© 1999 by Oxford University Press
Journal of the National Cancer Institute, Vol. 91, No. 1, 46-53,
January 6, 1999
© 1999 Oxford University Press
ARTICLES |
Reversal of Tamoxifen Resistance of Human Breast Carcinomas In Vivo by Neutralizing Antibodies to Transforming Growth Factor-ß
Affiliations of authors: C. L. Arteaga, Department of Medicine, Vanderbilt University School of Medicine, Vanderbilt Cancer Center, Department of Veteran Affairs Medical Center, Nashville, TN; K. M. Koli, T. C. Dugger, Department of Cell Biology, Vanderbilt University School of Medicine, Nashville; R. Clarke, Lombardi Cancer Center, Georgetown University, Washington, DC.
Correspondence to: Carlos L. Arteaga, M.D., Division of Medical Oncology, Vanderbilt University School of Medicine, 22nd Ave., S., 1956 TVC, Nashville, TN 37232-5536 (e-mail carlos.arteaga{at}mcmail.vanderbilt.edu).
 |
ABSTRACT |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionReferences |
|---|
BACKGROUND: Overexpression of transforming growth factor (TGF)-ß has been reported in human breast carcinomas resistant to antiestrogen tamoxifen, but the role of TGF-ß in this resistant phenotype is unclear. We investigated whether inhibition of TGF-ß2, which is overexpressed in LCC2 tamoxifen-resistant human breast cancer cells, could modify antiestrogen resistance. METHODS: TGF-ß2 expression was evaluated in LCC2 cells and tamoxifen-sensitive LCC1 cells by northern blot analysis. Secreted TGF-ß activity was quantified by use of an 125I-TGF-ß competitive radioreceptor assay. Sensitivity to tamoxifen was measured in a soft agarose colony-forming assay and in a xenograft model in nude and beige/nude mice. Natural killer (NK) cell cytotoxicity was measured by 51Cr release from LCC1 and LCC2 cell targets coincubated with human peripheral blood mononuclear cells. Decrease in TGF-ß2 expression in LCC2 cells was achieved by treatment with antisense oligodeoxynucleotides and confirmed by TGF-ß2 immunoblot analysis. RESULTS AND CONCLUSIONS: The proliferative response of LCC2 cells to tamoxifen in vitro was not altered by TGF-ß neutralizing antibodies. However, established LCC2 tumors in nude mice treated with tamoxifen plus TGF-ß antibodies failed to grow, whereas tumors treated with tamoxifen plus a control antibody continued to proliferate. This reversal of tamoxifen resistance by TGF-ß antibodies did not occur in beige/nude mice, which lack NK-cell function, suggesting that immune mechanisms may be involved in the antitumor effects of tamoxifen. Antisense TGF-ß2 oligodeoxynucleotides enhanced the NK sensitivity of LCC2 cells in the presence of tamoxifen. Finally, LCC1 tumors were markedly more sensitive to tamoxifen in NK-active than in NK-deficient mice. IMPLICATIONS: These data suggest that host NK function mediates, in part, the antitumor effect of tamoxifen and that TGF-ß2 may abrogate this mechanism, thus contributing to tamoxifen resistance.
|
INTRODUCTION |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Transforming growth factors (TGF)-ßs represent a large family of polypeptides involved in the regulation of cellular growth, differentiation, development, morphogenesis, and production of extracellular matrix (1,2). Three homologous mammalian isoforms have been reported: TGF-ß1, TGF-ß2, and TGF-ß3. Although these isoforms overall share similar receptor-binding properties and biologic effects in multiple experimental systems (1,2), the mouse knockout phenotypes for each isoform are different (3-5), suggesting that their effects in vivo may not fully overlap. Expression of TGF-ß1, -ß2, and -ß3 in human breast carcinoma cell lines and tumor tissues varies considerably [see(6)]. Several lines of data support the notion that mammary cell TGF-ßs are autocrine regulators of tumor and nontumor breast epithelial cells (7-9). Although growth inhibition of breast tumor cells with antiestrogens is associated with enhanced secretion of TGF-ß protein (9), transfection of a dominant negative type II TGF-ß receptor into MCF-7 breast cancer cells does not abrogate the response to tamoxifen (10), suggesting that autocrine TGF-ßs may not be universal mediators of tamoxifen action. On the other hand, several reports suggest that, in breast cancer cells, the TGF-ßs are important mediators of tumor progression by fostering critical tumor/host cell interactions like the enhancement of peritumoral angiogenesis and stroma and the inhibition of mechanisms of immune surveillance [reviewed in (6)]. In some cases, overexpression of TGF-ß1 has been temporally associated with estrogen independence and/or resistance to antiestrogens (11-15). Increased immunohistochemical staining for TGF-ß1 protein in breast tumor sections is associated with shorter postmastectomy survival independent of other prognostic factors (16). We have studied the expression and function of TGF-ßs in the tamoxifen-sensitive line LCC1 and the tamoxifen-resistant line LCC2 derived from MCF-7 human breast cancer cells. These cell lines were derived from MCF-7 mouse xenografts established in ovariectomized athymic mice. The LCC2 subline was selected in vitro in the presence of 4-hydroxytamoxifen. Both cell lines exhibit functional estrogen receptors (ERs) and progesterone receptors (PgRs) and form tumors in athymic nude mice in the presence or absence of added estradiol (17,18). This situation provides an experimental model to study mechanisms of tamoxifen in vivo resistance that do not involve loss of the ER.
|
METHODS |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Cell lines and reagents. The LCC1 (tamoxifen-sensitive) and LCC2 (tamoxifen-resistant) breast cancer lines were maintained in phenol red-containing improved minimal essential medium (IMEM; Life Technologies, Inc. [GIBCO BRL], Gaithersburg, MD) supplemented with 10% fetal calf serum (FCS; Hazleton Laboratories, Madison, WI) and 10 nM human insulin. Tamoxifen citrate was purchased from Sigma Chemical Co. (St. Louis, MO) and stored as a 1-mM stock solution in absolute ethanol. The 2G7 and 12H5 immunoglobulin G2 (IgG2) monoclonal TGF-ß-specific antibodies (provided by B. M. Fendly; Genentech Inc., South San Francisco, CA) were generated against human recombinant TGF-ß1 and have been described previously (19). The 12H5 IgG2 is devoid of TGF-ß neutralizing activity, while the 2G7 antibody blocks growth inhibition by TGF-ß1, -ß2, and -ß3 when tested against Mv1Lu mink lung epithelial cells (19).
RNA isolation and northern blot hybridization. Poly(A)+ RNA was purified from subconfluent monolayer cultures using oligo-dT cellulose chromatography as described (20). For northern blot analysis, 3 µg of messenger RNA (mRNA) was fractionated on 1.2% agarose gels containing formaldehyde and transferred to nylon membranes (Micron Separations Inc., Westboro, MA) by capillary transfer. Prehybridization and hybridization were performed at 42 °C in 50% formamide, 250 µg/mL single-stranded DNA, 1x Denhardt's solution, 50 µg/mL poly(A), 0.1% sodium dodecyl sulfate (SDS), and 5x standard saline citrate. Complementary DNA (cDNA) probes for TGF-ß1 (21), TGF-ß2 (22), and cyclophilin (23) were labeled with [32P]deoxycytidine triphosphate (>3000 Ci/mmol; Amersham Life Science Inc., Arlington Heights, IL) using a Rediprime labeling kit (Amersham Life Science Inc.).
Preparation of cell-conditioned medium and TGF-ß radioreceptor assay. Secreted TGF-ß bioactivity was measured in serum-free IMEM conditioned for 24 hours by adherent breast cancer cells as described previously (24). To activate secreted latent TGF-ß, the conditioned medium was acidified with 1 N HCl to pH 1.5 for 1 hour at 4 °C and then neutralized with 1 N NaOH before testing in a TGF-ß radioreceptor assay (24) utilizing AKR-2B (84A) mouse fibroblasts. Binding was performed in six-well plates in 1 mL/well binding buffer (24) containing 0.25 ng/mL. 125I-TGF-ß1 (specific activity, 173 µCi/µg; Du Pont NEN, Boston, MA) competed with variable volumes of conditioned medium. Human recombinant TGF-ß1 (Genentech Inc.) was used to generate a standard curve from which the receptor binding activity of conditioned medium was calculated and then standardized to the number of cells from which the conditioned medium was collected. In this binding assay with AKR-2B (84A) cells, TGF-ß1 and TGF-ß2 are equipotent in displacing 125I-TGF-ß1 binding (25). Addition of 1 µM tamoxifen directly to the binding buffer had no effect on TGF-ß1 binding in this assay (Arteaga CL: unpublished data).
Natural killer (NK) and lymphokine-activated killer (LAK) cell cytotoxicity assays. Sparse adherent cultures of LCC1 or LCC2 target cells (104 cells per well in a 96-well plate) were labeled for 4-6 hours at 37 °C with approximately 200 µCi/mL 51Cr (specific activity, 400-1200 Ci/g; Du Pont NEN) in IMEM supplemented with 10% FCS. Unbound radioactivity was removed after two washes with growth medium before the addition of effector cells. Human peripheral blood lymphocytes (PBLs) from healthy volunteers were prepared by Ficoll-Hypaque gradient centrifugation. After two washes, variable numbers of NK effector cells were added to the 51Cr-labeled breast cancer cell targets in a final volume of 0.2 mL/well in quadruplicate. To generate LAK effector cells, PBLs were incubated for 5 days at 37 °C in 5% CO2 with 1000 U/mL human recombinant interleukin 2 (Chiron Therapeutics, Emeryville, CA) and then added in ratios ranging from 5 : 1 to 100 : 1 to freshly labeled tumor cell targets. Spontaneous 51Cr release and PBL-induced (experimental) 51Cr release were measured in both assays after an overnight coincubation, and percent cytotoxicity was calculated as described previously (26) by the formula: ([experimental release - spontaneous release]/total release) x 100.
Mouse studies. Five- to 8-week-old female athymic (nude) mice (Harlan Sprague-Dawley, Inc., Madison, WI) or NIH-III beige/nude mice (Taconic Farms, Inc., Germantown, NY) were inoculated subcutaneously just caudal to the forelimb with approximately 5 x 106 LCC1 or LCC2 tumor cells in 0.25 mL of serum-free IMEM via a 22-gauge needle. Variable times after tumor cell inoculation, 25-mg, 60-day release tamoxifen pellets (Innovative Research, Toledo, OH) were implanted subcutaneously in the back at a distant site from the tumor via a 10-gauge trocar. In some experiments, animals received daily intraperitoneal injections of 2G7 or 12H5 antibodies (100 µg each) in a 0.2-mL volume via a 26-gauge needle. Tumor diameters were measured serially with calipers, and tumor volume was calculated by the formula: volume = width2 x length/2. At the completion of the experiments, some tumors were removed, fixed in 10% formalin, and paraffin-embedded, and 5-µm-thick sections were cut, stained with hematoxylin-eosin, and subjected to light microscopy. Care of all mice used in these studies was in accord with institutional guidelines.
Effects of antisense TGF-ß2 oligonucleotides and immunoblotting of TGF-ß2 secreted in cell-conditioned medium. TGF-ß2 antisense phosphorothioate oligodeoxynucleotides as well as a nonsense oligonucleotide sequence of similar length and with the same GC content (used as control) were synthesized by and purchased from Ransom Hill Bioscience, Inc. (Ramona, CA), on the basis of a recent report (27). A search in the European Molecular Biology Laboratory (EMBL) GenBank revealed no homologies between the 14-mer antisense sequence used in this study and the 5' coding region of any known gene sequence. For inhibition of TGF-ß2 expression, LCC2 cells were treated for 48 hours at 37 °C in serum-free IMEM with 3 µM antisense or nonsense oligonucleotides. Cell medium was then collected, cleared of cell debris by centrifugation, concentrated approximately 30-fold in a Centriprep 30 MW column (Millipore Corp., Bedford, MA), boiled in Laemmli sample buffer containing the reducing agent dithiothreitol (DTT), and resolved by 15% SDS-polyacrylamide gel electrophoresis. Following transfer to nitrocellulose, membranes were subjected to an immunoblot procedure that utilizes a TGF-ß2 polyclonal antiserum R & D Systems Inc., Minneapolis, MN). Detection of TGF-ß2 reactive bands in autoradiograms was performed with horseradish peroxidase-linked antirabbit immunoglobulins and enhanced chemiluminescence.
Statistical analyses. To analyze tumor growth curves, we used a repeated-measures analysis of variance (ANOVA) model with a serial correlation structure. The analysis was based on the logarithm of the tumor volume to stabilize the variance. To avoid strict assumptions about normality, the General Estimating Equation approach was used to fit the model, as implemented in SAS/PROC GENMOD software (SAS Institute, Inc., Cary, NC). The in vitro NK and LAK cytotoxicity data were analyzed using ANOVA, as implemented in SAS/PROC GLM software. For these cytotoxicity experiments, each data point represents the mean cpm of 51Cr released from labeled target cells in four wells. If the standard deviations of the quadruplicate determinations were less than 10% of the mean value (before the experimental release was subjected to the formula for calculating % cytotoxicity [above]), they were not included in the figures.
|
RESULTS |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
TGF-ß Expression and Immune Sensitivity of LCC1 and LCC2 Breast Cancer Cells
We first examined steady-state mRNA levels of TGF-ß1 and
TGF-ß2 in tamoxifen-sensitive and tamoxifen-resistant cells. LCC2
cells express similar low levels of TGF-ß1 mRNA (not shown) but
greater than 20-fold higher levels of TGF-ß2 mRNA relative to LCC1
cells (Fig. 1, A).
Protein levels in serum-free
medium conditioned by LCC2 cells, as measured in a
125I-TGF-ß1 radioreceptor assay, were 6.8 ng of TGF-ß
equivalents/106 cells per 24 hours in the absence of acid
activation in vitro, whereas LCC1 cells only secreted lower
levels of latent TGF-ß activity (Fig. 1, B). A 24-hour incubation
with 1 µM tamoxifen up-regulated TGF-ß2 mRNA levels in
LCC2 cells but not in LCC1 cells (Fig 1, A). By radioreceptor assay of
serum-free cell medium, tamoxifen (1 µM) treatment for 24
hours increased the secretion of (active) TGF-ß activity to 13.9
ng/106 cells per 24 hours, whereas, in LCC1 cell medium, this
bioactivity remained undetectable. To test the paracrine effects of
TGF-ß2 overexpression in the context of a multicellular
experimental system, we measured NK and LAK sensitivity of
51Cr-labeled LCC1 and LCC2 cells. NK and LAK cells are
potently suppressed by exogenous TGF-ß1 and TGF-ß2
(28). Consistent with these data and the overexpression of
TGF-ß2, LCC2 cells were statistically more resistant to NK
cell-mediated lysis (P = .0001) and LAK cell-mediated lysis
(P = .0002) than tamoxifen-sensitive LCC1 cells (Fig.
2).
|
|
TGF-ß Neutralizing Antibodies and Sensitivity to Tamoxifen In Vivo and In Vitro
To study the role of TGF-ß2 in tamoxifen resistance of LCC2
cells, we utilized the anti-TGF-ß 2G7 IgG2 antibody, which
neutralizes all three TGF-ß mammalian isoforms (19).
Established subcutaneous LCC2 tumors in nude mice bearing
60-day-release, 25-mg subcutaneously implanted tamoxifen pellets were
randomized to 100 µg/day of 2G7 or the 12H5 non-neutralizing
anti-TGF-ß control IgG2 by intraperitoneal injection. These
systemic doses of 2G7 result in detectable and sustained plasma levels
of an activity that blocks TGF-ß receptor binding when tested
ex vivo (26). In animals treated with 2G7, but not with the
12H5 control antibody, tumor growth was arrested (Fig. 3, A),
suggesting that tumor cell TGF-ß2 was mediating tamoxifen
resistance in this experimental system. A second experiment yielded
similar results. By light microscopy, both groups of LCC2 tumors were
poorly differentiated adenocarcinomas with no major histologic
differences between them. In the absence of tamoxifen treatment, 2G7
had no effect on LCC2 xenograft growth relative to 12H5-treated tumors,
suggesting that tumor cell TGF-ßs had no effect on basal LCC2 tumor
growth (not shown).
|
Previous studies (26-29) have shown that systemic administration of anti-TGF-ß antibodies can up-regulate NK function in nude mice that, in turn, exerts an antitumor effect. In addition, tamoxifen can induce and/or sensitize tumor cells to NK and LAK cells in vitro and in vivo (discussed below), thus implying that some of its antitumor effect is mediated by modulating this biologic response. These observations plus the high levels of NK activity present in athymic nude mice (30) and the relative resistance of LCC2 cells to NK activity (Fig. 2
), a known
cellular target for TGF-ß2, suggested to us a model in which
tamoxifen exerts some of its antitumor effect by up-regulating host NK
function in vivo. Overexpression of immunosuppressive
cytokines, like TGF-ß2, can then counteract this tamoxifen-mediated
host response and in part contribute to resistance to this
antiestrogen. To test this hypothesis, we repeated the experiment shown
in Fig. 3, A, in NK-deficient beige/nude mice.
Neutralization of TGF-ß2 with 2G7 did not restore tamoxifen
sensitivity to LCC2 tumors in NK-deficient mice (Fig. 3, B), suggesting
that, in this resistance model, NK host function is critical for the
antitumor effect of tamoxifen in vivo.
To further support the hypothesis that tumor host mechanisms were
involved in the restoration of tamoxifen sensitivity by neutralizing
antibodies, we examined the effect of 2G7 in tamoxifen-treated LCC2
cells in vitro. In this cell-autonomous experimental system,
LCC2 colony formation in the presence of 1 or 10 µg/mL 2G7 antibody
was similar in the presence or absence of 1 µM tamoxifen
(Fig. 4). Tamoxifen stimulated LCC2 colony formation
approximately 25% above ethanol (solvent for tamoxifen, used as
control) in the presence of 1 or 10 µg/mL of the 12H5 control
monoclonal antibody.
|
Antisense Oligodeoxynucleotide-Mediated Inhibition of Tumor Cell TGF-ß2 Expression and Enhancement of Tamoxifen-Induced NK Toxicity Against LCC2 Cells In Vitro
Several published data [see(31)] indicate that
tamoxifen
can stimulate immune effector cellular mechanisms in the tumor host as
well as sensitize tumor cell targets to cytotoxicity independent of
ER expression. Therefore, we first examined in a multicellular
experimental system whether this sensitization to NK action was
different in the tamoxifen-sensitive LCC1 versus tamoxifen-resistant
LCC2 cells. Consistent with their phenotype in vivo,
coincubation of NK effector cells with target cells with tamoxifen
in vitro resulted in sensitization of the LCC1 cells
(P = .0001). On the contrary, addition of tamoxifen reduced
the lower level of NK-mediated 51Cr release from LCC2
cells even further (P = .001; Fig. 5).
|
To test the contribution of TGF-ß2 to NK resistance in the presence of tamoxifen, we used antisense phosphorothioate oligodeoxynucleotides (27). Treatment of LCC2 cells with 3 µM antisense TGF-ß2 oligodeoxynucleotides for 48 hours abrogated the secretion of TGF-ß2, compared with nonsense oligodeoxynucleotides as measured by immunoblot of concentrated serum-free conditioned medium (Fig. 6).
Proliferation of LCC2 cells was not affected
during the 48-hour incubation with antisense oligonucleotides
compared with nonsense oligonucleotide-treated cultures. We then
examined the effect of antisense oligonucleotide-mediated inhibition
of TGF-ß2 expression in LCC2 cells on NK sensitivity in the
absence and presence of tamoxifen. Treatment of LCC2 cells with
TGF-ß2 antisense oligonucleotides enhanced their sensitivity to
NK effector cells in the presence of tamoxifen at all
effector-to-target ratios tested, but not in the absence of tamoxifen
(Fig. 7), suggesting that, in these
tamoxifen-resistant cells, TGF-ß2 mediates, in part, this relative
resistance to NK activity and that down-regulation of TGF-ß2
expression can unmask the NK-sensitizing effect of tamoxifen.
|
|
Antitumor Effect of Tamoxifen in Nude Mice Versus Beige/Nude Mice Against Tamoxifen-Sensitive LCC1 Xenografts
To test whether both the growth-inhibitory effect and the
sensitizing effect of tamoxifen on tumor targets to NK effector cells
were related phenomena, we concurrently examined the inhibitory effect
of tamoxifen on LCC1 tumors in nude mice, which exhibit elevated NK
cell function (30), as well as in beige/nude mice. In these
animals, the beige mutation markedly reduces NK function
(30). Tamoxifen pellets were implanted 2 days after
subcutaneous tumor cell inoculation. Within 4 weeks, 100% of LCC1
tumors formed in both nude and beige/nude mice without the need for
estrogen supplementation and in the absence of tamoxifen treatment
(Table 1). Four weeks after tumor inoculation, there
were no detectable LCC1 xenografts in tamoxifen-treated nude mice.
Conversely, five of eight NK-deficient beige/nude animals treated with
tamoxifen exhibited subcutaneous tumors that measured more than 4 mm in
diameter (>100 mm3). The formation of LCC1 tumors
suggested that, in this tumor model, host NK function was, in part,
mediating the growth-inhibitory effect of tamoxifen (Table 1). At 8
weeks, there were still no detectable LCC1 tumors in nude mice treated
with tamoxifen.
|
|
DISCUSSION |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
We have studied the role of TGF-ßs in the tamoxifen-resistant phenotype of LCC2 human breast cancer cells. Hormone-dependent breast carcinoma cells follow a predictable pattern from antiestrogen sensitivity to the acquisition of antiestrogen resistance. A number of molecular and cellular mechanisms are involved in the emergence of this phenotype, including the (infrequent) loss of ER expression, the selection of ER mutants with altered transcriptional responses, alterations in the intracellular pharmacology and/or binding of tamoxifen to breast cancer cells, and other non-ER-related mechanisms (see below). Several reviews (32,33) cover discussions on these mechanisms in more detail. The experimental models employing LCC1 and LCC2 cells were used in this study. These cells maintain high ER levels and up-regulate PgR levels in response to exogenous estradiol (17,18), thus providing a good system to study mechanisms of tamoxifen resistance that do not involve ER loss. Compared with other tamoxifen-resistant tumor models, LCC2 tumor growth is not stimulated by tamoxifen (18).
LCC2 tamoxifen-resistant cells markedly overexpress TGF-ß2 compared
with LCC1 tamoxifen-sensitive cells. Tamoxifen resistance in LCC2
tumors was abrogated by the coadministration of the antiestrogen with
neutralizing TGF-ß antibodies when tested in nude mice. This
response was not seen in beige/nude mice that lack NK activity.
Furthermore, tamoxifen sensitized the tamoxifen-sensitive LCC1 cells,
but not the resistant LCC2 cells, to NK effector cells in
vitro. Antisense oligonucleotide-mediated inhibition of TGF-ß2
secretion by LCC2 cells restored the ability of tamoxifen to sensitize
LCC2 cells to NK cell-induced cytolysis. Finally, tamoxifen exhibited a
greater antitumor effect against LCC1 tumor cells in NK-plus nude mice
compared with that seen in NK-deficient beige/nude mice. Taken
together, these data suggest that overexpression of TGF-ß2 by
breast tumor cells can mediate tamoxifen resistance in vivo.
Our results also suggest that host NK function is involved in the
antitumor effect of the antiestrogen tamoxifen and that overexpression
of potently immunosuppressive cytokines, like TGF-ß2, can
counteract this antiestrogenic effect in the host and thus contribute
to tamoxifen resistance. Of note, however, tamoxifen reduced the low
level of NK cell-mediated toxicity against LCC2 cells (Fig. 5). This
particular result may reflect the tamoxifen-mediated increase in
TGF-ß2 mRNA and the activation of secreted TGF-ß activity in
LCC2 cells (Fig. 1), which in turn can block NK cell function.
Our results would seem at odds with the dogma that proposes antiestrogen-induced up-regulation of autocrine TGF-ßs in breast carcinoma cells and tumor tissues as a mediator of tamoxifen's antitumor action (8,9), and that loss of this autocrine pathway can contribute to tamoxifen resistance. The LCC1/LCC2 model we used argues against this hypothesis in that tamoxifen-sensitive LCC1 cells bind TGF-ß1 poorly and do not respond to exogenous TGF-ß1. On the other hand, the LCC2 resistant cells express all three TGF-ß receptors in high levels (data not shown) and exhibit responses to exogenous TGF-ß1 in culture (34). Furthermore, a more recent study by Koli et al. (10) also challenges this dogma, in that transfection of a suppressible dominant-negative truncated type II TGF-ß receptor into antiestrogen- and TGF-ß-sensitive MCF-7 cells abrogated TGF-ß responses but not tamoxifen-mediated growth inhibition.
A rise in either the plasma levels of TGF-ß2 (35) or the tumor levels of TGF-ß2 mRNA (36) has been observed in patients with metastatic breast tumors receiving tamoxifen therapy. In one of these studies, TGF-ß1 and TGF-ß3 expression in tumors was not altered by tamoxifen therapy (36), suggesting a TGF-ß isoform-specific effect of tamoxifen. It was proposed in these studies that up-regulation of TGF-ß2 expression is a marker of clinical response to tamoxifen. However, the great majority of tamoxifen-sensitive mammary tumors will progress to a tamoxifen-resistant state, suggesting the possibility of a subsequent temporal association between breast tumor TGF-ß2 overexpression and tamoxifen resistance. This is consistent with our findings in the LCC2 cells. TGF-ß1 overexpression has also been temporally associated with antiestrogen resistance in human breast cancer cell lines (11,12) and mammary tumor tissues (13,15).
The greater inhibition of tamoxifen-sensitive LCC1 tumors in nude mice, which have elevated NK and normal LAK activities, compared with that seen in beige/nude NK-deficient mice, suggests that the antiproliferative effect of tamoxifen and the sensitization to host NK (and possibly LAK) function are related phenomena. Natural cytotoxicity, mediated by NK cells, is believed to play an important role in host antitumor surveillance mechanisms. NK cells are predominantly large granular lymphocytes, the majority of which express CD16 and CD56 cell surface antigens and represent overall 5%-8% of PBLs (37). NK cells can mediate cytolysis in the absence of major histocompatibility complex class 1 and class 2 antigen expression in target cells (37,38). In patients with breast cancer and in experimental systems, tamoxifen has been shown to enhance NK function and/or increase the sensitivity of tumor cell targets in an ER-independent fashion (31,39-42). However, in one report by Gottardis et al. (43), prolonged treatment with tamoxifen inhibited NK function in nude mice and stimulated growth of tamoxifen-resistant MCF-7 tumors. Nonetheless, NK activity is inversely related to the clinical stage of disease and/or the presence of axillary lymph node metastases in patients with breast cancer (44-46), implicating host natural cytotoxicity in the control of cancer progression. Accumulation of NK cells in tumors was temporally associated with a clinical spontaneous remission in a patient with metastatic breast carcinoma (47). Other reports (48-50) indicate that pharmacologically achievable concentrations of tamoxifen enhance immune cytolysis of tumor targets mediated by LAK cells and cytotoxic T cells. Overall and except for the observation by Gottardis et al. (43), these reports and our results are consistent with the hypothesis that host immune function may play a role in the antitumor effect of triphenylethylene antiestrogens, independent of the ER status of the tumor cell target. These reports also suggest the possibility of synergistic antitumor effects of tamoxifen with anti-TGF-ß antibodies in other ER-negative neoplasms. This possibility remains to be tested.
In vitro, anti-TGF-ß antibodies failed to reverse tamoxifen resistance in LCC2 cells. This result suggests that the observed reversal in vivo may not involve ER signaling in LCC2 cells. Other non-ER mechanisms of antiestrogen-mediated tumor inhibition that involve paracrine mechanisms have been reported. For example, tamoxifen and pure ER antagonists inhibit endothelial cell proliferation within mammary tumors and promote the apoptosis of angiogenesis-dependent breast tumors (51,52). Moreover, MCF-7 tumors transfected with fibroblast growth factor (FGF)-4 and FGF-1 develop hematogenous metastases in vivo and are unresponsive to tamoxifen (53-55), suggesting a role for angiogenic factors in ER-independent resistance to tamoxifen. Of note, TGF-ßs are potent inductors of angiogenesis (2,56), thus leading to a possible causal association between overexpression of TGF-ßs, high intratumoral microvessel density, and a tamoxifen-resistant phenotype. Whether enhanced TGF-ß2-mediated angiogenesis is involved in the antiestrogen resistance of TGF-ß2-overexpressing LCC2 tumors will require further experiments.
In summary, we have described a novel mechanism of antiestrogen resistance, as well as a potentially important immunologic component in tamoxifen response and acquired resistance in human breast carcinoma cells. It is interesting that the in vitro selection of LCC2 cells against tamoxifen produced a resistance mechanism that functions through an effect on cells or mechanisms (NK function) not involved in the initial selection. However, the greater effect of tamoxifen against LCC1 tumors in NK-competent versus NK-deficient mice, as well as the TGF-ß2 antisense oligonucleotide-mediated sensitization of LCC2 cells to NK effectors in vitro in the presence of tamoxifen, strongly argues in favor of this immunologic component. In addition, the published effects of tamoxifen therapy on the expression or content of TGF-ßs in patients' tumors and tumor models (13,15,36) suggest that tamoxifen-induced overexpression of TGF-ßs can also occur in vivo. Several practical implications could be derived from these data. First, the effects of tamoxifen on tumor host (endothelial, immune) cells will be missed in cell-autonomous in vitro screens and potentially lead to false-negative preclinical models. Second, ER-positive tumors that overexpress TGF-ßs may be clinically unresponsive de novo to tamoxifen and/or escape antiestrogens early and would, therefore, be candidates to alternative molecular therapies aimed at down-regulating the overexpressed immunosuppressive (or angiogenic) cytokines. Future prospective clinicoepidemiologic studies in patients should clarify the practical implications of these data and address how predictably the overexpression of TGF-ßs by ER-positive breast tumors may determine clinical resistance to tamoxifen and other antiestrogens in general.
|
NOTES |
|---|
Supported by Public Health Service grants R01CA58022 (R. Clarke), R01CA62212 (C. L. Arteaga), and CA65485 from the National Cancer Institute, National Institutes of Health, Department of Health and Human Services; Merit Review and Clinical Investigator grants from the Department of Veteran Affairs (C. L. Arteaga); and the Susan G. Komen Foundation (C. L. Arteaga).
We thank Dr. John Hanfelt (Lombardi Cancer Center) for his helpful comments and consultation.
|
REFERENCES |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
1 Roberts AB, Sporn MB. The transforming growth factor-ßs. In: Roberts AB, Sporn MB, editors. Handbook of experimental pharmacology. Heidelberg (Germany): Springer-Verlag; 1990. p. 419-72.
2 Moses HL. The biological action of transforming growth factor ß. In: Sara V, Hall K, Low H, editors. Growth factors from genes to clinical application. New York (NY): Raven Press; 1990. p. 141-55.
3 Shull MM, Ormsby I, Kier AB, Pawlowski S, Diebold RJ, Yin M, et al. Targeted disruption of the mouse transforming growth factor-ß1 gene results in multifocal inflammatory disease. Nature 1992;359:693-9.[CrossRef][Medline]cancerlit;93063258
4 Sanford LP, Ormsby I, Gittenberger-de Groot AC, Sariola H, Friedman F, Boivin GP, et al. TGFß2 knockout mice have multiple developmental defects that are non-overlapping with other TGFß knockout phenotypes. Development 1997; 124:2659-70.[Abstract]cancerlit;97359979
5 Proetzel G, Pawlowski SA, Wiles MV, Yin MY, Boivin GP, Howles PN, et al. Transforming growth factor-ß3 is required for secondary palate fusion. Nat Genet 1995;11:409-14.[CrossRef][ISI][Medline]cancerlit;96083588
6 Koli KM, Arteaga CL. Complex role of tumor cell TGF-ßs on breast carcinoma progression. J Mammary Gland Biol Neopl 1996;1:373-80.[CrossRef][Medline]
7
Pierce DF, Johnson MD, Matsui Y, Robinson SD, Gold LI,
Purchio AF, et al. Inhibition of mammary duct development but not alveolar outgrowth during
pregnancy in transgenic mice expressing active TGF-ß1. Genes Dev 1993;7:2308-17.
8 Knabbe C, Lippman ME, Wakefield LM, Flanders KC, Kasid A, Derynck R, et al. Evidence that transforming growth factor-ß is a hormonally regulated negative growth factor in human breast cancer cells. Cell 1987;48:417-28.[CrossRef][ISI][Medline]cancerlit;87102891
9
Butta A, MacLennan K, Flanders KC, Sacks NP, Smith I,
McKinna A, et al. Induction of transforming growth factor ß1 in human breast
cancer in vivo following tamoxifen treatment. Cancer Res 1992;52:4261-4.
10
Koli KM, Ramsey TT, Ko Y, Dugger TC, Brattain MG, Arteaga
CL. Blockade of transforming growth factor-ß signaling does not abrogate
antiestrogen-induced growth inhibition of human breast carcinoma cells. J Biol Chem 1997;272:8296-302.
11 Daly RJ, King RJB, Darbre PD. Interaction of growth factors during progression toward steroid independence in T-47-D human breast cancer cells. J Cell Biochem 1990;43:199-211.[CrossRef][ISI][Medline]cancerlit;90338162
12
Hermann ME, Katzenellenbogen B. Alterations in transforming
growth factor-ß production and cell responsiveness during the progression of MCF-7 human
breast cancer cells to estrogen-autonomous growth. Cancer Res 1990;54:5867-74.
13 Thompson AM, Kerr DJ, Steel CM. Transforming growth factor ß1 is implicated in the failure of tamoxifen therapy in human breast cancer. Br J Cancer 1991;63:609-14.[ISI][Medline]cancerlit;91214827
14 Arteaga CL, Carty-Dugger T, Moses HL, Hurd SD, Pietenpol JA. Transforming growth factor ß1 can induce estrogen-independent tumorigenicity of human breast cancer cells in athymic mice. Cell Growth Differ 1993;4:193-201.[Abstract]cancerlit;93222070
15 Baillie R, Coombes RC, Smith J. Multiple forms of TGF-ß1 in breast tissues: a biologically active form of the small latent complex of TGF-ß1. Eur J Cancer 1996;32A:1566-73.[CrossRef]cancerlit;97067771
16
Gorsch SM, Memoli VA, Stukel TA, Gold LI, Arrick BA.
Immunohistochemical staining for transforming growth factor ß1 associates with
disease progression in human breast cancer. Cancer Res 1992;52:6949-52.
17
Brunner N, Boulay V, Fojo A, Freter CE, Lippman ME, Clarke
R. Acquisition of hormone-independent growth in MCF-7 cells is accompanied by increased
expression of estrogen-regulated genes but without detectable DNA amplifications. Cancer Res 1993;53:283-90.
18 Brunner N, Frandsen TL, Holst-Hansen C, Bei M, Thompson EW, Wakeling AE, et al. MCF-7/LCC2: a 4-hydroxytamoxifen resistant human breast cancer variant that retains sensitivity to the steroidal antiestrogen ICI 182,780. Cancer Res 1993; 192;3:3229-32.
19 Lucas C, Bald LN, Fendly BM, Mora-Worms M, Figari IS, Patzer EJ, et al. The autocrine production of transforming growth factor ß1 during lymphocyte activation: A study with a monoclonal antibody-based ELISA. J Immunol 1990;145:1415-22.[Abstract]
20 Sambrook J, Fritsch EF, Maniatis TE. Molecular cloning: a laboratory manual. Cold Spring Harbor (NY): Cold Spring Harbor Laboratory; 1989.
21 Derynck R, Jarrett JA, Chen EY, Eaton DH, Bell JR, Assoian RK, et al. Human transforming growth factor-beta complementary DNA sequence and expression in normal and transformed cells. Nature 1985;316:701-5.[CrossRef][Medline]cancerlit;85296301
22 Madisen L, Webb NR, Rose TM, Marquardt H, Ikeda T, Twardzik D, et al. Transforming growth factor-ß2: cDNA cloning and sequence analysis. DNA 1988;7:1-8.[ISI][Medline]cancerlit;88166349
23 Danielson PE, Forss-Petter S, Brow MA, Calavetta L, Douglass J, Milner RJ, et al. p1B15: a cDNA clone of the rat mRNA encoding cyclophilin. DNA 1988;7:261-7.[ISI][Medline]
24 Arteaga CL, Coffey RJ Jr, Dugger TC, McCutchen CM, Moses HL, Lyons RM. Growth stimulation of human breast cancer cells with anti-transforming growth factor ß antibodies: evidence for negative autocrine growth regulation by transforming growth factor ß. Cell Growth Differ 1990;1:367-74.[Abstract]cancerlit;91669489
25 Lyons RM, Miller DA, Graycar JL, Moses HL, Derynck R. Differential binding of transforming growth factor-ß1, -ß2, and -ß3 by fibroblasts and epithelial cells measured by affinity cross-linking of cell surface receptors. Mol Endocrinol 1991;2:1887-96.
26 Arteaga CL, Hurd SD, Winnier AR, Johnson MD, Fendly BM, Forbes JT. Anti-transforming growth factor (TGF)-ß antibodies inhibit breast cancer cell tumorigenicity and increase mouse spleen natural killer cell activity. Implications for a possible role of tumor cell/host TGF-ß interactions in human breast cancer progression. J Clin Invest 1993;92:2569-76.cancerlit;94075600
27 Jachimczak P, Bogdahn U, Schneider J, Behl C, Meixensberger J, Apfel R, et al. The effect of transforming growth factor-ß2-specific phosphorothioate-anti-sense oligodeoxynucleotides in reversing cellular immunosuppression in malignant glioma. J Neurosurg 1993;78:944-51.[CrossRef][ISI][Medline]cancerlit;93253465
28 Ruscetti F, Varesio L, Ochoa A, Ortaldo J. Pleiotropic effects of transforming growth factor-ß on cells of the immune system. Ann N Y Acad Sci 1993;685:488-500.[Abstract]cancerlit;93370894
29
Wallick SC, Figari IS, Morris RE, Levinson AD, Palladino MA.
Immunoregulatory role of transforming growth factor ß (TGF-ß) in development of
killer cells: comparison of active and latent TGF-ß1. J Exp Med 1990;172:1777-84.
30 Clarke R. Human breast cancer cell line xenografts as models of breast cancer. The immunobiologies of recipient mice and the characteristics of several tumorigenic cell lines. Breast Cancer Res Treat 1996;39:69-86.[CrossRef][ISI][Medline]cancerlit;96327850
31 Baral E, Nagy E, Berczi I. The effect of tamoxifen on the immune response. In: Kellen JE, editor. Tamoxifen: beyond the antiestrogen. Boston (MA): Birkhauser; 1996. p. 137-78.
32 Wiebe VJ, Osborne CK, Fuqua SA, DeGregorio MW. Tamoxifen resistance in breast cancer. Crit Rev Oncol Hematol 1993;14:173-88.[ISI][Medline]cancerlit;94000415
33 Clarke R, Brunner N. Acquired estrogen independence and antiestrogen resistance in breast cancer. Estrogen receptor driven phenotypes? Trends Endocrinol Metab 1996;7:25-35.
34 Arteaga CL, Dugger TC, Clarke R, Koli KM. Blockade of tumor cell transforming growth factor (TGF)-ßs restores tamoxifen (tam) sensitivity to tam-resistant human breast cancer cells. Breast Cancer Res Treat 1996;41:220A.
35
Kopp A, Schmahl M, Knabbe C. Transforming growth factor
ß2 (TGF-ß2) levels in plasma of patients with metastatic breast cancer treated with
tamoxifen. Cancer Res 1995;55:4512-5.
36 MacCallum J, Keen JC, Bartlett JMS, Thompson AM, Dixon JM, Miller WR. Changes in expression of transforming growth factor ß mRNA isoforms in patients undergoing tamoxifen therapy. Br J Cancer 1996;74:474-8.[ISI][Medline]cancerlit;96316872
37 Trinchieri G. Biology of natural killer cells. Adv Immunol 1989;47:187-376.[ISI][Medline]cancerlit;90052881
38
Liao N, Bix M, Zijistra M, Jaenisch R, Raulet DH. MHC class I
deficiency: susceptibility to natural killer (NK) cells and impaired NK activity. Science 1991;253:199-202.
39 Screpanti I, Santoni A, Gulino A, Herberman RB, Frati L. Estrogen and antiestrogen modulation of the levels of mouse natural killer activity and large granular lymphocytes. Cell Immunol 1987;106:191-202.[CrossRef][ISI][Medline]
40 Mandeville R, Ghali SS, Chausseau JP. In vitro stimulation of human NK activity by an estrogen antagonist (tamoxifen) [letter]. Eur J Cancer Clin Oncol 1984;20:983-5.[CrossRef][ISI][Medline]cancerlit;84261668
41 Berry J, Green BJ, Matheson DS. Modulation of natural killer cell activity by tamoxifen in stage I post-menopausal breast cancer. Eur J Cancer Clin Oncol 1987;23:517-20.[CrossRef][ISI][Medline]cancerlit;88004602
42 Baral E, Nagy E, Berczi I. Modulation of natural killer cell-mediated cytotoxicity by tamoxifen and estradiol. Cancer 1995;75:591-9.[CrossRef][ISI][Medline]cancerlit;95112250
43
Gottardis MM, Wagner RJ, Borden EC, Jordan VC. Differential
ability of antiestrogens to stimulate breast cancer cell (MCF-7) growth in vivo and in vitro. Cancer Res 1989;49:4765-9.
44 An T, Sood U, Pietruk T, Cummings G, Hashimoto K, Crissman JD. In situ quantitation of inflammatory mononuclear cells in ductal infiltrating breast carcinoma. Relation to prognostic parameters.Am J Pathol 1987;128:52-60.[Abstract]cancerlit;87267986
45 Akimoto M, Ishii H, Nakajima Y, Iwasaki H. Assessment of host immune response in breast cancer patients. Cancer Detect Prev 1987;9:311-7.
46 Horst HA, Horny HP. Characterization and frequency distribution of lymphoreticular infiltrates in axillary lymph node metastases of invasive ductal carcinoma of the breast. Cancer 1987;60:3001-7.[CrossRef][ISI][Medline]cancerlit;88052502
47 Maiche AG, Jekunen A, Rissanen P, Virkkunen P, Halavaara J, Turunen JP. Sudden tumour regression with enhanced natural killer cell accumulation in a patient with stage IV breast cancer. Eur J Cancer 1994;30A:1642-6.[CrossRef]cancerlit;95134532
48 Baral E, Nagy E, Berczi I. Modulation of lymphokine-activated killer cell-mediated cytotoxicity by estradiol and tamoxifen. Int J Cancer 1996;66:214-8.[CrossRef][ISI][Medline]cancerlit;96179492
49 Baral E, Nagy E, Berczi I. Target cells are sensitized for cytotoxic T-lymphocyte-mediated destruction by estradiol and tamoxifen. Int J Cancer 1994;58:64-8.[ISI][Medline]cancerlit;94284044
50 Albertini MR, Gibson DF, Robinson SP, Howard SP, Tans KJ, Lindstrom MJ, et al. Influence of estradiol and tamoxifen on susceptibility of human breast cancer cell lines to lysis by lymphokine-activated killer cells. J Immunother 1992;11:30-9.cancerlit;92135176
51 Gagliardi A, Collins DC. Inhibition of angiogenesis by antiestrogens. Cancer 1994;53:533-5.
52
Haran EF, Maretzek AF, Goldberg I, Horowitz A, Degani H.
Tamoxifen enhances cell death in implanted MCF7 breast cancer by inhibiting endothelium
growth. Cancer Res 1994;54:5511-4.
53
McLeskey SW, Kurebayashi J, Honig SF, Zwiebel J, Lippman
ME, Dickson RB, et al. Fibroblast growth factor 4 transfection of MCF-7 cells produces cell
lines that are tumorigenic and metastatic in ovariectomized or tamoxifen-treated athymic nude
mice. Cancer Res 1993;53:2168-77.
54 Zhang L, Kharbanda S, Chen D, Bullocks J, Miller DL, Ding IY, et al. MCF-7 breast carcinoma cells overexpressing FGF-1 form vascularized, metastatic tumors in ovariectomized or tamoxifen-treated nude mice. Oncogene 1997;15:2093-108.[CrossRef][ISI][Medline]cancerlit;98031854
55
Zhang L, Kharbanda S, Hanfelt J, Kern FG. Both autocrine and
paracrine effects of transfected acidic fibroblast growth factor are involved in the
estrogen-independent and antiestrogen-resistant growth of MCF-7 breast cancer cells. Cancer Res 1998;58:352-61.
56
Roberts AB, Sporn MB, Assoian RK, Smith JM, Roche NS,
Wakefield LM, et al. Transforming growth factor type ß: rapid induction of fibrosis and
angiogenesis in vivo and stimulation of collagen formation in vitro. Proc
Natl Acad Sci U S A 1986;83:4167-71.
Manuscript received February 5, 1998; revised October 28, 1998; accepted November 2, 1998.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  B. A Teicher, J. M Yingling, and J. M McPherson TGF{beta} Blockade as Anticancer Therapy Am. Assoc. Cancer Res. Educ. Book, April 12, 2008; 2008(1): 71 - 81. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. E Burdette and T. K Woodruff Activin and estrogen crosstalk regulates transcription in human breast cancer cells Endocr. Relat. Cancer, September 1, 2007; 14(3): 679 - 689. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. P.V. Shekhar, S. Santner, K. A. Carolin, and L. Tait Direct Involvement of Breast Tumor Fibroblasts in the Modulation of Tamoxifen Sensitivity Am. J. Pathol., May 1, 2007; 170(5): 1546 - 1560. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. H. Al-Dhaheri and B. G. Rowan Protein Kinase A Exhibits Selective Modulation of Estradiol-Dependent Transcription in Breast Cancer Cells that Is Associated with Decreased Ligand Binding, Altered Estrogen Receptor {alpha} Promoter Interaction, and Changes in Receptor Phosphorylation Mol. Endocrinol., February 1, 2007; 21(2): 439 - 456. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. C Fleisch, C. A Maxwell, and M.-H. Barcellos-Hoff The pleiotropic roles of transforming growth factor beta in homeostasis and carcinogenesis of endocrine organs. Endocr. Relat. Cancer, June 1, 2006; 13(2): 379 - 400. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. I. Vanzulli, R. Soldati, R. Meiss, L. Colombo, A. A. Molinolo, and C. Lanari Estrogen or antiprogestin treatment induces complete regression of pulmonary and axillary metastases in an experimental model of breast cancer progression Carcinogenesis, June 1, 2005; 26(6): 1055 - 1063. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Troup, C. Njue, E. V. Kliewer, M. Parisien, C. Roskelley, S. Chakravarti, P. J. Roughley, L. C. Murphy, and P. H. Watson Reduced Expression of the Small Leucine-rich Proteoglycans, Lumican, and Decorin Is Associated with Poor Outcome in Node-negative Invasive Breast Cancer Clin. Cancer Res., January 1, 2003; 9(1): 207 - 214. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Clarke, F. Leonessa, J. N. Welch, and T. C. Skaar Cellular and Molecular Pharmacology of Antiestrogen Action and Resistance Pharmacol. Rev., March 1, 2001; 53(1): 25 - 72. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Lanari, I. Lüthy, C. A. Lamb, V. Fabris, E. Pagano, L. A. Helguero, N. Sanjuan, S. Merani, and A. A. Molinolo Five Novel Hormone-responsive Cell Lines Derived from Murine Mammary Ductal Carcinomas: In Vivo and in Vitro Effects of Estrogens and Progestins Cancer Res., January 1, 2001; 61(1): 293 - 302. [Abstract] [Full Text]
|
|
|
|
 |
|