Special Issue Research Article

Received: 18 December 2010, Revised: 2 February 2011, Accepted: 22 February 2011, Published online in Wiley Online Library: 28 April 2011

(wileyonlinelibrary.com) DOI: 10.1002/nbm.1717

Parahydrogen‐induced polarization (PHIP)hyperpolarized MR receptor imaging in vivo: apilot study of 13C imaging of atheroma in micePratip Bhattacharyaa*, Eduard Y. Chekmeneva,b, Wanda F. Reynoldsc,Shawn Wagnera, Niki Zachariasa, Henry R. Chana, Rolf Büngerd andBrian D. Rossa

MR techniques using hyperpolarized 13C have succ

NMR Biom

essfully produced examples of angiography and intermediarymetabolic imaging, but, to date, no receptor imaging has been attempted. The goal of this study was to synthesizeand evaluate a novel hyperpolarizable molecule, 2,2,3,3‐tetrafluoropropyl 1‐13C‐propionate‐d2,3,3 (TFPP), for thedetection of atheromatous plaques in vivo. TFPP binds to lipid bilayers and its use in hyperpolarized MR could prove tobe a major step towards receptor imaging. The precursor, 2,2,3,3‐tetrafluoropropyl 1‐13C‐acrylate‐d2,3,3 (TFPA), binds to1,2‐dimyristoylphosphatidylcholine lipid bilayers with a 1.6‐ppm chemical shift in the 19F MR spectrum. This moleculewas designed to be hyperpolarized through the addition of parahydrogen to the 13C‐acrylate moiety by parahydrogen‐induced polarization. TFPA was hyperpolarized to TFPP to an extent similar to that of the hydroxyethylacrylate tohydroxyethylpropionate transition: 17 ± 4% for TFPP versus 20% for hydroxyethylpropionate; T1 relaxationtimes (45±2 s versus 55±2 s) were comparable and the hyperpolarized properties of TFPP were characterized.Hydroxyethylacrylate, like TFPA, has a chemical structurewith an acrylatemoiety, but does not contain the lipid‐bindingtetrafluoropropyl functional group. Hyperpolarized TFPP binds to the lipid bilayer, appearing as a second, chemicallyshifted 13C hyperpolarizedMR signal with a further reduction in the longitudinal relaxation time (T1 = 21±1 s). In aortasharvested from low‐density lipoprotein receptor knock‐out mice fed with a high‐fat diet for 9months, and in whichatheroma is deposited in the aorta and heart, TFPP showedgreater binding to lipid on the intimal surface than in controlmice fed a normal diet. When TFPP was hyperpolarized and administered in vivo to atheromatous mice in a pilot study,increased binding was observed on the endocardial surface of the intact heart compared with normally fed controls.Hyperpolarized TFPP has bio‐sensing specificity for lipid, coupled with a 42 000‐fold sensitivity gain in the MR signal at4.7T. Binding of TFPP with lipids results in the formation of a characteristic second peak in MRS. TFPP therefore has thepotential to act as an in vivomolecular probe for atheromatous plaque imaging and may serve as a model of receptor‐targeted bio‐imaging with enhanced MR sensitivity. Copyright © 2011 John Wiley & Sons, Ltd.

Keywords: 13C; heart; receptor imaging; atheroma; hyperpolarization; TFPP; PHIP; MR

* Correspondence to: P. Bhattacharya, Enhanced Magnetic ResonanceLaboratory, Huntington Medical Research Institutes, 10 Pico Street, Pasadena,CA 91105, USA.E-mail: pratip@hmri.org

a P. Bhattacharya, E. Y. Chekmenev, S. Wagner, N. Zacharias, H. R. Chan, B. D. RossEnhanced Magnetic Resonance Laboratory, Huntington Medical ResearchInstitutes, Pasadena, CA, USA

b E. Y. ChekmenevRudi Schulte Research Institute, Santa Barbara, CA, USA

c W. F. ReynoldsSanford Burnham Medical Research Institute, La Jolla, CA, USA

d R. BüngerUniformed Services University of the Health Sciences, Bethesda, MD, USA

Abbreviations used: DMPC, 1,2‐dimyristoylphosphatidylcholine; DNP, dynamicnuclear polarization; HEP, 2‐hydroxyethyl 1‐13C‐propionate‐d2,3,3; LDLR, low‐density lipoprotein receptor; PHIP, parahydrogen‐induced polarization; TFPA,2,2,3,3‐tetrafluoropropyl 1‐13C‐acrylate‐d2,3,3; TFPP, 2,2,3,3‐tetrafluoropropyl 1‐

13C‐propionate‐d2,3,3.

102

INTRODUCTION

MRI can be used to measure noninvasively cardiac energetics andhas had an important impact on the diagnosis and understandingof cardiac function and disorders (1). The early promise ofMRS, as ameans of defining normal and disordered cardiac metabolism, hasnot been fully realized. The current practical problems of lowsensitivity and high costs of traditional MRI have hindered large‐scale applications of this technology in both basic and clinicalsciences. The MR signal intensity is a function of the product ofthe number of spins and the fractional nuclear spin polarization;the latter is only 10–6–10–5 for typical high‐field equilibrium initialconditions. This severely limits the utility of MRI and MRS inbiomedical diagnosis and research, especially for nuclei such as 13Cand 15N. Conventional MR is simply not sensitive enough –polarization is only a few parts per million, which translates into arelatively small number of nuclear ‘spins’ available for imagereconstruction. The difference in the spin population is propor-tional to the magnetic field strength, so that 3 T ‘buys’ twice thesignal available at 1.5 T, and so on. Conventional MR is also too

ed. 2011; 24: 1023–1028 Copyright © 2011 John Wiley & Sons, Ltd.

3

Figure 1. (a) Molecular cis addition of parahydrogen to 2,2,3,3‐tetrafluoropropyl 1‐13C‐acrylate‐d2,3,3 (TFPA) to produce 2,2,3,3‐tetrafluoropropyl 1‐13C‐propionate‐d2,3,3 (TFPP). The catalytic reaction wascarried out at 62 °C, but the hyperpolarized solution had cooled to 35–40 °Cby the time it was removed from the polarizer. (b) Quantification ofparahydrogen‐induced polarization (PHIP) 13C signal enhancement by 13Cspectroscopy of hyperpolarized 3.2mM TFPP at 4.7 T in vitro. Polarization of17%was achieved in this representative example. The inset spectrum showsa reference signal obtained from the natural abundance 4.5M ethanol; 13Cconcentration = 50mM per carbon site.

P. BHATTACHARYA ET AL.

1024

slow – an examination may take 30min or more. At relatively lowcost, hyperpolarization has the potential to solve many of theseproblems by amplifying the MR signal 10 000–100 000‐fold. Withthis sensitivity gain, hyperpolarization may enable in vivonanochemistry and metabolomics, decrease the time needed toobserve a metabolite concentration change, better extract thesignal of a target molecule against a background of low‐spinspecies in a mixture or in vivo (i.e. the 13C background is only 1.1%,and water protons decrease to zero in a hyperpolarized 13Cimaging study) and permit studies to be performed more rapidlyand at considerably lower cost.

Dynamic nuclear polarization (DNP) (2) and parahydrogen‐induced polarization (PHIP) (3,4) allow nuclear spin polarization toreach the order of unity, where nearly all nuclear magneticmoments contribute to the MR signal. These techniques constitutethe emerging field of hyperpolarization which increases the signal‐to‐noise ratio by a factor of 105 at reasonable costs. This has beensuccessfully demonstrated for 13C‐labeled compounds with spin–lattice relaxation times T1 of tens of seconds with demonstratedsensitivity enhancement factors of 104–105 for both DNP (5–9) andPHIP (10–12). These hyperpolarized 13C MR agents allow fastin vivo 13C imaging and spectroscopy. To date, a few biologicallyrelevant agents have been explored, including 1‐13C‐pyruvic acid(5–8) and bicarbonate (9) by DNP and 1‐13C‐succinic acid (10,11)by PHIP. These molecules were primarily designed to targetenzymatic biochemical processes at the cellular and tissue levelsby providing an instantaneous snapshot of the uptake andmetabolism of the 13C tracer. In addition, a small number ofangiographic reagents (12,13) have been designed and testedin vivo to provide real‐time coronary angiography, catheterpositioning and myocardial perfusion.

In this article, we present a novel class of 13C hyperpolarized‘receptor imaging’ molecular agents which have both enhancedMR sensitivity and a moiety capable of targeted bio‐sensing. Thisis exemplified by the –CH2–CF2–CF2H fluorocarbon group whichhas a lipophilic propensity in aqueous medium. This moietyfacilitates chemical exchange into lipids (14), shown here tomanifest as a unique 13C hyperpolarized spectroscopic signa-ture. The work described here is the first step towards the designof targeted enhanced MR contrast agents and an initial proof ofconcept in vivo. In particular, the lipophilic functionalitypresented here may potentially allow for the targeting of lipid‐rich atherosclerotic plaques (15–17) in vivo.

METHODS

Hyperpolarization

Parahydrogen addition and the transfer of spin order to 13C hasbeen described previously (18,19). The design and selection of thereagent have been described elsewhere (13,14). This molecularagent has two important moieties: the lipophilic fluorocarbonmotif –CH2–CF2–CF2H and the C=C–13C functionality for 13C PHIPhyperpolarization. A custom‐built PHIP polarizer (HuntingtonMedical Research Institutes, Pasadena, CA, USA) (20,21) wasemployed to hydrogenate the double bond of the acrylate moietyin 2,2,3,3‐tetrafluoropropyl 1‐13C‐acrylate‐d2,3,3 (TFPA) to yield2,2,3,3‐tetrafluoropropyl 1‐13C‐propionate‐d2,3,3 (TFPP) (Fig. 1a).The spin order was transferred from 1H nuclei to 13C at 1.76mTin the reactor by the heteronuclear pulse sequence describedby Goldman and Johannesson (19) using the hetero‐ andhomonuclear J coupling of 2‐hydroxyethyl 1‐13C‐propionate‐d2,3,3

wileyonlinelibrary.com/journal/nbm Copyright © 2011 Jo

(HEP). The resulting hyperpolarized TFPP product dissolved inaqueous ethanol (30%) was delivered to the 4.7‐T MR scanner(Bruker Avance, Bruker AG, Karlsruhe, Germany) and detected25–40 s following production, using a 1H/13C MR solenoid coil and1H/13C full‐body mouse volume coil (Doty Scientific, Columbia,SC, USA) (20).To quantify the degree of hyperpolarization, we employed the

reference of a single‐scan spectrum of thermally polarizednatural abundance ethanol at 4.7 T using the following formula:

%Pt¼detection ¼ ½reference�½polarized� �

signalðpolarizedÞsignalðreferenceÞ �

100%246 600

[1]

where 1/246 600 corresponds to the 13C nuclear polarization at298 K and 4.7 T, according to the Boltzmann distribution. Thedegree of hyperpolarization produced in the PHIP polarizer attime zero was back‐calculated using the delivery time and spin–lattice relaxation time T1 of the hyperpolarized agent as follows:

%Pt¼0 ¼ %Pt¼detection � expdelivery time

T1

� �[2]

The reported percentage hyperpolarization refers to %Pt = 0.

In vivo experiments in mice

The in vivo part of this study involved nonsurvival, intracardiacinjections of hyperpolarized TFPP into C57BL/6 mice. Three low‐density lipoprotein receptor (LDLR)‐deficient mice (JacksonLaboratories, Bar Harbor, ME, USA) were fed a high‐fat, Western‐type diet (Harlan Teklad, Placentia, CA, USA) for 9months. They

hn Wiley & Sons, Ltd. NMR Biomed. 2011; 24: 1023–1028

TOWARDS IMAGING PLAQUE IN VIVO USING 13C HYPERPOLARIZATION

reached 30–35g, whereas four normal control mice weighed only20–25g after 9months on a normal diet. Necropsy revealed thatthe LDLR‐deficientmice fed a high‐fat diet acquired atheroscleroticlesions in the aorta and aortic valve. For each scan, mice wereanesthetized with 1.5% isoflurane gas, 0.8 L/min oxygen perfacemask After being placed in a heated volume coil, the 4.7‐TBrukerMR scannerwas shimmed. Themousewas then taken out ofthe scanner for laparotomy, which allowed for a transdiaphrag-matic intracardiac injection of 1mL of 11–42mM TFPP under directvisualization using a 28‐gauge needle. Themousewas then quicklyre‐inserted into the scanner, with care, to maintain the shimgeometry. The time from injection to the start of the scan averaged15 s. After the scan, the mouse was immediately dissected toharvest the portion of the aorta that spanned the root to thediaphragm. All animal experiments were approved by theInstitutional Animal Care and Use Committees of HuntingtonMedical Research Institutes and the Burnham Institute. The 13Cnonlocalized spectra were acquired either as a single spectrumwith a large‐angle excitation pulse or multiple spectra (up to 512scans as shown in Figs 3g, h and 4, see Results and discussionsection) with a small‐angle excitation pulse.

Figure 2. 13C hyperpolarized spectra of 2,2,3,3‐tetrafluoropropyl 1‐13C‐propionate‐d2,3,3 (TFPP) in aqueous solutions acquired with a 10° excitationpulse at 35–40 °C: (a) 6mM TFPP, 4.5M ethanol (EtOH); (b) 2mM TFPP, 1.5M

EtOH, 20mM 1,2‐dimyristoylphosphatidylcholine (DMPC); (c) 2mM TFPP,1.8M EtOH in 60%porcine blood. 13C T1 value shown for each resonancewasmeasuredby recording the 13C hyperpolarized signal decaywith small‐angleexcitation pulses.

RESULTS AND DISCUSSION

Hyperpolarization of TFPP

The design and selection of the marker have been describedelsewhere (13,14). This new molecular agent contains two impor-tant moieties: the lipophilic fluorocarbon motif –CH2–CF2–CF2H(15,16) and theC=C–13C functionality for 13C PHIP hyperpolarization(13,14). Figure 1b shows the 13C spectrum of hyperpolarized TFPP: asingle carbonyl resonance at 177ppm. Analysis of the peak areas ofthe 13C spectrum in Fig. 1b yields a percentage of polarization attime zero of 17±4% for TFPP, which corresponds to a 42 000±10000‐fold enhancement of the TFPP 13C MR signal at 4.7 T.

102

Characterization of hyperpolarization

The percentage of hyperpolarization in TFPP is similar to thepolarization of approximately 20% obtained previously with HEP,which contains a similar PHIP hyperpolarization moiety (14). Thehyperpolarized signal of TFPP, like HEP, is not dependent on pH,temperature or osmolarity. In addition to a high level ofhyperpolarization, the lipophilic PHIP 13C reagent, TFPP, retainsthe important MR property of a long spin–lattice relaxation time(T1) on substitution of the 2‐hydroxyethyl group in HEP by the2,2,3,3‐tetrafluoropropyl group. A spin–lattice relaxation time of45±2 s was measured using hyperpolarized material at 4.7 T with7.5° excitation pulses. This is similar to the value of T1 = 55±2 s foraqueous HEP under identical conditions. When hyperpolarizedTFPP is mixed with synthetic 1,2‐dimyristoylphosphatidylcholine(DMPC) bilayers (Avanti Polar Lipids, Inc., Alabaster, AL, USA), anadditional 13C resonance at 174ppm is detected (Fig. 2b). Thissignificantly broader (full width at half‐maximum, 2.5 ppm) andchemically shifted (by 3ppm) resonance is interpreted as thefraction of TFPP signal bound to the DMPC membranes. A similarshift was observed for TFPA detected by 19F spectroscopy (14).Mixing hyperpolarized TFPP with DMPC also reduced the 13C spin–lattice relaxation time of hyperpolarized TFPP from 45 s toapproximately 20 s (Fig. 2b). As TFPP is a lipophilic reagent andmay be useful for distinguishing vulnerable plaques in vivo bysensing chemical exchange with lipids, it was also tested for

NMR Biomed. 2011; 24: 1023–1028 Copyright © 2011 John W

interactions with porcine blood (Sierra for Medical Science,Whittier, CA, USA) (Fig. 2c). Although 13C T1 is reduced to 15 s(Fig. 2c), there is no additional 13C resonance, as observed for TFPPinteraction with lipid membranes (Fig. 2c). This validates thehypothesis that the second TFPP 13C NMR signal at approximately174ppm is a signature of the interaction with lipids, but notwith blood.

Preliminary in vivo and in vitro studies

To study the feasibility of the model of in vivo lipid sensing withMR hyperpolarized reagents, we examined genetically modifiedmice with a deficiency in LDLR. These mice develop massiveatheroma deposits in the aorta and tricuspid and mitral valveswhen placed on a 9‐month, high‐fat Western diet (17,22). Wild‐type mice with normal LDLR do not develop atheroma underthese conditions. Specifically, we studied the interaction of TFPPwith aortas (ex vivo) and heart (in vivo) of LDLR‐deficient andcontrol mice. First, the animals were injected with 1mL (based onsurvival, mice tolerate intravenous loading with up to 1mL ofhyperpolarized reagent; H. Chan, N. Zacharias, unpublished work)of 11–42mM solution of hyperpolarized TFPP into the heart, and aseries of 13C spectra (Fig. 3g, h) was recorded 15± 5 s afterinjection. Representative in vivo 13C nonlocalized hyperpolarizedNMR spectra are shown in Fig. 3g (LDLR‐deficient mice) andFig. 3h (control mice). A representative in vivo time course ofsignal decay after intracardiac injection of 13C hyperpolarized

iley & Sons, Ltd. wileyonlinelibrary.com/journal/nbm

5

P. BHATTACHARYA ET AL.

1026

TFPP into LDLR‐deficient mice is shown in Fig. 4. In two of sevenexperiments, one control and one LDLR‐deficient mouse, fast 13Ctwo‐dimensional images (Fig. 3f) were acquired prior tothe spectroscopic acquisition of hyperpolarized TFPP in vivo.The co‐registration of 13C images with proton images revealedthat the 13C NMR signal was confined to the heart region in bothcases. This is consistent with the view that the detected 13C NMRspectra also originated from the heart (Fig. 3g, h). The ratio of thehyperpolarized NMR signal from TFPP molecules bound to lipidsto that in solution was markedly higher in LDLR‐deficientmice than in controls. This result is interesting given thefact that LDLR‐deficient mice are 50% larger, on average, thancontrol mice and the same absolute dosage was employed in thein vivo trials. This demonstrates a positive correlation of the

wileyonlinelibrary.com/journal/nbm Copyright © 2011 Jo

quantity of lipid in the aorta and heart with the experimentalhyperpolarized signal ratio of (lipid TFPP) : (total TFPP). We didnot obtain sufficient data to explain the dramatic loss of(lipid TFPP) : (total TFPP) at a high dosage of TFPP, but one canspeculate that the toxicity of the molecule may play a role. Adetailed study of the in vivo toxicity of TFPP is in progress in ourlaboratory. Moreover, nonlocalized proton spectra acquiredover the whole body of the two types of mice confirmed thepresence of an elevated lipid content in LDLR‐deficient micecompared with control mice (Fig. 3c, d). The in vivo 13C T1 value ofhyperpolarized TFPP varied between 14 and 22 s, which issufficiently long for imaging and spectroscopy experiments, asthe hyperpolarized signal persists up to 5 × T1 (i.e. approximately70–110 s for TFPP). The persistence of the hyperpolarized signalin vivo is also evident in the time course studies (Fig. 4).Following the in vivo experiments, the animal aortas were

harvested and washed with ice‐cold, 0.9% saline solution toremove blood from the endothelial surfaces of the vessel walls. Asmall piece of aorta (inset in Fig. 3e) from the animal was packed ina 4‐mm solid‐state NMR rotor and mixed with the same TFPPsolution as used for the in vivo experiments described above. Weutilized 19F solid‐state magic angle spinning NMR spectroscopy tomonitor TFPP lipid binding directly within the aorta. This wasmarkedly higher in LDLR‐deficient mice (Fig. 3a) than in controlmice (Fig. 3b). Thus, it seemed justified to rationalize that TFPP hassufficient chemical affinity to structural atheromatous lipids, andcan therefore undergo chemical exchange with the arterial wall.Binding with any blood constituents was minimal, as thecharacteristic chemical shift signature from the saline incubationswas also present in our 19F assays when the blood was removedfrom the aorta. If this rationale is valid, our data show that 19F TFPPMR is capable of sensing elevated lipid content in arterial andheart vessel walls in mice ex vivo (Fig. 3a, b).An additional NMR signal at 186ppm was observed in some

in vivo 13C hyperpolarized spectra with sufficiently high signal‐to‐noise ratio in both LDLR and control animals. There were alsotwo NMR signals (177 and 174ppm) corresponding to free and

Figure 3. 19F ex vivo NMR spectra of low‐density lipoprotein receptor(LDLR)‐deficient (a) and control (b) mouse aorta, demonstrating lipidbinding of 2,2,3,3‐tetrafluoropropyl 1‐13C‐propionate‐d2,3,3 (TFPP) from 30%ethanol aqueous solution. Spectra were recorded using magic anglespinning at 6–10 kHz without proton decoupling at 11.7 T. In vivononlocalized 1H spectra acquired in LDLR‐deficient (c) and control (d) miceconfirm excess lipid in LDLR‐deficientmice. (e) Photograph of the harvestedaorta. (f) Proton image [rapid acquisition with relaxation enhancement(RARE) factor, 4; two averages; in‐plane resolution, 0.3mm×0.3mm; 9‐minacquisition provides the anatomical location in which the proton spectrawere acquired]. In vivo 13C parahydrogen‐induced polarization (PHIP)hyperpolarized nonlocalizedNMR spectra acquired after a 1‐mL intracardiacinjection of 13mM TFPP in LDLR‐deficient (animal 2) (g) and control (animal4) (h) mice; 512 spectra (0.178 s each) were acquired with a 10° excitationpulse for each animal during a 76‐s acquisition time course. The average of64 individual scans acquired in 9.5 s is shown in (g) and (h). The detailedtime course is provided in Fig. 4. In a separate experiment, a fast 13C image[two‐dimensional fast imaging with steady precession (FISP) sequence;resolution, 3mm×3mm; one average; flip angle, 30°; acquisition, 0.12 s],shown as a red overlay over the proton image in (f), was acquired prior tohyperpolarized 13C spectroscopy acquisition. 13C and 1H in vivo spectra wereacquired using a 1H/13C double‐tuned volume coil at 4.7 T. (i) Distribution ofthe in vivo hyperpolarized 13C signal [(hyperpolarized lipid TFPP):(hyperpolarized total TFPP)] in LDLR‐deficient (blue) and control (red) micefor two different concentrations of injected hyperpolarized substrate.

hn Wiley & Sons, Ltd. NMR Biomed. 2011; 24: 1023–1028

Figure 4. In vivo time course of 13C hyperpolarized 2,2,3,3‐tetrafluoropropyl 1‐13C‐propionate‐d2,3,3 (TFPP) signal decay after intracardiac injection of 1mLof13mM TFPP in a low‐density lipoprotein receptor (LDLR)‐deficientmouse corresponding to Fig. 3g. Each spectrum (temporal resolution, 1.78 s) is an averageof 12 individual scans acquired using a 1H/13C double‐resonance volume coil (Doty Scientific) at 4.7 T. Acquisition parameters: excitation pulse, 10°; spectralwidth, 4000Hz; acquisition time, 128ms; repetition time, 148ms. MR scans 1 (first) and 300 (last) were recorded 15 and 60 s after injection, respectively.

TOWARDS IMAGING PLAQUE IN VIVO USING 13C HYPERPOLARIZATION

102

lipid‐bound TFPP. Furthermore, the solution 13C TFPPNMR signal at177ppm and 19F TFPP solution NMR signal (Fig. 3) exhibitedsplitting into two lines or the appearance of shoulders. Althoughthese additional resonances and shoulders were not identifiedin this proof‐of‐principle study, we hypothesize that there arepossibly more than two (free and lipid‐bound) exchange compart-ments. These could include chemical exchange with additionaldistribution spaces or compartments, such as endothelial cellmembranes, and/or sequestration into intercellular clefts withdesmosomal lipoprotein structures.It is also plausible that a small fraction of hyperpolarized TFPP

could interact with enzymes of the plasma or red and/or whiteblood cells. For example, the ester group in TFPP could behydrolyzed by the ubiquitous esterases in blood. This couldexplain the drastically different chemical shift of the resonanceat 186 ppm, which is within the range of chemical shifts ofcarboxylic acids in deprotonated form. We expect to assignthese additional minor resonances and shoulders in the future.

Limitations

Additional in vivo experimental work is necessary to determine theefficacy and specificity of hyperpolarized agents, such as TFPP, forthe sensing of lipids in atheromas and to make a comparison withthe published molecular imaging data on LDLR mouse models(23). The experiments described here were performed underconditions in which binding is not in equilibrium with the freetracer. These are all nonsteady‐state, transitional data. It remains tobe seen whether nonsteady‐state analysis, combined withappropriate corrections for T1 changes during the time of exposure,will enable the determination of the number of compartmentsthat contribute to the overall signal under the constraints of in vivohyperpolarization conditions. We are currently attempting tomodel the nonsteady‐state dynamics of hyperpolarized TFPP inin vivo situations. Furthermore, the experiments presented hererequired TFPP to be dissolved in 30% ethanol to achieve adequatedelivery to the animal. This is a problem, as almost any lipophilic

NMR Biomed. 2011; 24: 1023–1028 Copyright © 2011 John W

molecule could be potentially solubilized in ethanol and, wheninjected directly into the heart, might be expected to localize there.It is not clear whether ethanol will inactivate or even precipitateplasma proteins/enzymes, or possibly change the permeabilitycharacteristics and kinetics of cellular membranes, both within theatheroma and in the surrounding intact endothelial layers. Thisshortcoming may be overcome by developing a deliverymechanism devoid of ethanol solvent using, perhaps, albumin oranother common drug delivery system. Work is in progress in ourlaboratory to design, synthesize and test a family of other water‐soluble molecules that target lipid‐rich atheromas. Nanoparticularliposomal formulations are also under consideration.

CONCLUSIONS

We have demonstrated a novel strategy for the sensing of thelipid content on arterial walls in vivo employing 13C hyperpolarizedMR. This general strategy may represent a powerful method totarget biological receptors forMRI andMRS. As this novel use of 13Chyperpolarization satisfactorily generated an MR signal in vivo, itmay be possible to design hyperpolarized molecules with diversebinding properties, such as those currently employed in positronemission tomography.

Acknowledgements

We thank the following for funding: Tobacco Related DiseaseResearch Program (16KT‐0044 and 16RT‐0184), National Insti-tutes of Health/Nanotechnology Characterization Laboratory(NIH/NCI) (R01 CA 122513, 1R21 CA118509, 1K99CA134749‐01),Rudi Schulte Research Institute, James G. Boswell Fellowship,American Heart Association, American Brain Tumor Associationand Prevent Cancer Foundation. We thank Professors Daniel P.Weitekamp and Peter B. Dervan and Drs Valerie A. Norton, KeikoKanamori, Jan Hovener, Napapon Sailasuta, Kent Harris andSiu‐Kei Chow for useful discussions.

iley & Sons, Ltd. wileyonlinelibrary.com/journal/nbm

7

P. BHATTACHARYA ET AL.

1028

REFERENCES1. Hove MT, Neubauer S. Evaluating metabolic changes in heart disease

by magnetic resonance spectroscopy. Heart Metab. 2006; 32: 18–21.2. Abragam A, Goldman M. Principles of dynamic nuclear‐polarization.

Rep. Prog. Phys. 1978; 41: 395–467.3. Bowers CR, Weitekamp DP. Transformation of symmetrization order

to nuclear‐spin magnetization by chemical‐reaction and nuclear‐magnetic‐resonance. Phys. Rev. Lett. 1986; 57: 2645–2648.

4. Bowers CR, Weitekamp DP. Para‐hydrogen and synthesis allowdramatically enhanced nuclear alignment. J. Am. Chem. Soc. 1987;109: 5541–5542.

5. Ardenkjaer‐Larsen JH, Fridlund B, Gram A, Hansson G, Hansson L,Lerche MH, Servin R, Thaning M, Golman K. Increase in signal‐to‐noise ratio of >10,000 times in liquid‐state NMR. Proc. Natl. Acad. Sci.U.S.A. 2003; 100: 10 158–10 163.

6. Golman K, in’t Zandt R, Thaning M. Real‐time metabolic imaging.Proc. Natl. Acad. Sci. USA, 2006; 103: 11 270–11 275.

7. Golman K, in’t Zandt R, Lerche M, Pehrson R, Ardenkjaer‐Larsen JH.Metabolic imaging by hyperpolarized 13C magnetic resonance imagingfor in vivo tumor diagnosis. Cancer Res. 2006; 66: 10 855–10 860.

8. Day SE, Kettunen MI, Gallagher FA, Hu DE, Lerche M, Wolber J, GolmanK, Ardenkjaer‐Larsen JH, Brindle KM. Detecting tumor response totreatment using hyperpolarized (13)C magnetic resonance imagingand spectroscopy. Nat. Med. 2007; 13: 1382–1387.

9. Gallagher FA, Kettunen MI, Day SE, Hu DE, Ardenkjaer‐Larsen JH, in’tZandt R, Jensen PR, Karlsson M, Golman K, Lerche MH, Brindle KM.Magnetic resonance imaging of pH in vivo using hyperpolarized 13C‐labelled bicarbonate. Nature 2008; 453: 940–943.

10. Bhattacharya P, Chekmenev EY, Perman WH, Harris KC, Lin AP,Norton VA, Tan CT, Ross BD, Weitekamp DP. Towards hyperpolarized(13)C‐succinate imaging of brain cancer. J. Magn. Reson. 2007; 186:150–155.

11. Chekmenev EY, Hovener J, Norton VA, Harris K, Batchelder LS,Bhattacharya P, Ross BD, Weitekamp DP. PASADENA hyperpolariza-tion of succinic acid for MRI and NMR spectroscopy. J. Am. Chem.Soc. 2008; 130: 4212–4213.

12. Bhattacharya P, Harris K, Lin AP, Mansson M, Norton VA, Perman WH,Weitekamp DP, Ross BD. Ultra‐fast three dimensional imaging of

wileyonlinelibrary.com/journal/nbm Copyright © 2011 Jo

hyperpolarized C‐13 in vivo. Magn. Reson. Mater. Phys. Biol. Med.2005; 18: 245–256.

13. Bhattacharya P, Ross BD, Bunger R. Cardiovascular applications ofhyperpolarized contrast media and metabolic tracers. Exp. Biol. Med.2009; 234: 1395–1416.

14. Chekmenev EY, Chow SK, Tofan D, Weitekamp DP, Ross BD,Bhattacharya P. Fluorine‐19 NMR chemical shift probes molecularbinding to lipid membranes. J. Phys. Chem. B. 2008; 112: 6285–6287.

15. Choudhury RP, Fuster V, Fayad ZA. Molecular, cellular and functionalimaging of atherothrombosis. Nat. Rev. Drug Discov. 2004; 3: 913.

16. Lipinski MJ, Amirbekian V, Frias JC, Aguinaldo JGS, Mani V,Briley‐Saebo KC, Fuster V, Fallon JT, Fisher EA, Fayad ZA. MRI todetect atherosclerosis with gadolinium‐containing immunomicellestargeting the macrophage scavenger receptor. Magn. Reson. Med.2006; 56: 601–610.

17. Moore KJ, Freeman MW. Scavenger receptors in atherosclerosis beyondlipid uptake. Arterioscler. Thromb. Vasc. Biol. 2006; 26: 1702–1711.

18. Johannesson H, Axelsson O, Karlsson M. Transfer of para‐hydrogenspin order into polarization by diabatic field cycling. C. R. Physique,2004; 5: 315–324.

19. Goldman M, Johannesson H. Conversion of a proton pair para orderinto 13C polarization by rf irradiation, for use in MRI. C. R. Physique,2005; 6: 575–581.

20. Hövener JB, Chekmenev E, Harris K, Perman W, Robertson L, Ross BD,Bhattacharya P. PASADENA hyperpolarization of 13C biomolecules:equipmentdesign and installation.Magn. Reson.Mater. Phys. Biol. Med.2009; 22: 111–121.

21. Hövener JB, Chekmenev E, Harris K, Perman W, Tran T, Ross BD,Bhattacharya P. Quality assurance of PASADENA hyperpolarizationfor 13C biomolecules. Magn. Reson. Mater. Phys. Biol. Med. 2009; 22:123–134.

22. Castellani LW, Chang JJ, Wang XP, Lusis AJ, Reynolds WF. Transgenicmice express human MPO 2463G/A alleles at atherosclerotic lesions,developing hyperlipidemia and obesity in 2463G males. J. Lipid Res.2006; 47: 1366–1377.

23. Li D, Patel AR, Klibanov AL, Kramer CM, RuizM, Kang BY,Mehta JL, BellerGA, Glover DK, Meyer CH. Molecular imaging of atherosclerotic plaquestargeted to oxidized LDL receptor LOX‐1 using SPECT/CT andmagneticresonance. Circ. Cardiovasc. Imag. 2010; 3: 464–472.

hn Wiley & Sons, Ltd. NMR Biomed. 2011; 24: 1023–1028