Choline Transporter as a Novel Target for Molecular Imaging of Cancer

Abstract

Abnormalities of choline processing in cancer cells have been used as a basis for imaging of cancer with positron emission tomography and magnetic resonance spectroscopy. In this study, the transport mechanism for choline was investigated in cultured PC-3 prostate cancer cells. Furthermore, tritiated hemicholinium 3 (HC-3), a well-known inhibitor of choline transport, was studied as a prototypic molecular imaging probe in PC-3 cells and 9L glioma–bearing rats. [³H]Choline uptake by PC-3 cells was found to have both facilitative and nonfacilitative components. Facilitative transport was characterized by partial sodium dependence and intermediate affinity (K_M = 9.7 ± 0.8 μM). HC-3 inhibited choline with a K_I of 10.5± 2.2 μM. Ouabain (1 mM) caused a 94% reduction in choline uptake. At physiologic choline concentration, phosphocholine was the rapid and predominant metabolic fate. The binding of [³H]HC-3 to PC-3 cells was rapid and specific (competitively blocked with unlabeled HC-3). Biodistribution of [³H]HC-3 in 9L glioma–bearing rats showed the ranking of uptake to be kidney > lung > tumor > liver > skeletal muscle ≈ blood > brain. In comparison with [¹⁴C]choline, [³H]HC-3 showed over twofold higher tumor uptake and favorable uptake ratios of tumor to blood, tumor to muscle, tumor to lung, and tumor to liver. The data demonstrate the quantitative importance of an intermediate-affinity, partially sodium-dependent choline transport system on choline processing in PC-3 cancer cells. The biodistribution properties of [³H]HC-3 in tumor-bearing rats encourage the development of molecular imaging probes based on choline transporter binding ligands.

ALTERATIONS IN CHOLINE METABOLISM have come to be recognized as hallmarks of malignancy.¹ Consequently, intense interest has been given to the development of imaging techniques that evaluate choline processing for the purposes of cancer diagnosis,^1–6 staging,^7–9 and therapy monitoring.^1,2,10,11 Owing to their capabilities to evaluate metabolic processes, positron emission tomography (PET) and magnetic resonance spectroscopy (MRS) are the predominant modalities used for noninvasive imaging of choline metabolism. MRS studies of biopsied cancerous tissues and MRS imaging of in vivo tumors have generally shown elevated levels of choline-containing metabolites in cancer relative to normal tissue.^{1,3,4,10–16} Of these choline-containing metabolites, phosphocholine (PCho) has been found to most closely correlate with malignant transformation,^12,13 tumor growth rate,¹⁴ metastatic potential,¹⁵ and therapeutic response.^1,16

Tracer kinetic studies have shown that enhanced choline transport may play a role in the elevation of PCho levels in malignant cells.^17–20 High intracellular PCho to choline concentration ratios¹⁷ indicate that choline kinase activity levels exceed those needed to supply PCho for phosphatidylcholine synthesis along the Kennedy pathway.²¹ Indeed, metabolism of PCho to cytidine diphosphate (CDP)-choline, catalyzed by cytidine 5'-triphosphate (CTP):PCho cytidylyltransferase, appears to remain highly regulated in tumors.²¹ PET studies with ¹¹C- and ¹⁸F-labeled choline analogues have shown rapid clearance of radiolabeled choline from the blood owing to avid transport activity in kidney and liver.^2,7 Following bolus intravenous administrations, arterial blood concentrations of choline tracers fall to a few percent of peak values as early as 2 minutes after injection. Despite rapid blood clearance, choline tracer accumulations in several types of primary and metastatic tumors are moderate to high,^2,5,7–9,22 implying sufficient perfusion and choline transporter activity to support accumulation of tracer in the tumors. Furthermore, analysis of ¹⁸F-labeled metabolites in cultured prostate cancer cells after a 2-hour incubation with [¹⁸F]fluorocholine (FCH) showed the radiolabel to be predominantly in the form of phosphorylated FCH,²² indicating transport and phosphorylation to be important steps for FCH accumulation in prostate cancer cells.

Although there is growing evidence of the importance of choline transport in tumor choline processing, the nature and properties of the choline transporter(s) in the cell membrane of cancer cells are relatively poorly understood. Much work has been performed to genetically identify and clone the choline transporters present in brain,²³ liver,²⁴ kidney,²⁵ fibroblasts,²⁶ and various peripheral tissues,²⁷ but there are currently only limited data on the choline transport process in tumor cells.^28–35 Preliminary transport kinetic measurements in several human cancer cell lines showed a range of Michaelis constant (K_M) values (1–60 μM) for choline^28–35 that was in the same range as physiologic concentrations of choline in plasma (5–50 μM),²⁷ with most values falling between the neuronal high-affinity process (hCHT1) and the lower-affinity human choline transporter-like protein (hCTL) 1 and human organic cation transporter (hOCT) family of transporters present in peripheral tissues. To our knowledge, the choline transporter has not been characterized in prostate cancer.

In the present study, we performed mechanistic studies on the transport of choline in the PC-3 human prostate cancer cell line to provide further clarification of the properties of the choline transporter. We also propose, for the first time, the use of a radiolabeled analogue of the high-affinity choline transport inhibitor hemicholinium 3 (HC-3) as a molecular imaging probe for the choline transporter in cancer. HC-3 potently inhibits the neuronal high-affinity choline uptake system (CHT1) with enzyme inhibition constant (K_I) = 0.02 μM²³ and is also a weaker inhibitor of choline kinase (K_I = 500 μM).³⁶ Cellular binding of [³H]HC-3 is studied in PC-3 cells, whereas its in vivo biodistribution properties are evaluated in a 9L glioma–bearing rat model. The results demonstrate the importance of the transport step for choline processing in PC-3 human prostate cancer cells. Favorable tumor cell binding and biodistribution properties are found for the tritiated form of HC-3, encouraging further characterization and development of molecular imaging agents that target the choline transporter in cancer cells.

Methods

Purification of Commercially Available [Methyl-3H]Choline

[Methyl-³H]choline (Perkin Elmer Life Sciences, Wellesley, MA) was prone to decompose into [methyl -³H]betaine by autoxidation if it was dissolved in water and stored even at low temperatures. The autoxidation was prevented by the addition of ethanol (to give a final concentration of 50%). We found that commercial preparations of [³H]choline dissolved in absolute ethanol contained a few percent of [³H]betaine. [³H]Choline was therefore purified by thin-layer chromatography (TLC): TLC plate, aminopropyl–silica gel plate (NH TLC plate, Fuji-Silisia Chemical, Kasugai-City, Japan); solvent, methanol; Rf of [³H]choline, 0.25; Rf of [³H]betaine, 0.75. The purity of [³H]choline was confirmed by high-performance liquid chromatography (see below).

Kinetics of Choline Transport in PC-3 Cells In Vitro

The PC-3 androgen-independent human prostate cancer cell line was chosen for characterization of the choline transport process because of its widespread availability and use and because it has served as a primary screening tool for development of imaging probes for prostate cancer. PC-3 cells (0.5 × 10⁶ cells) were added to 2 mL of Roswell Park Memorial Institute (RPMI) 1640 + 10% fetal bovine serum (FBS) medium contained in 3.7 cm wells and grown under 5% CO₂ for 48 hours. The uptake study was performed after replacing the medium with the uptake medium (20 mM HEPES/Tris buffer, pH 7.4, 141 mM NaCl, 4 mM KCl, 2.8 mM CaCl₂, 1 mM MgCl₂, and 10 mM glucose), by incubating the cells with 5 to 500 μM choline (containing [³H]choline) at 37°C for 10 minutes. The uptake reaction was stopped by addition of ice-cold phosphate-buffered saline (PBS), followed by washing with ice-cold PBS three times. The sodium dependence of the uptake was examined by using the uptake medium containing 141 mM LiCl instead of 141 mM NaCl. In all cases, the tracer uptake period followed a 5-minute preincubation with medium lacking [³H]choline to allow for stabilization of the cells in the test medium. The effects of choline transport inhibitor HC-3 (10–100 μM as bromide salt; Sigma-Aldrich, St. Louis, MO) and metabolic inhibitors, ouabain (0.1–1.0 mM; Sigma-Aldrich) and dinitrophenol (0.1 hemicholinium-3 1.0 mM; Sigma-Aldrich), were tested by adding the inhibitors to the incubation medium during the preincubation and [³H]choline incubation periods. The radioactivity was measured by liquid scintillation counting. Total protein content in the cells was determined by the Lowry method. Five wells were used for calculation of each value.

For estimation of the cell to medium radioactivity concentration ratio, cell volume was measured by the hematocrit method using condensed cell suspensions. Approximately 10 μL of well-mixed cell suspension was used to fill a Mylar-wrapped, 75 mm hematocrit capillary tube (Drummond Scientific Company, Broomall, PA). The tube was then sealed with plaster and centrifuged at 1,100 rpm for 5 minutes at 4°C. The cell volume was calculated as the quotient of the hematocrit value and the known concentration of cells in the suspension. Using this methodology, 10⁶ PC-3 cells were found to occupy a volume of 5.5 μL. Assuming a constant PC-3 cell volume for the entire study, the volume of cells in each incubation well was calculated as the product of 5.5 μL and the number of million cells per well. The cell to medium ratio was calculated as
$\begin{array}{l} C e l l - t o - m e d i u m R a t i o = \\ \frac{C P M r e t a i n e d i n c e l l s p e r u n i t v o l u m e o f c e l l s}{C P M i n u p t a k e m e d i u m p e r u n i t v o l u m e o f m e d i u m} \end{array}$
(1)
where CPM is counts/minute of radioactivity.

The numerator of Equation 1 represents the radioactivity concentration in the cells, whereas the denominator represents radioactivity concentration in the uptake medium during the uptake period.

Preliminary experiments were done to show that cellular uptake of [³H]choline was linear with time over the first 20 minutes of incubation. Thus, the uptake rate (velocity) was determined as the choline uptake at 10 minutes divided by 10. The uptake rate was then fit by a nonlinear least-squares curve fitting technique³⁷ to the following equation including a nonfacilitative diffusion term:
$\begin{array}{l} Uptake Rate (nmol / mg protein / min) = \\ \frac{V_{max} [S]}{(K_{M} (1 + \frac{[I]}{K_{1}})) + [S]} + D [S] \end{array}$
(2)
where [S] (μM) is choline concentration, [I] (μM) is HC-3 concentration, V_max (nmol/mg protein/min) and K_M (μM) are the Michaelis-Menten constants for maximal transport velocity and concentration at half-maximal velocity, respectively, K_I (μM) is the inhibition constant for HC-3, and D (mL/mg protein/min) is the diffusion coefficient for nonfacilitative diffusion. The model assumes Michealis-Menten kinetics, competitive inhibition by HC-3 at the choline transporter,³² and first-order diffusion kinetics for the nonfacilitative diffusion component. The r² statistic was used as a test of goodness of fit:
$r^{2} = \frac{\sum_{i = 1}^{n} ({\hat{y}}_{i} - \overset{ˉ}{y})^{2}}{\sum_{i = 1}^{n} (y_{i} - \overset{ˉ}{y})^{2}}$
(3)
where y_i is the measured uptake rate, $\overset{ˉ}{y}$ is the mean uptake rate; and ${\hat{y}}_{1}$ is the model-predicted uptake rate at the ith measurement.

Analytic Method for Choline and Metabolites

Metabolism of [³H]choline in PC-3 cells was analyzed as follows: After the cells were washed three times with ice-cold PBS, the metabolism was stopped by the addition of methanol, and then hydrophilic and lipophilic compounds were separated by water-methanol-chloroform extraction according to Folch and colleagues.³⁸ Quantitation of hydrophilic compounds (choline, betaine, phosphorylcholine, CDP-choline, and glycerophosphorylcholine) was performed by a high-performance liquid chromatography–mass spectrometry system (Varian, Palo Alto, CA) using a silica gel column and gradient elution according to the method of Koc and colleagues.³⁹ The retention time of authentic standards of the choline metabolites was determined using mass spectroscopy with electrospray ionization. For measurements of ³H-labeled metabolites, fractions of eluant were collected every 30 seconds and counted in a liquid scintillation counter.

Binding of [3H]HC-3 to PC-3 Cells In Vitro

The binding of the radiolabeled choline transporter inhibitor [N-methyl-³H]HC-3 (Figure 1) (Perkin Elmer Life Sciences) to PC-3 cells was examined in vitro by adding 2 μCi of [³H]HC-3 (specific activity > 120 μCi/nmol) to each well that contained 0.56 × 10⁶ cells in 1 mL of uptake buffer. The final concentration of HC-3 in the well was < 15 μM. The cell volume measured by the hematocrit method was 4.26 μL/well. The protein content was 0.33 mg/well. After incubation at 37°C for 10 and 60 minutes, cells were quickly washed once with 3 mL of PBS. The binding of [³H]HC-3 was expressed by cell to medium ratio. In a second group, cells were preincubated with 100 μM nonradioactive HC-3 for 5 minutes, and then 2 μCi of [³H]HC-3 was added.

Figure 1.
Chemical structure of HC-3.

Biodistribution of [3H]HC-3 and [Methyl-14C]Choline in 9L Glioma–Bearing Rats

To understand the in vivo disposition of HC-3 in a tumor-bearing animal model, the biodistribution of [³H]HC-3 was investigated in Fischer 344 rats (n = 5) bearing subcutaneous 9L gliomas. We chose against using a PC-3 xenograft mouse model for this study because a rat model would allow for more practical collection of arterial blood samples and previous work with the PC-3 xenografts showed low accumulations of choline radiotracers indicative of poor perfusion of these tumors.⁷ Since elevation of choline uptake is found in a wide variety of tumors, including gliomas,^19,35 the model serves as a representative model for evaluating choline-based imaging probes. For comparison with tracer choline behavior, a separate group of rats (n = 8) was similarly studied with [methyl-¹⁴C]choline. The radiotracers (≈300 KBq) were administered as an intravenous bolus to a tail vein, and the animal was euthanized at 20 minutes postinjection. Prior to injection, the carotid artery was cannulated and arterial blood samples were collected over the duration of the experiment. After euthanization, the major tissues and organs were dissected, weighed, digested (SOLVABLE, Perkin-Elmer Life Sciences), and counted for ³H- or ¹⁴C-radioactivity in a liquid scintillation counter. Uptake in each tissue was expressed as a percentage of the injected dose per gram of tissue.

Statistical Methods

Values are expressed as mean ± SD. The statistical significance of differences of tracer uptakes in PC-3 cells with different conditions was determined by Student's t-test, with p < .05 considered significant. Differences in biodistribution uptake values between different tracers were treated likewise.

Results

Kinetics of Choline Transport in PC-3 Cells In Vitro

The dependence of [³H]choline uptake by PC-3 cells on choline concentration is shown in Figure 2. The uptake velocity into the cells was fit to a Michaelis-Menten model that included both facilitative and nonfacilitative diffusion transport terms (Equation 2). The facilitative component showed intermediate affinity (K_M = 9.7 ± 0.8 μM) (Table 1). Transport by nonfacilitative diffusion was also evident and became dominant at choline concentrations greater than 200 μM. Facilitative transport was maintained after replacement of sodium ions with lithium ions in the culture medium. Although V_max was unchanged, the K_M value increased to 28.2 ± 2.7 μM, indicating that the uptake mechanism was partially sodium dependent (see Figure 2).

The effect of HC-3, a potent inhibitor of choline transporter in the brain (CHT1) (K_I, ≈0.02 μM)²³ and an inhibitor of choline kinase (K_I, ≈500 μM),³⁶ on the choline uptake was examined. The cell culture conditions were the same as above. HC-3 inhibited transporter-facilitated choline uptake in a concentration-dependent fashion (K_I = 10.5 ± 2.2 μM) but did not inhibit free diffusion at higher concentrations of choline (see Table 1 and Figure 3). The fitting of our kinetic data to the model of Michaelis-Menten kinetics, competitive inhibition by HC-3, and nonfacilitative diffusion (see Equation 2) gave r² values of near-unity (see Table 1) that indicated excellent goodness of fit. The resultant K_I suggested that the potency of HC-3 to inhibit choline transport in PC-3 cells is intermediate between the higher potency described for hCHT1 inhibition (≈1 μM) and the markedly lower potency for hOCT1 or hOCT2 inhibition (≈250 μM).^24,25

Figure 2.
Kinetics of [³H]choline uptake in PC-3 cells and effects of lithium-for-sodium replacement. A, Standard medium containing 141 mM sodium. B, Lithium medium where sodium was replaced with 141 mM lithium. Lines show the fit of the modified Michaelis-Menten equation (Equation 2) to the data. Fitted parameter estimates are summarized in Table 1.

Table 1.
Parameter Estimates of Nonlinear Least-Squares Fit of Equation 2 to [³H]Choline Uptake in Cultured PC-3 Cells

Nonfacilitative

Facilitative Transport Parameters
HC-3 Inhibition K_I
Diffusion Parameter D
Goodness of Fit

Condition V_max(nmol/mg prot./min) K_M (μM) (μM) × 10⁻⁴ (mL/mg prot./min) r²

Control 0.214 ± 0.005 9.70 ± 0.78 — 9.8 ± 0.2 .999

Li⁺-for-Na⁺ replacement 0.244 ± 0.102 28.15 ± 2.70* — 8.8 ± 0.2* .999

0–100 μM HC-3 0.214 ± 0.011 8.33 ± 1.62 10.5 ± 2.2 11.8 ± 0.1 .998

HC-3 = hemicholinium 3; K_I = enzyme inhibition constant; K_M = Michaelis constant; V_max = maximum velocity.

*
p < .05 versus control.

To further characterize the mechanism of choline transport, it was of interest to determine the sensitivity of choline transport rates to inhibition of Na/K-dependent adenosine triphosphatase (ATPase) using ouabain (0.1–1 mM) and inhibition of cellular adenosine triphosphate (ATP) production using the oxidative phosphorylation uncoupler dinitrophenol (0.1–1 mM) (Table 2). It was anticipated that a diminishing of the negative membrane potential with ouabain would cause a decrease in choline transport rate because the positively charged choline molecule travels in the same direction as the charge gradient across the cell membrane. A second method used to decrease the membrane potential was replacement of sodium ions with potassium ions in the culture medium (see Table 2). Metabolic inhibition with the uncoupler dinitrophenol was anticipated to also decrease cell membrane choline transport, either by decreasing the availability of intracellular ATP needed by an active transport mechanism or by decreased activity of the Na/K-dependent ATPase and the consequent effects of this inhibition mentioned above. PC-3 cells were subject to the inhibitors during a preincubation period of 5 minutes (unless noted otherwise) followed by an incubation period with [³H]choline for 10 minutes. Ouabain inhibited both transporter-facilitated uptake (5 μM choline) and free diffusion (500 μM choline). The inhibition was dependent on ouabain concentration and was somewhat greater in magnitude at the low concentration of choline. The sensitivity of choline uptake to ouabain is consistent with the inhibition of uptake of positively charged choline molecules as the inwardly directed negative force of the membrane potential is diminished. Choline uptake was also decreased by replacement of sodium ions with potassium ions in the culture medium, although there was a slower onset of this effect than for ouabain: potassium-for-sodium replacement for a 60-minute preincubation period resulted in approximately the same level of decrease of choline uptake as 0.1 mM ouabain for a 5-minute preincubation period.

Figure 3.
Inhibition of [³H]choline uptake in PC-3 cells by hemicholinium 3 (HC-3). Lines show the fit of the modified Michaelis-Menten equation (Equation 2) to the data. Fitted parameter estimates are summarized in Table 1. Solid circles = control; empty squares = 10 μM HC-3; solid triangles = 50 μM HC-3; empty diamonds = 100 μM HC-3.

Table 2.
Effect of Other Inhibitors on the Uptake of [³H]Choline in Cultured PC-3 Cells

Choline Uptake (nmol/mg protein/10 min)

Incubation Medium At 5 μM Choline At 500 μM Choline

Control medium 0.96 ± 0.03 9.04 ± 0.43

+ 0.1 mM ouabain 0.49 ± 0.01* 6.39 ± 0.15*

+ 1 mM ouabain 0.058 ± 0.009* 3.14 ± 0.16*

Na⁺ replaced by K⁺, 0.70 ± 0.07* 8.58 ± 0.14

Na⁺ replaced by K⁺, after preincubation for 60 min 0.48 ± 0.04* 6.51 ± 0.18*

+ 0.1 mM 2,4-dinitrophenol 0.87 ± 0.05* 8.06 ± 0.19*

+ 1 mM 2,4-dinitrophenol 0.77 ± 0.05* 7.37 ± 0.22*

Control medium: 10mM HEPES/Tris buffer (pH 7.4), 141 mM NaCl, 4 mM KCl, 2.8 mM CaCl₂, 1 mM MgCl₂, 10 mM glucose.

*
p > .05 versus control.

Metabolic inhibition with the oxidative phosphorylation uncoupler dinitrophenol resulted in a concentration-dependent decrease in choline uptake, but the differences were relatively smaller than the inhibition with ouabain. These data confirmed the importance of a facilitative transport process for radiolabeled choline uptake into tumor cells that was most sensitively inhibited with a reduction in Na/K-dependent ATPase activity by ouabain.

Uptake and Metabolism of [3H]Choline in PC-3 Cells In Vitro

The time course of [³H]choline uptake was examined after the addition of 5 or 500 μM [³H]choline (Table 3). Figure 4 shows the distribution of radioactivity among the most prevalent choline metabolites as a function of incubation time. At low extracellular choline concentration, the intracellular radiolabel was primarily found as [³H]PCho, indicating rapid phosphorylation of transported [³H]choline (see Figure 4). The ratio of [³H]PCho to [³H]choline increases from approximately 7 to 24 over the interval of 10 to 60 minutes. At a low extracellular choline concentration of 5 μM, the cell to medium ratio increased from 11.3 ± 0.2 at 10 minutes to 39.8 ± 1.6, demonstrating the concentration of radiolabeled choline within cells resulting from intracellular phosphorylation. There was significant but minimal (> 4% over 60 minutes) metabolism of PCho to CDP-choline and lipophilic species. No radiolabeled betaine was detected in the cells or in the culture medium. At a high extracellular choline concentration of 500 μM, total choline uptake is approximately 10-fold increased owing to increased levels of nonfacilitative diffusion (see Figure 2). In this condition, the fraction of radiolabel found as nonmetabolized choline (≈50% at 10 minutes; see Figure 4) is markedly higher and the levels of PCho (≈46% at 10 minutes) are correspondingly lower than at a low extracellular choline concentration, demonstrating a flux limitation of choline kinase at the supraphysiologic extracellular choline concentration. The limited capacity of the cells to metabolically concentrate choline was also noted from cell to medium ratios that were approximately 10-fold lower than those at low extracellular choline concentration, as well as [³H]PCho to [³H]choline ratios that were similarly diminished. The limited velocity (V_max) of choline kinase resulted in insufficient phosphorylation of choline, leaving a large fraction of [³H]choline unphosphorylated. In this case, free [³H]choline (with positive charge) was retained within the cell, presumably by virtue of the negative charge of intracellular space maintained by the membrane potential. Radioactivity fractions in the CDP-choline fraction (4–5%) were somewhat increased relative to the low extracellular choline condition, possibly reflecting activation of choline-cytidyltransferase, whereas the fraction of radioactivity incorporating into lipids remained unchanged.

Figure 4.
Distribution of radioactivity among choline metabolites in PC-3 cell extracts as a function of extracellular choline concentration (A, 5 μM, or B, 500 μM) and incubation period with [³H]choline. Diamonds = choline; circles = phosphocholine; squares = cytidine diphosphate–choline; triangles = lipophilic metabolites.

Table 3.
[³H]Choline Uptake in PC-3 Cells

Time (min)

Choline 10 20 40 60

5 μM

Total [³H]choline uptake (nmol/mg protein) 0.76 ± 0.02 1.42 ± 0.04 2.28 ± 0.05 2.79 ± 0.1

Cell to medium ratio 11.3 ± 0.2 20.2 ± 0.6 32.5 ± 0.8 39.8 ± 1.6

500 μM

Total [³H]choline uptake (nmol/mg protein) 6.09 ± 0.11 9.80 ± 0.29 17.54 ± 0.65 23.39 ± 0.57

Cell to medium ratio 0.94 ± 0.02 1.51 ± 0.04 2.72 ± 0.10 3.62 ± 0.09

The cells were incubated with 5 or 500 μM [³H]choline.

Binding of [3H]HC-3 to PC-3 Cells In Vitro

The binding of the radiolabeled choline transporter inhibitor [³H]HC-3 to PC-3 cells was examined in vitro. The cell to medium ratio was 3.64 ± 0.49 after 10 minutes of incubation and did not change after 60 minutes of incubation (3.67 ± 0.33). This suggested that a binding equilibrium had been reached within 10 minutes. To determine the saturability of the binding process, [³H]HC-3 binding was evaluated in the presence of nonradioactive HC-3 (100 μM). The cell to medium-ratio at 10 minutes of incubation was reduced to 0.72 ± 0.19 (p < .001) in the presence of unlabeled HC-3, demonstrating saturability of binding. The cell to medium ratio was lower for [³H]HC-3 than for [³H]choline at all time points (see Table 3), consistent with the fact that [³H]HC-3 lacks a metabolical trapping mechanism to concentrate radioactivity in the cells.

Biodistribution of [3H]HC-3 in 9L Glioma–Bearing Rats

Blood clearance of radioactivity was rapid in the first 4 minutes following intravenous injection of [³H]HC-3 in the rat (Figure 5). Slower blood clearance is seen thereafter. The biodistribution data are shown in Table 4 in comparison with data from [¹⁴C]choline in the same experimental model. The ranking of uptake was found to be kidney > lung > tumor > atrioventricular node > liver > skeletal muscle ≈ blood. In comparison with [¹⁴C]choline, tritiated HC-3 showed over twofold higher tumor uptake (p < .05). The uptake ratios of tumor to blood, tumor to muscle, tumor to lung, tumor to liver, and tumor to kidney were trended higher for [³H]HC-3 relative to [¹⁴C]choline, although high intersubject variability precluded statistical significance of the differences. Uptake of HC-3 in the brain was very low, whereas excretion of radioactivity in urine was relatively high.

Figure 5.
Representative time course of radioactivity concentration in arterial plasma following bolus injection of [³H]HC-3 in a rat. The clearance pattern is biphasic, with an early rapid clearance phase within the first 5 minutes, followed by slow clearance thereafter.

Table 4.
Biodistribution of [³H]HC-3 and [¹⁴C]Choline in a 9L Glioma–Bearing Fischer Rat Model

Organ [³H]HC-3 (n = 5) [¹⁴C]Choline(n = 8)

Blood 0.133 ± 0.070 0.136 ± 0.076

Tumor 0.395 ± 0.176* 0.179 ± 0.098

Skeletal muscle 0.149 ± 0.030 0.121 ± 0.081

Liver 0.481 ± 0.144* 4.09 ± 1.78

Kidney 5.10 ± 2.85 7.43 ± 3.81

Lung 1.03 ± 0.34 1.36 ± 0.50

Brain 0.074 ± 0.035 0.063 ± 0.047

Heart regions

L atrium 0.339 ± 0.164

R atrium 0.401 ± 0.116

AV node 0.368 ± 0.143

Septum 0.298 ± 0.101

L ventricle 0.315 ± 0.135 0.595 ± 0.397

R ventricle 0.275 ± 0.109

Urine 36.9 ± 16.9* 0.01 ± 0.01

(% dose) (% dose)

Tumor:blood ratio 4.6 ± 3.7 1.9 ± 1.6

Tumor:liver 0.86 ± 0.35* 0.05 ± 0.033

Tumor:muscle 2.8 ± 1.7 1.03 ± 0.97

AV = atrioventricular; HC-3 = hemicholinium-3.

Values are % injected dose/g of each tissue at 20 minutes postadministration of tracer.

*
p < .05 versus [¹⁴C]choline.

Discussion

There is currently intense interest in the development of both PET and MRS imaging techniques to differentiate and localize malignancies based on abnormalities of choline processing in cancer cells. The present finding of rapid formation and persistent levels of [³H]PCho in PC-3 prostate cancer cells incubated with [³H]choline confirms the general consensus that is forming among choline imaging researchers that PCho accumulation in a variety of tumor cells is a hallmark of malignancy (recently reviewed in Gillies and Morse¹). Elevated PCho levels may result from one or more of the following changes:
Increased rate of extracellular choline uptake and phosphorylation by choline kinase

Decreased phosphatase activity acting on PCho

Decreased rate of metabolism of choline to CDP-choline by CTP to PCho cytidylyltransferase

Increased rate of phospholipase C catalyzed breakdown of phosphatidylcholine to PCho and diacylglycerol

Increased rate of phospholipase D catalyzed breakdown of phosphatidylcholine to choline and phosphatidic acid, followed by choline phosphorylation by choline kinase

Increased breakdown of phosphatidylcholine to choline through successive metabolism by phospholipase A₁, phospholipase A₂, and phosphodiesterase, followed by choline phosphorylation by choline kinase

Decreased oxidation of intracellular choline to betaine

Of the seven factors listed above, only the third can be considered implausible based on evidence showing that CTP to PCho cytidylyltransferase activity is tightly regulated to maintain membrane homeostasis.^40,41 The last factor regarding decreased choline oxidation may be relevant only if the normal tissue from which the tumor originated has significant levels of mitochondrial choline dehydrogenase and betaine aldehyde dehydrogenase, as would be the case for hepatocellular and renal carcinomas. The remaining five factors are considered to be plausible explanations for elevated PCho levels in neoplasms. Detailed tracer kinetic studies in breast cancer^17,18,20 and MRS studies in glioma cells¹⁹ indicate that the majority of accumulated PCho in cancer cells is derived from extracellular choline followed by phosphorylation. Also, a number of studies show a significant contribution of phospholipases to the elevation of PCho levels in cancer cells.^1,42–44 Although phospholipid turnover rates were not measured in the present study in PC-3 prostate cancer cells, the rapid accumulation of [³H]PCho following incubation with [³H]choline also suggests the importance of uptake and phosphorylation. An increase in the extracellular choline concentration from 5 to 500 μM resulted in a 10-fold increase in total [³H]choline uptake by the cells, whereas the [³H]PCho contribution increased approximately 4-fold at the earliest measurement point of 10 minutes (see Table 3). At the same time point, the majority of the increased total [³H]choline uptake was in the form of nonmetabolized tracer (see Figure 4) that was presumably retained within the cytoplasm by virtue of its positive charge in the presence of a negative membrane potential. This fraction was largely phosphorylated over the following hour but at a much slower rate than at 5 μM extracellular choline concentration. Thus, choline phosphorylation increased dramatically as the extracellular choline concentration was raised, yet there was an even greater enhancement of choline transport attributable to increased rate of transport by nonfacilitative diffusion (see Figure 2). From our data, the contribution of nonfacilitative diffusion to the overall transport rate is indicated to be minor (< 25%) at physiologic concentrations of choline in the plasma (5–50 μM).

Rather, choline transport rates in vivo would be expected to be governed by plasma choline levels, diffusion in the extracellular space, and expression and activity of the choline transporter(s). Importantly, this study demonstrates that at physiologic plasma choline concentrations, the capacity of choline kinase to phosphorylate choline was several-fold higher than the transport rate. Transport was a rate-determining step for tracer choline accumulation in the cells. Katz-Brull and colleagues estimated that the rate of choline phosphorylation was more than two orders of magnitude faster than the rate of its transport in cultured breast cancer cells, maintaining the ratio of concentrations of PCho to choline higher than 100.¹⁸

The significance of the role of choline transport also follows from the success of PET imaging of various cancers with ¹¹C- and ¹⁸F-labeled cholines despite an extremely rapid blood clearance of tracer (< 2 minutes).^2,5–7,22 We previously noted that the adequacy of uptake of FCH to allow visualization in PET images of primary and metastatic prostate cancer tumors in the presence of rapid blood clearance of tracer suggests a high dependence of tumor uptake on tumor perfusion and efficient transport from the blood into tumor cells.⁷ Intracellular phosphorylation is unlikely to be rate determining, although it plays an essential role in sequestering the PET choline tracers in the cell to provide high tumor to normal tissue contrast for imaging purposes. The radioactivity concentration in tumor at later times (> 10 minutes postinjection) may reflect intracellular metabolic events. For example, it has been speculated that the maintenance of radioactivity concentration in primary prostatic tumors in the presence of a slow clearance of radioactivity in neighboring normal prostate tissue may reflect lower prostatic phosphatase activity in tumor.⁵

Given the influence of the choline transporter(s) in tumor processing of choline and the important role it therefore plays in diagnostic imaging studies of choline metabolism with MRS and PET, there is an obvious need to identify and characterize this transporter in human neoplasms. Yet very little is known about the choline transporter(s) in human tumors. From the scant literature data available,^28–35 there may be significant differences in the transport characteristics in various tumor cell lines. The present study indicates that choline enters PC-3 prostate cancer cells via a carrier-mediated mechanism since it fulfills a number of well-established criteria.⁴⁵ Choline transport in PC-3 cells has a saturable, facilitative component apparent at low extracellular choline concentrations (> 50 μM) and a nonfacilitative diffusion component that becomes quantitatively important only at higher choline concentrations (> 50 μM). The facilitative transport system has the characteristics of intermediate affinity (K_M ≈10 μM) and partial sodium dependence, distinguishing it from the high-affinity, sodium-dependent system of the central nervous system (hCHT1) and the low-affinity, sodium-independent systems (hOCTs) in liver and kidney. There are substantial differences reported in the K_M values for choline transport in various cancer cell lines, although, in general, the K_M values are intermediate between those of hCHT1 and the hOCT family of transporters. The potency of HC-3 to inhibit choline uptake may also serve as a means to functionally characterize the transporter. HC-3 inhibits choline transport in PC-3 cells (K_I ≈ 10 μM), with a substantially higher potency than for choline transport in Krebs II ascites cells (K_I = 160 μM). HC-3's potency in PC-3 cells stands between that for hCHT1 and the hOCT family.^24,25

Choline transport is further characterized with respect to its sensitivity to inhibitors. In our study, we followed the approach Ribbes and colleagues used to characterize choline transport in Krebs II ascites cells.³¹ As found for ascites cells, ouabain, an inhibitor of Na/K-dependent ATPase, caused a concentration-dependent reduction in choline accumulation in the PC-3 cells. Our interpretation of this effect is that reductions in the amplitude of the negative membrane potential caused by ouabain results in decreased influx (via facilitative transport and nonfacilitative diffusion) and/or increased efflux of choline cations from the cell. Replacement of sodium with potassium caused modest reductions in choline accumulation in PC-3 cells, consistent with previous findings in Krebs II ascites cells.³¹ However, dinitrophenol, an oxidative phosphorylation uncoupler, caused modest reductions in choline uptake in our PC-3 cells but no change in ascites cells.³¹ As a whole, the data show that at physiologic concentrations of extracellular choline, the choline transport process in PC-3 cells is dominated by a carrier-mediated mechanism. The kinetic properties of the choline transporter(s) in PC-3 cells are distinct from those on neurons (hCHT1) and the known hOCT family of polyspecific organic cation transporters in liver and kidney. There are a number of similarities of characteristics of the choline transporter(s) in PC-3 cells with what is known of a newly discovered family of choline transporter–like (CTL) proteins (hCTL1–5) present in a variety of human tissues.²⁷ More extensive molecular and functional characterization studies have been done with CTL transporters in mice and rats,²⁷ showing CTL transporters to be sodium-independent (although there is some disagreement in the literature regarding the sodium dependency of hCTL), intermediate-affinity transporters of choline that can be completely inhibited by HC-3. Clearly, further work is required to genetically identify and characterize the choline transporter(s) in human cancers. As a first step, it will be important to evaluate the transport mechanism(s) in an androgen-sensitive prostate cancer cell line, such as LNCaP.

Given the important role of the choline transporter(s) in certain tumors, we proposed to develop molecular imaging probes that bind to the choline transporter(s) as potential tumor imaging agents for single-photon emission computed tomography (SPECT) or PET. It is anticipated that a noninvasive method of monitoring the expression and membrane presentation of the choline transporter(s) could be developed with an appropriate probe and corresponding imaging system. To test the feasibility of this approach, experiments were first performed with commercially available [³H]HC-3. HC-3 is a commonly employed high-affinity inhibitor of sodium-dependent choline transport of cholinergic neurons in the central nervous system (hCHT1). Its potency is dramatically lower for inhibition of hOCT family–mediated choline transport in the liver and kidney.^24,25 Previous studies in our laboratory showed that ¹⁸F-labeled choline uptake by PC-3 cells was 90% blocked by the addition of HC-3 in the culture medium, demonstrating a significant interaction of HC-3 with the choline transporter(s).⁷ In the same study, HC-3 had no effect on uptake of the glucose analogue fluorodeoxyglucose ([¹⁸F]FDG), demonstrating specificity of its action. Our preliminary experiments in cultured PC-3 cells demonstrated rapid binding of [³H]HC-3 that competed with unlabeled HC-3. An equilibrium was reached within 10 minutes of incubation. The lack of increase over the subsequent hour indicates that there is negligible cellular accumulation, at least within this time frame. Because of its bisquarternary structure, HC-3 is a highly polar molecule that would be expected to have slow diffusion across cell membranes. Although HC-3 has been shown to weakly inhibit choline kinase in in vitro enzyme assays,³⁶ its ability to dramatically lower PCho levels in cultured cancer cells³⁶ or xenografts following administration in host animals³⁶ may result from inhibition of choline transport in addition to potential effects on choline kinase activity. The much greater potency of HC-3 for inhibition of transport (K_I ≈10 μM) relative to choline kinase (K_I ≈500 μM) and the limitation of HC-3 permeation into cells argue for the mechanism of action being primarily at the transport step.

The biodistribution study in the 9L glioma–bearing Fischer rat model demonstrated moderate uptake of [³H]HC-3 in tumor and a mean tumor to blood concentration ratio of 4.6 at 20 minutes postinjection. Tumor uptake was significantly higher for [³H]HC-3 than for [¹⁴C]choline in the same model. The organs with highest uptake for both radiotracers were the liver and kidney. However, the normal uptake in the liver with [³H]HC-3 was approximately one-tenth that of [¹⁴C]choline. If a similar relationship is found in humans, positron-labeled HC-3 analogues may have a distinct advantage over radiolabeled choline analogues for evaluating liver lesions with PET. Liver lesions are not common with prostate cancer; they are more relevant to other cancer types, including breast cancer, colorectal cancers, and hepatocellular carcinoma. The higher hepatic accumulation of [¹⁴C]choline is likely due to its concentration in tissue through metabolic sequestration in the cells. [³H]HC-3 presumably lacks this property. The higher urinary excretion of [³H]HC-3 points to a potential disadvantage of a SPECT or PET probe of prostate cancer imaging based on HC-3, but the confounding of images by urinary radioactivity can be minimized by hydration of the patient prior to injection and voiding of the bladder prior to the commencement of the radiologic scan. This is a common practice in many PET centers for evaluation of pelvic tumors using the highly excreted glucose analogue [¹⁸F]FDG.

The biodistribution data corroborated previous work by Domino and colleagues, who performed whole-body autoradiography of mice at 10 minutes after intravenous administration of [methyl-¹⁴C]HC-3.⁴⁶ In their study, the highest radioactivity concentrations in the whole body were found in the solar plexus and bone marrow, organs not sampled in our study. The solar plexus (celiac plexus) is a great cluster of autonomic nerve ganglia located behind the stomach and contains a high density of presynaptic high-affinity choline transporters. Low accumulation of radioactivity in the brain (also rich in high-affinity choline transporters) indicates minimal transport of radiolabeled HC-3 across the blood-brain barrier. The high radioactivity in the bone marrow in the vertebrae and in the ribs represented the high uptake of [¹⁴C]HC-3 in rapidly proliferating hematopoietic cells. It will be important to follow up these observations in tumor-bearing animal studies of other cancer types, including prostate cancer and breast cancer.

These data support a novel approach to development of molecular imaging probes for cancer based on binding of an HC-3 analogue (or another choline transporter binding molecule) to the choline transporter(s). The chemical structure of HC-3 is amenable to labeling of its biphenyl moiety with radiohalogens or radiometal chelator conjugates for SPECT or PET, whereas the quarternary ammonium groups can be labeled with ¹¹C-methyl or ¹⁸F-methyl groups for PET. Labeling with optically active dyes could also be explored for use with optical imaging modalities. The potential for changes in the biochemical behavior caused by the addition of a different labeling group must be considered in designing and evaluating new HC-3 analogue tracers. Furthermore, the potential for mass effects at saturable choline transporter target sites and toxicity will place minimum limits on acceptable specific activities for the tracers. HC-3 has a median lethal dose of 0.12 mg/kg in mice, although slightly higher toxicities were seen with several HC-3 structural analogues.⁴⁷ Because of the high imaging sensitivity of PET, the mass levels associated with administration of PET probes (on the order of 1 μg per dose) will be much less than these indicated toxicity levels. Our laboratory is now developing ¹¹C- and ¹⁸F-labeled derivatives of HC-3 for evaluation as PET agents.

Conclusion

The facilitative transport system for choline in cultured PC-3 human prostate cancer cells was characterized by intermediate affinity (K_M = 9.7 ± 0.8 μM), partial sodium dependence, and an intermediate inhibition potency of HC-3 (K_I = 10.5 ± 2.2 μM). These characteristics distinguish it from the neuronal high-affinity, sodium-dependent choline transport system (hCHT1) and the low-affinity, sodium-independent choline transport systems (hOCT family) present in non-neuronal tissues such as the kidney and liver. Further work is needed to characterize the choline transport system in each tumor type and determine its possible relatedness to the newly discovered hCTL family of intermediate-affinity, sodium-independent choline transporters.²⁰ The high dependence of choline processing in tumors on the transport process suggests the choline transporter(s) as a novel target for molecular imaging. Preliminary evaluation studies of the in vivo distribution of the choline transport inhibitor [³H]HC-3 in a 9L glioma–bearing rat model demonstrated significant binding of the radiotracer in tumor and favorable tumor to muscle ratios. The results anticipate the development of a variety of appropriately labeled analogues of HC-3 (or other choline transporter binding molecules) as molecular probes for noninvasive imaging of choline transporter(s) in tumors using a variety of available molecular imaging modalities.

				Nonfacilitative
Control	0.214 ± 0.005	9.70 ± 0.78	—	9.8 ± 0.2	.999
Li⁺-for-Na⁺ replacement	0.244 ± 0.102	28.15 ± 2.70*	—	8.8 ± 0.2*	.999
0–100 μM HC-3	0.214 ± 0.011	8.33 ± 1.62	10.5 ± 2.2	11.8 ± 0.1	.998

	Choline Uptake (nmol/mg protein/10 min)
Control medium	0.96 ± 0.03	9.04 ± 0.43
+ 0.1 mM ouabain	0.49 ± 0.01*	6.39 ± 0.15*
+ 1 mM ouabain	0.058 ± 0.009*	3.14 ± 0.16*
Na⁺ replaced by K⁺,	0.70 ± 0.07*	8.58 ± 0.14
Na⁺ replaced by K⁺, after preincubation for 60 min	0.48 ± 0.04*	6.51 ± 0.18*
+ 0.1 mM 2,4-dinitrophenol	0.87 ± 0.05*	8.06 ± 0.19*
+ 1 mM 2,4-dinitrophenol	0.77 ± 0.05*	7.37 ± 0.22*

	Time (min)
5 μM
Total [³H]choline uptake (nmol/mg protein)	0.76 ± 0.02	1.42 ± 0.04	2.28 ± 0.05	2.79 ± 0.1
Cell to medium ratio	11.3 ± 0.2	20.2 ± 0.6	32.5 ± 0.8	39.8 ± 1.6
500 μM
Total [³H]choline uptake (nmol/mg protein)	6.09 ± 0.11	9.80 ± 0.29	17.54 ± 0.65	23.39 ± 0.57
Cell to medium ratio	0.94 ± 0.02	1.51 ± 0.04	2.72 ± 0.10	3.62 ± 0.09

Organ	[³H]HC-3 (n = 5)	[¹⁴C]Choline(n = 8)
Blood	0.133 ± 0.070	0.136 ± 0.076
Tumor	0.395 ± 0.176*	0.179 ± 0.098
Skeletal muscle	0.149 ± 0.030	0.121 ± 0.081
Liver	0.481 ± 0.144*	4.09 ± 1.78
Kidney	5.10 ± 2.85	7.43 ± 3.81
Lung	1.03 ± 0.34	1.36 ± 0.50
Brain	0.074 ± 0.035	0.063 ± 0.047
Heart regions
L atrium	0.339 ± 0.164
R atrium	0.401 ± 0.116
AV node	0.368 ± 0.143
Septum	0.298 ± 0.101
L ventricle	0.315 ± 0.135	0.595 ± 0.397
R ventricle	0.275 ± 0.109
Urine	36.9 ± 16.9*	0.01 ± 0.01
	(% dose)	(% dose)
Tumor:blood ratio	4.6 ± 3.7	1.9 ± 1.6
Tumor:liver	0.86 ± 0.35*	0.05 ± 0.033
Tumor:muscle	2.8 ± 1.7	1.03 ± 0.97

Footnotes

Acknowledgments

We wish to thank Shuyan Wang for her assistance with the biodistribution studies. We are grateful to Dr. Robert A. Harris for his helpful comments on the manuscript.

References

1.
Gillies RJ Morse DL . In vivo magnetic resonance spectroscopy in cancer. Annu Rev Biomed Eng 2005;7:287–326.

2.
Hara T Kosaka N Shinoura N Kondo T . PET imaging of brain tumor with [methyl-¹¹C]choline. J Nucl Med 1997;38:842–7.

3.
Moller-Hartmann W Herminghaus S Krings T . Clinical application of proton magnetic resonance spectroscopy in the diagnosis of intracranial mass lesions. Neuroradiology 2002;44:371–81.

4.
Kumar R Kumar M Jagannathan NR . Proton magnetic resonance spectroscopy with a body coil in the diagnosis of carcinoma prostate. Urol Res 2004;32:36–40.

5.
Kwee SA Wei H Sesterhenn I . Localization of primary prostate cancer with dual-phase ¹⁸F-fluorocholine PET. J Nucl Med 2006;47:262–9.

6.
Hara T Kosaka N Suzuki T . Uptake rates of ¹⁸F-fluorodeoxyglucose and ¹¹C-choline in lung cancer and pulmonary tuberculosis: a positron emission tomography study. Chest 2003;124:893–901.

7.
DeGrado TR Coleman RE Wang S . Synthesis and evaluation of ¹⁸F-labeled choline as an oncologic tracer for positron emission tomography: initial findings in prostate cancer. Cancer Res 2001;61:110–7.

8.
Yoshida S Nakagomi K Goto S . ¹¹C-Choline positron emission tomography in prostate cancer: primary staging and recurrent site staging. Urol Int 2005;74:214–20.

9.
de Jong IJ Pruim J Elsinga PH . Preoperative staging of pelvic lymph nodes in prostate cancer by ¹¹C-choline PET. J Nucl Med 2003;44:331–5.

10.
Lichy MP Bachert P Henze M . Monitoring individual response to brain-tumour chemotherapy: proton MR spectroscopy in a patient with recurrent glioma after stereotactic radiotherapy. Neuroradiology 2004;46:126–9.

11.
de Jong IJ Pruim J Elsinga PH . ¹¹C-choline positron emission tomography for the evaluation after treatment of localized prostate cancer. Eur Urol 2003;44:32–8.

12.
Ackerstaff E Pflug BR Nelson JB Bhujwalla ZM . Detection of increased choline compounds with proton nuclear magnetic resonance spectroscopy subsequent to malignant transformation of human prostatic epithelial cells. Cancer Res 2001;61:3599–603.

13.
Mori N Delsite R Natarajan K . Loss of p53 function in colon cancer cells results in increased phosphocholine and total choline. Mol Imaging 2004;3:319–23.

14.
Smith TA Eccles S Ormerod MG . The phosphocholine and glycerophosphocholine content of an oestrogen-sensitive rat mammary tumour correlates strongly with growth rate. Br J Cancer 1991;64:821–6.

15.
Singer S Souza K Thilly WG . Pyruvate utilization, phosphocholine and adenosine triphosphate (ATP) are markers of human breast tumor progression: a ³¹P- and ¹³C-nuclear magnetic resonance (NMR) spectroscopy study. Cancer Res 1995;55:5140–5.

16.
Evelhoch JL Keller NA Corbett TH . Response-specific Adriamycin sensitivity markers provided by in vivo ³¹P nuclear magnetic resonance spectroscopy in murine mammary adenocarcinomas. Cancer Res 1987;47:3396–401.

17.
Katz-Brull R Seger D Rivenson-Segal D . Metabolic markers of breast cancer: enhanced choline metabolism and reduced choline-ether-phospholipid synthesis. Cancer Res 2002;62:1966–70.

18.
Katz-Brull R Degani H . Kinetics of choline transport and phosphorylation in human breast cancer cells: NMR application of the zero trans method. Anticancer Res 1996;16:1375–80.

19.
Gillies RJ Barry JA Ross BD . In vitro and in vivo ¹³C and ³¹P NMR analyses of phosphocholine metabolism in rat glioma cells. Magn Reson Med 1994;32:310–8.

20.
Daly PF Lyon RC Faustino PJ Cohen JS . Phospholipid metabolism in cancer cells monitored by ³¹P NMR spectroscopy. J Biol Chem 1987;262:14875–8.

21.
Katz-Brull R Margalit R Degani H . Differential routing of choline in implanted breast cancer and normal organs. Magn Reson Med 2001;46:31–8.

22.
DeGrado TR Balwin SW Wang S . Synthesis and evaluation of ¹⁸F-labeled choline analogs as oncologic PET tracers. J Nucl Med 2001;42:1805–14.

23.
Okuda T Haga T Kanai Y . Identification and characterization of the high-affinity choline transporter. Nat Neurosci 2000;3:120–5.

24.
Zhang L Dresser MJ Gray AT . Cloning and functional expression of a human liver organic cation transporter. Mol Pharmacol 1997;51:913–21.

25.
Gorboulev V Ulzheimer JC Akhoundova A . Cloning and characterization of two human polyspecific organic cation transporters. DNA Cell Biol 1997;16:871–81.

26.
Schloss P Mayser W Niehuis A Betz H . Na⁺-dependent high-affinity uptake of choline into cultured fibroblasts. Biochem Biophys Res Commun 1994;199:1320–5.

27.
Michel V Yuan Z Ramsubir S Bakovic M . Choline transport for phospholipids synthesis. Exp Biol Med 2006;231:490–504.

28.
Lanks K Somers L Papirmeister B Yamamura H . Choline transport by neuroblastoma cells in tissue culture. Nature 1974;252:476–8.

29.
Yorek MA Dunlap JA Spector AA Ginsberg BH . Effect of ethanolamine on choline uptake and incorporation into phosphatidylcholine in human Y79 retinoblastoma cells. J Lipid Res 1986;27:1205–13.

30.
Plagemann PGW . Choline metabolism and membrane formation in rat hepatoma cells grown in suspension culture. III. Choline transport and uptake by simple diffusion and lack of direct exchange with phosphatidylcholine. J Lipid Res 1971;12:715–24.

31.
Ribbes G Hamza M Chap H Douste-Blazy L . Carrier-mediated choline uptake by Krebs II ascites cells. Biochim Biophys Acta 1985;818:183–90.

32.
Lloveras J Hamza M Chap H Douste-Blazy L . Action of hemicholinium-3 on phospholipids metabolism in Krebs II ascites cells. Biochem Pharmacol 1985;34:3987–93.

33.
Haeffner EW . Studies on choline permeation through the plasma membrane and its incorporation into phosphatidylcholine in Ehrlich-Lettre-ascites tumor cells in vitro. Eur J Biochem 1975;51:219–28.

34.
Doppler W Hoffmann J Maly K Grunicke H . Amiloride and 5-N,N-dimethylamiloride inhibit the carrier-mediated uptake of choline in Ehrlich ascites tumor cells. Biochem Pharmacol 1987;36:1645–9.

35.
Walum E . Uptake and release of choline in cultures of human glioma cells. Cell Mol Neurobiol 1981;1:389–99.

36.
Hernandez-Alcoceba R Saniger L Campos J . Choline kinase inhibitors as a novel approach for antiproliferative drug design. Oncogene 1997;15:2289–301.

37.
Marquardt DW . An algorithm for least squares estimation of nonlinear parameters. J Soc Industrial Appl Math 1963;2:431–41.

38.
Folch J Lees M Sloane Stanley GH . A simple method for the isolation and purification of total lipides from animal tissues. J Biol Chem 1957;226:497–509.

39.
Koc H Mar MH Ranasinghe A . Quantitation of choline and its metabolites in tissues and foods by liquid chromatography/electrospray ionization-isotope dilution mass spectrometry. Anal Chem 2002;74:4734–40.

40.
Wright PS Morand JN Kent C . Regulation of phosphatidylcholine biosynthesis in Chinese hamster ovary cells by reversible membrane association of CTP:phosphocholine cytidylyltransferase. J Biol Chem 1985;260:7919–26.

41.
Clement JM Kent C . CTP:phosphocholine cytidylyltransferase: insights into regulatory mechanisms and novel functions. Biochem Biophys Res Commun 1999;257:643–50.

42.
Iorio E Mezzanzanica D Alberti P . Alterations of choline phospholipids metabolism in ovarian tumor progression. Cancer Res 2005;65:9369–76.

43.
Negendank WG Padavic-Shaller KA Li CW . Metabolic characterization of human non-Hodgkin's lymphomas in vivo with the use of proton-decoupled phosphorous magnetic resonance spectroscopy. Cancer Res 1995;55:3286–94.

44.
Glunde K Jie C Bhujwalla ZM . Molecular causes of the aberrant choline phospholipid metabolism in breast cancer. Cancer Res 2004;64:4270–6.

45.
Wigler PW . Trans effects in cell membrane transport. J Theor Biol 1981;93:997–1008.

46.
Domino EF Cassino GB Placidi G-F . Autoradiographic distribution of ¹⁴C-hemicholinium-3 in mouse whole body and dog brain. J Pharmacol Exp Ther 1974;188:77–85.

47.
Benz FW Long JP . Investigations on a series of heterocyclic hemicholinium-3 analogs. J Pharmacol Exp Ther 1969;166:225–36.

				Nonfacilitative
	Facilitative Transport Parameters		HC-3 Inhibition K_I	Diffusion Parameter D	Goodness of Fit
Condition	V_max(nmol/mg prot./min)	K_M (μM)	(μM)	× 10⁻⁴ (mL/mg prot./min)	r²
Control	0.214 ± 0.005	9.70 ± 0.78	—	9.8 ± 0.2	.999
Li⁺-for-Na⁺ replacement	0.244 ± 0.102	28.15 ± 2.70*	—	8.8 ± 0.2*	.999
0–100 μM HC-3	0.214 ± 0.011	8.33 ± 1.62	10.5 ± 2.2	11.8 ± 0.1	.998

	Choline Uptake (nmol/mg protein/10 min)
Incubation Medium	At 5 μM Choline	At 500 μM Choline
Control medium	0.96 ± 0.03	9.04 ± 0.43
+ 0.1 mM ouabain	0.49 ± 0.01*	6.39 ± 0.15*
+ 1 mM ouabain	0.058 ± 0.009*	3.14 ± 0.16*
Na⁺ replaced by K⁺,	0.70 ± 0.07*	8.58 ± 0.14
Na⁺ replaced by K⁺, after preincubation for 60 min	0.48 ± 0.04*	6.51 ± 0.18*
+ 0.1 mM 2,4-dinitrophenol	0.87 ± 0.05*	8.06 ± 0.19*
+ 1 mM 2,4-dinitrophenol	0.77 ± 0.05*	7.37 ± 0.22*

	Time (min)
Choline	10	20	40	60
5 μM
Total [³H]choline uptake (nmol/mg protein)	0.76 ± 0.02	1.42 ± 0.04	2.28 ± 0.05	2.79 ± 0.1
Cell to medium ratio	11.3 ± 0.2	20.2 ± 0.6	32.5 ± 0.8	39.8 ± 1.6
500 μM
Total [³H]choline uptake (nmol/mg protein)	6.09 ± 0.11	9.80 ± 0.29	17.54 ± 0.65	23.39 ± 0.57
Cell to medium ratio	0.94 ± 0.02	1.51 ± 0.04	2.72 ± 0.10	3.62 ± 0.09