|Home | About | Journals | Submit | Contact Us | Français|
Mitogen-inducible gene 6 (Mig-6) is a putative tumor suppressor gene and prognostic biomarker in papillary thyroid cancer. We hypothesized that Mig-6 knockout would activate pro-oncogenic signaling in mouse thyrocytes.
We performed a thyroid-specific knockout using the Cre/loxP recombinase system.
Four knockout and 4 control mouse thyroids were harvested at 2 months of age. Immunoblotting confirmed Mig-6 ablation in knockout mice thyrocytes. Epidermal growth factor receptor (EGFR) and extracellular signal-regulated kinase (ERK) phosphorylation levels were increased in Mig-6 knockout compared to wild-type mice. Total EGFR levels were similar in knockout and wild-type mice. However, EGFR was absent in the caveolae-containing membrane fraction of knockout mice, indicating that Mig-6 depletion is associated with a change in the membrane distribution of EGFR. Although p65 localized to the nucleus in wild-type mice, it was distributed in both cytoplasm and nucleus in knockouts, suggesting that Mig-6 loss decreases p65 activity.
Our results confirm the feasibility of targeted, thyroid-specific gene knockout as a strategy for studying the relevance of specific genes in thyroid oncogenesis. We suggest that the loss of Mig-6 alters the membrane distribution of EGFR, which may limit receptor degradation and activate this oncogenic signaling pathway.
Mitogen-inducible gene 6 (Mig-6) is a multifunctional adaptor protein that inhibits epidermal growth factor receptor (EGFR) activity,1 directly down-regulates extracellular signal-regulated kinase (ERK) signaling,2 and up-regulates nuclear factor-κB (NF-κB) activity.3 Mig-6 levels are consistently lower in well-differentiated papillary thyroid cancers in comparison to normal matched thyroid tissue.4 In addition, Mig-6 knockdown in thyroid cancer cell lines results in increased levels of phospho-EGFR, and independently, increased phopho-ERK.4 Low Mig-6 expression is also associated with an increased risk of recurrence after thyroidectomy for papillary thyroid cancer.5 Altogether, the available body of evidence suggests that Mig-6 is a relevant tumor suppressor in thyroid cancer of follicular origin. However, no previous studies have investigated the loss of Mig-6 in thyrocytes in vivo.
Mig-6 germline null allele (Mig-6−/−) mice develop many abnormalities of the skin and solid organs. A 50% reduction in litter size is observed in newborn Mig-6−/− mice, and the remaining survivors typically die by the 6 months of age.6 Mig-6−/− mice suffer from severe joint deformities and tumorigenesis in various organs including the lung, gallbladder, bile ducts, uterus, endometrium, and gastrointestinal tract.2,7,8 However, previous studies have not examined the effects of Mig-6 knockout in the thyroid. In addition, embryonic lethality and multiorgan carcinogenesis in Mig-6−/− mice precluded characterization of the effects of Mig-6 loss in the thyroid.6,7,9 Therefore, thyroid-specific Mig-6 ablation is required to study the role of Mig-6 loss in vivo.
Homologous recombination of genes in embryonic stem cells is a useful tool for deletion of targeted genes in vivo. However, this system results in gene deletion in all tissues of the body. The Cre/loxP system is a powerful tool that allows gene deletion in a particular tissue or cell type. This system uses the site-specific recombinase Cre, which recognizes consensus loxP target sequences and excises any sequence placed between 2 loxP sites of the same relative orientation. By crossing a transgenic mouse expressing Cre under a cell-specific promoter with a mouse containing a gene flanked by loxP sites, one can obtain an animal with a cell-specific knockout of the particular gene of interest. This technique is useful tool to avoid embryonic lethality and multiorgan carcinogenesis.
We established thyroid-specific knockout transgenic mice (Mig-6flox/flox; TPO-Cre+) by breeding Mig-6flox/flox mice with mice expressing the Cre recombinase under the control of the human thyroid peroxidase (TPO) promoter. TPO is expressed exclusively in the thyroid when the mouse thyroid enters the final steps of differentiation.10 Target genes flanked by loxP sequences can be ablated in a thyroid-specific manner using a TPO-Cre transgenic mouse.11 We hypothesized that Mig-6 ablation in vivo would activate pro-oncogenic signaling in thyrocytes.
Mice were housed and maintained at the animal facility of Brigham and Women’s Hospital according to the institutional guidelines for the care and use of laboratory animals. Mig-6 floxed (Mig-6f/f) transgenic mice, harboring 2 loxP sites flanking Mig-6 exons 2 and 4 on a C57BL/6 background.12 One loxP site was inserted downstream of exon 4. The other loxP site was inserted upstream of exon 2. The 5′ ends of the targeting vector were constructed by cloning 2 kb of Mig-6 homologous sequence upstream of Mig-6 exon 2, whereas the 3′ end contains a 2-kb homologous sequence downstream of Mig-6 exon 4. The linearized targeting construct was transfected into R1 embroynic stem cells by electroporation. Correctly targeted clones were microinjected into blastocysts derived from C57BL/6 mice. Offspring mice were analyzed by polymerase chain reaction (PCR) genotyping to validate the germline transmission of the Mig-6–floxed allele. Genotyping was performed by PCR using primers: 5′-GGTCAGGGCTGTGCAG TCCGTAGA-3′ (Mig-6-F), 5′-CGATACCCCACCGA GACC-3′ (Neo-R), 5′-CTTCCCAAATCTAACACCC GACAC-3′ (Mig-6-R), yielding 460-bp and 212 bands for Mig-6f/f and wild-type mice, respectively. Transgenic mice in which the TPO promoter drives expression of Cre recombinase (TPO-Cre+) on a nearly pure FVB/N background were previously described.11 In brief, to establish the TPO-Cre transgenic construct, DNA 6.3 kb upstream of the human TPO gene was connected to the Cre recombinase gene and the construct was then flanked at both ends by 2 sets of insulator DNA. The linearized transgenic construct DNA was microinjected into pronuclei of mouse embryos, resulting in the production of 2 TPO-Cre transgenic founder lines. Northern Blots were performed to confirm the expression patterns for Cre recombinase, and therefore the line was chosen as TPO-Cre+ mouse line. Genotyping was performed by PCR using primers 5′-AGGTGTAGAGAAGGCACTTAGC-3′ and 5′-CTAATCGCCATCTTCCAGCAGG-3′, yielding a 412-bp band for TPO-Cre+ mice. Genotyping was performed by PCR as previously described.12 Mice of the genotype Mig-6d/d (Mig-6f/f; TPO-Cre+) were obtained by breeding Mig-6f/f and TPO-Cre+ mice. Mice with various genotypes were killed at 6 and 8 weeks of age. Thyroid tissue specimens were resected from mice, flash frozen, and then stored at −80°C. Tissue specimens were also fixed in 10% neutral-buffered formalin and embedded in paraffin. Contiguous sections were cut at 4 μm for hematoxylin–eosin staining and immunostaining with antibodies against human thyroid transcription factor-1 (TTF-1) and Ki-67. Hematoxylin–eosin and TTF-1 staining confirmed that the resected specimens were thyroid tissue.
Anti-mouse Mig-6 antibody was obtained from Sigma-Aldrich (St. Louis, MO). Anti-mouse EGFR, phospho-ERK, and clathrin antibodies were obtained from Cell Signaling (Beverly, MA). Anti-mouse ERK1/2 was purchased from Upstate Biotechnology (Lake Placid, NY). Anti-mouse Caveolin-1, NF-κB subunit p65, Hsp90, and OCT-1 were obtained from Santa Cruz Biotechnology (Santa Cruz, CA). Anti-actin antibody pan Ab-5 was purchased from Neomarker (Fremont, CA). Anti-phospho-tyrosine antibody was obtained from Calbiochem (San Diego, CA). Secondary horseradish peroxidase-conjugated anti-rabbit and -goat antibodies were purchased from Vector Laboratories (Burlingame, CA).
Immunoprecipitation was performed as previously described.13 Immunoprecipitated products were prepared and processed for Western blot analysis using a 10% sodium dodecyl sulfate polyacrylamide gel electrophoresis–separating gel under reduced conditions as previously described.4
Caveolae-containing membrane was isolated using detergent-free methods as previously described.14 In brief, mouse thyroid specimens were homogenized and sonicated in ice-cold buffer B (20 mM tricine pH 7.8, 1 mM ethylenediaminetetraacetic acid, and 250 mM sucrose) containing 1 mg/mL leupeptin, 1 mg/mL pepstatin A, 1 mg/mL soybean trypsin inhibitor, and 1 mg/mL benzamidine. A postnuclear supernatant fraction was prepared by spinning the lysate at 90 g for 10 minutes. The postnuclear supernatant was layered over 23 mL of ice-cold 30% Percoll in buffer B. After centrifugation at 85,000 g for 30 minutes, the cytosol and plasma membrane fractions were collected. The plasma membrane fraction was briefly sonicated, mixed with buffer C (50% w/v iodixanol in buffer B plus 40 mM sucrose) to a final iodixanol concentration of 23% and overlaid with 6 mL of linear (10–20%) gradient of iodixanol in buffer B. Samples were centrifuged at 53,000 g for 90 minutes. The bottom 4 mL of the gradient was pooled and designated noncaveolae membrane. The top 5 mL of the gradient was mixed with 4 mL of buffer C, overlaid with 1 mL of 15% w/v iodixanol in buffer B, and followed by 0.5 mL of 5% w/v iodixanol in buffer B. The gradients were centrifuged at 53,000 g for 90 minutes, and caveolae were collected from the 5%/15% interface (0.5 mL).
Mouse thyroid tissue specimens were homogenized, sonicated, and subjected to nuclear and cytosolic cell fractionation as previously described.4 The total protein concentration was measured using the BCA assay kit (Sigma-Aldrich) with bovine serum albumin as a standard, according to the manufacturer’s instructions. The protein extracts were stored in aliquots at −80°C until used for Western blot analysis. Films were scanned and band density was quantified using digital image densitometry analysis to determine the relative intensity of each band (ImageJ software; National Institutes of Health, Bethesda, MD).
To generate a model of Mig-6 loss in the thyroid, we crossed Mig-6f/f transgenic mice with transgenic mice expressing Cre recombinase under the control of the human TPO promoter. PCR genotyping analysis, using primers selectively amplifying the unrecombined or the recombined allele, showed that mice numbers 1, 9, 12, and 15 were Mig-6d/d (Mig-6f/f; TPO-Cre+) and that mice numbers 5, 10, 14, and 17 were Mig-6f/f (Mig-6f/f; TPO-Cre−) mice, while others were wild-type Mig-6 with or without the TPO-Cre gene (Fig 1, A) The level of Mig-6 expression was specifically ablated in Mig-6d/d but not Mig-6f/f or wild-type mice (Fig 1, B).
Mig-6d/d mice were born at the expected Mendelian ratios and developed normally, without displaying obvious physical or behavioral abnormalities. No significant differences in body weight among Mig-6d/d, Mig-6f/f, and wild-type mice were found in a 6- to 8-week follow-up (data not shown).
The histologic examination of these thyroid specimens did not reveal adenomatous changes or evidence of thyroid cancer in wild-type, Mig-6f/f, or Mig-6d/d mice. Less than 1% of cells of thyroid tissue resected from each mouse were Ki-67 positive (data not shown). Mig-6 loss alone is not sufficient to increase thyrocyte proliferation at 6 to 8 weeks of age.
To determine the effect of Mig-6 ablation on EGFR and ERK signaling in vivo, we compared the levels of phospho-EGFR and phospho-ERK in thyroid homogenates prepared from control and knockout mice. Our results showed that EGFR and ERK phosphorylation levels were undetectable in wild-type and Mig-6f/f mice. In comparison, EGFR and ERK phosphorylation levels were high in Mig-6d/d mice (Fig 2). These results show that Mig-6 is a negative inhibitor of EGFR and ERK signaling in thyrocytes in vivo.
Upon ligand binding, EGFR can be internalized by either clathrin-mediated endocytosis (CME) or nonclathrin endocytosis (NCE). CME is necessary for activation of EGFR signaling pathways, whereas NCE is thought to be initiated from caveolin-containing sterol-rich cell membrane domains and primarily leads to receptor degradation.13,15,16 Therefore, regulating the distribution of EGFR to caveolae may regulate its internalization and signaling. Because Mig-6 ablation activates EGFR, we next determined the impact of Mig-6 ablation on EGFR disassociation from caveolae. To verify if Mig-6 loss triggers EGFR trafficking in a NCE-dependent mechanism and leads to receptor degradation, we purified the caveolae-containing membrane fraction from mice thyroid, then determined EGFR levels by Western blot analysis to examine the association between EGFR and caveolae. We also collected total cell lysates then immunoblotted EGFR to examine if Mig-6 loss impacts EGFR protein degradation. EGFR levels were undetectable in caveolae-containing membranes purified from Mig-6d/d mouse thyroids. Abundant EGFR levels were detected in Mig-6f/f mice thyroid cell caveolae-containing membranes (Fig 3, A). We also observed that EGFR is not degraded in Mig-6d/d mouse thyroid (Fig 3, B). Therefore, Mig-6 may normally regulate EGFR activation by altering its distribution in the plasma membrane in a NCE-independent pathway and thereby control its pathway of internalization. To verify if Mig-6 loss triggers EGFR trafficking in a CME-dependent mechanism, we immunoprecipitated clathrin then immunoblotted EGFR to examine the association between EGFR and clathrin. Interaction of EGFR and clathrin was found in Mig-6d/d but not Mig-6f/f mouse thyroid (Fig 3, C). These findings suggested that Mig-6 loss in mouse thyroid triggers EGFR internalization in a CME-dependent mechanism.
Mig-6 promotes NF-κB activation by displacing p65 NF-κB from its inhibitor, IκB.3 To determine the impact of Mig-6 loss in the thyroid on NF-κB activity, we characterized the cell nuclear localization ratio of the p65, which is an indicator of NF-κB activity.17 p65 was evenly localized in the nucleus and cytoplasm in Mig-6f/f mouse thyroid tissue. In contrast, p65 preferentially localized to the cytoplasmic component in Mig-6d/d mice (Fig 4). These results suggest that Mig-6 promotes NF-κB activity in the thyroid.
Mig-6 ablation in the thyroid gland in vivo increases EGFR and ERK activity but decreases NF-κB activity. These effects on EGFR/ERK and NF-κB are consistent with our previous findings in thyroid cancer cell lines.4 However, we previously observed increased proliferation in Mig-6 knockdown cells in vitro. In contrast, the histologic examination of Mig-6d/d thyroids did not reveal adenomatous changes or evidence of increased proliferation by Ki-67 immunohistochemical staining, at least up to 8 weeks of age. This may be because of the young age of the mice at analysis, and more time may be necessary to observe the histologic evidence of hyperproliferation in Mig-6d/d mice.
A previous study revealed that the thyroid-specific loss of phosphatase and tensin homolog (PTEN), a known tumor suppressor, attenuated Akt signaling activation in 10-week-old mice. However, goiters and adenomas were observed only later in 10-month-old mice.18 PTEN knockdown, in addition to overexpression of PAX8/peroxisome proliferator-activated receptor γ fusion protein (PPFP), synergistically caused thyroid hyperplasia by 12 months of age.19 Perhaps Mig-6d/d mice may also develop hyperplastic changes in later stages of development.
Alternatively, it is possible that Mig-6 loss alone is not sufficient for tumorigenesis. Others have observed that the Mig-6 and PTEN loss together cause endometrial tumors in mice as early as 2 to 4 weeks of age.2 Skin-specific p53 knockout in addition to oncogene K-RAS activation synergistically causes squamous cell carcinoma in mice as early as by 2 weeks.20 Therefore, Mig-6 ablation alone may not be sufficient for the development of thyroid cancer; oncogenesis in Mig-6d/d mice may require the acquisition of other genetic defects, such as the loss of an additional tumor suppressor or oncogenic activation.
EGFR internalization and trafficking toward the cell cytoplasm and nucleus is a key mechanism for amplifying downstream signaling pathways, such as mitogen-activated protein kinase (MAPK)/extracellular signal-regulated kinase (ERK).21 However, EGFR internalization also promotes trafficking toward the endosome and finally results in EGFR degradation.22 EGFR signaling can either be terminated or amplified by EGFR internalization, and these events are critical for oncogenesis.23 Many tumor suppressors modulate EGFR downstream signaling through regulating EGFR internalization. Merlin has been shown to inhibit EGFR internalization, and it therefore attenuates downstream signaling pathway including MAPK in renal carcinoma cells.24 Our data also show that, in murine thyroid, Mig-6 blocks EGFR activation and attenuates ERK, a subunit of the MAPK signaling pathway. Mig-6 also appears to promote the distribution of EGFR to caveolin-containing membrane fractions and may thereby promote NCE leading to receptor degradation rather than activation of signaling pathways in response to ligand binding. In agreement with such a model, Mig-6 promotes EGFR internalization, results in receptor degradation, leading to attenuation of ERK activity.25 Unlike these previous studies, however, we do not observe an effect of Mig-6 on total levels of EGFR in mouse thyroid. This may reflect differences in the basal levels of EGFR receptor-mediated endocytosis in these distinct cell types or may suggest that Mig-6 attenuates EGFR signaling instead of affecting EGFR degradation in the thyroid.
In conclusion, our results show that Mig-6 loss in vivo increases EGFR and ERK activity and inactivates NF-κB in the mouse thyroid. We did not find histologic evidence of hyperproliferation in Mig-6 knockout mice; this may be related to the young age of the mice studied, or it may indicate that loss of Mig-6 alone is insufficient to cause tumorigenesis. Future studies should include more mature Mig-6d/d mice and should investigate the effects of adding other genetic defects in combination with Mig-6 loss on the thyroid.
Supported by an Osteen Junior Faculty Research Grant (to D.T.R) from the Department of Surgery at Brigham and Women’s Hospital.