Cancer Research 2010 AACR Elections 2010 Workshops
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Cancer Research Clinical Cancer Research
Cancer Epidemiology Biomarkers & Prevention Molecular Cancer Therapeutics
Molecular Cancer Research Cancer Prevention Research
Cancer Prevention Journals Portal Cancer Reviews Online
Annual Meeting Education Book Meeting Abstracts Online
Cancer Research 68, 1329, March 1, 2008. doi: 10.1158/0008-5472.CAN-07-5754
© 2008 American Association for Cancer Research
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Supplementary Data
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by van Veelen, W.
Right arrow Articles by Höppener, J. W.M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by van Veelen, W.
Right arrow Articles by Höppener, J. W.M.

Molecular Biology, Pathobiology, and Genetics

Synergistic Effect of Oncogenic RET and Loss of p18 on Medullary Thyroid Carcinoma Development

Wendy van Veelen1, Carola J.R. van Gasteren1, Dennis S. Acton2, David S. Franklin3, Ruud Berger1, Cornelis J.M. Lips2 and Jo W.M. Höppener1

1 Department of Metabolic and Endocrine Diseases, Division of Biomedical Genetics, and 2 Department of Clinical Endocrinology, Division of Internal Medicine and Dermatology, University Medical Center Utrecht, Utrecht, the Netherlands; and 3 Department of Biological Sciences and Purdue Cancer Center, Purdue University, West Lafayette, Indiana

Requests for reprints: Jo W.M. Höppener, P.O. Box 85090, 3508 AB, Utrecht, the Netherlands. Phone: 31-88-755-4987; Fax: 31-88-755-4295; E-mail: j.w.m.hoeppener{at}umcutrecht.nl.


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
Activating mutations in the RET proto-oncogene are associated with both familial and sporadic medullary thyroid carcinoma (MTC) development; however, the genetic mechanisms underlying MTC tumorigenesis remain largely unknown. Recently, we have identified somatic inactivating mutations in the cell cycle inhibitor gene P18 in human MTC, which coincided with activating RET mutations, suggesting a role for loss of P18 in combination with oncogenic RET in the multistep process of MTC development. Therefore, we crossed transgenic mice expressing oncogenic RET (RET2B) with mice lacking p18 (and p27, another cell cycle inhibitor) and monitored MTC development. RET2B;p18+/– mice and RET2B;p18–/– mice developed MTC with a highly increased incidence compared with their corresponding single mutant littermates. In addition, expression of oncogenic RET causes an earlier age of onset and larger MTCs in p18–/–;p27+/– mice. In a subset of MTCs of RET2B;p18+/–(;p27+/–) mice, p18Ink4c expression was completely lost. This loss of p18Ink4c expression correlated with higher proliferation rates as well as with larger MTCs, indicating that loss of p18 in combination with oncogenic RET not only increases the risk for MTC development but also enhances MTC progression. Our data strongly indicate that oncogenic RET and loss of p18 cooperate in the multistep tumorigenesis of MTC. [Cancer Res 2008;68(5):1329–37]


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
Multiple endocrine neoplasia type 2 (MEN2) is an autosomal dominantly inherited cancer syndrome, which is mainly characterized by a combination of medullary thyroid carcinoma (MTC) and adrenal pheochromocytoma. MEN2 can be subdivided into MEN2A, MEN2B, and familial MTC. Familial MTC patients solely develop MTC, whereas MEN2A and MEN2B patients may develop, in addition to MTC and pheochromocytoma, other tumors like parathyroid adenomas (for MEN2A) and mucosal ganglioneuromas (for MEN2B). MTC originates from the calcitonin-producing neuroendocrine C-cells in the thyroid gland (1).

MEN2 is caused by activating germ-line mutations in the RET proto-oncogene, which encodes a transmembrane receptor tyrosine kinase (2). RET mutations associated with MEN2 lead to constitutive activation of RET (3, 4). A strong genotype-phenotype correlation has been described for RET and MEN2 (5, 6). In MEN2A, the mutations affect cysteine residues in the extracellular domain of the protein (2). In MEN2B, the most common (95%) mutation results in a Met->Thr substitution at position 918 in the intracellular domain of the protein. The M918T mutation is also found as a somatic mutation in about 30% to 40% of sporadic MTC cases (2).

Previously, we generated transgenic mice expressing the human RET oncogene with the M918T mutation in the thyroid C-cells (RET2B; ref. 7). MTCs were detected in only 13% (3 of 23) of these mice at 11 to 24 months of age, whereas C-cell hyperplasia (CCH), a premalignant stage of MTC, was observed in 77% of these mice from 8 months onward. The incomplete penetrance and variable latency period for MTC development in these mice, together with the clinical observation that carriers of the same germ-line RET mutation, can develop MTC at widely varying ages, suggesting that in addition to the mutated RET gene, other tumorigenic events are required for the development of MTC.

In human MTC, both hereditary and sporadic, the most frequently detected chromosomal alteration is loss of a specific part or the entire short arm of chromosome 1, with the most common break point on 1p32 (8–10), where the candidate tumor suppressor gene CDKN2C (from here on indicated as P18) is located. Recently, we have detected several somatic inactivating mutations in P18 in a subset of human MTCs, indicating that P18 is a tumor suppressor gene involved in human MTC development.4

p18–/– mice develop CCH and MTC, the incidence of which is enhanced by additional loss of another cell cycle inhibitor (i.e., p27; ref. 11). Both cell cycle inhibitors p18Ink4c, a member of the INK4 family, and p27Kip1, a member of the CIP/KIP family, inhibit the formation of active cyclin-dependent kinase (CDK) complexes and thus prevent phosphorylation of Rb1, a major player in G1-S phase transition (12, 13). Interestingly, p27 knockout mice as well as Rb1 knockout mice also develop CCH and MTC, indicating that the CDK-Rb pathway is involved in the tumorigenesis of MTC (11, 14–16). Moreover, somatic RET mutations have been described in MTC of Rb1;p53 knockout mice (17), suggesting a cooperative effect of oncogenic RET and the Rb pathway in MTC tumorigenesis.

Because of the development of MTC in both RET2B transgenic mice and p18, p27 and p18;p27 knockout mice, together with the simultaneous occurrence of activating RET mutations and inactivating P18 mutations in human MTCs, we hypothesized that p18 and/or p27 may collaborate with oncogenic RET in the multistep tumorigenesis of MTC. To address this hypothesis, we crossed RET2B mice with p18+/–;p27+/– mice and monitored MTC development. Our results indicate a strong synergistic effect of oncogenic RET and loss of p18 on MTC development and progression.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Mouse strains and genotyping. All experiments with mice were done with the approval of the Animal Experimental and Ethics Committee of the University of Utrecht, Utrecht, the Netherlands. RET2B transgenic mice (C57BL/6 background) and p18;p27 knockout mice (C57BL/6 background) were previously described (7, 11, 18). For genotype verification, genomic tail-DNA was isolated with Prot-K lysis buffer [100 mmol/L Tris (pH 8), 5 mmol/L EDTA, 0.2% SDS, 0.2 mol/L NaCl, 100 µg/µL proteinase K] at 55°C. Separate PCRs were done for RET, p18, and p27 using primers listed in Supplementary Table S1. All PCRs were done at an annealing temperature of 55°C for 33 cycles.

Plasma calcitonin measurements. Blood was collected every 3 months via orbital puncture after isofloran anesthesia. Plasma was isolated from EDTA-blood by centrifugation. Plasma calcitonin concentrations were measured using a solid-phase, enzyme-labeled, two-site chemiluminescent immunometric assay (Siemens Diagnostics) with an Immulite 1000 analyzer according to the manufacturer's protocol. Interassay variability (measured over 18 assays) shows that this assay is very sensitive and robust: average concentration for assay-control 1, 11.2 pg/mL (SD, 0.27 pg/mL); variation coefficient, 2.42%; average concentration for assay-control 2, 209.5 pg/mL (SD, 3.99 pg/mL); variation coefficient, 1.90%. Plasma calcitonin levels were measured in counts per second (cps) and calculated to picograms according to the assay standard curve. Wild-type non–MTC-bearing mice have undetectable plasma calcitonin levels (~35,000 cps). Histologic analysis revealed that in our study, plasma calcitonin levels ≥1.4 pg/mL (≥100,000 cps) indicated microscopically detectable MTCs.

Tissue processing and immunohistochemistry. Isolated thyroid tissues were fixed in 4% formaldehyde (Klinipath) overnight, dehydrated, and embedded in paraffin. Immunohistochemistry [for calcitonin, p18Ink4c, p27Kip1, and proliferating cell nuclear antigen (PCNA)] was done on paraffin sections of 6 µm. For p27Kip1 and PCNA staining, antigen unmasking was done by boiling the slides in sodium citrate buffer (10 mmol/L sodium citrate trisodium salt dihydrate in distilled water, pH 6) for 15 min. Primary antibodies used were rabbit polyclonal anti-calcitonin antibody (DAKO), diluted 1:5,000; rabbit polyclonal anti-p18 antibody (M-168; Santa Cruz Biotechnology), diluted 1:50; mouse monoclonal anti-KIP1 antibody (BD Transduction Laboratories), diluted 1:500; and mouse monoclonal anti-PCNA antibody (PC10; Cell Signaling Technology), diluted 1:4,000. Secondary antibodies used were horseradish peroxidase (HRP)–conjugated swine anti-rabbit antibody (DAKO) and HRP-conjugated rabbit anti-mouse antibody (DAKO), diluted 1:100, at room temperature for 30 min. For p27Kip1 staining, goat anti-M/R/Ra IgG DVDPO-500H (Powervision; ImmunoLogic) was used as secondary antibody. After 3,3'-diaminobenzidine precipitation, a hematoxylin counterstaining was done. The anti-KIP1 antibody and Powervision were kindly provided by the Department of Pathology of the University Medical Center Utrecht, Utrecht, the Netherlands.

Laser capture microdissection. Laser capture microdissection was used to obtain cell populations of selected areas from paraffin-embedded tissue sections. Sections (10 µm thick) on 1-mm PEN Membrane Slides (P.A.L.M. Microlaser Technologies) were deparaffinized and lightly stained with hematoxylin. Using a P.A.L.M. Microbeam laser capture microdissection system (P.A.L.M. Microlaser Technologies), lesions of MTCs and normal tissues were separately isolated from the sections. DNA was isolated from the microdissected tissue samples with Prot-K lysis buffer [50 mmol/L Tris-HCl (pH 8), 0.5% Tween 20, 2 mg/mL Prot-K] at 55°C for 16 h.

Loss of heterozygosity analysis and somatic mutation analysis. For p18 loss of heterozygosity (LOH) analysis and p18 and RET mutation analysis, PCRs were done directly on DNA lysates isolated by laser capture microdissection, using Titanium Taq polymerase (Clontech). P18 primers are listed in Supplementary Table S1. RET primers were described in ref. 17. PCRs were done at an annealing temperature of 55°C (LOH) or 62°C (mutation analysis) for 33 cycles. For LOH analysis, all three listed primers, discriminating between wild-type and null alleles, were used in one PCR. On gel, wild-type and null alleles are discriminated by different sizes (Supplementary Table S1). As a control, tail-DNA from p18+/+, p18+/–, and p18–/– mice, as well as laser capture microdissection-DNA from paraffin sections of normal tissues of p18+/+ and p18–/– mice, was used. For mutation analysis, PCR products were purified with a JETquick PCR purification kit (Genomed). Direct sequence analysis PCRs were done on purified PCR products using BigDye Terminator (Applied Biosystems). As a control, tail-DNA of a wild-type mouse was used.

Statistical analysis. To compare MTC incidences between different genotype groups, the Fisher exact test was used. The Mann-Whitney test was used to compare plasma calcitonin levels of mice in different genotype groups. To compare proliferation indices of different MTCs, the Student t test was used. P < 0.05 was considered as significant.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
Generation of compound RET2B;p18, RET2B;p27 and RET2B;p18;p27 mice. To investigate the role of oncogenic RET, p18, and p27 in the multistep process of MTC development in vivo, we crossed RET2B mice (7) with p18+/–;p27+/– knockout mice (18). Mating between these mouse strains, both on a C57BL/6 background, yielded all expected genotypes at the anticipated Mendelian ratios. Reproduction was not affected in any of the generated genotypes. For this study, only mice from the third generation were used.

Wild-type mice (WT; n = 26), RET2B mice (n = 48), p18+/– mice (n = 25), p18–/– mice (n = 30), and p27+/– mice (n = 22) were generated. To study the effect of oncogenic RET with additional loss of either p18 or p27 on MTC development, compound RET2B;p18+/– mice (n = 44), RET2B;p18–/– mice (n = 31), and RET2B;p27+/– mice (n = 49) were generated. Franklin et al. (11) showed a synergistic effect of loss of both p18 and p27 on MTC development. To investigate a putative additional effect of oncogenic RET on loss of both p18 and p27, we generated p18+/–;p27+/– mice (n = 20), p18–/–;p27+/– mice (n = 24), RET2B;p18+/–;p27+/– mice (n = 44), and RET2B;p18–/–;p27+/– mice (n = 40). p27–/– mice were not generated because in combination with homozygous loss of p18, they have a low mean survival of 3.5 months (11, 18).

p18–/– mice with and without additional loss of p27 display gigantism from early ages onward (11, 18). We did not detect any apparent enhancement of the gigantism phenotype in the presence of oncogenic RET. All mice were monitored up to 12 months of age. In this period, survival was not affected in the different genotype groups, except for p18–/–;p27+/– mice, which died, or became moribund, at the age of 8 to 9 months due to previously described causes (11, 18). The survival of compound RET2B;p18–/–;p27+/– mice was comparable to the survival of p18–/–;p27+/– mice. Therefore, MTC development in the latter two genotype groups was monitored up to 9 months of age.

Synergistic effect of oncogenic RET and loss of p18 on MTC development. In RET2B mice, the expression of the oncogenic RET transgene is directed to the thyroid C-cells due to the used CALC-I promoter. These mice develop CCH and MTC without any other tumors involved in MEN2 (7). Here, we monitored the development of MTC in the different genotype groups by measuring plasma calcitonin levels of the mice. Plasma calcitonin levels are used as a specific tumor marker for the clinical diagnosis and follow-up of MTC patients (19). Previously, we showed that plasma calcitonin levels can also be used to monitor MTC development in mice (7). Blood was drawn every 3 months and when the mice became moribund. In this way, we were able to monitor MTC development and progression in time. At the time of sacrifice, thyroid glands were isolated for histologic analysis. Calcitonin immunostainings were done to identify the presence of a MTC. Bilateral and multifocal MTCs were detected in mice of different genotype groups. The level of plasma calcitonin in wild-type non–MTC-bearing mice was undetectable (Fig. 1A ). Histologic analysis revealed that in our study, plasma calcitonin levels of ≥1.4 pg/mL indicated the presence of a MTC (Fig. 1B). Most MTCs were detectable only by microscopy (Fig. 1B and C); however, some mice developed macroscopically detectable MTCs (Fig. 1D).


Figure 1
View larger version (68K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 1. MTC development in mice with elevated plasma calcitonin levels. A, calcitonin (CT) staining on the thyroid gland of a wild-type C57BL/6 mouse with undetectable plasma calcitonin level indicating normal C-cell amounts. Right, higher magnification of the boxed area in left. B, calcitonin staining on a thyroid gland of a mouse with a plasma calcitonin level of 1.7 pg/mL indicating a micro-MTC. Right, higher magnification of the boxed area in left. C, calcitonin staining on thyroid glands of mice with plasma calcitonin levels of 2.6, 10.4, and 158 pg/mL, indicating MTCs of increasing sizes. D, macroscopic bilateral MTC (white arrows) in a mouse with a plasma calcitonin level of 569 pg/mL. Black arrows, calcitonin-positive C-cells. f, thyroid follicle; t, trachea. Bar, 50 µm.

 
The numbers and percentages of mice of all different genotype groups that developed MTC within 12 months are summarized in Table 1 . Up to 12 months of age, no MTC was detected in the WT, RET2B, p18+/–, and p27+/– groups, whereas 11% (3 of 27) of p18–/– mice had developed MTC at this age. Forty-three percent (13 of 30) of compound RET2B;p18–/– mice displayed MTC at 12 months, which is a significantly increased incidence compared with both RET2B mice and p18–/– mice. The MTC incidence in RET2B;p18–/– mice was already significantly increased to 33% (10 of 30) at 9 months of age, indicating a strong synergistic effect of oncogenic RET and loss of p18 on MTC development.


View this table:
[in this window]
[in a new window]

  		Table 1. MTC incidence in mice of each genotype group up to 12 mo of age

 
Haploinsufficiency of p18, but not p27, in RET-induced MTC. Previously, it has been shown that treatment of heterozygous p18 knockout mice with a chemical carcinogen resulted in tumor development at an accelerated rate. The remaining wild-type allele of p18 was neither mutated nor silenced in tumors derived from heterozygotes, indicating that p18 is a haploinsufficient gene in tumorigenesis in mice (20). Although MTC developed predominantly in mice of homozygous p18 knockout groups, a subset of mice in the heterozygous p18 knockout groups did develop MTC as well. This offered an opportunity to address p18 haploinsufficiency in MTC development. As is shown in Table 1, loss of a single p18 allele (with and without additional loss of p27) did not result in MTC development in mice up to 12 months, indicating that heterozygous loss of p18 is not sufficient for MTC development within this period. However, compound RET2B;p18+/– mice displayed a significantly increased MTC incidence already at 9 months (21%; 9 of 42) as compared with both RET2B and p18+/– mice, suggesting that p18 is haploinsufficient in the presence of oncogenic RET (Table 1).

RET2B mice, which had lost a single p27 allele, did not show a significant increase in MTC incidence compared with RET2B mice and p27+/– mice: 4% (2 of 47) of RET2B;p27+/– mice developed MTC up to 12 months of age (Table 1), indicating that heterozygous loss of p27 does not affect RET-induced MTC development. These data suggest that RET and p18, but not RET and p27, cooperate in MTC development.

CCH in RET2B;p27+/– mice and p18+/–;p27+/– mice at 12 months of age. Up to 12 months of age, WT mice did not develop MTC, nor could we histologically detect CCH in these mice. The mean plasma calcitonin level of WT mice at the age of 12 months was indicated as 34,842 cps (SD, 6,466 cps; n = 25). To investigate whether mice of the different genotype groups that did not develop MTC (plasma calcitonin <100,000 cps) developed CCH, we analyzed whether they had significantly elevated plasma calcitonin levels at the age of 12 months (Table 2 ).


View this table:
[in this window]
[in a new window]

  		Table 2. Elevated plasma calcitonin levels indicating CCH in mice that did not develop MTC (plasma calcitonin <100,000 cps) at 12 mo of age

 
Plasma calcitonin levels of RET2B mice, p18+/– mice, and p27+/– mice did not significantly differ from those of WT mice, suggesting that these mice did not develop CCH at the age of 12 months. P18–/– mice, however, did develop a significant increase in plasma calcitonin at the age of 12 months (59,271 cps; SD, 17,043 cps; n = 24), suggesting CCH. In the three RET2B;p18 genotype groups in which MTC was also detected at 12 months of age, increased plasma calcitonin levels were identified in mice that did not develop MTC.

Interestingly, in the RET2B;p27+/– group and p18+/–;p27+/– group, in which MTC development could (almost) not be observed at 12 months, CCH was suggested by elevated plasma calcitonin levels compared with those of WT mice. Comparing plasma calcitonin levels of RET2B;p27+/– mice with plasma levels of RET2B mice revealed an increase that was not significant (P = 0.06), whereas plasma calcitonin levels of p18+/–;p27+/– mice were significantly (P < 0.001) increased compared with those of p18+/– mice. These data suggest that loss of p27 synergizes with loss of p18, but not with oncogenic RET, in MTC development.

Oncogenic RET causes earlier MTC onset in compound transgenic mice. The synergism between p18 and p27 could also be observed in p18–/–;p27+/– mice: 88% (15 of 17) of p18–/–;p27+/– double knockout mice developed MTC at 9 months of age, compared with 4% (1 of 28) of p18–/– mice (Table 1), indicating that simultaneous loss of p18 (homozygous) and p27 (heterozygous) is sufficient for MTC development in mice, as was also previously shown by Franklin et al. (11). We wondered whether oncogenic RET could increase the MTC incidence even further in the p18;p27 background. One hundred percent (26 of 26) of RET2B;p18–/–;p27+/– mice developed MTC at 9 months, which was not significantly increased compared with p18–/–;p27+/– mice at this age [88% (15 of 17); Table 1]. However, 27% (12 of 44) of RET2B;p18+/–;p27+/– mice displayed MTC at 12 months of age, which is a significant increase compared with 0% (0 of 18) of p18+/–;p27+/– mice (Table 1). In addition, a remarkable difference in MTC incidence between p18–/–;p27+/– mice and RET2B;p18–/–;p27+/– mice was observed at the age of 6 months: 31% (11 of 35) of RET2B;p18–/–;p27+/– mice developed MTC at this age compared with 0% (0 of 23) of p18–/–;p27+/– mice (Table 1), strongly indicating that oncogenic RET causes an earlier MTC onset in p18–/–;p27+/– mice. All mice that displayed MTC already at the age of 3 or 6 months (n = 15) belonged to groups with a combined RET2B and heterozygous or homozygous p18 knockout genotype (Table 1), confirming that the combination of oncogenic RET and loss of p18 strongly promotes (early) MTC development in mice.

Oncogenic RET induces larger MTCs in compound transgenic mice. As described above, oncogenic RET in combination with loss of p18 or loss of both p18 and p27 resulted in increased MTC incidence as well as decreased age of onset. Next, we investigated whether we could detect an effect on MTC growth of oncogenic RET, in addition to loss of p18 and p27, by comparing plasma calcitonin levels of mice with MTC. We found a correlation between plasma calcitonin levels and MTC size, as was also detected for human MTC patients (Fig. 1C; ref. 21). In Fig. 2 , plasma calcitonin levels of all mice with MTC from different compound genotype groups at 3, 6, 9, and 12 months of age are indicated.


Figure 2
View larger version (15K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2. Plasma calcitonin levels of mice with MTC (≥1.4 pg/mL) from compound genotype groups are indicated at 3, 6, 9, and 12 mo of age. A logarithmic scale was used to plot the plasma calcitonin levels. ND, not determined (mice were sacrificed at 9 mo).

 
At the age of 9 months, plasma calcitonin levels of compound RET2B;p18–/–;p27+/– mice with MTC (n = 24) were significantly higher (ranging from 1.4 to 569 pg/mL with a median concentration of 84.25 pg/mL; P < 0.01) compared with plasma calcitonin levels of p18–/–;p27+/– mice with MTC (n = 15; ranging from 2.1 to 261 pg/mL with a median concentration of 15.6 pg/mL; Fig. 2). This indicates that already at 9 months of age, RET2B;p18–/–;p27+/– mice have larger MTCs compared with p18–/–;p27+/– mice. This was also confirmed by histologic analysis.

To investigate whether we could detect a difference in growth rate in these MTCs, we analyzed proliferation in five MTCs of both p18–/–;p27+/– mice and RET2B;p18–/–;p27+/– mice. The MTCs were selected from mice of each genotype group, which matched according to their plasma calcitonin levels, and thus MTCs were of comparable sizes. As determined by PCNA immunostainings and subsequent proliferation index calculations, we did not identify a significant difference (P = 0.43) in the average proliferation rates between MTCs from the two different genotype groups: 17.7% (SD, 2.4%) for MTCs from p18–/–;p27+/– mice and 19.6% (SD, 4.6%) for MTCs from RET2B;p18–/–;p27+/– mice (Supplementary Table S2), suggesting that the larger MTCs detected in RET2B;p18;p27 mice are due to earlier age of onset rather than to increased growth rate.

Heterozygous loss of p27 increases MTC incidence in RET2B;p18–/–, but not RET2B;p18+/–, compound transgenic mice. As described above, heterozygous loss of p27 additionally affects MTC development in combination with complete loss of p18, but not with oncogenic RET. At 9 months of age, RET2B;p18–/–;p27+/– mice displayed a significantly higher MTC incidence (100%; 26 of 26) compared with RET2B;p18–/– mice [33% (10 of 30); Table 1]. In addition, RET2B;p18–/–;p27+/– mice with MTC also displayed significantly higher (P < 0.001) plasma calcitonin levels compared with RET2B;p18–/– mice with MTC (n = 10; ranging from 1.7 to 420 pg/mL with a median concentration of 2.2 pg/mL), indicating an additional effect of heterozygous loss of p27 on RET2B;p18–/––induced MTC development (Fig. 2).

Comparing plasma calcitonin levels of RET2B;p18+/–;p27+/– mice that did not develop MTC at the age of 12 months with plasma calcitonin levels of RET2B;p18+/– mice revealed an increase that was not significant (P = 0.07), suggesting that additional loss of p27 does not strongly enhances CCH in RET2B;p18+/– mice (Table 2). Surprisingly, RET2B;p18+/–;p27+/– mice (7%; 3 of 44) displayed a significantly lower MTC incidence at the age of 9 months compared with RET2B;p18+/– mice (21%; 9 of 42), suggesting a protective effect of loss of p27 on MTC development (Table 1). However, five of these nine RET2B;p18+/– mice displayed plasma calcitonin levels of 1.4 to 2 pg/mL, indicating very small MTCs (Fig. 2). Moreover, at 12 months of age, these genotype groups displayed equal frequencies of MTC development [26% (10 of 39) for RET2B;p18+/– mice and 27% (12 of 44) for RET2B;p18+/–;p27+/– mice; Table 1] and no significant differences (P = 0.52) in plasma calcitonin levels (Fig. 2), indicating that it is not likely that loss of p27 is protective for MTC development.

p27 is known to be a haploinsufficient tumor suppressor gene (22). To investigate haploinsufficiency of p27 in (RET2B;)p18–induced MTC, a p27Kip1 immunostaining was done on 41 MTCs from heterozygous p27 knockout mice (21 RET2B;p18–/–;p27+/– mice, 7 RET2B;p18+/–;p27+/– mice, 11 p18–/–;p27+/– mice, and 2 RET2B;p27+/– mice). As a positive control, we stained MTC sections of two p27+/+ mice. In none of the 41 MTCs complete loss or strongly reduced p27Kip1 expression could be observed (not shown), indicating that the remaining p27 allele is still intact. These results indicate that p27 functions as a haploinsufficient tumor suppressor gene in p18-induced MTC tumorigenesis.

Somatic alterations in p18 or RET in murine MTCs. Next, we investigated whether we could detect loss of p18 as a somatic event in MTCs from mice without a germ-line defect in p18. We stained MTCs from seven RET2B mice (>12 months old) and from two RET2B;p27+/– mice for p18Ink4c. As a negative control, a p18Ink4c immunostaining was done on MTC sections from a p18–/– mouse (Fig. 3A ). Eight MTCs of RET2B(;p27+/–) mice stained positive for p18Ink4c, whereas in one MTC of a RET2B transgenic mouse a patchy p18Ink4c expression pattern was detected (Fig. 3B). We could not detect homozygous loss of p18 in the DNA isolated from this tumor (Supplementary Fig. S1). Subsequently, we carried out a mutation analysis on DNA isolated from all nine p18+/+ MTCs. In none of these MTCs a mutation in the coding region of p18 could be detected.


Figure 3
View larger version (120K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 3. p18Ink4c immunostaining of paraffin sections of murine MTCs. A, negative control staining of p18–/– MTC (RET2B;p18–/–;p27+/– mouse). B, p18Ink4c immunostaining of p18+/+ MTCs (left, positive staining of a MTC of a RET2B;p27+/– mouse; right, patchy staining of a MTC of a RET2B mouse). C, p18Ink4c immunostaining of p18+/– MTCs (left, positive staining of a MTC of a RET2B;p18+/–;p27+/– mouse; middle, negative staining of a MTC of a RET2B;p18+/– mouse; right, patchy staining of a MTC of a RET2B;p18+/– mouse). Bar, 50 µm.

 
Coxon et al. (17) detected somatic RET mutations in four of nine MTCs from Rb;p53 double knockout mice. These RET mutations corresponded with human MEN2A-associated RET mutations. To investigate whether such mutations were acquired in MTCs from p18;p27 knockout mice as well, we carried out nucleotide sequence analysis on DNA isolated from MTCs of seven p18–/–;p27+/– mice. In none of these MTCs a somatic RET mutation could be identified.

Loss of p18Ink4c expression in MTCs from p18+/– mice correlates with MTC progression. To assess whether p18Ink4c expression was lost in MTCs of heterozygous p18 knockout mice, p18Ink4c expression was determined in MTCs of RET2B;p18+/– (n = 9) and RET2B;p18+/–;p27+/– mice (n = 8). In 7 of these 17 MTCs of heterozygous p18 knockout mice, p18Ink4c expression could be detected, comparable with p18Ink4c expression in MTC of p18+/+ mice. However, in 7 of these 17 MTCs, p18Ink4c expression was completely lost, and in 3 MTCs a patchy p18Ink4c expression pattern was observed (Fig. 3C; Table 3 ), indicating that somatic loss of p18Ink4c expression occurs frequently in murine MTC.


View this table:
[in this window]
[in a new window]

  		Table 3. Loss of p18Ink4c expression in MTCs of RET2B;p18+/–(;p27+/–) knockout mice correlates with plasma calcitonin levels and proliferation rates

 
Next, we investigated whether this somatic loss of p18Ink4c expression was due to LOH or somatic mutations in the p18 gene. In none of the MTCs that showed (partial) loss of p18Ink4c expression could we detect loss of the wild-type p18 allele(s) (Supplementary Fig. S1), nor could we detect mutations in the coding region of p18 in MTCs from these 17 RET2B;p18+/–(;p27+/–) mice, indicating that other mechanisms are involved in the loss of p18Ink4c expression in these MTCs.

Furthermore, we investigated whether loss of p18Ink4c expression affected MTC growth by comparing the plasma calcitonin levels of RET2B;p18+/–(;p27+/–) mice with MTCs with (partial) loss of p18Ink4c expression (n = 10; ranging from 6.2 to 522 pg/mL with a median concentration of 97 pg/mL) to plasma calcitonin levels of RET2B;p18+/–(;p27+/–) mice with MTCs with p18Ink4c expression (n = 7; ranging from 1.7 to 19.5 pg/mL with a median concentration of 4.8 pg/mL; Table 3). The plasma calcitonin levels of mice with MTCs without p18Ink4c expression were significantly higher (P = 0.001) compared with those of mice with MTCs expressing p18Ink4c, indicating that MTCs without p18Ink4c expression were larger compared with those expressing p18Ink4c.

Next, we determined the proliferation rates of MTCs of RET2B;p18+/–(;p27+/–) mice with and without p18Ink4c expression. The average proliferation index of MTCs that had (partially) lost p18Ink4c expression [n = 10; 17.9% (SD, 4.6)] resembled the average proliferation index of MTCs of RET2B;p18–/–(;p27+/–) mice (19.6%; SD, 4.6%) as shown in Supplementary Table S2. This was significantly higher (P = 0.007) compared with the average proliferation index of MTCs of RET2B;p18+/–(;p27+/–) mice that express p18Ink4c [n = 6; 11.5% (SD, 2.5%); Table 3]. In conclusion, loss of p18Ink4c expression correlates with larger MTCs and higher proliferation rates, indicating that loss of p18Ink4c enhances MTC progression.


   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
In this article, we provide experimental evidence that p18 inactivation functionally collaborates with oncogenic RET in murine MTC development. As compared with the single mutant mouse strains, we have found an increased MTC incidence in both RET2B;p18+/– and RET2B;p18–/– mice from 9 months onward. Heterozygous loss of p27 did not result in increased MTC incidence in RET2B mice up to 12 months of age. This indicates that loss of p18, but not loss of p27, cooperates with oncogenic RET in MTC development. The observed earlier age of onset of MTC, as well as the larger MTCs, in compound RET2B;p18(;p27) mice provides further evidence for synergism between oncogenic RET and loss of p18. Furthermore, somatic loss of p18Ink4c expression in MTCs of heterozygous p18 knockout mice resulted in higher proliferation rates and larger MTCs. Loss of p18 in the presence of oncogenic RET greatly enhances MTC incidence as well as MTC progression, which shows that loss of p18 is a frequent and severe additional oncogenic hit in RET-induced MTC tumorigenesis in mouse.

Interestingly, our results are consistent with a previously reported study about the role of p18 and p27 in MEN1 (23). p18, but not p27, collaborates with Men1 to suppress tumor formation in pituitary, testis, thyroid, parathyroid, and pancreatic islets (23). The synergistic effect of loss of p18 on tumor formation in both Men1 knockout mice and RET2B transgenic mice indicates a general tumor suppressor role for p18 in endocrine tissues. It has been proposed that INK4 proteins regulate the cell cycle in a cell lineage–specific manner due to tissue-specific differences in expression patterns (24). In contrast to most other INK4 proteins, p18Ink4c is expressed during mouse embryogenesis as well as in most adult tissues like testis, spleen, kidney, skeletal muscle, and lung (25). Therefore, the tissue-specific effect of loss of p18 on endocrine tumorigenesis is not likely due to an endocrine-specific expression pattern of p18Ink4c. More likely, a lack of redundancy involving the other INK4 proteins in endocrine tissues might explain the particular importance of p18 in these tissues.

Previously, biochemical studies have revealed that loss of menin, the Men1 gene product, down-regulates p18Ink4c and p27Kip1 expression (26, 27). Previously, we have detected that expression of oncogenic RET results in down-regulation of p18Ink4c and p27Kip1 expression, leading to increased proliferation (28). Therefore, the observed (partial) loss of p18Ink4c expression in a subset of MTCs from p18+/– mice without loss of the remaining p18 allele might be caused by oncogenic RET signaling. However, this oncogenic RET signaling did not result in loss of p27Kip1 expression in MTCs from p27+/– mice.

Several reports have shown that p18Ink4c and p27Kip1 cooperate in cell cycle arrest of different cell types (29, 30). Loss of inhibition both early and late in G1 (e.g., by loss of p18Ink4c and p27Kip1, respectively) affects cell cycle progression more severely compared with loss of inhibition early in G1 alone (31), which could explain the differential effects observed in RET2B;p18+/– mice and RET2B;p27+/– mice. Different double knockout mouse models, in addition to p18;p27 knockout mice, like Rb;p53 knockout mice (16, 32), p18;p53 knockout mice, p27;p53 knockout mice (33), and Rb;p27 knockout mice (14), have revealed a synergism between cell cycle regulatory genes in MTC development. This is the first time that synergism between a receptor tyrosine kinase (RET) and a CDK inhibitor (p18Ink4c) is reported to be involved in the multistep process of MTC development. Loss of p27 in compound RET2B;p18–/– mice further increases MTC incidence, which is not associated with loss of p27Kip1 expression, indicating that p27 is a haploinsufficient tumor suppressor gene in p18- and RET2B;p18–induced MTC development.

Involvement of oncogenic RET in human MEN2 and MTC development is already well established. Our data suggest a cooperative role for P18 and RET in human MTC tumorigenesis. Germ-line P18 mutations have not yet been described in man, and somatic alterations in the P18 gene have rarely been detected in human cancer. However, loss of chromosome 1p32, the location of the P18 gene, is associated with both sporadic and familial MTC (8–10). Recently, we have detected somatic inactivating P18 mutations in sporadic as well as hereditary MTCs, with a relatively high frequency compared with other types of human cancer investigated.4 Interestingly, all somatic P18 mutations detected in human MTC coincided with germ-line or somatic RET mutations, suggesting that a cooperation between oncogenic RET and inactive P18 would also contribute to human MTC tumorigenesis.

It has been suggested that p18 is a haploinsufficient gene in mice. Haploinsufficiency of p18 has only been detected in the presence of carcinogen treatment (20) or in combination with loss of p27 or Patched (11, 34). In our mouse models, we show that p18 functions as a haploinsufficient tumor suppressor gene in the presence of oncogenic RET. However, the haploinsufficiency of p18 is not absolute because loss of both p18 alleles resulted in a higher MTC incidence in all genotype groups and loss of p18Ink4c expression resulted in enhanced MTC growth in p18+/– mice. Taken together, we show that loss of p18 greatly contributes to RET-induced MTC incidence as well as MTC progression. We propose that loss of p18 (and p27) is an additional oncogenic hit in the multistep process of RET-induced MTC tumorigenesis.


   Acknowledgments
 
Grant support: Grant UU2712 from the Dutch Cancer Society.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

We thank H. Boere, S. Versteeg, and H. Avezaat of the Animal Facility of the University of Utrecht, Utrecht, the Netherlands for assistance with the mouse experiments; H. Compagner, I. Maitimu, and Dr. E. Lentjes of the Department of Endocrinology of the University Medical Center Utrecht, Utrecht, the Netherlands for assistance with plasma calcitonin measurements; and several colleagues for critical reading of the manuscript.


   Footnotes
 
Note: Supplementary data for this article are available at Cancer Research Online (http://cancerres.aacrjournals.org/).

Current address for D.S. Acton: MucoVax B.V., Leiden, the Netherlands. Current address for D.S. Franklin: Department of Biochemistry, Tulane University School of Medicine, New Orleans, LA. Current address for W. van Veelen: Department of Gastroenterology and Hepatology, Erasmus Medical Center, Rotterdam, the Netherlands.

4 W. Van Veelen, M. Gloerich, C.J.R. Van Gasteren, E. Kalkhoven, R. Berger, C.J.M. Lips, J.W.M. Höppener, D.S. Acton. P18 is a tumor suppressor gene involved in human medullary thyroid carcinoma and pheochromocytoma development. Submitted for publication. Back

Received 10/ 5/07. Revised 12/18/07. Accepted 1/ 7/08.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Sizemore GW, Health H III, Carney JA. Multiple endocrine neoplasia type 2. Clin Endocrinol Metab 1980;9:299–315.[Medline]
  2. Eng C, Mulligan LM. Mutations of the RET proto-oncogene in the multiple endocrine neoplasia type 2 syndromes, related sporadic tumours, and hirschsprung disease. Hum Mutat 1997;9:97–109.[CrossRef][Medline]
  3. Asai N, Iwashita T, Matsuyama M, Takahashi M. Mechanism of activation of the ret proto-oncogene by multiple endocrine neoplasia 2A mutations. Mol Cell Biol 1995;15:1613–9.[Abstract/Free Full Text]
  4. Santoro M, Carlomagno F, Romano A, et al. Activation of RET as a dominant transforming gene by germline mutations of MEN2A and MEN2B. Science 1995;267:381–3.[Abstract/Free Full Text]
  5. Machens A, Gimm O, Hinze R, Hoppner W, Boehm BO, Dralle H. Genotype-phenotype correlations in hereditary medullary thyroid carcinoma: oncological features and biochemical properties. J Clin Endocrinol Metab 2001;86:1104–9.[Abstract/Free Full Text]
  6. Mulligan LM, Marsh DJ, Robinson BG, et al. Genotype-phenotype correlation in multiple endocrine neoplasia type 2: report of the International RET Mutation Consortium. J Intern Med 1995;238:343–6.[Medline]
  7. Acton DS, Velthuyzen D, Lips CJ, Hoppener JW. Multiple endocrine neoplasia type 2B mutation in human RET oncogene induces medullary thyroid carcinoma in transgenic mice. Oncogene 2000;19:3121–5.[CrossRef][Medline]
  8. Marsh DJ, Theodosopoulos G, Martin-Schulte K, et al. Genome-wide copy number imbalances identified in familial and sporadic medullary thyroid carcinoma. J Clin Endocrinol Metab 2003;88:1866–72.[Abstract/Free Full Text]
  9. Mathew CG, Smith BA, Thorpe K, et al. Deletion of genes on chromosome 1 in endocrine neoplasia. Nature 1987;328:524–6.[CrossRef][Medline]
  10. Yang KP, Nguyen CV, Castillo SG, Samaan NA. Deletion mapping on the distal third region of chromosome 1p in multiple endocrine neoplasia type IIA. Anticancer Res 1990;10:527–33.[Medline]
  11. Franklin DS, Godfrey VL, O'Brien DA, Deng C, Xiong Y. Functional collaboration between different cyclin-dependent kinase inhibitors suppresses tumor growth with distinct tissue specificity. Mol Cell Biol 2000;20:6147–58.[Abstract/Free Full Text]
  12. Sherr CJ, Roberts JM. Inhibitors of mammalian G1 cyclin-dependent kinases. Genes Dev 1995;9:1149–63.[Free Full Text]
  13. Sherr CJ, Roberts JM. CDK inhibitors: positive and negative regulators of G1-phase progression. Genes Dev 1999;13:1501–12.[Free Full Text]
  14. Park MS, Rosai J, Nguyen HT, Capodieci P, Cordon-Cardo C, Koff A. p27 and Rb are on overlapping pathways suppressing tumorigenesis in mice. Proc Natl Acad Sci U S A 1999;96:6382–7.[Abstract/Free Full Text]
  15. Takahashi C, Contreras B, Bronson RT, Loda M, Ewen ME. Genetic interaction between Rb and K-ras in the control of differentiation and tumor suppression. Mol Cell Biol 2004;24:10406–15.[Abstract/Free Full Text]
  16. Williams BO, Remington L, Albert DM, Mukai S, Bronson RT, Jacks T. Cooperative tumorigenic effects of germline mutations in Rb and p53. Nat Genet 1994;7:480–4.[CrossRef][Medline]
  17. Coxon AB, Ward JM, Geradts J, Otterson GA, Zajac-Kaye M, Kaye FJ. RET cooperates with RB/p53 inactivation in a somatic multi-step model for murine thyroid cancer. Oncogene 1998;17:1625–8.[CrossRef][Medline]
  18. Franklin DS, Godfrey VL, Lee H, et al. CDK inhibitors p18(INK4c) and p27(Kip1) mediate two separate pathways to collaboratively suppress pituitary tumorigenesis. Genes Dev 1998;12:2899–911.[Abstract/Free Full Text]
  19. Lips CJ, Hoppener JW, Thijssen JH. Medullary thyroid carcinoma: role of genetic testing and calcitonin measurement. Ann Clin Biochem 2001;38:168–79.[CrossRef][Medline]
  20. Bai F, Pei XH, Godfrey VL, Xiong Y. Haploinsufficiency of p18(INK4c) sensitizes mice to carcinogen-induced tumorigenesis. Mol Cell Biol 2003;23:1269–77.[Abstract/Free Full Text]
  21. Cohen R, Campos JM, Salaun C, et al. Preoperative calcitonin levels are predictive of tumor size and postoperative calcitonin normalization in medullary thyroid carcinoma. Groupe d'Etudes des Tumeurs a Calcitonine (GETC). J Clin Endocrinol Metab 2000;85:919–22.[Abstract]
  22. Polyak K. The p27Kip1 tumor suppressor gene: still a suspect or proven guilty? Cancer Cell 2006;10:352–4.[CrossRef][Medline]
  23. Bai F, Pei XH, Nishikawa T, Smith MD, Xiong Y. p18Ink4c, but not p27Kip1, collaborates with Men1 to suppress neuroendocrine organ tumors. Mol Cell Biol 2007;27:1495–504.[Abstract/Free Full Text]
  24. Roussel MF. The INK4 family of cell cycle inhibitors in cancer. Oncogene 1999;18:5311–7.[CrossRef][Medline]
  25. Zindy F, Quelle DE, Roussel MF, Sherr CJ. Expression of the p16INK4a tumor suppressor versus other INK4 family members during mouse development and aging. Oncogene 1997;15:203–11.[CrossRef][Medline]
  26. Karnik SK, Hughes CM, Gu X, et al. Menin regulates pancreatic islet growth by promoting histone methylation and expression of genes encoding p27Kip1 and p18INK4c. Proc Natl Acad Sci U S A 2005;102:14659–64.[Abstract/Free Full Text]
  27. Milne TA, Hughes CM, Lloyd R, et al. Menin and MLL cooperatively regulate expression of cyclin-dependent kinase inhibitors. Proc Natl Acad Sci U S A 2005;102:749–54.[Abstract/Free Full Text]
  28. Joshi PP, Kulkarni MV, Yu BK, et al. Simultaneous down-regulation of CDK inhibitors p18(Ink4c) and p27(Kip1) is required for MEN2A-RET-mediated mitogenesis. Oncogene 2007;26:554–70.[CrossRef][Medline]
  29. Myers TK, Andreuzza SE, Franklin DS. p18INK4c and p27KIP1 are required for cell cycle arrest of differentiated myotubes. Exp Cell Res 2004;300:365–78.[CrossRef][Medline]
  30. Swarbrick A, Lee CS, Sutherland RL, Musgrove EA. Cooperation of p27(Kip1) and p18(INK4c) in progestin-mediated cell cycle arrest in T-47D breast cancer cells. Mol Cell Biol 2000;20:2581–91.[Abstract/Free Full Text]
  31. Delaval B, Birnbaum D. A cell cycle hypothesis of cooperative oncogenesis [review]. Int J Oncol 2007;30:1051–8.[Medline]
  32. Harvey M, Vogel H, Lee EY, Bradley A, Donehower LA. Mice deficient in both p53 and Rb develop tumors primarily of endocrine origin. Cancer Res 1995;55:1146–51.[Abstract/Free Full Text]
  33. Damo LA, Snyder PW, Franklin DS. Tumorigenesis in p27/p53- and p18/p53-double null mice: functional collaboration between the pRb and p53 pathways. Mol Carcinog 2005;42:109–20.[CrossRef][Medline]
  34. Uziel T, Zindy F, Xie S, et al. The tumor suppressors Ink4c and p53 collaborate independently with Patched to suppress medulloblastoma formation. Genes Dev 2005;19:2656–67.[Abstract/Free Full Text]



This article has been cited by other articles:


Home page
 J Mol EndocrinolHome page
A. Cerrato, V. De Falco, and M. Santoro
Molecular genetics of medullary thyroid carcinoma: the quest for novel therapeutic targets
J. Mol. Endocrinol., October 1, 2009; 43(4): 143 - 155.
[Abstract] [Full Text] [PDF]

Home page
 J. Clin. Endocrinol. Metab.Home page
L. Ye, L. Santarpia, G. J. Cote, A. K. El-Naggar, and R. F. Gagel
High Resolution Array-Comparative Genomic Hybridization Profiling Reveals Deoxyribonucleic Acid Copy Number Alterations Associated with Medullary Thyroid Carcinoma
J. Clin. Endocrinol. Metab., November 1, 2008; 93(11): 4367 - 4372.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Supplementary Data
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by van Veelen, W.
Right arrow Articles by Höppener, J. W.M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by van Veelen, W.
Right arrow Articles by Höppener, J. W.M.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Cancer Research Clinical Cancer Research
Cancer Epidemiology Biomarkers & Prevention Molecular Cancer Therapeutics
Molecular Cancer Research Cancer Prevention Research
Cancer Prevention Journals Portal Cancer Reviews Online
Annual Meeting Education Book Meeting Abstracts Online