| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Studies |
The Institute of Genetics (M.D., A.A., G.B., B.G., E.F.) and the Oncogenetics Unit (E.F.) Chaim Sheba Medical Center, Tel-Hashomer, Israel; and the Department of Neurosurgery, Kopfklinikum (E.F.A., M.B., R.F.), Erlangen, Germany
Address all correspondence and requests for reprints to: Eitan Friedman, M.D., Ph.D., The Suzanne Levy Gertner Oncogenetics Laboratory, Institute of Genetics, Chaim Sheba Medical Center, Tel-Hashomer 52621, Israel. E-mail: feitan@post.tau.ac.il, or eitan211{at}netvision.net.il
| Abstract |
|---|
|
|
|---|
5q23), 9p (9p21
9pter),
13q (13q21
13q32), and 17q were detected in 1030% of the tumors.
Noteworthy, no tumor displayed deletion of 11q, the MEN1
gene locus. These findings suggest that genes localized to previously
undescribed chromosomal regions play a role in the tumorigenesis of
nonfunctioning pituitary adenomas. | Introduction |
|---|
|
|
|---|
To the best of our knowledge, the only genetic alterations described in nonfunctioning pituitary tumors were gsp mutations (9) and allelic loss at 11q13 (12). By and large, the genetic alterations that were sought, and frequently not found, in pituitary neoplasms were in candidate genes (such as p53, ras, and retinoblastoma) or candidate chromosomal regions [11q (at the MEN1 gene locus) (13) and 13q (at the retinoblastoma locus (14)]. The difficulty in ascertaining which chromosomes harbor genes that are involved in pituitary tumorigenesis is partially attributed to the technical difficulty in obtaining tumor-specific karyotypes. Comparative genomic hybridization (CGH), provides an opportunity to scan the entire human genome and localize genetic alterations of medium size to a specific chromosomal region in a single experiment (15). It has been estimated that an amplicon as small as 100 kilobases can be detected by CGH, whereas the detection limits of deletions are about 12 megabases (16). In this study, we applied CGH to 23 nonfunctioning pituitary tumors to define which chromosomal regions are putatively involved in pituitary tumor devel-opment.
| Materials and Methods |
|---|
|
|
|---|
High mol wt DNA was isolated from the resected pituitary tumor at the same time when primary pituitary cell cultures were set up, as previously described in detail (17). DNA isolation from the tumor (i.e. test DNA) was performed by standard techniques. Reference DNA was extracted from peripheral blood of karyotypically normal males, using standard protocols.
Pathological features, immunocytochemistry, and hormone assays
All tumors were examined by a neuropathologist using separate formalin-fixed, paraffin-embedded, hematoxylin-eosin-stained sections. All immuncytochemical studies were performed using a standard peroxidase-antiperoxidase method (18) with rabbit antibodies against PRL, GH, ACTH, TSH, LH, and FSH (Dako, Munich Germany). Cryostat sections of 510 µm were prepared and fixed in cold acetone for 15 min. Endogenous peroxidase activity was blocked with 3% H2O2 in phosphate-buffered saline for 5 min, followed by staining with rabbit antiserum. Sections were developed with 0.03% 3,3-diaminobenzidine tetrahydrochloride containing 0.01% H2O2 in Tris buffer.
Preoperative serum levels of hormones were determined by the ELISA method, using commercially available kits. The kits for GH and PRL were purchased from ELIAS (Freiburg, Germany), and those for LH, FSH, TSH and ACTH were obtained from Ares-Serono (Geneva, Switzerland).
Genomic DNA probes and labeling procedure
One microgram from each test and reference DNA was directly labeled with spectrum-Green-deoxy-UTP and spectrum-Red-deoxy-UTP (Vysis, Downers Grove, IL), respectively, by nick translation using the Vysis kit, according to the manufacturers recommendations. Fragment lengths after nick translation ranged from 500-2000 bp, as measured after separation on 1.5% agarose gels (19).
CGH
CGH was performed as previously described (20) with minor modifications. High quality metaphase spreads were prepared from a karyotypically normal male according to standard protocols. Slides were prepared at least 1 day before hybridization, and were kept at room temperature for a maximal period of 2 weeks. Slides were denatured in 70% formamide and 2 x standard saline citrate (SSC) at 70 C for 2 min and dehydrated in a series of immersions in ice-cold ethanol. One microgram each of spectrum-Green-labeled test DNA and spectrum-Red-labeled reference DNA, and 40 µg unlabeled Cot-1 DNA (Life Technologies, Gaithersburg, MD) were mixed, ethanol precipitated, and resuspended in 14 µL hybridization buffer (50% formamide and 10% dextran sulfate in 2 x SSC). This probe mixture was denatured for 5 min in 75 C, preannealed for 30 min at 37 C, and hybridized on slides containing denatured chromosomes using a glass coverslip sealed with rubber cement. Hybridization took place in a moist chamber at 37 C for 72 h. Slides were washed in 0.4 x SSC at 75 C for 2 min, followed by 0.1% Nonidet P-40 in 2 x SSC for 2 min at room temperature. Target chromosomes were counterstained with 4,6-diamino-2-phenylindole (DAPI; Sigma Chemical Co., St. Louis, MO) and mounted with Vectashield (Vector Laboratories, Burlingame, CA).
Digital image analysis
CGH results were analyzed using an epifluorescence microscope (Zeiss Axioscope, Jena, Germany) equipped with a 100-watt mercury lamp and a cooled charge-coupled device camera (Photometrics, Tucson, AZ) controlled by a Cytovision image analysis system (Applied Imaging International, Tyne and Wear, UK). Analysis was performed as previously described (21). Briefly, 1015 metaphases were chosen for image analysis with high fluorescence intensity and uniform hybridization. Gray level images were captured separately for each fluorochrome (DAPI, spectrum-Green, and spectrum-Red). Karyotyping performed based on DAPI banding by inverting the DAPI image. Green and red fluorescence intensities were determined along each chromosome medial axis at 1-pixel intervals. Green to red fluorescence intensity ratio profiles were calculated after background correction and normalization of the green to red ratio for each metaphase to 1.0. Mean ratio profiles for each chromosome were determined after data from all analyzed metaphases were combined. Trisomies or partial chromosome gains were defined as having a green/red ratio greater than 1.25. Monosomies or partial chromosome losses were defined as having a green/red ratio less than 0.75 (22).
| Results |
|---|
|
|
|---|
Each patients age and sex, the tumor size [as determined by
Wilsons staging and grading system (23)], peripheral blood hormone
profile of patients, and results of immunohistochemical analyses and
immunocytochemical analyses are summarized in Table 1
. There were 12 men and 11 women, with
an age range of 3378 yr. None of the 23 patients with tumors
clinically classified as nonfunctioning had detectable hormonal
hypersecretion in the serum, but in 18 tumors there was
immunohistochemical evidence or in vitro culture evidence of
the presence of gonadotropins (Table 1
). Tumor diameters ranged from
1345 mm, and invasion of the pituitary tumors into the surrounding
anatomical structures was evident in 6 cases. Three patients (no. 920,
925, and 936) underwent reoperation within 6 months after
transsphnoidal resection because of local tumor recurrence.
|
The results of CGH analysis are summarized in an ideogram (Fig. 1
) and in Table 2
, and examples of chromosomal gains and
losses are shown in Fig. 2
. There were
five tumors in which no genetic abnormality was detected by this
analysis (no. 907, 911, 936, 946, and 964). At the other extreme, there
were five tumors (no. 910, 914, 926, 967, and 980) that displayed
chromosomal aberrations involving seven or more chromosomes.
Interestingly, one of the tumors with no detectable genetic abnormality
(no. 936) required reoperation for local recurrence, whereas none of
the tumors displaying multiple aberrations required such a procedure.
Noteworthy, the tumors displaying multiple genetic aberrations had
unique patterns of abnormalities: four tumors in this group displayed a
gain of the long arm of chromosome 4 not exhibited by any other tumor
sample. In addition, multiple gains of entire chromosomes were noted
specifically in this group of tumors; three of five tumors had
overrepresentation of at least three different chromosomes. This
phenomenon was not detected in any of the other tumor samples (Fig. 1
).
|
|
|
4q31.1, 5q at 5q13
5q23, and
13q at 13q21
13q32, were found in five samples each (21.7%). Gain of
chromosome 9p (9p12
9pter) was found in four samples (17.4%), and
gains in chromosomes 1q, 8p, and 16q were each found in three samples
(13%). These latter amplifications were consistently localized to a
well defined, relatively small, subchromosomal region. No tumor
displayed loss of the long arm of chromosome 11 (the MEN1
gene locus) (13), and only one had loss of an entire copy of chromosome
13 (the Rb gene locus) (14). In addition, one tumor displayed
amplification of 20q at band 13.3, distal to the chromosomal region
harboring the gene for the
-subunit of the stimulatory GTP-binding
protein, Gs
(24). | Discussion |
|---|
|
|
|---|
Two types of genetic alterations can be detected by CGH: amplifications and deletions. DNA amplification is a well recognized genetic abnormality in tumors, traditionally revealed by Southern blot analysis. The amplified DNA presumably contains direct acting oncogenes that are overexpressed, resulting in uncontrolled cellular proliferation and tumor development (27). The amplifications of 4q, 5q, 8p, 9p, 13q 16q, and 17 detected in nonfunctioning pituitary tumors seem to be nonrandom, affecting four or more tumors, or the specific amplicon is localized to a relatively small, subchromosomal region. Some of these chromosomal regions are amplified in a variety of other tumors, as determined by CGH: 5q amplification in hereditary renal cell carcinoma (28) 13q in uveal melanomas (29), and 17q in breast cancer (19). The long arm of chromosome 17 contains several known genes that have been implicated in tumorigenesis; c-erbB-2 plays a significant role in the pathogenesis of breast cancer (30), ovarian cancer (31), and neuroblastoma (32), but apparently not in pituitary tumors (33). A region on the long arm of chromosome 5 is duplicated in renal cell carcinomas (34), and the distal region of chromosome 9p was reported in primitive neuroectodermal tumors (35).
Chromosomal losses and deletions are indicative of the involvement of tumor suppressor genes in tumorigenesis (36). Despite the relative insensitivity of CGH in detecting losses, nonrandom deletions affecting several chromosomal loci were identified in prostate cancer (37). Furthermore, in that study there was a 76% concordance between the CGH results and allelic loss genotype determined by a PCR-based method. Thus, the chromosomal loss of 18q seen in our series might reflect bona fide allelic loss of chromosome 18, targeting the DCC gene (38) or other tumor suppressor genes that are apparently involved in prostate cancer (37).
Genetic alteration involving the X chromosome in a substantial number of tumors (34.7%) in our series is interesting, as it harbors no obvious candidate genes, and its involvement in tumorigenesis has been reported sporadically and anecdotally (37). Unexpectedly, 11q was not deleted in any tumor, and only one tumor displayed allelic loss of chromosome 13. These regions contain the MEN1 (13) and retinoblastoma (14) genes, respectively, which have been implicated in pituitary tumorigenesis (11). Indeed, somatic mutations of the recently cloned MEN1 gene (menin) (39) have been detected in a subset of pituitary tumors, as direct evidence of their involvement in pituitary neoplasia (40). It is possible that genetic changes exist in these regions that are below the CGH resolution and play a role in the development of pituitary tumors. Alternatively, our findings might indicate the infrequent involvement of these genes in the tumorigenesis of nonfunctioning pituitary tumors. By the same token, other genes localizing to chromosomal regions displaying amplification or losses are acting as oncogenes or tumor suppressor genes, respectively, and are probably involved in pituitary tumor development.
The possibility that the genetic alterations detected by CGH in this series are secondary events, subclonal expansions of the monoclonal tumor, needs to be considered. However, this seems unlikely because all analyzed tumors were benign, and only a fraction of them recurred. Secondary genetic events are primarily observed in malignant invasive tumors with a rapid proliferation rate. Furthermore, the nonrandom distribution coupled with the subchromosomal localization of the genetic alterations strongly support the idea that these abnormalities are relevant to and associated with the neoplastic process.
In summary, nonrandom genetic alterations affecting previously undescribed chromosomal regions in pituitary tumorigenesis are detected by CGH. This finding further supports the monoclonal origin of these neoplasms and points out that genes other than the currently appreciated candidate genes may be involved in pituitary neoplasia.
| Footnotes |
|---|
Received December 17, 1997.
Revised January 28, 1998.
Accepted February 4, 1998.
| References |
|---|
|
|
|---|
gene in non-functioning
pituitary tumors. J Clin Endocrinol Metab. 77:765769.[Abstract]
This article has been cited by other articles:
![]() |
A. Abu, M. Frydman, D. Marek, E. Pras, C. Stolovitch, A. Aviram-Goldring, S. Rienstein, H. Reznik-Wolf, and E. Pras Mapping of a Gene Causing Brittle Cornea Syndrome in Tunisian Jews to 16q24 Invest. Ophthalmol. Vis. Sci., December 1, 2006; 47(12): 5283 - 5287. [Abstract] [Full Text] [PDF] |
||||
![]() |
W E Farrell Pituitary tumours: findings from whole genome analyses. Endocr. Relat. Cancer, September 1, 2006; 13(3): 707 - 716. [Abstract] [Full Text] [PDF] |
||||
![]() |
P. Vigano, E. Somigliana, I. Chiodo, A. Abbiati, and P. Vercellini Molecular mechanisms and biological plausibility underlying the malignant transformation of endometriosis: a critical analysis Hum. Reprod. Update, January 1, 2006; 12(1): 77 - 89. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. Zhao, D. Dahle, Y. Zhou, X. Zhang, and A. Klibanski Hypermethylation of the Promoter Region Is Associated with the Loss of MEG3 Gene Expression in Human Pituitary Tumors J. Clin. Endocrinol. Metab., April 1, 2005; 90(4): 2179 - 2186. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Ezzat Pituitary Tumor Pathogenesis--The Hunt for Novel Candidate Genes Continues J. Clin. Endocrinol. Metab., November 1, 2003; 88(11): 5116 - 5118. [Full Text] [PDF] |
||||
![]() |
J. Gogusev, J. B. de Joliniere, L. Telvi, M. Doussau, A. Stojkoski, and M. Levradon Cellular and Genetic Constitution of Human Endometriosis Tissues Reproductive Sciences, March 1, 2000; 7(2): 79 - 87. [Abstract] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |