[CANCER
RESEARCH
45, 3155-3160.
July 1985]
Defective Facilitated Diffusion of Nucleosides, a Primary Mechanism of
Resistance to 5-Fluoro-2'-deoxyuridine in the HCT-8 Human Carcinoma
Line1
Alberto F. Sobrero, Robyn D. Moir, Joseph R. Berlino, and Robert E. Handschumacher2
Department of Pharmacology,
Yale University School ol Medicine, New Haven, Connecticut 06510
vitro development of resistance of other mammalian cells to
FdUrd3 has been documented to be due to decreased levels or
ABSTRACT
In vitro resistance of HCT-8 cells to 5-fluoro-2'-deoxyundine
(FdUrd) has been obtained after a stepwise increase (up to 1 ¿tu)
in the concentration of the nucleoside in the culture medium over
a period of 6 months. With a clonogenic assay, the toxicities of
17 antineoplastic agents on HCT-8-sensitive and -resistant cells
were compared. Resistant cells were 700-fold resistant to FdUrd
and showed different degrees of cross-resistance to several
purine and pyrimidine nucleoside analogues; no cross-resistance
was noted to base analogues and other cytotoxic drugs. The
activities of FdUrd phosphorylase, 5'-fluorouridine kinase, 5fluorouridine phosphorylase, 5-fluorouracil phosphoribosyltransferase, and thymidylate synthase were not significantly different
in the sensitive and resistant cell lines. Mixing experiments
indirectly excluded the possible elevation of the level of cytoplasmic phosphatases. The activity of FdUrd kinase in sensitive
cell extracts was no more than twice that of resistant cells, and
the affinities of this enzyme for FdUrd and thymidine at 0.1 to 50
UM were similar in both cell lines. However, cultures of this line
failed to accumulate 5-fluoro-2'-deoxyuridylate
at concentrations
of FdUrd that resulted in substantial accumulation of the nucleotide in the sensitive line. These contrasting data suggested a
defect in the facilitated diffusion of the analogue. The entrance
of free nucleoside and its subsequent phosphorylation were
compared in the two lines over short (2 to 40 s) and longer time
periods at 25°Cand at 4°Cover a range of extracellular FdUrd
concentrations (0.1 to 10 //M). Rapid entrace of the nucleoside
into sensitive cells was observed, but entry was not detectable
in resistant cells. Dipyridamole and nitrobenzylthioinosine inhibi
tion as well as high-performance liquid chromatography analysis
confirmed that data obtained from the sensitive cell line during
the first 40 s primarily reflected facilitated diffusion of free nu
cleoside.
MATERIALS AND METHODS
Chemicals. FUra, FUrd, FdUrd, dThd, 6-mercaptopurine, 6-methylmercaptopurine riboskte, dipyridamole, and phosphoribosyl-1 -pyrophosphate were purchased from Sigma Chemical Co., St. Louis, MO; hypoxanthine was from P-L Biochemicals, Inc., Milwaukee, Wl; and 5-iododeoxyuridine was from Nutritional Biochemicals Corp., Cleveland, OH.
1-/3-D-Arabinofuranosylcytosine, cisplatin, mitomycin C, doxorubicin, vincristine, and methyl-1-(2-chloroethyl)-3-cyclohexyl-1-nitrosourea
were
the materials available for clinical use. MTX and pyrazofurin were gen
erously supplied by Lederte Laboratories and Eli Lilly and Co., respec
tively. [14C]FdUrd (56 mCi/mmol), [14C]FUrd (56 mCi/mmol), [3H]FdUrd
(18 Ci/mmol), [3H]FUrd (18 Ci/mmol), and [3H]dThd (65 Ci/mmol) were
purchased from Moravek Biochemicals, City of Industry, CA; [14C]sucrose (0.4 Ci/mmol) and 3H2O (1 mCi/ml) were purchased from New
England Nuclear, Boston, MA; and [14C]FUra (52 mCi/mmol) was from
Amersham/Searte Corp., Arlington Heights, IL. NBMPR and [3H]NBMPR
were generously supplied by Dr. A. R. P. Paterson.
Methods. The colon adenocarcinoma line HCT-8 (19) was obtained
from Dr. W. A. F. Tompkins and demonstrated to
after repeated cultures with antibiotics. Cells were
contamination every 6 months and consistently
during the course of these experiments. Cultures
be free of Mycoplasma
tested for Mycoplasma
found to be negative
were maintained in 25-
sq cm sterile plastic flasks (Costar, Cambridge, MA) as a monolayer in
RPMI Medium 1640 (Grand Island Biological Co., Grand Island, NY)
supplemented with 10% HS and were subcultured weekly after 3-min
INTRODUCTION
Nucleoside transport in mammalian cells is known to occur
through a facilitated diffusion system that accepts a wide variety
of physiological purine and pyrimidine nucleosides as well as
their analogs (15). Given the low specificity of this system, cells
might be expected to become resistant to these analogs by
losing their capacity to transport nucleosides. In fact, only with
a S 49 mouse T cell lymphoma cell line has convincing evidence
for a nucleoside carrier defect been presented (6). In vivo and in
"Supported
lack of the enzyme TK (13, 20), increased levels of the target
enzyme TS (2,16), altered affinity of TS for the substrate analog
FdUMP (1), and possibly increased activity of cytoplasmic nucleotide phosphatases inactivating FdUMP (9). We describe for
the first time a clone of human cells (HCT-8) with minimal changes
in TS and TK levels, that are resistant in vitro to FdUrd by virtue
of impaired nucleoside transport, and discuss the clinical poten
tial of this finding in view of the origin of this line from a human
colon carcinoma.
by Grants CA 08010, CA 24887, and 28852 and by American
Cancer Society Grant CH 67.
* To whom requests for reprints should be addressed.
Received 7/27/84; revised 3/14/85; accepted 3/19/85.
CANCER
incubation with 5% trypsin in 0.9% NaCI solution and after repeated
pipetting to dissociate cell aggregates. Under these conditions, the
doubling time was 18 h, and the cloning efficiency was about 40%.
In vitro resistance of HCT-8 cells to FdUrd was achieved by a stepwise
increase (up to 1 ¿/
M) in the concentration of the nucleoside over a period
of 6 months. Soft agar cloning of 104 resistant cells in 30 ml of RPMI
Medium 1640 containing 10% HS, 0.01% agar and 1 UM FdUrd allowed
"The abbreviations used are: FdUrd, 5-fluorc-2'-<Jeoxyuridine; TK, thymidine
kinase; TS, thymidylate synthase; FdUMP, 5-fluoro-2'-deoxyuridylate;
FUra, 5fluorouraäl; FUrd, 5-fluorouridine; dThd, thymidine; MTX, methotrexate; NMBPR,
6-{(4-nitrobenzyl)thio]-9--y-D-ribofuranosylpurine;
HS, horse serum; PBS, phos
phate-buffered saline; DHFR, dihydrofolate reducÃ-ase; FUMP, 5-fluorouridine 5'monophosphate; TCA, trichtoroacetic acid; HPLC, high-performance liquid chro
matography.
RESEARCH VOL. 45 JULY 1985
3155
Downloaded from cancerres.aacrjournals.org on June 15, 2017. © 1985 American Association for Cancer Research.
DEFECTIVE
TRANSPORT
IN HCT-8 CELLS
the isolation of a rapidly growing clone of these cells that was dissociated
by several passages through a 25-gauge needle and grown as a monolayer for 2 weeks. Cells from this clone were then cultured in a 96-well
and 10 to 50 /¿I
of cell extract.
Reactions were initiated by the addition of cell extract and incubated
at 37°C; 5 ¡Awere removed at each time point, applied to thin-layer
plate at low density (10 cells/ml, 0.1 ml/well) (Costar); the most rapidly
growing subline with a cloning efficiency of 10% and a doubling time of
18 h was isolated, expanded, and used in all subsequent experiments.
Stock cultures were maintained in RPMI Medium 1640 supplemented
with 10% HS and 1 MM FdUrd. Cells used for determination of enzyme
activities as well as uptake and cloning experiments were subcultured in
the same medium without FdUrd for 7 to 10 days.
For the monolayer clonogenic assay, stock cultures were trypsinized
for 3 min, and an essentially monocellular dispersion was obtained by
passing the trypsinized cells through a 25-gauge needle. Since the
cloning efficiency of the resistant line was about % that of the sensitive
line, 4 x 102 sensitive and 2 x 103 resistant cells in 5 ml of medium
containing 10% HS were dispensed into sterile 60-mm Retri dishes
(Costar) and incubated at 37°C and 100% humidity with 7.5% CO2.
chromatography plates (Merck; silica gel with 60 F254 fluorescent indi
cator) with unlabeled standards of FUra, FUrd, FdUrd, and FUMP, and
immediately dried at 80°Cto stop the reaction. The plates were devel
Eighteen h later, when the cells were attached to the bottom of the Retri
dish but had not yet divided, 0.1 ml of an appropriate dilution of drug in
PBS (0.8% NaCI:0.02% KCI:0.12% NaHPO4:0.02% KH2PO4) was added
to each dish. Control dishes received the same volume of PBS. After
incubation as indicated, the medium was decanted, the dishes were
washed twice with 5 ml of PBS, and 5 ml of fresh medium were added.
The sensitivity of both sensitive and resistant cells to the antineoplastic
agents was determined under conditions of continuous exposure to the
drugs. Clonal growth was determined after staining with orcein. Colonies
containing more than 200 cells were scored at x10 magnification using
a dissecting microscope. Under these clonal growth conditions, more
than 95% of the colonies from both sensitive and resistant cells contained
more than 200 cells. Each experimental point was determined in triplicate
with 4 replicate controls; experiments were repeated at least twice.
For outgrowth assays, 5 x 104 and 10x10*
sensitive and resistant
folate; 0.9 mM formaldehyde;
cells, respectively, in 10 ml of RPMI Medium 1640 containing 10% HS
were plated in 25-sq cm tissue culture flasks. Duplicate cultures were
collected daily for 9 days by 30-min trypsinization, and appropriate
oped in a chloroform:methanol:acetic
acid (17:3:1) solvent. The stand
ards were visualized by UV light, and radioactivity in each spot was
measured in a Beckman liquid scintillation spectrometer. Assays were
performed at 2 enzyme concentrations so that rates were derived from
less than 25% substrate conversion, and the rate was linear for at least
2 time points (15 and 30 min). Activity is expressed as nmol of product
formed per mg of protein per h. Protein concentration was estimated by
the method of Bradford (4).
The displacement of tritium from 2'-[5-3H]deoxyuridylate was used to
determine TS activity (18). The stock reaction mixture, in a total volume
of 40 ul, contained: 110 MM [3H]deoxyuridylate; 1 mM (±)-i_-tetrahydro62.5 mM 2-mercaptoethanol;
50 mM NaF;
130 mM phosphate buffer (pH 7.5); and the enzyme (5 and 10 n\). After
5-min incubation at 37°C, the reaction was started by addition of the
enzyme and was stopped after 60 min by addition of 200 n\ of a charcoal
(Norit) suspension (100 mg/ml) in 2% TCA. After centrifugation, 100 ^
of supernatant fluid were added to 4.5 ml of Liquiscint and counted in a
Beckman liquid scintillation counter. Results are expressed as nmol of
3H released per h per mg of protein. All enzyme assays were linear with
respect to time and enzyme concentration at the substrate concentra
tions used.
DHFR was analyzed spectrophotometrically
by measuring the de
crease in absorbance at 340 nm (11) using a Gilford Model 2400
spectrophotometer. A molar extinction of 12,000 at 340 nm was used
for the calculation of DHFR activity and expressed as /¿molof substrate
converted per mg of protein per h.
Cell suspensions for studies of FdUrd uptake were prepared by 3-min
trypsinization followed by appropriate washes. The final concentration
of [3H]FdUrd was 0.1 MM(1 to 2 Ci/ml), unless otherwise specified. Short-
dilutions of the cell suspensions were counted using a Model ZBI Coulter
Counter (Coulter Electronics, Hialeah, FL). Both cell lines have a doubling
time of approximately 18 h during the phase of logarithmic growth, but
resistant cells have a lag phase of 2 to 3 days, compared to 1 day for
sensitive cells (data not shown).
Using the same technique, it was possible to study the effect of MTX
alone or in combination with hypoxanthine or dThd or both of these
compounds on the growth of the 2 cell lines. In these experiments, the
drugs were added in 0.1- to 5-ml cultures (25 sq cm) immediately after
term uptake (<40 s) was measured by a rapid sampling technique (3)
using 100 /J of oil [84 parts of 125 DC 550 Hi Phenyl Silicon oil (NYE,
Inc., New Bedford, MA) plus 16 parts of paraffin oil] layered on top of
100 M!of 15% TCA in a 0.5-ml polypropylene microcentrifuge tube in an
Eppendorf centrifuge, Brinkmann Model 3200. [3H]FdUrd uptake was
initiated in these tubes by addition of 100 ¡aof cell suspension (5 x 106
plating and remained in the culture medium for the entire period of
incubation (7 days).
To prepare cytosol for enzyme assays, exponentially growing cells in
600-sq cm tissue culture dishes (Nunc, Roskilde, Denmark) were washed
twice with 50 ml of ice-cold 0.9% NaCI solution and harvested, scraping
with a rubber blade into 2 ml of 50 mM Tris-HCI, pH 7.4, with 4 nM NaF
and 3 mM dithtothreitol at 0°C.Cell suspensions were homogenized for
FdUrd solution, two 1-ml syringes with 22-gauge needles were inserted
into a 0.3-mm inside-diameter Tygon tubing with a mixing volume of 2
5 s (Tissumizer; Tekmar Co., Cincinnati, OH), and the extracts were
frozen and thawed twice in dry ice and methanol; no intact cells could
be seen by light microscopy. The suspensions were centrifugea at
100,000 x g for 60 min at 4°C, and the supernatant was used for
enzyme assays.
Enzymes of fluoropyrimidine metabolism were assayed in 100-Ml re
action mixtures containing 50 mM Tris-HCI, pH 7.4. The reaction mixtures
for the different enzymes were supplemented as follows: for FUrd and
FdUrd phosphorylase, with 2 mM inorganic phosphate, 50 MM[14C]FUrd,
or [14C]FdUrd and 10 to 50 n\ of cell extract; for FUrd kinase, with 2 mM
ATP, 2 mM MgCI2, 50 MM [14C]FUrd, and 10 to 50 n\ of cell extract; for
FdUrd kinase and TK, with 2 mM ATP, 2 mM MgCI2, 0.1 to 50 MM [3H]FdUrd, or 0.1 to 50 MM[3H]dThd, and cell extract sufficient to achieve 5
cells/ml) and 100 n\ of RPMI Medium 1640 with 10% HS, containing
twice the final volume concentration of labeled nucleoside and terminated
by centrifugation for 15 s. To achieve good mixing of cells with the [3H]-
to 3 /il and the appropriate volume (0.2 ml) mixed by rapid injection. For
uptake experiments over longer time periods (up to 40 min), cells and
radioactive substrates were stirred gently with a small magnetic stirring
bar to prevent clumping. Cell suspensions (200 //I) were transferred by
pipet onto the oil layer, and the tubes were centrifuged as above. The
centrifugea tubes were frozen in dry ice and methanol, the lower layer
containing the cell lysate was removed by slicing and transferred to
plastic vials containing 0.2 ml of H2O and vortexed, and total radioactivity
was determined in a Beckman liquid scintillation counter with 5 ml of
Liquiscint.
The total water volume of the suspended cells and the amount of
extracellular medium carried into the pellet were determined by incuba
tion for up to 2 min with [14C]sucrose and 3H2O. The internal water
volume was then calculated by the difference between the pellet water
volume and the external water space as measured by 14C.
Samples were prepared for HPLC analysis by removing the cell lysates
from the bottom layer of the microfuge tubes and washing the tubes
with 0.25 ml of 5% TCA. The lysate and washes were heated at 80°C
to 20% product formation; and for FUra phosphoribosyltransferase,
with
200 MM phosphoribosyl pyrophosphate, 2 mM MgCI2, 50 ut* [14C]FUra,
CANCER
RESEARCH
for 20 min to convert nucleotides to the corresponding monophosphates.
The samples were mixed with an equal volume of Freon containing 0.5
VOL. 45 JULY 1985
3156
Downloaded from cancerres.aacrjournals.org on June 15, 2017. © 1985 American Association for Cancer Research.
DEFECTIVE
TRANSPORT
IN HCT-8 CELLS
Table 2
Table 1
Effect of FdUrd and dThd concentrations on kinase activity
Reaction conditions and analysis as described in "Materials and Methods."
Activity of enzymes of pyrimidine metabolism
Reaction conditions are described in 'Materials and Methods.*
protein/h)FdUrd
TS"
kinase
FUrd kinase
FUra phosphoribosyltransferase
FUrd phosphorylase
FdUrd phosphorylase
DHFRdSensitive15.6
Activity (nmol/mg
protein/h)Sensitive51.1
(nmol/mg
Substrate concentrations
(MM)FdUrd
±10.7"
±8.4
30.2 ±7.9
27.2 ±13.7
40.8 ±7.8
49.8 ±12.2
6.9 + 4.1
5.9 + 1.6
0.7 + 1.4
2.1 ±1.5
1.0 + 1.7
3.4 ±0.8
13.1 ±6.1Resistant10.2
11.6 ±3.9
50
1.00.3
4.8
1.7
0.7112.8
0.1dThd
* Mean ±SD of at least 3 experiments.
50
1,0
0.3
0.1^52.0ta<5
6 Measured by the tritium displacement assay; see "Materials and Methods."
c Activity is expressed as jimol/mg protein/h.
4.5
2.1
0.7-Xt
M trioctylamine, vortexed for 10 s, and centrif uged to separate the organic
phase (lower) from the aqueous solution of nucleotides.
FdUrd and FdUMP were separated by HPLC on a strong anión
exchanger (Whatman Partisil SAX; 10) using isocratic elution with 15 rriM
sodium phosphate, pH 3.3.4 The column temperature was maintained at
35°C, and sufficient amounts of standard FdUrd and FdUMP were
coinjected with samples to identify their elution positions based on
absorbance at 280 nm.
Binding of [3H]NBMPR (30 Ci/mol) was determined by incubation with
suspensions of sensitive and resistant cells at 0.1 and 1.0 nw as
described by Cass ef al. (6). The specificity of binding was determined
by incubation with 6 nw of 6-nitrobenzylthioguanosine.
4.8
1.50.799.37.1
1.9o^r
o^^Seiwitivex"Resistant\
1.5p
x^/*x^,IOResistant33.6
O^
1.0oO.0.50uo-X^"
°"!
'
-,D
RESULTS
\I5
The line of HCT-8 cells selected for resistance to FdUrd and
used in all subsequent experiments was 700-fold resistant (50%
effective dose, 1.5 A/M),and resistance was stable after passage
in the absence of drug for 3 months.
Minimal differences (<2-fold) were observed between the ac
tivities of TS and DHFR in sensitive and resistant cells, and the
activities of FdUrd phosphorylase, FUra kinase, FUrd phospho
rylase, and FUra phosphoribosyltransferase were indistinguish
able in the 2 lines (Table 1). All enzymes of pyrimidine nucleoside
metabolism, including FdUrd kinase, were measured using sat
urating concentrations of substrates (50 JÕM),
and in none of the
6 experiments was the activity of the kinase in sensitive cells
greater than twice that of resistant cells (Table 1). Since a
reduced affinity of FdUrd kinase could cause resistance at the
concentrations of drug encountered in cultures, the enzyme was
assayed using both FdUrd and dThd at lower concentrations
(Table 2); the kinase activity in both cell lines was similar at every
concentration of both substrates.
To determine if nucleotide phosphatase activity was a contrib
uting factor to resistance, the activity of reaction mixtures con
taining equal amounts of extract from sensitive and resistant
cells was assayed for kinase activity. Since the activity was the
algebraic sum of the activity in each extract, such a mechanism
of resistance is indirectly excluded.
Despite the unchanged activity of FdUrd kinase in the resistant
cells, a culture of this line incubated with [3H]FdUrd failed to
accumulate FdUMP and other 3H derivatives at a concentration
of FdUrd that resulted in substantial accumulation of nucleotides
in the sensitive line over a 20-min period (Chart 1). In the extracts
of sensitive cells, the proportion of total intracellular 3H present
4 Me Guire and J. R. Bertino, unpublished method.
20
5Activity
Minutes
Chart 1. FdUMP formation by HCT-8 cells from [3H]FdUrd (0.1 MM). After ex
posure to 0.1 ,<M[3H]FdUrd for 1, 5, 10, and 20 min, 5X105 cells in 200 pi of
medium were centrifugea in microcentrifuge tubes containing 100 ¡Aof silicon:paraffin oil on 100 rf of 15% TCA (see "Materials and Methods"). After
appropriate extraction, the cell lysates were analyzed for FdUrd and FdUMP as
outlined in "Materials and Methods."
as [3H]FdUMP increased from less than 10% at 12.5 s to 65%
at 10 min and approximately 80% at 20 min (Chart 2, A and B).
Consideration of both the enzyme and whole-cell phosphorylation data suggested a defect in the entrance of FdUrd into the
resistant cell line by the nonspecific facilitated diffusion mecha
nism for nucleosides. When entrance of free nucleoside and its
phosphorylation were compared in both lines over short time
periods (2 to 40 s), a marked difference was observed (Chart 2,
A and B). The percentage of [3H]FdUMP as a function of the
total radioactivity in the acid-soluble fractions of the 2 lines
exposed to 0.1 MM[3H]FdUrd for up to 42 s was sufficiently small
to permit these data to be interpreted as a measure of entry of
free nucleoside into the cells. It was possible to retard the entry
of nucleoside by reducing the temperature to 4°C,thus expand
ing the time course of the initial phase of nucleoside entry (Chart
2D). These data are entirely consistent with those observed at
25°C.
Proof that the entry of nucleoside in the sensitive line at the
early time periods reflected facilitated diffusion was obtained by
measurement of the extracellular fluid carried through the oil
using [14C]sucrose and the total cellular water by 3H2O. Further
more, this process was sensitive to blockade by both dipyridamole and NBMPR, potent inhibitors of facilitated diffusion (14)
(Chart 3). In the resistant line, even after 20-min incubation,
CANCER RESEARCH VOL. 45 JULY 1985
3157
Downloaded from cancerres.aacrjournals.org on June 15, 2017. © 1985 American Association for Cancer Research.
DEFECTIVE
TRANSPORT
IN HCT-8 CELLS
4°C
25°C
(22X1
'Stntitivt
in
O
RttitMnt
R«ii*tont
Chart 2. Transport and anabotism of [3H]FdUrd in HCT-8 cells. Cells (5 x 10«/ml)
were
incubated with 0.1 w [3H]FdUrd for s (short
term) or min (long term) at 25°Cor 4°Cand
assayed for FdUrd uptake as described in "Ma
terials and Methods." Numbers in parentheses,
total percentage of [3H]FdUMPover the total
intracellular 3H as determined by HPLC analy
sis; the balance was present as [3H]-
o.
u
IO
20
30
40
Seconds
20
40
60
Seconds
FdUrd.
Minutes
20
TabteS
Inhibition of colony growth of sensitive and resistant HCT-8 cells by a variety of
antineoplasticagents
Colony growth ED»"
—
Magnitude of
Sensitive
Resistant
resistance
Drug
(HM)
(,,M)
(-fold)
[3H]FdUrd.O.I/iM
15
IO
N
'o
FdUrd5'-lododeoxyuridineFUrdS'-Bromodeoxyuridine5'-Deoxy-5-fluorouridine1
IO
20
Seconds
30
40
-/3-o-ArabinofuranosylcytosinedWPyrazofurin6-Methylmercaptopurine
Chart 3. Effect of nucleosidetransport inhibitors on trie initial [3H]Fdllrd uptake
in sensitive HCT-8 cells. Cells (5 x 10«/ml)
were incubated with 0.1 MMpHJFdUrd
(1 to 2 Ci/ml) at 25°Cwith or without the inhibitor and assayed for FdUrd uptake
as described in 'Materials and Methods." Ü,sucrose (extracellular medium that is
ribosideFUra6-MercaptopurineMTXCisplatinMitomycin
carried into the pellet); •H2Ospace (total pellet water).
FdUrd did not achieve concentrations by the oil stop method that
exceed the amount of extracellular water that is carried into the
pellet as measured by 114C(sucrose. The virtual absence of any
CDoxorubicinMethyl-CCNUvuotatine0.0034.00.0056.64.50.0161200.050.0422.1130.0162.80.01
entry in the resistant cell line was confirmed not only at 0.1 UM
FdUrd but also at 1.0 and 10 »M(data not shown).
Saturation of [3H]NBMPR binding to the sensitive cell line, a
8 ED»,concentration of drug causing 50% reduction in colony number after
continuous exposure (10 days) compared to untreated controls; Methyl-CCNU,
methyl-1-(2-chloroethyl)-3-cyclohexyl-1-nitrosourea.
* P value, 0.01.
measure of the functional state of the transporter, was observed
at approximately 0.3 nw. The number of binding sites on the
sensitive cell line (2 x 10s NBMPR per cell) is approximately the
same number noted with the S-49 lymphoma line (6). Nitroben-
resistanceto a varietyof nucleosideanaloguesshould be ob
servedwith the resistantcell line despitedifferentmechanisms
of activationand target enzymes.Indeed,there was extensive
zylthioguanosine (6 fiM) completely prevented binding in the
cross-resistanceto otherpyrimidinedeoxynucleosideanalogues
sensitive cell line. In identical experiments with the resistant cell
line, binding of [3H]NBMPR was undetectable (<3 x 103 sites/
(Table3). Cross-resistanceof a lesserdegreeto FUrd,pyrazofurin,andeven6-methylmercaptopurine
ribosidewas alsonoted
cell).
The defect in nucleosidetransport suggested that cross- despitetheiractivationby differentkinases.Thelowerdegreeof
CANCER
RESEARCH
VOL. 45 JULY 1985
3158
Downloaded from cancerres.aacrjournals.org on June 15, 2017. © 1985 American Association for Cancer Research.
DEFECTIVE
TRANSPORT
cross-resistance to very high concentrations of dThd as well as
5'-deoxy-5-fluorouridine
may reflect nonfacilitated entry inde
pendent of a carrier-mediated transport.
Consistent with a primary nucleoside transport defect in this
cell line was its lack of cross-resistance to all other cytotoxic
agents tested, including purine and pyrimidine analogues.
A defect in nucleoside transport should render cells insensitive
to dThd protection against MTX toxicity in the presence of
hypoxanthine. Either hypoxanthine (100 /ÃŒM)
or dThd (10 ftu)
alone afforded only minor protection of sensitive cells from a
concentration of MTX that produces complete arrest of cell
growth (Chart 4). The 2 agents together allowed a normal growth
rate of the sensitive line until they had completely utilized the
components that circumvent the metabolic blockade (Chart 4A).
Conversely, the growth of resistant cells was completely inhibited
by the same concentration of MTX despite the presence of
hypoxanthine and dThd (Chart 48). Increasing the concentration
of hypoxanthine to 300 /IM and dThd to 30 »Mcaused only slight
increase in growth over that with 100 ^M and 10 MM.respectively
(data not shown).
DISCUSSION
It is remarkable
that
acquisition
of 700-fold
resistance,
achieved by exposure to increasing concentrations of FdUrd,
was associated with minimal differences in any of the activating
or target enzymes for this analogue. However, the linear increase
in intracellular FdUMP in sensitive cells incubated with [3H]FdUrd
and the absence of nucleotide formation by the resistant cells
reflect an almost quantitative difference between the cell lines.
IN HCT-8 CELLS
Three possibile changes in resistant cells could explain this
discrepancy: (a) rapid breakdown of FdUMP by elevated cytoplasmic nucleotide phosphatase activity; (b) reduced affinity of
FdUrd kinase for the substrate; and (c) impaired facilitated diffu
sion of the nucleoside. The first hypothesis was excluded by
mixing experiments, and the second, by assaying kinase activity
at a range of substrate concentrations. However, the rapid "oil
stop" data clearly indicate that resistant cells have an essentially
complete deletion of the facilitated diffusion mechanism for entry
of FdUrd. Even more remarkable is the undetectable permeability
of these cells to FdUrd for as long as 20 min. Evidence that the
resistant cell line has a greatly reduced binding capacity for
[3H]NBMPR gives further credence to this mechanism of resist
ance and may reflect either a deletion of the carrier or the
synthesis of an altered form that is incapable of binding NBMPR.
Plageman and Wohlhueter (15) have reviewed the topic of
"presumptive cell clones defective in (nucleoside) transport."
There are numerous reports about cell lines that behave in a
manner consistent with a defective nucleoside carrier (5,10,12),
but convincing evidence for such a mutant has perhaps been
produced only for a mutant subline of the S-49 mouse T-cell
lymphoma line isolated by Cohen ef al. (7) by a single-step
selection in the presence of an adenosine deaminase inhibitor.
Characterization of the deletion of nucleoside transport system
in this line has been presented by Cass ef al. (6). The current
data present the first direct evidence of acquired resistance to
FdUrd and impaired nucleoside transport as a mechanism of
resistance to FdUrd in a colon carcinoma cell line. These results
may be clinically very relevant in view of the human origin of
these HCT-8 cells. Of interest is the preliminary observation that
I0r
IO1
A. Sensitive
B. Resistant
•Control
8 Control
I0f
(A
O
£ 10
ai
o
A MTX+dTM
»Hyp
SMTX + H«.
MTX»dThd
IO4
MTX
2345
23456
Days
Days
Chart 4. dThd:hypoxanthine (Hyp) protection of sensitive and resistant HCT-8 cells from MTX cytotoxicity. MTX (0.1 nu), dThd (10 <IM),and hypoxanthine (100 MM)
alone or in different combinations were added at the time of plating and remained in the culture for the entire period of incubation. Duplicate cultures were collected daily
by 30-min trypsimzation, and the cell number was determined as described in "Materials and Methods."
CANCER
RESEARCH
VOL. 45 JULY 1985
3159
Downloaded from cancerres.aacrjournals.org on June 15, 2017. © 1985 American Association for Cancer Research.
DEFECTIVE TRANSPORT
2 other HCT-8 lines, independently selected for high levels of
resistance to FdUrd in this laboratory, may also be transport
mutants with minimal enzyme changes. It could be hypothesized
that colon cancer cells have a unique tendency to develop
resistance to nucleoside analogues by deletion of the mechanism
for facilitated diffusion under selection pressure with FdUrd. This
may relate to the major physiological role of colonie epithelial
cells in nutrient transport and the potentially high sensitivity of
this portion of the genome to mutation under selective pressure.
In progress is a survey of a variety of renal and colon carcinomas
for the rate at which resistance selection in these lines is asso
ciated primarily with a transport defect. The in vitro conditions
used to select for FdUrd resistance (stepwise increase in the
concentration of the nucleoside over a period of 6 months) may
find its clinical equivalent in those patients treated with intrahepatic arterial infusion of FdUrd for liver métastasesfrom colorectal cancer. This new modality of regional chemotherapy admin
istration (8) involves the intermittent (2 weeks on and 2 weeks
off) long-term infusion (up to 1 year) of FdUrd via an impiantatale
pump. This clinical situation in many respects mimics the means
used in this paper to select the resistant cell line and could
generate similar resistant clones in patients.
Cross-resistance to purine and pyrimidine nucleoside ana
logues, but not to base analogues, was expected and confirmed
the mechanism of resistance to FdUrd. The somewhat lower
cross-resistance to the arabinoside and riboside analogues may
possibly reflect a degree of selectivity in the transport mecha
nism; however, earlier evidence for such specificity is minimal.
The lipophilic quality of 5'-deoxy-5-fluorouridine may permit pas
sive diffusion through the membrane and account for the lower
degree of cross-resistance to this agent.
It may seem difficult to reconcile the very great difference
between the ability of low concentrations of dThd (<1.0 UM) in
combination with hypoxanthine to confer protection from MTX
cytotoxicity in the sensitive and resistant lines, as reported in the
following paper, with the minimal increase in the very high
concentrations of dThd needed to cause cytotoxicity in both
lines (120 and 330 A/M)(Chart 4). It is possible, however, that at
these very high concentrations, passive diffusion may overcome
the lack of facilitated diffusion or that other mechanisms of
cytotoxicity at extracellular sites may intervene.
It appears that the mode of resistance reported for this cell
line is not associated with the nonspecific resistance to several
anticancer drugs of large dimension that express the membrane
"P-glycoprotein" as reported by Riordan and Ling (17), since no
cross-resistance
to colchicine, vinblastine, doxorubicin and acti-
nomycin D was seen.
Although collateral sensitivity to any of the 18 agents tested
was not observed, selective kill of the cells that display this
resistant phenotype can be achieved by exposure to agents
whose cytotoxicity can be reversed by nucleosides; sensitive cell
lines would be protected under these conditions. The prevention
of this potential to protect against MTX toxicity by dThd in
combination with hypoxanthine in FdUrd-sensitive cells might
also be achievable in normal marrow or intestinal cells of patients,
while any colon carcinoma cells resistant by virtue of deletion of
nucleoside transport or even dThd kinase would remain fully
CANCER RESEARCH
IN HCT-8 CELLS
sensitive. The potential therapeutic consequences of this finding
as well as other selective drug combinations for human colon
cancer cells resistant to FdUrd are more fully discussed in the
following companion paper.
ACKNOWLEDGMENTS
We are greatly indebted to Dr. A. R. P. Paterson for helpful discussions of the
transport experiments and a generous supply of S-[3H]nitrobenzylthioinosine. The
experimental participation of Dr. J. Damowski and Chris Holdndge in the binding
experiments has been invaluable, and Dr. M. Jastreboff kindly supplied the other
HCT-8 resistant cell lines for comparison.
REFERENCES
1. Bapat, A., Zarow, C., and Danenberg, P. V. Human teukemic cells resistant to
5-fluoro-2'-deoxyuridine contain a thymidylate synthase with lower affinity for
nucteotides. J. Biol. Chem., 258: 4130-4136,1983.
2. Baskin, F., Carlin, S. C., Kraus, P., Friedkin, M., and Rosenberg, R. N.
Experimental chemotherapy of neuroblastoma. II. Increased thymidylate syn
thetase activity in 5-fluorodeoxyuridine-resistant
variant of mouse neuroblas
toma. Mol. Pharmacol., 11:105-117,1975.
3. Belt, J. A. Heterogeneity of nucleoside transport in mammalian cells. Two
types of transport activity in L1210 and other cultured neoplastic cells. Mol.
Pharmacol., 24: 479-484,1983.
4. Bradford, M. M. A rapid and sensitive method for the quantitation of microgram
quantities of protein utilizing the principle of protein-dye binding. Anal.
Btochem., 72: 248-254,1976.
5. Breslow. R. E., and Goldsby, R. A. Isolation and characterization of thymidine
transport mutants of Chinese hamster cells. Exp. Cell Res., 55: 339-346,
1969.
6. Cass, C. E., Kolassa, N., Uehara, Y., Dahlig-Hartey, E., Hariey, E., and
Paterson, A. R. P. Absence of binding sites for the transport inhibitor nitrobenzylthioinosine on nucleoside transport-deficient
mouse lymphoma cells.
Biochim. Biophys. Acta, 649: 769-777,1981.
7. Cohen, A., Ullman, B., and Martin, D. W., Jr. Characterization of a mutant
mouse lymphoma cell with deficient transport of purine and pyrimidine nucleo
sides. J. Btol. Chem., 254:112-116,1979.
8. Ensminger, W. D. Therapeutic advances using regional chemotherapy. Proc.
Am. Soc. Cancer Res., 25: 410-412,1984.
9. Femandes, D. J., and Carroll, C. C. Multiple modes of resistance of human
CCRF-CEM teukemic cells to 5-fluoro-2'-deoxyuridine. Proc. Am. Soc. Cancer
Res., 24: 283, 1983.
10. Freed, J. J., and Mezger-Freed, L. Origin of thymidine kinase deficient (TIC)
haploid frog cells via an intermediate thymidine transport deficient (TT~) phe
notype. J. Cell Phystol., 82: 199-212,1973.
11. Gupta, S. V., Greenfield, N. J., Poe, M., Makylu, D. R., Williams, M. N.,
Moroson, B. A., and Benino, J. R. Dihydrotolate reducÃ-ase from a resistant
subline of the L1210 lymphoma. Purification by affinity chromatography and
ultraviolet difference spectrophotometric and circular dichroic studies. Bio
chemistry, 76: 3073-3079,1977.
12. Lynch, T. P., Cass, C. E., and Paterson, A. R. P. Defective transport of
thymidine by cultured cells resistant to 5-bromodeoxyuridine. J. Supramol.
Struct., 6. 363-374,1977.
13. Mulkins, M. A., and Heidelberger, C. Biochemical characterization of fluoropyrimidine-resistant murine teukemic cell lines. Cancer Res., 42: 965-973,
1982.
14. Paterson, A. R. P., Law, E. Y., Dahlig. E., and Cass, C. E. A common basis
for inhibition of nucleoside transport by dipyndamole and nitrobenzylthioincsine? Mol. Pharmacol., 78: 40-44,1980.
15. Placeman, P. G. W., and Wohlhueter, R. M. Permeation of nucleosides, nucleic
acid bases, and nucleotides in animal cells. Curr. Top. Membr. Transp., 14:
225-330,1980.
16. Priest, D. G., and Ledford, B. E. Increased thymidylate synthetase in 5fluorodeoxyuridine resistant cultured hepatoma cells. Btochem. Pharmacol.,
24: 1544-1553,1980.
17. Riordan, J. R., and Ling, V. Purification of P-glycoprotein from plasma mem
brane vesicles of Chinese hamster ovary cell mutants with reduced colchicine
permeability. J. Biol. Chem., 254:12701-12705,1979.
18. Roberts, D. An isotopie assay for thymidylate synthetase. Biochemistry, 5:
3546-3548,1955.
19. Tompkins, W. A. F., Watrach, A. M., Shmate, J. D., Shultz, R. M., and Harris.
J. A. Cultural and antigenic properties of newly established cell strains derived
from adenocarcmomas of the human colon and rectum. J. Nati. Cancer Inst.,
52: 1101-1110,1974.
20. Umeda, M., and Heidelberger, C. Comparative studies of fluorinated pyrimidines with various cell lines. Cancer Res., 28: 2529-2538,1968.
VOL. 45 JULY 1985
3160
Downloaded from cancerres.aacrjournals.org on June 15, 2017. © 1985 American Association for Cancer Research.
Defective Facilitated Diffusion of Nucleosides, a Primary
Mechanism of Resistance to 5-Fluoro-2 ′-deoxyuridine in the
HCT-8 Human Carcinoma Line
Alberto F. Sobrero, Robyn D. Moir, Joseph R. Bertino, et al.
Cancer Res 1985;45:3155-3160.
Updated version
E-mail alerts
Reprints and
Subscriptions
Permissions
Access the most recent version of this article at:
http://cancerres.aacrjournals.org/content/45/7/3155
Sign up to receive free email-alerts related to this article or journal.
To order reprints of this article or to subscribe to the journal, contact the AACR Publications
Department at [email protected].
To request permission to re-use all or part of this article, contact the AACR Publications
Department at [email protected].
Downloaded from cancerres.aacrjournals.org on June 15, 2017. © 1985 American Association for Cancer Research.