| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Division of Metabolism, Endocrinology, and Diabetes, Department of Internal Medicine (T.E., T.J.G., G.D.H.), and Department of Pathology (T.J.G.), University of Michigan Health System, Ann Arbor, Michigan 48109-2200
Address all correspondence and requests for reprints to: Gary D. Hammer, University of Michigan Medical School, 109 Zina Pitcher Place, BSRB 1502, Ann Arbor, Michigan 48109-2200. E-mail: ghammer{at}umich.edu; or Tobias Else, University of Michigan Medical School, 109 Zina Pitcher Place, BSRB 1860, Ann Arbor, Michigan 48109-2200. E-mail: telse{at}umich.edu.
 |
Abstract |
|---|
 Top Abstract IntroductionPatients and MethodsResults and DiscussionReferences |
|---|
Objective: Tissue samples of diagnostically known adrenocortical neoplasms were evaluated for parameters of known TMMs, telomerase activity (TA), and alternative telomere lengthening (ALT).
Design: The study analyzed retrospectively collected frozen adrenocortical tissue samples from the University of Michigan Health System.
Patient Samples: Samples included 24 ACCs, 11 adrenocortical adenomas (ACAs), and three normal adrenal tissues.
Main Outcome Measures: Telomerase activity (telomerase activity protocol assay), alternative telomere lengthening (telomere restriction fragment analysis, telomere associated promyelocyte leukemia bodies) were measured.
Results: A total of 22 of 24 ACCs (92%) could be definitively assigned to a TMM. The TMM classification was: 19 of 24 TA (79%), two of which displayed very long telomeres, one of 24 ALT (4%) and two of 24 (8%) TA and ALT. Results of two of 24 (8%) were inconclusive (one negative for TA and positive in one ALT assay, one negative in all assays). None of the normal adrenal tissues (none of three) or ACA (none of 11) samples had signs of an active TMM.
Conclusions: TA is the main TMM in the majority of ACCs, but subsets of ACCs additionally or exclusively exhibit signs of ALT. Determination of telomere maintenance mechanisms in diagnostically challenging adrenocortical tumors might be of additional diagnostic value in the pathological diagnosis of malignant vs. benign lesions.
|
Introduction |
|---|
|
TopAbstractIntroductionPatients and MethodsResults and DiscussionReferences |
|---|
Telomerase is a ribonucleoprotein that elongates the 3' end of chromosomes to circumvent the end replication problem that leads to a loss of telomeric sequences over consecutive cell cycles and eventually to a removal of cells with critically short telomeres from the proliferating pool by the induction of apoptosis or senescence. Normal human somatic cells lack telomerase activity (TA), and only some constantly proliferating or self-renewing tissues such as germ cells, skin, and the lymphoid system display significant TA (7, 8, 9). Proliferating tumor cell clones often reactivate telomerase expression and activity. Roughly 90% of malignant human tumors make use of this telomere maintenance mechanism (TMM) (10, 11). Unfortunately, the few studies that have surveyed TA as a potential marker for malignancy in the adrenal gland reported conflicting results, presumably due to the limitations of the following: 1) small sample sizes (fewer than eight ACCs), 2) lack of conclusive final pathological diagnosis, 3) questionable tissue quality, or 4) differences in assay systems used (12, 13, 14, 15, 16).
Recently telomerase-independent alternative telomere maintenance mechanisms defined as alternative telomere lengthening (ALT) have been described in several human tumors, mainly sarcoma-derived tumor cell lines (17, 18, 19, 20). In a subset of ALT cell lines, homologous recombination between sister chromatids is used as a TMM, leading to an increased overall telomere length as well as length variation, compared with TA cell populations (21, 22). Although the exact mechanisms of ALT need to be elucidated, the presence of telomere-associated promyelocyte leukemia (PML) bodies correlates well with the existence of ALT (23). PML nuclear bodies are multifunctional subnuclear components, which have been shown to play a role in tumor-suppressive functions such as p53-mediated apoptosis and senescence or Rb-mediated transcriptional repression (24, 25). In ALT cell lines a bright telomere fluorescence in situ hybridization (FISH) signal colocalizes with PML immunostaining.
In a study investigating TMMs (TA and ALT) in liposarcomas, TMMs could be identified in approximately half of the neoplasms. A total of 25.9% of tumors had TA, 26.6% were classified as ALT, and a small number was positive for both TMMs. This study shows that different TMMs are not mutually exclusive and that the underlying molecular mechanism needs to be further elucidated (19). The primary purpose of our study was to evaluate the different TMMs in ACCs, using the high-quality tissues collected at the University of Michigan Health System. A pathologically defined large sample set of ACCs (24 samples) and control tissues were scored for TMMs, using the telomere repeat amplification protocol (TRAP) assay as a direct measurement of TA and immunostaining for telomere-associated PML bodies and telomere-length restriction fragment (TRF) analysis as surrogate parameters for ALT.
|
Patients and Methods |
|---|
|
TopAbstractIntroductionPatients and MethodsResults and DiscussionReferences |
|---|
Patient samples (Table 1
) were collected at the University of Michigan via the Tissue Procurement Service with institutional review board approval. All samples were obtained from primary surgery except samples 20, 21, and 28, which were from surgery of recurrent ACC. Tissue sections were reviewed by an endocrine pathologist, and representative parts were processed to paraffin tissue array blocks containing normal adrenocortical tissues (NATs), adrenocortical adenomas (ACAs), ACCs, and other reference organ tissues. Tissue samples were frozen in optimum cutting temperature compound and stored at –80 C. To select representative tissue samples, a section of tissue (NAT, ACA, or ACC) was marked on a standard hematoxylin and eosin-stained frozen section. The corresponding tissue was then microdissected from the tissue block and further kept at –80 C or on dry ice. This method of tissue preparation avoids multiple freeze and thaw cycles and ensures tissue integrity as well as tissue pathology of the obtained tissue sample. Three NATs, 11 ACAs, and 24 ACCs were analyzed for telomere length, TA, and telomere-FISH/PML staining. Initial samples (two NATs, two ACAs, and 24 ACCs) were blind coded for the scoring researcher. The number of ACAs and NATs was increased for revision of this publication, and additional samples were therefore not blind coded.
|
An adrenocortical-specific tissue array was constructed using 1-mm cores from paraffin-embedded blocks of the University of Michigan archives. Combined telomere-FISH and immunohistochemistry was carried using the protocol of Meeker et al. (26) with slight modifications. Briefly, deparaffinized slides were hydrated through a graded ethanol series, transferred to ddH2O, and then boiled in citrate buffer [10 mM sodium citrate, 0.05% Tween 20 (pH 6.0)]. Slides were then washed in PBST (PBS, 0.05% Tween 20), rinsed in ddH2O, transferred to 95% ethanol, and air dried. The hybridization mixture [0.3 µg/ml peptide nucleic acid probe Cy3(CCCTAA)3, 70% formamide, 10 mM Tris, (pH 7.5), 0.5% blocking reagent] was applied, and slides were heated to 83 C on a glass plate in a prewarmed hybridization oven. Hybridization was carried out for at least 2 h at room temperature in the dark. Slides were washed twice in 70% formamide, 10 mM Tris (pH 7.5), and 0.1% BSA and once in PBST and then incubated overnight o/n at 4 C with anti-PML (no. 05-718; Upstate, Lake Placid, NY) 1:100 in PBST. Primary antibody was detected with a goat antimouse Alexa-Fluor 486-coupled antimouse IgG (no. A11029; Invitrogen, Carlsbad, CA; 1:200 in PBST, 1 h, room temperature), counterstained with 4',6'-diamino-2-phenylindole (Sigma, St. Louis, MO; 1:2000, PBS), and washed twice in PBS. Cells grown on slides were subjected to the same protocol but not air dried before hybridization. Images were analyzed using an Optiphot-2 microscope (Nikon, Melville, NY) with a DP-70 camera and software system (Olympus, Hauppauge, NY). Only samples with a bright nuclear telomere-FISH signal (detectable at much lower exposure time than the usual telomere signal) colocalizing with anti-PML staining were counted as ALT positive in this procedure (>8 colocalization/high-power field HPF).
Cell culture
Cell culture media used in this study were as follows: NCIh295A: RPMI 1640, 2.5% bovine serum, and 1% insulin/transferrin/selenium; U2OS: McCoy’s 5a medium, 10% fetal bovine serum (FBS); HeLa: DMEM, 10% FBS; SW13: DMEM, 10% FBS; RL251: RPMI 1640 10% FBS (27, 28). For immunohistochemical analysis, cells were grown on fibronectin (Sigma; 1:40 in serum-free media)-coated slides overnight, fixed in 3.7% formaldehyde in PBS for 10 min, and permeabalized 2 min at room temperature in 0.5% Nonidet P-40 in PBS. Slides were stored in 100% methanol at –20 C.
TRAP
Telomerase activity was determined by the TRAP using a commercially available kit (no. S7700; Chemicon, Temecula, CA) according to the manufacturer’s recommendation. Tissue samples were gently crushed on liquid nitrogen and resuspended in [(3-cholamidopropyl) dimethyl-ammonio]-1-propane-sulfonate buffer. One microgram of protein was used for the TRAP assay. For cell lines and controls, the equivalent of 5 x 103 and 5 x 102 cells was used. A 32P-end labeled primer was used in the TRAP reaction and dried gels were exposed to film for an average of 4 h. Sample lanes with a clearly visible ladder-type pattern were assessed as telomerase positive. All samples were run with a heat-inactivated negative control, and every run was controlled by a sample of 5 x 102 HeLa cells. Negative samples were spiked with the equivalent of 5 x 102 HeLa cells and reevaluated to exclude telomerase or polymerase inhibitors interfering with the assay.
TRF length assay
TRF analysis used published protocols (29) with modifications. Tissues (
100 µg) gently ground in liquid nitrogen as well as the nonsoluable fraction of the TRAP protocol [(3-cholamidopropyl) dimethyl-ammonio]-1-propane-sulfonate lysis were resuspended in 100 µl PBS and mixed with 100 µl 2% LMP-agarose (SeaPlaque; Lonza, Rockland, ME). Plugs of this mixture were digested for 36 h in 2% sarcosyl, 500 mM EDTA, and 1 mg/ml proteinase K. After three washes in Tris/EDTA and preincubation in 500 µl Dpn II buffer (NEB, Ipswich, MA), genomic DNA was digested with 60 U Dpn II in fresh buffer for 36 h. Plugs and a low range PFG marker (NEB) were run in 1% agarose (Seakem, Lonza) using a CHEF mapper (Bio-Rad, Hercules, CA). The automatic algorithm was set to a range of 1–80 kb. The DNA gels were depurinated (0.25 N HCl), denatured (0.5 N NaOH, 1.5 M NaCl), neutralized [0.5 M Tris, 1.5 M NaCl (pH 6)], and transferred to 
-probe GT membrane (Bio-Rad). The membrane was baked 2 h at 80 C, prehybridized [4x Denhardts, 6x saline sodium citrate (SSC), 0.1% sodium dodecyl sulfate (SDS)[, and hybridized with purified 32P-end labeled (TTAGGG)3 [5 µl 32P-ATP, 2 µl T4 polynucleotide kinase buffer, 1 µl PNK (NEB), 12 µl ddH2O, 30 min at 37 C, and cleaned through G25 spin columns (Amersham, Piscataway, NJ)] in 6x SSC and 0.1% SDS. Membranes were washed three times for 1 h in 6x SSC and 0.1% SDS and exposed to film for up to 4 d at –80 C. Scanned films were analyzed using the telemetric software (http://bioinformatics.fccc.edu/software/opensource/telometric/telometric.shtml).
|
Results and Discussion |
|---|
|
TopAbstractIntroductionPatients and MethodsResults and DiscussionReferences |
|---|
Twenty-four ACCs, 11 ACAs, and three NATs were analyzed for TA. Of the ACCs, 21 of 24 (88%) displayed clearly detectable TA, and three of 24 (12%) were TA negative (Tables 2 and 3 and Fig. 1). None of the ACAs and NATs had detectable TA. Samples negative in the primary TRAP assay were spiked with an extract of the equivalent of 5 x 102 HeLa cells and reassayed in the TRAP to exclude inhibitors of TA in the sample preparation. All but one of the samples negative in the primary assay then displayed significant TA (not significantly different from the HeLa cell positive control). The remaining sample showed a slightly diminished but clearly detectable TA. This excludes the presence of a telomerase inhibitor, and these samples were therefore classified as TA negative. Extracts obtained from all adrenocortical cell lines, NCIh295A, SW13, and RL251, also displayed significant telomerase activity.
|
|
|
To evaluate telomere length, we ran Dpn II-digested genomic DNA of ACCs, ACAs, and NATs in a pulsed-field gel (Table 2
and Fig. 2). In this analysis, three of three NATs and 11 of 11 ACAs displayed a homogeneous length distribution with a median telomere length less than 15 kb (7–14 kb) in accordance with the reported length for human somatic cells. The ACC samples displayed a much broader variation between samples (3–78 kb) with five of 24 samples showing a TRF pattern reminiscent of ALT with a very long telomere length (median > 19 kb) (18). Of the adrenocortical cell lines tested, only NCIh295A had long telomeres (median 23 kb), whereas telomere length of HeLa, SW13, and RL251 were in a range less than 10 kb.
|
Tissue arrays including all samples used for TRF and TRAP analysis as well as additional ACA, NAT, and control tissue samples (24 ACCs, 22 ACAs, four NATs, and 16 normal control tissues) were subjected to a FISH procedure with a Cy3-labeled peptide nucleic acid telomere probe followed by immunohistochemistry for PML bodies. The association of a telomere signal with PML bodies has been shown to define tumors using ALT (23). Using this procedure, four of 24 ACC samples displayed a striking colocalization of PML staining and a bright telomere signal, which was detectable at exposure times significantly lower ( one of six) than that necessary for the visualization of the usual telomeric telomere signal (Table 2 and Fig. 3). All but one of these samples were also determined to be ALT positive in the TRF analysis. Finally, two of the three ALT-positive ACCs were assayed positive for TA. None of the other 20 ACCs, 22 ACAs, four NATs, or control tissue samples was positive in this assay. Some samples showed a positive PML signal but lacked bright telomere signals (Fig. 3). None of the array samples displayed a bright telomere signal in the absence of positive PML staining. We also tested the adrenocortical cell line NCIh295A for telomere-associated PML bodies as well as U2OS cells, a defined ALT cell line, and HeLa cells, a well-described TA cell line, as a positive and negative control, respectively (Fig. 4). Despite their telomerase activity, NCIh295A cells displayed a classically colocalized ALT staining, indicative of dual TMMs in these cells. We did not observe a typical ALT staining in either SW13 or RL251 cells.
|
|
and 3). The majority of ACC samples were positive for TA (21 of 24, 88%), with 19 of 24 (79%) using TA as their only TMM. In the ALT analysis, three ACCs showed definitive signs of ALT (positive in the immunohistochemical assay and long telomeres >19 kb), with two of three having TA as well and one of three being definitively categorized as ALT. Two samples were inconclusive in data interpretation. One sample was ALT positive in the telomere-associated PML body staining but ALT negative in the TRF analysis (length <15 kb) and negative for TA. The other sample was negative for TA and ALT-associated immunostaining, whereas telomere length was uninterpretable due to the lack of DNA in this lane. In this current analysis, TMMs could be determined in the vast majority of samples. This is in contrast to analyses of other tumor entities, in which nearly half of the samples could not be definitively assigned to a class of TMM and underscores the importance of active TMMs in ACCs (17, 19). The fact that a relatively high percentage of our ACC samples were TA positive is in contrast to previous studies and might be due to differences in assays, sample quality, and tissue integrity. In this study tissues were handled as safely as possible from sample site through storage to final analysis. For the latter two stages, thawing did not occur. Although we tried to ensure to capture the right sample area from tissue blocks guided by hematoxylin and eosin-stained mirror slides, the sample lacking TA activity as well as interpretable TRF analysis may represent an autolytic, partially degraded sample. Also, tissue underlying the target area on the frozen section slide may have been of extremely low cellularity or necrosis. The variability of percentage of TA-positive samples between studies may also reflect epidemiological heterogeneity between different studies.
These data identify TA as the major TMM in ACCs and provide evidence for ALT as a TMM in a small subset of ACCs as well as for the coexistence of telomerase-dependent and -independent mechanisms in the same tumor. Indeed, the existence of ALT is in accordance with an earlier study, which showed long telomere length in two of six ACCs (18). Interestingly, our findings in the adrenocortical cell line NCIh295A support the usage of TA as well as ALT mechanisms. It has recently been shown that NCIh295R cells, another subclone of the original parental NCIh295 cell line, uses ALT but is TA negative, which highlights the variability of TMM in the same cell line over passages and subcloning procedures (30). These findings are well in accordance with data from liposarcomas showing a coexistence of both mechanisms in the same tumor presumably due to the existence of different subpopulations using different, or even both, mechanisms (17, 19, 31).
Although our current study is too small to correlate TMMs with patient and disease characteristics, it is remarkable that both patients from which the samples lacking TA were derived showed increased survival and have a long standing history of ACC (108 and 61 months, both alive). This suggests a survival advantage for patients with tumors lacking TA, regardless of other TMM.
There is good evidence from other studies that the existence of TMMs generally correlates with a worse prognosis. In accordance with our results, in one study investigating TMM in astrocytoma IV, length of patient survival correlated with the presence of ALT, whereas in two studies of osteosarcomas and liposarcomas, any kind of TMM was correlated with a worse prognosis (19, 32, 33).
In transformed human and bovine adrenocortical cells, TA is not necessary for the development of a malignant phenotype but is required for immortalization and maintenance of an in vivo malignant tumor growth over extended passage numbers. In a study conducted by Sun et al. (34), cultures of transformed TA-negative bovine or human adrenocortical cells did not maintain the ability of indefinite clonal expansion but progressed to senescence after a limited number of serial transplantations in nude mice. Only after the additional expression of telomerase did transformed adrenocortical cells gain the characteristic of maintaining a malignant phenotype over an extended number of serial xenograft transplantation (34).
The analyses of TMMs presented in this study were conducted on extremely small tissue samples. As little as less than 5 mg can be sufficient for these procedures. This fact makes the determination of TMM interesting for samples from procedures, in which only very small tissue samples can be obtained, such as fine-needle biopsies. A biopsy approach is currently not part of the general diagnostic work-up of incidentally discovered adrenocortical lesion of any size, and for large adrenal lesions, a surgical approach is warranted (35, 36, 37). Future studies need to address the safety of biopsy procedures and determine the incidence of associated periprocedural risks such as hemorrhage or needle-track metastasis. However, because an increasing number of studies reports the feasibility of transgastral ultrasound and computed tomography-guided fine-needle aspirates or biopsies, these tissues may become more commonly available (38, 39, 40). TMM analysis could be used as a molecular diagnostic tool to aid in differentiation of benign and malignant lesions and may complement simple morphological and immunocytochemical analysis of these samples. However, the validation of this diagnostic procedure in a larger sample set or a prospective clinical trial is warranted.
The high percentage of samples positive for TMM in general and TA specifically identifies telomeres and the ribonucleoprotein telomerase as a potential molecular target for ACC therapy. In preclinical studies, several telomerase inhibitors as well as immunological approaches using telomerase as an antigen show promising results in cancer therapy models (41). Further motivation to use TMM as a therapeutic target comes from the observation that transformed adrenocortical cells need TA (or potentially any other kind of TMM) for maintenance of a malignant phenotype (34).
In summary, our study demonstrates the existence of different TMMs and telomerase-dependent (TA) as well as telomerase-independent (ALT) mechanisms in ACCs, but neither of these is evident in NAT or ACA. These findings suggest that the analysis of TMM may not only be of potential diagnostic value but also represent a novel molecular therapeutic target in ACCs for which new therapeutic strategies are desperately needed.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Disclosure Statement: T.E., T.J.G., and G.D.H. have nothing to declare.
First Published Online January 15, 2008
Abbreviations: ACA, Adrenocortical adenoma; ACC, adrenocortical cancer; ALT, alternative telomere lengthening; FBS, fetal bovine serum; FISH, fluorescence in situ hybridization; NAT, normal adrenocortical tissue; PBST, PBS and Tween 20; PML, promyelocyte leukemia; SDS, sodium dodecyl sulfate; SSC, saline sodium citrate; TA, telomerase activity; TMM, telomere maintenance mechanism; TRAP, telomere repeat amplification protocol; TRF, telomere-length restriction fragment.
Received August 17, 2007.
Accepted January 7, 2008.
|
References |
|---|
|
TopAbstractIntroductionPatients and MethodsResults and DiscussionReferences |
|---|
alleviates the transcriptional repression mediated by tumor suppressor Rb. J Biol Chem 276:43491–43494This article has been cited by other articles:
 |
|
 |
|
  T. Else Telomeres and telomerase in adrenocortical tissue maintenance, carcinogenesis, and aging J. Mol. Endocrinol., October 1, 2009; 43(4): 131 - 141. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. C. Kim, F. M. Barlaskar, J. H. Heaton, T. Else, V. R. Kelly, K. T. Krill, J. O. Scheys, D. P. Simon, A. Trovato, W.-H. Yang, et al. In Search of Adrenocortical Stem and Progenitor Cells Endocr. Rev., May 1, 2009; 30(3): 241 - 263. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Bielinska, H. Parviainen, S. Kiiveri, M. Heikinheimo, and D. B. Wilson REVIEW PAPER: Origin and Molecular Pathology of Adrenocortical Neoplasms Vet. Pathol., March 1, 2009; 46(2): 194 - 210. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. H. S. Costa, A. C. Latronico, R. M. Martin, A. S Barbosa, M. Q Almeida, C. F. P. Lotfi, H. P Lima Valassi, M. Y. Nishi, A. M. Lucon, S. A. Siqueira, et al. Expression profiles of the glucose-dependent insulinotropic peptide receptor and LHCGR in sporadic adrenocortical tumors J. Endocrinol., February 1, 2009; 200(2): 167 - 175. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. J. Giordano, R. Kuick, T. Else, P. G. Gauger, M. Vinco, J. Bauersfeld, D. Sanders, D. G. Thomas, G. Doherty, and G. Hammer Molecular Classification and Prognostication of Adrenocortical Tumors by Transcriptome Profiling Clin. Cancer Res., January 15, 2009; 15(2): 668 - 676. [Abstract] [Full Text] [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 |
