 685. 18. Marshall, R. C.; Powers-Risius, P.; Reutter, B. W.; O’Neil, J. P.; La Belle, M.; Huesman, R. H.; VanBrocklin, H. F. J. Nucl. Med. 2004, 45, 1950.

19. Marshall, R. C.; Powers-Risius, P.; Reutter, B. W.; Taylor, S. E.; VanBrocklin, H. F.; Huesman, R. H.; Budinger, T. F. J. Nucl. Med. 2001, 42, 272. 20. Yalamanchili, P.; Wexler, E.; Hayes, M.; Yu, M.; Bozek, J.; Kagan, M.; Radeke, H. S.; Azure, M.; Purohit, A.; Casebier, D. S.; Robinson, S. P. J. Nucl. Cardiol. 2007, 14, 782. 21. Purohit, A.; Benetti, R.; Hayes, M.; Guaraldi, M.; Kagan, M.; Yalamanchili, P.; Su, F.; Azure, M.; Mistry, M.; Yu, M.; Robinson, S.; Dischino, D. D.; Casebier, D. Bioorg. Med. Chem. Lett. 2007, 17, 4882. 22. O’Neil, J. P.; VanBrocklin, H. F. WO 2007/137135 A2, 2007; PCT Int. Appl. 2007. 23. Yu, M.; Guaraldi, M. T.; Mistry, M.; Kagan, M.; McDonald, J. L.; Drew, K.; Radeke, H.; Azure, M.; Purohit, A.; Casebier, D. S.; Robinson, S. P. J. Nucl. Cardiol. 2007, 14, 789. 24. Yu, M.; Guaraldi, M.; Kagan, M.; Mistry, M.; McDonald, J.; Bozek, J.; Yalamanchili, P.; Hayes, M.; Azure, M.; Purohit, A.; Radeke, H.; Casebier, D. S.; Robinson, S. P. Eur. J. Nucl. Med. Mol. Imaging 2009, 36, 63. 25. Higuchi, T.; Nekolla, S. G.; Huisman, M. M.; Reder, S.; Poethko, T.; Yu, M.; Wester, H. J.; Casebier, D. S.; Robinson, S. P.; Botnar, R. M.; Schwaiger, M. J. Nucl. Med. 2008, 49, 1715. 26. Maddahi, J.; Czernin, J.; Lazewatsky, J.; Huang, S.-C.; Dahlbom, M.; Schelbert, H.; Sparks, R.; Ehlgen, A.; Crane, P.; Zhu, Q.; Devine, M.; Phelps, M. J. Nucl. Med. 2011, 52, 1490. 27. Huisman, M. C.; Higuchi, T.; Reder, S.; Nekolla, S. G.; Poethko, T.; Wester, H. J.; Ziegler, S. I.; Casebier, D. S.; Robinson, S. P.; Schwaiger, M. J. Nucl. Med. 2008, 49, 630. 28. Mou, T.; Jing, H.; Yang, W.; Fang, W.; Peng, C.; Guo, F.; Zhang, X.; Pang, Y.; Ma, Y. Bioorg. Med. Chem. 2010, 18, 1312. 29. Mou, T.; Zhao, Z.; Fang, W.; Peng, C.; Guo, F.; Liu, B.; Ma, Y.; Zhang, X. J. Nucl. Med. 2012, 53, 472. 30. Purohit, A.; Radeke, H.; Azure, M.; Hanson, K.; Benetti, R.; Su, F.; Yalamanchili, P.; Yu, M.; Hayes, M.; Guaraldi, M.; Kagan, M.; Robinson, S.; Casebier, D. J. Med. Chem. 2008, 51, 2954. [31]. Kronauge, J. F.; Chiu, M. L.; Cone, J. S.; Davison, A.; Holman, B. L.; Jones, A. G.; Piwnica-Worms, D. Nucl. Med. Biol. (Int. J. Radiat. Appl. Instrum. Part B) 1992, 19, 141. 32. Babich, J. W.; Maresca, K. P. WO 03/086476 A1, 2003; PCT Int. Appl. 2003. 33. Wei, L.; Bensimon, C.; Lockwood, J.; Yan, X.; Fernando, P.; Wells, G. R.; Duan, Y.; Chen, Y. X.; Redshaw, R.; Covitz, P. A.; Ruddy, T. D. Bioorg. Med. Chem. 2013, 21, 2903. 34. Blasko, G.; Shieh, H. L.; Pezzuto, J. M.; Cordell, G. A. J. Nat. Prod. 1989, 52, 1363. 35. Fang, N.; Casida, J. E. J. Agric. Food Chem. 1999, 47, 2130. 36. Coll, J. J. Agric. Food Chem. 2005, 53, 3749. 37. Negherbon, W. O. Insecticides In Handbook of Toxicology; W.B. Saunders: Philadephia, 1959; Vol. III, p 661. 38. Tomlin, C. D. S. Rotenone. In The Pesticide Manual, 11th ed.; British Crop Protection Council: Farnham, UK, 1997; p 1097. 39. Walker, J. E. Q. Rev. Biophys. 1992, 25, 253. 40. Carroll, J.; Fearnley, I. M.; Shannon, R. J.; Hirst, J.; Walker, J. E. Mol. Cell. Proteomics 2003, 2, 117.