| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Department of Internal Medicine (J.N.V., N.Y., J.J.W.), Section of Endocrinology and Metabolism, and Departments of Pathology (D.R., J.D.) and Surgery (R.U.), Yale University School of Medicine, New Haven, Connecticut 06520; and Center for Molecular Medicine and Division of Endocrinology and Metabolism (A.A.), University of Connecticut School of Medicine, Farmington, Connecticut 06303-3101
Address all correspondence and requests for reprints to: John J. Wysolmerski, M.D., Section of Endocrinology and Metabolism, Department of Medicine, Yale University School of Medicine, TAC S131, 333 Cedar Street, New Haven, Connecticut 06520-8020.
| Abstract |
|---|
|
|
|---|
Objective, Design, Setting, Patients, Interventions, and Main Outcome Measures: We report a case of severe hyperparathyroidism resulting from the ectopic production of PTH by a pancreatic malignancy. Cells from the primary tumor (PEPP1 cells) were established in culture to examine the etiology of ectopic PTH gene expression in this patient.
Results and Conclusions: We failed to find amplification or rearrangement of the PTH gene but documented hypomethylation of the PTH promoter in tumor tissue. We found that PEPP1 cells support expression of a reporter gene containing regulatory sequences from the human PTH gene promoter. Therefore, this is the first report documenting ectopic PTH production by a tumor as the result of transactivation of the PTH gene. PEPP1 cells may be useful for future studies aimed at elucidating the details of PTH gene regulation.
| Introduction |
|---|
|
|
|---|
| Subject and Methods |
|---|
|
|
|---|
A 74-yr-old woman presented to an outside hospital with progressive nausea, fatigue, polyuria, and polydipsia. Her serum calcium was 15.3 mg/dl (8.810.2); 5 months earlier it had been 9.3 mg/dl. Her serum level of intact PTH was elevated at 399 pg/ml (1065). Computed tomography and positron emission tomography scans of the abdomen revealed pancreatic and liver masses. Despite treatment for hypercalcemia, she developed progressive obtundation and was transferred to Yale-New Haven Hospital for treatment of hypercalcemic crisis.
Upon admission to YNHH, she was unresponsive with a nonfocal neurological exam. Her medical history included type 2 diabetes, coronary artery disease, gastroesophageal reflux disease and hypertension. She had undergone mastectomy for breast cancer 18 yr previously and hysterectomy and bilateral salpingoophorectomy for benign uterine disease 9 yr previously. Admission laboratories revealed a serum calcium level of 18 mg/dl (8.810.2), an albumin level of 2.2 g/dl (3.55.0), an ionized calcium level of 10.1 mg/dl (4.124.95),and a biointact PTH level of 2,310 pg/ml (640).
A neck ultrasound revealed thyromegaly without evidence of a parathyroid mass. A parathyroid sestamibi scan demonstrated increased uptake in the left thyroid bed. It was thought that the patient likely had decompensated primary hyperparathyroidism, perhaps due to parathyroid carcinoma as well as an intraabdominal malignancy. However, the diagnosis of malignancy-associated hypercalcemia due to ectopic PTH production was also considered.
Neck exploration revealed a left thyroid adenoma and five normal parathyroid glands including a left intrathymic parathyroid gland that was resected. The intraoperative PTH assay was used to obtain bilateral internal jugular and peripheral vein samples, which were notable for the absence of a gradient, with all samples ranging from 1363 to 1518 pg/ml. The following day, a computed tomography-guided liver biopsy confirmed a high-grade malignancy consistent with either metastatic small-cell lung cancer or poorly differentiated neuroendocrine tumor. Immunohistochemical staining of the tumor cells was positive for PTH, suggesting a diagnosis of malignancy-associated hypercalcemia due to tumor production of PTH. In addition, a serum PTHrP level returned elevated at 11.7 pg/ml (<1.3 pmol/liter). Despite vigorous treatment with iv bisphosphonates, calcitonin, and forced saline diuresis, her serum calcium level could be reduced only transiently. Because her tumor was unresectable and her condition grave, the family opted for comfort measures and the patient died several days later from refractory hypercalcemia and multiorgan failure. An autopsy performed immediately after death revealed a poorly differentiated neuroendocrine tumor arising in the body of the pancreas with metastatic disease to the liver and retroperitoneal lymph nodes. Fresh tumor was harvested.
Isolation of tumor cells
All procedures were carried out with the approval of Yale Universitys Institutional Review Board. Fresh tumor tissue was minced and incubated for 4 h at 37 C in DMEM/F12 (Invitrogen, Carlsbad, CA) containing 0.2% dispase, 0.2% collagenase (Worthington, Lakewood, NJ), 5% fetal bovine serum, 50 µg/ml gentamicin, and 500 ng/ml amphotericin-B (Sigma, St. Louis, MO). Cells were washed twice with DMEM/F12 and then plated in growth media (DMEM/F12 with 10% fetal bovine serum, 50 µg/ml gentamicin, and 500 ng/ml amphotericin-B). Cells derived from the pancreatic tumor were designated PEPP1 cells for pancreatic, ectopic PTH production.
Immunohistochemistry
Immunohistochemistry was performed using standard techniques on formalin-fixed, paraffin-embedded sections of the primary tumor and the liver metastasis; the thymic parathyroid served as a positive control. We used a polyclonal goat antihuman PTH (134) antiserum donated by A. Karaplis (McGill University, Montréal, Québec) and the Vectastain Elite ABC kit (Vector Laboratories, Burlingame, CA).
Analysis of gene expression
Total cellular RNA was isolated using TRIzol reagent (Invitrogen) and was further purified over RNEasy minicolumns (QIAGEN, Valencia, CA) with on-column DNase digestion. Quantitative RT-PCR (QRT-PCR) was performed by standard methods using the OpticonII DNA engine (MJ Research, Waltham, MA). The following human gene expression assays (Applied Biosystems, Foster City, CA) were used: PTH (Hs00174888_m1), PTHrP (Hs00174969_m1), homeo box-A3 (Hs00399177_m1), paired box transcription factor (Pax)-1 (Hs00196352_m1), Pax9 (Hs00196354_m1), eyes absent (Eya)1 (Hs00166804_m1), T-box1 (Tbx1) (Hs00271949_m1), specificity protein (Sp)-3 (Hs01595808_m1), calcium-sensing receptor (Hs00173436_m1), and glyceraldehyde-3-phosphate-dehydrogenase (GAPD, Hs99999905_m1). The following RT2 PCR primer sets (SuperArray, Frederick, MD) were used for SYBR-Green-based QRT-PCR: Sp1 (PPH01482A), nuclear factor Y (NFY)-
subunit (PPH07751A), NFYß (PPH00436A), NFY
(PPH16009A), and GAPD (PPH00150A). Probe-based QRT-PCR was performed with Brilliant QRT-PCR master mix (Stratagene, La Jolla, CA), and SYBR-Green-based QRT-PCR was performed with Brilliant SYBR-Green QRT-PCR master mix (Stratagene). Normal parathyroid RNA (Promega, Madison, WI) served as a positive control for PTH expression and a calibrator sample for relative quantitation of expression by the 2-
CT method (6). Samples were run in triplicate.
PTH secretion
Intact PTH was measured in conditioned media using a two-site immunoradiometric assay (Immutopics, San Clemente, CA). Three samples of conditioned media collected on different days were analyzed in duplicate.
PTH gene copy number
Genomic DNA was isolated from PEPP1 cells and HEK 293T cells using DNAzol (Molecular Research Center, Cincinnati, OH). Fifty nanograms of DNA were amplified by QRT-PCR using primers for the human PTH gene (5'-TCGAAG TGGGGAGCTAATGGGAA-3' and 5'-CTCCAAGGGCAACAAAATTGTTGCA-3') and the human GAPD gene (5'-CGTGGAGTCCACTGGCGTCTCACC-3' and 5'-GTCATACTTCTC ATGGTTCACACCC-3'). The number of copies of the PTH gene in pancreatic tumor cells was calculated by the 2-
CT method, using GAPD as an endogenous reference gene, and HEK 293T cells, which are known to be hypotriploid (7), were used as a calibrator sample.
Southern blotting
Genomic DNA was extracted from tumor and control normal tissues and Southern blot analysis was preformed as previously described (8, 9). Twenty micrograms DNA were digested at 37 C for 2 h for each enzyme, BamHI, HindIII, EcoRI, MspI (New England Biolabs, Beverly, MA), and Hpa II (NEB) at 10U/µl. The probe was a 775-bp BglII fragment from the human PTH gene (8, 10).
PTH gene promoter assays
We used a reporter construct (hPTHp-GFP) in which green fluorescent protein (GFP) expression is driven by a previously described 5.5-kb fragment of the human PTH gene (11). Cells were transfected with hPTHp-GFP or pCDNA3.I (to control for background fluorescence of transfected cells) using the GenePORTER transfection reagent (Genlantis, San Diego, CA) and were cotransfected with pGL3-control (Promega) to control for transfection efficiency. After 48 h, fluorescence intensity and luciferase activity were measured in the same cells. hPTHp-GFP fluorescence was first corrected for background fluorescence (cells transfected with pCDNA3.1) and then normalized to luciferase activity to correct for transfection efficiency. Each transfection included nine replicates per cell type and was repeated three times.
| Results |
|---|
|
|
|---|
The primary tumor was a high-grade neuroendocrine carcinoma of the pancreas (Fig. 1A
). Immunohistochemistry documented PTH protein expression in the pancreatic tumor (Fig. 1D
) and liver metastasis (Fig. 1C
). As expected, the patients thymic parathyroid also stained for PTH (Fig. 1B
), whereas her thymus tissue did not (not shown), validating the specificity of the PTH antibody. QRT-PCR revealed that the primary tumor and liver metastasis expressed PTH mRNA at a level approximately two to three times that of normal human parathyroid tissue (Fig. 1E
). PTH mRNA expression was undetectable in adjacent normal pancreas and liver. PEPP1 cells, cultured from the primary tumor, grew readily for more than 10 passages and retained expression of PTH mRNA (Fig. 1E
). Furthermore, conditioned media from these cells contained 12.6 ± 2.2 pM PTH (124.8 ± 21.8 pg/ml). Control PANC-1 pancreatic carcinoma cells, HEK 283T human embryonic kidney cells, and 8701BC breast cancer cells did not express PTH mRNA, and their conditioned media did not contain PTH protein. Because the patients circulating level of PTHrP was also elevated, we assayed her tumor for PTHrP gene expression. As expected, the primary tumor tissue, metastatic liver tumor, and PEPP1 cells expressed PTHrP mRNA by QRT-PCR, whereas the normal pancreas tissue and normal liver tissue did not (data not shown). We also assayed the pancreatic tumor, liver metastasis, and PEPP1 cells for expression of the calcium-sensing receptor, which was negative.
|
We found that the PTH gene was not amplified in cultured tumor cells (Fig. 2A
). Furthermore, these cells had a normal diploid karyotype (not shown). Southern analysis using multiple restriction enzyme digests (not shown) of the pancreatic and liver tumor samples revealed no evidence of rearrangement involving a 16-kb span of DNA containing the entire PTH gene and more than 5 kb of upstream flanking sequences. However, paired digestion using MspI and the methylation sensitive enzyme, HpaII, did reveal a pattern consistent with hypomethylation of the PTH gene in both the pancreatic and liver tumors but not in the normal liver control sample (Fig. 2B
). Such hypomethylation would not be expected in nonparathyroid tissues and is indicative of active PTH gene transcription in the tumor tissues (12).
|
Given the lack of structural abnormalities in the PTH gene in the tumor, we next asked whether ectopic PTH expression was mediated through a trans-acting mechanism. We assayed the activity of a reporter construct consisting of a 5.5-kb fragment of the human PTH gene promoter placed upstream of the green fluorescent protein cDNA. This portion of the PTH promoter has been shown to contain regulatory sequences necessary for proper parathyroid-specific transgene expression in mice (11). As shown in Fig. 3
, fluorescence was negligible in all cell types tested except for the PEPP1 cells, suggesting that these cells contained transcription factors capable of activating tissue-specific regulatory elements of the PTH gene promoter. Therefore, we next examined the expression of 11 transcription factors implicated in either parathyroid gland development (Tbx1, Hoxa3, paired box transcription factor 1, Pax9, Eya1, and Gcm2) (13) or binding to the human PTH promoter (Sp1, Sp3, NFY
, NFYß, and NFY
) (5). Each of these transcription factors was either undetectable in tumor tissue and/or PEPP1 cells or was expressed at similar levels in both the tumor and adjacent normal tissues (data not shown). Thus, none of these transcription factors were likely responsible for PTH gene expression in this tumor.
|
| Discussion |
|---|
|
|
|---|
Rearrangements of the PTH gene promoter with the cyclin D1 gene in parathyroid adenomas suggest that DNA sequences necessary for tissue-specific expression of the PTH gene reside in the 5' regulatory region of the gene (24). In addition, the 5.5-kb fragment of the human PTH gene used in this study has previously been shown to be sufficient to direct specific expression of cyclin D1(11) and GFP (Arnold, A., unpublished data) to the parathyroid glands of transgenic mice. The ability of our patients tumor cells to express the hPTHp-GFP reporter gene suggests that her tumor expressed a transcription factor(s) that was able to interact with this portion of the PTH promoter to activate PTH gene expression. Despite the specific absence of parathyroid glands in glial cells missing 2 (Gcm2)-null mice and patients with loss-of-function mutations in Gcm2 (3, 4), this transcription factor was not expressed in the tumor cells, confirming that whereas Gcm2 may be necessary for parathyroid development, it is not necessary for PTH gene expression. Likewise, none of the other transcription factors previously implicated in parathyroid development (13) or binding to the PTH promoter (5) explained the activation of the gene in these tumor cells. Study of the transcriptional regulation of the PTH gene has been hampered by the lack of parathyroid cell lines. PEPP1 cells, derived from our patients tumor, may prove useful for the study of PTH gene expression.
| Acknowledgments |
|---|
| Footnotes |
|---|
The authors have nothing to declare.
First Published Online November 1, 2005
Abbreviations: GAPD, Glyceraldehyde-3-phosphate-dehydrogenase; GFP, green fluorescent protein; hPTHp, human PTH gene promoter; NFY, nuclear factor Y; Pax, paired box transcription factor; QRT-PCR, quantitative RT-PCR; Sp, specificity protein.
Received September 20, 2005.
Accepted October 26, 2005.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
L. J. Suva PTH Expression, Not Always where You Think ... J. Clin. Endocrinol. Metab., February 1, 2006; 91(2): 396 - 397. [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 |