Download Treatment with Drugs Able to Reduce Iodine Efflux Significantly

0021-972X/06/$15.00/0
Printed in U.S.A.
The Journal of Clinical Endocrinology & Metabolism 91(6):2389 –2395
Copyright © 2006 by The Endocrine Society
doi: 10.1210/jc.2005-2480
Treatment with Drugs Able to Reduce Iodine Efflux
Significantly Increases the Intracellular Retention Time
in Thyroid Cancer Cells Stably Transfected with
Sodium Iodide Symporter Complementary
Deoxyribonucleic Acid
Rossella Elisei, Agnese Vivaldi, Raffaele Ciampi, Pinuccia Faviana, Fulvio Basolo, Ferruccio Santini,
Claudio Traino, Furio Pacini, and Aldo Pinchera
Department of Endocrinology and Metabolism, Section of Endocrinology (R.E., A.V., R.C., F.S., A.P.); Department of
Oncology, Division of Pathology III (P.F., F.B.); and Service of Radiation Safety, South Chiara Hospital (C.T.), University
of Pisa, 56124 Pisa, Italy; Department of Internal Medicine, Endocrinology and Metabolism, and Biochemistry, University of
Siena (F.P.), 53100 Siena, Italy; and AMBISEN Center, High Technology Center for the Study of the Environmental
Damage of the Endocrine and Nervous Systems, University of Pisa (A.P.), 56100 Pisa, Italy
Context: One of the major limits of gene therapy with sodium iodide
symporter (NIS), which enables cells to be subjected to radioiodine
therapy, is that NIS-transfected cells rapidly release the intracellular
iodine.
Methods: We transfected human anaplastic (FRO) and medullary
(TT) thyroid cancer-derived cell lines that were unable to take up
iodine with human NIS cDNA. The possibility of increasing the iodine
retention time by treating the transfected clones with myricetin,
lithium, 17-(allylamino)-17-demethoxygeldanamycin (17-AAG), and
4,4⬘-diisothiocyanatostilbene-2,2⬘-disulfonic acid (DIDS) was
explored.
Results: We obtained 19 FRO and 16 TT clones stably transfected
with NIS. Twelve of 19 FRO and nine of 16 TT clones expressed the
full-length NIS mRNA; 11 of 12 FRO and four of nine TT clones were
O
NE OF THE main functions of the thyroid follicular cell
is its ability to take up iodine, which is fundamental
for the thyroid hormone synthesis (1). Sodium iodide symporter (NIS) is the plasma membrane protein involved in the
uptake and active transportation of iodine from the blood
into the follicular cell (2). Because this property may be also
retained after malignant transformation, the ability to take
up iodine represents the major tool for both diagnosis and
treatment of thyroid cancer (3).
Two types of well-differentiated thyroid cancer, papillary
and follicular [differentiated thyroid cancer (DTC)], arise
from follicular cell. Although the overall 5-yr survival rate for
DTC is high, about 20% of patients develop an aggressive
First Published Online March 14, 2006
Abbreviations: 17-AAG, 17-(Allylamino)-17-demethoxygeldanamycin; AIT, apical iodide transporter; DIDS, 4,4⬘-diisothiocyanatostilbene2,2⬘-disulfonic acid; DTC, differentiated thyroid cancer; FRTL-5, Fisher
rat thyroid L-5; h, human; hsp90, 90-kDa heat shock protein; NIS, sodium iodide symporter; PDS, pendrin.
JCEM is published monthly by The Endocrine Society (http://www.
endo-society.org), the foremost professional society serving the endocrine community.
able to take up radioiodine and correctly expressed NIS protein on the
plasma membrane. Kinetic analysis of iodide uptake in the two clones
(FRO-19 and TT-2) with the highest uptaking activity revealed that
the plateau was reached after 30 min by FRO-19 and after 60 min by
TT-2. The t1/2 of the iodide efflux was 9 min in FRO-19 and 20 min in
TT-2. The treatment of the two cell lines with four different drugs
revealed that DIDS and 17-AAG, but not myricetin and lithium,
significantly increased the intracellular iodide retention time in FRO19, but not in TT-2.
Conclusions: We showed that 17-AAG and DIDS prolong the retention time of 131I in NIS-transfected thyroid tumoral cells, thus reinforcing the hope of using this approach for future clinical application,
especially in patients with thyroid carcinoma who are no longer responsive to conventional therapy. (J Clin Endocrinol Metab 91:
2389 –2395, 2006)
disease, with local recurrence and distant metastases, as a
consequence of their progressive inability to take up radioiodine (4). During the dedifferentiation process, the tumor
progressively loses the expression of the differentiation
genes, starting with the loss of NIS and continuing with the
loss of thyroglobulin, thyroperoxidase, and TSH receptor (5).
Follicular cancer, poorly DTC, and undifferentiated or anaplastic thyroid cancer have a much worse prognosis. Intermediate clinical behavior is observed in medullary thyroid
cancer originating from parafollicular C cells. Conventional
chemotherapy and radiotherapy are poorly effective for the
treatment of these advanced and aggressive thyroid tumors
(6, 7).
NIS gene cloning (8, 9) has raised the possibility of transfecting NIS cDNA in cancer cells with the aim to induce
functional NIS expression and, as a consequence, the ability
to concentrate iodine. Radioiodine treatment of NIS-transfected cells could thus be reconsidered (10, 11). The major
problem of this therapeutic approach is that these cells are
deprived of the machinery responsible for iodine organification (12), and intracellular iodine is rapidly released
(13–15).
2389
2390
J Clin Endocrinol Metab, June 2006, 91(6):2389 –2395
A way to overcome this problem might be to increase
radioiodine retention time within the cell by blocking iodine
efflux. Several drugs have been previously shown to be able
to reduce iodine efflux in normal or malignant transformed
thyroid cells (16 –19). In this study we tested four of these
drugs [myricetin (20), lithium chloride (21), 17-(allylamino)17-demethoxygeldanamycin (17-AAG) (22), and 4,4⬘-diisothiocyanatostilbene-2,2⬘-disulfonic acid (DIDS) (23)] for their
ability to reduce the iodine efflux in both normal and tumoral
follicular cells. To this purpose, we transfected two human
thyroid cancer-derived cell lines, anaplastic (FRO) and medullary (TT), with human NIS cDNA. After selecting the
clones with the highest iodine uptake, the possibility of increasing the iodine retention time by treating the cells with
myricetin, lithium, 17-AAG, and DIDS was explored.
Materials and Methods
Human (h) NIS-FL cDNA stable transfection
FRO and TT cell lines were used for transfection experiments. Cell
culture conditions were described previously (12). For transfection, cells
were plated at the density of 1 ⫻ 105 in six-well plates and stably
transfected with full-length hNIS cDNA in pcDNA3 vector (gift from Dr.
S. Jhiang, Departments of Physiology and Internal Medicine, Ohio State
University, Columbus, OH) and with the empty vector as a control.
Transfection was performed with a lipid mixture, Lipofectamine (Invitrogen Life Technologies, Inc., Milan, Italy). After selection with 400
␮g/ml neomycin for 15 d, well-isolated clones were picked up by the
cylinder technique, isolated, then cultured for additional experiments.
Stably transfected cell culture was continued in 400 ␮g/ml neomycin
(Invitrogen Life Technologies, Inc.)-supplemented medium.
RNA extraction, RT-PCR, and NIS real-time
quantitative RT-PCR
Total RNA was isolated and reverse transcribed in cDNA as previously described (12). cDNA was then amplified by PCR using specific
primers for a ubiquitous gene (a portion of n-Ras) as a control and for
NIS gene. The primers and conditions used for PCR are reported in Table
1. PCR products were electrophoresed in a 2% agarose gel, then transferred to a nylon membrane. Each filter was hybridized with internal
probes (Table 1) specific for each amplified fragment and revealed using
a chemiluminescent method (ECL-CDP star detection, Amersham Biosciences, Milan, Italy).
To perform quantitative RT-PCR for NIS-mRNA, we used Real-Time
Sequence Detection System 7700 (Applied Biosystems, Foster City, CA)
(24). A serial dilution of an NIS-containing plasmid was used as the
standard curve. Two primer pairs were used, one specific for the amino
and one specific for the carboxyl terminus of NIS protein to ensure that
the full-length mRNA was present. The ubiquitous gene glyceraldehyde-3-phosphate dehydrogenase was also amplified for normalization
of the samples. Primers for glyceraldehyde-3-phosphate dehydrogenase
were obtained from Applied Biosystems; the NIS primers are reported
in Table 1. Samples omitting reverse transcriptase and cDNA were
Elisei et al. • Iodine Uptake in NIS-Transfected Thyroid Cells
included in each run as controls of potential laboratory and/or assay
contamination.
NIS immunocytochemistry
Cells were detached by trypsin, centrifuged, fixed by formalin, and
included by paraffin in blocks. Sections (5 ␮m) were cut, deparaffined,
and rehydrated. All slides were subjected to antigen retrieval using 10%
citrate buffer. Washes were performed with PBS for 5 min. Endogenous
peroxide activity was blocked with 5% H2O2 for 15 min.
Sections were incubated with the purified hNIS antibody (gift from
Brahms Diagnostica GmbH, Berlin, Germany) at a 1:1000 dilution at
room temperature for 1 h, then subjected to avidin and biotin block for
20 min each, to streptavidin peroxidase for 10 min, and to 3,3⬘-diaminobenzidine substrate chromogen for 5 min. The sections were counterstained with hematoxylin.
Iodine uptake and efflux
All transfected clones positive for NIS mRNA expression and negative controls were analyzed for their ability to take up iodine according
to a method currently used in the laboratory (12). Radioactivity in each
sample was counted in a ␥-counter as counts per minute. Cell number
was also determined, and iodine uptake was expressed as picomoles of
I⫺ per 106 cells. Clones able to take up iodine were then studied to
monitor both the kinetics and the specificity of iodine uptake by repeating the iodine uptake experiments at different times (0, 2, 5, 10, 15,
30, 45, and 60 min) and in the presence of 10 ␮m sodium perchlorate.
Iodine efflux was studied by incubating cells for 45 min at 37 C in
Hanks’ balanced salt solution incubation buffer plus 0.1 ␮Ci Na125I and
1 ␮m sodium iodide. After the incubation, radioactive buffer was replaced with nonradioactive buffer every 3 min for 30 min, and all
supernatants were collected. Cells were lysed after the last supernatant
removal. Radioactivity was counted in both supernatants and lysed
cells. Total radioactivity at the beginning of efflux (100%) was calculated
by adding the counts per minute found in each supernatant to those
found in the lysed cells.
Iodine efflux was measured both before and after treatment with
myricetin, lithium, 17-AAG, and DIDS in two of the positive FRO and
TT clones transfected with NIS and able to take up iodine. Myricetin (10
mm stock solution), lithium chloride (1 m stock solution), 17-AAG (3 mm
stock solution), and DIDS (100 mm stock solution; all from SigmaAldrich Corp., Milan, Italy) were dissolved in 20 ␮l dimethylsulfoxide/
800 ␮l absolute ethanol, in distilled water, in dimethylsulfoxide, and in
0.1 potassium bicarbonate, respectively. Cells were treated with myricetin 50 ␮m for 96 h, with 10 mm lithium chloride for 24 h, 17-AAG 3
␮m for 24 h, and with 1 mm DIDS for 1 h. The Fisher rat thyroid L-5
(FRTL-5) cell line was used as a control in all efflux experiments. FRTL-5
cells were cultured following standard conditions (25). All experiments
were performed in triplicate and repeated three times.
Dosimetry
Because the in vivo therapy is usually performed by administrating
I, we calculated the transfected cell-absorbed dose per unit of administered activity (Gy/MBq) for 131I based on the evidence that the
physic half-life of radioiodine has no influence on radioiodine kinetics
at cellular levels.
131
TABLE 1. PCR primers and probes used for the qualitative RT-PCR of N-ras and NIS mRNA (rows 1 and 2) and for quantitative realtime RT-PCR of both amino-terminal (row 3) and carboxyl-terminal (tow 4) portions of NIS mRNA
Gene
Primer forward
Primer reverse
Probe
N-ras
ATGACTGAGTACAAACTG
AGGAAGCCTTCGCCTGTCCT
CCTCATCCTGAACCAAGTGACC
NIS
CTTCTGAACTCGGTCCTCAC
TCCAGAATGTATAGCGGCTC
CCTCATCCTGAACCAAGTGACC
NIS-NH2
GCTCTTCATGCCCGTCTTCT
GGCTGAAGCGCATCTCCA
FAM-TGGGCCTCACCAGCACCTACGAGTACTAMRA
NIS-COOH CCATCCTGGATGACAACTTGG AAAAACAGACGATCCTCATTG FAM-AGAACTCCCCACTGGAAACAAGAAGCCCTAMRA
a
Conditions
Annealing temperature,
55 C; 1.5 mM MgCl2
Annealing temperature,
56 C; 1 mM MgCl2
a
a
Quantitative real-time PCR following standard conditions. FAM, 6-Carboxyfluorescein; TAMRA, tetramethylrhodamine.
Elisei et al. • Iodine Uptake in NIS-Transfected Thyroid Cells
The calculation of the cell absorbed dose per unit of administered
activity, DC(Gy/MBq), was made using the equation: DC ⫽ SC4C␶,
where SC-C are the S-factors for 7 ␮m diameter cells (26). ␶ was calculated
taking into account that the radioiodine kinetics in the cells can be
considered monoexponential, according to the following equation:
A(t) ⫽ A0 exp(t/␶), where A(t) is the radioiodine activity in the cell, A0
is the administered activity, t is time, and ␶ is the cumulated activity per
unit of administered activity.
It is worth noting that the calculation of DC(Gy/MBq) was based on
the following hypotheses: 1) radioiodine activity is homogeneously distributed in the cell culture; 2) dose is calculated taking into account only
the self-irradiation of the cell, neglecting the effect of the surrounding
radioactivity; and 3) radioiodine activity is considered uniformly distributed into the cell, i.e. the different distributions of the activity in
nucleus, cytoplasm, and cell surface are neglected.
Statistical analysis
Statistical analysis was performed by ANOVA for repeated measures
test and Student’s t test with Bonferroni correction using StatView 4.5
software (Abacus Concepts, Inc., Berkeley, CA). Results were considered
statistically significant at P ⬍ 0.05.
Results
Nineteen FRO and 16 TT clones transfected with hNIS and
two FRO and six TT clones transfected with the empty vector
were isolated after NIS stable transfection and neomycin
selection. Qualitative and quantitative RT-PCR of the transfected clones, performed with specific primers for the amino
and carboxyl termini of the NIS gene, revealed that 12 of 19
FRO and nine of 16 TT clones expressed the full-length NIS
mRNA. As expected, hNIS mRNA was not detected in clones
transfected with the empty vector.
All hNIS mRNA-expressing clones and clones transfected
with the empty vector were then subjected to radioiodine
uptake analysis; 11 of 12 FRO and four of nine TT clones
showed the ability to take up iodide (FRO clones from 2,800
to 56,000 cpm/106 cells and TT clones from 7,200 and 55,000
cpm/106 cells; Fig. 1). As a control, iodine uptake of the
FRTL-5 cell line was also measured (⬃10,000 cpm/106 cells).
As expected, none of the clones transfected with the empty
vector showed iodine uptake. The clones FRO-19 and TT-2,
J Clin Endocrinol Metab, June 2006, 91(6):2389 –2395
2391
which showed the highest levels of iodine uptake (56,000 and
55,000, respectively) and coincided with clones with the highest copy number of NIS mRNA, were selected for additional
experiments after performing a perchlorate inhibition test, demonstrating that the iodine uptake was NIS mediated.
The immunocytochemistry with anti-NIS antibody
showed strong positive staining on the plasma membrane of
FRO and TT clones positive for both NIS mRNA expression
and iodine uptake (Fig. 2, A and D). Weak cytoplasmic staining was detected in clones that were positive for NIS mRNA
expression but unable to take up iodine (Fig. 2, B and E).
Negative results were obtained in negative controls (wildtype FRO and TT cells; Fig. 2, C and F).
Iodine uptake reached the steady-state level (plateau) after
about 30 and 60 min in FRO-19 and TT-2 cells, respectively
(Fig. 3A). A rapid efflux of iodine was observed in FRO-19
cells, because about 50% of the cellular radioactivity was
released into the medium in about 9 min (t1/2 ⫽ 9 min),
whereas a slower efflux was detected in TT-2 cells (t1/2 ⫽ 20
min; Fig. 3B).
As shown in Table 2, treatment with 17-AAG and DIDS
caused a significant increase in iodine uptake at a plateau in
FRTL-5 (1.5- and 1.5-fold increases, respectively; P ⬍ 0.05).
In FRO-19 cells, the same effect was observed after treatment
with DIDS (1.2-fold increase). Total iodine uptake was not
affected by treatment with any of the four drugs in TT-2 cells.
As shown in Table 3, myricetin, lithium, 17-AAG, and
DIDS significantly reduced iodine efflux in FRTL-5 cells (P ⫽
0.003 to P ⬍ 0.0001). Thirty minutes after removal of 125I,
intracellular radioactivity was increased 20% when cells
were treated with lithium and 17-AAG, 25% when cells were
treated with DIDS, and 51% when cells were treated with
myricetin (compared with untreated cells).
When drugs were tested for their ability to affect iodine
efflux in cell lines transfected with NIS cDNA, iodine retention was significantly (P ⬍ 0.0001) increased by both DIDS
and 17-AAG in FRO-19 cells (Fig. 4). The intracellular iodine
content was significantly higher as early as 15 min after
FIG. 1. Iodide uptake in wild-type and transfected FRO and TT clones. Eleven of 12 FRO and four of nine TT clones showed the ability to take
up iodide (FRO clones from 2,800 to 56,000 cpm/106 cells and TT clones from 7,200 and 55,000 cpm/106 cells). As a control, iodine uptake of
the FRTL-5 cell line was also measured. Clones FRO-19 (A) and TT-2 (B) were chosen for additional experiments.
2392
J Clin Endocrinol Metab, June 2006, 91(6):2389 –2395
Elisei et al. • Iodine Uptake in NIS-Transfected Thyroid Cells
FIG. 2. Immunocytochemistry with
anti-NIS antibody of FRO-19 and TT-2
clones stably transfected with fulllength NIS cDNA. Clones able to express NIS mRNA and take up iodide (A
and D) showed correct localization of
NIS protein on the plasma membrane;
clones able to express NIS mRNA, but
not to take up iodide (B and E), showed
weak cytoplasmic staining; nontransfected clones (C and F) did not show any
staining.
removal of radioiodine from the medium compared with that
in untreated cells (⫹29% and ⫹23% in DIDS- and 17-AAGtreated cells, respectively), and it was almost doubled at 30
min (⫹103% and ⫹69%, respectively; Table 3). No significant
effect on iodine retention was produced by myricetin or
lithium in FRO-19 cells. No effect was produced by any drug
in the TT-2 cell line.
As shown in Table 4, calculation of the administered activity in the FRO-19 cell line (control vs. cells treated with
FIG. 3. Iodide kinetics of FRO-19 and TT-2 clones. A, Time
course of iodide uptake, showing a plateau at 30 min for
FRO-19 and at 60 min for TT-2. B, Iodide efflux, showing
a rapid efflux of iodine in FRO-19 cells (t1/2 ⫽ 9 min) and
a slower efflux in TT-2 cells (t1/2 ⫽ 20 min). Values are the
means of three replicate determinations.
17-AAG and DIDS) showed a significantly higher dose of
radioiodine in treated cells vs. controls (P ⬍ 0.01).
Discussion
Although NIS gene therapy is an attractive approach for
the treatment of dedifferentiated and/or undifferentiated
thyroid carcinomas, there are two major limits that reduce
the possibility of using radioiodine as a therapeutic agent (13,
Elisei et al. • Iodine Uptake in NIS-Transfected Thyroid Cells
J Clin Endocrinol Metab, June 2006, 91(6):2389 –2395
2393
TABLE 2. Increase in iodine uptake, expressed as picomoles per 106 cells, in cells treated with four different drugs
FRTL-5
FRO-19
TT-2
Untreated
Myricetin
17-AAG
Lithium
DIDS
48.7 ⫾ 9.3
325.5 ⫾ 3.3
909.6 ⫾ 80.8
60.0 ⫾ 5.4 (1.2)
263.3 ⫾ 13.2 (0.8)
1045.3 ⫾ 53.3 (1.1)
72.7 ⫾ 5.3 (1.5)a
281.3 ⫾ 12.0 (0.8)
1059.3 ⫾ 219.9 (1.1)
42.3 ⫾ 1.2 (0.9)
343.2 ⫾ 14.6 (1.0)
661.5 ⫾ 5.4 (0.7)
70.9 ⫾ 7.1 (1.5)a
401.3 ⫾ 16.1 (1.2)a
754.6 ⫾ 61.9 (0.8)
The fold increases are given in parentheses. All experiments were performed in triplicate; the results of one of the three experiments are
reported.
a
P ⬍ 0.05.
27): 1) the poor efficiency of NIS transfection and incorrect
localization of the encoded protein; and 2) the short time of
iodide retention within the cell. Indeed, in vivo studies have
shown that radioiodine treatment does not significantly
change the volume of tumors induced in mice after sc injection of NIS-transfected thyroid cells (14).
Regarding the problem of transfection efficiency, we observed a quite important difference in the two cell lines, FRO
and TT. This phenomenon is likely to be dependent on the
different cell types and suggests that the cell system to be
transfected should be accurately chosen for clinical applications. In this regard, literature data are limited. However, a
transfection efficiency of about 16% has been reported in the
transfection of both prostate and thyroid cancer cell lines (28,
29), which is much lower than those obtained by us in FRO
(11 of 19; 58%) and TT (four of 16; 25%) cells. It is likely that
a different delivery system of the transfected gene (e.g. adenovirus or retrovirus) might be more efficient (15). Furthermore, we observed that in several NIS mRNA-expressing
clones, the NIS protein was not on the plasma membrane and
was not functional. We can speculate that clones expressing
NIS mRNA but unable to take up iodide were subjected to
posttranscriptional gene silencing. This phenomenon is very
common in transgenic plants and is due to a mechanism also
present in animal cells that has the final aim of defending the
cell from foreign DNA (30, 31). Although we think that these
technical problems should be taken into account and considered as possible limiting factors of this approach, several
promising data on NIS gene transfer from various groups
have been reported in different tumor cell lines (13, 32, 33).
Regarding the problem of the short retention time of iodine
in transfected cells, we demonstrated for the first time that
treatment with 17-AAG and DIDS of FRO cells, derived from
a human anaplastic thyroid carcinoma and transfected with
NIS cDNA, is able to significantly reduce iodide efflux, producing an increase in the iodide retention time that improves
the killing potential of radioiodine treatment. The same
drugs were not able to increase iodide retention in TT cells
derived from a human medullary thyroid carcinoma and
transfected with NIS cDNA. The reason for this discrepancy
may be that FRO and TT cell lines derive from different
thyroid tumor histotypes.
The mechanism of iodide efflux is still not fully understood, and several unknown proteins are likely to be involved (34). At present, only pendrin (PDS) and the recently
characterized protein apical iodide transporter (AIT) have
TABLE 3. Effect of treatment of FRTL-5 and FRO-19 cells with the four drugs on radioiodide efflux
Time (min)
FRTL-5 cells
3
6
9
12
15
18
21
24
27
30
P value
FRO-19 cells
3
6
9
12
15
18
21
24
27
30
P value
% Intracellular radioactivity
Untreated
Myricetin
Lithium
17-AAG
DIDS
49.7 ⫾ 3.5
31.6 ⫾ 0.3
23.5 ⫾ 2.3
19.0 ⫾ 3.6
14.3 ⫾ 2.2
11.7 ⫾ 2.4
9.5 ⫾ 2.3
7.9 ⫾ 2.3
6.0 ⫾ 2.1
4.4 ⫾ 2.1
55.7 ⫾ 3.3 (12)
36.1 ⫾ 2.6 (14)
27.5 ⫾ 2.7 (17)
22.5 ⫾ 4.8 (18)
18.9 ⫾ 5.2 (32)
15.8 ⫾ 5.0 (34)
13.5 ⫾ 4.9 (41)
10.8 ⫾ 4.5 (37)
8.7 ⫾ 4.6 (45)
6.6 ⫾ 4.0 (51)
⬍0.0001
55.0 ⫾ 4.1 (10)
36.8 ⫾ 1.7 (16)
27.6 ⫾ 1.7 (17)
22.1 ⫾ 3.4 (16)
17.9 ⫾ 3.9 (25)
15.2 ⫾ 4.5 (30)
12.5 ⫾ 4.4 (31)
10.3 ⫾ 4.3 (31)
7.1 ⫾ 1.4 (18)
5.3 ⫾ 1.6 (20)
⬍0.0001
54.2 ⫾ 6.3 (9)
37.1 ⫾ 4.2 (17)
28.2 ⫾ 2.9 (20)
22.7 ⫾ 3.0 (19)
19.0 ⫾ 3.5 (32)
14.1 ⫾ 1.4 (20)
11.8 ⫾ 1.9 (24)
9.4 ⫾ 2.1 (20)
7.6 ⫾ 2.2 (27)
5.3 ⫾ 2.1 (20)
0.0004
54.2 ⫾ 5.5 (9)
35.4 ⫾ 2.5 (12)
25.0 ⫾ 1.6 (6)
19.9 ⫾ 0.5 (5)
16.2 ⫾ 1.7 (13)
13.4 ⫾ 2.4 (15)
11.3 ⫾ 2.4 (18)
9.3 ⫾ 2.7 (18)
7.3 ⫾ 2.8 (21)
5.8 ⫾ 3.0 (25)
0.0033
80.0 ⫾ 1.4
62.7 ⫾ 1.9
48.7 ⫾ 2.8
38.2 ⫾ 2.3
29.7 ⫾ 2.1
23.2 ⫾ 1.7
17.9 ⫾ 1.9
13.9 ⫾ 1.9
10.5 ⫾ 1.8
7.5 ⫾ 1.7
76.1 ⫾ 4.7 (⫺5)
58.9 ⫾ 5.9 (⫺6)
45.4 ⫾ 6.2 (⫺6)
35.0 ⫾ 5.2 (⫺8)
27.1 ⫾ 4.5 (⫺9)
20.7 ⫾ 4.1 (⫺10)
15.6 ⫾ 3.2 (⫺13)
11.9 ⫾ 2.6 (⫺14)
8.9 ⫾ 2.5 (⫺15)
6.6 ⫾ 2.3 (⫺13)
77.3 ⫾ 3.4 (⫺3)
61.1 ⫾ 4.7 (⫺2)
48.1 ⫾ 5.1 (⫺1)
37.7 ⫾ 3.1 (⫺1)
30.0 ⫾ 2.4 (0)
23.6 ⫾ 2.1 (1)
18.3 ⫾ 1.8 (2)
14.3 ⫾ 1.3 (3)
11.0 ⫾ 1.2 (4)
8.1 ⫾ 1.4 (7)
79.7 ⫾ 4.3 (0)
65.4 ⫾ 5.3 (4)
54.0 ⫾ 5.4 (11)
44.9 ⫾ 4.0 (17)
36.7 ⫾ 3.9 (23)a
30.3 ⫾ 3.1 (30)
24.9 ⫾ 2.7 (39)
20.2 ⫾ 2.4 (45)
16.1 ⫾ 2.5 (53)
12.7 ⫾ 2.6 (69)
⬍0.0001
82.3 ⫾ 2.0 (3)
67.9 ⫾ 4.6 (8)
56.3 ⫾ 5.3 (16)
46.6 ⫾ 4.7 (22)
38.6 ⫾ 4.5 (29)a
32.1 ⫾ 4.7 (38)
26.8 ⫾ 4.9 (50)
22.2 ⫾ 4.9 (60)
18.5 ⫾ 4.8 (76)
15.3 ⫾ 4.8 (103)
⬍0.0001
The values are the mean ⫾ SD of three experiments, each performed in triplicate. The increase (%) in intracellular radioactivity in treated
vs. untreated cells is shown in parentheses.
a
The intracellular iodine content became significantly higher in treated vs. untreated cells as early as 15 min.
2394
J Clin Endocrinol Metab, June 2006, 91(6):2389 –2395
Elisei et al. • Iodine Uptake in NIS-Transfected Thyroid Cells
FIG. 4. Iodide efflux of FRO-19 cells under basal conditions (red line)
and after treatment with 17-AAG (green line) and DIDS (blue line).
Iodine retention was significantly increased by both DIDS and 17AAG. The values are the means of three experiments, each performed
in triplicate (SD are indicated by bars).
been recognized as passive iodide transporters (35, 36). It has
been shown that DIDS directly inhibits iodide efflux by interfering with the iodide-specific channel located at the apical pole of thyroid cells (19). Unfortunately, that study was
performed before cloning of the genes coding for the two
above-mentioned proteins (PDS and AIT), and we cannot
exclude or confirm that, generally speaking, DIDS might act
through the inhibition of PDS and/or AIT. However, on the
basis of our previous demonstration that the FRO cell line
does not express the PDS gene (12), we can exclude an involvement of this gene. Because DIDS is a well-known inhibitor of anionic channels, we can also speculate that the
iodide efflux of FRO-19 cells might be mediated by other
anionic channels, not specific, but permeable to iodide,
which have been demonstrated to be expressed in thyroid
cells (37, 38).
17-AAG treatment has recently been demonstrated to decrease the iodide efflux in normal rat thyroid cells, PCCL3
(18). This action seems to be specific for thyroid cells, because
it was not observed in other cell lines. We obtained similar
results in a cell line derived from normal rat thyroid cells
(FRTL-5) and in FRO-19 cells, derived from a human anaplastic thyroid carcinoma and transfected with NIS cDNA.
The negative result obtained with the TT-2 cell line treated
with 17-AAG confirms a certain specificity of 17-AAG action
in decreasing iodide efflux in follicular cells. 17-AAG is also
a known inhibitor of the 90-kDa heat shock protein (hsp90),
and it is likely that through the inhibition of hsp90, it acts as
TABLE 4. Cumulative activity and dose per unit of administered
activity of radioiodine (131I) in FRO-19 cells untreated (C) and
treated with 17-AAG and DIDS
FRO-19 cells
C
17-AAG
DIDS
␶ (min) ⫾
SD
a
12.4 ⫾ 0.3
14.9 ⫾ 0.4
15.8 ⫾ 0.7
D (Gy/MBq) ⫾
SD
b
0.232 ⫾ 0.006
0.279 ⫾ 0.007c
0.300 ⫾ 0.01c
c
a
␶ values for the three experiences were obtained by fitting the
experimental data using the following equation: A(t) ⫽ A0关exp(⫺1/
␶)t兴.
b
D(Gy/MBq) ⫽ SC-C ␶.
c
P ⬍ 0.01
an antineoplastic agent, both by reducing cellular proliferation and by inducing apoptosis. hsp90 is a chaperone for a
group of proteins, several of which are involved in cancer
proliferation (22). Recently, one of the most important oncogenes for papillary thyroid cancer (RET/PTC1) has been
demonstrated to be a client protein of hsp90 (18). The combination of antineoplastic activity with the ability to reduce
iodide efflux, as shown by our data, makes 17-AAG a good
candidate for the treatment of thyroid cancer, especially
when patients are losing the ability to take up iodine.
17-AAG and DIDS were effective in reducing iodide efflux
in FRO-19 as well as in FRTL-5 cells. The same approach does
not seem to be applicable in the TT cell line, because none of
the tested drugs was able to interfere with iodide transport
in this cell line. However, by analyzing the efflux kinetics, it
appears evident that iodide transport is different in the two
cell systems; in fact, FRO-19 cells release half the iodide after
9 min, whereas TT-2 cells release it after about 20 min. This
finding is in agreement with our previous report of the expression of the thyroperoxidase gene in TT cells (12) and the
recent observation of iodide organification in NIS-transfected TT cells (39). Although we did not evaluate the organification of iodide in the cells, it is conceivable to hypothesize that iodide efflux in TT-2 cells is slower than that
in FRO-19 cells because of a partial iodide organification in
the first cell type, but not in the latter. As a consequence, in
this cell system, it is more conceivable to improve the rate of
iodine organification, as previously suggested (40), than to
reduce iodine efflux. Nevertheless, other drugs might be
explored in the future.
As far as the in vivo therapeutic effect is concerned, it
could be of interest to determine the lowest absorbed dose
per administered activity able to induce cell death and/or
damage. To our knowledge, this information is still unknown. However, there are a few studies clearly demonstrating that 131I treatment of mice with xenografted tumors, obtained using NIS-transfected cells, reduced the
growth rate and volume of the tumor (13, 32) and/or
increased the survival rate of the animal models (33) despite the rapid iodide efflux. Based on these results, we
may assume that in our case also a therapeutic effect of 131I
treatment could be obtained in an experimental model and
could be improved by the simultaneous treatment with
17-AAG or DIDS, as demonstrated by the statistically significant increase in the absorbed dose of 131I. However,
because the iodide efflux in vivo might be significantly
different from that in vitro, our data have to be confirmed
in xenografted animals before hypothesizing about any
human therapeutic protocol.
In conclusion, in this study we showed that NIS gene
therapy, although very promising, may have some technical
limitations to be considered before its in vivo application.
However, our results indicated that once these limitations
have been overcome, certain drugs (i.e. 17-AAG and DIDS)
could be used to prolong the retention time of 131I in thyroid
tumoral cells transfected with the NIS gene. Thus, our data
reinforce the hope of using this approach for future clinical
application, especially in those patients with DTC who are no
longer responsive to conventional therapy.
Elisei et al. • Iodine Uptake in NIS-Transfected Thyroid Cells
Acknowledgments
We thank Dr. Sissy Jhiang, codirector of the Department of Physiology and Department of Internal Medicine, Ohio State University (Columbus, OH), for the kind gift of NIS full-length cDNA; Dr. Jennifer
Cummings for revising the language of the manuscript; and Dr. Lucio
Masserini for revising statistics.
Received November 14, 2005. Accepted March 3, 2006.
Address all correspondence and requests for reprints to: Dr. Rossella
Elisei, Department of Endocrinology and Metabolism, Via Paradisa 2,
University of Pisa, 56124 Pisa, Italy. E-mail: [email protected].
This work was supported in part by grants from Ministero
dell’Istruzione Universitaria e Ricerca Scientifica and Associazione Italiana per la Ricerca sul Cancro.
R.E., A.V., R.C., P.F., F.B., F.S., C.T., F.P., and A.P. have nothing to
declare.
References
1. Braverman, LE, Utiger DL 2000 Werner and Ingbar’s the thyroid: a fundamental and clinical text, 8th Ed. Philadelphia: Lippincott Williams & Wilkins
2. Jhiang SM 2000 Regulation of sodium iodine symporter. Rev Endocr Metab
Disord 1:205–215
3. Schlumberger M, Pacini F 1999 Thyroid tumors. Paris: Nucleon Editions;
117–129
4. Arturi F, Russo D, Schlumberger M, Du Villard JA, Caillou B, Vigneri P 1998
Iodine symporter gene expression in human thyroid tumors. J Clin Endocrinol
Metab 83:2193–2196
5. Elisei R, Pinchera A, Romei C, Gryczynska M, Pohl V, Maenhaut C, Fugazzola L, Pacini F 1994 Expression of thyrotropin receptor (TSH-R), thyroglobulin, thyroperoxidase and calcitonin messenger ribonucleic acid in thyroid
carcinoma: evidence of TSH-R gene transcript in medullary histotype. J Clin
Endocrinol Metab 78:867
6. Haugen BR 1999 Management of the patient with progressive radioiodine
non-responsive disease. Semin Surg Oncol 16:34 – 41
7. Santini F, Bottici V, Elisei R, Montanelli L, Mazzeo S, Basolo F, Pinchera A,
Pacini F 2002 Cytotoxic effects of carboplatinum and epirubicin in the setting
of an elevated serum thyrotropin for advanced poorly differentiated thyroid
cancer. J Clin Endocrinol Metab 87:4160 – 4165
8. Smanik PA, Liu Q, Furminger TL, Ryu K, Xing S, Mazzaferri EL, Jhiang SM
1996 Cloning of the human sodium iodide symporter. Biochem Biophys Res
Commun 226:339 –345
9. Dai G, Levy O, Carrasco N 1996 Cloning and characterization of the thyroid
iodide transporter. Nature 379:458 – 460
10. Spitzweg C, Harrington KJ, Pinke LA, Vile RG, Morris JC 2001 The sodium
iodide symporter and its potential role in cancer therapy. J Clin Endocrinol
Metab 86:3327–3335
11. Chung JK 2002 Sodium iodide symporter: its role in nuclear medicine. J Nucl
Med 43:1188 –1200
12. Elisei R, Vivaldi A, Agate L, Ciampi R, Molinaro E, Piampiani P, Romei C,
Faviana P, Basolo F, Miccoli P, Capodanno A, Collecchi P, Pacini F, Pinchera
A 2005 All-trans retinoic acid treatment inhibits the growth of retinoic acid
receptor ␤ messenger ribonucleic acid expressing thyroid cancer cell lines but
does not reinduce the expression of thyroid specific genes. J Clin Endocrinol
Metab 90:2403–2411
13. Smit JW, Schroder-van der Elst JP, Karperien M, Que I, Stokkel M, van der
Heide D, Romijn JA 2002 Iodide kinetics and experimental 131I therapy in a
xenotransplanted human sodium iodide symporter transfected human follicular thyroid carcinoma cell line. J Clin Endocrinol Metab 87:1247–1253
14. Shimura H, Haraguchi K, Miyazaki A, Endo T, Onaya T 1997 Iodide uptake
and experimental 131I therapy in transplanted undifferentiated thyroid cancer
cells expressing the Na⫹/I⫺ symporter gene. Endocrinology 138:4493– 4496
15. Haberkorn U, Henze M, Altmann A, Jiang S, Morr I, Mahmut M, Peschke
P, Kubler W, Debus J, Eisenhut M 2001 Transfer of the human NaI symporter
gene enhances iodide uptake in hepatoma cells. J Nucl Med 42:317–325
16. Schroder-van der Elst JP, van der Heide D, Romijin JA, Smit JA 2004 Differential effects of natural flavonoids on growth and iodine content in a human
Na⫹/I⫺ symporter-transfected follicular thyroid carcinoma cell line. Eur J
Endocrinol 150:557–564
17. Koong SS, Reynolds JC, Movius EG, Keenan AM, Ain KB, Lakshman MC,
Robbins J 1999 Lithium as a potential adjuvant to 131I therapy of metastatic,
well differentiated thyroid carcinoma. J Clin Endocrinol Metab 84:912–916
J Clin Endocrinol Metab, June 2006, 91(6):2389 –2395
2395
18. Marsee DK, Venkateswaran A, Tao H, Vadysirisack D, Zhang Z, Vandre DD,
Jhiang SM 2004 Inhibition of heat shock protein 90, a novel RET/PTC1associated protein, increases radioiodine accumulation in thyroid cells. J Biol
Chem 279:43990 – 43997
19. Amphoux-Fazekas T, Samih N, Hovsepian S, Aouani A, Beauwens R, Fayet
G 1998 DIDS (4,4⬘-diisothiocyanatostilbene-2,2⬘-disulfonic acid) increases iodine trapping, inhibits thyroperoxydase and antagonizes the TSH-induced
apical iodine efflux in porcine thyroid cells. Mol Cell Endocrinol 141:129 –140
20. Ong KC, Khoo HE 1997 Biological effects of myricetin. Gen Pharmacol 29:
121–126
21. Urabe M, Hershman JM, Pang XP, Murakami S, Sugawara M 1991 Effect of
lithium on function and growth of thyroid cells in vitro. Endocrinology 129:
807– 814
22. Bagatell R, Whitesell L 2004 Altered Hsp90 function in cancer: a unique
therapeutic opportunity. Mol Cancer Ther 3:1021–1030
23. Cabantchik ZI, Greger R 1992 Chemical probes for anion transporters of
mammalian cell membranes. Am J Physiol 262:C803–C827
24. Elisei R, Vivaldi A, Agate L, Molinaro E, Nencetti C, Grasso L, Pinchera A,
Pacini F 2004 Low specificity of blood thyroglobulin messenger ribonucleic
acid assay prevents its use in the follow-up of differentiated thyroid cancer
patients. J Clin Endocrinol Metab 891:33–39
25. Ambesi Impiombato FS, Parks LAM, Coons HG 1980 Culture of hormonedependent epithelial cells from rat thyroids. Proc Natl Acad Sci USA 77:3455
26. Goddu SM, Rao DV, Howell RW 1994 Multicellular dosimetry for micrometastases: dependence of self-dose versus cross-dose to cell nuclei on type
and energy of radiation and subcellular distribution of radionuclides. J Nucl
Med 35:521–530
27. Lee WW, Lee B, Kim SJ, Jin J, Moon DH, Lee H 2003 Kinetics of iodide uptake
and efflux in various human thyroid cancer cells by expressing sodium iodide
symporter gene via a recombinant adenovirus. Oncol Rep 10:845– 849
28. Spitzweg C, Zhang S, Bergert ER, Castro MR, McIver B, Heufelder AE,
Tindall DJ, Young CYF, Morris JC 1999 Prostate-specific antigen (PSA) promoter-driven androgen-inducible expression of sodium iodide symporter in
prostate cancer cell lines. Cancer Res 59:2136 –2141
29. Smith JWA, Schroder-van der Elst JP, Karperien M, Que I, van der Pluijm
G, Goslings B, Romijn JA, van der Heide D 2000 Reestablishment of in vitro
and in vivo iodide uptake by transfection of the human sodium iodide symporter (hNIS) in a hNIS defective human thyroid carcinoma cell line. Thyroid
10:939 –943
30. Almeida R, Allshire RC 2005 RNA silencing and genome regulation. Trends
Cell Biol 15:251–258
31. Fagard M, Vaucheret H 2000 (Trans)gene silencing in plants: how many
mechanisms? Annu Rev Plant Physiol Plant Mol Biol 51:167–194
32. Spitweg C, O’Connor MK, Bergert ER, Tindall DJ, Young CYF, Morris JC
2000 Treatment of prostate cancer by radioiodine therapy after tissue-specific
expression of the sodium iodide symporter. Cancer Res 60:6526 – 6530
33. Cho JY, Shen DHY, Yang W, Williams B, Buckwalter TLF, La Perle KMD,
Hinkle G, Pozderac R, Kloos R, Nagaraja HN, Barth RF, Jhiang SM 2002 In
vivo imaging and radioiodine therapy following sodium iodide symporter
gene transfer in animal model of intracerebral gliomas. Gene Ther 9:1139 –1145
34. Rousset BA, Dunn JT 2004 Thyroid hormone synthesis and secretion. In: The
thyroid and its diseases, Chap 2. http://www.thyroidmanager.org
35. Royaux IE, Suzuki K, Mori A, Katoh R, Everett LA, Kohn LD, Green ED 2000
Pendrin, the protein encoded by the Pendred syndrome gene (PDS), is an apical
porter of iodide in the thyroid and is regulated by thyroglobulin in FRTL-5
cells. Endocrinology 141:839 – 845
36. Rodriguez AM, Perron B, Lacroix L, Caillou B, Leblanc G, Schlumberger M,
Bidart JM, Pourcher T 2002 Identification and characterization of a putative
human iodide transporter located at the apical membrane of thyrocytes. J Clin
Endocrinol Metab 87:3500 –3503
37. Armstrong JW, Cragoe Jr EJ, Bourke JR, Huxham GJ, Manley SW 1992
Chloride conductance of apical membrane in cultured porcine thyroid cells
activated by cyclic AMP. Mol Cell Endocrinol 88:105–110
38. Devuyst O, Golstein PE, Sanches MV, Piontek K, Wilson PD, Guggino WB,
Dumont JE, Beauwens R 1997 Expression of CFTR in human and bovine
thyroid epithelium. Am J Physiol 272:C1299 –C308
39. Cengic N, Baker CH, Schutz M, Goke B, Morris JC, Spitzweg C 2005 A novel
therapeutic strategy for medullary thyroid cancer based on radioiodine therapy following tissue-specific sodium iodide symporter gene expression. J Clin
Endocrinol Metab 90:4457– 4464
40. Huang M, Batra RK, Kogai T, Lin YQ, Hershman JM, Lichtenstein A, Sharma
S, Zhu LX, Brent GA, Dubinett SM 2001 Ectopic expression of the thyroperoxidase gene augments radioiodide uptake and retention mediated by the
sodium iodide symporter in non-small cell lung cancer. Cancer Gene Ther
8:612– 618
JCEM is published monthly by The Endocrine Society (http://www.endo-society.org), the foremost professional society serving the
endocrine community.