Sage Journals: Discover world-class research

Abstract

The insulin-like growth factor I receptor (IGF1-R) system has long been implicated in cancer and is a promising target for tumor therapy. Besides in vitro screening assays, the discovery of specific inhibitors against IGF-1R requires relevant cellular models, ideally applicable to both in vitro and in vivo studies. With this aim in mind, the authors generated an inducible cell line using the tetracycline-responsive gene expression system to mimic the effects of therapeutic inhibition of the IGF-1R both in vitro and on established tumors in vivo. Inducible overexpression of IGF-1R in murine embryonic fibroblasts was achieved and resulted in the transformation of the cells as verified by their ability to grow in soft agar and in nude mice. Continuous repression of exogenous IGF-1R expression completely prevented outgrowth of the tumors. Furthermore, induced repression of IGF-1R expression in established tumors resulted in regression of the tumors. Interestingly, however, IGF-1R–independent relapse of tumor growth was observed upon prolonged IGF-1R repression. The IGF-1R cell line generated using this approach was successfully employed to test reference small-molecule inhibitors in vitro and an IGF-1R–specific inhibitory antibody, EM164, in vivo. Besides efficacy as a read-out, phospho-AKT could be identified as a pharmacodynamic biomarker, establishing this cell line as a valuable tool for the preclinical development of IGF-1R inhibitors.

Keywords

IGF-1R tet-off transformation EM164 preclinical drug discovery

Deregulation of the insulin-like growth factor I receptor (IGF-1R) has long been linked to tumorigenesis. Early experiments demonstrated that overexpression of IGF-1R in mouse embryo fibroblasts (MEFs) resulted in cell transformation.¹ Transformation occurred in a dose-dependent manner, as shown in cells deleted for their endogenous IGF-1R copy.² Activation of the IGF-1R mediates cell proliferation and survival but also transformation, angiogenesis, and metastasis.^3,4 The receptor is activated by its ligands IGF-1 and -2, which result in receptor autophosphorylation and the stimulation of downstream signaling cascades including the IRS-1/PI-3K/PKB/S6K and Grb2/Sos/Ras/MAPK pathways. Interference with its activity by either gene disruption,⁵ antisense oligonucleotides,⁶ dominant negative mutants,^7,8 neutralizing antibodies,^9,10 or small-molecule inhibitors¹¹ was found to interfere with cell growth and proliferation. Thus, the IGF-1R is a promising drug target for the development of novel anticancer drugs.¹²

The discovery of new IGF-1R–targeting compounds requires suitable model systems, both for initial screens and the further characterization of potential lead compounds. In vitro enzyme assays, which are fast, reliable, and comparably easy to set up, may serve for the initial hit generation. However, to characterize compound selectivity and potency and to allow for the measurement of efficacy and identification of potential biomarkers, cellular and in vivo models are required that are much more complex to establish. Cellular models for transforming receptor tyrosine kinases (RTKs) may readily be generated via stable transfection of the corresponding cDNA into untransformed cell lines (e.g., mouse fibroblasts) or using tumor cell lines that endogenously express the gene. However, in either case, it is difficult to prove that the given RTK is indeed the single transforming principle. Furthermore, when used for animal tumor models, it is usually impossible to distinguish between a requirement of the RTK for original cellular transformation or tumor maintenance. Stable transformation of a given gene may alter the general gene expression pattern of the cell leading to the expression of other oncogenes. The situation is even more complex in cells that express the gene of interest endogenously.

We therefore chose to establish an inducible cell system based on the tet-repressor that allows for reversible expression of a given gene.¹³ This approach has been successfully used for the ErbB2 receptor.^14,15 Thus, here we describe the generation of an inducible cellular model system for the IGF-1R receptor using an MEF cell line harboring a plasmid encoding the tet-repressor. The resulting cell line was tested for IGF-1 responsiveness in vitro and for transformation via anchorage-independent growth in soft agar and tumor growth in vivo. Finally, the established cell line was used to characterize the effect of known and putative IGF-1R inhibitors in vitro as well as in vivo and the identification of phospho-AKT as a potential downstream biomarker.

Materials and Methods

Cell Lines and Antibodies

The cell line expressing the anti-myc–specific monoclonal antibody 9E10 was purchased from Developmental Studies Hybridoma Bank (Iowa City, IA). The anti-IGF-1R β (C-20; sc-713), PY99 (sc-7020B), and anti-AKT1 (C-20) antibodies were obtained from Santa Cruz Biotechnology (Santa Cruz, CA). The antiphospho-AKT antibody (9275) was from Cell Signaling Technologies (New England Biolabs, Hitchin, Herts, UK). MEFTOFF (Clontech, Heidelberg, Germany) were kept in DMEM (61965-026; Life Technologies, Darmstadt, Germany) supplemented with 10% FCS/penicillin/streptomycin at 37 °C/10% CO₂.

Generation of an MEF Cell Line Inducibly Expressing the IGF-1R

The cDNA encoding wild-type human IGF-1R, supplemented with a C-terminal 2myc6His tag, was inserted into pTRE (Clontech), and that construct was co-transfected with the plasmid pSV Hygro into the tet-off MEF (MEFTOFF) cell line. Successfully transfected cells were selected using 200 µg/mL hygromycin.

FACS/Western/Phospho-ELISA Analysis

For the fluorescence-activated cell sorting (FACS)/Western/phospho-enzyme-linked immunoadsorbent assay (ELISA) measurement, cells kept in the presence or absence of doxycycline for >48 h were plated at 750 000 cells/well into six 10-cm dishes each, keeping the respective doxycycline status. Twenty-four hours after plating, five plates were washed twice with phosphate-buffered saline (PBS), and the medium was replaced with DMEM containing 0.1% FCS only. Another 24 h later, 0, 5, 50, 100, and 200 ng IGF-1 (I3769, Sigma, Hamburg, Germany) was added, and the cells were incubated for an additional 24 h. On the following day, the medium supernatant was removed, the plates were washed with PBS, and the cells trypsinized and combined with the medium supernatant. The cells were collected by centrifugation and resuspended in 500 mL PBS. Two hundred fifty microliters of the cell suspension was fixed with methanol, rehydrated in PBS, stained with 25 µg/mL propidium iodide in PBS containing 100 µg/mL RNAse H, and analysed in a FACSCalibur (Becton Dickenson, Heidelberg, Germany). The remainder of the cells was pelleted and lysed in buffer A (20 mM TRIS pH 8/135 mM NaCl/1% NP-40/10% glycerol/1.5 mM sodium vanadate) and 1 tablet complete protease inhibitors (Roche, Mannheim, Germany). The protein content was determined using the Bradford method,¹⁶ and the lysates were used for ELISA and Western blot analysis. For the ELISA, 100 µg protein was applied into 96-well plates precoated with anti-myc (9E10) antibody. PY-99 antibody was used to detect phosphorylated IGF-1R. For the Western blots, 30 µg of protein was blotted onto nitrocellulose membranes and probed with anti-myc (9E10), anti–IGF-1R, anti-AKT1, and a phospho-AKT antibody (all at 1:200).

Soft Agar and Immunofluorescence Assays

A soft agar assay as well as immunofluorescence detection of IGF-1R were performed using cells that had been grown for >48 h in the presence or absence of doxycycline. For soft agar assays, the cells were embedded in 0.4% agarose containing DMEM, FCS, keeping the respective doxycycline status, and plated onto a layer prepared of 0.8% agarose. For immunofluorescence analysis, the cells were plated onto cover slips, fixed in 4% paraformaldehyde, permeabilized, and the IGF-1R was detected using the 9E10 antibody followed by an FITC-labeled secondary anti-mouse antibody. DAPI was used as a nuclear counterstain.

Cellular Phosphorylation Assays

Cells kept for >48 h in the presence or absence of doxycycline were seeded in 48-well plates. After serum starvation for 18 h, the serially diluted compounds (in DMSO, final concentration of DMSO: 1%) were added for 90 min followed by IGF-1 for 10 min. The cells were then lysed and the ELISA was performed as described above.

Animal Experiments

All animal experiments were performed according to German Animal License Regulations (Tierschutzgesetz) identical to UKCCCR Guidelines for the welfare of animals in experimental neoplasia. NMRI nude mice were obtained from Harlan Winkelmann GmbH, Germany.

The injection of 4 × 10⁷ cells in 0.2 mL aliquots into the subcutaneous space of the left flank was performed using a 29G needle syringe. Volumes of primary tumors were taken macroscopically by calipering and multiplying the distances of all three dimensions. Therapy or doxycycline treatment (2 mg/mL in drinking water containing 1% sucrose) started after the primary tumor had reached a volume of about 200 mm³, or as indicated in the figure legend. During the course of the study, animal weights were measured three times per week. Animal behavior was monitored daily. The criteria to evaluate the antitumor activity of the IGF-1R antibody were tumor regressions:

Complete regression (CR) = regression below the limit of palpation (<63 mm)

Partial regression (PR) = regression greater than 50% reduction in tumor mass (CR was included in the PR)

For BrdU labeling of sections, the animals were injected intraperitoneally with 100 to 200 µL of a 10 mg/ml solution of BrdU in sterile 1X DPBS (BD Pharmingen, Heidelberg, Germany) 24 h before necropsy.

IGF-1R Antibody EM164

The murine anti–IGF-1R monoclonal antibody, EM164,^17,18 was prepared by Immunogen Inc. (Waltham, MA) and provided by Sanofi (94400 Vitry-sur-Seine, France). EM164 was formulated in PBS without Ca²⁺ and Mg²⁺ and administered intraperitoneally twice a week.

IGF-1R Protein Expression in MEF/IGF-1R Tumor Samples

Tumors were cut into small pieces and then grinded mechanically by bead beating in a Qiagen Mill Mixer MM301 in ice-cold lysis buffer (1% Triton X-100, 100 mM NaCl, 10 mM Tris-HCl [pH 7.5], 1 mM EDTA [pH 8], 1 mM EGTA [pH 8], 1 mM NaF, 20 mM Na₄P₂O₇, 1 mM Na₃VO₄, 10% glycerol, and one tablet protease inhibitors; Roche Diagnostics, Mannheim, Germany). After a 2-h incubation at 4 °C, the tumor lysates were centrifuged at 16 000 g for 15 min at 4 °C, and the supernatants were transferred to fresh tubes on ice. The total protein concentration was determined using a Bradford assay¹⁶ according to the manufacturer’s instructions (BIO-RAD protein assay). The lysates were stored at −80 °C until analysis. IGF-1R expression and AKT phosphorylation in tumor lysates were determined using a commercially available ELISA assay (DY391; R&D Systems, Lille CEDEX, France) according to the manufacturer’s instructions.

Immunohistochemistry

Seven-micrometer cryosections were prepared from snap-frozen tumors and fixed using 4% paraformaldehyde. Endogenous peroxidases were blocked using 3% H₂O₂ in methanol. Exogenous and total IGF-1R were detected using a biotinylated 9E10 antibody followed by a streptavidin-horseradish peroxidase (PO) conjugate or a rabbit anti–IGF-1R β antibody followed by a secondary PO-labeled anti-rabbit antibody, respectively. TUNEL and BrdU stainings were done according to manufacturers’ instructions (Roche and BD Pharmingen, respectively).

Results

Generation of a Tet-Repressable IGF-1R MEF Cell Line

The insulin-like growth factor receptor has been a promising cancer drug target for more than a decade. However, although small-molecule inhibitors as well as antibodies against IGF-1R are in clinical development, to date no therapy against IGF-1R has been approved.

The discovery of novel molecules requires relevant testing systems. So to generate a tool for the cellular and in vivo testing of IGF-1R antagonists, we planned to generate a mouse fibroblast cell line with an inducible IGF-1R allele. An MEF cell line expressing the Tet-Off transactivator protein was transfected with a vector carrying the full-length human IGF-1R cDNA under the control of a tet-sensitive minimal CMV promoter.¹³ After selection and clonal expansion, cell lines were tested for tet-repressable expression of the IGF-1R target.

Cell clone 28 showed robust expression of the receptor in the absence and complete suppression in the presence of the repressor doxycycline. Expression of the IGF-1R transgene is detectable 24 h after removal of the repressor (Suppl. Fig. 1A).

IGF-1R activation induces autophosphorylation of tyrosine residues on the C-terminal cytoplasmic tail of the kinase, which triggers the binding of signaling molecules that mediate, among others, proliferation and protection from apoptosis.^19,20

In the presence of 10% FCS, the growth behavior of the cells was independent of the presence of the IGF-1R. However, upon serum starvation and subsequent supplementation with increasing amounts of IGF-1, a stimulation of cellular proliferation and a concomitant reduction of apoptosis were observed, dependent on IGF-1R expression (Suppl. Fig. 1B, first and second panels). The addition of 100 or 200 ng of IGF-1 induced tyrosine-phosphorylation of the receptor tyrosine kinase, as detected in a phospho-specific ELISA (Suppl. Fig. 1B, third panel). Again, the presence of the repressor doxycycline completely abolished IGF-1R expression (Suppl. Fig. 1B, bottom panel).

The parental MEF cell line is nontumorigenic in nude mice, but the IGF-1R has been reported to transform mouse fibroblasts. As a test for anchorage-independent growth, a hallmark of cellular transformation, we first checked whether the expression of the IGF-1R enables the cell clone 28 to grow in soft agar. Because colonies were detectable, albeit few (data not shown), the cells were next implanted into nude mice and their growth monitored over time. Growth was very slow in the beginning, but exponential growth was observed after day 40 (Suppl. Fig. 1C).

However, despite these promising results, the slow growth of the cells in the mouse model and the resulting rather large tumor size differences precluded the use of this cell line in a robust animal model.

Generation of Subclones 28.1 and 28.3 with Improved Growth Characteristics

Picking cell clones that grow in soft agar is one way to isolate cells with better growth characteristics in a setting selective for transformed cells. Subclones from clone 28 were expanded after isolation from soft agar, and two clones, 28.1 and 28.3, were characterized as described in the previous section.

Western blot analysis demonstrated that both subclones retained their doxycycline responsiveness, with similar kinetics as clone 28 ( Fig. 1 ). In the presence of 10% FCS, IGF-1R expression did not affect proliferation characteristics of the cell lines, although a slight decrease of apoptotic cells was apparent ( Fig. 2A , B ). However, serum starvation induced massive apoptosis and some reduction of the number of proliferating cells. Addition of IGF-1 to the cells kept in the absence of doxycycline fully reversed this phenotype. Higher IGF-1 concentrations (100 and 200 ng/mL) stimulated proliferation above even the levels found in the control grown in 10% FCS, whereas apoptosis levels dropped below ( Fig. 2A , B ). No (28.1) or a much less pronounced (28.3) IGF-1–induced effect was observed if the cells had been kept in the presence of doxycycline ( Fig. 2A , B ). Levels of IGF-1 that induced cellular proliferation and rescued cells from apoptosis also led to high levels of receptor tyrosine kinase autophosphorylation ( Fig. 2C ).

Figure 1.

Insulin-like growth factor I receptor (IGF1-R) Western blot analysis of two sublines, 28.1/3, generated from clone 28. (A) Cells grown in the presence of doxycycline were washed and then resuspended in medium with or without repressor. Cells were lysed after 24, 48, and 72 h and the lysates tested for the presence of the recombinant IGF-1R via 9E10 Western blot.

Figure 2.

Fluorescence-activated cell sorting (FACS), enzyme-linked immunoadsorbent assay (ELISA), and Western blot analysis of insulin-like growth factor I receptor (IGF-1R). Cells from the IGF-1R expressing lines 28.1 and 28.3 were grown in the presence or absence of doxycycline and serum starved for 24 h before increasing amounts of IGF-1 were added to the medium. Forty-eight hours later, cells were either fixed and stained for cell cycle FACS analysis or lysed for Western blot analysis. (A) The percentage of cells in G2/M (FACS analysis). (B) The percentage of apoptotic cells (FACS analysis). (C) The degree of IGF-1R autophosphorylation (ELISA). (D) IGF-1R (9E10), AKT phosphorylation, and total AKT (loading control) levels as measured in Western blots.

Western blot analysis confirmed the absence of the IGF-1R in the doxycycline-treated cells ( Fig. 2D ). Phosphorylation of AKT-1, as a downstream target of the IGF-1R, and a potential pharmacodynamic marker for IGF1-R inhibition, could be detected in cells treated with the two highest IGF-1 concentrations. Total AKT levels served as a loading control ( Fig. 2D , bottom panel).

Immunofluorescence analysis was used to verify the expression of IGF-1R on a single cell level. Most cells expressed the transgene at easily detectable levels, with the highest levels apparently in cells piling on top of each other. Addition of doxycycline completely abolished the signal ( Fig. 3A ). Finally, the cells were checked for their ability to grow anchorage independently. In contrast to the parental clone 28, both subclones formed large colonies in soft agar ( Fig. 3B ).

Figure 3.

Immunofluorescence and soft agar assays. (A) Immunofluorescence staining of cells grown on cover slips either in the presence or absence of the repressor. The cells were fixed after 72 h, and the insulin-like growth factor I receptor (IGF-1R) expressed by the cells were detected using the 9E10 myc antibody and counterstained using DAPI. (B) Cells were embedded in soft agar and grown in the presence or absence of doxycycline.

Next, growth of subclones 28.1 and 28.3 was tested in nude mice.

Reversible Growth In Vivo of Subclones 28.1 and 28.3

Cells from both subclones were implanted into the left flanks of a total of nine mice each, along with a negative control using the original, nontransformed MEF cell line. Three mice bearing 28.1 or 28.3 cells, respectively, received doxycycline directly after implantation, as a control for the IGF-1R dependency of the tumor growth. Of the other six animals, three animals received doxycycline as an anti–IGF-1R treatment after about 21 d, when the tumors reached an average size of about 200 to 300 mm³ (compared with parental clone 28: 40 d; Supp. Fig. 1C). A control group of three animals was left untreated (clone 28.1) or received doxycycline at tumor sizes of more than 1000 mm³, representing late-stage tumors (clone 28.3).

Tumors from untreated mice reached a size of about 2000 mm³ after 45 d (3/3; Fig. 4A , left panel, black line). Doxycycline treatment of tumors at a size of 200 to 300 mm³ induced regression, and tumors were undetectable after 12 d ( Fig. 4A , dark gray line). However, when doxycycline was removed, tumors relapsed and exponential growth was observed after another 3 to 4 wk (3/3, both cell lines; Fig. 4A , dark gray line). If, after 5 wk, doxycycline was removed from animals that had been treated from day 0, two of three animals bearing the 28.1 cell line developed a tumor (28.3: 0/3; Fig. 4A , left panel, light gray line). Last but not least, tumors that had been treated with doxycycline at a late stage (>1000 mm³) also regressed ( Fig. 4A , left panel, light gray line; right panel, black line). But, after about 8 d, relapse was observed even in the presence of doxycycline ( Fig. 4A , right panel, black line).

Figure 4.

In vivo growth of the sublines 28.1/28.3 and growth inhibition by doxycycline. (A) Into the flanks of nude mice, 2 × 10⁶ cells were injected subcutaneously. Tumor growth was regularly checked via calipering, and at given tumor sizes, insulin-like growth factor I receptor (IGF-1R) expression was repressed by exposing the mice to doxycycline in their drinking water (downward arrows). In some cases, doxycycline was removed after some time (upward arrows). Growth curves from 28.1 (left panel) and 28.3 (right panel) as well as the parental cell line (MEFTOFF) are shown. (B) Immunohistochemistry and TUNEL staining of sections from tumors of animals kept in the absence (–doxy) or presence (+doxy) of doxycycline. Left panels, anti-myc staining (brown) and hematoxylin nuclear counterstain (blue); right panels, TUNEL stain (green: TdT-FITC–labeled apoptotic cells; red: PI, nuclear counterstain).

To confirm that doxycycline treatment shut down IGF-1R expression in vivo, tissue sections were prepared from tumors of the 28.1-untreated control group ( Fig. 2A , left panel, black line) and the late-stage treated group ( Fig. 4A , left panel, light gray line). Besides the exogenous IGF-1R receptor, which was identified by its myc-tag, a potential induction of apoptosis as a consequence of IGF-1R repression was analyzed via TUNEL staining.

Doxycycline treatment completely abolished expression of IGF-1R in vivo. Whereas a robust, relatively homogenous staining was obtained from sections of untreated tumors, doxycycline treatment reduced the signal from the myc-tagged transgene to undetectable levels ( Fig. 4B ). Furthermore, TUNEL-positive areas that were absent from sections of untreated tumors were readily detected in those of their treated counterparts ( Fig. 4B ).

The relapse of tumor growth observed in the presence of doxycycline was puzzling ( Fig. 4A , right panel, black line). Potential explanations for this unexpected finding are, among others, that (1) it was an isolated event, specific for clone 28.3, (2) the cells had acquired resistance against doxycycline, or (3) the cells had overcome their dependence on the IGF-1R for survival. To distinguish between these possibilities, cells of the clone 28.1 were implanted into nude mice and left either untreated or treated with doxycycline. Although some tumors of the latter group were harvested during regression, others were kept under treatment to check whether a relapse of tumor growth could be observed. Sections of the tumors were then stained either for myc, to determine the level of exogenously expressed protein, or IGF-1R, to test whether endogenously expressed IGF-1R might complement for the loss of the exogenously expressed protein. Furthermore, the proliferation status of the tumors was determined via BrdU incorporation.

Three exemplary tumor growth curves are shown in Figure 5 : one control tumor, which grew to a size of approximately 2500 mm³ in approximately 50 d ( Fig. 5 , black line), and two tumors that had been treated with doxycycline: one isolated while in regression ( Fig. 5 , light gray line) and the other treated until growth resumption ( Fig. 5 , dark gray line). Regrowth of two other tumors that received continued doxycycline treatment was also observed, indicating that the relapse was a reproducible event (data not shown). Immunohistochemical staining of tumor sections from the respective samples taken after doxycycline treatment demonstrated a complete elimination of the myc-signal ( Fig. 5 , left immunohistochemistry [IHC] panels). This result was in line with a net reduction of total IGF-1R, as demonstrated using an antibody recognizing both human and mouse IGF-1R protein ( Fig. 5 , middle IHC panels). No induction of endogenous IGF-1R expression was observed from tumors that regrew in the presence of doxycycline ( Fig. 5 , compare the middle IHC panels), indicating that the tumors activated an alternative survival pathway.

Figure 5.

In vivo growth of the sublines 28.1/28.3 and growth inhibition by doxycycline. Repeat of Figure 4 , but only one animal per group is shown (total n = 3 per group). Animals were sacrificed at given time points during tumor growth, regression, or relapse, and the tumors were analyzed with immunohistochemistry. 9E10 (exogenous), as well as an insulin-like growth factor I receptor (IGF-1R) antibody (total IGF-1R) and BrdU stains are shown. A quantification of the results of the BrdU staining is shown in Table 1 .

Table 1.

Effect of Insulin-like Growth Factor I Receptor Repression on Tumor Cell Proliferation In Vivo

	BrdU-Positive Cells^a	p Value^b
−Doxycycline	251.7 ± 118.0
+Doxycycline day 43	118.5 ± 48.3	0.03
+Doxycycline day 43	9.0 ± 4.0	<0.0001

Per mm².

With respect to control.

The rate of proliferation as determined via BrdU incorporation, as expected, mirrored the position of the tumor in the growth curve. The noninhibited exponentially growing tumor had the highest percentage of BrdU-positive cells, followed by the relapsing tumor. The tumor in remission was essentially depleted of S-phase cells ( Fig. 5 , right IHC panels; Table 1 ).

Thus, the repression of the IGF-1R as the original transforming principle led to a significant, but only transient, tumor regression in this cell system. The reason for the relapse was not a leakiness of the tet-system but most likely a switch to another survival pathway. Nevertheless, the original clear dependence of the cell lines on the IGF-1R signaling for survival and proliferation under starvation conditions in vitro and the rapid induction of tumor remission (even if transitory) make this a valuable cellular test system for IGF-1R inhibitors in vitro and in vivo.

Characterization of Agents Directed against IGF-1R In Vitro and In Vivo

Two kinase inhibitors were tested on the cell line in vitro: NVP-AEW541 and staurosporine. NVP-AEW541 is a small molecule that inhibits Ins-R and IGF-1R in in vitro kinase assays with similar IC₅₀ values (0.15 µM) but distinguishes between the two highly similar receptor tyrosine kinases in cellular assays (IC₅₀s of 2.3 and 0.086 µM, respectively¹¹). Staurosporine is a rather general kinase inhibitor, with IC₅₀s in the nanomolar range for many kinases.²¹ The 28.1 cells were serum starved followed by a preincubation with the inhibitor and a subsequent stimulation by IGF-1.

The IC₅₀ for NVP-AEW541, 54 nM, corresponded very well with the literature value. Staurosporine was 10 times less effective with an IC₅₀ of 650 nM ( Fig. 6 ).

Figure 6.

Cellular assays using 28.1. (A) Enzyme-linked immunoadsorbent assay (ELISA)–based measurement of NVP-AEW541 and staurosporine-mediated inhibition of cellular insulin-like growth factor I receptor (IGF-1R) autophosphorylation. ELISA signals expressed as percentage of untreated control (z′ = 0.66).

A murine anti–IGF-1R monoclonal antibody, EM164, served to test the model system in vivo. EM164 is a potent antagonist of human IGF-1R that blocks binding of IGF-1 to the receptor, inhibits receptor activation, promotes degradation of IGF-1R, and exhibits activity against tumor cell lines in vitro and in vivo.^17,18 Mice engrafted subcutaneously with #28.1 cells were randomized into six groups at day 30 when the tumor volumes reached sizes of ≥200 mm³. Group 1 received vehicle, group 2 doxycycline (2 mg/mL) in their drinking water as a positive control (days 30–54), and groups 3 to 6 increasing doses of EM164 (0.4, 1, 4, and 40 mg/kg/injection) administered intraperitoneally twice weekly for 3 wk (six injections; days 30–48).

Tumor growth curves demonstrated a nice dose-dependent effect of EM164, although doxycycline was the most effective treatment ( Fig. 7A ). Five days after onset of treatment, doxycycline induced 100% complete tumor regressions, confirming that tumor growth was IGF-1R dependent ( Fig. 5B ; ***p < 0.0001). At the same time, antitumor activity of EM164 was observed at 40 mg/kg (11/11 PR, 6/11 CR; ***p < 0.0001), 4 mg/kg (1/12 PR, 1/12 CR; not significant; if the outlier, which exceeded 450 mm³ at the time of randomization, is omitted, **p = 0.009), and 1 mg/kg (3/12 PR, 2/12 CR; *p = 0.03), whereas the lower dose was inactive ( Fig. 7B , right upper panel). When compared at the time of necropsy (day 54 [some animals: day 51, see figure legend]), doxycycline (***p < 0.001), as well as the two higher doses of EM164 (*p = 0.04 for the 4 mg/kg group and ***p < 0.001 for the 40 mg/kg group, respectively), led to significantly reduced tumor sizes when compared with the untreated control ( Fig. 7C ). Taken together, these findings established EM164 as a dose-dependent inhibitor of IGF-1R function that suppresses tumor growth in vivo.

Figure 7.

In vivo inhibition of 28.1 tumor growth by the anti–insulin-like growth factor I receptor (IGF-1R) antibody EM164. (A) Tumor sizes for the in vivo growth curve were measured via calipering three times per week, with the curves representing the mean ± SEM for each group (upper left-hand panel). The dotted vertical gray line marks the randomization day. (B, C) Scatter graphs showing tumor sizes as calipered 6 d after treatment onset at day 36 and as measured at necropsy (day 54; however, tumor sizes from animals sacrificed due to ethical reasons [ulcerating tumors] on day 51 were included: control group, four animals; EM164 0.4 mg/mL, two animals; 1 mg/mL, one animal; 4 mg/mL, two animals). (D) Bar graphs from measurements in tumor lysates of IGF1-R expression (left) and AKT phosphorylation levels, displayed as percentage of control.

As a biomarker for the effect of EM164, IGF-1R down-regulation and a potential reduction of phospho-AKT levels as an important downstream target of IGF-1R were measured in tumor lysates. In untreated tumors, human IGF1R was expressed at very high levels (1094 ± 87.3 ng/mg of total protein), also when compared to tumor cell lines (e.g., 6.45 ng/mg of total protein in pancreas BXPC-3 or 6.02 ng/mg in multiple myeloma L363). Doxycycline treatment essentially abolished IGF-1R expression (4.4 ± 1.01 ng/mg of total protein; –99.6%), but EM164, albeit only at the highest dose of 40 mg/kg, also significantly reduced the levels of IGF-1R (–64%; Fig. 7D ). Lower doses showed increased rather than decreased IGF-1R levels, possibly indicating a compensatory effect by the tumor. Phosphorylation of AKT was also strongly reduced by doxycycline and the highest EM164 dose, indicating that the IGF-1R signaling cascade was effectively inhibited ( Fig. 7D ).

Discussion

The present study characterizes MEF cell lines, which are reversibly transformed due to a regulatable overexpression of the receptor tyrosine kinase IGF-1R. The original clone, 28, showed good overexpression of the transgene and clear IGF-1 dependency when grown in the absence of serum. However, when embedded into soft agar, or implanted into nude mice, growth was very slow, indicating that the transforming potential of the IGF-1R had not yet been fully exploited. Sublines generated from clone 28 via soft agar selection yielded cell lines that still depended on IGF-1R expression but grew much faster both in soft agar and in the mouse model.

The molecular differences between clone 28 and its sublines 28.1 and 28.3 have not yet been fully elucidated. No significant difference was found between the parental and the sublines with respect to IGF-1R expression levels (compare Supp. Fig. 1A with Fig. 1A; the Western blots had been developed side by side) or cell proliferation (compare the upper panels of Supp. Fig. 1B with Fig. 1B), but the sublines appeared much more sensitive to serum starvation, with apoptosis reaching levels of up to 30% (Supp. Fig. 1B and Fig. 1B, second panel). This effect was somewhat less pronounced with clone 28.3, but that clone also responded to IGF-1 stimulation at IGF-1R levels undetectable by Western blotting. An increased sensitivity toward growth factor deprivation is often found in fast proliferating tumor cells.²²

The relapse of tumor growth in the mouse in the presence of doxycycline was an unexpected finding ( Fig. 4 ). However, this was not due to leakiness of the tet-system, because exogenous IGF-1R levels—as detected via 9E10 IHC—remained low. Also, total IGF-1R levels remained much lower than in the induced state, indicating that elevated expression of endogenous IGF-1R did not compensate for the loss of human IGF-1R expression ( Fig. 5 ). Similar observations were also made in a comparable model using ErbB2 as a doxycycline-controlled transforming receptor tyrosine kinase.¹⁵ The expression and/or activation of other receptors might provide a potential explanation for this phenomenon. EM164 treatment of hepatocellular carcinoma was shown to lead to EGFR/HER3 activation, and a synergy was found between the antibody and the EGFR inhibitor gefitinib.²³ Similar observations were reported for a small-molecule IGF-1R inhibitor, BMS-536924.²⁴ Interestingly, the reverse also seems to work, that is, IGF-1R appears to mediate resistance to the HER-2 antibody trastuzumab.²⁵ However, at least in colon cancer, EGFR treatment resistance does not appear to sensitize tumors to IGF-1R treatment.²⁶ Thus, a similar scenario may be responsible for the doxycycline resistance/IGF-1R independence observed in our MEF cell lines.

The window of IGF-1R dependency is, however, large enough, and the cell line described here has been successfully employed in a preclinical discovery campaign to establish the IGF-1R–mediated antitumor activity of EM164, a murine IGF-1R–specific antibody.^17,18 Treatment with EM164 resulted in a dose-dependent tumor regression, demonstrating the principle that EM164 can block IGF-1R function and suppress in vivo the growth of IGF-1R–dependent tumors. IGF-1R protein levels as well as AKT phosphorylation as pharmacodynamic biomarkers were strongly reduced upon EM164 treatment, establishing that the antibody reached its target in the tumor cells and blocked IGF-1R signaling by inducing its degradation ( Fig. 7A , B ).

Moreover, a proof-of-concept study demonstrated that this cell line is a valuable tool to discover blood-based biomarker via secretome proteomics. Vascular endothelial growth factor and osteopontin were differentially detected by LC-MS/MS in vitro, and osteopontin could indeed be detected via ELISA in the plasma of mice bearing IGF-1R–expressing tumors but not their doxycycline-treated counterparts. The increase of osteopontin levels paralleled tumor growth, qualifying the molecule as a potential plasma biomarker for IGF-1R–dependent tumors.²⁷

In conclusion, the cell lines described here represent important tools for IGF-1R–related research and drug discovery.

Footnotes

Supplementary material for this article is available on the Journal of Biomolecular Screening Web site at .

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

The authors received no financial support for the research, authorship, and/or publication of this article.

References

Kaleko

Rutter

W. J.

Miller

A. D.

Overexpression of the Human Insulinlike Growth Factor I Receptor Promotes Ligand-Dependent Neoplastic Transformation. Mol. Cell. Biol. 1990, 10, 464–473.

Rubini

Hongo

D’Ambrosio

Baserga

The IGF-I Receptor in Mitogenesis and Transformation of Mouse Embryo Cells: Role of Receptor Number. Exp. Cell Res. 1997, 230, 284–292.

Baserga

The Insulin-like Growth Factor I Receptor: A Key to Tumor Growth?

Cancer Res. 1995, 55, 249–252.

Sachdev

Zhang

Matise

Gaillard-Kelly

Yee

The Type I Insulin-like Growth Factor Receptor Regulates Cancer Metastasis Independently of Primary Tumor Growth by Promoting Invasion and Survival. Oncogene 2010, 29, 251–262.

Sell

Rubini

Rubin

Liu

J. P.

Efstratiadis

Baserga

Simian Virus 40 Large Tumor Antigen Is Unable to Transform Mouse Embryonic Fibroblasts Lacking Type 1 Insulin-like Growth Factor Receptor. Proc. Natl. Acad. Sci. U. S. A. 1993, 90, 11217–11221.

White

P. J.

Fogarty

R. D.

Werther

G. A.

Wraight

C. J.

Antisense Inhibition of IGF Receptor Expression in HaCaT Keratinocytes: A Model for Antisense Strategies in Keratinocytes. Antisense Nucleic Acid Drug Dev. 2000, 10, 195–203.

Prager

H. L.

Asa

Melmed

Dominant Negative Inhibition of Tumorigenesis In Vivo by Human Insulin-like Growth Factor I Receptor Mutant. Proc. Natl. Acad. Sci. U. S. A. 1994, 91, 2181–2185.

D’Ambrosio

Ferber

Resnicoff

Baserga

A Soluble Insulin-like Growth Factor I Receptor That Induces Apoptosis of Tumor Cells In Vivo and Inhibits Tumorigenesis. Cancer Res. 1996, 56, 4013–4020.

Arteaga

C. L.

Osborne

C. K.

Growth Inhibition of Human Breast Cancer Cells In Vitro with an Antibody against the Type I Somatomedin Receptor. Cancer Res. 1989, 49, 6237–6241.

10.

De Leon

D. D.

Wilson

D. M.

Powers

Rosenfeld

R. G.

Effects of Insulin-like Growth Factors (IGFs) and IGF Receptor Antibodies on the Proliferation of Human Breast Cancer Cells. Growth Factors 1992, 6, 327–336.

11.

Garcia-Echeverria

Pearson

M. A.

Marti

. In Vivo Antitumor Activity of NVP-AEW541—A Novel, Potent, and Selective Inhibitor of the IGF-IR Kinase. Cancer Cell 2004, 5, 231–239.

12.

Zha

Lackner

M. R.

Targeting the Insulin-like Growth Factor Receptor-1R Pathway for Cancer Therapy. Clin. Cancer Res. 2010, 16, 2512–2517.

13.

Gossen

Freundlieb

Bender

Muller

Hillen

Bujard

Transcriptional Activation by Tetracyclines in Mammalian Cells. Science 1995, 268, 1766–1769.

14.

Baasner

von Melchner

Klenner

Hilgard

Beckers

Reversible Tumorigenesis in Mice by Conditional Expression of the HER2/c-erbB2 Receptor Tyrosine Kinase. Oncogene 1996, 13, 901–911.

15.

Schiffer

I. B.

Gebhard

Heimerdinger

C. K.

. Switching ff HER-2/neu in a Tetracycline-Controlled Mouse Tumor Model Leads to Apoptosis and Tumor-Size-Dependent Remission. Cancer Res. 2003, 63, 7221–7231.

16.

Bradford

M. M.

A Rapid and Sensitive Method for the Quantitation of Microgram Quantities of Protein Utilizing the Principle of Protein-Dye Binding. Anal. Biochem. 1976, 72, 248–254.

17.

Maloney

E. K.

McLaughlin

J. L.

Dagdigian

N. E.

. An Anti-Insulin-like Growth Factor I Receptor Antibody That Is a Potent Inhibitor of Cancer Cell Proliferation. Cancer Res. 2003, 63, 5073–5083.

18.

Bladt

Vrignaud

Chiron

. Pre-clinical Evaluation of the Anti-tumor Activity of the IGF1R-Specific Antibody AVE1642. In 97th AACR annual meeting; AACR: Washington, DC, 2006.

19.

Valentinis

Baserga

IGF-I Receptor Signalling in Transformation and Differentiation. Mol. Pathol. 2001, 54, 133–137.

20.

LeRoith

Roberts

C. T.

Jr.

The Insulin-like Growth Factor System and Cancer. Cancer Lett. 2003, 195, 127–137.

21.

Karaman

M. W.

Herrgard

Treiber

D. K.

. A Quantitative Analysis of Kinase Inhibitor Selectivity. Nat. Biotechnol. 2008, 26, 127–132.

22.

Evan

G. I.

Wyllie

A. H.

Gilbert

C. S.

. Induction of Apoptosis in Fibroblasts by c-myc Protein. Cell 1992, 69, 119–128.

23.

Desbois-Mouthon

Baron

Blivet-Van Eggelpoel

M. J.

. Insulin-like Growth Factor-1 Receptor Inhibition Induces a Resistance Mechanism via the Epidermal Growth Factor Receptor/HER3/AKT Signaling Pathway: Rational Basis for Cotargeting Insulin-like Growth Factor-1 Receptor and Epidermal Growth Factor Receptor in Hepatocellular Carcinoma. Clin. Cancer Res. 2009, 15, 5445–5456.

24.

Huang

Greer

Hurlburt

. The Mechanisms of Differential Sensitivity to an Insulin-like Growth Factor-1 Receptor Inhibitor (BMS-536924) and Rationale for Combining with EGFR/HER2 Inhibitors. Cancer Res. 2009, 69, 161–170.

25.

Zhao

Mascarenhas

Pollak

Insulin-like Growth Factor-I Receptor Signaling and Resistance to Trastuzumab (Herceptin). J. Natl. Cancer Inst. 2001, 93, 1852–1857.

26.

Reidy

D. L.

Vakiani

Fakih

M. G.

. Randomized, Phase II Study of the Insulin-like Growth Factor-1 Receptor Inhibitor IMC-A12, with or without Cetuximab, in Patients with Cetuximab- or Panitumumab-Refractory Metastatic Colorectal Cancer. J. Clin. Oncol. 2010, 28, 4240–4246.

27.

Piersma

S. R.

Fiedler

Span

. Workflow Comparison for Label-Free, Quantitative Secretome Proteomics for Cancer Biomarker Discovery: Method Evaluation, Differential Analysis, and Verification in Serum. J. Proteome Res. 2010, 9, 1913–1922.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.14 MB

Generation of a Conditionally Transformed Murine Embryonic Fibroblast Cell Line Using Doxycycline-Dependent IGF-1R Overexpression

Abstract

Keywords

Materials and Methods

Cell Lines and Antibodies

Generation of an MEF Cell Line Inducibly Expressing the IGF-1R

FACS/Western/Phospho-ELISA Analysis

Soft Agar and Immunofluorescence Assays

Cellular Phosphorylation Assays

Animal Experiments

IGF-1R Antibody EM164

IGF-1R Protein Expression in MEF/IGF-1R Tumor Samples

Immunohistochemistry

Results

Generation of a Tet-Repressable IGF-1R MEF Cell Line

Generation of Subclones 28.1 and 28.3 with Improved Growth Characteristics

Reversible Growth In Vivo of Subclones 28.1 and 28.3

Characterization of Agents Directed against IGF-1R In Vitro and In Vivo

Discussion

Footnotes

References

Supplementary Material