Fax: (011) +81-22-717-8047
Frequent nuclear accumulation of β-catenin in pituitary adenoma
Article first published online: 4 JAN 2001
Copyright © 2001 American Cancer Society
Volume 91, Issue 1, pages 42–48, 1 January 2001
How to Cite
Semba, S., Han, S.-Y., Ikeda, H. and Horii, A. (2001), Frequent nuclear accumulation of β-catenin in pituitary adenoma. Cancer, 91: 42–48. doi: 10.1002/1097-0142(20010101)91:1<42::AID-CNCR6>3.0.CO;2-7
- Issue published online: 4 JAN 2001
- Article first published online: 4 JAN 2001
- Manuscript Accepted: 28 AUG 2000
- Manuscript Revised: 11 JUL 2000
- Manuscript Received: 3 FEB 2000
- Ministry of Education, Science, Sports, and Culture of Japan
- β-catenin (CTNNB1);
- pituitary adenoma;
- nuclear accumulation;
- somatic mutation
β-catenin (CTNNB1) is known to be a member of the cadherin-catenin superfamily and to function in cell-cell adhesion. However, it also has been reported that CTNNB1 plays an important role in carcinogenesis. In the current study, the authors observed expression of the CTNNB1 protein in primary pituitary adenomas to investigate the role of CTNNB1 in the development of pituitary adenomas.
A total of 37 pituitary adenomas were analyzed. Expression of CTNNB1 and the cell proliferation marker Ki-67 were observed immunohistochemically. In addition, the authors performed direct sequencing to detect somatic mutations of exon 3 of the CTNNB1 gene.
Twenty-one of 37 pituitary adenomas (57%) demonstrated abnormal nuclear accumulation of CTNNB1. It is interesting to note that tumors with an accumulation of CTNNB1 in the nucleus showed a statistical tendency toward an association with increased immunoreactivity of Ki-67 (P < 0.05) whereas no significant correlation was detected between the status of CTNNB1 and other clinicopathologic features. Missense mutations in exon 3 of the CTNNB1 gene also were detected in the cases with abnormal nuclear accumulation of the CTNNB1 protein.
The results of the current study suggest that up-regulation of the Wnt signaling pathway, including accumulation of mutant CTNNB1 in the nuclei, plays an important role in the tumorigenesis and development of adenoma in the pituitary gland. Cancer 2001;91:42–8. © 2001 American Cancer Society.
Worldwide, the incidence rate of pituitary neoplasms among intracranial tumors is approximately 10%.1 Adenomas arising from the anterior lobe of the pituitary gland have been reported to be associated with synthesis and the release of hormones such as growth hormone (GH), prolactin (PRL), thyroid-stimulating hormone (TSH), adrenocorticotropic hormone (ACTH), follicle-stimulating hormone (FSH), and luteinizing hormone (LH), and there is clinical and biochemical evidence of a characteristic hypersecretory syndrome in 70% of cases. To our knowledge several aspects of genetic alteration in pituitary adenomas have been reported to date. Heterozygous Rb-knockout mice develop pituitary adenomas at a high frequency, suggesting that the Rb pathway is involved in pituitary tumorigenesis.2, 3 However, mutation of the Rb gene itself is infrequent in pituitary adenomas.4, 5 Conversely, p27-knockout mice develop adenomas in the intermediate lobe of the pituitary gland, suggesting that transcriptional or translational inactivation of p27 also may contribute to tumorigenesis in some organs.6–8 Patients with multiple endocrine neoplasia type 1 (MEN1), a familial neoplastic syndrome with adenomas arising in the pituitary gland, parathyroid gland, and pancreas, have germline mutations of the MEN1 gene.9 However, to our knowledge no mutations of the MEN1 gene have been detected in sporadic pituitary tumors. The p53 gene plays an important role in cell cycle checkpoint regulation, DNA repair, and apoptosis, but its mutation is rare in pituitary tumors.10–12
Cell-cell adhesion is important in cell growth regulation, and the cadherin-catenin complex plays an important role. The intracellular domain of each cadherin family member forms a complex with catenin family members such as α-, β-, and γ-catenins.13 Down-regulation of cadherin-catenin complexes has been observed in various human tumors and is associated closely with tumor progression, metastasis, and aggressiveness. Moreover, recent reports have demonstrated that CTNNB1 plays a key role in the Wnt signal transduction pathways regulating cell growth; adenomatous polyposis coli (APC) and glycogen synthase kinase 3-β (GSK3β) promote ubiquitin-mediated CTNNB1 degradation through phosphorylation of specific serine/threonine residues of CTNNB1.14–16 Mutated CTNNB1 protein is believed to stabilize and translocate itself from the cell membrane to the nucleus, and to bind to transcription factors Lef and/or Tcf, resulting in the induction of expression of the c-myc and/or cyclin D genes.17, 18
In the current study we observed the nuclear accumulation of CTNNB1 in pituitary adenomas and analyzed the relation between the CTNNB1 status and clinicopathologic findings. Furthermore, we investigated genetic alterations of the CTNNB1 gene and Ki-67 expression to evaluate the importance of the nuclear accumulation of CTNNB1 protein in the course of tumorigenesis of the pituitary gland.
MATERIALS AND METHODS
Tissue Samples and DNA Extraction
We analyzed a total of 37 randomly selected tissues from Japanese patients with pituitary adenomas that were removed at Tohoku University Hospital, Sendai, Japan). No tumors were classified as malignant histopathologically. The clinicopathologic findings are summarized in Table 1; there were 15 males and 22 females, with an average age of 34 years (range, 27–74 years). The average size of the tumors was 34 mm in greatest dimension (range, 7–66 mm). There were 20 plurihormonal adenomas, 12 gonadotroph adenomas, 4 mixed GH and PRL cell adenomas (including PRL cell adenomas), and 1 silent ACTH adenoma. Each resected specimen was divided in half and 1 part of the tumor was frozen immediately in liquid nitrogen and stored at –80 °C until use whereas the remaining tumor was fixed in 10% buffered formalin or ethanol and embedded in paraffin for histologic diagnosis and immunohistochemical analysis. DNA was extracted according to methods described previously.19
|Case no.||Age (yrs)||Gender||Tumor sizea||Clinical diagnosisb||Histologic diagnosisc||Hormone production (immunohistochemically)d||Ki-67 labeling indexe||β-catenin (CTNNB1)f||Expression patterng||Somatic mutation of the CTNNNB1geneh|
|2||27||F||24||GH||GH & PRL||GH, PRL||6.0%||(+++)||(−)||(+++)||Homo||None|
|3||56||M||66||PRL||Pluri||GH, PRL, LH, FSH, α-SU||3.0%||(+++)||(−)||(+++)||Homo||None|
|8||46||F||10||Cushing||Pluri||ACTH, PRL, LH, α-SU||1.2%||(++)||(−)||(++)||Homo||None|
|9||46||M||22||PRL||Pluri||ACTH, PRL, TSH, LH, FSH, α-SU||2.9%||(++)||(++)||(+++)||Hetero||None|
|10||42||M||37||Cushing||Pluri||ACTH, LH, FSH||2.7%||(++)||(++)||(+++)||Hetero||codon 45: TCT(Ser) to TCC(Pro)|
|11||38||F||7||Cushing||Pluri||ACTH, LH, FSH||0.8%||(++)||(++)||(+++)||Hetero||None|
|12||70||M||32||Non||Pluri||ACTH, FSH||1.0%||(++)||(+)||(+++)||Hetero||codon 43: GCT(Ala) to CCT(Pro)|
|13||52||M||34||Non||Pluri||ACTH, GH, PRL, LH, FSH, α-SU||0.8%||(++)||(+++)||(+++)||Hetero||None|
|14||57||F||63||Non||Pluri||ACTH, GH, LH, FSH||14.4%||(++)||(+)||(+++)||Hetero||None|
|16||33||F||26||Non||Pluri||ACTH, PRL, TSH, LH, FSH||1.9%||(+++)||(++)||(+++)||Hetero||None|
|17||31||F||16||Cushing||Pluri||ACTH, PRL, LH||1.4%||(+)||(++)||(++)||Hetero||None|
|18||48||F||32||Cushing||Gonado||ACTH, LH, FSH, α-SU||0.2%||(++)||(++)||(+++)||Hetero||codon 45: TCT(Ser) to TAT(Tyr)|
|19||50||M||31||Non||Gonado||LH, FSH, α-SU||1.7%||(++)||(++)||(+++)||Hetero||codon 43: TCT(Ser) to ACT(Thr)|
|20||45||F||42||Non||Gonado||ACTH, LH, α-SU||0.3%||(++)||(+)||(+++)||Hetero||None|
|22||34||F||32||GH||Pluri||GH, LH, FSH, α-SU||21%||(+)||(++)||(−)||None (PCR-SSCP)|
|23||32||M||52||PRL||Pluri||ACTH, PRL, LH, FSH, α-SU||4.2%||(+)||(++)||(−)||None (PCR-SSCP)|
|24||45||M||39||TSH||Pluri||TSH, α-SU||5.0%||(+)||(++)||(+/−)||None (PCR-SSCP)|
|25||52||F||50||Cushing||Pluri||ACTH, PRL, LH, FSH, α-SU||0.6%||(+)||(++)||(−)||None (PCR-SSCP)|
|26||52||M||25||Non||Pluri||LH, FSH, α-SU||1.8%||(++)||(++)||(+/−)||None (PCR-SSCP)|
|27||61||M||37||Non||Pluri||ACTH, GH, LH, FSH, α-SU||21%||(++)||(++)||(+/−)||None (PCR-SSCP)|
|28||43||F||42||Non||Pluri||ACTH, PRL, TSH, LH, FSH||0.3%||(+)||(++)||(−)||None (PCR-SSCP)|
|29||51||F||47||Non||Gonado||LH, α-SU||1.0%||(+)||(++)||(+/−)||None (PCR-SSCP)|
|30||67||F||21||Non||Gonado||LH, FSH, α-SU||1.2%||(+)||(++)||(+/−)||None (PCR-SSCP)|
|31||74||M||23||Non||Gonado||FSH, α-SU||0.3%||(+)||(++)||(+/−)||None (PCR-SSCP)|
|32||50||M||45||Non||Gonado||LH, FSH, α-SU||0.3%||(+)||(++)||(−)||None (PCR-SSCP)|
|33||65||F||55||Non||Gonado||ACTH, LH, FSH||0.6%||(+)||(++)||(+/−)||None (PCR-SSCP)|
|34||41||M||30||Non||Gonado||ACTH, LH, FSH, α-SU||1.8%||(+)||(++)||(+/−)||None (PCR-SSCP)|
|35||66||M||38||Non||Gonado||LH, FSH||0.9%||(+)||(++)||(+/−)||None (PCR-SSCP)|
|36||46||F||28||GH||GH & PRL||GH, PRL||0.5%||(+)||(++)||(+/−)||None (PCR-SSCP)|
|37||56||F||20||GH||GH & PRL||GH, PRL||0.7%||(+)||(++)||(+/−)||None (PCR-SSCP)|
Expression of CTNNB1 and Ki-67 Proteins
Monoclonal antibodies to the CTNNB1 and Ki-67 proteins were purchased from PharMingen (San Diego, CA) and Dakopatts (Glostrup, Denmark), respectively. A modification of the immunoglobulin (Ig) enzyme bridge technique (avidin-biotin complex method) was used as described elsewhere.20 Deparaffinized tissue sections were immersed in methanol containing 0.03% hydrogen peroxide for 30 minutes to block the endogenous peroxide activity. For the detection of CTNNB1 and Ki-67, microwave pretreatment in citrate buffer was performed for 15 minutes to retrieve the antigenicity. The sections then were incubated with normal horse serum (diluted 1:20) for 30 minutes to block the nonspecific antibody binding sites. The sections were treated consecutively at room temperature with anti-CTNNB1 (diluted 1:200) and anti-Ki-67 (diluted 1:100) monoclonal antibodies for 90 minutes, biotinylated antimouse IgG horse serum (diluted 1:100) for 30 minutes, and avidin DH-biotinylated horseradish peroxide complex (Vecstain ABC kit; Vector Laboratories, Burlingame, CA) for 30 minutes. Peroxide staining was performed for 10–15 minutes using a solution of 3,3′-diaminobenzidine tetrahydrochloride (DAB) in 50 mM Tris-HCl (pH 7.5) containing 0.001% hydrogen peroxide. The sections were counterstained with 0.1% methyl green. All the immunostained slides were observed by two investigators (S.S. and H.I.) independently to make the grading as objective as possible. We then analyzed other clinicopathologic parameters. The immunoreactivity of the andti-CTNNB1 antibody was graded as (−) to (+++) according to the number of stained cells, and the staining intensity in individual cells was graded as follows: (−): nearly no positive cells; (+): 5–25% of tumor cells showed weak to moderate immunoreactivity; (+ +): 25–50% of tumor cells showed moderate immunoreactivity or 10–50% of tumor cells showed intense immunoreactivity; and (+ + +): > 50% of tumor cells showed intense immunoreactivity.20 Tumors graded as (+ +) and (+ + +) were regarded as strongly positive. The Ki-67 labeling index was determined by counting the number of positive cells in a total of ≥1000 tumor cells observed in ≥ 10 representative high-power fields (×400).
Analysis of Hormone Production in Pituitary Adenoma
We also used immunohistochemical analysis to analyze the hormones produced in pituitary adenomas. The antibodies used and their dilutions were as follows: monoclonal anti-ACTH (1-39) antibody (Dakopatts) (1:75); monoclonal anti-PRL antibody (Immunotech, Westbrook, ME) (1:300); polyclonal anti-GH antibody (Dako Company, Carpinteria, CA) (1:1); monoclonal anti-TSH antibody (CosmoBio Co. Ltd., Tokyo, Japan) (1:60); monoclonal anti-LH antibody (Immunotech) (1:70); and monoclonal anti-FSH antibody (CosmoBio Co. Ltd.) (1:60).
The relations between the results of the immunohistochemical study and clinicopathologic parameters were investigated using Spearman rank correlation analysis. When a significant difference was observed, further characterization using the nonparametric Mann–Whitney U test was performed. Multivariate analysis used the logistic model with multiple regression in the Statview-J 4.02 software package (SAS Institute, Cary, NC). A P value < 0.05 was considered to be statistically significant. Values are presented as the mean ± the standard deviation.
Mutation Analysis of the CTNNB1 Gene
The cases with nuclear accumulation of CTNNB1 were analyzed for genetic alterations by direct sequencing of exon 3 in the gene using polymerase chain reaction (PCR)-amplified products. Primers used for PCR amplification and sequencing were as follows: for β-f, it was 5′-TTAGTCACTGGCAGCAACAG-3′ and for β-r, it was 5′-CTCTTCCTCAGGATTGCCTT-3′. Each 15-μL reaction mixture containing 10 ng of DNA, 6.7 mM Tris-HCl (pH 8.8), 16.6 mM (NH4)2SO4, 10 mM β-mercaptoethanol, 6.7 μM ethylenediamine tetraacetic acid, 6.7 mM MgCl2, 1 μM of the primer pair, 1.5 mM of each deoxynucleotide, 10% (volume/volume) dimethylsulfoxide, and 0.75 units of Taq DNA polymerase was amplified for 40 cycles using the following regimen: denaturation at 94 °C for 30 seconds, annealing at 55 °C for 30 seconds, and extension at 72 °C for 30 seconds. Nucleotide sequencing was performed using a Thermo-Sequenase dye terminator cycle sequencing premix kit (Amersham, Little Chalfont, UK) and the ABI PRIZM 310 Genetic Analyzer (Perkin-Elmer, Foster City, CA) according to methods previously described.21 Both sense and antisense strands were sequenced to confirm the results. For cases with tumors with negative immunostaining, PCR-single strand conformation polymorphism (SSCP) analysis was performed to screen for mutations in the CTNNB1 gene, as previously described.22
We first analyzed the expression of CTNNB1 in pituitary adenomas immunohistochemically. Examples are shown in Figure 1. In the normal pituitary gland, CTNNB1 was expressed mainly at the cell membrane but showed weak expression in the cytoplasm and nucleus (Fig. 1A). Case 16 showed remarkable nuclear accumulation of CTNNB1 whereas a marked decrease in expression was noted at the cell junction (Fig. 1B). We considered this pattern of expression to be homogenous. Conversely, remarkable accumulation of CTNNB1 was observed in the nuclei as well as at the cell membrane (Case 2) (Fig. 1C). We considered this pattern of expression to be heterogeneous. These results are summarized in Table 1. Among the 37 pituitary tumors, 21 (57%) cases demonstrated abnormal nuclear accumulation of the CTNNB1 protein. In addition, cytoplasmic CTNNB1 was detected in all 21 cases that showed nuclear CTNNB1 accumulation. Thirteen of these 21 CTNNB1-positive cases (62%) showed heterogeneous expression whereas 8 cases (38%) showed homogeneous expression.
We then analyzed the correlation between nuclear CTNNB1 accumulation and clinicopathologic features. Abnormal accumulations of CTNNB1 protein were detected in 13 of 20 plurihormonal adenomas (65%), 5 of 12 gonadotroph adenomas (42%), and 2 of 4 mixed GH and PRL cell adenomas (50%). The one case of silent ACTH adenoma also showed nuclear accumulation of CTNNB1. Although we also examined the relation between the expression pattern of CTNNB1 protein (heterogeneous or homogeneous) and the clinicopathologic findings, no statistical correlation was observed with regard to gender, age, tumor size, or clinical symptoms including recurrence.
We next analyzed the immunoreactivity of Ki-67 and examined the relation between nuclear staining of CTNNB1 and proliferative activity in pituitary adenomas. It was not possible to examine normal pituitary glands as the control because normal tissues were not available. Therefore we scored Ki-67 immunoreactivity based on the number of positive cells per 1000 tumor cells. An example of Ki-67 expression is shown in Figure 1D (Case 16). The tumor cells showed weak to moderate levels of expression. In total, 21 of 37 cases (57%) showed > 1.0% Ki-67 immunoreactivity (Table 1). Cases 14 and 21 showed > 10% Ki-67 immunoreactivity. The sizes of these tumors were 63 mm and 39 mm, respectively, at the time of surgery, but there were no signs of malignancy histopathologically. In these cases, radiation therapy also was added but one of these two cases recurred after the initial surgery (Case 14). Using the Mann–Whitney U test, a statistically significant difference (P < 0.05) was observed between the expression of Ki-67 (> 1.0%) and CTNNB1 nuclear accumulation.
To collect evidence regarding whether nuclear transport might be caused by a somatic mutation, we further investigated genetic alterations of CTNNB1 at exon 3, which harbors hot spots for activating mutations of the gene. Nucleotide sequencing analyses were performed in all tumors that showed immunohistochemically abnormal accumulations of CTNNB1 in their nuclei. Examples of the mutational analyses are shown in Figure 2. Four pituitary adenomas had somatic mutations at exon 3 of the CTNNB1 gene, including two plurihormonal adenomas (Cases 10 and 12) and the gonadotroph adenomas (Cases 18 and 19). Missense mutations at codon 45 encoding a serine residue were detected in Cases 10 and 18 whereas missense mutations involving codon 43, encoding alanine, were observed in Cases 12 and 19 (Table 1) (Fig. 2). We further analyzed the nucleotide sequences of corresponding normal tissues and found that these were somatic mutations. It is interesting to note that all these samples exhibited heterogeneous expression of the CTNNB1 protein. In tumors without abnormal accumulations of CTNNB1, no abnormalities were detected by PCR-SSCP (data not shown).
It has been reported that nuclear accumulation of CTNNB1 is detected in several human malignancies, including tumors of the gastrointestinal tract, breast, endometrium, and thyroid gland.23–28 However, to our knowledge the current study is the first report demonstrating the abnormal nuclear accumulation of CTNNB1 in pituitary adenomas; 57% of pituitary adenomas showed accumulations of CTNNB1 in the nucleus, suggesting that abnormal nuclear accumulation of CTNNB1 plays an important role in pituitary tumorigenesis. We also detected several somatic mutations of the CTNNB1 gene in the pituitary adenomas examined. Missense mutations at serine/threonine residues in exon 3 of this gene, including sites phosphorylated by GSK3β, are believed to permit escape of the CTNNB1 protein from ubiquitin-mediated protein degradation, and finally to contribute to its stabilization.16, 27 Missense mutations at codon 45 encoding Ser were observed in two cases. In addition, mutations at codon 43 encoding alanine also were observed in two cases; to our knowledge mutations in this codon have never previously been reported. It is not clear whether each of the mutations at codon 43 contributes to nuclear translocation and/or to stabilization of the CTNNB1 protein. Alternatively, a CTNNB1 somatic mutation at codon 43 may cause escape from CTNNB1 protein phosphorylation by GSK3β coupled with ubiquitin-mediated CTNNB1 proteolysis.28, 29
We classified the accumulation pattern of CTNNB1 into two groups: heterogeneous staining and homogeneous staining. The tumors classified in the former group displayed CTNNB1 expression at the junction region (cell membrane) as well as the nucleus; membranous staining was decreased markedly in the latter group. There might be several possible explanations for these two different expression patterns: 1) there could be wild-type CTNNB1 expression in the group with the heterogeneous expression pattern but not in the group with the homogeneous expression pattern; 2) some unknown mechanisms may exist to suppress normal CTNNB1 expression in the tumors with the homogenous expression pattern, such as DNA methylation of the promoter region of CTNNB1 and/or an abnormality in the ubiquitin-mediated proteolytic system; or 3) CTNNB1 protein could not be located at the cell membrane because of loss or abnormalities in the APC or cadherin proteins.15, 30, 31
The percentage of abnormal accumulation of the CTNNB1 protein in the current series of pituitary adenomas is surprisingly high. However, clinical and biologic behaviors, including recurrence, were not found to be correlated with the presence or absence of nuclear accumulation of the CTNNB1 protein. These results may suggest that abnormality of CTNNB1 is a rather early event in pituitary tumorigenesis.
Protein expression levels of α- and CTNNB1 were down-regulated by dexamethasone in rat 235-1 pituitary tumor cells that produce high levels of PRL, and these dexamethasone-treated cells showed altered cell-cell interaction.32 Loss of expression of the E-cadherin protein was observed in all five pituitary adenomas examined by Schwechheimer et al.33 However, Kawamoto et al. observed the expression of E-cadherin in all 30 pituitary adenomas they examined, and they reported no correlations between E-cadherin expression and cavernous sinus invasion.34 The results of the current study, which showed no correlation between clinical and pathologic features and abnormal expression of CTNNB1, are in agreement with this report. An abnormality in the CTNNB1 protein most likely is one of the most important events in the molecules of the cadherin-catenin superfamily that play a role in pituitary tumorigenesis.
With regard to other endocrine tissues, frequent nuclear accumulation of CTNNB1 and somatic mutation of this gene also have been reported in thyroid anaplastic carcinoma.35 This tumor is well recognized as one of the most aggressive and it has been suggested that its malignant behavior may associated with mutant CTNNB1 acting as an oncoprotein. We also have reported the frequent accumulation of CTNNB1 protein in carcinoid tumors arising from the digestive tract and islet cell tumors of the pancreas, which have potentially malignant behavior.36 As described earlier, accumulation of mutant CTNNB1 may induce malignant tumor behavior, although no obvious findings were detected in the current series of pituitary adenomas. Further investigations will be needed for the clarification of these issues.
Expression of Ki-67 antigen was considered to be a proliferative marker because the cell population detected by the Ki-67 antibody is referred to by its growth fraction (early G1, S-, G2, and M-phase).37 In the current study, nuclear accumulation of CTNNB1 was associated with high Ki-67 expression. This may suggest that abnormally expressed CTNNB1 plays an important role in pituitary tumorigenesis through up-regulation of cell proliferation and tumor progression. Previous studies have reported that the complex of mutant CTNNB1 and Tcf may induce expression of the c-myc and/or cyclin D genes in colon carcinoma cell lines.14, 17, 18 In light of this connection, it is of interest to investigate the status of these genes in pituitary adenomas.
The authors thank Dr. Barbara Lee Smith Pierce (the Life Science Coordinator for the University of Maryland Asian Division) for her editorial work in the preparation of this article.
- 19Analysis and cloning of eukaryotic genomic DNA #9.14. Isolation of high-molecular weight DNA from mammalian cells. In: SambrookJ, FritshEF, ManiatisT, editors. Molecular cloning. 2nd edition. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press, 1989: 16–23., , .
- 36Nuclear accumulation of β-catenin in human endocrine tumors: association with Ki-67 (MIB-1) proliferative activity. Endocr Pathol. In press., , , .