5-(2-18F-Fluoroethoxy)-L-Tryptophan as a Substrate of
System L Transport for Tumor Imaging by PET
Stefanie D. Kra ¨mer1, Linjing Mu2, Adrienne Mu ¨ller1, Claudia Keller1, Olga F. Kuznetsova1,3, Christian Schweinsberg2,
Dominic Franck1, Cristina Mu ¨ller4, Tobias L. Ross1, Roger Schibli1,4, and Simon M. Ametamey1
1Center for Radiopharmaceutical Sciences ETH-PSI-USZ, Institute of Pharmaceutical Sciences ETH, Zurich, Switzerland;2Center
for Radiopharmaceutical Sciences ETH-PSI-USZ, Department of Nuclear Medicine, University Hospital Zurich, Zurich, Switzerland;
3N.P. Bechtereva Institute of Human Brain RAS, Russian Academy of Science, St. Petersburg, Russia; and4Center for
Radiopharmaceutical Sciences ETH-PSI-USZ, Paul Scherrer Institute, Villigen, Switzerland
Large neutral L-amino acids are substrates of system L amino
acid transporters. The level of one of these, LAT1, is increased
in many tumors. Aromatic L-amino acids may also be substrates
of aromatic L-amino acid decarboxylase (AADC), the level of
which is enhanced in endocrine tumors. Increased amino acid
uptake and subsequent decarboxylation result in the intracellular
accumulation of the amino acid and its decarboxylation product.
18F- and11C-labeled neutral aromatic amino acids, such as L-3,4-
dihydroxy-6-18F-fluorophenylalanine (18F-FDOPA) and 5-hydroxy-
L-[b-11C]tryptophan, are thus successfully used in PET to image
endocrine tumors. However, 5-hydroxy-L-[b-11C]tryptophan has
a relatively short physical half-life (20 min). In this work, we eval-
uated the in vitro and in vivo characteristics of the18F-labeled
tryptophan analog 5-(2-18F-fluoroethoxy)-L-tryptophan (18F-
L-FEHTP) as a PET probe for tumor imaging. Methods:18F-L-
FEHTP was synthesized by no-carrier-added18F fluorination
of 5-hydroxy-L-tryptophan. In vitro cell uptake and efflux of
18F-L-FEHTP and18F-FDOPA were studied with NCI-H69 en-
docrine small cell lung cancer cells, PC-3 pseudoendocrine
prostate cancer cells, and MDA-MB-231 exocrine breast
cancer cells. Small-animal PET was performed with the re-
spective xenograft-bearing mice. Tissues were analyzed for
potential metabolites. Results:18F-L-FEHTP specific activity
and radiochemical purity were 50–150 GBq/mmol and greater
than 95%, respectively. In vitro cell uptake of18F-L-FEHTP
was between 48% and 113% of added radioactivity per milli-
gram of protein within 60 min at 37?C and was blocked by
greater than 95% in all tested cell lines by the LAT1/2 inhib-
itor 2-amino-2-norboranecarboxylic acid.18F-FDOPA uptake
ranged from 26% to 53%/mg. PET studies revealed similar
xenograft-to-reference tissue ratios for
18F-FDOPA at 30–45 min after injection. In contrast to the
18F-FDOPA PET results, pretreatment with the AADC inhibi-
tor S-carbidopa did not affect the18F-L-FEHTP PET results.
No decarboxylation products of18F-L-FEHTP were detected
in the xenograft homogenates. Conclusion:
accumulates in endocrine and nonendocrine tumor models
via LAT1 transport but is not decarboxylated by AADC.18F-L-
FEHTP may thus serve as a PET probe for tumor imaging and
quantification of tumor LAT1 activity. These findings are of
interest in view of the ongoing evaluation of LAT1 substrates
and inhibitors for cancer therapy.
Key Words: PET; tumor imaging; LAT; SLC7A; tryptophan
J Nucl Med 2012; 53:434–442
Most malignant lesions have an increased demand for
amino acids. This property makes18F- and11C-labeled
a-amino acids good candidates for tumor imaging by PET.
Unlike18F-FDG, the current gold standard for cancer imag-
ing by PET, amino acids do not generally accumulate in the
healthy brain or inflamed tissues. Amino acid–based PET
probes thus provide good tumor-to-background ratios in
the brain and allow differentiation between tumors and in-
flammation. Their accumulation in tumors is a consequence
of increased amino acid uptake by cancer cells. In particular,
the level of the heterodimer transport system for large neutral
amino acids, LAT1/4F2hc (SLC7A5/SLC3A2; LAT1), is in-
creased in many types of human tumors and may correlate
with the malignancy of the lesion (1). The system L amino
acid transporter (LAT) substrates L-[methyl-11C]methionine
and O-(2-18F-fluoroethyl)-L-tyrosine (18F-FET) are used for
brain tumor imaging (1,2).
The aromatic L-amino acids L-3,4-dihydroxyphenylalanine,
L-tryptophan, and 5-hydroxy-L-tryptophan are substrates not
only of LAT1 but also of aromatic L-amino acid decarbox-
ylase (AADC), the level of which is enhanced in tumor cells
with an endocrine character (3). Decarboxylation of those
amino acids produces biogenic amines, which can subse-
quently be trapped in secretory vesicles. The combination of
amino acid uptake and decarboxylation has been designated
the APUD (amine precursor uptake and decarboxylation) con-
cept (4). The PET probes L-3,4-dihydroxy-6-18F-fluorophe-
nylalanine (18F-FDOPA) and 5-hydroxy-L-[b-11C]tryptophan
(11C-5HTP) accumulate according to the APUD concept in
endocrine tumors (1,5).
Received Aug. 8, 2011; revision accepted Oct. 25, 2011.
For correspondence contact:
Radiopharmaceutical Sciences ETH-PSI-USZ, Institute of Pharmaceutical
Sciences, Wolfgang-Pauli-Strasse 10, CH-8093 Zurich, Switzerland.
Published online Feb. 13, 2012.
COPYRIGHT ª 2012 by the Society of Nuclear Medicine, Inc.
434THE JOURNAL OF NUCLEAR MEDICINE • Vol. 53 • No. 3 • March 2012
We aimed at synthesizing and evaluating an18F-labeled
alternative to11C-5HTP for tumor imaging by PET because
the longer physical half-life of18F (110 min) has several
advantages over the 20-min half-life of11C (6). We hypoth-
(Fig. 1) is a substrate of LATs and possibly of AADC, sim-
ilar to11C-5HTP, and therefore would be a good PET probe
for imaging endocrine tumors and possibly other tumors.
During our evaluation of18F-L-FEHTP, the synthesis of
18F-L-FEHTP and the first PET images and biodistribution
results obtained with18F-L-FEHTP in S180 fibrosarcoma–
bearing mice were published by Li et al. (7).
In this work, we describe the synthesis of18F-L-FEHTP
and compare its biochemical and pharmacologic character-
istics in cell cultures and in small-animal PET with those of
18F-FDOPA. For the invitro and invivo evaluations, we chose
an endocrine tumor model, the small cell lung cancer cell line
NCI-H69 with high AADC activity (8); the so-called “pseu-
doendocrine” prostate cancer cell line PC-3, which has been
shown to express AADC at the messenger RNA level (9); and
exocrine breast cancer cell line MDA-MB-231, which dis-
plays no AADC activity. LAT1 expression has been shown
to occur in all 3 cell lines (10–12).
MATERIALS AND METHODS
18F-L-FEHTP was prepared with a 2-step radiolabeling approach
similar to that described recently (7) and as depicted in Figure 1.
The disodium salt of 5-hydroxy-L-tryptophan was prepared by re-
acting 5-hydroxy-L-tryptophan with 2 equivalent sodium methoxide
(0.5N) in dry methanol at room temperature. No-carrier-added
18F-fluoride was produced via the18O(p,n)18F nuclear reaction by
irradiation of enriched18O-water in a cyclotron (Cyclone 18/9;
IBA).18F-fluoride was immobilized on an anion-exchange cartridge
(QMA Light; Waters) preconditioned with K2CO3(5 mL, 0.5 M)
and then 5–10 mL of water. The activity was eluted with tetrabu-
tylammonium hydroxide (0.6 mL, 0.18 mM) and dried under vac-
uum with a stream of nitrogen at 110?C. Azeotropic drying was
repeated 3 times with 1 mL of acetonitrile each time. To the dried
18F-fluoride complex (typically 20 GBq), ethylene glycol ditosylate
(5–6 mg in 0.9 mL of CH3CN) was added, and the mixture was
heated at 105?C for 6 min. The reaction mixture was cooled, diluted
with 35% ethanol in water (10 mL), and passed through a LiChrolute
EN cartridge (Merck). The18F-labeled product was eluted with di-
methyl sulfoxide (DMSO; 0.8–0.9 mL) to the second reactor, which
had been preloaded with 5-hydroxy-L-tryptophan disodium salt (5–10
mg) in a mixture of DMSO (0.20 mL) and water (0.15 mL). The
reaction mixture was heated at 120?C for 9 min. After cooling, the
reaction mixture was diluted with water (1.8–2.0 mL) and purified by
semipreparative high-performance liquid chromatography (HPLC)
(Supplemental Material) (supplemental materials are available online
only at http://jnm.snmjournals.org). The18F-L-FEHTP product was
collected at about 14 min and neutralized with sodium hydrogen
carbonate (8.4%). The solution was passed through a sterile filter
and used for in vitro and in vivo studies.
18F-DL-FEHTP was produced by the same procedure but with
the racemic 5-hydroxy-tryptophan disodium salt.
18F-FDOPA was obtained at the University Hospital Zurich
from routine production for clinical use (13). The specific activity
was 9–120 GBq/mmol at the end of synthesis.
In Vitro Cell Studies
PC-3, NCI-H69, and MDA-MB-231 cells were purchased from
the German Collection of Microorganisms and Cell Cultures, Cell
Lines Service, and the American Type Culture Collection, re-
spectively. Cells were grown to subconfluence in 48-well plates
(Costar; Corning) or 5-mL plastic tubes (for NCI-H69). Cells were
washed and incubated for 1 h at 37?C with Earle balanced salt
solution containing Ca21and Mg21(EBSS; Invitrogen). The ir-
reversible monoamine oxidase A and B inhibitors clorgyline (Sigma)
and pargyline (Acros Organics), respectively, and—if indicated—
the LAT inhibitor 2-amino-2-norboranecarboxylic acid (BCH)
were added at final concentrations of 0.1, 0.1, and 10 mM,
respectively. At time zero, about 20 kBq of18F-L-FEHTP or
18F-FDOPA were added, and the cells were incubated at 37?C or
on ice. The final tracer concentrations were 0.3–4 nM18F-L-
FEHTP and 0.3–4 mM18F-FDOPA, that is, less than typical
Michaelis-Menten constants KM values for LAT1 and AADC
(14). Total uptake was less than 15% of the added tracer after 1
h. At various time points, the cells were washed twice with ice-cold
EBSS, and adherent cells were detached with trypsin–ethylenedia-
minetetraacetic acid (Invitrogen). All cells were transferred to
Eppendorf tubes, and the radioactivity was quantified in a g-counter
(Wizard 1480; PerkinElmer). To inhibit AADC, we added 80 mM
S-carbidopa (Santa Cruz Biotechnology) in DMSO–100 mM
phosphate-buffered saline 30 min before tracer addition. The final
DMSO concentration was 0.3%. Cell viability was 80%–90% after
60 min for all conditions, as tested with trypan blue staining.
For efflux assays, cells were preincubated for 1 h as described
earlier, and then the cultures were washed twice with ice-cold
EBSS and incubated with EBSS or EBSS containing 0.8 mM
L-leucine (time zero). At various time points, the cells were
washed twice with ice-cold EBSS, detached as described earlier,
generation of18F-L-FEHTP. ACN 5 aceto-
nitrile; TBAF 5 tetra-n-butylammonium
fluoride; Ts 5 tosyl.
Radiosynthesis scheme for
18F-L-FEHTP FOR PET TUMOR IMAGING • Kra ¨mer et al. 435
and analyzed in the g-counter. Protein was quantified with the DC
Protein Assay (Bio-Rad) after cell lysis with 2% sodium dodecyl
sulfate. Bovine serum albumin was used for calibration.
In Vivo PET Experiments
Animal care and experiments were conducted in accordance
with Swiss Animal Welfare legislation and were approved by the
Veterinary Office of Canton Zurich, Zurich, Switzerland. Five-
week-old female NMRI nude and BALB/c mice were supplied by
Charles River. They were allowed free access to water and food.
Six-week-old mice were inoculated subcutaneously in the right
shoulder with 2 · 106PC-3 or MDA-MB-231 cells in 100 mL of
Matrigel (BD Biosciences) or 5 · 106NCI-H69 cells in 100 mL of
phosphate-buffered saline with Ca21and Mg21(Invitrogen). At 3–5
wk after inoculation, when the tumor volumes reached 0.2–1.5 cm3,
the mice were injected in a tail vein with 3.5–18 MBq of18F-L-
FEHTP or18F-FDOPA. For experiments with AADC inhibition,
mice were injected intraperitoneally with S-carbidopa at 25 mg/kg
60 min before tracer injection. For transport competition experi-
ments, mice were injected intraperitoneally with L-tryptophan at
25 mg/kg 10 min before tracer injection. Anesthesia with 2%–3%
isoflurane in oxygen–air was initiated 10 min before the PET scan,
and animals were monitored as described previously (15).
PET scans were performed with a VISTA eXplore small-animal
PET/CT camera (GE Healthcare) in list mode for dynamic analysis or
static mode for whole-body scans (2 bed positions) (16). Static scans
with 2 bed positions were started with the anterior body part contain-
ing the tumor 30 min after tracer injection. The complete 2 bed
position scans lasted 30 min (15 min per bed position). Data were
reconstructed with the 2-dimensional ordered-subsets expectation
maximization protocol and analyzed with PMODVersion 3.2 software
(PMOD Technologies Ltd.). A region of interest for the xenograft,
a reference region of equal shape and volume on the contralateral side,
and a region of interest for the brain were drawn in PMOD on the
basis of the PET images. Tissue radioactivity was expressed as the
standardized uptake value (SUV), that is, the decay-corrected radio-
activity per cubic centimeter divided by the injected radioactivity dose
per gram of body weight.
Ex Vivo Biodistribution, Metabolite Studies, and In
Vitro Plasma Albumin Binding
After the PET scans, mice were sacrificed by decapitation under
isoflurane anesthesia, and tissues were removed, weighed, and
analyzed in the g-counter. The accumulated radioactivity per gram
of tissue was calculated as the decay-corrected radioactivity per
gram of tissue divided by the injected radioactivity dose per gram
of body weight.
For ex vivo metabolite studies, 14–32 MBq of18F-L-FEHTP or
18F-FDOPAwere injected via a lateral tail vein into xenograft-bearing
9- to 12-wk-old NMRI nude mice. Blood samples were withdrawn
via the opposite tail vein, and animals were sacrificed by decapitation
under isoflurane anesthesia at various time points. Xenografts were
excised and homogenized in equal volumes of phosphate-buffered
saline (Invitrogen) with a Polytron (Kinematica, Inc.). Proteins in
the xenograft homogenates, plasma, and urine were precipitated with
equal volumes of ice-cold acetonitrile and centrifugation. Superna-
tants of the xenograft homogenates were extracted again, and all
supernatants were filtered and analyzed by reversed-phase thin-layer
chromatography and ultra-performance liquid chromatography (Sup-
plemental Material). Serum albumin binding was determined by equi-
librium dialysis as described in the Supplemental Material.
Radiosynthesis of18F-L-FEHTP and18F-DL-FEHTP
Chiral HPLC analysis demonstrated that an L-enantio-
meric pure product was obtained when an L-enantiomeric
pure precursor was used for radiolabeling. In a typical exper-
iment, a radiochemical yield of about 23% (decay corrected,
corresponding to 15% not decay corrected) was achieved
with a radiochemical purity of greater than 95%. The specific
activities were in the range of 50–150 GBq/mmol at the end
of synthesis. When 5-hydroxy-DL-tryptophan disodium salt
was used as the precursor, similar radiochemical yield and
radiochemical purity were obtained.
In Vitro Cell Uptake and Efflux Studies
Figure 2 shows the rapid influx of18F-L-FEHTP and18F-
FDOPA into NCI-H69 endocrine cells at 37?C and 4?C,
reaching between 37% and 52% of the total added radioac-
tivity per milligram of protein within 5 min. After the fast
initial uptake,18F-L-FEHTP radioactivity increased steadily,
whereas18F-FDOPA radioactivity decreased at 37?C but not
When no amino acids were added to the incubation buffer,
the efflux of18F-L-FEHTP was slow compared with the
uptake at 37?C and was negligible at 4?C (Fig. 2B). The
uptake (B and D) of18F-L-FEHTP (A and B) and18F-FDOPA (C
and D) in NCI-H69 cell cultures at 37?C (•and n) and 4?C (s
and h). (E)18F-FDOPA uptake in presence of 80 mM S-carbidopa
(: and n) and 0.3% DMSO as control (nand h) at 37?C (: andn)
and 4?C (n and h). (F)18F-L-FEHTP efflux in presence of 0.8 mM L-
leucine in incubation buffer at 37?C (•) and 4?C (s). Symbols rep-
resent averages from 3 independent experiments each (except for 2
independent experiments with L-leucine). Error bars show SDs, ex-
cept for experiments with L-leucine, for which error bars indicate
values from 2 independent experiments.
(A–D) Uptake (A and C) and efflux after 60 min of
436THE JOURNAL OF NUCLEAR MEDICINE • Vol. 53 • No. 3 • March 2012
release of18F-FDOPA or its metabolites continued as ob-
served during uptake at 37?C and was negligible at 4?C
(Fig. 2D). The decrease in radioactivity after the initial fast
uptake of18F-FDOPAwas almost abolished in the presence of
80 mM S-carbidopa (Fig. 2E). At this concentration, intracel-
lular AADC is partially inhibited without a significant loss of
viability of endocrine tumor cells (5,17,18). Figure 2F shows
the rapid efflux of18F-L-FEHTP at 37?C and 4?C after the
addition of 0.8 mM L-leucine, a LAT1 substrate.
The accumulation (average 6 SD) of18F-L-FEHTP and
18F-FDOPA in PC-3 cells reached 113% 6 19%/mg and
53% 6 15%/mg, respectively, of added radioactivity at 37?
C and 60 min (Fig. 3). Efflux was negligible at 4?C but was
significant at 37?C. As observed for NCI-H69 cells, the ad-
dition of 0.8 mM L-leucine resulted in the rapid efflux of
18F-L-FEHTP and18F-FDOPA within 5 min. The uptake of
racemic18F-DL-FEHTP was between 50% and 70% of that
of18F-L-FEHTP, indicating a lower level of uptake of the
D-isomer than of the L-isomer (Fig. 3A).
Figure 4 shows the accumulation of18F-labeled amino
acids in MDA-MB-231 exocrine cells.
18F-FDOPA uptake reached 48% 6 3%/mg and 26% 6
4%/mg, respectively, of added radioactivity after 60 min of
incubation at 37?C. Both amino acids also showed significant
uptake at 4?C. When no amino acids were added to the in-
cubation buffer, both amino acids showed moderate efflux at
37?C and no significant net efflux at 4?C.
The LAT competitive inhibitor BCH blocked18F-L-FEHTP
uptake in all cell lines by greater than 95% at 10 mM. Block-
ing was less efficient for18F-FDOPA, with 13% 6 6% and
10% 6 6% residual uptake at 37?C in NCI-H69 cells and PC-
3 cells, respectively (Supplemental Material).
In Vivo Metabolism
18F-FDOPA and 5-hydroxy-L-tryptophan are decarbox-
ylated in humans and laboratory animals (19). We found
only radioactive metabolites and no18F-FDOPA in the
blood, brain, and tumor 60 min after18F-FDOPA adminis-
tration (data not shown). In contrast, mainly parent18F-L-
FEHTP was detected in the blood, brain, and xenografts 15
and 60 min after18F-L-FEHTP administration. Some radio-
active metabolites of unknown identity were detected in the
urine, and traces were detected in the blood. The results of
reversed-phase thin-layer chromatography of the tissue
extracts are shown in Figure 5. The results were confirmed
by ultra-performance liquid chromatography with 2 differ-
ent mobile phases (data not shown).
Small-Animal PET with Xenograft-Bearing Mice
On the basis of the favorable uptake kinetics of18F-L-
FEHTP in the cell experiments, we expected significant up-
take into xenografts in vivo. Pilot dynamic PET scans with
NMRI nude mice showed the accumulation of both18F-L-
18F-FDOPA in PC-3 xenografts. The
FEHTP SUV ratios in the xenografts and reference tissues
decreased with time between 30 and 150 min after injection.
The18F-FDOPA SUV ratios did not decrease, in accordance
with the APUD concept (Fig. 6). We chose the time window
of 30–45 min after injection for static PET scans. No signif-
icant uptake of18F-L-FEHTP into bone was observed, indi-
cating that in vivo defluorination was negligible.
Figure 7A shows PET images (static) of NMRI nude mice
bearing NCI-H69, PC-3, and MDA-MB-231 xenografts.
18F-FDOPA was administered after S-carbidopa preadminis-
tration. The levels of radioactivity accumulation in the xeno-
L-FEHTP (A and B,•at 37?C and s at 4?C connected with solid lines),
at 37?C and•at 4?C connected with solid lines). Efflux of18F-L-FEHTP
(B) and18F-FDOPA (D) was also studied with 0.8 mM L-leucine in in-
cubation buffer (symbols connected with dashed lines). Symbols rep-
resent averages from 3 independent experiments each (except for 2
independent experiments with L-leucine and 1 experiment with18F-DL-
FEHTP). Error bars show SDs, except for experiments with L-leucine,
for which error bars indicate values from 2 independent experiments.
Uptake (A and C) and efflux (B and D) in PC-3 cells of18F-
18F-DL-FEHTP (A,¤at 37?Cand ♢at 4?C),and18F-FDOPA (CandD,n
(A and B) and18F-FDOPA (C and D) in MDA-MB-231 cells at 37?C
(•and n) and 4?C (s and h). Symbols and error bars represent
averages and SDs from 3 independent experiments each.
Uptake (A and C) and efflux (B and D) of18F-L-FEHTP
18F-L-FEHTP FOR PET TUMOR IMAGING • Kra ¨mer et al. 437
grafts and reference regions were similar for the 2 tracers.
Figures 7B and 7C show the respective SUVs and SUV ratios.
The MDA-MB-231 xenografts had a large gelatinous core
(Supplemental Material), resulting in relatively low levels of
average tumor uptake of18F-L-FEHTP and18F-FDOPA. How-
ever, PET sectional reconstruction (Fig. 7A) showed tracer
accumulation in the periphery of the MDA-MB-231 xeno-
18F-DL-FEHTP was tested in a PC-3 xenograft–bearing
mouse. The xenograft and reference tissue SUVs were about
half those of18F-L-FEHTP, resulting in a similar xenograft-
to-reference tissue SUV ratio as found with18F-L-FEHTP
(data not shown).
Next, we investigated whether plasma L-tryptophan levels
had an influence on18F-L-FEHTP PET images. After the
administration of tryptophan at 25 mg/kg, the xenograft SUVs
32 MBq) was injected into xenograft-bearing NMRI nude mice. Tis-
sues were extracted and analyzed by reversed-phase thin-layer
chromatography.18F-L-FEHTP reference sample had same reten-
tion time as main spot in all lanes. Minutes indicate time of sacrifice
after18F-L-FEHTP injection. No metabolites were detected in xeno-
grafts, and only traces of metabolites were detected in blood.
In vivo metabolism of18F-L-FEHTP.18F-L-FEHTP (14–
and18F-FDOPA (squares) in PC-3 xenografts (black), brain (gray), and
reference tissue (no fill). (B) Ratios of PC-3 xenograft activity to refer-
ence tissue activity (black) and brain activity to reference tissue activity
(gray) for18F-L-FEHTP (circles) and18F-FDOPA (squares). Connected
symbols represent time–activity curves from 1 mouse.18F-FDOPA
was used without S-carbidopa.
(A) Time–activity curves (SUVs) for18F-L-FEHTP (circles)
FEHTP and18F-FDOPA. (A) Identical NCI-H69 (top row), PC-3 (middle
row), and MDA-MB-231 (bottom row) xenograft–bearing NMRI nude
mice were each scanned with18F-L-FEHTP (left of scale bar) and18F-
FDOPA (right of scale bar). At 1 h before18F-FDOPA scans, S-carbi-
dopa (25 mg/kg) was injected intraperitoneally. Shown are transversal
and coronal sections (SUV color scale) and maximum-intensity pro-
jections (MIP, gray scale). Tumors are indicated by arrows. (B) SUVs of
xenografts and reference regions after18F-L-FEHTP and18F-FDOPA
injections. PC-3, w/o carbid. 5 PC-3 xenograft–bearing BALB/c mice
without S-carbidopa pretreatment (images not shown). (C) Ratios of
xenograft SUVs to reference region SUVs for18F-L-FEHTP and18F-
FDOPA. Data are averages and SDs from 3–7 animals each. There
were no significant differences between18F-L-FEHTP and18F-FDOPA
(P . 0.05;18F-FDOPA after S-carbidopa pretreatment). There were
significant differences between18F-DOPA SUVs with and without
S-carbidopa pretreatment (P , 0.01) but not between the xenograft-
to-reference tissue SUV ratios with and without S-carbidopa pretreat-
ment. Note that MDA-MB-231 tumors had gelatinous core.
PET analysis of xenograft-bearing mice with18F-L-
438THE JOURNAL OF NUCLEAR MEDICINE • Vol. 53 • No. 3 • March 2012
and xenograft-to-reference tissue SUV ratios were lower by
trend but without significance (P . 0.05). The SUV ratios
for NCI-H69 xenografts were 1.71 6 0.26 (with L-tryptophan
preadministration, n 5 4) versus 1.95 6 0.30 (no L-tryptophan
preadministration, n 5 3, from Fig. 7C) and for PC-3 xeno-
grafts 1.57 6 0.05 (with L-tryptophan preadministration, n 5 4)
versus 1.80 6 0.39 (no L-tryptophan preadministration, n 5 7,
from Fig. 7C).
Influence of AADC Inhibition on18F-FDOPA PET
The18F-FDOPA PET scans shown in Figure 7A were per-
formed after S-carbidopa pretreatment. Four PC-3 xenograft-
bearing BALB/c mice were also scanned with18F-FDOPA in
the absence of S-carbidopa. The18F-FDOPA SUVs in the
xenografts and reference tissues were about 50% those after
S-carbidopa pretreatment (Fig. 7B). However, the average
xenograft-to-reference tissue SUV ratios in the 2 protocols
were not significantly different (Fig. 7C). The xenograft and
background18F-L-FEHTP SUVs in the same mice were not
significantly different from those in the NMRI nude mice
(Fig. 7B). The
18F-L-FEHTP xenograft-to-reference tissue
SUV ratio in the PC-3 xenograft–bearing BALB/c mice was
2.1 6 0.3 (n 5 4). S-carbidopa had no influence on18F-L-
FEHTP PET images (Supplemental Material).
Ex Vivo Biodistribution and In Vitro Serum
Table 1 shows average uptake values (SUV) for18F-L-
FEHTP 70 min after injection. The levels of uptake (reported
as percentage injected dose per gram of body weight [%ID/
g]) were highest in the pancreas (29–55 %ID/g) and then the
kidneys (6–18 %ID/g) and xenografts (5–9 %ID/g). The pre-
administration of tryptophan at 25 mg/kg had no significant
influence on the biodistribution of18F-L-FEHTP but resulted
in a generally lower level of uptake in tissues with high LAT1
expression, that is, NCI-H69 xenografts, brain, and kidneys.
A high level of plasma protein binding could result in a high
level of background radioactivity because of slow clearance
from the blood. Plasma tryptophan is associated to 80%–90%
with albumin (20). The bound fraction of18F-L-FEHTP in 4%
bovine serum albumin at 37?C was negligible, that is, 0.05 6
0.01 (n 5 4 dialysis cells). No binding was detected with
human or rat plasma diluted 1/20 (data not shown).
18F-L-FEHTP synthesis was similar to the procedure re-
cently described by Li et al. (7), with similar radiochemical
yield and purity. The complete radiolabeling process was
established in a fully automated module, resulting in repro-
ducible quality of the product and reducing the radioactiv-
ity burden on the radiochemist. The crude product was
purified by semipreparative HPLC with 35 mM acetate
buffer containing 8% ethanol. The collected HPLC fraction
was directly used for in vitro and in vivo experiments after
neutralization with 8.4% sodium bicarbonate.
18F-L-FEHTP accumulated in cancer cells of endocrine,
pseudoendocrine, and exocrine phenotypes in vitro and in
vivo. Decarboxylation or another metabolic step was not
involved in the accumulation, confirming that increased
transport alone is sufficient for tumor imaging by PET with
labeled large neutral amino acids (2).
Because18F-L-FEHTP uptake into the cells was observed
not only at 37?C but also at 4?C and because efflux was
accelerated by the addition of L-leucine, we concluded that
accumulation was mediated by the exchange of one or more
Ex Vivo Biodistribution* Data for18F-L-FEHTP 70 Minutes After Injection
18F-L-FEHTP biodistribution (SUV) in:
mice (n 5 3)
mice given tryptophan at 25 mg/kg (n 5 3)
mice (n 5 4)
Intestines with contents
1.58 6 0.48
0.74 6 0.14
0.75 6 0.08
0.84 6 0.32
0.79 6 0.15
0.78 6 0.11
0.78 6 0.33
2.05 6 0.59
0.81 6 0.10
0.73 6 0.13
8.50 6 1.64
0.88 6 0.20
0.77 6 0.15
1.24 6 0.21
1.00 6 0.50
0.79 6 0.05
0.67 6 0.05
1.04 6 0.67
0.79 6 0.05
0.84 6 0.07
1.73 6 0.14
0.83 6 0.05
0.78 6 0.03
8.06 6 0.32
0.82 6 0.09
0.81 6 0.05
1.40 6 0.35
0.87 6 0.22
0.77 6 0.08
0.68 6 0.06
1.06 6 0.43
0.78 6 0.07
0.64 6 0.06
2.00 6 0.69
0.85 6 0.15
0.74 6 0.12
8.15 6 1.32
0.83 6 0.08
0.74 6 0.07
*Accumulated radioactivity per gram of tissue divided by radioactivity dose per gram of body weight (SUV). Data are reported as
average 6 SD unless otherwise indicated.
See Ishiwata et al. (41) for18F-FDOPA tissue distribution.
18F-L-FEHTP FOR PET TUMOR IMAGING • Kra ¨mer et al.439
amino acids rather than unilateral or concentrative trans-
porters. The low temperature sensitivity of the cell uptake
of18F-L-FEHTP was in line with the results of an early
study on L-tryptophan uptake into Ehrlich ascites tumor
cells (21). Exchange transporters carrying neutral aromatic
amino acids are SLC7A5 (LAT1), SLC7A6 (Y1LAT2),
SLC7A7 (Y1LAT1), SLC7A8 (LAT2), SLC7A9 (B(0,1)
AT), SLC43A1 (LAT3), and SLC43A2 (LAT4) (14,22).
Na1-dependent symporters that may also be involved in
the transport of L-tryptophan analogs are SLC1A5 (ATB0,
ASCT2) and SLC6A14 (ATB01) (14).
The cell uptake of18F-L-FEHTP was inhibited to greater
than 95% by 10 mM BCH in all tested cell lines. BCH is
a substrate and competitive inhibitor of Na1-independent
transport of large neutral amino acids, with typical trans-
porter-inhibitor constants Kivalues of less than 1 mM for
SLC7A5 (LAT1) and SLC7A8 (LAT2) (23,24). SLC7A6,
SLC7A7, SLC1A5, and SLC6A14 were not directly
inhibited by BCH, and no amino acid efflux was observed
for SLC7A7 after the addition of
SLC7A9, SLC43A1, and SLC43A2 were only moderately
inhibited by 10 mM BCH (30–32).
Considering the highly efficient transport inhibition by 10
mM BCH in our experiments with 3 tumor cell lines of
different phenotypes, we concluded that18F-L-FEHTP was
almost exclusively taken up by LAT1 or LAT2. The latter
is, however, not typically overexpressed in cancer cells (33).
18F-FDOPA influx into MDA-MB-231 nonendocrine cells
was almost completely blocked by BCH, also suggesting
LAT1 or LAT2 as the major uptake mechanism. These results
were in agreement with the findings of Neels et al. (5).
18F-FDOPA radioactivity was released from NCI-H69 en-
docrine cells during uptake experiments at 37?C but not at 4?C,
suggesting an energy-driven mechanism. According to the
APUD concept, a higher level of accumulation of18F-FDOPA
radioactivity would be expected at 37?C than at 4?C because of
accumulation via enzymatic decarboxylation. The unexpected
energy-dependent efflux of radioactivity was inhibited by the
AADC inhibitor S-carbidopa, indicating that it was related to
18F-FDOPA decarboxylation. The observed AADC-dependent
efflux may have resulted from the release of the decarboxyl-
ation product 6-18F-fluoro-L-dopamine. In contrast to the
18F-FDOPA results, the kinetics and extent of18F-L-FEHTP
uptake were temperature independent, excluding the involve-
ment of any energy-dependent mechanism, such as AADC.
18F-L-FEHTP and18F-FDOPA accumulated significantly
in NCI-H69 and PC-3 xenografts, with no significant differ-
ence between the tracers or the tumor models at scan times
between 30 and 45 min after injection. MDA-MB-231 xeno-
grafts were not ideal for quantitative comparisons because of
their large gelatinous core. However, both tracers accumu-
lated in the intact periphery of the xenografts.
Our data indicated that18F-L-FEHTP was not decarbox-
ylated in vivo. In contrast to the findings obtained with
18F-FDOPA, the AADC inhibitor S-carbidopa had no in-
18F-L-FEHTP PET images. Furthermore, no
metabolites of L-FEHTP were detected in tumor homoge-
nates, and only traces of metabolites were found in blood.
Although the18F-FDOPA SUV ratios for PC-3 xenografts
and reference tissues did not decrease between 30 and
150 min, in line with the APUD concept, the respective
SUV ratios of18F-L-FEHTP decreased with time. The ob-
served metabolic stability of18F-L-FEHTP was in contrast to
the findings obtained with11C-5HTP (5,19). The latter was
almost quantitatively decarboxylated within minutes in hu-
man carcinoid (endocrine) liver tumors and metastases (19).
The methylation of L-3,4-dihydroxyphenylalanine at cat-
echol 3-OH resulted in the loss of its properties as a sub-
strate or inhibitor of AADC, suggesting that binding to the
active site was abolished (34). On the basis of these findings
and our observations with18F-L-FEHTP, it may be specu-
lated that O-alkylation in the aromatic system of AADC
substrates changes their electronic or steric properties in
a way that disfavors binding to the active site. Structure–
activity relationship, site-directed mutagenesis, or molecu-
lar modeling studies are required to elaborate the structural
properties of AADC substrates.
Our findings indicated that the mechanism of18F-L-FEHTP
accumulation was more similar to that of18F-FET than to that
of11C-5HTP.18F-FET is taken up by transport systems but is
not metabolized in tumor cells (2). In mice with peripheral
xenografts,18F-L-FEHTP and18F-FET showed similar distri-
bution and kinetic behaviors, including brain uptake, with
higher relative kidney radioactivity of18F-L-FEHTP than of
18F-FET (35). In vitro18F-FET uptake was less efficiently in-
hibited by BCH than we observed for18F-L-FEHTP (2), in-
dicating that the latter may be more selective for LAT1 than
18F-FET. However, only a direct comparison will provide a
L-tryptophan is a substrate of indoleamine 2,3-dioxygenase
(IDO). Intracellular IDO activity may result in the trapping of
polar18F-L-FEHTP metabolites or in indirect18F-L-FEHTP
accumulation via intracellular L-tryptophan depletion. Signif-
icant IDO activity was shown for MDA-MB-231 cells after
interferon-g stimulation (12). We did not detect any radio-
metabolites in MDA-MB-231 xenografts 15 min after injec-
tion and therefore did not further investigate this potential
route of18F-L-FEHTP radioactivity accumulation.
Inview of the evaluation of LAT1 substrates and inhibitors
as potential anticancer drugs (36–38),18F-L-FEHTP is a can-
didate PET tracer for staging tumors according to their LAT1
expression and activity and therefore their susceptibility to
LAT1-targeted drugs. In addition, the in vivo efficiency of
LAT1 inhibitors may be monitored directly by PET. Further
studies are required to show the correlation between tumor
LAT1 activity and18F-L-FEHTP uptake kinetics in vitro and
in vivo and to evaluate the direct and indirect effects of other
amino acid transporters and IDO on this correlation.
We found relatively high levels of18F-L-FEHTP and
18F-FDOPA radioactivity in all tissues, with notably high
activity in the pancreas. Serum albumin binding was
excluded as a cause of high levels of18F-L-FEHTP back-
440THE JOURNAL OF NUCLEAR MEDICINE • Vol. 53 • No. 3 • March 2012
ground activity. The latter may be assigned to the relatively
high LAT1 expression in healthy rodent tissue. Mice have
relatively high LAT1 expression levels in most tissues, with
particularly high levels in the pancreas. Humans have high
LAT1 expression levels in protective endothelial and epi-
thelial barriers and in proliferating cells but negligible
expression in other tissues (22,25,39,40). Taking these spe-
cies differences into account, we expect lower levels of
background radioactivity in humans and therefore higher
SUV ratios between tumors with increased LAT1 activity
and healthy tissues.
We have characterized the biochemical and pharmacologic
properties of18F-L-FEHTP, an18F-labeled aromatic L-amino
acid analog with promise for tumor imaging. We identified
18F-L-FEHTP as a substrate for LAT1/2 transport but not for
decarboxylation by AADC. Therefore,18F-L-FEHTP is a po-
tent PET probe for the LAT1 activity of malignant lesions,
independent of the AADC activity of tumors and healthy
tissues. It may be used in the future to stage tumors for
susceptibility to LAT1-targeted drugs.
The costs of publication of this article were defrayed in
part by the payment of page charges. Therefore, and solely
to indicate this fact, this article is hereby marked “adver-
tisement” in accordance with 18 USC section 1734.
We thank Petra Wirth for assistance with animal experi-
ments and Bruno Mancosu, Judith Fra ¨ssdorf, Anass Johayem,
and Zoran Vujicic for assistance with tracer development and
for tracer production. We thank Olga S. Fedorova and Holger
Siebeneicher for their contributions to tracer development.
We thank Sandra Borkowski, Sabine Zitzmann-Kolbe, Keith
Graham, and Ludger Dinkelborg from Bayer Schering
Pharma for fruitful discussions. No potential conflict of in-
terest relevant to this article was reported.
1. Plathow C, Weber WA. Tumor cell metabolism imaging. J Nucl Med. 2008;49
2. Langen KJ, Hamacher K, Weckesser M, et al. O-(2-[18F]fluoroethyl)-L-tyrosine:
uptake mechanisms and clinical applications. Nucl Med Biol. 2006;33:287–294.
3. Sundin A, Eriksson B, Bergstrom M, et al. PET in the diagnosis of neuroendocrine
tumors. Ann N Y Acad Sci. 2004;1014:246–257.
4. Pearse AG. The APUD concept and hormone production. Clin Endocrinol
5. Neels OC, Koopmans KP, Jager PL, et al. Manipulation of [11C]-5-hydroxytrypto-
phan and 6-[18F]fluoro-3,4-dihydroxy-L-phenylalanine accumulation in neuroendo-
crine tumor cells. Cancer Res. 2008;68:7183–7190.
6. Ametamey SM, Honer M, Schubiger PA. Molecular imaging with PET. Chem
7. Li R, Wu SC, Wang SC, et al. Synthesis and evaluation of L-5-(2-[18F]fluoroethoxy)
tryptophan as a new PET tracer. Appl Radiat Isot. 2010;68:303–308.
8. Carney DN, Gazdar AF, Bepler G, et al. Establishment and identification of small cell
lung-cancer cell-lines having classic and variant features. Cancer Res. 1985;45:2913–
9. Avgeris M, Koutalellis G, Fragoulis EG, Scorilas A. Expression analysis and clin-
ical utility of L-Dopa decarboxylase (DDC) in prostate cancer. Clin Biochem.
10. Oka S, Okudaira H, Yoshida Y, et al. Transport mechanisms of trans-1-amino-3-
fluoro[1-14C]cyclobutanecarboxylic acid in prostate cancer cells. Nucl Med Biol.
11. Miko E, Margitai Z, Czimmerer Z, et al. miR-126 inhibits proliferation of small
cell lung cancer cells by targeting SLC7A5. FEBS Lett. 2011;585:1191–1196.
12. Travers MT, Gow IF, Barber MC, et al. Indoleamine 2,3-dioxygenase activity
and L-tryptophan transport in human breast cancer cells. Biochim Biophys Acta.
13. Namavari M, Bishop A, Satyamurthy N, et al. Regioselective radiofluorodestannyla-
tion with [18F]F2and [18F]CH3COOF: a high yield synthesis of 6-[18F]fluoro-L-dopa.
Int J Rad Appl Instrum A. 1992;43:989–996.
14. UniProt Consortium. UniProt Web site. Available at: www.uniprot.org. Accessed
January 19, 2012.
15. Honer M, Bru ¨hlmeier M, Missimer J, et al. Dynamic imaging of striatal D2re-
ceptors in mice using quad-HIDAC PET. J Nucl Med. 2004;45:464–470.
16. Wang Y, Seidel J, Tsui BM, et al. Performance evaluation of the GE Healthcare
eXplore VISTA dual-ring small-animal PET scanner. J Nucl Med. 2006;47:1891–1900.
17. Verbeek MM, Geurtz PB, Willemsen MA, Wevers RA. Aromatic L-amino acid de-
carboxylase enzyme activity in deficient patients and heterozygotes. Mol Genet Metab.
18. Gilbert JA, Frederick LM, Ames MM. The aromatic-L-amino acid decarboxylase
inhibitor carbidopa is selectively cytotoxic to human pulmonary carcinoid and small
cell lung carcinoma cells. Clin Cancer Res. 2000;6:4365–4372.
19. Sundin A, Eriksson B, Bergstro ¨m M, et al. Demonstration of [11C] 5-hydroxy-L-
tryptophan uptake and decarboxylation in carcinoid tumors by specific positioning
labeling in positron emission tomography. Nucl Med Biol. 2000;27:33–41.
20. Yuwiler A, Oldendorf WH, Geller E, Braun L. Effect of albumin binding and amino
acid competition on tryptophan uptake into brain. J Neurochem. 1977;28:1015–1023.
21. Jacquez JA, Sherman JH, Terris J. Temperature dependence of amino acid transport in
Ehrlich ascites cells: with results which bear on the A-L distinction. Biochim Biophys
22. Verrey F. System L: heteromeric exchangers of large, neutral amino acids in-
volved in directional transport. Pflugers Arch. 2003;445:529–533.
23. Stevens BR, Ross HJ, Wright EM. Multiple transport pathways for neutral amino
acids in rabbit jejunal brush-border vesicles. J Membr Biol. 1982;66:213–225.
24. Kudo Y, Boyd CA. Characterisation of L-tryptophan transporters in human placenta:
a comparison of brush border and basal membrane vesicles. J Physiol. 2001;531:405–
25. Bro ¨er A, Wagner CA, Lang F, Broer S. The heterodimeric amino acid transporter
4F2hc/y1LAT2 mediates arginine efflux in exchange with glutamine. Biochem J.
26. Pfeiffer R, Rossier G, Spindler B, et al. Amino acid transport of y1L-type by
heterodimers of 4F2hc/CD98 and members of the glycoprotein-associated amino
acid transporter family. EMBO J. 1999;18:49–57.
27. Kanai Y, Fukasawa Y, Cha SH, et al. Transport properties of a system y1L
neutral and basic amino acid transporter: insights into the mechanisms of sub-
strate recognition. J Biol Chem. 2000;275:20787–20793.
28. Sloan JL, Mager S. Cloning and functional expression of a human Na1and
Cl2-dependent neutral and cationic amino acid transporter B(01). J Biol Chem.
29. Kirchhoff P, Dave MH, Remy C, et al. An amino acid transporter involved in
gastric acid secretion. Pflugers Arch. 2006;451:738–748.
30. Taylor PM, Hundal HS, Rennie MJ. Transport of glutamine in Xenopus laevis
oocytes: relationship with transport of other amino acids. J Membr Biol.
31. Babu E, Kanai Y, Chairoungdua A, et al. Identification of a novel system L
amino acid transporter structurally distinct from heterodimeric amino acid trans-
porters. J Biol Chem. 2003;278:43838–43845.
32. Bodoy S, Martin L, Zorzano A, et al. Identification of LAT4, a novel amino acid
transporter with system L activity. J Biol Chem. 2005;280:12002–12011.
33. Haase C, Bergmann R, Fuechtner F, et al. L-type amino acid transporters LAT1 and
LAT4 in cancer: uptake of 3-O-methyl-6-18F-fluoro-L-dopa in human adenocarci-
noma and squamous cell carcinoma invitro and invivo. J Nucl Med. 2007;48:2063–
34. Soares-da-Silva P, Parada A, Serrao P. The O-methylated derivative of L-DOPA, 3-
O-methyl-L-DOPA, fails to inhibit neuronal and non-neuronal aromatic L-amino
acid decarboxylase. Brain Res. 2000;863:293–297.
35. Wester HJ, Herz M, Weber W, et al. Synthesis and radiopharmacology of O-(2-[18F]
fluoroethyl)-L-tyrosine for tumor imaging. J Nucl Med. 1999;40:205–212.
36. Imai H, Kaira K, Oriuchi N, et al. Inhibition of L-type amino acid transporter 1 has
antitumor activity in non-small cell lung cancer. Anticancer Res. 2010;30:4819–4828.
18F-L-FEHTP FOR PET TUMOR IMAGING • Kra ¨mer et al.441
37. Fan X, Ross DD, Arakawa H, et al. Impact of system L amino acid transporter 1 Download full-text
(LAT1) on proliferation of human ovarian cancer cells: a possible target for
combination therapy with anti-proliferative aminopeptidase inhibitors. Biochem
38. Lin J, Raoof DA, Thomas DG, et al. L-type amino acid transporter-1 overexpres-
sion and melphalan sensitivity in Barrett’s adenocarcinoma. Neoplasia. 2004;6:
39. Heiss P, Mayer S, Herz M, et al. Investigation of transport mechanism and uptake
kinetics of O-(2-[18F]fluoroethyl)-L-tyrosine in vitro and in vivo. J Nucl Med.
40. Blomquist L. Species differences in the accumulation of 4-iodophenylalanine in
the exocrine pancreas. Comp Biochem Physiol. 1969;28:777–782.
41. Ishiwata K, Kubota K, Kubota R, et al. Selective 2-[18F]fluorodopa uptake for
melanogenesis in murine metastatic melanomas. J Nucl Med. 1991;32:95–101.
442THE JOURNAL OF NUCLEAR MEDICINE • Vol. 53 • No. 3 • March 2012