ICANCER RESEARCH 37, 2529-2537,
June 15. 19971
Increased Facilitated Transport of Dehydroascorbic Acid without Changes in
Sodium-dependent Ascorbate Transport in Human Melanoma Cells'
Charles Spielholz, David W. Golde, Alan N. Houghton, Francisco Nualart, and Juan Carlos Vera2
Programs in Molecular Phannacology and Therapeutics (C. S., D. W. G., J. C. V.1 and Immunology (A. N. H.], Memorial Sloan-Kettering
10021, and Departamento de Histologia y Embriologia, Facultad de Ciencias, Universidad de Concepción, Concepcidn, Chile (F. N.J
ABSTRACT
Many cell types transport vitamin C solely In its oxidized form, dehy.
droascorbic acid, through facilitative glucose transporters. These cells
accumulate large intracellular concentrations of vitamin C by reducing
dehydroascorbic acid to ascorbate, a form that is trapped Intracellularly.
Certain specialized cells can transport Vitamin C in its reduced form,
ascorbate, through a sodium-dependent cotransporter. We found that
normal human melanocytes and human malignant melanoma cells are
able to transport vitamin C using both mechanisms. Melanoma cell lines
‘
Cancer Center. New York, New York
Vitamin C has been proposed to play a role in melanin synthesis
(8—10)and to affect growth, differentiation, and metabolism in mela
nocytes and melanoma (11—23).
Most mammals synthesize vitamin C in the liver from six carbon sugar
precursors and then transport it to different cells and tissues (24). Some
mammals, such as primates and guinea pigs, have lost the capacity to
synthesize
ascorbic
acid, so the vitamin
must be obtained
in the diet (7,
25). Although the ability to transport and trap vitamin C is important to
human cells, the mechanisms underlying these processes are just being
transported
dehydroascorbic
acid at a rate that was at least 10 times
investigated. Our laboratory has shown that in many cell types, including
greater than the rate of transport by melanocytes,whereas both mela fresh neutrophils (26), cultured myeboid leukemia (HL-60) cells (27), and
nomacellsandmebanocytes
transportedascorbatewithsimilarefficiency.
normal and malignant breast cells,3 vitamin C is transported through
Dehydroascorbic acid transport was inhibited by deoxyglucose and cy.
facilitative glucose transporters in its oxidized form, dehydroascorbic
tochalasin B, Indicating the direct participation of facilitative glucose
transporters In the transport of oxidized vitamin C. Melanoma cells acid. Once inside the cell, dehydroascorbic acid is rapidly converted into
the reduced form of vitamin C, ascorbic acid (27). Reduced ascorbate is
accumulated intracellular
vitamin C concentrations
that were up to 100
trapped inside these cells because it is not transportable through facilita
times greater than the corresponding extracellular dehydroascorbic acid
concentrations, whereas Intracellular accumulation of vitamin C by mela
tive glucose transporters (27, 28). Conversion of dehydroascorbic acid to
nocytes never exceeded the extracellular level of dehydroascorbic acid. reduced ascorbate keeps the intracellular concentration of dehydroascor
Melanoma cells transported dehydroascorbic acid through at least two bic acid low, resulting in the net movement ofdehydroascorbic acid from
different transporters, each with a distinct Km,a finding that agreed well outside the cell to the inside. The concentration of intracellular ascorbic
with the presence of several glucose transporter
isoforms m these cells.
acid can then rise above the extraceblular concentrations of the vitamin.
Only one kinetic component of ascorbate uptake was Identified in both
Our
laboratory has also shown that the process of dehydroascorbic acid
melanocytes and melanoma cells, and ascorbate transport was sodium
transport and the subsequent accumulation of ascorbate inside the cell are
dependent and Inhibited by ouabain. Both cell types were able to accu
two kinetically distinct processes (28). The molecular identity and the
mulate intracellular concentrations of vitamin C that were greater than
functional characteristics of the cellular components responsible for the
the extracellular ascorbate concentrations. The data Indicate that mela
noma cells and normal melanocytes transport vitamin C using two dii
conversion of dehydroascorbic
acid to ascorbate are still controversial
ferent transport systems. The transport of dehydroascorbic acid is medi
(29—35).
ated by a facilitated mechanism via glucose transporters, whereas
A second mechanism of vitamin C transport has also been described.
transport
ofascorbic
acid involves a sodium-ascorbate
cotransporter.
The
Selected cell types may transport vitamin C in the form ofascorbate using
differential capacity of melanoma cells to transport the oxidized form of
a sodium-dependent cotransporter. These cells include osteoblasts (36),
vitamin C reflects the increased expression of facilitative transporters
kidney cells (37—39),small intestine enterocytes (40), ciliary epithelial
associated with the malignant phenotype.
cells (41), astrocytes (42), and adrenomedullary chromaffin cells (43).
Many of these reports, as well as others (44, 45), have been difficult to
interpret because they did not control for the oxidation of ascorbate to
INTRODUCTION
dehydroascorbic acid under their experimental conditions (27, 46, 47). In
The role of vitamin C as a coenzyme in collagen and proteoglycan
addition, a sodium-dependent cotransporter of reduced ascorbate has not
synthesis and its importance in the prevention of scurvy has been
yet been molecularly characterized.
known since the early part of this century, but its functions and
Tumor cells have vastly increased requirements for glucose (48).
physiology in humans are still incompletely understood (1—3).Vita
This fundamental metabobite is supplied from extracellular sources
mm C is found in most, if not all, tissues and in particularly high
and is transported to the interior of the cell by facilitative glucose
concentrations in the adrenal gland, pituitary gland, eye lens, brain,
transporters down a concentration gradient (49). Glucose is then
pancreas, liver, and spleen (4). In addition to its function as a general
trapped inside the cell by conversion to glucose-6-phosphate
by
antioxidant (3), vitamin C may play a role in maintaining certain
appropriate hexokinases, the first step of glucose metabolism. To
enzymes in their reduced form (3) and may be involved in the
compensate for their increased need to transport glucose, tumor cells
synthesis of carnitine (5) and certain neuroendocrine peptides (6).
express increased levels of facilitative glucose transporters (50, 51).
There is also evidence that vitamin C is involved in male fertility (7).
Because efficient uptake of dehydroascorbic acid occurs through
facilitative glucose transporters, the question arises as to whether
Received12/2/96;accepted4/22/97.
tumor
cells also display increased uptake of vitamin C.
The costs of publication of this article were defrayed in part by the payment of page
We report here that cultured normal human melanocytes and mel
charges. This article must therefore be hereby marked advertisement
in accordance
with
18 U.S.C. Section 1734 solely to indicate this fact.
anoma cells use two distinct mechanisms to transport and accumulate
I Supported
by Grants
R0l
CA30388,
ROl HL42107,
and P30 CA08748
from
the NIH
vitamin C. Both cell types are able to transport ascorbate unidirec
and
grants
from
Memorial
Sloan-Kettering
Cancer
Center
Institutional
Funds,
the
Schultz
Foundation, the DeWitt Wallace Clinical Research Foundation, and Grant Diuc
tionally through a sodium-dependent, active cotransporter and are also
96035001-1 1 from the Direccibn de Investigacibn, Universidad de Concepcibn, Chile.
2 To
whom
requests
for
reprints
should
be addressed,
at Memorial
Sloan-Kettering
Cancer Center, 1275 York Avenue, New York, NY 10021. Fax: (212) 772-8550.
3 Unpublished
observations.
2529
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1997 American Association for Cancer Research.
VITAMIN C UPTAKE BY MELANOMA
able to transport dehydroascorbic acid through facilitative glucose
transporters. We show that melanoma cells accumulate up to 100-fold
more vitamin C than normal mebanocytes because of a substantial
increase in transport of dehydroascorbic acid through facilitative
glucose transporters. Melanoma cells show no change in the net
transport of reduced ascorbate relative to normal melanocytes. The
neoplastic phenotype of the melanoma cells is characterized by a
major up-regulation in the ability to take up dehydroascorbic acid via
facilitative transport without modulation of sodium-dependent co
transport of ascorbate.
CELLS AND MELANOCYTES
ylgiucose until equilibrium
inside
was reached (27). The amount of methylglucose
the cells at equilibrium
can be used to estimate
the internal
volume
of
the cells that is free to exchange with the extracellular medium. All cells
reached equilibrium accumulation of methylglucose within 2 h, and the inter
nal volume was estimated to be 1 @l
for 106melanoma cells and 0.5 ,.d for 106
melanocytes. Intracellular concentrations are calculated based on the amount
of uptake and the internal volume of the cells and are expressed in molar units.
The Michaelis constant, Km,the concentration of metabolite required to reach
one-half maximal velocity of transport or accumulation, was calculated using
a Lineweaver-Burk
analysis (57). The IC@, the concentration
of inhibitor
required to produce a 50% decrease in transport or accumulation of metabo
bites, was calculated according to standard methods (57).
Immunodetection
MATERIALS AND METHODS
of glucose
grown on Lab-Tek chamber
transporter
microscope
types
was
performed
on cells
slides (Nunc, Naperville,
IL) as
Deoxyglucose,4 L-glucose,D-fructose,methylglucose, methyl a-D-glucopy
ranoside, L-ascorbicacid (sodium salt), ouabain, cytochabasinB, and cytocha
lasin E were obtained from Sigma Chemical Co. (St. Louis, MO). l-'4C-
described (58) using the peroxidase-antiperoxidase system to visualize the
immune reaction. The antibodies for GLUT1, GLUT2, GLUT3, and GLUT4
LAscorbic
cabs, Southbridge,
acid
(4.75—8.20
mCi/mmol),
[l,2-3H(N)]-deoxyglucose
are well characterized,
(greater
than 30 Ci/mmol), and 3-O-[me:hyl-3H]methylglucose(86.70 Cilmmol) were
obtained from DuPont NEN (Boston, MA). Fetal bovine serum was obtained
from Gemini (Calabasas, CA). Other cell culture reagents were obtained from
Life Technologies, Inc. (Gaithersburg, MD) unless noted otherwise. Prepara
tions of ascorbic
acid were oxidized
to dehydroascorbic
acid by the addition
at 265 nM and HPLC (52). For assays that required concentrations
acid
greater
than
1 msi,
dehydroascorbic
acid
from
Aldrich Chemical Co. (Milwaukee, WI) was used to reach the final dehy
droascorbic acid concentration. Preparations of ascorbic acid were kept in the
reduced form by the addition of 1 mMDTT to 6.1 mMascorbic acid in 30 mM
HEPES (pH 5.5) as described (27).
The human melanoma cell lines, clone 3.44 of SK-MEL-l3l and clone 1-5
of SK-MEL-b3l , SK-MEL-l88, SK-MEL-23, SK-MEL-22A, and SK-MEL
29, were grown
in RPMI
1640 supplemented
with
10% fetal bovine
serum,
penicillin, and streptomycin (53, 54). Strains of normal human epidermal
melanocytes (neonatal) were obtained from Clonetics (San Diego, CA) and
were grown to confluency according to directions using their media and
supplements. In addition, two strains of normal human epidermal melanocytes,
FS 373 and FS 374, were obtained from Dr. Anthony Albino (Sloan-Kettering
Institute)
and were grown
as described
(55). Although
we have not noted any
differences in the uptake of vitamin C by cells in different stages of confluence,
all uptake experiments were performed in 12-well plates with approximately
80,000—100,000cells/well. Tumor cell viability was greater than 95%, as
determined by trypan blue exclusion. Viability of normal cells was 100%. All
cells were deprived of fetal bovine serum, growth factors, and other media
additives for 16 h before any uptake assays were performed.
The uptake of ascorbic acid, dehydroascorbic acid, deoxyglucose, and
methylglucose was measured by a modification of procedures described pre
viously (27, 28). Cells were transferred to incubation media [15 mMHEPES
(pH 7.4), 135 mMNaCI, 5 m@iKCI, 1.8 mrsiCaCl2,0.8 mMMgCI2,and 0. 1 msi
DTFJ containing labeled ascorbic acid, dehydroascorbic acid, deoxyglucose, or
methylglucose as described in the figure legends. The final specific activity of
ascorbic acid or dehydroascorbic acid used in all uptake assays ranged from
2500 to 5000 cpm/nmol. After the labeling period, cells were washed four
times in ice-cold PBS (0.9% NaCI, 25 mrsiKPO@,pH 7.4) and solubilized in
0.5
ml
of
0.2%
SDS
in
10 mai
Tris
(pH
8.0).
Cellular
uptake
of
antibodies
(East Acres Biologi
Melanocytes and Melanoma Cells Transport Reduced and Ox
of dehy
obtained
available
RESULTS
of
0.02 unit of ascorbate oxidase (Sigma) per ml of 1 @M
ascorbic acid in the
presence of0.l mMDli' as described (27, 28). The conversion of ascorbic acid
to dehydroascorbic acid was confirmed spectrophotometrically by absorbance
droascorbic
commercially
MA).
idized Vitamin C but Melanoma Cells Show Preferential Trans
port of Dehydroascorbic
Acid. Time course analysis of vitamin C
uptake revealed that melanoma cells and melanocytes were able to take
up both the reduced form, ascorbic acid, and the oxidized form, dehy
droascorbic acid (Fig. 1, A and B). However, they showed clear quanti
tative differences in their respective abilities to take up reduced or
oxidized vitamin C. Melanoma cells (SK-MEL-l3l, clone 3.44) took up
approximately 6 nmol of vitamin C per l0@ cells in 30 mm when
provided with 50 @tM
dehydroascorbic acid (Fig. 1A). After 30 rain, the
melanoma cells continued to take up dehydroascorbic acid, reaching
nearly 9 nmol per l0@cells by 60 mm. In contrast, uptake ofascorbic acid
occurred very slowly, with the result that the melanoma cells were able
to take up only 300 pmol of the vitamin per 106 cells in 30 mm when
provided with 50 @.1.M
ascorbic acid (Fig. 1A). At 60 mm, uptake ap
proached only 500 pmol per l0@cells. Thus, although melanoma cells
showed a remarkable capacity to take up dehydroascorbic acid, they had
only a modest ability to take up ascorbic acid.
In contrast to melanoma cells, normal melanocytes showed only a
modest capacity to take up both forms of vitamin C, and they showed
preferential uptake of ascorbic acid as opposed to dehydroascorbic
acid (Fig. 1B). Uptake of dehydroascorbic acid occurred very slowly;
mebanocytes (strain 2486, from Clonetics) took up less than 200 pmol
of the vitamin per 106 cells in 30 mm when provided with 50 p.M
dehydroascorbic acid (Fig. 1B). Uptake leveled off at 30 mm, with no
further increase in uptake observed for the remainder of the experi
ment (Fig. 1B). Mebanocytes showed a greater capacity to take up
ascorbic acid than dehydroascorbic acid; they were able to take up
approximately 600 pmol per 106 cells in 30 ruin when provided with
50
p.M ascorbic
acid
(Fig.
1B).
Although
there
was
a clear
decrease
in
the rate of uptake after the 30-mm period, the melanocytes continued
to take up reduced ascorbate in a linear fashion, reaching approxi
labeled
molecules was measured by scintillation spectroscopy. In some experiments,
cells were also treated with cytochalasin B or E, deoxyglucose, L-glucose, or
other metabolites during the uptake assay as described in the figure legends.
mately 1 nmol of vitamin C per 106 cells at 60 mm. Thus, although
melanocytes and melanoma cells have approximately equal capacities
Transport of dehydroascorbic
more dehydroascorbic acid than melanocytes.
Uptake of dehydroascorbic acid by cells involves at least two steps
(28). The rate of the first step, the transport of dehydroascorbic acid,
can be estimated from the slope of the early linear phase of the time
course of uptake (28). If the rate of the second step, the intracellular
accumulation of ascorbic acid, is lower than that of transport, it can be
estimated from the slope of the linear phase of the batter part of the
time course (28). The time course experiments showed that dehy
ions was accomplished
acid or ascorbic acid in the absence of sodium
by replacing
the NaCI in the incubation
media
with
either LiCl, choline chloride, or sucrose. For all time points tested, the total
time that the cells were in media lacking sodium was the same. Hill coeffi
cients (nH) were calculated by standard methods (56).
Cell volumes were estimated by incubation of cells with 1 mM [3H]meth
4 The abbreviations
used are: deoxyglucose,
2-deoxy-o-glucose;
methylglucose,
methyl-o-glucopyranose: HPLC. high-performance liquid chromatography.
3-0-
to take up ascorbicacid, melanomacells were able to take up 10 times
2530
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1997 American Association for Cancer Research.
V@@AMIN
C UPTAKE BY MELANOMA CELLS AND MELANOCYTES
(1)
8
Fig. 1. Uptake ofdehydroascorbic acid (DHA) or
ascorbic acid (AA) by clone 3.44 melanoma cells
and melanocyte strain 2486. A, time course of ac
cumulationof5Ogssiofeitherdehydroascorbic
acid
(•)
or ascorbicacid (0) by clone3.44melanoma
cells. B, time course of accumulation of 50 p@iof
either dehydroascorbic
CD
0
0
E
C
a
B
.@
0.
acid (•)or ascorbic acid
(0) by melanocytes.C, timecourseof transportof
50 @5M
dehydroascorbicacid by melanocytes(0)
and melanoma cells (•).D, time course of transport
of 50 psi ascorbicacid by melanocytes(0) and
melanoma cells (•).E, transport of increasing con
centrations of dehydroascorbic
acid by melanoma
cells(•)
andmelanocytes(0) duringa 30-sassay.
F, Lineweaver-Burk
E
@61
•1
@-.10
analysis of the low Km com
ponentof dehydroascorbic
acidtransportby mela
noma cells. G, Lineweaver-Burkanalysisof the
@
0
high Km component of dehydroascorbic
acid trans
port by melanoma cells. H, transport of increasing
C
:!
B J
Melanoma
concentrations of ascorbic acid by melanocytes (0)
and melanoma cells (•)during a 1-mm assay.
@
.5
:@0@
8
0
12
10
DHA (mM)
droascorbic acid uptake by melanoma cells and mebanocytes consisted
of at least two distinct kinetic components, as revealed by the biphasic
nature of the slope of the curve representing the uptake data (Fig. 1,
A and B). The first component, transport of dehydroascorbic acid,
occurred within the first 2.5 mm of the time course for both melano
cytes and melanoma cells (Fig. 1, C and D). The rates of dehy
droascorbic acid transport were 220 and 14 pmobper mm per 106cells
for melanoma cells and melanocytes, respectively, indicating that
melanoma cells transported dehydroascorbic acid at a rate that was
approximately 15 times faster than that of melanocytes. The rate of
accumulation was estimated from the linear phase of the uptake curve
from 5 to 30 mm (Fig. 1, A and B). Melanoma cells showed a rate of
accumulation of 140 pmol/min per 106 cells, which corresponds to
about 67% of the rate of transport. For melanocytes, the rate of
accumulation was approximately 6 pmol/min per 106 cells, about 42%
of the rate of transport. Therefore, melanoma cells, in addition to
transporting dehydroascorbic acid at a higher rate than melanocytes,
accumulated vitamin C via dehydroascorbic acid transport at a rate
that was 25 times greater than melanocytes.
Uptake of ascorbic acid by melanocytes and melanoma cells did not
show biphasic kinetics (Fig. 1, A—D).Following the model for dehy
droascorbic acid uptake, we estimated the rate of ascorbate transport
using the first 2.5 rain of the time course (Fig. lD) and the rate of
ascorbic acid accumulation using the linear portion of the time course
after 5 mm (Fig. 1, A and B). For melanocytes, the rate of ascorbic
acid transport and accumulation were similar, 22 and 21 pmol/min per
106 cells, respectively. For melanoma cells, the rates of ascorbic acid
transport and accumulation were 9 and 11 pmol/min per 106 cells,
respectively. Thus, melanocytes transported and accumulated ascorbic
acid at a rate that was approximately 2 times faster than that of
melanoma cells. Total accumulation of vitamin C in melanocytes
incubated in the presence of ascorbic acid or dehydroascorbic acid
was, however, only a fraction of the accumulation of vitamin C by
melanoma cells incubated in the presence of dehydroascorbic acid.
Next, we performed uptake experiments using two other melano
cyte strains (FS 373 and FS 374) and five other melanoma cell lines
(clone 1-5 of SK-MEL-l3l
and SK-MEL-l88, SK-MEL-23, SK
20
-0.2 0.0 0.2 0.4
1/DHA (1/mM)
0.1
0.2
AA (mM)
MEL-22A, and SK-MEL-29; Table 1). These experiments confirmed
and extended the previous observations, indicating that cultured mel
anocyte strains were able to take up more ascorbic acid than dehy
droascorbic acid, and that melanoma cell lines showed a greater
capacity to take up dehydroascorbic acid than ascorbic acid. On
average, the mebanocyte cell strains were abbe to transport and accu
mulate ascorbic acid at a rate that was at least 2-fold greater than that
of dehydroascorbic acid (Table 1). In fact, one melanoma cell line,
clone 1-5 of SK-MEL-l3l, did not take up any measurable ascorbic
acid. On the other hand, the melanoma cell lines were able to take up
greater quantities of dehydroascorbic acid than the melanocyte cell
strains, with a marked increase in the rates of both transport and
accumulation (Table 1). Although experiments to be described next
were performed in several of the cell lines, for simplicity, represent
ative data will be presented for melanoma cells clone 3.44 of
SK-MEL- 13 1 and melanocyte strain 2486.
Dose-response studies using 30-s assays for melanoma cells revealed
that the transport of dehydroascorbic acid saturated at about 10 mi@i(Fig.
1L). Further analysis of the dehydroascorbic acid transport data in mel
anoma cells revealed the presence of two kinetic components, one satu
rating at nearby 1 mistdehydroascorbic acid and a second one approaching
saturation at 10 [email protected] analysis of the data confirmed the
presence of two kinetic components (Fig. 1, F and G). The apparent Km
values were 0.38 and 3.84 msi and V,,,@values were 1.8 and 13 mrsi/min,
respectively, for the transport of dehydroascorbic acid.
Dose-response
studies
using
1-mm
assays
revealed
that
the
trans
port of dehydroascorbic acid by mebanocytes saturated at about 10 mt@i
(Fig. bE). The 1-mm assays were necessary in the melanocyte system
to increase the sensitivity of the measurements.Because of the low
levels of dehydroascorbic acid transport by melanocytes, detailed
Lineweaver-Burk analysis of the data was not possible. However, this
study did confirm that, at all concentrations tested, melanoma cells
(Fig. bE) transported more dehydroascorbic acid than melanocytes.
A detailed analysis of the transport of reduced ascorbate by both
melanoma cells and melanocytes was difficult because of the low
levels of uptake at short incubation times. Dose-response studies using
1-mn assays revealed that the transport of ascorbate by both cell types
2531
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1997 American Association for Cancer Research.
VITAMIN
C UPTAKE
BY MELANOMA
CELLS AND MELANOCYTES
melanocvtes―Transport
Table I Transport. accumulation
rates. and total accumulation of s'itamin C in melanoma cell lines and normal
rate
(pmol/min/l06
DHACelltype
DHAMelanocyte
strains
190±8FS373
2486―
370±30FS374
260±10Melanoma
Accumulation rate
cells)
(pmol/min/106 cells)Total
AA
DHA
AA
22±2
14±2
21±1
6±1640±70
18±6
17±2
16±1
8±1440±80
15±2
15±4
13±1
6±1430±60
accumulation
mm)AA
(pmol/106 cells in 30
linesSK-MELcell
800SK-MEL-l31,
I31, clone 3.44―
640SK-MEL-188clone 1-5
9 ±1
NAC
120SK-MEL-23
ND
1640SK-MEL-22A
ND
880SK-MEL-29
ND
ND
200a
All calculations
b Calculated
1.C
NA,
not
were done from three determinations.
220 ±30
ND
11 ±1
NA
ND
630 ±50
AA, ascorbic
140 ±16330
370 ±100
±70
12±2
140±8340
±60
4140 ±
6 ±2
460 ± 30170
±30
14160±
ND
6 ±I
260 ±20160
±30
7820±
ND
19±I1
330±8560
±60
9960 ±
acid; DHA,
dehydroascorbic
5600 ±
10950 ±
acid.
from data in Fig.
applicable;
ND,
not
determined.
tamed a value of 10 at the highest extracellular dehydroascorbic acid
concentration of 10 mt@i(Fig. 2B). In contrast, normal melanocytes
failed to accumulate high intracellular ascorbic acid concentrations
when incubated in the presence of extracellular dehydroascorbic acid
(Fig. 2A). The ratio of intracellular ascorbic acid versus extracellular
dehydroascorbic acid never reached a value greater than 5 (Fig. 2B).
For example, intracellular ascorbate was only 1.5 mt@iat 3 mt@iextra
Melanoma Cells and Melanocytes Accumulate Vastly Different cellular dehydroascorbic acid. The data indicate that melanoma cells
Intracellular Vitamin C Concentrations. The time course data and melanocytes have marked differences in their ability to accumu
showed that melanoma cells and melanocytes each obtain vitamin C
late intracellular ascorbic acid. At 3 p.M external dehydroascorbic
acid, melanoma cells accumulated approximately 35-fold more ascor
using two different transport systems. One system selectively trans
ports oxidized vitamin C (the dehydroascorbic acid pathway), and a bic acid than melanocytes, and this difference increased to 60-fold at
second system transports reduced vitamin C (the ascorbic acid path
10 m@@i
extracelbular dehydroascorbic acid (Fig. 2C).
In contrast to the marked difference in vitamin C accumulation
way). We examined the kinetics of the uptake process in more detail
to betterdefine the mechanismsinvolved andthe relativecontribution between melanocytes and melanoma cells incubated in the presence of
of each transport system to intracellular vitamin C availability in dehydroascorbic acid, both cell lines accumulated similar intracellular
melanoma cells and melanocytes.
concentrations of ascorbic acid when presented with extracellular
We showed previously that cells that transport dehydroascorbic acid
ascorbic acid (Fig. 2D). Melanocytes accumulated about three times
accumulate intracellular concentrations of ascorbic acid that are much
more ascorbic acid than melanoma cells via the ascorbic acid pathway
at all concentrations tested (Fig. 2, D—F).For example, at 3 p.M
higher than the respective extraceblular concentrations of dehydroascorbic
acid (27, 28). We verified the identity of the chemical form of radiola
extracelbular ascorbate, melanocytes and melanoma cells accumulated
beled vitamin C present in both the incubation medium and cell extracts
intracellular concentrations of 160 and 55 p.M ascorbic acid, respec
by HPLC. Analysis of the incubation media showed that more than 98%
tively, a 3-fold difference in favor of melanocytes. An extracellular
ofthe radioactivityelutedin thepositionofdehydroascorbicacid,andthe ascorbic acid concentration of 3 misi yielded intracellular concentra
peak disappeared and eluted with authentic ascorbic acid when the
tions equal to 1.6 and 0.52 mm@i
for melanocytes and melanoma cells,
samples were treated with 1 m@iD1T before the HPLC separation.
respectively.
These experiments
also showed that, although melano
Greater than 95% of the radioactive material taken up by the cells
cytes and melanoma cells were able to accumulate vitamin C intra
incubated in the presence of dehydroascorbic acid eluted in the position
cellularly through the ascorbic acid pathway, their ability to do so was
corresponding to ascorbic acid. The remaining radioactive material eluted
limited compared to that of melanoma cells transporting dehy
in several small peaks that did not coincide with the elution of ascorbic
droascorbic acid (Fig. 2E). At 3 p.M extracellular ascorbic acid,
acid or dehydroascorbic acid. The ascorbic acid peak disappeared, and
melanocytes and melanoma cells accumulated 55- and 20-fold more
the radioactive material ebuted in the position corresponding to dehy
intracellular vitamin C than the amount expected at equilibrium,
droascorbic acid when the cellular extracts were treated with ascorbic
respectively, but the ratio of intracellular versus extracellular ascorbic
acid oxidase before the HPLC separation, confirming the identity of the
acid decreased rapidly as the extracellular ascorbic acid concentration
accumulated vitamin C as ascorbic acid.
increased and was less than 1 at external ascorbic acid concentrations
Accumulation studies using 20-mm uptake assays revealed that
greater than 300 p.M (Fig. 2E).
melanoma cells possessed a remarkable capacity to accumulate high
Lineweaver-Burk analysis of the 20-mm dose-response studies
intracellular concentrations of ascorbic acid (Fig. 2A). Increasing the
revealed one kinetic component for each cell type when studying the
external concentration of dehydroascorbic acid from 1 p.M to 10 mr@i accumulation of ascorbic acid in melanoma cells and melanocytes
bed to a parallel increase in intracellular ascorbic acid, from 90 p.Mto
through the dehydroascorbic acid pathway (Table 2). The apparent
85 mt@i(Fig. 2.4). At all extracellular dehydroascorbic acid concentra
Km5 were very similar:
1 mM for melanoma
cells and 0.45 mi@i for
tions, the ratio of intracellular ascorbic acid to extraceblular dehy
melanocytes. In contrast, melanoma cells and melanocytes showed a
droascorbic acid was always greater than unity (Fig. 2B). A ratio of
difference of two orders of magnitude in their respective Vm@ of
100 or greater was observed at external dehydroascorbic acid concen
accumulation of ascorbic acid through the dehydroascorbic acid path
trations from 1 to 300 p.M. Although the ratio decreased as extracel
way, with values of 5.6 and 0.07 rnt@i/min,respectively (Table 2). The
lular dehydroascorbic acid increased to 1 mr@ior more, it still main
data indicate that the enormous capacity of melanoma cells to accu
saturated at concentrations of less than 100 p.M(Fig. 110. The data did
not suggest multiple kinetic components of ascorbate transport, and
we estimated the apparent Km and Vm@values for ascorbate transport
to be 20 and S p.M/mm for melanoma cells and 20 and 30 p.t@i/minfor
melanocytes, respectively. These observations clearly show the rela
tive inefficiency of ascorbate transport when compared to dehy
droascorbic acid transport by melanoma cells.
2532
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1997 American Association for Cancer Research.
VITAMIN C UPTAKE BY MELANOMA CELLS AND MELANOCYTES
100
A@
C
E
@
@B
.2120
I
0
75
5
Melanoma
Fig. 2. Accumulation of vitamin C by melanoma cells and melano
cytes by the dehydroascorbic acid (DIM) pathway and the ascorbic acid
(AA) pathway during 20-mm uptake times. A, accumulation of intracel
lular vitamin C by melanoma cells (I) and melanocytes (0) when
provided with increasing concentrations of dehydroascorbic acid in the
extracellular media. B, ratio of the intracellular concentration of vitamin
concentration
:@
@.
I
).0010.01 0.1 1
DHA (mM)
of vitamin C in melanoma cells to that of
Melanocytes
1(
0.0010.01 0.1
1
0
0.1 01 0.01 0.1
10
DHA(mM)
DHA(mM)
.@. 60
droascorbic acid in the extracellular media. D, accumulation of intracel
C
lular vitamin C by melanoma cells (I) and melanocytes (0) when
@
provided with increasing concentrations of ascorbic acid in the extracel
lular media. E, ratio of the intracellular concentration of vitamin C to the
@
@
extracellularconcentrationof ascorbicacid in melanomacells(I) and
@
melanocytes (0) using the data presented in A. F, ratio of intracellular
a
.5
E
LA
E@
‘ll
Q@4@F
I
>@@‘u
3 I
Melanocytes
@3
of vitamin C in melanoma cells to that of melanocytes
extracellular media.
:@‘I
OC
.DB
aa
mulate ascorbic acid through the dehydroascorbic acid pathway com
pared to melanocytes is the result of a dramatic increase in the Vm@
without
major
changes
in the
respective
Km5.
A similar analysis revealed the presence of a unique kinetic component
in each cell type associated with the accumulation of vitamin C through
the ascorbic acid pathway. The apparent Km and V,,@ for accumulation
were 13 and 16 p.M/mm for melanoma cells and 47 and 72 p.M/mm for
melanocytes (Table 2). Thus, the results of the kinetic data were consist
ent with the observation that melanocytes accumulated more vitamin C
than the melanoma cells through the ascorbic acid pathway.
Transport
of Dehydroascorbic
Acid into Melanocytes
and Mel
anoma Cells Occurs through Facilitative Glucose Transporters.
We showed previously that facilitative glucose transporters are efficient
transporters of dehydroascorbic acid (26). We performed inhibition stud
ies with cytochalasin B and competition studies with deoxyglucose to
establish whether melanoma cells and melanocytes transported dehy
droascorbic acid through facilitative glucose transporters. Cytochalasin B
and deoxyglucose inhibited, in a dose-dependent manner, the uptake of
dehydroascorbic acid by melanoma cells and melanocytes, whereas their
inactive analogues, cytochalasin E and L-glucose, had no effect (Fig. 3,
A—D).Although cytochalasin B was able to completely abolish dehy
droascorbic acid uptake in melanoma cells (Fig. 3A), it inhibited only
approximately 65% of uptake in melanocytes (Fig. 3B). Similarly, de
oxyglucose totally blocked dehydroascorbic acid uptake in melanoma
cells (Fig. 3C) but inhibited only 50% ofuptake in melanocytes (Fig. 3D).
The IC50 for the inhibition of dehydroascorbic acid uptake for both
melanoma cells and melanocytes was approximately 10 russfor cytocha
lasin B and 3 mM for deoxyglucose. Because cytochalasin B and deoxy
Table2 Kmand V,,.@determinationsof vitaminC accumulationby melanomacell
AA
KmMelanocytes'@
Celltype
Melanoma
@sMa
cells'
All accumulation
Km
V,,,@
0.07 msi/min
0.45 mM
72 p@i/min
47
5.60 msi/min
1.00 ms@
16 @.u.t/min
13
data are calculated
@M
from 20-ruin uptake times and are derived from
the,plots in Fig. 2.
Melanocyte strain 2486.
C SK-MEL-l31,
clone
AA(mM)
glucose did not completely abolish uptake of dehydroascorbic acid in
melanocytes, we used fructose and methyl a-r-glucopyranoside to test
for the presence of additional transporters not sensitive to the action of
cytochalasin B and deoxyglucose. Neither of these compounds were able
to inhibit dehydroascorbic acid uptake in melanocytes, indicating that
dehydroascorbic
acid uptake does not involve a fructose transporter
such
as GLUT5 (49) or a sodium-glucose cotransporter (59). Replacement of
NaCl in the media with choline chloride, LiCb, or sucrose did not affect
the uptake of dehydroascorbic acid by melanoma cells or mebanocytes
(Fig. 3, E and F) nor did it enhance
the partial inhibitory
effect of
cytochalasin B or deoxyglucose on dehydroascorbic acid uptake by
mebanocytes (data not shown). Similarly, ouabain did not affect the
uptake of dehydroascorbic acid by melanoma cells or melanocytes (data
not shown). Thus, transport of dehydroascorbic acid by melanoma cells
and melanocytes is a process that does not require sodium and is not
dependent on a sodium-potassium ATPase (60).
The data are compatible with the participation of glucose transport
ers in dehydroascorbic acid transport in melanoma cells and melano
cytes. To further analyze this point, we studied the ability of mela
noma cells to transport deoxyglucose, a substrate of facilitative hexose
transporters. These experiments demonstrated that melanoma cells
have a remarkable capacity to transport deoxyglucose, transporting at
least 50-fold more deoxyglucose than melanocytes (Fig. 3G). We,
therefore, examined the possibility that melanoma cells showed in
creased glucose transporter expression. Immunodetection experiments
using antibodies against four different glucose transporters (GLUT 1,
GLUT2,
GLUT3,
and
pressed
the transporters
GLUT3
(Fig.
4,
A—D).
GLUT4)
revealed
GLUT1,
On
the
other
that
GLUT2,
hand,
melanoma
cells
and GLUT4
melanocytes
cx
but not
expressed
smaller amounts of GLUT! and GLUT2 than melanoma cells and
failed to express the transporters GLUT3 and GLUT4 (Fig. 4, E—H).
Taken together, the data show that the increased dehydroascorbic acid
transport observed in melanoma cells compared to melanocytes relates to
increased expression of facilitative glucose transporters. In addition, the
expression of several glucose transporter isoforms in melanoma cells may
explain the presence of high- and low-affinity dehydroascorbic acid
transporters in these cells. GLUTI and GLUT4 are high-affinity trans
porters (49), and GLUT2 is a low-affinity transporter (49).
lines and melanocytes°DHA
normal
@
I
11I
01
I I 1I
0.001 0.01 0.1
@o010@01
01 1
AA(rnM)
AA (mM)
I
@a 2r
@.
@
@
I
I
/
°@@ma@oma
15
@
process,
III
45
whenprovidedwith increasingconcentrationsof ascorbicacid in the
of the
I
.2
melanocyteswhen providedwith increasingconcentrationsof dehy
concentration
.2
.@
Melan@;@a
Melanocytes
C totheextracellularconcentrationofdehydroascorbic
acidinmelanoma
cells(I) andmelanocytes(0) usingthedatapresentedinA. C, ratioof
intracellular
=@60
@250
>
3.44.
2533
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1997 American Association for Cancer Research.
VITAMIN C UPTAKE BY MELANOMA CELLS AND MELANOCYTES
Fig. 3. Transport of dehydroascorbic acid through facili
tative glucose transporters. A and B. inhibition of uptake of
100 @ssi
dehydroascorbic acid during a 5-mm uptake assay by
increasing concentrations of cytochalasin B (Cyt B. I) but
not cytochalasin E (Cyt E. 0) in melanoma cells (A) and
melanocytes (B). C and D, inhibition of uptake of 100 @u.i
dehydroascorbic
acid during a 5-mm uptake assay by increas
ing concentrations of deoxyglucose (DOG. I) but not i
@
glucose (L-Gk, 0) in melanoma cells (C) and melanocytes
(D). E, effect of replacing NaCI (0) with choline chloride
(I), LiCI (s), or sucrose(A) on the transportof 50
dehydmascorbic
@
acid by melanoma
cells. F, effect of replac
ing NaCI (0) with choline chloride (I) on the accumulation
of 50 p@,idehydroascorbic acid by melanocytes. G, transport
of 50 psi deoxyglucoseby melanomacells (I) and melano
cytes (0).
600
Cl)
G
Cl)
a
a
0
0
CD
0
CD
0
0
Melanoma
400
E
E
C
0.
a
a
a 200
.@
Il)
0.
0.
:@
Melanocytes
0
I
0
0
0
0
0 30 60 90120150
Time(sec)
Time (mm)
Evidence for Transport
of Ascorbic Acid into Melanoma
Cells
and Melanocytes through a Sodium-Ascorbate Cotransporter.
We next examined the functional characteristics of the transporter of
ascorbic acid present in melanoma cells and melanocytes. Cytochalasin B
did not inhibit the uptake of ascorbic acid by melanocytes or melanoma
cells in contrast to their effect on dehydroascorbic acid uptake (Fig. 5, A
and B). Similarly, deoxyglucose, fructose, and methyb-a-o-glucopyrano
side had no effect on the transport of ascorbic acid by melanocytes (data
not shown), indicating that the transport of ascorbic acid by both meba
nocytes and melanoma cells occurs by a route other than facilitative
glucose transporters. We, therefore, explored the possibility that the
transport
of ascorbic
acid
by melanocytes
could
be mediated
by a sodi
um-ascorbate cotransporter. To test the sodium dependence of ascorbic
acid transport and accumulation, we replaced the NaCl in the incubation
buffer with choline chloride, LiCl, or sucrose. Transport of ascorbic acid
by melanocytes was decreased by at least 80% in the absence of sodium,
an observation indicating the participation of a sodium cotransporter in
the transport of ascorbic acid by these cells (Fig. SC). The accumulation
of vitamin C via the ascorbic acid pathway was also dependent on the
presence of sodium ions during the uptake assay (Fig. SD). Identical
results, almost total inhibition of ascorbic acid transport and accumula
tion, were obtained when melanoma cells were incubated with ascorbate
in buffer backing sodium ions (data not shown). Dose-dependence exper
iments revealed that ascorbic acid uptake in melanocytes increased as a
hyperbolic function of the extracellular sodium concentration (Fig. SE),
with maximum uptake observed at approximately 60 my@NaCI. Increas
ing the extracellubar sodium concentration above 60 mist up to 135 m@i
failed to cause a further increase in the uptake of ascorbic
acid. We
determined Hill coefficients of 1.0 and 1.1 for the effect of different
sodium concentrations on the transport of 10 and 50 p.isiascorbic acid,
respectively (Fig. SF'), which is consistent with the concept that one
sodium ion is cotransported into the melanocytes for each ascorbate
molecule transported.
Because sodium cotransport processes require that the sodium
transported into the cell eventually be transported back out of the cell,
we tested the dependence of ascorbic acid transport on the activity of
the sodium-potassium ATPase using ouabain, an inhibitor of the
enzyme (60). When melanocytes and melanoma cells were incubated
in the presence of ouabain, ascorbic acid uptake was decreased by
more than 60% (Fig. 5, G and H). The specificity of the effect of
ouabain was evidenced by the observation that it did not inhibit
dehydroascorbic acid uptake in the same cells (data not shown).
DISCUSSION
Our data define two mechanisms of vitamin C transport and accumu
bation in melanocytes and melanoma cells. A model summarizing both
mechanisms is presented schematically in Fig. 6. One mechanism in
volves the efficient transport of the oxidized form of vitamin C, dehy
droascorbic acid, through glucose transporters by an energy-independent
facilitative diffusion process. Because transport of dehydroascorbic acid
through the glucose transporter can occur in either direction (27, 28), the
cell must trap the vitamin C molecule by converting dehydroascorbic acid
to its reduced form, ascorbate. This mechanism has been shown in our
laboratory to function in normal neutrophils (26) and HL-60 cells (27);
these cells do not take up reduced vitamin C.
Cancer cells transport more glucose than nontumor cells through
increased expression ofglucose transporters (50, 51), although the mech
anism by which cancer cells increase expression of these transporters is
not clearly understood. The ability of melanoma cells to transport large
amounts
of dehydroascorbic
acid relative
to melanocytes
is due to the
marked increase in expression of GLUTI and GLUT2 and the appear
ance of GLUT4. The presence of more than one facilitative glucose
transporter in melanoma cells also explains why melanoma cells display
a high-affinity (Km, 0.38 fliM) and a low-affinity component (Km, 3.84
mM) for the transport
of dehydroascorbic
acid.
Melanoma cells showed a remarkable ability to accumulate intracel
lular concentrations of vitamin C via the dehydroascorbic acid transport
pathway. These cells were capable of reducing intracellular concentra
tions of nearly 100 mrviof dehydroascorbic acid to ascorbate in 20 mm
and accumulated intracellular concentrations of vitamin C greater than
100 times higher than the extracellular dehydroascorbic acid concentra
tions. Melanocytes do not have this reducing capacity.
The mechanism by which melanoma cells reduce dehydroascorbic
2534
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1997 American Association for Cancer Research.
@
,@
VITAMIN C UPTAKE BY MELANOMA
CELLS AND MELANOCYTES
B@
@C
r
v@
cli,
E
0
@
@L
;@@b
@
‘@
-,
Fig. 4. Immunodetection of facilitative glucose
transporters
in melanoma
h
cells and melanocytes.
Melanomacells(A-D)andmelanocytes(E-H)were
processed for immunoperoxidase staining with anti
H
G
GLUTI (A and E), anti-GLUT2(B and F), anti
GLUT3 (C and G), and anti-GLUT4
A
(D and H)
antibodies.
.!
4r,
@,
4i,
‘Ø@
c
GLUT1
GLUT2
acid is not known. Potential mechanisms invoking glutathione and van
ous reducing enzymes have been proposed as candidates in the intercon
version of oxidized and reduced vitamin C (29-35), and melanomas are
known to have high levels of glutathione synthetase activity (61). One
conclusion we can draw from our data is that the enzyme(s) involved in
the intracellular conversion ofdehydroascorbic acid to ascorbate must be
increased in number and/or specific activity in melanoma cells because
they show a great increase in their ability to accumulate vitamin C when
provided with dehydroascorbic acid relative to melanocytes.
Melanocytes transported dehydroascorbic acid very slowly and also
had a limited capacity to reduce it to ascorbate. These cells did not
accumulate substantial intracellular concentrations of vitamin C during
GLUT3
20-rain accumulation assays. These observations indicate that unlike the
melanoma cells, the dehydroascorbic acid pathway is not a significant
pathway of vitamin C uptake by normal melanocytes.
The second mechanism of vitamin C uptake by melanocytes and
melanoma cells involves the sodium-dependent cotransport of reduced
ascorbate. In addition to its dependency on sodium, the uptake of ascor
bate occurs by a mechanism distinct from the uptake of dehydroascorbic
acid because uptake of ascorbate is inhibited by ouabain (whereas the
uptake of dehydroascorbic acid is not affected) and is not inhibited by
cytochalasin B or deoxyglucose. The inhibition by ouabain indicates a
robe for a functional sodium-potassium ATPase in the uptake of ascorbate
through the sodium-ascorbate cotransporter. In addition, the two mech
Melanocytes
Melanoma
48
125
CytB
@
I
Cyt E
@. 75
@
@.LM
ascorbic
acid
during
a 5-mm
uptake
assay
chloride
(I),
LiC1 (I),
C
H
-
a
0
A
400
D
0
75
CD
32
CytE
-
0
0
E
50
a
Chohne
16
-
CholIne
25
@25
in
C+
(1)
.@
IA,
0.01 0.1
1
, rBCI I I
10 0 0.01 0.1
Cytochalasmn (jiM)
melanocytes (B). C and D. effect of replacing NaCI
with choline
II
Cyt B
@5o
melanomacells and melanocytes.Increasingcon
centrationsof cytochalasinB (Cyt B, I) and cy
tochalasinE (CytE, 0) in melanomacells(A)and
(0)
•
0
@100
Fig. 5. Transport of ascorbic acid through a
sodium cotransporter system. A and B. effect of
inhibitors of glucose uptake on the uptake of 100
GLUT4
1
10
I
0
3@ @09@1@01@C 01
Time(see)
or sucrose
(A)on thetransport(C)andaccumulation(D)of 50
@LM
ascorbic acid by melanoma cells. E, effect of
sodium ion concentration on the accumulation of 10
g.LM(0)
and 50
@isi(I)
ascorbic
(13
acid by melano
cytes. F, Hill analysis ofthe data shown in E. G and
H, effect of 10 psi ouabain (I) on the accumulation
CD
0
of ascorbatein melanocytes(G) and melanoma
0
cells (If). Untreated cells (0) were used as controls.
0.
I Sucrose
10
20
30
Time (mm)
E
a
a
0.
2535
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1997 American Association for Cancer Research.
VITAMIN C UPTAKE BY MELANOMA
Glucose-Dehydroascorbic
r
CELLS AND MELANOCYTES
of ascorbate inside the cell: one pool from the reduction of dehy
droascorbic acid, and the other pool from the sodium-dependent
cotransport of ascorbate.
From our data, we are able to extrapolate what the relative
contributions of the dehydroascorbic acid and ascorbate pathway to
vitamin C uptake in vivo might be. In the melanocyte, the ascorbate
pathway is critical to vitamin C uptake. The Km of the process is
in the micromolar range, which is within the expected physiobog
ical concentrations of vitamin C in the blood (4, 24). The dehy
droascorbic acid pathway would not be expected to be of signifi
cance to melanocytes because they have a highly limited capacity
to reduce dehydroascorbic acid to ascorbate and because the Km of
transport by this pathway is milbimolar. On the other hand, mela
noma cells potentially use both pathways in vivo. They can use the
ascorbate pathway because the Km is in the micromolar range. The
dehydroascorbic acid pathway may be very active, although its Km
is in the millimobar range, even at low dehydroascorbic
acid
concentrations by virtue of the high Vmax of accumulation (5.6
mM/mm). Melanoma cells require only 5 p.M dehydroascorbic acid
to accumulate the same intracellular concentrations of vitamin C as
Acid
Transporter (GLUT)
Overexpressed
‘I inmelanoma
cells
I Inhibited
byCytochalasin
B
LComieted
byDeoxyglucose
Ascorbate
Ascorbate
Na@
45—50
p.M
ascorbic
acid.Themajor
unknown
inthissituation
isthe
+ .—.-,
a
I
Ascorbate
[email protected]
Cotra@sporter
r
Transport is
Llahibi@@ by Ouabain
Fig. 6. Model for vitamin C transport by melanocytes and melanoma cells. Vitamin C
can be transported
into melanocytes
and melanoma
cells by two different
mechanisms.
In
the dehydroascorbic acid pathway, dehydroascorbic acid is transported through facilitative
glucose transporters. This pathway is inhibited by cytochalasin B and inhibited compet
itively by the glucose analogue, deoxyglucose, but not by ouabain. Once inside the cell.
dehydroascorbic acid is reduced to ascorbate. In the ascorbic acid pathway, ascorbic acid
is transported through a sodium-ascorbate cotransporter. Transport along the ascorbic acid
pathway is inhibited by ouabain but not by deoxyglucose or cytochalasin B. The molecular
identity of the sodium-dependent
cotransporter is not known, and no direct inhibitors have
been identified. It is not clear if ascorbic acid that is transported into the cell via the
sodium-ascorbate
cotransporter
is kept in a separate pool from ascorbic acid that results
pericellular concentration
of dehydroascorbic
acid. Melanoma
cells have strong prooxidant activities (64) and may be able to
generate substantial concentrations of dehydroascorbic
acid out
side the cell.
A central question raised by these observations is why meba
noma cells develop the ability to transport and trap vitamin C. One
possibility is that vitamin C plays a role in the differentiation of
mebanocytic cells, the main function of which is synthesis of the
pigment melanin. Intermediates
in melanin synthesis include
highly reactive molecules that create an oxidative environment
inside the cell. Thus the antioxidant properties of vitamin C could
be important to protect the cell as it synthesizes melanin (8—10,65,
from the reduction of dehydroascorbic acid. Melanoma cells, like other tumor cells,
66).Ourobservations
suggestthatvitaminC transportandtrapping
express increased numbers of facilitative glucose-dehydroascorbic
are not simply coregulated with the ability of melanocytic cells to
synthesize pigment. The transport of vitamin C by the dehy
droascorbic acid pathway did not correlate with pigmentation in
melanoma cells (53); in particular, two melanoma cell lines, clone
1-S of SK-MEL-l3 1 and SK-MEL-22A, are amebanotic melanomas
that do not synthesize pigment but continue to transport and trap
barge amounts of vitamin C.
We propose that increased transport and trapping of vitamin C is
regulated during the process of malignant transformation. The marked
increase in transport of dehydroascorbic acid by melanoma cells
compared to normal melanocytes supports this notion. Regulation of
facilitative glucose transporters appears to be the underlying mecha
nism that leads to increased uptake of vitamin C in its oxidized form.
This mechanism requires an additional step of reduction of dehy
droascorbic acid to ascorbic acid before cells can accumulate vitamin
C. The close association of increased vitamin C uptake with the
malignant phenotype suggests that intracellular accumulation of vita
mm C could be related to other characteristics of transformed mela
nocytes, including high rates of cellular metabolism, the ability to
metastasize widely, or resistance to cytotoxic therapies such as radi
ation and chemotherapy (67).
acid transporters. This
allows for increased transport of dehydroascorbic acid by melanoma cells relative to
melanocytes. Melanoma cells also have a great capacity to reduce large quantities of
intracellular dehydroascorbic acid to ascorbic acid. This results in increased accumulation
of vitamin C by melanoma cells.
anisms show vastly different kinetics, with the Km of ascorbate transport
and accumulation in the micromolar range and the Km for dehydroascor
bic acid in the low millimobar range. Thus far, melanocytes and mela
noma cells are the only family ofcells that we have identified that can use
two different mechanisms of vitamin C transport. Unlike the dehy
droascorbic acid pathway, the ascorbate pathway functions at approxi
mately the same level in both cell types; it is approximately 3-fold more
active in melanocytes than melanoma cells.
Transport of ascorbate through a sodium-dependent cotransporter
has been described for osteoblasts (36), adrenomedublary chromaffin
cells (43), and astrocytes (42). It is interesting that melanocytes,
adrenomedullary chromaffin cells, and astrocytes are all neural crest
derivatives. The relationship between melanocytes and adrenomedul
lary chromaffin cells becomes even more intriguing when one con
siders that some of the reactions of catecholamine synthesis in ad
renomedullary chromaffin cells are similar to those for melanin
synthesis in melanocytes (62, 63).
One common feature of both the dehydroascorbic acid pathway and
the ascorbate pathway of vitamin C uptake in melanocytes and mel
anoma cells is that both pathways end up with reduced ascorbate
inside the cell. We do not know, however, whether the intracellular
ascorbate generated from each pathway enters into the same compart
ment. In other words, it is not known whether there are separate pools
REFERENCES
I. Englard, S., and Seifter, S. The biochemical functions of ascorbic acid. Annu. Rev.
Nutr., 6: 365—406,1986.
2. Padh, H. Cellular functions of ascorbic acid. Biochem. Cell Biol., 68: 1166—1
173, 1990.
3. Sauberlich, H. E. Pharmacology of vitamin C. Annu. Rev. Nutr., 14: 371—391,1994.
4. Omaye, S. T., Schaus, E. E., Kutnink, M. A., and Hawkes, W. C. Measurement of
vitamin C in blood components by high-performance
liquid chromatography.
1mph
cation in assessing vitamin C status. Ann. NY Acad. Sci., 498: 389-401, 1987.
2536
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1997 American Association for Cancer Research.
VITAMIN C UPTAKE BY MELANOMA
5. Rebouche, C. J. Ascorbic acid and carnitine biosynthesis. Am. J. Clin. Nutr., 54:
ll47S—ll52S,
1991.
6. Eipper, B. A., and Mains, R. E. The role of ascorbate in the biosynthesis of
neuroendocrine peptides. Am. J. Clin. Nutr., 54: 1l53S—l156S, 1991.
7. Dawson, E. B., Harris, W. A., and Powell, L. C. Relationship between ascorbic acid
and male fertility. World Rev. Nutr. Diet, 62: 1—26,1990.
8. Kojima, S., Yamaguchi, H., Morita, K., and Ueno, Y. Inhibitory effect of sodium
5,6-benzyhidene
ascorbate(SBA)ontheelevationof melaninbiosynthesisinducedby
ultraviolet-A (UV-A) light in cultured B-l6 melanoma cells. Biol. Pharm. Bull., 18:
1076—1080, 1995.
9. Kameyama, K., Sakai, C., Kondoh, S., Yonemoto, K., Nishiyama, S., Tagawa, M.,
Murata, T., Ohnuma, T., Quigley, J., Dorsky, A., Bucks, D., and Blanock, K.
Inhibitory effect of magnesium i-ascorbyl-2-phosphate (VC-PMG) on melanogenesis
in vitro and in vivo. J. Am. Acad. Dermatol., 34: 29—33,1996.
10. Laskin, J. D., and Piccinini, L. A. Tyrosinase isozyme heterogeneity in differentiating
B16/C3 melanoma. I. Biol. Chem., 261: 16226—16235, 1986.
I 1. Varga, J. M., and Airoldi, L. Inhibition of transplantable melanoma tumor develop
(glutaredoxin) and protein disulfide isomerase have dehydroascorbate reductase ac
tivity.
J.Biol.Chem.,265: 15361—15364,
1990.
35. Winkler, B. S. Unequivocal evidence in support of the nonenzymatic redox coupling
between glutathione/glutathione
disulfide and ascorbic acid/dehydroascorbic
acid.
Biochim. Biophys. Acta, 1117: 287—290,1992.
36. Wilson, J. X., and Dixon, S. J. High-affinity sodium-dependent uptake of ascorbic
acid by rat osteoblasts. J. Membr. Biol., 11 1: 83—91
, I989.
37. Dyer,
38.
39.
40.
mentin miceby prophylacticadministrationof Ca-ascorbate.LifeSci.,32: 1559— 41.
1564, 1983.
12. Parsons, P. 0., and Morrison, L. E. DNA damage and selective toxicity of dopa and
ascorbate:copper in human melanoma cells. Cancer Res., 42: 3783—3788,1982.
13. Pierson, H. F., and Meadows, G. G. Sodium ascorbate enhancement
42.
of carbidopa
levodopa methyl ester antitumor activity against pigmented Bl6 melanoma. Cancer
43.
Res.,43: 2047—2051,
1983.
14. Beam, S., Froussard, P., Guichard, M., Jasmin, C., Augery, Y., Sinoussi-Barre, F., and
Wray, W. Vitamin C preferential toxicity for malignant melanoma cells. Nature
(Lond.), 284: 629—631, 1980.
15. Gardiner, N. S., and Duncan, J. R. Enhanced prostaglandin synthesis as a mechanism
for inhibition of melanoma cell growth by ascorbic acid. Prostaglandins Leukotrienes
Essent. Fatty Acids, 34: 119—126,1988.
44.
45.
16. Gardiner,N.S.,andDuncan,J. R. Inhibitionof murinemelanomagrowthbysodium
ascorbate. J. Nutr., 119: 586—590, 1989.
17. Meadows, G. 0., Pierson, H. F., and Abdallah, R. M. Ascorbate in the treatment of
experimentaltransplantedmelanoma.Am.J. Clin.Nuts.,54: 1284S—l29lS,
1991.
18. Dc Pauw-Gillet, M. C., Siwek, B., Pozzi, 0., Sabbioni, E., and Bassleer, R. J. Effects
of FeSO4on 816 melanomacellsdifferentiationandproliferation.AnticancerRes.,
10: 1029—1033, 1990.
19. Dc Pauw-Gillet, M. C., Siwek, B., Pozzi, 0., Sabbioni, E., and Bassleer, R. J. Control
46.
47.
48.
49.
of B16 melanomacellsdifferentiationand proliferationby CuSO4and vitaminC.
Anticancer Res., 10: 391—395,1990.
20. Osmak, M., Eckert-Maksic, M., Pavelic, K., Maksic, Z. B., Spaventi, R., Beketic, L.,
Kovacek, I., and Suskovic, B. 6-Deoxy-6-bromo-ascorbic
acid inhibits growth of
mouse melanoma cells. Res. Exp. Med., 190: 443-449, 1990.
21. Stoll, K. E., and Duncan, J. R. The effect of ascorbic acid on arachidomc acid and
prostaglandin E2 metabolism in Bl6 murine melanoma cells. Prostaghandins Leukot
rienes Essent. Fauy Acids, 47: 307—312,1992.
22. Stoll, K. E., and Duncan, J. R. The effect of ascorbate on essential fatty acid
composition in B16 melanoma cells. Prostaglandins Leukotrienes Essent. Fatty Acids,
49: 771—776,1993.
23. Stoll, K. E., Ottino, P., and Duncan, J. R. Interrelationship of ascorbate, arachidonic
acid and prostaglandin E2 in B16 melanoma cells. Prostaglandins Leukotrienes
Essent.FattyAcids,50: 123-131,1994.
24. Rose, R. C. Transport of ascorbic acid and other water-soluble vitamins. Biochim.
Biophys. Acta, 947: 335—366,1988.
25. Nishikimi. M., Fukuyama, R., Minoshima, S., Shimizu, N., and Yagi, K. Cloning and
chromosomal mapping of the human nonfunctional gene for L-gulono-gamma-lactone
oxidase, the enzyme for L-ascothic acid biosynthesis missing in man. J. Biol. Chem.,
269:13685—13688,
1994.
26. Vera, J. C., Rivas, C. I., Fischbarg, J., and Golde, D. W. Mammalian facilitative
hexose transporters mediate the transport of dehydroascorbic acid. Nature (Lond.),
364: 79—82,1993.
27. Vera, J. C., Rivas, C. I., Thang, R. H., Farber, C. M., and Golde, D. W. Human HL-6O
myeloidleukemiacells transportdehydroascorbic
acid via the glucosetransporters
and accumulate reduced ascorbic acid. Blood, 84: 1628—1634, 1994.
28. Vera, J. C., Rivas, C. I., Velasquez, F. V., Zhang, R. H., Concha, I. I., and Golde,
D. W. Resolution of the facilitated transport of dehydroascorbic acid from its
intracellular accumulation as ascorbic acid. J. Biol. Chem., 270: 23706—23712, 1995.
29. Del Bello, B., Maellaro, E., Sugherini, L., Santucci, A., Comporti, M., and Casini,
A. F. Purification of NADPH-dependent dehydroascorbate reductase from rat liver
and its identification with 3 a-hydroxysteroid dehydrogenase. Biochem. J., 304:
385—390,1994.
30. Farber, C. M., Liebes, L. F., Kangarns, D. N.. and Silber, R. Human B lymphocytes
show greater susceptibility to H202 toxicity than T lymphocytes. J. Immunol., 132:
2543—2546,
1984.
31. Maellaro, E., Del Bello, B., Sugherini, L., Santucci, A., Comporti, M., and Casini,
A. F. Purification and characterization of glutathione-dependent dehydroascorbate
50.
51 .
52.
53.
54.
55.
D. L., Kanai,
Y., Hediger,
M. A.. Rubin,
S. A., and Said, H. M. Expression
of
a rabbit renal ascorbic acid transporter in Xenopus laevis oocytes. Am. J. Physiol.,
267: C30l—C306, 1994.
Bowers-Komro, D. M., and McCormick, D. B. Characterization of ascorbic acid
uptake by isolated rat kidney cells. J. Nutr., 121: 57—64,1991.
Rose, R. C. Ascorbic acid transport in mammalian kidney. Am. J. Physiol.. 250:
F627—F632, 1986.
Sihiprandi, L., Vanni, P., Kessler. M., and Semenza, G. Na@-dependent, electroneutral
L-ascorbate transport across brush border membrane vesicles from guinea pig small
intestine. Biochim. Biophys. Acta, 552: 129—142,1979.
Helbig, H., Korbmacher, C., Wohlfarth, J., Berweck, S., Kuhner, D., and Wiederholt,
M. Electrogenic Na@-ascorbate cotransport in cultured bovine pigmented ciliary
epithehial cells. Am. J. Physiol., 256: C44—C49. 1989.
Wilson, J. X. Ascorbic acid uptake by a high-affinity sodium-dependent mechanism
in cultured rat astrocytes. J. Neurochem., 53: 1064—1071, 1989.
Diliberto, E. J., Jr., Heckman, G. D., and Daniels, A. J. Characterization of ascorbic
acid transport by adrenomedullary chromaffin cells. J. Biol. Chem., 258: 12886—
12894,1983.
Washko, P., Rotrosen, D., and Levine, M. Ascorbic acid transport and accumulation
in human neutrophils. J. Biol. Chem., 264: 18996—19002, 1989.
Welch, R. w., Bergsten, P., Butler, J. D., and Levine, M. Ascorbic acid accumulation
and transport in human fibroblasts. Biochem. J., 294: 505—510,1993.
Bode, A. M., Cunningham, L., and Rose, R. C. Spontaneous decay of oxidized
ascorbic acid (dehydro-i-ascorbic
acid) evaluated by high-pressure liquid chroma
tography. Clin. Chem., 36: 1807—1809,1990.
Winkler,
B. S. In vitro
oxidation
of ascorbic
acid
and
its prevention
by GSH.
Biochim. Biophys. Acta, 925: 258—264, 1987.
Warburg, 0. On the origin of cancer cells. Science (Washington DC), 123: 309—314,
1956.
Baldwin, S. A. Mammalian passive glucose transporters: members of an ubiquitous
family of active and passive transport proteins. Biochim. Biophys. Acta, 1154:
17—49,1993.
Younes, M., Lechago, L. V., Somoano. J. R., Mosharaf, M., and Lechago, J. Wide
expression of the human erythrocyte glucose transporter Glutl in human cancers.
CancerRes.,56:
1164—1167, 1996.
Brown, R. S., and Wahi, R. L. Overexpression of Glut-I glucose transporter in human
breast cancer. An immunohistochemical study. Cancer (Phila.), 72: 2979—2985,1993.
Liebes, L. F., Kuo, S., Krigel, R., Pelle. E., and Silber, R. Identification and
quantitation of ascorbic acid in extracts of human lymphocytes by high-performance
liquid chromatography. Anal. Biochem., 118: 53—57,1981.
Houghton. A. N., Real, F. X., Davis, L. J., Cordon-Cardo, C., and Old, L. J.
Phenotypic heterogeneity of melanoma: relation to the differentiation program of
melanoma cells. J. Exp. Med., 164: 812—829, 1987.
Real, F. X., Fhiegal, B., and Houghton, A. N. Surface antigens of human melanoma
cells cultured in serum-free medium: induction of expression of major histocompat
ibihity complex class II antigens. Cancer Res., 48: 686—693, 1988.
Eisinger. M., and Marko, 0. Selective proliferation of normal human melanocytes in
vitro in the presence of phorbol ester and cholera toxin. Proc. Nati. Acad. Sci. USA,
79:2018—2022.
1982.
56. Freifelder, D. Physical Biochemistry: Applications to Biochemistry and Molecular
Biology, Ed. 2, pp. 654—684. San Francisco: W. H. Freeman and Company, 1982.
57. Schulz, A. R. Enzyme Kinetics, Ed. 1. New York: Cambridge University Press. 1994.
58. Zamora-Leon, S. P., Golde, D. W., Concha, I. I., Rivas, C. I., Delgado-Lopez, F..
Baselga, J., Nulart, F., and Vera, J. C. Expression of the fructose transporter GLUTS
in human breast cancer. Proc. NatI. Acad. Sci. USA, 93: 1847—1852,1996.
59. Hediger, M. A., and Rhoads, D. B. Molecular physiology of sodium-glucose cotrans
porters. Physiol. Rev., 74: 993—1026, 1994.
60. Lingrel, J. B., and Kuntzweiler, T. Na@, K(+)-ATPase. J. Biol. Chem., 269: 19659—
19662,1994.
61. Benathan, M., Alvero-Jackson, H., Mooy, A. M., Scaletta, C., and Frenk. E. Rela
tionship between melanogenesis. glutathione levels and melphalan toxicity in human
melanoma cells. Melanoma Res., 2: 305-314, 1992.
62. McEwan, M., and Parsons, P. G. Inhibition of melanization in human melanoma cells
by a serotonin uptake inhibitor. J. Invest. Dermatol., 89: 82—86,1987.
63. Faray. B. A., Lawson, D. H., and Nixon, D. w. Melanoma detection by enzyme
radioimmunoassay of L-dopa. dopamine, and 3-O-methyldopamine in urine. Chin.
Chem., 27: 108—112,1981.
64. Szatrowski, T. P., and Nathan, C. F. Production of large amounts of hydrogen
peroxide by human tumor cells. Cancer Res.. 51: 794—798. 1991.
65. Ando, S., Ando. 0., Suemoto, Y., and Mishima, Y. Tyrosinase gene transcription and
reductase from rat liver. Biochem. J., 301: 471—476.1994.
32. Martensson, J., Jam, A., Stole, E., Frayer, W., Auld, P. A.. and Meister, A. Inhibition
of glutathionesynthesisin the newbornrat: a modelfor endogenouslyproduced
oxidativestress.Proc.NatI.Acad.Sci.USA,88: 9360—9364,
1991.
33. Stahl, R. L., Liebes, L. F., and Silber, R. A reappraisal ofleukocyte dehydroascorbate
reductase. Biochim. Biophys. Acts, 839: 119—121,1985.
34. Wells, W. W., Xu, D. P., Yang, Y. F., and Rocque, P. A. Mammalian thioltransferase
CELLS AND MELANOCYTES
its control
by melanogenic
inhibitors.
J. Invest.
Dermatol.,
100:
150S—155S,
1993.
66. Ros, J. R., Rodriguez-Lopez, J. N., and Garcia-Canovas, F. Effect of L-ascorbic acid
on the monophenolase activity of tyrosinase. Biochem. i., 295: 309—312, 1993.
67. Balch, C., Houghton, A. N., Sober, A. S., Soong, S-i., and Milton, G. W. Cutaneous
Melanoma, Ed. 2. Philadelphia: J. B. Lippincots, 1992.
2537
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1997 American Association for Cancer Research.
Increased Facilitated Transport of Dehydroascorbic Acid
without Changes in Sodium-dependent Ascorbate Transport in
Human Melanoma Cells
Charles Spielholz, David W. Golde, Alan N. Houghton, et al.
Cancer Res 1997;57:2529-2537.
Updated version
E-mail alerts
Reprints and
Subscriptions
Permissions
Access the most recent version of this article at:
http://cancerres.aacrjournals.org/content/57/12/2529
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 17, 2017. © 1997 American Association for Cancer Research.