Imaging of branched chain amino acid ... - Wiley Online Library

10 downloads 0 Views 568KB Size Report
Tomchick DR, Chuang DT. Structure of rat BCKD kinase: nucleotide-induced domain communication in a mitochondrial protein kinase. Proc Natl Acad Sci USA.
IJC International Journal of Cancer

Imaging of branched chain amino acid metabolism in tumors with hyperpolarized 13C ketoisocaproate Magnus Karlsson1, Pernille R. Jensen1, Rene´ in ’t Zandt1, Anna Gisselsson1, Georg Hansson1, Jens Ø. Duus2, Sebastian Meier2 and Mathilde H. Lerche1 1 2

Imagnia AB, Box 8225, 200 41 Malmo¨, Sweden Carlsberg Laboratory, Gamle Carlsberg Vej 10, 2500 Valby, Denmark

Modern genetic tools have markedly improved the understanding of cancer as a genetic disease by linking the development, progression and remission of cancer to underlying genetic changes.1,2 These approaches have also revealed the genetic heterogeneity of tumors. Thus, an understanding of the molecular signatures of the disease with noninvasive techniques would be highly desirable in order to define molecular targets for a tumor-specific or even personalized diagnosis and treatment.3 This can be achieved by hyperpolarized chemical shift imaging (CSI), which is a recently devised imaging modality for the visualization of molecular processes in vivo.4–6 The method relies on a signal enhancement of the inherently weak nuclear magnetic resonance (NMR) signal by several orders of magnitude in a process termed dynamic nuKey words: 13C NMR, BCAT, a-ketoisocaproate, leucine, DNP-NMR Abbreviations: BCAT: branched chain amino acid transferase; CSI: chemical shift imaging; DNP: dynamic nuclear polarization; DOTA: 1,4,7,10-tetraazacyclododecane-N,N0 ,N,N0 -tetraacetic acid; ECG: electrocardiogram; EL4: murine lymphoma; FOV: field of view; KIC: keto-isocaproate; NMR: nuclear magnetic resonance; R3230AC: rat mammary adenocarcinoma; RF: radiofrequency; TR: repetition time Grant sponsor: GE Healthcare DOI: 10.1002/ijc.25072 History: Received 21 Sep 2009; Accepted 19 Oct 2009; Online 3 Dec 2009 Correspondence to: Mathilde H. Lerche, Imagnia AB, Box 8225, 200 41 Malmo¨, Sweden, Tel: þ46-40-82760, Fax: þ46-40-82761, þ45-33-274708, E-mail: [email protected]

C 2009 UICC Int. J. Cancer: 127, 729–736 (2010) V

clear polarization (DNP). This process increases the magnetization of nuclear spins ex situ to generate a ‘hyperpolarized’ molecule, which is injected as a nonradionuclide imaging marker.7 The enhancement of detectable NMR signal by several orders of magnitude renders imaging experiments with a variety of substrates possible. As NMR is a high-resolution spectral technique, the method bears a unique potential to monitor chemical modifications of the hyperpolarized molecule in vivo. Previous studies using hyperpolarized pyruvate have detected cancer tissues by their increased anaerobic metabolism.4,5,8 Notably, the method has also shown the potential to measure early tumor responses to therapy.9 Cellular amino acid metabolism is regulated by the activity, organ distribution and cellular compartmentalization of metabolic enzymes including branched chain amino acid transferase (BCAT).10–12 The current study shows the use of hyperpolarized a-keto-[1-13C]isocaproate (KIC) as a novel marker for imaging tissue state in vivo. KIC is metabolized to its transamination product leucine by the branched chain amino acid transaminase (BCAT) in a reaction that concomitantly converts glutamate to a-ketoglutarate (Fig. 1a). BCAT was originally identified as an overexpressed gene product in a mouse teratocarcinoma cell line.13 Evidence has accumulated suggesting that BCAT is a useful marker for the grading and genetic characterization of tumors.14–19 We employ hyperpolarized [1-13C]KIC in a preclinical study of the tissue-specific conversion of KIC to leucine. Metabolism of hyperpolarized [1-13C]KIC yields unprecedented MRI contrast between a murine lymphoma (EL4) tumor and surrounding healthy tissue. A rat mammary adenocarcinoma

Short Report

Powerful analytical tools are vital for characterizing the complex molecular changes underlying oncogenesis and cancer treatment. This is particularly true, if information is to be collected in vivo by noninvasive approaches. In the recent past, hyperpolarized 13C magnetic resonance (MR) spectroscopy has been employed to quickly collect detailed spectral information on the chemical fate of tracer molecules in different tissues at high sensitivity. Here, we report a preclinical study showing that a-ketoisocaproic acid (KIC) can be used to assess molecular signatures of tumors with hyperpolarized MR spectroscopy. KIC is metabolized to leucine by the enzyme branched chain amino acid transferase (BCAT), which is found upregulated in some tumors. BCAT is a putative marker for metastasis and a target of the proto-oncogene c-myc. Very different fluxes through the BCAT-catalyzed reaction can be detected for murine lymphoma (EL4) and rat mammary adenocarcinoma (R3230AC) tumors in vivo. EL4 tumors show a more than 7-fold higher hyperpolarized 13C leucine signal relative to the surrounding healthy tissue. In R3230AC tumor on the other hand branched chain amino acid metabolism is not enhanced relative to surrounding tissues. The distinct molecular signatures of branched chain amino acid metabolism in EL4 and R3230AC tumors correlate well with ex vivo assays of BCAT activity.

730

Amino acid metabolism in tumors

Figure 1. (a) Reaction scheme for the metabolic production of leucine from a-ketoisocaproic acid. The reaction is catalyzed by BCAT, which catalyzes the transamination of leucine, isoleucine and valine to the respective a-ketoacids. (b) Metabolic production of [1-13C]leucine (176.8 ppm) in an EL4 mouse model after injecting 175 ll of 20 mM hyperpolarized [1-13C]KIC over 6 sec. MR spectra were recorded on 20 mm image slices with repetition every 2 sec using 10 flip angles. The spectra shown here monitor metabolism inside a tumor volume element. (c) Time course of signal amplitudes in the spectra shown in (b), corrected for polarization losses from previous experiments

Short Report

[cosn(10 ) for n previous experiments]. Chemical shift images were recorded 20–32 sec after KIC injection.

(R3230AC) tumor model on the other hand does not show such strongly enhanced metabolism of hyperpolarized [1-13C]KIC inside the tumor as compared to surrounding tissue. The detected [1-13C]leucine signal in the two models correlates well with ex vivo measurements of the BCAT activity in tissue samples from the two tumors. Thus, hyperpolarized [1-13C]KIC is identified as a biomarker with the potential to advance the molecular understanding of genetic and metabolic differences between tumors by assessing tissue BCAT activity in vivo by means of the [1-13C]leucine signal.

Material and Methods Cells

The mouse lymphoma cell line EL4 was obtained from DeEn Hu, Cambridge University. These cells were grown in suspension with 90% RPMI 1640 medium from GTF (Gothenburg, Sweden) þ 10% fetal bovine serum in a 5% CO2 atmosphere to a concentration of 1–1.5  106/ml and harvested by centrifugation. The rat mammary carcinoma cell line R3230AC, clone D was obtained from Professor Gianni Bussolati, University of Turin. The rat mammary cell lines

were grown adherently in 90% RPMI 1640 medium þ 10% fetal bovine serum in a 5% CO2 atmosphere to confluence and harvested by trypsination. Tumor preparation

After harvesting, EL4 cells were washed in phosphate buffered saline and resuspended to 50  106/ml in phosphate buffered saline. Female c57Bl/6 mice were sedated and 100 ll of the phosphate buffered saline suspension (5  106 EL4 cells) were subcutaneously injected into the right flank. Tumors were grown for 9–11 days before the mice were imaged. Tumors from R3230AC cells were grown for two passages in female Fisher rats according to a literature procedure.20 The rats were imaged when tumors reached a size of roughly 1 cm3 (circa 20 days of the second passage). All animal experiments were approved by the local ethical committee. Imaging substance

[1-13C]KIC was purchased as sodium salt from Cambridge Isotope Laboratories (Andover, MA) and converted to the acid form by acidifying an aqueous solution of the salt to pH C 2009 UICC Int. J. Cancer: 127, 729–736 (2010) V

Karlsson et al.

< 1 with sulfuric acid followed by an extraction of the acidified solution with diethylether. After drying the ether phase with anhydrous magnesium sulfate, the solvent was removed in vacuum. The product [1-13C]KIC was a colorless oil. A DNP-preparation of [1-13C]KIC was established by dissolving the trityl radical (tris(8-carboxyl-2,2,6,6-tetra(2(1hydroxyethyl))-benzo[1,2-d:4,5-d0 ]bis(1,3)dithiole-4-yl)methyl sodium in the acid to 11 mM concentration. A trimeric GdDOTA complex was added to a concentration of 0.4 mM trimer.21 Addition of the Gd complex enhances the solid state polarization.22 The negative effect on the relaxation time constant of the hyperpolarized substance in solution is negligible at the concentrations used. As a neat acid, 7 M [1-13C]KIC vitrifies readily in the polarizer. This DNP-preparation of [1-13C]KIC was hyperpolarized in a polarizer as described.7 The hyperpolarized sample was subsequently dissolved in an aqueous solution of sodium hydroxide and phosphate buffer (40 mM, osmolality adjusted with NaCl to 210 mOsm) to provide a 20 mM solution of hyperpolarized [1-13C]KIC at pH of 7.4 6 0.1, an osmolality of 290 mOsm 6 10, T1{9.4 T} of 55 6 3 sec and a polarization of 32% 6 3% at the time of administration. A dose of 0.175 mmol/kg was infused during 6 sec. Twenty seconds after start of the infusion, the CSI sequence was started.

731

was connected to the catheter to allow for accurate dosing. Considering the dead volume in the tail vein catheter the volume administered was 175 ll. In the rat R3230AC experiments the same procedure was followed but the administered volume of the hyperpolarized imaging substance was 1500 ll and gas mixture flow was increased to 400 ml/min. Body temperature of the animals was controlled to vary within 37.0 6 0.2 C. ECG, breathing and rectal temperature were monitored with SA Instruments 1025 MR compatible animal monitoring system. Injection of KIC did not lead to any abnormal effects on the physiological parameters that were monitored. Processing of imaging data

Metabolic images were calculated from CSI data sets with time domain fitting in jMRUI.23 After phasing of the spectra, amplitudes were determined assuming constant phase, identical line width and a fixed frequency difference (110 Hz) of KIC and leucine. Results were further processed with homewritten Matlab (Mathworks, Natick, MA) scripts. The surface coil with its inhomogeneous reception profile affects the 13C image appearance to a certain extent. The distribution of KIC signal in the picture is therefore partly dependent on the sensitivity of the surface coil used and therefore the very edge of the tumor appears to have less KIC than the centre.

MR experiments

C 2009 UICC Int. J. Cancer: 127, 729–736 (2010) V

Fluorometric and spectrophotometric assays

BCAT activities in the different cell lines and tissue extracts were determined spectrophotometrically according to Schadewaldt and Adelmeyer.24 The protein content was determined spectrophotometrically with a commercial BCA-kit (BCA1; Sigma-Aldrich, St. Louis, MO). Metabolite concentrations were determined by adapting a fluorescence assay by Goldberg et al. originally developed for determination of malate.25

Results In vivo assessment of BCAT reaction

[1-13C]KIC was hyperpolarized to 30% polarization, equaling a signal enhancement on the order of 105 relative to the equilibrium 13C spin polarization in state-of-the-art medical scanners of 3 T magnetic field at 310 K (0.0003%). Upon [1-13C]KIC injection into an EL4 mouse, ample 13C signal is detectable in mouse tissue within few seconds in vivo, both for [1-13C]KIC (d13C ¼ 172.6 ppm) and its transamination product [1-13C]leucine (d13C ¼ 176.8 ppm) (Fig. 1). No other reaction product than [1-13C]leucine gets quickly detectable, presumably due to the lower activity of branched chain a-ketoacid dehydrogenase, which catalyzes the rate limiting oxidative decarboxylation of a-ketoacids in the degradation of branched chain amino acids.26 Hyperpolarized [1-13C]KIC thus is a biomarker that specifically reports on metabolism by BCAT in vivo. Sufficient sensitivity for in vivo assays results from high signal enhancement and efficient [1-13C]KIC transport. Efficient KIC uptake into normal and neoplastic mammalian tissue is plausible, as KIC has been

Short Report

Animals (n ¼ 5 mice and n ¼ 4 rats) were anesthetized with a gas mixture of 96% air and 4% isoflurane. A catheter was placed in the tail vein and saline with heparin 5 IE/ml was injected at a volume corresponding to the imaging experiments. The animals were positioned in a 2.35 T Bruker (Fa¨llanden, Switzerland) Biospec Avance II MR scanner. Anestesia was maintained in the MR system by a gas mixture of oxygen and nitrous oxide with 2% isoflurane. Gas flow was 200 ml/min O2 and 200 ml/min N2O. MR imaging was performed using a dual-tuned 1H-13C birdcage resonator (inner diameter of 72 mm) and a standard proton MR imaging sequence to retrieve anatomic information including the location of the tumor. The 13C spectra and images were acquired with a 20 mm surface coil placed around the tumor and the surface coil was decoupled from the dual-tuned 1H-13C birdcage resonator. Imaging of the hyperpolarized substances in the tumor was performed with a CSI sequence with centric circular sampling. A delay trigger was added on the SA (Edison, NJ) instrument (100 ms, length 900 ms) resulting in a total acquisition time of 12 s. Other MR parameters were FOV 35  35 mm2  10 mm, matrix size of 16  16, 10 RF pulse, TR of 35 ms. This CSI sequence was started 20 sec after injection of KIC. Two imaging slices were acquired with a thickness of 5 mm each. Thus, the spatial resolution was 2.2  2.2  5 mm3. In the mouse EL4 experiments, a catheter with a volume of 300 ll was filled with solution of the hyperpolarized imaging substance and the line was connected to the tail catheter. The animal was repositioned into the magnet. A Hamilton syringe filled with 200 ll saline solution

732

Amino acid metabolism in tumors

Figure 2. Imaging of BCAT activity in vivo. The anatomical 1H image of an EL4 mouse is shown on the top left. Chemical shift images of [1-13C]leucine and [1-13C]KIC after injection of hyperpolarized [1-13C]KIC are overlaid onto the anatomical image. The tumor position is indicated by a dashed line. [1-13C]leucine is specifically observed inside the tumor. 1D 13C spectra for volume elements of the tumor and intestine (blue and red dots) demonstrate the difference in [1-13C]leucine signal between tumor and surrounding tissue. The position of the

Short Report

surface coil is indicated by the two white dots on the top left.

described as a substrate of various fast monocarboxylate transporters including MCT1, which is abundant in muscle tissue and transformed cell lines.27–29 A representative time series of in vivo MR spectra taken from a tumor specific tissue slice in EL4 tumor bearing mice is shown in Figure 1b. The hyperpolarization of 13C labeled compounds allows detecting the signals of interest without interference from the nonlabeled, nonhyperpolarized background. The time series shows that the intracellular BCAT activity leads to maximum [1-13C]leucine signal 20 sec after injection of hyperpolarized [1-13C]KIC (Fig. 1c). Both [1-13C]KIC and [1-13C]leucine signals have lost the vast majority of their enhanced 13C spin polarization after 100 sec in vivo (Fig. 1c). Information on the distribution and metabolism of [1-13C]KIC can thus be sensitively sampled within the first minute after injection. Hyperpolarized [1-13C]leucine imaging in EL4 murine lymphoma

Murine lymphoma [1-13C]KIC and [1-13C]leucine signals were imaged in 5 mm slices through the tumor 20–32 sec after intravenous injection of hyperpolarized [1-13C]KIC (n ¼ 5). Resultant 13C chemical shift images were overlaid with anatomical 1 H images (Fig. 2) to yield functional images of [1-13C]KIC distribution and [1-13C]leucine synthesis. Images obtained in this way show that [1-13C]KIC has been distributed to the gastrointestinal tract and tumor within the first 20 sec after injec-

tion. After injection of 4 lmol [1-13C]-KIC, the signal-to-noise ratio of [1-13C]-leucine in the EL4 lymphoma model is determined to be 13.3 6 6.3 (mean 6 SD). Most notably, [1-13C]leucine signal is detected within the tumor at a high contrast of 6.9 6 1.0 relative to the highest [1-13C]leucine signal in surrounding tissues. The [1-13C]KIC images indicate that [1-13C]KIC signal is not specifically increased inside EL4 tumor tissue. This suggests that substrate uptake is of minor importance for enhanced flux through the BCAT catalyzed reaction inside the neoplasm and thus for [1-13C]leucine contrast between tumor and healthy tissue. Influence of [1-13C]KIC uptake on the significant contrast in [1-13C]KIC metabolism between EL4 tumor and healthy tissue was additionally assessed by optical tests ex vivo (see ‘‘Material and Methods’’ section for details). To this end, KIC uptake into tumor and normal tissue during in vivo DNPNMR assays was determined by freeze clamping tumor and muscle tissue before or 25 sec after injection of 4 lmol KIC into mice. Total KIC pools before injection were 0.05 6 0.05 and 0.17 6 0.05 lmol per gram tissue for tumor and muscle, respectively, while pools after injection were increased to 0.67 6 0.04 and 0.61 6 0.05 lmol/g tissue (n ¼ 5 for all groups) for tumor and muscle, respectively. As BCAT has a low KM (0.14 mM) for KIC,24 levels of the substrate after injection of hyperpolarized [1-13C]KIC are not limiting to the BCAT catalyzed reaction either in tumor or muscle tissue, and substrate C 2009 UICC Int. J. Cancer: 127, 729–736 (2010) V

Karlsson et al.

tumor (red) and normal (blue) tissue of a R3230AC rat. Spectra are compared to the in vivo tumor spectrum of the EL4 mouse (gray) as shown in Figure 2, to display lower branched chain amino acid metabolism in R3230AC tumor.

availability is not expected to contribute significantly to tissue contrast of branched chain amino acid metabolism between tumor and surrounding tissue. Muscle was chosen for comparison with tumor, as it is among the tissues with the highest BCAT activity in healthy mammals and thus is a determinant of the tissue contrast of branched chain amino acid metabolism between tumor and healthy tissue. Hyperpolarized [1-13C]leucine imaging in R3230AC rat mammary adenocarcinoma

Previous studies have indicated that BCAT gene expression is significantly increased in a variety of tumor cell lines but that this over expression is not a general feature of neoplasms.14 In vivo imaging experiments using hyperpolarized [1-13C]KIC were therefore conducted also in a rat breast cancer model to assess tumor specific differences in branched chain amino acid metabolism in vivo. As for the lymphoma model, the hyperpolarized [1-13C]KIC signal is evenly distributed in the rat adenocarcinoma tumor model over the entire imaging slice. Turnover of [1-13C]KIC to [1-13C]leucine in the R3230AC model is, however, minor (Fig. 3). Substantial [1-13C]KIC signal in tumor volume elements indicates that [1-13C]leucine synthesis is not limited by tissue perfusion in the rat model. Further, KIC pools were assayed ex vivo also for the R3230AC model, and substantial tracer uptake to concentrations above the BCAT Michaelis constant KM for KIC was found both for muscle (0.07 6 0.003 lmol per gram tissue before KIC infusion, 151 6 14 lmol/g tissue after KIC infusion) and tumor tissue (0.05 6 0.01 lmol/g tissue before KIC infusion, 199 6 56 lmol/g tissue after KIC infusion). To validate ample tissue perfusion of rat R3230AC tumor tissue with marker molecules, hyperpolarized [1-13C]pyruvate was employed. Hyperpolarized [1-13C]pyruvate is a well established tumor marker due to the increased levels of anaerobic glycolysis and resultant high lactate dehydrogenase catalyzed C 2009 UICC Int. J. Cancer: 127, 729–736 (2010) V

flux of 13C signal between [1-13C]pyruvate and [1-13C]lactate in neoplastic tissue.5,9 Therefore, [1-13C]pyruvate was used as a reference substrate both in EL4 and R3230AC tumors. Use of hyperpolarized [1-13C]pyruvate shows the expected high [1-13C]lactate formation in both tumor models. This confirms high tissue perfusion and efficient cellular monocarboxylic acid uptake in both tumor models and indicates that higher [1-13C]leucine formation in EL4 tumors results from higher BCAT activity or co-substrate availability rather than from higher tissue perfusion and marker uptake. This underlines the use of hyperpolarized [1-13C]pyruvate as a general in vivo tumor marker, while hyperpolarized [1-13C]KIC provides a discriminative marker for the non-invasive profiling of branchedchain amino acid metabolism in tumors. Mechanistic assays ex vivo

BCAT activity and reactant pools were assayed ex vivo with optical tests on tissue extracts in order to rationalize high [1-13C]leucine contrast between EL4 tumor and healthy tissue. Tissue extracts of mouse EL4 and rat R3230AC tumor were compared to extracts of mouse and rat muscle. Total pools of the reactants KIC, leucine, glutamate and a-ketoglutarate (Fig. 1a) determined in this way show significantly increased total glutamate and leucine pools both in EL4 and R3230AC tumor tissue compared to muscle tissue, supposedly due to the recruitment of amino acids for tumor proliferation (Fig. 4a).30 Concentrations of the corresponding a-ketoacids on the other hand are low, both in tumor and muscle. In addition to metabolite pools, BCAT activities of different tissues were assayed ex vivo. BCAT activities were comparable in rat muscle, R3230AC tumor and mouse muscle, but were substantially increased in solid EL4 tumors (Fig. 4b). Spectrophotometric assays of BCAT activity were repeated on cell cultures of EL4 and R3230AC cell lines to minimize the possible effects of the tumor environment on cell differentiation.31 BCAT activities determined in this way are in agreement with the data from hyperpolarized in vivo NMR and spectrophotometric ex vivo tumor assays, showing significantly higher BCAT activity in EL4 (16.6 6 1.7 U/g protein) than in R3230AC cell suspensions (3.7 6 1.8 U/g protein).

Discussion We describe a pilot study showing that hyperpolarized markers are suitable for the profiling of tumors at the single gene level. To this end, hyperpolarized ketoisocaproate is introduced as a novel imaging modality for tumors with high BCAT activity. In mammals, BCAT occurs as a cytosolic and a mitochondrial isoform, both of which have been reported overexpressed in cancer cell lines and gastric cancer biopsies.14,15,17 The cytosolic BCAT gene Bcat1 was derived from a subtraction/co expression strategy with Myc-induced tumors and Bcat1 was shown to be a direct target of the c-myc oncogene.32–34 Bcat1 and c-myc mRNA levels show, however, no direct simple correlation between Bcat1 and c-myc gene expression.14–16 The Bcat1 gene is further linked to the oncogene Kras-2 in mouse and human

Short Report

Figure 3. In vivo magnetic resonance spectra of KIC metabolism in

733

734

Amino acid metabolism in tumors

Figure 4. (a) Metabolite pools in EL4 mouse tumor and muscle (MM) tissue, R3230AC rat tumor and muscle (RM) tissue. Amino acid concentrations are increased in tumor tissues, but similar in different tumors (n ¼ 5 for all groups; white bars represent signal below the detection limit). Data show the mean along with the standard deviation. (b) Tissue BCAT activity ex vivo verifies 10-fold increased activity in

Short Report

solid EL4 tumors relative to R3230AC tumors and muscle.

genomes17,18 and SV40 large T antigen-induced tumors show amplified Bcat1 expression.19 More recently, mRNA expression levels and immunohistochemistry have shown that Bcat1 over expression is a sensitive marker of metastases in medullablastoma16 and colorectal cancer.35 Fast in vivo assays of branched chain amino acid metabolism should be of interest for cancer diagnosis and prognosis. Our finding of largely increased branched chain amino acid metabolism in EL4 but not in R3230AC tumors in vivo is in principal agreement with previous reports of increased BCAT mRNA transcription in specific tumor lines.14–16 Recent transcriptome and proteome analysis of biopsy samples have pointed to problems in correlating RNA and protein levels of BCAT1.16 Thus, rapid functional in vivo assays with natural substrates as introduced by BCAT assays with hyperpolarized [1-13C]KIC seem highly prospective for delineating actual biochemical pathway activities in normal and abnormal tissue. [1-13C]KIC is a well-suited in vivo marker for the BCAT catalyzed reaction due to its efficient tissue uptake. Our data suggest that tissue contrast of the [1-13C]leucine signal after [1-13C]KIC injection originates from higher BCAT activity and increased glutamate pools in EL4 tumor (Fig. 4). Various studies indicate that increased glutamate concentrations are common in tumors.36 As BCAT has a particularly high KM of 6.6 mM for glutamate (when compared with 0.14 mM for KIC),24 increased glutamate pools in tumors support increased [1-13C]leucine formation. High pool sizes of the KIC transamination product leucine itself in tumor can additionally assist the flux of 13C signal from [1-13C]KIC into the cellular [1-13C]leucine pool by increasing the overall turnover of the enzymatic reaction.9 Comparable

pools of leucine and glutamate are found in EL4 and R3230AC tumors (Fig. 4). Nevertheless, R3230AC tumors exhibit low flux through the BCAT catalyzed reaction in agreement with low BCAT enzyme activities as determined ex vivo. This indicates the potential of hyperpolarized [1-13C]KIC for the molecular profiling of tumors. Our in vivo and ex vivo assays do not ultimately rule out different cellular [1-13C]KIC uptake into normal and pathological tissue, but the strong correlation between [1-13C]KIC metabolism in vivo and BCAT enzymatic activity as determined ex vivo suggests that BCAT rather than transporter activity is assayed in vivo. The tissue contrast between neoplasm and normal tissue in the current study ultimately depends on BCAT activities in the surrounding tissues and on the signal amplitude obtained in the tumor. Use of high-field hyperpolarization in a 4.6 T polarizer shows that a further 2-fold increase in tracer polarization and corresponding in vivo signal amplitudes are within the realms of possibility when using the latest generation of DNP polarizers.26 This implies a further increase of tissue contrast for the instances, where nontumor leucine signal is below the noise level. Increased in vivo signal will alternatively allow further reduction of tracer concentrations. The capacity to reduce tracer concentrations is supported by the finding that injection of hyperpolarized [1-13C]KIC in our setup results in KIC tissue concentrations above the Michaelis constant KM of BCAT towards KIC. Under these in vivo experimental conditions, the enzyme is therefore close to saturation with KIC. Variation of the KIC dose given in the mouse EL4 tumor experiments between 10 lmol and 2 lmol accordingly did not alter the tumor leucine C 2009 UICC Int. J. Cancer: 127, 729–736 (2010) V

Karlsson et al.

735

signal. This suggests that injection of hyperpolarized [1-13C]KIC at even lower concentrations may be viable for biocompatible and little cost-intensive assays with clinical potential.

Acknowledgements This work was financially supported by a research grant from GE Healthcare to M. Karlsson, P.R. Jensen, R. in ’t Zandt, G. Hansson, A. Gisselsson, M.H. Lerche. We thank Jan H. Ardenkjær-Larsen for providing access to the 4.6 T polarizer.

1.

Vogelstein B, Kinzler KW. The multistep nature of cancer. Trends Genet 1993;9: 138–41. 2. DeRisi J, Penland L, Brown PO, Bittner ML, Meltzer PS, Ray M, Chen Y, Su YA, Trent JM. Use of a cDNA microarray to analyse gene expression patterns in human cancer. Nat Genet 1996;14:457–60. 3. Brindle K. New approaches for imaging tumour responses to treatment. Nat Rev Cancer 2008;8:94–107. 4. Golman K, in ’t Zandt R, Thaning M. Real-time metabolic imaging. Proc Natl Acad Sci USA 2006;103:11270–5. 5. Golman K, Zandt RI, Lerche M, Pehrson R, Ardenkjaer-Larsen JH. Metabolic imaging by hyperpolarized 13C magnetic resonance imaging for in vivo tumor diagnosis. Cancer Res 2006;66:10855– 60. 6. Schroeder MA, Cochlin LE, Heather LC, Clarke K, Radda GK, Tyler DJ. In vivo assessment of pyruvate dehydrogenase flux in the heart using hyperpolarized carbon13 magnetic resonance. Proc Natl Acad Sci USA 2008;105:12051–6. 7. Ardenkjær-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 USA 2003;100:10158–63. 8. Albers MJ, Bok R, Chen AP, Cunningham CH, Zierhut ML, Zhang VY, Kohler SJ, Tropp J, Hurd RE, Yen YF, Nelson SJ, Vigneron DB, et al. Hyperpolarized 13C lactate, pyruvate, and alanine: noninvasive biomarkers for prostate cancer detection and grading. Cancer Res 2008;68:8607–15. 9. Day SE, Kettunen MI, Gallagher FA, Hu DE, Lerche M, Wolber J, Golman K, Ardenkjaer-Larsen JH, Brindle KM. Detecting tumor response to treatment using hyperpolarized 13C magnetic resonance imaging and spectroscopy. Nat Med 2007;13:1382–7. 10. Burns KH, Owens GE, Ogbonna SC, Nilson JH, Matzuk MM. Expression profiling analyses of gonadotropin responses and tumor development in the absence of inhibins. Endocrinology 2003;144:4492–507. 11. Suryawan A, Hawes JW, Harris RA, Shimomura Y, Jenkins AE, Hutson SM. A molecular model of human branched-chain amino acid metabolism. Am J Clin Nutr 1998; 68:72–81. 12. Jensen PR, Karlsson M, Meier S, Duus JØ, Lerche MH. Hyperpolarized amino acids

13.

14.

15.

16.

17.

18.

19.

20.

21. 22.

C 2009 UICC Int. J. Cancer: 127, 729–736 (2010) V

for in vivo assays of transaminase activity. Chem Eur J 2009;15:10010–2. Niwa O, Kumazaki T, Tsukiyama T, Soma G, Miyajima N, Yokoro K. A cDNA clone overexpressed and amplified in a mouse teratocarcinoma line. Nucleic Acids Res 1990;18:6709. Ben-Yosef T, Yanuka O, Halle D, Benvenisty N. Involvement of Myc targets in c-myc and N-myc induced human tumors. Oncogene 1998;17:165–71. Kato T, Goto H, Niwa Y, Ohmiya N, Hayakawa T, Murakami T, Nakai N, Hutson SM, Shimomura Y. Expression of mRNAs for c-myc and branched-chain aminotransferases in human gastric cancer cells and tissues. J Clin Biochem Nutr 2000; 29:29–36. de Bont JM, Kros JM, Passier MM, Reddingius RE, Sillevis Smitt PA, Luider TM, den Boer ML, Pieters R. Differential expression and prognostic significance of SOX genes in pediatric medulloblastoma and ependymoma identified by microarray analysis. Neuro Oncol 2008;10:648–60. Rodriguez S, Jafer O, Goker H, Summersgill BM, Zafarana G, Gillis AJ, van Gurp RJ, Oosterhuis JW, Lu YJ, Huddart R, Cooper CS, Clark J, et al. Expression profile of genes from 12p in testicular germ cell tumors of adolescents and adults associated with i(12p) and amplification at 12p11.2-p12.1. Oncogene 2003;22:1880–91. Manenti G, Galbiati F, Gianni-Barrera R, Pettinicchio A, Acevedo A, Dragani TA. Haplotype sharing suggests that a genomic segment containing six genes accounts for the pulmonary adenoma susceptibility 1 (Pas1) locus activity in mice. Oncogene 2004;23:4495–504. Weggen S, Preuss U, Pietsch T, Hilger N, Klawitz I, Scheidtmann KH, Wiestler OD, Bayer TA. Identification of amplified genes from SV40 large T antigen-induced rat PNET cell lines by subtractive cDNA analysis and radiation hybrid mapping. Oncogene 2001;20:2023–31. Sapino A, Papotti M, Sanfilippo B, Gugliotta P, Bussolati G. Tumor types derived from epithelial and myoepithelial cell lines of R3230AC rat mammary carcinoma. Cancer Res 1992;52:1553– 60. GE Healthcare. US Patent WO2007/ 064226. Johannesson H, Macholl S, ArdenkjaerLarsen JH. Dynamic Nuclear Polarization

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

of [1-13C]pyruvic acid at 4.6 Tesla. J Magn Reson 2009;197:167–75. Vanhamme L, Sundin T, Hecke PV, Huffel SV. MR spectroscopy quantitation: a review of time-domain methods. NMR Biomed 2001;14:233–46. Schadewaldt P, Adelmeyer F. Coupled enzymatic assay for estimation of branched-chain L-amino acid aminotransferase activity with 2-Oxo acid substrates. Anal Biochem 1996;238:65–71. Bergmeyer HU. Methods of enzymatic analysis, vol. 1 Weinheim: Verlag Chemie GmbH, 1974. Machius M, Chuang JL, Wynn RM, Tomchick DR, Chuang DT. Structure of rat BCKD kinase: nucleotide-induced domain communication in a mitochondrial protein kinase. Proc Natl Acad Sci USA 2001;98:11218–23. Broer S, Broer A, Schneider HP, Stegen C, Halestrap AP, Deitmer JW. Characterization of the high-affinity monocarboxylate transporter MCT2 in Xenopus laevis oocytes. Biochem J 1999;341 (Pt 3):529–35. Broer S, Schneider HP, Broer A, Rahman B, Hamprecht B, Deitmer JW. Characterization of the monocarboxylate transporter 1 expressed in Xenopus laevis oocytes by changes in cytosolic pH. Biochem J 1998;333 (Pt 1):167–74. Kilberg MS, Gwynn MB. Plasma membrane transport of 2-ketoisocaproate by rat hepatocytes in primary culture. J Biol Chem 1983;258:11524–7. Jager PL, Vaalburg W, Pruim J, de Vries EG, Langen KJ, Piers DA. Radiolabeled amino acids: basic aspects and clinical applications in oncology. J Nucl Med 2001; 42:432–45. Rottiers P, Verfaillie T, Contreras R, Revets H, Desmedt M, Dooms H, Fiers W, Grooten J. Differentiation of EL4 lymphoma cells by tumoral environment is associated with inappropriate expression of the large chondroitin sulfate proteoglycan PG-M and the tumor-associated antigen HTgp-175. Int J Cancer 1998;78: 503–10. Benvenisty N, Leder A, Kuo A, Leder P. An embryonically expressed gene is a target for c-Myc regulation via the c-Myc-binding sequence. Genes Dev 1992;6:2513–23. Ben-Yosef T, Yanuka O, Benvenisty N. ECA39 is regulated by c-Myc in human and by a Jun/Fos homolog, Gcn4, in yeast. Oncogene 1996;13:1859–66.

Short Report

References

736

35. Yoshikawa R, Yanagi H, Shen CS, Fujiwara Y, Noda M, Yagyu T, Gega M, Oshima T, Yamamura T, Okamura H, Nakano Y, Morinaga T, et al. ECA39 is a novel distant metastasis-related biomarker in colorectal cancer. World J Gastroenterol 2006;12:5884–9.

36. Taylor BS, Pal M, Yu J, Laxman B, KalyanaSundaram S, Zhao R, Menon A, Wei JT, Nesvizhskii AI, Ghosh D, Omenn GS, Lubman DM, et al. Humoral response profiling reveals pathways to prostate cancer progression. Mol Cell Proteomics 2008;7:600–11.

Short Report

34. Schuldiner O, Eden A, Ben-Yosef T, Yanuka O, Simchen G, Benvenisty N. ECA39, a conserved gene regulated by cMyc in mice, is involved in G1/S cell cycle regulation in yeast. Proc Natl Acad Sci USA 1996;93:7143–8.

Amino acid metabolism in tumors

C 2009 UICC Int. J. Cancer: 127, 729–736 (2010) V