AKT Activation Promotes Metastasis in a Mouse Model of Follicular Thyroid Carcinoma
http://www.100md.com
《内分泌学杂志》
Laboratory of Molecular Biology (C.S.K., Y.K., S.-Y.C.) and Experimental Immunology Branch (M.K.), Center for Cancer Research, National Cancer Institute, Bethesda, Maryland 20892
Divisions of Endocrinology and Oncology, Ohio State University School of Medicine, and Arthur G. James Cancer Center (V.V.V., M.S., M.D.R.), Columbus, Ohio 43210
Abstract
The phosphatidylinositol 3-kinase/AKT pathway is crucial to many cell functions, and its dysregulation in tumors is a common finding. The molecular basis of follicular thyroid cancer metastasis is not well understood but may also be influenced by AKT activation. We previously created a knockin mutant mouse that expresses a mutant thyroid hormone receptor- gene (TRPV mouse) that spontaneously develops thyroid cancer and distant metastasis similar to human follicular thyroid cancer. In this study, we investigated whether our mouse model exhibits similar AKT activation as human follicular thyroid cancer. Western blot analysis on thyroids from both wild-type and TRPV/PV mice revealed elevation of activated AKT in TRPV/PV mice. Immunohistochemistry and confocal microscopy reveal activated AKT in both the thyroid and metastatic lesions of TRPV/PV mice. Whereas all three AKT isoforms were overexpressed in primary tumors from TRPV/PV mice in the cytoplasm of thyroid cancer cells, only AKT1 was also found in the nucleus, matching the localization of activated AKT in a pattern similar to human follicular thyroid cancer. In the metastases, all AKT isoforms correlated with phosphorylated AKT nuclear localization. We created primary thyroid cell lines derived from TRPV/PV mice and found reduction of phosphorylated AKT levels or AKT downstream targets diminishes cell motility. Activated AKT is common to both human and mouse follicular thyroid cancer and is correlated with increased cell motility in vitro and metastasis in vivo. Thus, TRPV/PV mice could be used to further dissect the detailed pathways underlying the progression and metastasis of follicular thyroid carcinoma.
Introduction
ACTIVATION OF THE phosphatidylinositol 3-kinase (PI3-K)/AKT signaling pathway has been associated with multiple human cancers. AKT (protein kinase B), a serine/threonine kinase, was initially identified as a viral oncogene and is composed of three isoforms, AKT1, AKT2, and AKT3, with differing levels of tissue expression and function (1, 2). PI3-K activation of AKT is a multistep process beginning with PI3-K phosphorylating phosphatidylinositol-4,5-bisphosphate to phosphatidylinositol-3,4,5-triphosphate. AKT’s binding to these phosphoinositides results in its relocation from the cytoplasm to the plasma membrane (3). Phosphorylation of AKT at two distinct sites leads to its activation and nuclear translocation (3, 4). Downstream targets of the PI3-K/AKT pathway affect many cellular processes such as growth, differentiation, motility, and apoptosis. Dysregulation of the PI3-K/AKT pathway has been seen with mutations in the PI3-K regulatory domain, PI3-K amplification, AKT overexpression, or loss of function of the tumor suppressor gene, phosphatase and tensin homolog deleted on chromosome 10 (PTEN) (5).
Metastasis is a complex process whereby tumor cells have acquired the abilities to detach from the primary site through alterations in cell-cell adhesion molecules, degrade the extracellular matrix, invade other sites through altered cell motility, and survive in a new environment (6). AKT1 has been shown to regulate cell motility and that its overexpression in tumors may enhance their metastatic potential (7). AKT1 overexpression is associated with increased activity of matrix metalloproteinases through its activation of nuclear factor-B binding to the matrix metalloproteinase promoter (7, 8), and its ability to up-regulate angiogenesis, through VEGF, may also contribute to tumor survival (9). Other AKT isoforms have been reported to promote invasiveness through up-regulation of 1-integrins and other pathways (10). The PI3-K/AKT pathway therefore has an important role in the metastatic phenotype.
Thyroid cancer, derived from follicular epithelial cells, has many different subtypes, with the most common being papillary and follicular cancers. Follicular thyroid cancer (FTC) is typically a well-differentiated cancer but has a greater predilection than papillary to metastasize to distant sites. The pathways resulting in FTC metastasis are not well understood, and the known genetic alterations responsible for FTC tumorigenesis, including activating ras mutations, inactivating mutations of PTEN, and Pax8/PPAR rearrangements, account for only a minority of cases (11). Recent studies have suggested that the thyroid hormone receptors (TRs) may play a role in cancer development (12). TRs are ligand-dependent nuclear transcription factors that regulate gene expression by interacting with the thyroid hormone response element of its target genes. Two separate TR genes exist, TR and TR, which may be alternatively spliced to form four distinct TR isoforms that bind to T3. TR1 was originally discovered as the counterpart to a viral oncogene, v-erbA (13). V-erbA is a mutated TR that is unable to bind to T3 but may still bind to thyroid hormone response elements to repress transcription; its dominant-negative activity is sufficient to promote tumorigenesis in transgenic mice (14).
In addition, TR may also function as a tumor suppressor. A knockin mouse containing the dominant-negative TR1 mutant, PV (TRPV mouse) exhibits an extreme phenotype of the human syndrome, resistance to thyroid hormone. TRPV/PV mice are homozygous for the mutant TR, which lacks any ability to bind to T3 and loss of transcriptional activity. TRPV/PV mice have dysfunction of the pituitary-thyroid axis and display inappropriately high levels of TSH with high thyroid hormone levels. Interestingly, TRPV/PV mice spontaneously develop FTC beginning with hyperplasia and then capsular and vascular invasion, with progression to distant metastasis and transformation to anaplastic thyroid cancer (15). Further work with TRPV mice showed that in the absence of a wild-type allele, a single mutated TR, can cause FTC, signifying that TR can act as tumor suppressor (16). TRPV/PV mice provide an in vivo model for studying the pathways contributing to FTC metastasis.
Recent studies of primary human thyroid cancer specimens by several groups showed AKT overexpression and overactivation, particularly in FTC cases (17, 18). In addition, in our studies, AKT activity is most predominant in thyroid cancer cells invading tumor capsules, compared with those localized to central, less invasive regions (19). These invasive cells in the primary tumors were also characterized by predominantly nuclear colocalization of phosphorylated AKT (pAKT) and AKT1. In vitro studies using a poorly differentiated thyroid cancer cell line, NPA, revealed that the ability of the cells to migrate was associated with nuclear localization of pAKT and that PI3-K inhibitors reduced cell motility, thus suggesting a functional role for this pathway in thyroid cancer motility (19).
Considering the current data regarding AKT1 activation in human FTC and its association with motility and the interaction of TR and the PI3-K/AKT pathway, it is important to determine what occurs in vivo (20, 21). This study investigates, using the TRPV mouse model, whether there is a correlation of AKT activation and tumor formation and progression by comparing AKT expression and activity levels in spontaneously developing FTC tumors from the primary site and distant lung metastatic sites. In addition, we evaluate whether there is any particular pattern of subcellular localization of AKT activity in regions of invasion or metastases. We also present evidence that inhibition of PI3-K in primary thyroid cell lines from TRPV/PV mice results in down-regulation of AKT with consequent decrease in cell motility in vitro. The TRPV/PV mouse model recapitulates human FTC biology with an increase in pAKT and association of pAKT and the ability to invade.
Materials and Methods
Mouse strains
The care and handling of the animals used in this study were approved by the National Cancer Institute Animal Care and Use Committee. Mice harboring the TRPV gene (TRPV) were prepared via homologous recombination, as previously described (22). Genotyping was carried out using the PCR method described previously (22).
Western blot analysis
TRPV/PV mice and their gender- and age-matched wild-type control mice were killed, and the thyroid gland was removed and frozen in liquid nitrogen. Preparation of the thyroid was modified from a previously described protocol of preparation of protein from the pituitary gland (23). The thyroid gland was homogenized on ice, in lysis buffer containing 50 mM Tris, 100 mM HCl, 0.1% Triton X-100, 0.2 μM okadaic acid, 100 mM NaF, and 2 mM Na3VO4 and a proteinase inhibitor tablet (Complete Mini EDTA free; Roche, Mannheim, Germany), followed by incubation on ice for 10 min with occasional vortexing. The lysate was centrifuged for 5 min at 20,000 x g at 4 C, and the supernatant was collected. For Western blot analysis of primary thyroid cells, cells were harvested by adding scraping buffer [150 mM NaCl, 41.8 mM Tris (pH 7.4), and 1.044 mM EDTA], scraped with a plastic spatula, and centrifuged at 2600 x g at 4 C for 3 min, and then the supernatant was removed. The cells were incubated in lysis buffer [50 mM Tris, 250 mM NaCl, 5 mM EDTA, 0.5% Nonidet P-40, 0.2 μM okadaic acid, and a proteinase inhibitor tablet (Complete Mini EDTA-free; Roche)] for 10 min on ice, vortexed, and spun at 16,000 x g at 4 C for 10 min.
The protein concentration for each lysate was determined by the Bradford assay (Pierce Chemical Co., Rockford, IL) using BSA (Pierce) as the standard. Protein sample (50 μg) was loaded and separated by SDS-PAGE. After electrophoresis, the protein was electrotransferred to a polyvinylidene difluoride membrane (Immobilon-P; Millipore Corp., Bedford, MA). Antibodies used according to the manufacturers’ manuals include phospho-AKT (Ser473) no. 9271 (Cell Signaling Technologies, Beverly, MA), AKT no. 9272 (Cell Signaling Technologies) at 1:1000 dilution, hemagglutinin (HA) sc-7392 (Santa Cruz Biotechnology, Santa Cruz, CA) at 1:200, phospho-mammalian target of rapamycin (mTOR) (Ser2448) no. 2971 (Cell Signaling Technologies) at 1:1000, and phospho-p70 S6K (Thr421/Ser424) no. 9204 (Cell Signaling Technologies) at 1:1000 dilution. These antibodies are documented by Cell Signaling Technologies to recognize the above-mentioned proteins from mice, excluding HA. To verify the specificity of the AKT and pAKT antibodies, we performed Western blot analysis comparing lysate from mouse thyroids with cell lysate prepared from MCF-7 cells serum starved for 16 h, before and after a 30-min incubation of 100 ng/ml IGF-I (Pepro-Tech, Rocky Hill, NJ).
Secondary antibodies used were horseradish peroxidase-conjugated goat antimouse or antirabbit IgG (Amersham Biosciences, Piscataway, NJ) and detected using the Western Lightning chemiluminescence reagent plus system (PerkinElmer Life Sciences, Boston, MA). The blots were stripped with Re-Blot Plus (Chemicon, Temecula, CA) and reprobed with rabbit polyclonal antibodies to protein disulfide isomerase at 0.5 μg/ml (Sigma, St. Louis, MO). Three independent experiments were performed using male age-matched mice and one with female age-matched TRPV/PV and wild-type mice.
Immunohistochemistry
Histological sections were prepared as previously described (19). Phosphorylated AKT (Ser473, immunohistochemistry specific) no. 9277 (Cell Signaling Technologies), AKT1 sc-5298 (Santa Cruz Biotechnology), AKT2 06-606 (Upstate Biotechnology, Lake Placid, NY), AKT3 sc-11520 (Santa Cruz Biotechnology), and AKT no. 9272 (Cell Signaling Technologies) were used. Vectastain Universal Quick kit, antigen unmasking solution, and 3,3'-diaminobenzidine substrate kit were purchased from Vector Laboratories (Burlingame, CA). Negative controls not including primary antibody were performed for each experiment. Five low- and five high-power fields were independently examined by three investigators (M.D.R., M.S., and V.V.V.), including one pathologist (V.V.V.). Figures are representative of those examinations.
Confocal microscopy
Confocal images of thyroid and lung tissue samples from both wild-type and TRPV/PV mice were collected using a Zeiss LSM 510 microscope equipped with a META spectral detector (Carl Zeiss Inc., Jena, Germany). Primary antibodies used include AKT1 sc-5298 (Santa Cruz Biotechnology), AKT2 06–606 (Upstate Biotechnology), AKT3 sc-11520 (Santa Cruz Biotechnology), and phospho-AKT (Ser473, immunohistochemistry specific) no. 9277 (Cell Signaling Technologies). Images were collected as previously described (19). Briefly, images from tissue samples that were unlabeled, from Alexafluor 488-conjugated secondary antibody alone (Molecular Probes Inc, Eugene, OR) or tissue samples immunolabeled with primary and secondary antibody, were collected using the same acquisition configuration. Using the Alexafluor 488 antibody-only sample and unlabeled specimens, reference spectra were calculated at regions of interest to determine the emission signatures for Alexafluor 488 and background autofluorescence of the samples. Separation of specific fluorescence emission signal was obtained using a linear unmixing algorithm included in the Zeiss operating software (version 3.0). The resultant images were exported as TIF files and prepared as figures using Adobe Photoshop (version 5.02; Adobe Systems Inc., San Jose, CA).
Primary thyroid culture
Using a method modified from a previously described protocol (24), thyroid tissue was placed in sterile PBS and minced using a razor blade. The tissue was spun at 950 x g for 5 min at 4 C and the supernatant was removed. The tissue was then placed in a solution containing MEM Eagle salts (Sigma), 112 U type I collagenase (Sigma), and 1.2 U/ml Dispase 1 (Roche) and incubated in a 37 C shaker for 2 h. The tissue was then sheared using a 19-gauge then 21-gauge needle (10 times). The tissue was centrifuged at 1500 x g for 5 min, and the supernatant was removed. The cells were then washed and centrifuged in primary cell media. The culture medium was prepared as follows. NuSerum 4 (BD Biosciences, Bedford, MA) was diluted 2.5 times with Ham’s F-12 media (Sigma). This diluted NuSerum 4 was supplemented with somatostatin (Sigma) at a concentration of 10 ng/ml, glycyl-L-histidyl-L-lysine acetate (Sigma) at a concentration of 2 ng/ml, TSH (Sigma) at a concentration of 5 mIU/ml, and antibiotic-antimycotic (Gibco, Invitrogen, Carlsbad, CA) at a concentration of 200 U/ml. The cells were then plated and grown at 37 C in a 5% CO2, 90% humidity incubator.
Cell motility assay
The method used was modified from the original protocol described previously (25). Motility assays were performed in 8-μm-pore transwells (6.5 mm; Costar, Corning, NY) in triplicate. Primary cells were detached by trypsin, and 5 x 104 cells were placed in 200 μl Ham’s F-12 media. Bottom wells contained 500 μl of the primary cell media. Cells were incubated at 37 C for 2 h. The wells were then decanted and the cells fixed in 1% glutaraldehyde in PBS (Sigma) and stained with 0.1% crystal violet (Sigma) in water for 30 min. The wells were destained by rinsing in deionized water. Using separate cotton-tipped applicators soaked in 0.2% Triton X-100 (Sigma), both the nonmigrating cells on the upper side of the well and the migrating cells on the lower side were removed and placed in 1 ml of 0.2% Triton X-100. These samples were kept at 4 C overnight. The solution of each sample was measured at A590 using a spectrophotometer. Rate of motility was calculated by the absorbance of the migrating cells divided by the absorbance of the nonmigrating cells.
PI3-K inhibition
Primary TRPV/PV thyroid cells were grown in primary cell media until 50–60% confluent and then incubated in 10 μM LY294002 (Thomas Scientific, Swedesboro, NJ) or continuous media for 24 h. Afterward the cells were harvested for Western blot analysis or cell motility assay as described above.
Adenovirus infection
Primary TRPV/PV thyroid cells were seeded at 1.5 x 105 overnight in 60-mm dishes. The following morning cells were infected with replication-defective adenoviral constructs consisting of either an HA-tagged dominant-negative mutant of rat AKT1 (DNAKT) under the control of the CAG eukaryotic promoter or containing the gene for -galactosidase (-gal) (26, 27). These viruses were previously shown to infect mouse primary smooth muscle cells (28). Earlier determination of multiplicity of infection (MOI) was carried out using primary thyroid cells infected with various titers of the -gal adenovirus for 4 h and then 48 h later, staining for -gal using the -gal staining kit (Roche). Cells infected at a MOI of 100 for 4 h at 37 C resulted in more than 95% infection based on -gal staining. After incubation of primary thyroid cells with either the adenovirus for -gal or DNAKT for 4 h, media were replaced with fresh primary cell media. Forty-eight hours after infection, the cells were harvested for either Western blot analysis or cell motility assay as described above.
Statistical analysis
Data are expressed as mean ± SEM. Cell motility experiments were analyzed by unpaired t test with StatView 5.0 (Abacus Concepts, Inc., San Diego, CA) where P < 0.05 was considered significant.
Results
Overexpression of pAKT in the thyroids of TRPV/PV mice
TRPV/PV mice spontaneously develop FTC with distant metastasis similarly to human FTC. The apparent role of AKT overexpression and overactivation in human FTC development and behavior suggests that AKT activity may also have a role in our model of FTC (19). To evaluate this possibility, we performed Western blots for total AKT, pAKT, and a control protein from age- and gender-matched wild-type and TRPV/PV mice at different ages. Representative results are shown in Fig. 1. We found, similar to human FTC, that TRPV/PV thyroid glands are characterized by AKT overactivation, compared with gender- and age-matched wild-type thyroids (Fig. 1). As TRPV/PV mice age, the incidence of FTC and distant metastasis increases (15, 16). Whereas at 3 months of age, all TRPV/PV mice develop thyroid hyperplasia, they may not have yet developed FTC (15). In Fig. 1, the increased level of pAKT is marked by 6 months of age and persists through 11 months. At the age of 11 months, the expression of total AKT protein was increased as compared with wild-type mice. The band intensities were quantified and analysis of pAKT to AKT ratios indicate an age-dependent increase of pAKT/AKT as TRPV/PV mice age (Fig. 1B), suggesting an increase activity of AKT as thyroid carcinogenesis progresses. The increased pAKT expression in thyroid tumor cells rather than associated stromal cells was further confirmed by immunohistochemistry. Compared with wild-type mice (Fig. 2A), strong immunoactive staining of pAKT was detected in thyroid cancer cells of TRPV/PV mice (Fig. 2B, shown by arrows). Thus, the abundance of pAKT protein is elevated in thyroid cancer cells of TRPV/PV mice.
In contrast to the overexpression and overactivation of AKT in the primary and metastatic thyroid tumor cells of TRPV/PV mice, another major signaling pathway, the ERK-MAPK pathway, which could also be involved in thyroid tumorigenesis and can influence cell motility (29), was not similarly overexpressed or overactivated in the primary tumors or metastases (data not shown). These findings are consistent with our past observations that in human FTC, we did not find evidence of enhanced activation of MAPK in invasive vs. normal cells (19). Thus, increased activation of the Ras-ERK-MAPK pathway, unlike the PI3-K/AKT signaling pathway, is not associated with progression and metastasis of follicular thyroid cancer in TRPV/PV mice. These data demonstrate that thyroid tumorigenesis in the TRPV/PV mice is associated with enhanced activation of AKT.
Expression of pAKT in the metastatic lesions of TRPV/PV mice
To determine whether pAKT levels may also be correlated with metastases, the metastatic lung lesions from TRPV/PV mice were isolated and evaluated for pAKT staining by immunohistochemistry. As shown in Fig. 2D, the metastatic lesions in the lung retain high levels of pAKT. The wild-type lung section does not contain any metastatic lesions and reveals that there is little background pAKT in normal lung tissue (Fig. 2C).
Overexpression of all AKT isoforms in the thyroid cancers of TRPV/PV mice
There are tissue-specific expression patterns of the three isoforms of AKT. AKT1 is present in almost all tissues at high levels, whereas high levels of AKT2 are observed in the muscle and reproductive organs, and AKT3 is highly expressed in the brain and testis (30, 31, 32, 33). Whereas the AKT isoforms demonstrate substrate specificity, it is also thought that there is functional redundancy (34) and that subcellular localization of AKT may also play a role in determining its downstream effects. As shown above, as the TRPV/PV mice age, thyroid cancer develops, and there is overexpression of total AKT and overactivity of AKT. We evaluated the expression profiles of the three AKT isoforms in the thyroids of older (i.e. > 10 months old) TRPV/PV mice by immunohistochemistry using antibodies that are isoform specific but recognize both activated phosphorylated and inactive forms. We found that all isoforms of AKT are overexpressed in the thyroid cancers of older TRPV/PV mice (Fig. 3, A1, B1, and C1;) in comparison with age-matched wild-type controls (Fig. 3, A, B, and C, respectively). These findings are similar to human FTC in which AKT1 and AKT2 were overexpressed, compared with normal thyroid tissue, and AKT3 results were more variable (20).
Subcellular localization of AKT isoforms and pAKT in primary FTC lesions
It has been reported that whereas the majority of AKT is located in the cytoplasm, AKT is also translocated into the nucleus in which it interacts with specific nuclear targets, suggesting that subcellular localization of AKT may be an important determinant of biological activity (34). Previous work associated higher levels of nuclear pAKT in human FTC cells that were in invasive regions of primary tumors on the periphery of the tumor mass and that AKT1 colocalized with pAKT in those regions in comparison with AKT2 or AKT3 (19). Figure 3A1 shows that AKT1 is localized to both the nucleus and cytoplasm in the primary thyroid tumors (shown by an arrow), whereas AKT2 and AKT3 are predominantly cytoplasmic (Fig. 3, B1 and C1, respectively, open arrows). Immunoactive pAKT was localized to both compartments (Fig. 3, D1). In contrast, in metastasis, all AKT isoforms and immunoactive pAKT localized primarily to the nucleus (Fig. 3, A2, B2, C2, D, and E). These immunohistochemical data were confirmed by subsequent confocal immunofluorescence studies (Fig. 4, A and B). Figure 4A, a and c, highlights AKT1’s presence in both the nucleus and cytoplasm in the primary tumors (arrowhead and arrows, respectively), whereas AKT2 and AKT3 remain in the cytoplasm (Fig. 4A, f and i). In metastasis, activated AKT is predominantly localized to the nucleus (Fig. 4B, a and c).
Reduced motility is associated with inhibition of PI3-K/AKT signaling pathway
To determine whether the PI3-K/AKT pathway might play a functional role in TRPV/PV metastasis, we created primary cultures from the thyroid tumors of the TRPV/PV mice. Using Boyden chamber migration assays, we evaluated cell motility in the presence or absence of PI3-K inhibitor, LY294002. Figure 5A shows that in the presence of LY294002 (10 μM), the extent of activation of pAKT was reduced 50–60% as compared with the control. Concomitant with the reduction of pAKT, cell motility was also reduced approximately 57% (Fig. 5B). To confirm the critical role of AKT in motility in these primary cells, we transiently infected cells with adenovirus containing either a HA-tagged, dominant-negative AKT or -gal as a control. This mutant AKT1 has alanine substituted at the known phosphorylation sites (Thr308 and Ser473), resulting in the loss of its kinase activity (26). Kitamura et al. (26) also found that this construct exhibits dominant-negative activity; infection of Chinese hamster ovary and 3T3-L1 adipocyte cells resulted in loss of insulin-stimulated endogenous AKT activity. Stabile et al. (28) showed that primary mouse smooth muscle cells infected with this dominant-negative AKT have a reduction in AKT activity as shown by a decrease in phosphorylation of a downstream target, p70 S6K. As inhibition of PI3K activity reduces primary thyrocyte motility, the selective reduction of AKT activity, via transient expression of this dominant-negative AKT, should also show similar effects on motility. Figure 6A shows that infection of the primary cells with the DNAKT have decreased phosphorylation of two downstream AKT targets, mTOR and p70 S6K, consistent with inhibition of AKT activity. Phosphorylated AKT levels were not evaluated because the mutant AKT cross-reacts with pAKT antibodies (personal communication, Dr. W. Ogawa, Kobe University, Japan). The resultant decrease in AKT activity is associated with decreased cell motility (Fig. 6B). Taken together, these data suggest that the association between enhanced AKT activation and metastasis in the TRPV/PV mice may be functionally important.
Discussion
The mechanism for FTC metastasis is not well understood but may include signaling pathways already implicated in promoting metastasis in other cancers. The PI3-K/AKT pathway is a logical point of investigation as patients with the autosomal dominant Cowden’s syndrome who have mutations in PTEN, resulting in AKT overactivation, develop FTC (35). However, it is often difficult to obtain sufficient quantities of human metastatic FTC tissues for profiling the genetic differences between the primary and metastatic lesion. The recent creation of the TRPV/PV mouse model that spontaneously develops FTC with metastasis provides an opportunity to elucidate the pathways contributing to the metastatic phenotype.
Prior work has shown human FTC to have elevated pAKT levels, compared with normal tissue (17, 18), and also demonstrate nuclear localization of pAKT the more invasive regions of the tumor (i.e. periphery) in association with AKT1 localization (19). Functional data have demonstrated that human thyroid cancer cell motility is dependent on PI3-K activity (19). In this study, it is striking how similar the patterns of AKT activation and localization in primary thyroid cancers in TRPV/PV mice are to human FTC. Specifically, pAKT and AKT1’s presence in both the cytoplasm and nucleus in primary thyroid tumors in TRPV/PV mice resembles human FTC. The lack of enhanced ERK activation in the thyroid cancers of our mouse model also parallels human FTC. The association between nuclear AKT activity and metastasis in this model is strengthened by the greater predilection for pAKT and all three AKT isoforms to localize to the nucleus in the metastatic thyroid cancer cells. Finally, a functional role for PI3-K/AKT signaling pathway in metastasis in this model is strongly suggested by the in vitro data demonstrating that pharmacological inhibition of PI3K activity and specific inhibition of AKT both result in the reduction of TRPV/PV thyroid tumor cell motility. These results suggest that both AKT activity in general as well as its nuclear localization may play an important role in the development of FTC metastasis in TRPV/PV mice.
That the PI3-K/AKT signaling pathway is activated in the thyroids of TRPV/PV mice harboring a dominant-negative TR mutant is in contrast of recent findings by Cao et al. (20). Whereas T3 is considered to exert its effects via TR and at the level of transcription, Cao et al. showed that T3 could activate the PI3-K/AKT pathway via nongenomic effect. These authors found that in a human fibroblast system, viral-mediated overexpressed TR1 is coimmunoprecipitated with p85, the regulatory subunit of PI3-K (20). mTOR, a downstream effector of AKT, is also activated in the presence of T3. PI3-K inhibitors reduced T3 activation of mTOR, suggesting that the PI3-K/AKT pathway is responsible for mTOR activation by T3. It was further shown that a dominant-negative TR mutant, G345R, although able to bind to p85, was unable to activate the PI3-K/AKT signaling (20).
Whereas the molecular mechanisms by which the PV mutant mediates the activation of PI3-K/AKT signaling pathway in TRPV/PV mice is currently being elucidated in our laboratory, there are clear differences between the systems reported by Cao et al. (20) and the present study. First, Cao et al. used cultured fibroblasts in which TR1 or TR1G345R is overexpressed, whereas the present study is an in vivo system with physiological levels of TRs or PV. The PV mutation in TRPV mice is a targeted knockin mutation; hence, its regulation of expression is the same as the wild-type TR gene (22). It is reasonable to expect that the regulation and functions of TRs and TR mutants at the overexpressed levels could differ from those at the physiological level. Second, TR mutations are different in these two studies. The PV mutation is a frameshift mutation in the C-terminal 16 amino acids, whereas TR1G345R is a point mutation. There are precedents to suggest that different mutations of the TR lead to different functional consequences. Mice that harbor TR1337T display neurological phenotype (36), whereas TRPV mice have no such abnormalities. TRPV/PV mice developed follicular thyroid carcinoma (15, 16) and pituitary adenomas (37), but no such defects are reported in TR1337T knockin mice. Thus, it is possible that PV and TR1G345R affect differently on the PI3-K/AKT signaling pathway. Clearly, this issue would need to be clarified in future studies.
The consistent pattern of AKT activation and predilection for nuclear localization of AKT1 in primary tumors and all AKT isoforms in metastatic lesions from TRPV/PV mice suggests that FTC metastasis may be in part dependent on altered activity and cellular localization of the PI3-K/AKT signaling pathway. Whereas AKT1 and AKT2 are more ubiquitously expressed, compared with AKT3, the percentage of functional overlap among the isoforms is unknown. There appears to be differing levels of AKT isoform activation, depending on the type of tumor. Dufour et al. (38) reported that AKT isoforms have distinct functional consequences for survival of human intestinal epithelial cells along with different upstream activating signals. Whether the activation of all AKT isoforms in the primary thyroid lesion may provide a future diagnostic approach to distinguish benign vs. malignant tumor is unknown.
The predominant nuclear localization of all AKT isoforms and pAKT in the metastatic FTC lesions suggests that its nuclear distribution may be relevant to both initiating and sustaining metastasis. Nuclear AKT is known to affect many proteins, including the Forkhead box, class O (FOXO) family proteins, cAMP response element binding protein, and p21Cip/WAF1 (39, 40, 41). In particular, AKT phosphorylation inactivates the FOXO subfamily of forkhead transcription factors, leading to decreased levels of proapoptotic genes (42). Identification of the downstream targets of AKT that promote cell motility, migration, and angiogenesis is critical for not only the understanding of the molecular basis underlying metastasis of FTC but also the identification of potential molecular targets for treatment of FTC.
It is important to recognize that the nuclear changes associated with tumorigenesis in these tumors may promote a more nuclear appearance of immunohistochemical staining. However, the confocal microscopic studies’ more precise ability to detect differences in subcellular localization of AKT isoforms and pAKT between the primary cancer cells and metastatic cancer cells corroborate the immunohistochemistry results. The specific mechanisms responsible for the nuclear predominant localization of activated AKT of the metastatic thyroid cancer cells are uncertain. It is possible that the localization is intrinsically regulated by alterations in the tumor cells; it is also possible that interactions between the tumor cells and the microenvironment regulate localization.
In summary, in the present paper, we have demonstrated that that the PI3-K/AKT signaling pathway is activated similarly in the TRPV/PV mouse model in comparison with human FTC. Its activation is associated with tumor formation and nuclear localization of immunoactive pAKT is associated with tumor progression. In vitro, TRPV/PV primary thyrocyte motility is dependent on PI3-K and AKT signaling. Taken together, these data suggest that the TRPV/PV mouse model could be used to further dissect the pathways and identify key players that contribute to the metastasis of FTC.
Footnotes
This research was supported in part by the Intramural Research Program of the National Institutes of Health, National Cancer Institute, Center for Cancer Research.
Abbreviations: DNAKT, Dominant-negative mutant of rat AKT1; FTC, follicular thyroid cancer; -gal, -galactosidase; HA, hemagglutinin; MOI, multiplicity of infection; mTOR, mammalian target of rapamycin; pAKT, phosphorylated AKT; PI3-K, phosphatidylinositol 3-kinase; PTEN, phosphatase and tensin homolog deleted from chromosome 10; TR, thyroid hormone receptor.
References
Bellacosa A, Testa JR, Staal SP, Tsichlis P 1991 A retroviral oncogene, akt, encoding a serine-threonine kinase containing an SH2-like region. Science 254:274–277
Coffer PJ, Jin J, Woodgett JR 1998 Protein kinase B (c-AKT): a multifunctional mediator of phosphatidylinositol 3-kinase activation. Biochem J. 335:1–13
Franke TF, Hornik CP, Segev L, Shostak GA, Sugimoto C 2003 PI3K/AKT and apoptosis: size matters. Oncogene. 22:8983–8998
Feng J, Park J, Cron P, Hess D, Hemmings BA 2004 Identification of a PKB/Akt hydrophobic motif Ser-473 kinase as DNA-dependent protein kinase. J Biol Chem. 279:41189–41196
Luo J, Manning BD, Cantley, LC 2003 Targeting the PI3K-AKT pathway in human cancer: rationale and promise. Cancer Cell 4:257–262
Hanahan D, Weinberg RA 2000 The hallmarks of cancer. Cell 100:57–70
Park BK, Zeng X, Glazer RI 2001 Akt1 induces extracellular matrix invasion and matrix metalloproteinase-2 activity in mouse mammary epithelial cells. Cancer Res. 61:7647–7653
Kim D, Kim S, Koh H, Yoon SO, Chung AS, Cho KS, Chang J 2001 Akt/PKB promotes cancer cell invasion via increased motility and metalloproteinase production. FASEB J. 15:1953–1962
Jiang BH, Zheng JZ, Aoki M, Vogt PK 2000 Phosphatidylinositol 3-kinase signaling mediates angiogenesis and expression of vascular-endothelial growth factor in endothelia cells. Proc Natl Acad Sci USA. 97:1749–1753
Fujita S, Watanabe M, Kubota T, Teramoto T, Kitajima M 1995 Alteration of expression in integrin 1-subunit correlates with invasion and metastasis in colorectal cancer. Cancer Lett. 91:145–149
Nikiforova MN, Biddinger PW, Caudill CM, Kroll TG, Nikiforov YE 2002 PAX8 PPAR rearrangement in thyroid tumors: RT-PCR and immunohistochemical analyses. Am J Surg Pathol. 26:1016–1023
Cheng SY 2003 Thyroid hormone receptor mutations in cancer. Mol Cell Endocrinol. 213:23–30
Sap J, Munoz A, Damm K, Goldberg Y, Ghysdael J, Leutz A, Beug H, Vennstrom B 1986 The c-erb-A protein is a high-affinity receptor for thyroid hormone. Nature 324:635–640
Barlow C, Meister B, Lardelli M, Lendahl U, Vennstrom B 1994 Thyroid abnormalities and hepatocellular carcinoma in mice transgenic for v-erbA. EMBO J. 13:4241–4250
Suzuki H, Willingham MC, Cheng SY 2002 Mice with a mutation in the thyroid hormone receptor gene spontaneously develop thyroid carcinoma: a mouse model of thyroid carcinogenesis. Thyroid 12:963–969
Kato Y, Ying H, Willingham MC, Cheng SY 2004 A tumor suppressor role for thyroid hormone receptor in a mouse model of thyroid carcinogenesis. Endocrinology 145:4430–4438
Ringel MD, Hayre N, Saito J, Saunier B, Schuppert F, Burch H, Bernet V, Burman KD, Kohn LD, Saji M 2001 Overexpression and overactivation of Akt in thyroid carcinoma. Cancer Res. 61:6105–6111
Miyakawa M, Tsushima T, Murakami H, Wakai K, Isozaki O, Takano K 2003 Increased expression of phosphorylated p70S6 kinase and Akt in papillary thyroid cancer tissues. Endocr J. 50:77–83
Vasko V, Saji M, Hardy E, Kruhlak M, Larin A, Savchenko V, Miyakawa M, Isozaki O, Murakami H, Tsushima T, Burman KD, De Micco C, Ringel MD 2004 Akt activation and localization correlate with tumour invasion and oncogene expression in thyroid cancer. J Med Genet. 41:161–170
Cao X, Kambe F, Moeller LC, Refetoff S, Seo H 2005 Thyroid hormone induces rapid activation of AKT/protein kinase B-mammalian target of rapamycin-p70s6K cascade through phosphatidylinositol 3-kinase in human fibroblasts. Mol Endocrinol. 19:102–112
Saito J, Kohn AD, Roth RA, Noguchi Y, Tatsumo I, Hirai A, Suzuki K, Kohn LD, Saji M, Ringel MD 2001 Regulation of FRTL-5 thyroid cell growth by phosphatidylinositol (OH) 3 kinase-dependent Akt-mediated signaling. Thyroid 11:339–351
Kaneshige M, Kaneshige K, Zhu X-G, Dace A, Garrett L, Carter TA, Kazlauskaite R, Pankratz DG, Wynshaw-Boris A, Refetoff S, Weintraub B, Willingham MC, Barlow C, Cheng SY 2000 Mice with a targeted mutation in the thyroid hormone receptor gene exhibit impaired growth and resistance to thyroid hormone. Proc Natl Acad Sci USA. 97:13209–13214
Heaney AP, Singson R, McCabe, CJ, Nelson V, Nakashima M, Melmed S 2000 Expression of pituitary-tumour transforming gene in colorectal tumours. Lancet 355:716–719
Jeker LT, Hejazi M, Burek CL, Rose NR, Caturegli, P 1999 Mouse thyroid primary culture. Biochem Biophys Res Commun. 257:511–515
Yu Y, Merlino G 2002 Constitutive c-met signaling through an autocrine mechanism promotes metastasis in a transgenic transplantation model. Cancer Res. 62:2951–2956
Kitamura T, Ogawa W, Sakaue H, Hino Y, Kuroda S, Takata M, Matsumoto M, Maeda T, Konishi H, Kikkawa U, Kasuga M 1998 Requirement for activation of the serine-threonine kinase Akt (protein kinase B) in insulin stimulation of protein synthesis but not of glucose transport. MCB. 18:3708–3717
Takata M, Ogawa W, Kitamura T, Hino Y, Kuroda S, Kotani K, Klip A, Gingras A-C, Sonenberg N, Kasuga M 1999 Requirement for Akt (protein kinase B) in insulin-induced activation of glycogen synthase and phosphorylation of 4E-BP1 (PHAS-1). J Biol Chem. 274:20611–20618
Stabile E, Zhou YF, Saji M, Castagna M, Shou M, Kinnaird TD, Baffour R, Ringel MD, Epstein SE, Fuchs S 2003 Akt controls vascular smooth muscle cell proliferation in vitro and in vivo by delaying G1/S exit. Circ Res. 93:1059–1065
Janda E, Lehmann K, Killisch I, Jechlinger M, Herzig M, Downward J, Beug H, Grunert S 2002 Ras and TGF() cooperatively regulate epithelia cell plasticity and metastasis: dissection of Ras signaling pathways. J Cell Biol. 156:299–313
Coffer PJ, Woodgett JR 1991 Molecular cloning and characterisation of a novel putative protein-serine kinase related to the cAMP-dependent and protein kinase C families. Eur J Biochem. 201:475–481
Kim YB, Peroni OD, Franke TF, Kahn BB 2000 Divergent regulation of Akt1 and Akt2 isoforms in insulin target tissues of obese Zucker rats. Diabetes 49:847–856
Nakatani K, Sakaue H, Thompson DA, Weigel RJ, Roth R 1999a Identification of a human Akt3 (protein kinase B) which contains the regulatory serine phosphorylation site. Biochem Biophys Res Commun. 257:906–910
Franke TF 2000 Assays for AKT. Methods Enzymol. 322:400–410
Ahmed NN, Franke TF, Baellacosa A, Datta K, Gonzalez-Portal ME, Taguchi T, Testa JR, Tsichlis PN 1993 The proteins encoded by c-akt and v-akt differ in post-translational modification, subcellular localization and oncogenic potential. Oncogene. 8:1957–1963
Eng C 2003 PTEN: one gene, many syndromes. Hum Mutat. 22:183–198
Hashimoto K, Curty FH, Borges PP, Lee CE, Abel ED, Elmquist JK, Cohen RN, Wondisford FE 2001 An unliganded thyroid hormone receptor causes severe neurological dysfunction. Proc Natl Acad Sci USA. 98:3998–4003
Furumoto H, Ying H, Chandramouli GV, Zhao L, Walker RL, Meltzer PS, Willingham MC, Cheng SY 2005 An unliganded thyroid hormone receptor activates the cyclin D1/cyclin-dependent kinase/retinoblastoma/E2F pathway and induces pituitary tumorigenesis. Mol Cell Biol. 25:124–135
Dufour G, Demers M-J, Gagne D, Dydensborg AB, Teller IC, Bouchard V, Degongre I, Beaulieu J-F, Cheng JQ, Fujita N, Tsuruo T, Vallee K, Vachon PH 2004 Human intestinal epithelial cell survival and anoikis: differentiation state-distinct regulation and roles of protein kinase B/Akt isoforms. J Biol Chem. 279:44113–44122
Brunet A, Bonnin A, Zigmond MJ, Lin MZ, Juo P, Hu LS, Anderson MJ, Arden KC, Blenis J, Greenberg ME 1999 Akt promotes cell survival by phosphorylating and inhibiting a Forkhead transcription factor. Cell 96:857–868
Du K, Montminy M 1998 CREB is a regulatory target for the protein kinase Akt/PKB. J Biol Chem. 273:32377–32379
Shioi T, Kang PM, Douglas PS, Hampe J, Yballe CM, Lawitts J, Catley LC, Izumo S 2000 The conserved phosphoinositide 3-kinase pathway determines heart size in mice. EMBO J. 19:2537–2548
Gilley J, Coffer PJ, Ham J 2003 FOXO transcription factors directly activate bim gene expression and promote apoptosis in sympathetic neurons. J Cell Biol. 162:613–622(Caroline S. Kim, Vasily V)
Divisions of Endocrinology and Oncology, Ohio State University School of Medicine, and Arthur G. James Cancer Center (V.V.V., M.S., M.D.R.), Columbus, Ohio 43210
Abstract
The phosphatidylinositol 3-kinase/AKT pathway is crucial to many cell functions, and its dysregulation in tumors is a common finding. The molecular basis of follicular thyroid cancer metastasis is not well understood but may also be influenced by AKT activation. We previously created a knockin mutant mouse that expresses a mutant thyroid hormone receptor- gene (TRPV mouse) that spontaneously develops thyroid cancer and distant metastasis similar to human follicular thyroid cancer. In this study, we investigated whether our mouse model exhibits similar AKT activation as human follicular thyroid cancer. Western blot analysis on thyroids from both wild-type and TRPV/PV mice revealed elevation of activated AKT in TRPV/PV mice. Immunohistochemistry and confocal microscopy reveal activated AKT in both the thyroid and metastatic lesions of TRPV/PV mice. Whereas all three AKT isoforms were overexpressed in primary tumors from TRPV/PV mice in the cytoplasm of thyroid cancer cells, only AKT1 was also found in the nucleus, matching the localization of activated AKT in a pattern similar to human follicular thyroid cancer. In the metastases, all AKT isoforms correlated with phosphorylated AKT nuclear localization. We created primary thyroid cell lines derived from TRPV/PV mice and found reduction of phosphorylated AKT levels or AKT downstream targets diminishes cell motility. Activated AKT is common to both human and mouse follicular thyroid cancer and is correlated with increased cell motility in vitro and metastasis in vivo. Thus, TRPV/PV mice could be used to further dissect the detailed pathways underlying the progression and metastasis of follicular thyroid carcinoma.
Introduction
ACTIVATION OF THE phosphatidylinositol 3-kinase (PI3-K)/AKT signaling pathway has been associated with multiple human cancers. AKT (protein kinase B), a serine/threonine kinase, was initially identified as a viral oncogene and is composed of three isoforms, AKT1, AKT2, and AKT3, with differing levels of tissue expression and function (1, 2). PI3-K activation of AKT is a multistep process beginning with PI3-K phosphorylating phosphatidylinositol-4,5-bisphosphate to phosphatidylinositol-3,4,5-triphosphate. AKT’s binding to these phosphoinositides results in its relocation from the cytoplasm to the plasma membrane (3). Phosphorylation of AKT at two distinct sites leads to its activation and nuclear translocation (3, 4). Downstream targets of the PI3-K/AKT pathway affect many cellular processes such as growth, differentiation, motility, and apoptosis. Dysregulation of the PI3-K/AKT pathway has been seen with mutations in the PI3-K regulatory domain, PI3-K amplification, AKT overexpression, or loss of function of the tumor suppressor gene, phosphatase and tensin homolog deleted on chromosome 10 (PTEN) (5).
Metastasis is a complex process whereby tumor cells have acquired the abilities to detach from the primary site through alterations in cell-cell adhesion molecules, degrade the extracellular matrix, invade other sites through altered cell motility, and survive in a new environment (6). AKT1 has been shown to regulate cell motility and that its overexpression in tumors may enhance their metastatic potential (7). AKT1 overexpression is associated with increased activity of matrix metalloproteinases through its activation of nuclear factor-B binding to the matrix metalloproteinase promoter (7, 8), and its ability to up-regulate angiogenesis, through VEGF, may also contribute to tumor survival (9). Other AKT isoforms have been reported to promote invasiveness through up-regulation of 1-integrins and other pathways (10). The PI3-K/AKT pathway therefore has an important role in the metastatic phenotype.
Thyroid cancer, derived from follicular epithelial cells, has many different subtypes, with the most common being papillary and follicular cancers. Follicular thyroid cancer (FTC) is typically a well-differentiated cancer but has a greater predilection than papillary to metastasize to distant sites. The pathways resulting in FTC metastasis are not well understood, and the known genetic alterations responsible for FTC tumorigenesis, including activating ras mutations, inactivating mutations of PTEN, and Pax8/PPAR rearrangements, account for only a minority of cases (11). Recent studies have suggested that the thyroid hormone receptors (TRs) may play a role in cancer development (12). TRs are ligand-dependent nuclear transcription factors that regulate gene expression by interacting with the thyroid hormone response element of its target genes. Two separate TR genes exist, TR and TR, which may be alternatively spliced to form four distinct TR isoforms that bind to T3. TR1 was originally discovered as the counterpart to a viral oncogene, v-erbA (13). V-erbA is a mutated TR that is unable to bind to T3 but may still bind to thyroid hormone response elements to repress transcription; its dominant-negative activity is sufficient to promote tumorigenesis in transgenic mice (14).
In addition, TR may also function as a tumor suppressor. A knockin mouse containing the dominant-negative TR1 mutant, PV (TRPV mouse) exhibits an extreme phenotype of the human syndrome, resistance to thyroid hormone. TRPV/PV mice are homozygous for the mutant TR, which lacks any ability to bind to T3 and loss of transcriptional activity. TRPV/PV mice have dysfunction of the pituitary-thyroid axis and display inappropriately high levels of TSH with high thyroid hormone levels. Interestingly, TRPV/PV mice spontaneously develop FTC beginning with hyperplasia and then capsular and vascular invasion, with progression to distant metastasis and transformation to anaplastic thyroid cancer (15). Further work with TRPV mice showed that in the absence of a wild-type allele, a single mutated TR, can cause FTC, signifying that TR can act as tumor suppressor (16). TRPV/PV mice provide an in vivo model for studying the pathways contributing to FTC metastasis.
Recent studies of primary human thyroid cancer specimens by several groups showed AKT overexpression and overactivation, particularly in FTC cases (17, 18). In addition, in our studies, AKT activity is most predominant in thyroid cancer cells invading tumor capsules, compared with those localized to central, less invasive regions (19). These invasive cells in the primary tumors were also characterized by predominantly nuclear colocalization of phosphorylated AKT (pAKT) and AKT1. In vitro studies using a poorly differentiated thyroid cancer cell line, NPA, revealed that the ability of the cells to migrate was associated with nuclear localization of pAKT and that PI3-K inhibitors reduced cell motility, thus suggesting a functional role for this pathway in thyroid cancer motility (19).
Considering the current data regarding AKT1 activation in human FTC and its association with motility and the interaction of TR and the PI3-K/AKT pathway, it is important to determine what occurs in vivo (20, 21). This study investigates, using the TRPV mouse model, whether there is a correlation of AKT activation and tumor formation and progression by comparing AKT expression and activity levels in spontaneously developing FTC tumors from the primary site and distant lung metastatic sites. In addition, we evaluate whether there is any particular pattern of subcellular localization of AKT activity in regions of invasion or metastases. We also present evidence that inhibition of PI3-K in primary thyroid cell lines from TRPV/PV mice results in down-regulation of AKT with consequent decrease in cell motility in vitro. The TRPV/PV mouse model recapitulates human FTC biology with an increase in pAKT and association of pAKT and the ability to invade.
Materials and Methods
Mouse strains
The care and handling of the animals used in this study were approved by the National Cancer Institute Animal Care and Use Committee. Mice harboring the TRPV gene (TRPV) were prepared via homologous recombination, as previously described (22). Genotyping was carried out using the PCR method described previously (22).
Western blot analysis
TRPV/PV mice and their gender- and age-matched wild-type control mice were killed, and the thyroid gland was removed and frozen in liquid nitrogen. Preparation of the thyroid was modified from a previously described protocol of preparation of protein from the pituitary gland (23). The thyroid gland was homogenized on ice, in lysis buffer containing 50 mM Tris, 100 mM HCl, 0.1% Triton X-100, 0.2 μM okadaic acid, 100 mM NaF, and 2 mM Na3VO4 and a proteinase inhibitor tablet (Complete Mini EDTA free; Roche, Mannheim, Germany), followed by incubation on ice for 10 min with occasional vortexing. The lysate was centrifuged for 5 min at 20,000 x g at 4 C, and the supernatant was collected. For Western blot analysis of primary thyroid cells, cells were harvested by adding scraping buffer [150 mM NaCl, 41.8 mM Tris (pH 7.4), and 1.044 mM EDTA], scraped with a plastic spatula, and centrifuged at 2600 x g at 4 C for 3 min, and then the supernatant was removed. The cells were incubated in lysis buffer [50 mM Tris, 250 mM NaCl, 5 mM EDTA, 0.5% Nonidet P-40, 0.2 μM okadaic acid, and a proteinase inhibitor tablet (Complete Mini EDTA-free; Roche)] for 10 min on ice, vortexed, and spun at 16,000 x g at 4 C for 10 min.
The protein concentration for each lysate was determined by the Bradford assay (Pierce Chemical Co., Rockford, IL) using BSA (Pierce) as the standard. Protein sample (50 μg) was loaded and separated by SDS-PAGE. After electrophoresis, the protein was electrotransferred to a polyvinylidene difluoride membrane (Immobilon-P; Millipore Corp., Bedford, MA). Antibodies used according to the manufacturers’ manuals include phospho-AKT (Ser473) no. 9271 (Cell Signaling Technologies, Beverly, MA), AKT no. 9272 (Cell Signaling Technologies) at 1:1000 dilution, hemagglutinin (HA) sc-7392 (Santa Cruz Biotechnology, Santa Cruz, CA) at 1:200, phospho-mammalian target of rapamycin (mTOR) (Ser2448) no. 2971 (Cell Signaling Technologies) at 1:1000, and phospho-p70 S6K (Thr421/Ser424) no. 9204 (Cell Signaling Technologies) at 1:1000 dilution. These antibodies are documented by Cell Signaling Technologies to recognize the above-mentioned proteins from mice, excluding HA. To verify the specificity of the AKT and pAKT antibodies, we performed Western blot analysis comparing lysate from mouse thyroids with cell lysate prepared from MCF-7 cells serum starved for 16 h, before and after a 30-min incubation of 100 ng/ml IGF-I (Pepro-Tech, Rocky Hill, NJ).
Secondary antibodies used were horseradish peroxidase-conjugated goat antimouse or antirabbit IgG (Amersham Biosciences, Piscataway, NJ) and detected using the Western Lightning chemiluminescence reagent plus system (PerkinElmer Life Sciences, Boston, MA). The blots were stripped with Re-Blot Plus (Chemicon, Temecula, CA) and reprobed with rabbit polyclonal antibodies to protein disulfide isomerase at 0.5 μg/ml (Sigma, St. Louis, MO). Three independent experiments were performed using male age-matched mice and one with female age-matched TRPV/PV and wild-type mice.
Immunohistochemistry
Histological sections were prepared as previously described (19). Phosphorylated AKT (Ser473, immunohistochemistry specific) no. 9277 (Cell Signaling Technologies), AKT1 sc-5298 (Santa Cruz Biotechnology), AKT2 06-606 (Upstate Biotechnology, Lake Placid, NY), AKT3 sc-11520 (Santa Cruz Biotechnology), and AKT no. 9272 (Cell Signaling Technologies) were used. Vectastain Universal Quick kit, antigen unmasking solution, and 3,3'-diaminobenzidine substrate kit were purchased from Vector Laboratories (Burlingame, CA). Negative controls not including primary antibody were performed for each experiment. Five low- and five high-power fields were independently examined by three investigators (M.D.R., M.S., and V.V.V.), including one pathologist (V.V.V.). Figures are representative of those examinations.
Confocal microscopy
Confocal images of thyroid and lung tissue samples from both wild-type and TRPV/PV mice were collected using a Zeiss LSM 510 microscope equipped with a META spectral detector (Carl Zeiss Inc., Jena, Germany). Primary antibodies used include AKT1 sc-5298 (Santa Cruz Biotechnology), AKT2 06–606 (Upstate Biotechnology), AKT3 sc-11520 (Santa Cruz Biotechnology), and phospho-AKT (Ser473, immunohistochemistry specific) no. 9277 (Cell Signaling Technologies). Images were collected as previously described (19). Briefly, images from tissue samples that were unlabeled, from Alexafluor 488-conjugated secondary antibody alone (Molecular Probes Inc, Eugene, OR) or tissue samples immunolabeled with primary and secondary antibody, were collected using the same acquisition configuration. Using the Alexafluor 488 antibody-only sample and unlabeled specimens, reference spectra were calculated at regions of interest to determine the emission signatures for Alexafluor 488 and background autofluorescence of the samples. Separation of specific fluorescence emission signal was obtained using a linear unmixing algorithm included in the Zeiss operating software (version 3.0). The resultant images were exported as TIF files and prepared as figures using Adobe Photoshop (version 5.02; Adobe Systems Inc., San Jose, CA).
Primary thyroid culture
Using a method modified from a previously described protocol (24), thyroid tissue was placed in sterile PBS and minced using a razor blade. The tissue was spun at 950 x g for 5 min at 4 C and the supernatant was removed. The tissue was then placed in a solution containing MEM Eagle salts (Sigma), 112 U type I collagenase (Sigma), and 1.2 U/ml Dispase 1 (Roche) and incubated in a 37 C shaker for 2 h. The tissue was then sheared using a 19-gauge then 21-gauge needle (10 times). The tissue was centrifuged at 1500 x g for 5 min, and the supernatant was removed. The cells were then washed and centrifuged in primary cell media. The culture medium was prepared as follows. NuSerum 4 (BD Biosciences, Bedford, MA) was diluted 2.5 times with Ham’s F-12 media (Sigma). This diluted NuSerum 4 was supplemented with somatostatin (Sigma) at a concentration of 10 ng/ml, glycyl-L-histidyl-L-lysine acetate (Sigma) at a concentration of 2 ng/ml, TSH (Sigma) at a concentration of 5 mIU/ml, and antibiotic-antimycotic (Gibco, Invitrogen, Carlsbad, CA) at a concentration of 200 U/ml. The cells were then plated and grown at 37 C in a 5% CO2, 90% humidity incubator.
Cell motility assay
The method used was modified from the original protocol described previously (25). Motility assays were performed in 8-μm-pore transwells (6.5 mm; Costar, Corning, NY) in triplicate. Primary cells were detached by trypsin, and 5 x 104 cells were placed in 200 μl Ham’s F-12 media. Bottom wells contained 500 μl of the primary cell media. Cells were incubated at 37 C for 2 h. The wells were then decanted and the cells fixed in 1% glutaraldehyde in PBS (Sigma) and stained with 0.1% crystal violet (Sigma) in water for 30 min. The wells were destained by rinsing in deionized water. Using separate cotton-tipped applicators soaked in 0.2% Triton X-100 (Sigma), both the nonmigrating cells on the upper side of the well and the migrating cells on the lower side were removed and placed in 1 ml of 0.2% Triton X-100. These samples were kept at 4 C overnight. The solution of each sample was measured at A590 using a spectrophotometer. Rate of motility was calculated by the absorbance of the migrating cells divided by the absorbance of the nonmigrating cells.
PI3-K inhibition
Primary TRPV/PV thyroid cells were grown in primary cell media until 50–60% confluent and then incubated in 10 μM LY294002 (Thomas Scientific, Swedesboro, NJ) or continuous media for 24 h. Afterward the cells were harvested for Western blot analysis or cell motility assay as described above.
Adenovirus infection
Primary TRPV/PV thyroid cells were seeded at 1.5 x 105 overnight in 60-mm dishes. The following morning cells were infected with replication-defective adenoviral constructs consisting of either an HA-tagged dominant-negative mutant of rat AKT1 (DNAKT) under the control of the CAG eukaryotic promoter or containing the gene for -galactosidase (-gal) (26, 27). These viruses were previously shown to infect mouse primary smooth muscle cells (28). Earlier determination of multiplicity of infection (MOI) was carried out using primary thyroid cells infected with various titers of the -gal adenovirus for 4 h and then 48 h later, staining for -gal using the -gal staining kit (Roche). Cells infected at a MOI of 100 for 4 h at 37 C resulted in more than 95% infection based on -gal staining. After incubation of primary thyroid cells with either the adenovirus for -gal or DNAKT for 4 h, media were replaced with fresh primary cell media. Forty-eight hours after infection, the cells were harvested for either Western blot analysis or cell motility assay as described above.
Statistical analysis
Data are expressed as mean ± SEM. Cell motility experiments were analyzed by unpaired t test with StatView 5.0 (Abacus Concepts, Inc., San Diego, CA) where P < 0.05 was considered significant.
Results
Overexpression of pAKT in the thyroids of TRPV/PV mice
TRPV/PV mice spontaneously develop FTC with distant metastasis similarly to human FTC. The apparent role of AKT overexpression and overactivation in human FTC development and behavior suggests that AKT activity may also have a role in our model of FTC (19). To evaluate this possibility, we performed Western blots for total AKT, pAKT, and a control protein from age- and gender-matched wild-type and TRPV/PV mice at different ages. Representative results are shown in Fig. 1. We found, similar to human FTC, that TRPV/PV thyroid glands are characterized by AKT overactivation, compared with gender- and age-matched wild-type thyroids (Fig. 1). As TRPV/PV mice age, the incidence of FTC and distant metastasis increases (15, 16). Whereas at 3 months of age, all TRPV/PV mice develop thyroid hyperplasia, they may not have yet developed FTC (15). In Fig. 1, the increased level of pAKT is marked by 6 months of age and persists through 11 months. At the age of 11 months, the expression of total AKT protein was increased as compared with wild-type mice. The band intensities were quantified and analysis of pAKT to AKT ratios indicate an age-dependent increase of pAKT/AKT as TRPV/PV mice age (Fig. 1B), suggesting an increase activity of AKT as thyroid carcinogenesis progresses. The increased pAKT expression in thyroid tumor cells rather than associated stromal cells was further confirmed by immunohistochemistry. Compared with wild-type mice (Fig. 2A), strong immunoactive staining of pAKT was detected in thyroid cancer cells of TRPV/PV mice (Fig. 2B, shown by arrows). Thus, the abundance of pAKT protein is elevated in thyroid cancer cells of TRPV/PV mice.
In contrast to the overexpression and overactivation of AKT in the primary and metastatic thyroid tumor cells of TRPV/PV mice, another major signaling pathway, the ERK-MAPK pathway, which could also be involved in thyroid tumorigenesis and can influence cell motility (29), was not similarly overexpressed or overactivated in the primary tumors or metastases (data not shown). These findings are consistent with our past observations that in human FTC, we did not find evidence of enhanced activation of MAPK in invasive vs. normal cells (19). Thus, increased activation of the Ras-ERK-MAPK pathway, unlike the PI3-K/AKT signaling pathway, is not associated with progression and metastasis of follicular thyroid cancer in TRPV/PV mice. These data demonstrate that thyroid tumorigenesis in the TRPV/PV mice is associated with enhanced activation of AKT.
Expression of pAKT in the metastatic lesions of TRPV/PV mice
To determine whether pAKT levels may also be correlated with metastases, the metastatic lung lesions from TRPV/PV mice were isolated and evaluated for pAKT staining by immunohistochemistry. As shown in Fig. 2D, the metastatic lesions in the lung retain high levels of pAKT. The wild-type lung section does not contain any metastatic lesions and reveals that there is little background pAKT in normal lung tissue (Fig. 2C).
Overexpression of all AKT isoforms in the thyroid cancers of TRPV/PV mice
There are tissue-specific expression patterns of the three isoforms of AKT. AKT1 is present in almost all tissues at high levels, whereas high levels of AKT2 are observed in the muscle and reproductive organs, and AKT3 is highly expressed in the brain and testis (30, 31, 32, 33). Whereas the AKT isoforms demonstrate substrate specificity, it is also thought that there is functional redundancy (34) and that subcellular localization of AKT may also play a role in determining its downstream effects. As shown above, as the TRPV/PV mice age, thyroid cancer develops, and there is overexpression of total AKT and overactivity of AKT. We evaluated the expression profiles of the three AKT isoforms in the thyroids of older (i.e. > 10 months old) TRPV/PV mice by immunohistochemistry using antibodies that are isoform specific but recognize both activated phosphorylated and inactive forms. We found that all isoforms of AKT are overexpressed in the thyroid cancers of older TRPV/PV mice (Fig. 3, A1, B1, and C1;) in comparison with age-matched wild-type controls (Fig. 3, A, B, and C, respectively). These findings are similar to human FTC in which AKT1 and AKT2 were overexpressed, compared with normal thyroid tissue, and AKT3 results were more variable (20).
Subcellular localization of AKT isoforms and pAKT in primary FTC lesions
It has been reported that whereas the majority of AKT is located in the cytoplasm, AKT is also translocated into the nucleus in which it interacts with specific nuclear targets, suggesting that subcellular localization of AKT may be an important determinant of biological activity (34). Previous work associated higher levels of nuclear pAKT in human FTC cells that were in invasive regions of primary tumors on the periphery of the tumor mass and that AKT1 colocalized with pAKT in those regions in comparison with AKT2 or AKT3 (19). Figure 3A1 shows that AKT1 is localized to both the nucleus and cytoplasm in the primary thyroid tumors (shown by an arrow), whereas AKT2 and AKT3 are predominantly cytoplasmic (Fig. 3, B1 and C1, respectively, open arrows). Immunoactive pAKT was localized to both compartments (Fig. 3, D1). In contrast, in metastasis, all AKT isoforms and immunoactive pAKT localized primarily to the nucleus (Fig. 3, A2, B2, C2, D, and E). These immunohistochemical data were confirmed by subsequent confocal immunofluorescence studies (Fig. 4, A and B). Figure 4A, a and c, highlights AKT1’s presence in both the nucleus and cytoplasm in the primary tumors (arrowhead and arrows, respectively), whereas AKT2 and AKT3 remain in the cytoplasm (Fig. 4A, f and i). In metastasis, activated AKT is predominantly localized to the nucleus (Fig. 4B, a and c).
Reduced motility is associated with inhibition of PI3-K/AKT signaling pathway
To determine whether the PI3-K/AKT pathway might play a functional role in TRPV/PV metastasis, we created primary cultures from the thyroid tumors of the TRPV/PV mice. Using Boyden chamber migration assays, we evaluated cell motility in the presence or absence of PI3-K inhibitor, LY294002. Figure 5A shows that in the presence of LY294002 (10 μM), the extent of activation of pAKT was reduced 50–60% as compared with the control. Concomitant with the reduction of pAKT, cell motility was also reduced approximately 57% (Fig. 5B). To confirm the critical role of AKT in motility in these primary cells, we transiently infected cells with adenovirus containing either a HA-tagged, dominant-negative AKT or -gal as a control. This mutant AKT1 has alanine substituted at the known phosphorylation sites (Thr308 and Ser473), resulting in the loss of its kinase activity (26). Kitamura et al. (26) also found that this construct exhibits dominant-negative activity; infection of Chinese hamster ovary and 3T3-L1 adipocyte cells resulted in loss of insulin-stimulated endogenous AKT activity. Stabile et al. (28) showed that primary mouse smooth muscle cells infected with this dominant-negative AKT have a reduction in AKT activity as shown by a decrease in phosphorylation of a downstream target, p70 S6K. As inhibition of PI3K activity reduces primary thyrocyte motility, the selective reduction of AKT activity, via transient expression of this dominant-negative AKT, should also show similar effects on motility. Figure 6A shows that infection of the primary cells with the DNAKT have decreased phosphorylation of two downstream AKT targets, mTOR and p70 S6K, consistent with inhibition of AKT activity. Phosphorylated AKT levels were not evaluated because the mutant AKT cross-reacts with pAKT antibodies (personal communication, Dr. W. Ogawa, Kobe University, Japan). The resultant decrease in AKT activity is associated with decreased cell motility (Fig. 6B). Taken together, these data suggest that the association between enhanced AKT activation and metastasis in the TRPV/PV mice may be functionally important.
Discussion
The mechanism for FTC metastasis is not well understood but may include signaling pathways already implicated in promoting metastasis in other cancers. The PI3-K/AKT pathway is a logical point of investigation as patients with the autosomal dominant Cowden’s syndrome who have mutations in PTEN, resulting in AKT overactivation, develop FTC (35). However, it is often difficult to obtain sufficient quantities of human metastatic FTC tissues for profiling the genetic differences between the primary and metastatic lesion. The recent creation of the TRPV/PV mouse model that spontaneously develops FTC with metastasis provides an opportunity to elucidate the pathways contributing to the metastatic phenotype.
Prior work has shown human FTC to have elevated pAKT levels, compared with normal tissue (17, 18), and also demonstrate nuclear localization of pAKT the more invasive regions of the tumor (i.e. periphery) in association with AKT1 localization (19). Functional data have demonstrated that human thyroid cancer cell motility is dependent on PI3-K activity (19). In this study, it is striking how similar the patterns of AKT activation and localization in primary thyroid cancers in TRPV/PV mice are to human FTC. Specifically, pAKT and AKT1’s presence in both the cytoplasm and nucleus in primary thyroid tumors in TRPV/PV mice resembles human FTC. The lack of enhanced ERK activation in the thyroid cancers of our mouse model also parallels human FTC. The association between nuclear AKT activity and metastasis in this model is strengthened by the greater predilection for pAKT and all three AKT isoforms to localize to the nucleus in the metastatic thyroid cancer cells. Finally, a functional role for PI3-K/AKT signaling pathway in metastasis in this model is strongly suggested by the in vitro data demonstrating that pharmacological inhibition of PI3K activity and specific inhibition of AKT both result in the reduction of TRPV/PV thyroid tumor cell motility. These results suggest that both AKT activity in general as well as its nuclear localization may play an important role in the development of FTC metastasis in TRPV/PV mice.
That the PI3-K/AKT signaling pathway is activated in the thyroids of TRPV/PV mice harboring a dominant-negative TR mutant is in contrast of recent findings by Cao et al. (20). Whereas T3 is considered to exert its effects via TR and at the level of transcription, Cao et al. showed that T3 could activate the PI3-K/AKT pathway via nongenomic effect. These authors found that in a human fibroblast system, viral-mediated overexpressed TR1 is coimmunoprecipitated with p85, the regulatory subunit of PI3-K (20). mTOR, a downstream effector of AKT, is also activated in the presence of T3. PI3-K inhibitors reduced T3 activation of mTOR, suggesting that the PI3-K/AKT pathway is responsible for mTOR activation by T3. It was further shown that a dominant-negative TR mutant, G345R, although able to bind to p85, was unable to activate the PI3-K/AKT signaling (20).
Whereas the molecular mechanisms by which the PV mutant mediates the activation of PI3-K/AKT signaling pathway in TRPV/PV mice is currently being elucidated in our laboratory, there are clear differences between the systems reported by Cao et al. (20) and the present study. First, Cao et al. used cultured fibroblasts in which TR1 or TR1G345R is overexpressed, whereas the present study is an in vivo system with physiological levels of TRs or PV. The PV mutation in TRPV mice is a targeted knockin mutation; hence, its regulation of expression is the same as the wild-type TR gene (22). It is reasonable to expect that the regulation and functions of TRs and TR mutants at the overexpressed levels could differ from those at the physiological level. Second, TR mutations are different in these two studies. The PV mutation is a frameshift mutation in the C-terminal 16 amino acids, whereas TR1G345R is a point mutation. There are precedents to suggest that different mutations of the TR lead to different functional consequences. Mice that harbor TR1337T display neurological phenotype (36), whereas TRPV mice have no such abnormalities. TRPV/PV mice developed follicular thyroid carcinoma (15, 16) and pituitary adenomas (37), but no such defects are reported in TR1337T knockin mice. Thus, it is possible that PV and TR1G345R affect differently on the PI3-K/AKT signaling pathway. Clearly, this issue would need to be clarified in future studies.
The consistent pattern of AKT activation and predilection for nuclear localization of AKT1 in primary tumors and all AKT isoforms in metastatic lesions from TRPV/PV mice suggests that FTC metastasis may be in part dependent on altered activity and cellular localization of the PI3-K/AKT signaling pathway. Whereas AKT1 and AKT2 are more ubiquitously expressed, compared with AKT3, the percentage of functional overlap among the isoforms is unknown. There appears to be differing levels of AKT isoform activation, depending on the type of tumor. Dufour et al. (38) reported that AKT isoforms have distinct functional consequences for survival of human intestinal epithelial cells along with different upstream activating signals. Whether the activation of all AKT isoforms in the primary thyroid lesion may provide a future diagnostic approach to distinguish benign vs. malignant tumor is unknown.
The predominant nuclear localization of all AKT isoforms and pAKT in the metastatic FTC lesions suggests that its nuclear distribution may be relevant to both initiating and sustaining metastasis. Nuclear AKT is known to affect many proteins, including the Forkhead box, class O (FOXO) family proteins, cAMP response element binding protein, and p21Cip/WAF1 (39, 40, 41). In particular, AKT phosphorylation inactivates the FOXO subfamily of forkhead transcription factors, leading to decreased levels of proapoptotic genes (42). Identification of the downstream targets of AKT that promote cell motility, migration, and angiogenesis is critical for not only the understanding of the molecular basis underlying metastasis of FTC but also the identification of potential molecular targets for treatment of FTC.
It is important to recognize that the nuclear changes associated with tumorigenesis in these tumors may promote a more nuclear appearance of immunohistochemical staining. However, the confocal microscopic studies’ more precise ability to detect differences in subcellular localization of AKT isoforms and pAKT between the primary cancer cells and metastatic cancer cells corroborate the immunohistochemistry results. The specific mechanisms responsible for the nuclear predominant localization of activated AKT of the metastatic thyroid cancer cells are uncertain. It is possible that the localization is intrinsically regulated by alterations in the tumor cells; it is also possible that interactions between the tumor cells and the microenvironment regulate localization.
In summary, in the present paper, we have demonstrated that that the PI3-K/AKT signaling pathway is activated similarly in the TRPV/PV mouse model in comparison with human FTC. Its activation is associated with tumor formation and nuclear localization of immunoactive pAKT is associated with tumor progression. In vitro, TRPV/PV primary thyrocyte motility is dependent on PI3-K and AKT signaling. Taken together, these data suggest that the TRPV/PV mouse model could be used to further dissect the pathways and identify key players that contribute to the metastasis of FTC.
Footnotes
This research was supported in part by the Intramural Research Program of the National Institutes of Health, National Cancer Institute, Center for Cancer Research.
Abbreviations: DNAKT, Dominant-negative mutant of rat AKT1; FTC, follicular thyroid cancer; -gal, -galactosidase; HA, hemagglutinin; MOI, multiplicity of infection; mTOR, mammalian target of rapamycin; pAKT, phosphorylated AKT; PI3-K, phosphatidylinositol 3-kinase; PTEN, phosphatase and tensin homolog deleted from chromosome 10; TR, thyroid hormone receptor.
References
Bellacosa A, Testa JR, Staal SP, Tsichlis P 1991 A retroviral oncogene, akt, encoding a serine-threonine kinase containing an SH2-like region. Science 254:274–277
Coffer PJ, Jin J, Woodgett JR 1998 Protein kinase B (c-AKT): a multifunctional mediator of phosphatidylinositol 3-kinase activation. Biochem J. 335:1–13
Franke TF, Hornik CP, Segev L, Shostak GA, Sugimoto C 2003 PI3K/AKT and apoptosis: size matters. Oncogene. 22:8983–8998
Feng J, Park J, Cron P, Hess D, Hemmings BA 2004 Identification of a PKB/Akt hydrophobic motif Ser-473 kinase as DNA-dependent protein kinase. J Biol Chem. 279:41189–41196
Luo J, Manning BD, Cantley, LC 2003 Targeting the PI3K-AKT pathway in human cancer: rationale and promise. Cancer Cell 4:257–262
Hanahan D, Weinberg RA 2000 The hallmarks of cancer. Cell 100:57–70
Park BK, Zeng X, Glazer RI 2001 Akt1 induces extracellular matrix invasion and matrix metalloproteinase-2 activity in mouse mammary epithelial cells. Cancer Res. 61:7647–7653
Kim D, Kim S, Koh H, Yoon SO, Chung AS, Cho KS, Chang J 2001 Akt/PKB promotes cancer cell invasion via increased motility and metalloproteinase production. FASEB J. 15:1953–1962
Jiang BH, Zheng JZ, Aoki M, Vogt PK 2000 Phosphatidylinositol 3-kinase signaling mediates angiogenesis and expression of vascular-endothelial growth factor in endothelia cells. Proc Natl Acad Sci USA. 97:1749–1753
Fujita S, Watanabe M, Kubota T, Teramoto T, Kitajima M 1995 Alteration of expression in integrin 1-subunit correlates with invasion and metastasis in colorectal cancer. Cancer Lett. 91:145–149
Nikiforova MN, Biddinger PW, Caudill CM, Kroll TG, Nikiforov YE 2002 PAX8 PPAR rearrangement in thyroid tumors: RT-PCR and immunohistochemical analyses. Am J Surg Pathol. 26:1016–1023
Cheng SY 2003 Thyroid hormone receptor mutations in cancer. Mol Cell Endocrinol. 213:23–30
Sap J, Munoz A, Damm K, Goldberg Y, Ghysdael J, Leutz A, Beug H, Vennstrom B 1986 The c-erb-A protein is a high-affinity receptor for thyroid hormone. Nature 324:635–640
Barlow C, Meister B, Lardelli M, Lendahl U, Vennstrom B 1994 Thyroid abnormalities and hepatocellular carcinoma in mice transgenic for v-erbA. EMBO J. 13:4241–4250
Suzuki H, Willingham MC, Cheng SY 2002 Mice with a mutation in the thyroid hormone receptor gene spontaneously develop thyroid carcinoma: a mouse model of thyroid carcinogenesis. Thyroid 12:963–969
Kato Y, Ying H, Willingham MC, Cheng SY 2004 A tumor suppressor role for thyroid hormone receptor in a mouse model of thyroid carcinogenesis. Endocrinology 145:4430–4438
Ringel MD, Hayre N, Saito J, Saunier B, Schuppert F, Burch H, Bernet V, Burman KD, Kohn LD, Saji M 2001 Overexpression and overactivation of Akt in thyroid carcinoma. Cancer Res. 61:6105–6111
Miyakawa M, Tsushima T, Murakami H, Wakai K, Isozaki O, Takano K 2003 Increased expression of phosphorylated p70S6 kinase and Akt in papillary thyroid cancer tissues. Endocr J. 50:77–83
Vasko V, Saji M, Hardy E, Kruhlak M, Larin A, Savchenko V, Miyakawa M, Isozaki O, Murakami H, Tsushima T, Burman KD, De Micco C, Ringel MD 2004 Akt activation and localization correlate with tumour invasion and oncogene expression in thyroid cancer. J Med Genet. 41:161–170
Cao X, Kambe F, Moeller LC, Refetoff S, Seo H 2005 Thyroid hormone induces rapid activation of AKT/protein kinase B-mammalian target of rapamycin-p70s6K cascade through phosphatidylinositol 3-kinase in human fibroblasts. Mol Endocrinol. 19:102–112
Saito J, Kohn AD, Roth RA, Noguchi Y, Tatsumo I, Hirai A, Suzuki K, Kohn LD, Saji M, Ringel MD 2001 Regulation of FRTL-5 thyroid cell growth by phosphatidylinositol (OH) 3 kinase-dependent Akt-mediated signaling. Thyroid 11:339–351
Kaneshige M, Kaneshige K, Zhu X-G, Dace A, Garrett L, Carter TA, Kazlauskaite R, Pankratz DG, Wynshaw-Boris A, Refetoff S, Weintraub B, Willingham MC, Barlow C, Cheng SY 2000 Mice with a targeted mutation in the thyroid hormone receptor gene exhibit impaired growth and resistance to thyroid hormone. Proc Natl Acad Sci USA. 97:13209–13214
Heaney AP, Singson R, McCabe, CJ, Nelson V, Nakashima M, Melmed S 2000 Expression of pituitary-tumour transforming gene in colorectal tumours. Lancet 355:716–719
Jeker LT, Hejazi M, Burek CL, Rose NR, Caturegli, P 1999 Mouse thyroid primary culture. Biochem Biophys Res Commun. 257:511–515
Yu Y, Merlino G 2002 Constitutive c-met signaling through an autocrine mechanism promotes metastasis in a transgenic transplantation model. Cancer Res. 62:2951–2956
Kitamura T, Ogawa W, Sakaue H, Hino Y, Kuroda S, Takata M, Matsumoto M, Maeda T, Konishi H, Kikkawa U, Kasuga M 1998 Requirement for activation of the serine-threonine kinase Akt (protein kinase B) in insulin stimulation of protein synthesis but not of glucose transport. MCB. 18:3708–3717
Takata M, Ogawa W, Kitamura T, Hino Y, Kuroda S, Kotani K, Klip A, Gingras A-C, Sonenberg N, Kasuga M 1999 Requirement for Akt (protein kinase B) in insulin-induced activation of glycogen synthase and phosphorylation of 4E-BP1 (PHAS-1). J Biol Chem. 274:20611–20618
Stabile E, Zhou YF, Saji M, Castagna M, Shou M, Kinnaird TD, Baffour R, Ringel MD, Epstein SE, Fuchs S 2003 Akt controls vascular smooth muscle cell proliferation in vitro and in vivo by delaying G1/S exit. Circ Res. 93:1059–1065
Janda E, Lehmann K, Killisch I, Jechlinger M, Herzig M, Downward J, Beug H, Grunert S 2002 Ras and TGF() cooperatively regulate epithelia cell plasticity and metastasis: dissection of Ras signaling pathways. J Cell Biol. 156:299–313
Coffer PJ, Woodgett JR 1991 Molecular cloning and characterisation of a novel putative protein-serine kinase related to the cAMP-dependent and protein kinase C families. Eur J Biochem. 201:475–481
Kim YB, Peroni OD, Franke TF, Kahn BB 2000 Divergent regulation of Akt1 and Akt2 isoforms in insulin target tissues of obese Zucker rats. Diabetes 49:847–856
Nakatani K, Sakaue H, Thompson DA, Weigel RJ, Roth R 1999a Identification of a human Akt3 (protein kinase B) which contains the regulatory serine phosphorylation site. Biochem Biophys Res Commun. 257:906–910
Franke TF 2000 Assays for AKT. Methods Enzymol. 322:400–410
Ahmed NN, Franke TF, Baellacosa A, Datta K, Gonzalez-Portal ME, Taguchi T, Testa JR, Tsichlis PN 1993 The proteins encoded by c-akt and v-akt differ in post-translational modification, subcellular localization and oncogenic potential. Oncogene. 8:1957–1963
Eng C 2003 PTEN: one gene, many syndromes. Hum Mutat. 22:183–198
Hashimoto K, Curty FH, Borges PP, Lee CE, Abel ED, Elmquist JK, Cohen RN, Wondisford FE 2001 An unliganded thyroid hormone receptor causes severe neurological dysfunction. Proc Natl Acad Sci USA. 98:3998–4003
Furumoto H, Ying H, Chandramouli GV, Zhao L, Walker RL, Meltzer PS, Willingham MC, Cheng SY 2005 An unliganded thyroid hormone receptor activates the cyclin D1/cyclin-dependent kinase/retinoblastoma/E2F pathway and induces pituitary tumorigenesis. Mol Cell Biol. 25:124–135
Dufour G, Demers M-J, Gagne D, Dydensborg AB, Teller IC, Bouchard V, Degongre I, Beaulieu J-F, Cheng JQ, Fujita N, Tsuruo T, Vallee K, Vachon PH 2004 Human intestinal epithelial cell survival and anoikis: differentiation state-distinct regulation and roles of protein kinase B/Akt isoforms. J Biol Chem. 279:44113–44122
Brunet A, Bonnin A, Zigmond MJ, Lin MZ, Juo P, Hu LS, Anderson MJ, Arden KC, Blenis J, Greenberg ME 1999 Akt promotes cell survival by phosphorylating and inhibiting a Forkhead transcription factor. Cell 96:857–868
Du K, Montminy M 1998 CREB is a regulatory target for the protein kinase Akt/PKB. J Biol Chem. 273:32377–32379
Shioi T, Kang PM, Douglas PS, Hampe J, Yballe CM, Lawitts J, Catley LC, Izumo S 2000 The conserved phosphoinositide 3-kinase pathway determines heart size in mice. EMBO J. 19:2537–2548
Gilley J, Coffer PJ, Ham J 2003 FOXO transcription factors directly activate bim gene expression and promote apoptosis in sympathetic neurons. J Cell Biol. 162:613–622(Caroline S. Kim, Vasily V)