Regulator of G-protein signaling–GAIP-interacting protein COOH terminus (GIPC) is involved in protein trafficking, endocytosis, and receptor clustering and is associated with insulin-like growth factor I receptor (IGF-IR), a receptor important for proliferation and anchorage-independent growth. Here, we described GIPC expression in different human pancreatic adenocarcinoma (PCA) cell lines and we examined the role of GIPC in the regulation of IGF-IR protein levels in PCA. Interestingly, inhibition of GIPC expression by RNA interference led to reduced IGF-IR protein levels and a subsequent decrease in proliferation of PCA cells. We also determined that the PDZ domain of GIPC is essential for the post-translational regulation and the binding of IGF-IR. The importance of GIPC in pancreatic cancer development and progression is supported by tissue microarray data of 300 pancreatic cancer specimens where GIPC is highly expressed in PCA. Taken together, our data suggest that GIPC is a central molecule for the stability of IGF-IR and could be a target for future therapy. (Cancer Res 2006; 66(21): 10264-8)
Ductal adenocarcinoma of the pancreas is an aggressive disease characterized by its invasiveness, rapid progression, and resistance to treatment (1). There is currently no effective treatment option. The main goal of our study is to elucidate the molecular mechanism of pancreatic cancer development. The insulin-like growth factor I (IGF-I) receptor (IGF-IR) is an important tyrosine kinase involved in anchorage-independent growth and the progression of pancreatic adenocarcinoma (PCA; ref. 2). Recently, GIPC, an regulator of G-protein signaling (RGS)-GAIP binding protein, was found to bind IGF-IR (3, 4). In cultured cells, GIPC is found on small vesicles near the plasma membrane, which contain transferrin and is associated with Myosin VI (5). Furthermore, in vivo, GIPC is localized in clathrin-rich invaginations and endocytic compartments (6). Because this evidence supports a possible role for GIPC in protein trafficking and endocytosis, its association with IGF-IR raises the question of whether GIPC plays an important role in regulating IGF-IR expression and, subsequently, pancreatic cancer development. In this study, we show that GIPC is highly expressed in human PCA and knocking down GIPC leads to a decreased proliferation rate. Furthermore, we showed that GIPC is important for tumor growth by promoting the stability of IGF-IR and preventing its degradation.
Materials and Methods
Cell lines. Human primary pancreatic ductal adenocarcinoma cell lines (American Type Culture Collection, Manassas, VA) PANC1, MIA-PaCa2, AsPC1, Su86.86, CFPAC, and Capan-2 were grown in RPMI 1640 or DMEM supplemented with 10% fetal bovine serum, penicillin, and streptomycin and incubated at 37°C with 5% CO2.
Western blot and immunoprecipitation. Whole-cell lysates were prepared as described previously (7) and separated by SDS-PAGE. Goat polyclonal antibodies against the NH2 terminus of GIPC (clone N19, Santa Cruz Biotechnology, Santa Cruz, CA), rabbit polyclonal antibodies against IGF-IRβ (clone C20, Santa Cruz Biotechnology), and antibodies against β-actin and FLAG (Sigma, St. Louis, MO) were used for immunodetection followed with an horseradish peroxidase (HRP)–conjugated secondary antibody (Santa Cruz Biotechnology) and the SuperSignal West Pico substrate (Pierce Biotechnology, Rockford, IL). Immunoprecipitations were done as described previously (7). Specifically, 1,000 μg cell lysate was incubated with 2 to 6 μg of a rabbit polyclonal antibody against human GIPC (Proteus BioSciences, Ramona, CA) or FLAG. Proteasome inhibitor I (Calbiochem, San Diego, CA) was used at a concentration of 25 μmol/L.
RNA interference, transfection, and proliferation assay. After a 24-hour incubation with antibiotic-free medium, cells were transfected with GIPC small interfering RNA (siRNA) using the Transfection Reagent 2 (Dharmacon, Lafayette, CO). Sixty hours after transfection, GIPC knockdown was confirmed by Western blot analysis. Proliferation was measured at different time points using a colorimetric 3-(4-5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium salt (MTS) assay (Promega, Madison, WI) and thymidine incorporation was also done 60 hours after transfection. Plasmid transfection was done using the Effectene kit (Qiagen, Valencia, CA) according to the recommendation of the manufacturer.
Immunofluorescence. Cells (2 × 104) were seeded on chamber slides in antibiotic-free medium for 24 hours. Cells were then transfected with GIPC siRNA or scramble siRNA (Dharmacon). The medium was changed 48 hours after transfection, and after 96 hours, slides were washed and fixed with 4% paraformaldehyde. After blocking with 10% goat serum for 15 minutes, the slides were stained with primary antibodies against IGF-IR or GIPC (1:100) for 2 hours in 1% goat serum. Slides were then incubated with secondary antibodies labeled with Alexa Fluor 488 and Alexa Fluor 633 or Alexa 456 (1:200; Molecular Probes, Eugene, OR) for 1 hour. After mounting with Vectashield containing 4′,6-diamidino-2-phenylindole (Vector Laboratories, Burlingame, CA), confocal microscopy was done. For colocalization studies, the cells were permeabilized with 0.2% Triton X-100 at room temperature for 5 minutes.
Reverse transcription-PCR analysis. RNA was isolated from cells using the RNeasy Mini kit (Qiagen). The sequences for human GIPC and IGF-IR are the following: GIPC, 5′-GCAGCGTGATCGACCAGAT-3′ (forward) and 5′-GCAGGCTCTGCCCGTTAA-3′ (reverse); IGF-IR, 5′-CATCGACATCCGCAACGA-3′ (forward) and 5′-CCCTCGATCACCGTGCA-3′ (reverse). RNA (2 μg) was used for reverse transcription and the resulting cDNA (2 μL) was used for PCR with 1 mmol/L of each primer pair and the SYBR Green PCR Master Mix (Applied Biosystems, Foster City, CA). The conditions for the PCR are as follows: 40 cycles of 95°C for 15 seconds and 60°C for 1 minute. The housekeeping gene 36B4 was used to normalize all results.
Immunohistochemistry. Paraffin-embedded specimens were washed thrice in xylene and once with 100% ethanol. Specimens were rehydrated. Endogenous peroxidases were quenched with 3% H2O2 in methanol for 30 minutes. Antigen retrieval was done with 1 mmol/L EDTA for 30 minutes at room temperature. Specimens were incubated with the primary rabbit anti-GIPC antibody (1:1,000; gift from M. Farquhar, Department of Cellular and Molecular Medicine, UCSD, La Jolla, CA; ref. 8) for 30 minutes at room temperature and then developed using the RbMACH3/HRP method and an autostainer.
Tissue microarray analysis and patient data. Of the 128 patients included in this study, 73 were male. By International Union Against Cancer staging, 23 samples were stage I, 29 were stage II, 59 were stage III, and 14 were stage IV. With grading according to WHO, 38 tumors were grade 1, 79 were grade 2, and 10 were grade 3. Per patient, 3 cores were prepared, although 84 of these cores could not be evaluated. GIPC expression was scored semiquantitatively by two pathologists (T.S. and M.H.M.) in respect to intensity (scale: 0-3) and extent (scale: 0, 0% positive; 1, 1-25% positive; 2, 26-50% positive; 3, 51-75% positive; and 4, 76-100% positive) of staining using a 12-point scale (product of intensity and extent).
Statistical analysis. The cases were then grouped as follows: “low grade/normal” consisting of normal tissue, pancreatic intraepithelial neoplasia (PanIN) tissue, and G1 adenocarcinomas and “high grade” consisting of G2 and G3 cases. To determine whether the categorical/ordinal variables were associated with each other, the Pearson's λ2 test and rank-sum tests were used and Ps < 0.01 were considered significant.
GIPC protein levels were highly detectable in all cell lines. To assess the role of GIPC in PCA, we investigated six pancreatic tumor cell lines for GIPC expression and found that it was highly expressed (Fig. 1A). To study whether GIPC depletion affects tumor cell proliferation or cell survival, we knocked down GIPC by RNA interference (RNAi; Fig. 1B). Our results showed that a significant decrease in cellular proliferation as measured by either thymidine incorporation (Fig. 1C) or the MTS assay (Fig. 1D) had been exhibited by all tested cell lines with decreased GIPC protein expression. These results suggest that GIPC expression is correlated with PCA cell proliferation.
GIPC regulates IGF-IR protein expression in PCA cells. To evalulate the molecular mechanism of GIPC-mediated PCA proliferation, we postulated that GIPC might control IGF-IR expression in PCA cells. As shown by Western blot, IGF-IR levels significantly decreased in PCA cells after treatment with GIPC siRNA in a dose-dependent manner (Fig. 2A). Similar results were also observed in confocal microscopy when immunostaining was done in PCA cells with anti-IGF-IR antibody (red) in the presence of GIPC siRNA (Fig. 2B). Furthermore, we identified that GIPC and IGF-IR are in the same immunocomplex as examined by immunoprecipitation (Fig. 2C) and confocal microscopy (Fig. 2D). These immunocomplexes appeared to be transient as they were not always detectable.
We next tested whether GIPC regulates IGF-IR protein expression at the transcriptional level. Using quantitative PCR with IGF-IR primers, we found only minor changes in the level of IGF-IR mRNA after GIPC knockdown, suggesting the influence of GIPC on a different pathway (Fig. 3A). As GIPC is involved in protein trafficking and receptor clustering, we postulated that GIPC might protect IGF-IR from proteasomal degradation. After treatment with a proteasomal inhibitor, we were able to regain the expression of IGF-IR in GIPC siRNA-treated samples, indicating a role for GIPC in preventing the proteasomal degradation of IGF-IR (Fig. 3B). We then examined whether the PDZ domain of GIPC was important for IGF-IR stabilization. To evaluate the significance of the PDZ domain, we transiently transfected AsPC1 and PANC1 cells with a FLAG-tagged GIPC expression plasmid lacking the PDZ domain and IGF-IR was consistently down-regulated in both PCA (Fig. 3C). These data suggest that GIPC lacking the PDZ domain functions as a dominant-negative mutant. To further show the importance of the PDZ domain for the binding of GIPC to IGF-IR, we transfected pancreatic cancer cells with a FLAG-tagged wild-type (WT) GIPC expression plasmid and a plasmid lacking the PDZ domain of GIPC (GIPCΔPDZ). After 2 hours of proteasome inhibitor treatment to avoid the degradation of IGF-IR, we could detect WT GIPC and IGF-IR in the same immunocomplex but not GIPCΔPDZ and IGF-IR (Fig. 3D), clearly showing the importance of the PDZ domain.
GIPC is highly expressed in the cytoplasm of carcinoma samples. To determine GIPC protein expression in primary human pancreatic ductal adenocarcinoma, tissue microarrays of normal and tumor tissues were screened with immunohistochemistry (Fig. 4A-D). Expression was observed mainly on the luminal portion of the nonmalignant ductal epithelial cells and only a few acinar pancreatic cells expressed GIPC. GIPC expression was confined to epithelial cells and did not stain the underlying stroma. Importantly, there was significantly higher cytoplasmic GIPC expression in the higher grade patient samples compared with that of normal pancreatic tissue, PanINs, and low-grade cancers (Supplementary Table). For statistical evaluation, the Pearson's λ2 test and rank-sum tests were used with a significant Ps < 0.01. These results suggest that GIPC expression correlates with disease progression and subsequent IGF-IR expression.
Using tissue microarrays and immunohistochemistry, our results show a high level of GIPC in human tissue samples of PCA in comparison to normal acinar and ductal tissue, supporting the results of previous studies (9). Using pancreatic cell lines as an in vitro model for PCA, our data clearly indicate that increased expression is due to the regulatory effects of GIPC on IGF-IR.
Previous reports have shown that IGF-IR is important for pancreatic cancer cell survival (10–12). In our study, we were able to reduce IGF-IR expression in PCA cells by specific knockdown of GIPC. We showed that the knockdown of GIPC leads to a proliferative decrease in four different PCA cell lines, a result that could be attributed to IGF-IR down-regulation. This down-regulation of IGF-IR has been described to lead to an increase in apoptosis. Interestingly, the apoptotic pathway seems to only be partially affected as most of the cell lines tested did not show any increase in apoptosis after GIPC reduction. Two possibilities are that other mechanisms are possibly counteracting the effect of IGF-IR down-regulation or the time points for apoptosis evaluation are too early. In contrast, the metastatic Su86.86 pancreatic cell line showed a clear apoptotic increase as evaluated with Annexin V staining and cleaved poly(ADP-ribose) polymerase detection (data not shown) after GIPC knockdown. Our data further validates the previously reported association between IGF-IR and GIPC via immunoprecipitation and confocal microscopy. It seems that GIPC bridges an important connection between IGF-IR and G-protein signaling as one of its binding partners is a RGS (8). Possibly, this connection is also important for the regulation of IGF-IR levels in PCA cells, although this will have to be determined in future studies.
The main molecular mechanism by which GIPC regulates IGF-IR expression seems to be the stabilization of the IGF-IR protein itself. Using a proteasome inhibitor, IGF-IR expression could be recovered after RNAi treatment against GIPC. Furthermore, there is only a minor reduction of IGF-IR messenger after GIPC knockdown, indicating a role for GIPC in regulating IGF-IR expression at the post-translational level. Although GIPC is an adaptor molecule for Myosin VI and GAIP (13), these molecules have very important roles in retrograde intracellular protein transport and endocytosis. This links GIPC to protein trafficking (13, 14) and proposes that GIPC prevents IGF-IR from degradation by connecting it to the transport machinery of the cells. Another important function of GIPC is the clustering of receptors on the cell surface (15). In the absence of GIPC, the clusters could potentially become unstable and the receptors could undergo degradation. Despite these observations, we cannot disregard other potent binding partners to GIPC that are involved in proliferation and survival, such as integrins (16), because it is not yet determined what effect GIPC down-regulation has on these proteins.
Several reports have indicated that protein-protein interactions in tumor and normal cells are essential elements in organizing protein complexes. One of the most common domains for protein interaction is the PSD95/DlgA/ZO-1 homology protein interaction domain or simply the PDZ domain. PDZ domains were first identified in the neuron expressed postsynaptic density protein (PSD95), its Drosphila homologue disc large tumor suppressor (DlgA) gene product (17), and in the tight junction protein zonula occludens-1 (ZO1). PDZ domains are composed of ∼95 residues that act as modules and scaffolds for protein-protein interactions. Interestingly, our laboratory showed that the PDZ domain of GIPC is important for neuropilin-1-mediated signaling and possibly transmits neuropilin-1 signaling during angiogenesis (7). Studies in Xenopus and mammalian cells have shown that the binding of GIPC to IGF-IR requires this PDZ domain, although additional domains are also involved (3, 4). Although we have shown that the PDZ domain of GIPC is important for the regulation of IGF-IR, further studies need to be carried out to understand the molecular mechanism of how this PDZ domain stabilizes IGF-I, although a recently published study emphasized the importance of GIPC in retrograde protein trafficking (18).
This study has shown that a new therapeutic approach to pancreatic cancer is feasible by targeting the PDZ domain of GIPC. Many published reports have focused on blocking IGF-IR function by preventing ligand binding or by disturbing the receptor tyrosine kinase activity but these approaches have not proven to be successful yet. From our results, GIPC expression can be correlated with disease progression, thereby suggesting that the blocking of GIPC is a viable therapeutic option against PCA.
Note: Supplementary data for this article are available at Cancer Research Online (http://cancerres.aacrjournals.org/).
Grant support: NIH grants CA78383, HL072178, and HL70567, Mayo Clinic Pancreatic Cancer Specialized Programs in Research Excellence Pilot grant, and American Cancer Society grant (D. Mukhopadhyay). M.H. Muders is a German Research Foundation Fellow (Deutsche Forschungsgemeinschaft; grant MU 2687/1-1). D. Mukhopadhyay is a Scholar of the American Cancer Society.
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
We thank Dr. Marilyn G. Farquhar for the GIPC antibody and Dr. Ascoli for the GIPC expression plasmids. We acknowledge Prof. Dr. G. Baretton, Dresden, Germany for his great support and advice.