Annals of Surgical Oncology Sign the Guestbook
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
10.1245/ASO.2005.03.103
Annals of Surgical Oncology 12:412-416 (2005)
© 2005 Society of Surgical Oncology
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
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
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Rosen, J. E.
Right arrow Articles by Libutti, S. K.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Rosen, J. E.
Right arrow Articles by Libutti, S. K.

Original Article

Gland Size Is Associated With Changes in Gene Expression Profiles in Sporadic Parathyroid Adenomas

Jennifer E. Rosen, MD1, Nick G. Costouros, MD1, Dominique Lorang, PhD1, A. Lee Burns, MD1, H. Richard Alexander, MD1, Monica C. Skarulis, BA2, Craig Cochran, BA2, James F. Pingpank, MD1, Stephen J. Marx, MD2, Allen M. Spiegel, MD2 and Steven K. Libutti, MD1

1 Surgery Branch, National Cancer Institute, Building 10, Room 4W-5940, 10 Center Drive, Bethesda, Maryland 20892
2 Metabolic Diseases Branch, National Institute of Diabetes and Digestive and Kidney Diseases, National Institutes of Health, Bethesda, Maryland 20892

Correspondence: Address correspondence and reprint requests to: Steven K. Libutti, MD; E-mail: steven_libutti{at}nih.gov.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Background: Sporadic parathyroid adenomas (SPAs) are benign neoplasms responsible for most cases of primary hyperparathyroidism (pHPT). The molecular pathways responsible for the variations in clinical severity of pHPT are unknown. We studied gene expression profiles in patients with SPAs and pHPT to determine associations between these changes and clinical parameters.

Methods: We selected 10 patients with solitary SPAs and nonfamilial, non-MEN1 pHPT treated with surgery from 2001 to 2003. Pathologic and clinical data were reviewed. At operation, tissues from SPAs were frozen in liquid nitrogen; total RNA was obtained from sections, and the diagnosis was confirmed with hematoxylin and eosin staining. Control normal parathyroid RNA was age- and sex-matched. RNA was amplified, labeled, and hybridized to a microarray of 22,272 human oligonucleotides. Cluster analysis of gene expression, analysis of expression ratios, and comparison of clinical parameters were performed.

Results: All patients were cured; all specimens were consistent with SPAs. K means clustering divided the 10 patients into 2 distinct 5-patient gene expression groups by using uncentered correlation based on gene subgrouping. Of the clinical parameters, only the mean gland volume was significantly different between group 1 (390 ± 160 mm3) and group 2 (1080 ± 615 mm3; P = .032 by Mann-Whitney test). Seventy-five genes were significantly upregulated or downregulated (with a ratio of <.33 or >3) compared with controls. These genes included the v-fos viral oncogene homolog and six calcium ion-binding signaling proteins.

Conclusions: Differential expression of a few critical genes may contribute to differences in gland volume in SPAs. A better understanding of these pathways may help to define the pathophysiology of pHPT.

Key Words: Parathyroid adenoma • Gene expression profile • Microarray analysis • Primary hyperparathyroidism


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Sporadic parathyroid adenomas (SPAs) are benign neoplasms responsible for most cases of primary hyperparathyroidism (pHPT). The clinical presentation of pHPT varies, and it may even be diagnosed incidentally before the patient has experienced osteoporosis, nephrolithiasis, or any of the other manifestations of the derangement of calcium homeostasis. The development of minimally invasive techniques and advanced imaging has helped improve the surgical management of these patients. However, the underlying molecular pathways responsible for the variations in the clinical severity of pHPT are still relatively unknown.1 Schachter et al.2 found alterations in various kinase genes involved in angiogenesis and apoptosis between adenoma and hyperplastic parathyroid tissue, although their sample size was limited to two patients. Stojadinovic et al.3 evaluated tissue microarray-based molecular profiling of various cell-cycle regulatory proteins and validated a molecular phenotype based on cell-cycle regulatory proteins that correlated with clinicopathologic severity. However, their study was not based exclusively on solitary parathyroid adenomas.

We studied gene expression profiles from SPAs from patients with pHPT to determine associations between these changes and clinical parameters. With the human genome sequencing completed, oligonucleotide arrays based on these sequences allow a window into the cell’s transcriptional activity. Neoplastic processes result from changes in gene expression patterns.4,5 The ability of expression genomics to analyze a large number of genes simultaneously across multiple samples may lead to our being able to look at parathyroid tumorigenesis as a pattern or cascade and then to identify key genes that are altered more significantly than others.6–9

In this study, we set out to determine whether gene expression profiling and cluster analysis could lead us to identify new tumor-related genes and to determine whether differential expression of these genes correlated with clinical parameters.10,11 Understanding these gene pathways may also lead us to understanding more important processes in the normal parathyroid, including calcium homeostasis.


   METHODS
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Patients
Patients with parathyroid tumors were identified from our database of patients operated on for pHPT on National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK) Institutional Review Board-approved protocols. Because patients with a variety of parathyroid disorders are seen at our institution, only patients with sporadic, nonfamilial pHPT were included as part of this study. All patients were treated and followed up at the National Institutes of Health (NIH) from 2001 to 2003. All histopathologic diagnoses were made by the Surgical Pathology Department at the NIH. Signs and symptoms were determined by a complete, thorough history and physical examination performed both by the NIDDK and the Surgery Branch National Cancer Institute staK. Ten age- and sex-matched RNA samples from normal parathyroid tissue were pooled and used as a control (Clinomics Biosciences, Pittsfield, MA). A decrease in parathyroid hormone of ≥50% from the higher value of the basal and pre-excision measurements was used to predict successful parathyroidectomy in pHPT.

Parathyroid Hormone Measurement
Blood samples were drawn after anesthesia, just before excision of a single adenoma in pHPT, and at 5, 10, and 15 minutes after excision. The parathyroid hormone level was measured with the Nichols QuiCk-IntraOperative Intact parathyroid hormone assay (Nichols Institute Diagnostics, San Clemente, CA).

Specimen Collection
Parathyroid tissues collected under the NIDDK Institutional Review Board–approved protocols were snap-frozen in liquid nitrogen at the time of surgery and stored at –80°C until use. Sections were stained with hematoxylin and eosin to confirm the diagnosis.

Clinicopathologic Categories
Clinical data included the patient’s date of birth, age at diagnosis, sex, presenting signs and symptoms, preoperative and postoperative serum ionized calcium level, preoperative and postoperative parathyroid hormone level, and duration of disease. Pathologic data included gland volume and gland location.

Tissue Processing and Isolation of RNA
At the time of surgery, all glands were measured in three dimensions: length, width, and thickness. Frozen sections of 176 to 1980 mg of tissue were collected in test tubes. Twelve- micrometer serial sections of these specimens were cut on a standard cryostat with a clean blade, placed on uncharged microslides, and stored at –80°C until use. Five sections per specimen were lysed and homogenized, and total RNA was extracted by using a selective binding silica gel-based membrane protocol (RNeasy; Qiagen, Valencia, CA). Approximately 2 µg of total RNA was obtained from each tumor sample. The total RNA was then subjected to two rounds of amplification by following the modified Eberwine method. This resulted in approximately 60 µg of amplified messenger RNA.12–14 Our second round of in vitro transcription used a 1:1 molar ratio of aminoallyl-labeled uridine triphosphate to unlabeled uridine triphosphate (Ambion Inc., Austin, TX). The quality of the extracted RNA was tested by spectrophotometry and by evaluations on the RNA 6000 Labchip with an Agilent Bioanalyzer 2100 (Agilent Technologies, Palo Alto, CA).

Microarray Analysis
Hybridization was performed on 22,272 human oligonucleotide microarrays, Hs-Operon2, produced by the National Cancer Institute/NIH (ATC, Gaithersburg, MD). Comparisons were made for each tumor with the same pooled normal parathyroid control. Fluorescent marker dyes (Cy5 and Cy3) were used to label 4 µg of the test and control samples, respectively. The respective dyes and samples were also switched to test for any labeling bias. The mixture of the two populations of RNA species was then hybridized to the same microarray and incubated for 16 hours at 42°C. The 10 oligonucleotide microarrays (1 for each patient sample) were then washed and scanned by using GenePix 4000B (Axon Instruments Inc., Union City, CA) to capture fluorescence; images were analyzed with GenePix software version 5.0. For each sample, data containing the image of the array and an Excel (Microsoft, Redmond, WA) file containing the expression ratio values for each gene were uploaded onto the Madb Web site (National Center for Biotechnology Information/NIH; http://nciar-ray.nci.nih.gov) for further analysis. To accurately compare measurements from different experiments, the data were normalized, and the ratio (signal Cy5/ signal Cy3) was calculated so that the median (ratio) was 1.0 and was then converted to log 2 numerical values.

Statistical Analysis
Cluster analysis of gene expression, analysis of expression ratios, and comparison of clinical parameters were performed. K means clustering was performed by using uncentered correlation. Non-parametric comparison of the medians of the two clusters was performed by using the Mann-Whitney test. Mean gland volume was calculated on the basis of measurements in three dimensions: length, width, and thickness. We excluded from the analysis all data from microarray spots labeled bad or not found, those with a low signal to background ratio (defined as <2), those with a target diameter between 5 and 300 µm, and those with a signal <500 intensity (quanta). We used the log 2 of the background-subtracted, normalized ratio of the mean Cy5 and mean Cy3 expression values. Arrays were normalized to the 50th percentile (median) by using all spot-filtered genes. K means clustering is a tool that identifies groupings of specimens based on related patterns of gene expression. All analyses were performed by using GraphPad InStat version 3.0a for Macintosh (GraphPad Software, San Diego, CA). Data are expressed as mean ± SD unless otherwise indicated.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
Ten patients were identified for study (six men and four women; mean age, 44.1 years). All patients were cured, and all specimens were consistent with SPAs. All of the patients with pHPT underwent successful parathyroidectomy and had normal or reduced calcium levels after operation and in follow-up studies ranging from 1 to 36 months. There were no intra-operative or postoperative complications. The application of K means clustering divided the 10 patients into 2 distinct gene expression groups of 5 patients by using uncentered correlation based on gene subgrouping (Fig. 1Go). The K means clustering used 225 upregulated and 227 downregulated genes to generate the clusters. Cluster integrity testing showed 100% robustness and 0% discrepancy after 100 perturbations, thus validating the reproducibility of the assignments. Clinical parameters were compared between these two groups (Table 1Go). Only mean gland volume was significantly different between group 1 (390 ± 160 mm3) and group 2 (1080 ± 615 mm3; P = .032 by Mann-Whitney test).


Figure 1
View larger version (7K):
[in this window]
[in a new window]
  		FIG. 1. Dendrogram illustrating the clustering of 10 patients with parathyroid adenoma into 2 families of 5 members each according to patterns of gene expression.

 

View this table:
[in this window]
[in a new window]
  		TABLE 1. Patient demographics and significance by Mann-Whitney U-testa
 
We also identified genes that were altered in their expression in parathyroid adenomas compared with normal parathyroid by applying the following selection criteria: gene expression in the tumor must have differed from that in normal control by at least 3-fold, and that difference must have been detected in at least 25% of the arrays performed. Seventy-five genes met these criteria and were significantly upregulated or downregulated (with a ratio of <.33 or >3) compared with controls. These genes included the v-fos viral oncogene homolog and seven calcium ion-binding signaling proteins (Table 2Go).


View this table:
[in this window]
[in a new window]
  		TABLE 2. Genes significantly upregulated or downregulated in parathyroid adenomas
 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
Our goal in this study was to correlate clinical information with molecular profiling to provide insight into pathways involved in the pathophysiology of parathyroid adenomas. Analysis of parathyroid adenoma RNA compared with pooled normal parathyroid RNA identified two dominant clustered gene expression patterns; patients with a significantly larger mean gland volume had a uniquely different pattern of gene expression. The choice of an appropriate control was important to compare individual samples against a common reference standard. We also chose to evaluate only one tumor type, the solitary parathyroid adenoma. This allows us to evaluate gene changes based on a more uniform population. It is important to note here that the posited multistep process of carcinogenesis has not been proven in the development of parathyroid carcinoma, and in fact we saw patients with longstanding symptoms without evidence of carcinoma.

The fact that upregulation of the v-fos Finkel-Biskis-Jinkins (FBJ) murine osteosarcoma viral oncogene homolog (v-fos) was seen in the adenomas when compared with controls is of particular interest: v-fos is a well-known transcription factor that is key to the developmental regulation of the human cytoskeleton.15,16 Both overexpression and underexpression of FOS have been detected in patients with various bone diseases.17 Altered regulation of SPARC-like 1 has been demonstrated in various tumors of epithelial origin,18 and the plastin 3 T-iso-form has differential expression in solid tissue malignancies.19 Several of these calcium-binding proteins also have vascular-related effects: the two members of the annexin family are known anticoagulants,20 and thrombospondin 2 is known to be an inhibitor of tumor growth and angiogenesis.21 Calcyclin (S100 calcium-binding protein A6) expression is dysregulated in both cytoskeletal abnormalities and human neoplasias.22,23 These genes hold interest for future exploration once validated through the use of real-time quantitative polymerase chain reaction.

In summary, this strategy demonstrates the power of genomic analysis as a technique for studying the underlying pathways responsible for the pathophysiology of neuroendocrine tumors. Further evaluation and linkage of clinical data to molecular profiling may allow for a better understanding of tumor pathogenesis, or even of normal parathyroid function and calcium homeostasis.

Received for publication March 29, 2004. Accepted for publication December 29, 2004.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Miedlich S, Krohn K, Paschke R. Update on genetic and clinical aspects of primary hyperparathyroidism. Clin Endocrinol 2003;59:539–54.[CrossRef][Medline]
  2. Schachter PP, Ayesh S, Schneider T, Laster M, Czerniak A, Hochberg A. Expression of kinase genes in primary hyperparathyroidism: adenoma versus hyperplastic parathyroid tissue. Surgery 2002;132:1094–9.[CrossRef][Medline]
  3. Stojadinovic A, Hoos A, Nissan A, et al. Parathyroid neoplasms: clinical, histopathological, and tissue microarray-based molecular analysis. Hum Pathol 2003;34:54–64.[CrossRef][Medline]
  4. Staudt LM. Gene expression profiling of lymphoid malignancies. Annu Rev Med 2002;53:303–18.[CrossRef][Medline]
  5. Van de Vijver MJ, He YD, van’t Veer LJ, et al. A gene-expression signature as a predictor of survival in breast cancer. N Engl J Med 2002;347:1999–2009.[Abstract/Free Full Text]
  6. Schulze A, Downward J. Navigating gene expression using microarrays—a technology review. Nat Cell Biol 2001;3:E190–5.[CrossRef][Medline]
  7. Sauter G, Simon R. Predictive molecular pathology. N Engl J Med 2002;347:1995–6.[Free Full Text]
  8. Simon R, Radmacher MD, Dobbin K, McShane LM. Pitfalls in the use of DNA microarray data for diagnostic and prognostic classification. J Natl Cancer Inst 2003;95:14–8.[Free Full Text]
  9. Golub TR, Slonim DK, Tamayo P, et al. Molecular classification of cancer: class discovery and class prediction by gene expression monitoring. Science 1999;286:531–7.[Abstract/Free Full Text]
  10. Miller LD, Long P, Wong L, Mukherjee S, McShane LM, Liu ET. Optimal gene expression analysis by microarrays. Cancer Cell 2002;2:353–61.[CrossRef][Medline]
  11. Raychaudhuri S, Sutphin PD, Chang JT, Altman RB. Basic microarray analysis: grouping and feature reduction. Trends Biotechnol 2001;19:189–93.[CrossRef][Medline]
  12. Eberwine J. Amplification of mRNA populations using aRNA generated from immobilized oligo(dT)-T7 primed cDNA. Biotechniques 1996;20:584–91.[Medline]
  13. Feldman AL, Costouros NG, Wang E, et al. Advantages of mRNA amplification for microarray analysis. Biotechniques 2002;33:906–12.[Medline]
  14. Wang E, Miller LD, Ohnmacht GA, Liu ET, Marincola FM. High-fidelity mRNA amplification for gene profiling. Nat Biotechnol 2000;18:457–9.[CrossRef][Medline]
  15. Grigoriadis AE, Wang ZQ, Cecchini MG, et al. C-fos: a key regulator of osteoclast-macrophage lineage determination and bone remodeling. Science 1994;266:443–8.[Abstract/Free Full Text]
  16. Grigoriadis AE, Wang ZQ, Wagner EF. Fos and bone cell development: lessons from a nuclear oncogene. Trends Genet 1995;11:436–41.[CrossRef][Medline]
  17. Dony C, Gruss P. Proto-oncogene c-fos expression in growth regions of fetal bone and mesodermal web tissue. Nature 1987;328:711–4.[CrossRef][Medline]
  18. Isler SG, Schenk S, Bendik I, et al. Genomic organization and chromosomal mapping of SPARC-like 1, a gene down regulated in cancers. Int J Oncol 2001;18:521–6.[Medline]
  19. Lin CS, Park T, Chen ZP, Leavitt J. Human plastin genes. Comparative gene structure, chromosome location, and differential expression in normal and neoplastic cells. J Biol Chem 1993;268:2781–92.[Abstract/Free Full Text]
  20. Tait JF, Sakata M, McMullen BA, et al. Placental anticoagulant proteins: isolation and comparative characterization of four members of the lipocortin family. Biochemistry 1988;27:6268–76.[CrossRef][Medline]
  21. Streit M, Riccardi L, Velasco P, et al. Thrombospondin-2: a potent endogenous inhibitor of tumor growth and angiogenesis. Proc Natl Acad Sci U S A 1999;96:14888–93.[Abstract/Free Full Text]
  22. Komatsu K, Murata K, Kameyama M, et al. Expression of S100A6 and S100A4 in matched samples of human colorectal mucosa, primary colorectal adenocarcinomas and liver metastases. Oncology 2002;63:192–200.[CrossRef][Medline]
  23. Hsieh HL, Schafer BW, Cox JA, Heizmann CW. S100A13 and S100A6 exhibit distinct translocation pathways in endothelial cells. J Cell Sci 2002;115:3149–58.[Abstract/Free Full Text]



This article has been cited by other articles:


Home page
 J. Med. Genet.Home page
J V Warner, D R Nyholt, F Busfield, M Epstein, J Burgess, S Stranks, P Hill, D Perry-Keene, D Learoyd, B Robinson, et al.
Familial isolated hyperparathyroidism is linked to a 1.7 Mb region on chromosome 2p13.3-14.
J. Med. Genet., March 1, 2006; 43(3): e12 - e12.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
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
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Rosen, J. E.
Right arrow Articles by Libutti, S. K.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Rosen, J. E.
Right arrow Articles by Libutti, S. K.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS