!" #$%&$ '((%$

$)$ *+ %(+)$)+ #$&+%*+ ##$&$ ,--, Dissertation for the Degree of Doctor of Medical Science in Surgery presented at Uppsala University in 2002

ABSTRACT Correa, P. 2002. Vitamin D and its in Parathyroid Tumors. Acta Universitatis Upsaliensis. Comprehensive Summaries of Uppsala Dissertations from the Faculty of Medicine, 1186, 49 pp. Uppsala. ISBN 91-554-5410-0.

Hyperparathyroidism (HPT) is characterized by tumor development in the parathyroid glands and excessive production of parathyroid hormone. Parathyroidectomy is the only considered therapy for the majority of patients. LOH (loss of heterozygosity) analysis revealed putative tumor suppressor on regions 1p and 11q in tumors from patients with truly mild hypercalcemia. Active vitamin D [1,25(OH)2D3] and its receptors, the (VDR), are essential regulators of the calcium homeostasis and are involved in HPT development. The VDR-FokI polymorphism, coupled to bone mineral density, was found not to be associated to development of primary HPT (pHPT). The total VDR mRNA levels is reduced in adenomas of pHPT as well as in hyperplastic glands of secondary HPT (sHPT). The VDR exon 1f transcripts were exclusively downregulated in the adenomas of pHPT, suggesting default regulation of the tissue-specially expressed VDR 1f promoter. The cytochrome P450 enzymes responsible for synthesis and degradation of 1,25(OH)2D3, namely vitamin D3 25-hydroxylase (25- hydroxylase), 25-hydroxyvitamin D3 1α-hydroxylase (1α-hydroxylase) and 25- hydroxyvitamin D3 24-hydroxylase (24-hydroxylase) were found to be expressed in normal and pathological parathyroid glands. Tumors of pHPT and sHPT demonstrated increased 1α-hydroxylase and reduced 24- and 25-hydroxylase expression, suggesting an augmented local production of active vitamin D. In contrast, parathyroid carcinomas displayed reduced expression of all three hydroxylases. The gained knowledge of vitamin D metabolism and catabolism in parathyroid tumors may indicate possibilities for novel treatment of sHPT and perhaps pHPT.

Key words: Hyperparathyroidism, Loss of heterozygosity, Vitamin D, Vitamin D Receptor, Polymorphism, Hydroxylase, Genetics, Tumorigenesis.

Pamela Correa, Department of Surgery, University Hospital, S-751 85 Uppsala, Sweden.

 Pamela Correa 2002 ISSN 0282-7476 ISBN 91-554-5410-0 Printed in Sweden by Uppsala University, Tryck och medier, Uppsala 2002 “ a little bit less conversation a bit more action… “ Contents ______

LIST OF PUBLICATIONS……………………………………………………. 6

ABBREVIATIONS…………………………………………………………….….. 7

INTRODUCTION………………………………………………………………….. 8 Clinical Presentation of Hyperparathyroidism………………………………… 8 Calcium Homeostasis……………………………………………………………. 9 Vitamin D Metabolism…………………………………………………………… 11 The Cell Cycle……………………………………………………………………. 14 Cancer Genetics…………………………………………………………………. 15 Molecular Genetics of Hyperparathyroidism………………………………….. 17 Common Features……………………………………………………….. 17 Primary Hyperparathyroidism……………………………………………18 Secondary Hyperparathyroidism……………………………………….. 20 Parathyroid Carcinomas………………………………………………… 21

AIM OF THE INVESTIGATION……………………………………………. 22

SUBJECTS AND METHODS……………………………………………….. 23 Subjects…………………………………………………………………………… 23 LOH Analysis……………………………………………………………………... 23 Genotype Analysis……………………………………………………………….. 24 Citrate and Calcium Clamp……………………………………………………… 25 Measurements of mRNA Levels………………………………………………... 25 Ribonuclease Protections Assay……………………………………….. 25 Real-time Quantitative PCR…………………………………………….. 26 Cell Culture……………………………………………………………………….. 27 RT-PCR Analysis………………………………………………………………… 28 Immunohistochemistry…………………………………………………………... 28

4

Statistical Analysis……………………………………………………………….. 29

RESULTS AND DISCUSSION……………………………………………...30

LOH studies in mild and severe primary hyperparathyroidism – Paper I….. 30 The vitamin D receptor start codon polymorphism in primary hyper- parathyroidism – Paper II……………………………………………………….. 31 Vitamin D receptor exon 1f transcripts in primary hyperparathyroidism – Paper III…………………………………………………………………………… 33

25-hydroxyvitamin D3 1α-hydroxylase, 25-hydroxyvitamin D3 24-

hydroxylase and vitamin D3 25-hydroxylase in parathyroid tumors– Paper IV-V………………………………………………………………………………… 35

GENERAL DISCUSSION…………………………………………………….. 39

ACKNOWLEDGEMENTS…………………………………………………….. 41

REFERENCES……………………………………………………………………... 42

5 List of publications ______

This thesis is based on the following papers, which will be referred to in the text by their Roman numerals.

I. Pamela Correa, C Juhlin, J Rastad, G Åkerström, G Westin, T Carling (2002). Allelic loss in clinically and screening-detected primary hyperparathyroidism. Clinical Endocrinology, 56:113-117

II. Pamela Correa, J Rastad, P Schwarz, G Westin, A Kindmark, E Lundgren, G Åkerström, T Carling (1999). The vitamin D receptor (VDR) start codon polymorphism in primary hyperparathyroidism and parathyroid VDR messenger ribonucleic acid levels. Journal of Clinical Endocrinology & Metabolism, 84:1690-1694

III. Pamela Correa, G Åkerström, G Westin (2002). Exclusive underexpression of vitamin D receptor (VDR) exon 1f transcripts in tumors of primary hyperparathyroidism. European Journal of Endocrinology, 147:1-5

IV. Ulrika Segersten*, Pamela Correa*, M Hewison, P Hellman, H Dralle, T Carling, G Åkerström, G Westin (2002). 25-hydroxyvitamin D3-1α-hydroxylase expression in normal and pathological parathyroid glands. Journal of Clinical Endocrinology & Metabolism, 87:2967-2972

V. Pamela Correa*, U Segersten*, P Hellman, G Åkerström, G Westin (2002). Increased 25-hydroxyvitamin D3 1α-hydroxylase and reduced 25-hydroxyvitamin D3 24-hydroxylase expression in parathyroid tumors- new prospects for treatment of hyperparathyroidism with vitamin D. Journal of Clinical Endocrinology & Metabolism, in press

*These authors contributed equally. Reprints were made with the permission of the publishers.

6

Abbreviations ______

1α-hydroxylase 25-hydroxyvitamin D3 1α-hydroxylase 1,25 (OH)2D3 1,25 dihydroxyvitamin D3 24-hydroxylase 25-hydroxyvitamin D3 24 hydroxylase 25-hydroxylase vitamin D3 25-hydroxylase 25(OH)D3 25-hydroxyvitamin D3 BMD bone mineral density bp Ca2+ calcium CAR calcium sensing receptor CDK cyclin-dependent kinases cDNA complementary deoxyribonucleic acid CGH comparative genomic hybridization CKI cyclin dependent kinase inhibitor FHH familial benign hypocalciuric hypercalcemia G gaps GAPDH glyceraldehyde 3-phosphate dehydrogenase HPT hyperparathyroidism INK4 inhibitor of cdk4 KIP kinase inhibitor LOH loss of heterozygosity M mitosis phase megalin/LRP-2 low density lipoprotein receptor- related 2 MEN1 multiple endocrine neoplasia type 1 MEN2A multiple endocrine neoplasia type 2A MMR mismatch repair system mRNA messenger ribonucleic acid NSHPT neonatal severe hyperparathyroidism PCR polymerase chain reaction pHPT primary hyperparathyroidism PTH parathyroid hormone RB RXR retinoic X receptor s serum S DNA synthesis phase sHPT secondary hyperparathyroidism TGFβ transforming growth factor β TSG tumor suppressor VDR vitamin D receptor VDRE vitamin D response element

7 Introduction ______

Clinical Presentation of Hyperparathyroidism Hyperparathyroidism (HPT) refers to states with excessive production of parathyroid hormone (PTH) and an altered set-point of the calcium regulated PTH secretion. Primary HPT (pHPT) is characterized by tumor development in the parathyroid glands leading to pronounced release of PTH. In secondary HPT (sHPT) hyperplastic parathyroid glands develop in response to hypocalcemia most commonly renal failure. Female sex, age, irradiation to the neck (Tisell et al., 1995) and lithium salts (Abdullah et al., 1999) predispose to pHPT. Solitary adenomas account for roughly 85% of the disease and hyperfunction in multiple glands, mainly chief cell hyperplasia, for the remaining part (Mallette et al., 1994). Water clear cell hyperplasia and parathyroid carcinoma are extremely rare entities and affect <1% of patients explored for pHPT (Shane et al., 2001). Historically pHPT was associated with prominent hypercalcemia, rapid progress and overt symptoms such as recurrent nephrolithiasis, symptomatic bone disease and severe muscle weakness (Rastad et al., 2001). Today the majority of patients exhibits mild hypercalcemia and vague symptoms including fatigue, bone pain, and cognitive symptoms (Lundgren et al., 1998, Rastad et al., 2001). Although sporadic pHPT is the major cause of the disease, the disorder exists in several autosomal dominant inherited disorders such as multiple endocrine neoplasia type 1 (MEN1), MEN 2A, HPT jaw syndrome and familial isolated HPT. pHPT may present in all age groups but is particularly frequent in postmenopausal women demonstrating a prevalence of 2-3% (Lundgren et al., 1997), where the disease is associated with mild hypercalcemica and less obvious symptoms. Bisphosphonates and calcitonin are used for treatment of hypercalcemic crisis but surgery, i.e. parathyroidectomy is for the majority of patients the only considered therapy. Surgery for pHPT results in normocalcemia in roughly 95-99% of the patients (Socialstyrelsen 2001).

8

Initial hypocalcemia, hyperphosphatemia, decreased 1,25 dihydroxyvitamin D3

[1,25(OH)2D3], abnormal parathyroid function and skeletal resistance to PTH distinguish sHPT. Most patients with renal failure develop parathyroid hyperplasia to some extent as the hypocalcemic state lasts longer than in other hypocalcemic disorders such as deficiency or malabsorption of vitamin D. In sHPT the parathyroid glands exhibit general enlargement histologically characterized by hyperplasia of either nodular or diffuse pattern (Mallette et al., 1994). The majority of the patients are on hemodialysis and present with variable frequency and severity of symptoms. Improved medical treatments with calcium and vitamin D replacement have resulted in fewer patients eventually requiring parathyroidectomy. In this thesis sHPT will refer to uremic sHPT.

Calcium Homeostasis In mammalians calcium has important extracellular and intracellular functions (Table 1). The extracellular calcium concentration is maintained within a narrow interval by the calciotrophic hormones PTH and 1,25(OH)2D3 (Figure 1). Alterations in systemic level of these hormones or the PTH-related peptide are the major causes of aberrant extracellular fluid calcium concentration.

FORM LOCATION FUNCTIONS intracellular cytosol, nucleus, endoplasmatic reticule, second messenger, action potentials, nerve mitochondria conduction, contraction, motility, cytoskeletal function, cell division, secretion, storage extracellular extracellular fluids, bones, teeth blood clotting, exocytosis, contraction, intercellular adhesion, mineral storage

Table 1. The intra- and extracellular functions of calcium in mammalians.

A calcium sensing receptor denoted CAR is located on the surface of the parathyroid chief cells. This receptor was initially cloned from bovine parathyroid (Brown et al., 1993) and subsequently in humans. The CAR protein is encoded by a gene on chromosome 3q13 and is a G protein coupled receptor. Apart from the parathyroid glands CAR is expressed in other human calcium sensing organs

9 2 C N 2

2 2 C C

G- Figure 1. Schematic repre- sentation of the regulatory COO PL system maintaining calcium

I 3 homeostasis PK PKC protein kinase C, 2

Regulatio 2 C2 PLC phospholipase C, [Ca C n i 2 C IP3 inositoltriphosphate Parathyroid Systemic BON KIDNE

2 C 2 C

3 P 4 2 1,25(OH) 3 C 25(OH) 3

Vitamin

SMALL LIVE such as kidney, keratinocytes, certain cerebral cells and thyroid C-cells (Brown et al., 2001). Heterozygous mutations in CAR causes familial benign hypocalciuric hypercalcemia (FHH) and the homozygous form gives raise to neonatal severe HPT (NSHPT) with need for parathyroid surgery (Pollak et al., 1993). Monoclonal antibodies against human parathyroid cells recognized another protein named megalin/LRP-2 (Juhlin et al., 1987) and these antibodies also inhibited calcium induced rise in intracellular calcium and caused reduction in PTH release (Juhlin et al., 1988a). However, a presumptive calcium sensing function of megalin/LRP- 2 is unclear. Recently it was shown that megalin/LRP-2 was essential for endocytic uptake of 25-hydroxyvitamin D3 [25(OH)D3] in renal tubule cells (Nykjaer et al., 1999), indicating involvement of the receptor in the vitamin D metabolism. In response to hypocalcemia the 84 amino acid peptide PTH is synthesized and released by the chief cells in the parathyroid glands (Mundy et al., 1999). Recently the thymus was described as an auxiliary source of PTH production (Günther et al., 2000) in parathyroid deficient mice but the extent of this PTH

10

production or its physiological significance is unknown. The amino terminal end of the PTH molecule binds to the PTH receptor and enhances osteoclastic bone resorption and calcium reabsorption from the distal tubule cells. Furthermore,

PTH induces renal 25-hydroxyvitamin D3 1α-hydroxylase (1α-hydroxylase) which augments 1,25(OH)2D3 and increases intestinal calcium absorption. There exist well-defined feedback loops between 1,25(OH)2D3, calcium and PTH (Figure 1). Calcitonin is a 34 amino acid peptide synthesized and secreted by the parafollicular cells of the thyroid gland. Hypercalcemia causes calcitonin secretion and lowers serum calcium levels by inhibiting osteoclast activity. In vertebrates the precise biological role of calcitonin in the overall scheme of calcium homeostasis is uncertain.

Vitamin D Metabolism

Vitamin D3 is produced from 7-dehydrocholestrol in the skin by sun exposure or obtained from dietary sources. The initial step in the process of making active vitamin D is 25-hydroxylation of vitamin D3 (Jones et al., 1999a). This reaction is catalyzed by the enzyme vitamin D3 25-hydroxylase (25-hydroxylase). The liver is the principal site of hydroxylation of vitamin D3 to 25(OH)D3 though substantial activity occurs in other tissues such as kidney, osteoblasts and endothelial cells (Axén et al., 1995, Ichikawa et al., 1995, Reiss et al., 1997). There are controversies if more than one enzyme is capable of 25-hydroxylation, and the physiological significance of 25-hydroxylase is presently unclear. 25(OH)D3 is the major circulating form of vitamin D in humans (Jones et al., 1999a).

Hypocalcemia results in 1α-hydroxylation of 25(OH)D3 and as pointed out above renal 1α-hydroxylase is strongly upregulated by PTH. 1α-hydroxylase expression has not only been demonstrated in the proximal convoluted tubules but also in more distal areas in the nephron namely the thick ascending loop of Henle, the distal convoluted tubule and cortical collecting ducts (Zehnder et al., 1999). Recently 1α-hydroxylase expression has been reported in non-renal tissues such as keratinocytes (Bikle et al., 1986, Fu et al., 1997), testis (Fu et al., 1997), brain (Fu et al., 1997, Zehnder et al., 2001), cultured bone cells (Howard et al., 1981),

11 macrophages (Overbergh et al., 2000, Monkawa et al., 2000), placenta (Delvin et al., 1987, Diaz et al., 2000), prostate cells (Schwartz et al., 1998), colon adenocarcinoma (Cross et al., 1997, Tangpricha et al., 2001), non-small cell lung carcinomas (Jones et al., 1999b), and islet cells of the pancreas (Zehnder et al., 2001). The function of 1α-hydroxylase in these tissues has yet to be fully defined but appears to involve local production of 1,25(OH)2D3 for autocrine or paracrine regulation. 1,25(OH)2D3 elicits most of its biological effects by binding to its high affinity vitamin D receptor (VDR) (Haussler et al., 1997, Jones et al., 1999a) in a number of target tissues including the parathyroids. After binding to VDR the complex either form homodimers or a heterodimer with the retinoic X receptor (RXR) or other steroid receptors, subsequent binding to vitamin D response elements (VDRE) promotes activation of target genes (Figure 2). A list of gene products known to be up- or downregulated at the transcriptional level is shown in

1,25(OH)2D

VDR RXR

coactivator RNA polymerase II

VDRE

Figure 2. Mechanism of 1,25(OH)2D3 action. After binding to VDR the complex either forms homo- or heterodimers with for example the retinoic X receptor. Binding to VDRE attracts coactivators and initiate transcription of target genes.

Table 2. Inactivation of 1,25(OH)2D3 begins with 24-hydoxylation by the enzyme

25-hydroxyvitamin D3 24 hydroxylase (24-hydroxylase). This enzyme is not only restricted to traditional vitamin D responsive tissues like kidney and intestine but is rather ubiquitously expressed (Jones et al., 1999a). 1,25(OH)2D3 upregulates

12

the 24-hydroxylase gene and its promoter contains two VDREs (Ohyama et al., 1994, Jones et al., 1999a). The 1α-hydroxylase and 24-hydroxylase are tightly and reciprocally regulated by PTH and 1,25(OH)2D3. PTH regulates 24- hydroxylase by altering the messenger ribonucleic acid (mRNA) stability (Zierold et al., 2001).

UPREGULATED GENES DOWNREGULATED GENES Table 2. Example caldinin PTH of genes regulated at transcriptional osteocalcin PTH relatide peptide level by osteopontin collagen type 1 1,25(OH)2D3. 24-hydroxylase c- p21 IL-2

β3-integrin 1α-hydroxylase

1,25(OH)2D3 demonstrates antiproliferative and differentiating effects in several tissues. Localization of 1,25(OH)2D3 and VDR in the parathyroid glands were reported decades ago (Jones et al., 1999a). 1,25(OH)2D3 executes suppression of PTH synthesis at genomic level by inhibiting PTH gene transcription and subsequently PTH secretion (Silver et al., 1999). Moreover

1,25(OH)2D3 is capable to suppress parathyroid cell proliferation (Nygren et al., 1988, Szabo et al., 1989). The human VDR gene spans more than 60 kb, contains 14 exons and is transcribed from at least three promoters (Baker et al., 1988, Crofts et al., 1998, Miyamato et al., 1997, Byrne et al., 2000). The VDR gene contains several known polymorphisms in exon 2, exon 8, intron 9 and the poly(A) region. When the nucleotide sequence of the human VDR complementary DNA was reported, two potential translation initiation (ATG) sites were disclosed in exon 2 (Baker et al., 1988). The T/C polymorphism detected by the FokI restriction enzyme resides within the first potential start site. The resulting difference in VDR length by three amino acids may affect the function of the protein (Arai et al., 1997, Gross et al., 1998). This contrasts to the other described polymorphisms since they do not alter the amino acid sequence (Morrison et al., 1994). Recently four novel

13 upstream exons were identified denoted 1f, 1e, 1d and 1b (Crofts et al., 1998). These give raise to several VDR transcripts that vary in the 5’ part of the mRNA (Miyamato et al., 1997, Crofts et al., 1998). The 5’ terminal exons 1a, 1d and 1f are found in 5, 5 and 4 different transcripts respectively, of which two exon 1d transcripts potentially encode proteins with N-terminal extensions of 50 or 23 amino acids. Most of the 14 VDR transcripts have appeared to be expressed in one analyzed parathyroid adenoma (Crofts et al., 1998). Expression of all four transcripts that originate from exon 1f seems to be restricted to calcium regulating tissues such as kidney, parathyroid and an intestinal carcinoma cell line, which suggests that the distal promoter is cell type-specifically regulated (Crofts et al., 1998).

The Cell Cycle As summarized in Figure 3 the cell cycle is characterized by a DNA synthesis phase where initiation and completion of DNA replication is achieved (S) and a mitosis phase (M). In between these phases there are gaps (G). The most important check point in mammalian cells is the restriction point in late G1 phase when beyond this point mitosis will irrevocably take place. Cell cycle progression is governed by sequential formation, activation and inactivation of cyclins, regulating subunit, and their cyclin-dependent kinases (CDK), catalytic subunit, (Hunter et al., 1994, Sherr et al., 1995). The formation of these complexes depends on cell cycle regulated expression of cyclins that assemble pre-existing CDKs. The major cyclin/CDK complexes are shown in Figure 3. Cyclin/CDK complexes are positively and negatively regulated by phosphorylation. The ability of cyclin/CDK complexes to phosphorylate retinoblastoma protein (RB) and other substrates are regulated by cyclin-dependent kinase inhibitors (CKIs). Passage through restriction point requires phosphorylation and inactivation of the RB, CKIs and the tumor suppressor gene product . In humans there are two families of CKIs namely the kinase inhibitor proteins (KIP) and the inhibitor of cdk4 (INK4). The KIP group includes p21/waf1/cip1, p27/kip1 and p57/kip2 that are homologous in the amino terminus and have broad CDK specificity. The INK4 group consists of p15, p16, p18 and p19 and has more restricted CDK specificity. 14

restriction

oncogene G1 phase cyclin s cyclin

cdk cdk cdk cyclin p p G0 pR pR

M cyclin

cyclin cyclin cdk2 p cdk cdk S

G2

Figure 3. Schematic representation of the cell cycle. When cells enter into early G1 phase. The cyclin D1/CDK4 complex phosphorylates the RB protein leading to sequential phosphorylation by cyclin E/CDK2 and induction of S phase.

During tumorigenesis normal cell cycle regulation is abandoned and restriction point abnormalities such as overexpression of cyclins and loss of function of CKIs become common features. 1,25(OH)2D3 performs its action on fundamental cellular processes including proliferation, differentiation and apoptosis.

1,25(OH)2D3 can induce arrest in the G1 phase of the cell cycle and this has at least partly been associated with increased CKI p21 levels in hematopoietic and pancreatic cancer cell lines and decreased p21 levels in squamous cell carcinoma (Liu et al., 1996, Kawa et al., 1997, Hershberger et al., 1999).

Cancer Genetics Only certain tumor types can appear after impairment of one specific gene for example the MEN1 syndrome. Usually damage to one gene is not sufficient for cellular transformation and four to seven genetic events are estimated to be needed for tumorigenesis in epithelial cancers (Renan et al., 1993). Series of genetic and epigenetic events are required for cancer formation ensuing in either unregulated proliferation or apoptosis. Two different types of genes may be involved namely proto-oncogenes and tumor suppressor genes (TSGs). Damage of a proto-oncogene results in an activated oncogene. In normal cells proto-

15 oncogenes are responsible for cellular growth control, proliferation and differentiation by regulating protein phosphorylation, gene transcription or signal transduction. Inversions or translocations, point mutations or amplification activate proto-oncogenes. TSGs enhance tumor development when becoming inactivated since their normal task is to restrain cell proliferation. Loss of function may be demonstrated and normal phenotype can be restored by introducing the wild-type gene. Point mutation or deletion in both alleles of the TSG results in loss of function. Insufficiency of TSGs can be inherited or occur somatically. Some candidate TSGs does not fulfil the original criteria for TSGs and some of these genes are believed to be involved in DNA repair and genomic stability. Cells possess DNA repair systems and one of these systems is denoted the mismatch repair system (MMR). MMR accounts for the majority of DNA reparation and mutations do not only cause hereditary non-polyposis colorectal cancer (Kinzler et al., 1997) but are found to be important for sporadic tumor development in other cell types as well. DNA methylations of CpG islands in promoter regions is a common mechanism of epigenetic silencing of genes in somatic cells (Baylind et al., 2000). Promoter silencing by methylation of one or both alleles has been shown in several TSGs. Today it is unknown if monoclonal expansion is due to methylation in some especially predisposed promoters or if methylation occurs randomly. Genetic polymorphism studies gained acceptability with the development of the Project and has been extensively used to analyze genetic determinants of complex diseases such as coronary heart disease, osteoporosis and diabetes mellitus. The study strategy is often case-control-, associations- and linkage disequilibrium analyses, in which the presence of one allele on a chromosome in a diseased patient may indicate high probability that a particular allele will be present at a neighbouring site on the same chromosome, with intermediate traits. Although polymorphism studies are still in their infancy controversies remain. For example most of the polymorphisms in association studies are non-functional and sample heterogeneity (ethnic groups, sex) or statistical errors by using inadequate study groups are some pitfalls.

16

Molecular Genetics of Hyperparathyroidism Common Features The inferior and superior parathyroid glands develop from the third and fourth pharyngeal pouches, respectively. They detach from the pharyngeal wall and migrate to their proper anatomical position on the medial half of the posterior surface of each thyroid lobe. Little is known about what forces direct the parathyroid organogenesis. In mouse two master-regulatory genes encoding transcription factors are essential for parathyroid gland development namely the genes Gcm2 and Hoxa3. Mice who lack Gcm2 or Hoxa3 (Günther et al., 2000, Manley et al., 1998) do not develop parathyroid glands at all. The human parathyroid glands have recently been shown to be the major non-neural expression site for Gcm2 and parathyroid adenomas demonstrate reduced Gcm2 mRNA expression possibly implying a role for Gcm2 in parathyroid tumorigenesis (Correa et al., 2002a).

normal parathyroid

polyclonal monoclona expansion l

hyperplasia (pHPT or parathyroi d hyperplastic parathyroi noduli d

Figure 4. Model of development if parathyroid tumors. Polyclonal expansion is triggered by alterations in serum levels of calcium, 1,25(OH)2D3 and phosphate and aberrant expression of VDR and CAR. Monoclonal expansion requires somatic mutations causing proliferation. Some polyclonal lesions will eventually become monoclonal since their high proliferation rate is prone to mutations.

Most parathyroid tumors are preceded by parathyroid cell proliferation engaging all glands followed by a monoclonal expansion i.e. they arise from one single precursor cell, which has obtained selective growth advantage. The monoclonality is expected to represent certain genetic events that characterize

17 the clonal tumorigenesis (Figure 4). By X-chromosome inactivation analysis using a DNA polymorphism approach (Arnold et al., 1988) the majority of parathyroid adenomas, parathyroid tumors of MEN1 (Friedman et al., 1989) and also a large portion of primary and secondary hyperplastic glands (Arnold et al., 1995) were found to be of monoclonal nature. Hyperplastic glands of sHPT, which histologically demonstrate nodular hyperplasia, are more likely to be monoclonal than those that display a diffuse growth pattern (Tominaga 1999a). In most cases the initial parathyroid cell proliferation is believed to be driven by potent parathyroid cell proliferation stimulators such as hypocalcemia, hyperphosphatemia and vitamin D deficiency. Reduced VDR, CAR and megalin/LRP-2 expression on both mRNA and protein levels are common features for parathyroid adenomas, hyperplasias of primary and secondary origin and carcinomas (Juhlin et al., 1988b, Farnebo et al., 1997b, Gogusev et al., 1997, Farnebo et al., 1998a, Carling et al., 2000, Sudhaker Rao et al., 2000). The reduced receptor expressions are thought to reflect gained resistance to

1,25(OH)2D3 and calcium regulation contributing to the tumorigenesis. Exactly what causes the decreased expression levels is unclear since no somatic mutation has been demonstrated (Cetani et al., 1999, Brown et al., 2000). Furthermore, polymorphisms in exon 8, intron 8 and exon 9 of the VDR gene (Carling et al., 1997a) and in exon 7 of the CAR gene (Cole et al., 1999) have been associated to development of pHPT or levels of serum (s)-calcium in diseased patients respectively. VDR polymorphisms have also been related to the severity of sHPT (Nagaba et al., 1998).

Primary Hyperparathyroidism In 1988 a chromosomal rearrangement that separated the 5’ regulatory region of the PTH gene from its coding exons was described (Motokura et al., 1991) (Figure 5).This was particularly interesting since the tissue-specific in the parathyroid chief cells could activate a proto-oncogene by this rearrangement. The adjacent DNA sequence was cloned and called PRAD, later cyclin D1, and was shown to be dramatically overexpressed in the adenoma from where it had been cloned. The up-regulation of cyclin D1 in the adenomas seems

18

PTH coding Figure 5. Activation of cyclin D1. PTH promoter A) Normal chromosome 11 with the PTH gene located on the short arm and the cyclin D1 centromere gene on the other chromosome arm. B) DNA rearrangement causing constant activation when cyclin cyclin D1 D1 is put under control of the PTH gene promoter.

A) normal B) inverted to be driven by DNA sequences in the upstream PTH gene vicinity. The rearrangement has only been found in a subset of unusually large parathyroid adenomas. However, cyclin D1 may be overexpressed also in the absence of the specific rearrangement since 18-39% of investigated parathyroid adenomas overexpress cyclin D1 (Hsi et al., 1996, Vasef et al., 1999). Furthermore transgenic mice who overexpress cyclin D1 develop pHPT (Imanishi et al., 2001). No somatic mutation in the cyclin D1 gene other than the gene rearrangement has been found in parathyroid adenomas suggesting that cyclin D1 overexpression of the wild type sequence rather than mutational activation may be the predominant cyclin D1 oncogenic mechanism (Hosokawa et al., 1995). The RET proto-oncogene, which is responsible for tumorigenesis in the MEN2A syndrome, has failed to demonstrate somatic mutations in sporadic pHPT (Padberg et al., 1995). Somatic loss of a tumor suppressor gene allele often involves loss of chromosomal material ranging from modest deletions to almost an entire chromosome. By comparing polymorphic loci in DNA from blood and tumor in the same individual loss of heterozygosity (LOH) analysis with microsatellite markers and also comparative genomic hybridization (CGH) have been used for characterizing deletions and gains of DNA, discovering putative TSGs and proto- oncogenes respectively. In parathyroid adenomas candidate TSGs have been identified at gene loci 1p, 1q, 6q, 9p, 11p, 11q and 15 q (Tahara et al., 1996a, Farnebo et al., 1997a) and possible proto-oncogenes at 16p and 19p

19 (Palanisamy et al., 1998). ~30% of parathyroid adenomas show LOH at 11q13, the locus of the MEN1 gene (Chandrasekharappa et al., 1997), and roughly half of these show MEN1 gene mutations (Heppner et al., 1997, Carling et al., 1998a, Farnebo et al., 1998b). This finding makes the MEN1 gene the most commonly mutated gene in parathyroid adenomas and also implicates presence of other putative TSGs at 11q13. A D418D polymorphism in exon 9 in the MEN1 gene has recently been associated with development of pHPT (Correa et al., 2002b). The MEN1 protein, menin, was initially shown to bind the Ap-1 JunD and represses JunD-mediated transcriptional activity (Agarwal et al., 1999, Gobl et al., 1999). Recently menin was implicated in tumorigenesis when menin inactivation was shown to inhibit transforming growth factor β (TGFβ)-mediated cell growth inhibition through Smad3 (Kaji et al., 2001). Several novel interaction partners PEM, Nm23, glial fibrillary acidic protein and vimentin have been identified (Lemmens et al., 2001, Ohkura et al., 2001, Lopez-Egido et al., 2002). Furthermore, the MEN1 gene knock-out mice develop a MEN1-like syndrome (Crabtree et al., 2001). Other candidate TSGs and proto-oncogenes such as RB (Dotzenrath et al., 1996), p53 (Hakim et al., 1994), ras (Friedman et al., 1990), p15 (Tahara et al., 1996b), p16 (Tahara et al., 1996b), p18 (Tahara et al., 1997), RAD54 (Carling et al., 1999a) and RAD51 (Carling et al., 1999b) do not appear to contribute to the tumorigenesis in the parathyroid. However double mutant mice lacking p18 and p27 or p18 and p21 develop parathyroid adenoma/hyperplasia in 38% and 14% respectively (Franklin et al., 2000) suggesting that distinct tissue- specific CKI collaborations might contribute to parathyroid tumorigenesis.

Secondary Hyperparathyroidism As mentioned above most hyperplasias are monoclonal lesions. Nodular hyperplasia shows significantly greater expression of cyclin D1, RB and Ki67 compared to diffuse hyperplasia (Tominaga et al., 1999b). LOH studies have implicated allelic loss at chromosome 2, 3q, 7p and 18 q (Farnebo et al., 1997a, Chudek et al., 1998, Nagy et al., 2001). Occasional loss at 11q and sporadic allelic losses for at least one chromosomal arm are frequent. Both the CAR gene at 3q (Degenhardt et al., 1998) and the MEN1 gene at 11q13 (Tahara et al.,

20

2000) failed to show somatic mutations indicating that other genes must be responsible for the monoclonal expansion. These genes are still to be identified.

Parathyroid Carcinomas Parathyroid carcinomas commonly demonstrate allelic gains on chromosome 1q, 5q, 9q, 16p, 19p and Xq and LOH on 1p, 3q, 4q,13q, 17 and 21q (Agarwal et al., 1998, Kytölä et al., 2000, Shane et al., 2001). Notably allelic loss at 11q, PTH gene arrangements or ras gene mutations are not frequent in the carcinomas. Furthermore LOH at the RB locus and abnormal expression of RB protein have been suggested to be specific for parathyroid carcinomas (Cryns et al., 1994). Ki- 67 a marker for proliferative activity, has been reported as a useful indicator of parathyroid malignant disease (Abbona et al., 1995, Farnebo et al., 1999). An increased risk for parathyroid carcinoma is associated with the hereditary HPT jaw syndrome recently localized to 1q21-31 (Shane et al., 2001). Most chromosomal regions frequently lost in adenomas are seldom or never lost in carcinomas, supporting that carcinomas arise de novo rather than from pre- existing adenomas.

21 Aim of the investigation

______

The aim of this investigation was explore the following specific questions and thereby improve our knowledge of parathyroid tumorigenesis.

1) To relate clinical characteristics in patients demonstrating pHPT to LOH in the chromosome regions 1p, 6q, and 11q13

2) To determine the VDR-FokI genotype frequencies in postmenopausal women with pHPT and control subjects, any associations with clinical signs of the disorders, as well as parathyroid VDR and PTH mRNA expression

3) To study which of the different VDR exon 1a, 1d and 1f transcripts causes the reduced VDR expression level in parathyroid tumors and investigate how

1,25(OH)2D3 affects the CKI p21

4) To assess whether 1α-hydroxylase is expressed in the normal and pathological parathyroid gland and quantify the 1α-hydroxylase mRNA expression in parathyroid tumors

5) To analyze if parathyroid glands express 25- and 24-hydroxylase and unravel a potential involvement in tumors of pHPT and sHPT

22

Subjects and methods

______

Subjects (I-V) The patients and controls used in the genetic association study of the VDR- FokI polymorphism were partly recruited by a population-based screening for pHPT (Lundgren et al., 1997). 23 Danish pHPT patients were also recruited since they had been subjected to measurements of the calcium-controlled PTH secretion in vivo. Two parathyroid carcinomas were provided by prof Henning Dralle, Germany. All other tissue specimens and clinical data were recruited from the clinical routine at Uppsala University Hospital. Informed consent and approval of ethical committee was achieved.

LOH Analysis (I) Leukocyte DNA was prepared by standard methods or using a genomic DNA isolation kit (Promega). Parathyroid tumors tissue was intraoperatively snap frozen in liquid nitrogen, stored at –70 °C and genomic DNA was extracted by standard procedures. The highly polymorphic microsatellite markers located at 1p; D1S243, D1S214, D1S244, D1S228, D1S170 and D1S199, at 6q; D6S269, D6S286, D6S287, D6S292, D6S311, D6S290, D6S305, D6S264, and D6S297, and at 11q13; PYGM (CA), INT-2, and D11S906 were 32P-labelled before the PCR. The PCR reaction (10 µl) contained 20 ng of genomic DNA, 0.2 U Taq polymerase in a 1x PCR buffer (Life Technologies), 1.5 mM MgCl2, 0.1 mM of each dNTP, and 2 pmol of each primer. PCR conditions included an initial denaturation at 95 °C for 2 min, followed by 27 cycles at 95 °C for 30 s, 55 –62 °C for 30 s and 72 °C for 30 s with a final extension at 72 °C for 7 min. PCR products were mixed with 10 µl formamide gel loading buffer, heat denatured at 80 °C for 5 min, electrophoresed in a denaturing 4.5 % polyacrylamide gel, and quantified on a PhosphorImager (Molecular Dynamics). Additional mapping with fluorescent

23 labelled microsatellite markers D1S468, D1S2893, D1S2870, D1S253, D1S2694, D1S503 and D1S450 on ABI 877 (Applied Biosystems) were performed. PCR reactions contained 10 ng of genomic DNA, 2 pmol of each primmer (labelled with HEX, 6-FAM or TET), 0.2 mM of each dNTP, 1x PCR buffer, 1.5 mM MgCl2 and 0.2 U Taq Platinum DNA polymerase in a final volume of 5 µl. PCR parameters used an initial denaturation at 95 °C for 2 min followed by 94 °C for 45 s, 57 °C for 45 s and 72 °C for 60 s for 30 cycles and a final extension of 72 °C for 6 min. PCR products were analysed on an ABI 310 semi-automated sequencer together with the size marker GeneScan 350 TAMRA and quantified using GeneScan software. LOH was identified, for either radioactive or fluorescent labelled makers, as total absence or reduction of >50% of the signal intensity of an allele in the tumor DNA versus constitutional DNA.

Genotype Analysis (II) Leukocyte DNA was prepared by standard methods or by using a genomic DNA isolation kit (Promega). The primers VDR2a and VDR2b and 2 µl of the

VDR exon

VDR

...GGACGGAGGCA ...GGATGGAGGCAATG Fo F f

Figure 6. Depiction of the FokI restriction site in the first ATG, the site (bold) is present in the lower DNA sequence (f) and absent in the upper (F). leukocyte DNA was used for PCR to amplify a 265-bp fragment containing the start codon polymorphism (Gross et al., 1996). The PCR products were digested with FokI at 37 °C for 3 h, followed by electrophoresis in a 1.5% agarose-gel containing ethidium bromide. Thus, the genotypes FF ( 265 bp), Ff (265, 196, 69 bp) and ff (196 and 69 bp) could be identified (Figure 6).

24

Citrate and Calcium Clamp (II) Twenty-three pHPT patients underwent a CiCa clamp technique protocol to evaluate the PTH dynamics to sequential induction of hypo-and hypercalcemia (Schwarz et al., 1993). Based on this method the calcium set-point (the blood Ca2+ concentration causing 50% inhibition of maximal PTH secretion) of each patient was determined.

Measurements of mRNA Levels Ribonuclease Protection Assay (II) VDR and PTH mRNA levels were determined by the ribonuclease protection assay (RNase protection assay) using total RNA from parathyroid adenomas of pHPT patients (Carling et al., 1998b) (Figure 7). In short, fragments of human

Figure 7. Principle of RNase protection assay.

VDR and PTH cDNA was subcloned into pBluescript II KS (Stratagene). Radio- labelled antisense RNA for VDR, PTH and glyceraldehyde 3-dehydrogenas (GAPDH) (Ambion) was produced by in vitro transcription from linearized plasmid

25 utilizing [α-32P] UTP (Amersham Biosciences) and T3 or T7 RNA polymerase (Stratagene). The RNA probes were purified on a 6% 7 M urea polyacrylamide gel and eluted overnight in a buffer containing 0.5 M NaAc (pH 7.0), 1 mM EDTA and 0.2% SDS. Total RNA from HeLa cells or parathyroid adenomas were subjected to RNase protection assay with RNase A (8 µg/ml) and RNase T1 (16 U/ml) (Forsberg et al., 1991). Excess 32P-labelled VDR, PTH and GAPDH antisense RNA was hybridized overnight to 10 µg total RNA. The samples were run on a 6% 7 M urea polyacrylamide gel after RNase digestion, Proteinase K treatment, phenol/chloroform extraction and ethanol precipitation. After overnight exposure the bands were quantified by PhosphorImager (Molecular Dynamics) analysis.

Real-time Quantitative PCR (III-V)

The principle of real-time quantitative PCR is based on the use of fluorescent- labelled probes designed to hybridize to the gene target sequence of the two PCR primers (Heid et al., 1996) (Figure 8). Each probe has a fluorescent reporter dye and a quencher dye attached at the 5’ and the 3’ ends, respectively. For

R Q 5´ 3 ´ 3 ´ 5´

5´ 3 ´ 5´

R Q 5´ 3 ´ 3 ´ 5´

5´ 5´

Q R 3 ´

5´ 3 ´ 5´

5´ 5´

Figure 8. The principle of real-time quantitative RT-PCR. R reporter, Q quencher.

26

intact probe the presence of quencher inhibits reporter dye emission by quenching energy emission. During the PCR extension phase, if the target of interest is present, the annealed probe is cleaved by the 5’ to 3’ exonuclease activity of Taq DNA polymerase. The cleavage produces an increase of reporter dye fluorescence emission. This occurs in every PCR cycle only if probe is annealed to the target sequence, which results in an increase of fluorescence proportional to the concentrations of target sequences in the initial sample. To normalize for pipetting errors and volume changes the reporter fluorescence is divided by the fluorescence of the passive reference to determine the normalized reporter signal. The fluorescence detection is performed with the ABI PRISM 7700 Sequence Detector (Applied Biosystems). A charge coupled device camera monitors the emission data every few seconds and the software analyzes the data. Quantification of the amount of target gene in unknown sample is accomplished by using a standard curve. We measured the mRNA levels of different VDR exons (1a, 1d, 1f and 9), 1α-hydroxylase, 24-hydroxylase and 25- hydroxylase in tumor and normal cDNAs and compared with the corresponding levels for GAPDH mRNA. The PCR mixtures contained 5 µl of cDNA template, 1x

TaqMan buffer A, 5.5 mM MgCl2 , 200 µM of dATP, dCTP, dGTP and 400 µM dUTP, 100 nM probe, 200 nM of each primer, 0.01 U AmpERase UNG¡ and 0.05 U ampliTaq Gold√. All regents were applied in the TaqMan PCR core Reagent Kit (Applied Biosystems). Each cDNA sample was analyzed in triplicate. Standard curves were established by amplifying a purified PCR fragment covering the sites for probes and primers.

Cell Culture (III) Biopsies of parathyroid adenomas were obtained from patients undergoing parathyroidectomy. After removal of visible fat and connective tissue, the adenomas was minced with scissors. Cell suspensions were prepared by digestion in 1 mg/ml collagenase (Sigma), 0.05 mg/ml DNaseI (Sigma), 1.5% bovine serum albumin (Sigma) and 1.25 mM Ca2+ as previously described

27 (Rudberg et al., 1986). After digestion in a shaking incubator for 45 min, the suspensions were filtered through nylon mesh (125 µm) and exposed to 1 mM EGTA in 25 mM Hepes buffer (pH 7.4) containing 142 mM NaCl and 6.7 mM KCl. Debris and dead cells were removed by centrifugation through 25% and 75% isotonic Percoll (Amersham Biosciences). Cell viability, as determined by the Trypan blue exclusion test, exceeded 95%. Cells were placed in 6-well dishes, with 106 cells/well, and cultured for 24 hours in DME/Hamm’s F-12 (50:50), 1 mM total calcium, 4% fetal calf serum, 15 mM Hepes, 100 IU/ml penicillin, 100 µg/ml streptomycin, 5 µg/ml insulin, 2 mM glutamine and 1% nonessential amino acids. Medium was then replaced with the same as above, with the exception for 1 mg/ml bovine serum albumin instead of calf serum, addition of holo-transferrin to

-8 5 µg/ml and 0.1% ethanol or 10 M 1,25(OH)2D3 (MacDonald et al., 1994). The serum-free medium was replenished once, 24 hours before the harvest. After 60 hours of hormone treatment cells were harvested and total RNA was prepared.

RT-PCR analysis (IV-V)

Oligonucleotide primers used for RT-PCR of 1α-hydroxylase, 24- and 25- hydroxylase were designed using the published human sequences in GeneBank. The primers generated fragments of 252 bp, 289 bp, and 250 bp respectively. Sequence analysis of the PCR fragments were undertaken on ABI 373A using ABI prism dye terminator cycle sequencing ready reaction kit (Applied Biosystems). RT-PCR analysis of GAPDH was performed as a control to verify the amount of integrity of mRNA in RNA preparations from the different tissues.

Immunohistochemistry (IV-V) Tissues intended for immunohistochemistry were peroperatively snap frozen and stored at -70 °C. Cryosections of 6 µm were fixed in acetone followed by quenching of endogenous peroxidase activity in 0.3 % H2O2 in methanol for 15 min, and blocked with normal rabbit serum (1:10) or with an avidin-biotin blocking kit (Vector Lab). The 1α-hydroxylase (Zehnder et al., 1999) polyclonal peptide

28

antisera (1:300), the megalin/LRP-2 (Juhlin et al., 1987) monoclonal antibody E11 (10 µg/ml) or polyclonal 25-hydroxylase antibody were applied to the tissue sections and incubated for 90 min at room temperature (the antisera was diluted in 0.1M Tris, pH 7.4, containing 10% normal swine serum). The slides were exposed during 30 minutes to rabbit anti-mouse IgG (1:40) or biotin-labelled donkey anti-sheep IgG (1:500), after which a mouse peroxidase anti-peroxidase complex (1:250) or an avidin-biotin complex (Vector Lab) was applied. Slides were developed using 3-amino-9-ethylcarbazole and counterstained with Mayer's hematoxylin. Control sections included use of non-immune mouse serum or of primary antiserum preincubated with an excess of immunizing peptide.

Statistical Analysis (I-V) All statistical analyses were accomplished with StatView 5.1. Values are presented as the mean ± SEM. Statistical analyses comprised Chi2-test (paper I- II), unpaired t test (paper I,-II, IV-V) and ANOVA followed by Fischer’s PLSD (paper II-III). p<0.05 was considered significant.

29 Results and discussion

______

LOH studies in mild and severe primary hyperparathyroidism – Paper I As described previously allelic losses are common in parathyroid adenomas, recognising 1p, 6q, and 11q13 as chromosomal hotspots for these changes. LOH studies rely on Knudson two hit hypothesis, namely two inactivated gene copies result in loss of function and thereby tumor formation. Comparison of clinically and screening-detected pHPT is of interest since this may identify which tumors are determinant to progress or correlate to a more severe clinical picture. We examined 56 patients with pHPT, of which 21 were recruited from a population- based screening, for allelic losses. We used 6, 9 and 3 microsatellite markers for chromosome 1p, 6q and 11q13 respectively. All patients were informative i.e. heterozygous for at least two microsatellite markers used for each chromosomal region. 27%, 23% and 23% of the tumors showed LOH at chromosome 1p, 6q, and 11q13. The LOH pattern of allelic loss at both 1p and 11q13 was more common in the tumors of screening-detected pHPT patients compared to those recruited from the clinical routine (38% versus 20%; p=0.02) and (43% versus 11%; p=0.001), while loss at 6q was more prevalent in the clinically recruited group (31% versus 10%; p=0.001). We performed a deletion mapping study of chromosome 1p using 7 additional markers in the tumors displaying LOH in the first round of analysis. Although scattered LOH could be detected using all makers, the most frequent chromosome region demonstrating LOH was an approximately 6 cM region between D1S214 and D1S503. At 6q the most common allelic loss involved the region spanning D6S292 to D6S305 and all three 11q13 markers showed the same frequency of allelic losses. There was no apparent relationship between presence and absence of LOH at either chromosome and clinical characteristics such as glandular weight, s-calcium or PTH.

30

This study provided a possibility to study the different patterns of allelic loss in asymptomatic patients with truly mild pHPT, with s-calcium and s-PTH levels within the reference range or only slightly elevated and with particular small parathyroid tumors. We speculate that the increased prevalence of LOH at 1p and 11q13 may be associated with tumor gene inactivation causing increase in parathyroid cell proliferation but despite this not causing a clinically overt hypercalcemic syndrome. This hypothesis is consistent with the demonstration of MEN1 gene mutations in parathyroid tumors of truly mild pHPT (Carling et al., 1998a). On the contrary LOH at 6q, which was more common in the patients recruited from the clinical routine, could be associated with gene inactivation resulting in more pronounced parathyroid abnormality. Furthermore, this study narrowed a region on chromosome 1p36.31 between D1S214 and D1S503 containing about 10 to 12 genes. We believe that future studies characterizing the molecular events causing variants of pHPT may become important for the clinical management of these patients.

The vitamin D receptor start codon polymorphism in primary hyperparathyroidism – Paper II VDR alleles comprise risk factors for pHPT and for a number of pathological states such as osteoarthritis (Uitterlinden et al., 1997), Graves’ disease (Ban et al., 2000), Crohn's disease (Simmons et al., 2000), malignant melanoma (Hutchinson et al., 2000) and rectal cancer (Speer et al., 2000). Since the polymorphisms in intron 8 and exon 9 of the VDR gene do not alter the amino acid sequence it has been suggested that they are in linkage disequilibrium with other polymorphisms of the VDR gene or other genes with possible impact on parathyroid tumorigenesis. These suggestions may also explain why silent polymorphisms appear to be related to less calcium sensitive PTH release, increased PTH mRNA levels and decreased VDR mRNA levels (Carling et al., 1997b, Carling et al.,1998b). The VDR-FokI polymorphism changes the translation start site of the VDR mRNA, resulting in a three amino acid shorter VDR protein in the FF variant (Baker et al., 1988). The extent of vitamin D-

31 dependent transcriptional activation of a reporter construct under control of a VDRE in transfected cells was almost two fold greater for the FF variant in comparison to the ff variant (Arai et al., 1998). However another recent study could not detect significant differences in ligand affinity, DNA binding or transactivation activity between f-VDR and F-VDR forms (Gross et al., 1998). The ff variant had previously been associated with reduced bone mineral density (BMD) in postmenopausal women (Gross et al., 1996, Harris et al., 1997). We genotyped 182 postmenopausal women with sporadic pHPT and matched controls. There was a tendency to underrepresentation of the ff genotype in the screening-detected and all pHPT patients compared to the controls (p=0.07 and 0.09 respectively). However, the allele frequencies in all pHPT for the f and F alleles were 38% and 62% versus 45% and 54% in the controls (p=0.05). Remarkably it was the F allele that was weakly related to pHPT and not the f allele, which previously has been associated to reduced BMD, implicating that the f-VDR variant may be more important for bone turnover than parathyroid function. To further explore the functional importance of VDR-FokI genotypes we compared genotypes not only to parathyroid VDR and PTH expression but also to pHPT patients subjected to subquential induction of hypo- and hypercalcemia (CiCa-clamp technique). We were unable to demonstrate any significant differences in mRNA expression and although all patients showed the expected right-shift in the calcium set-point the calcium set-points did not differ significantly among the FF, Ff and ff genotypes. The VDR-FokI polymorphism was not in linkage disequilibrium with the BsmI, ApaI and TaqI polymorphisms either. VDR-FokI polymorphism seems to be at most weakly associated with pHPT, as lately confirmed by others (Sosa et al., 2000), and the variant of the protein does not seem to influence the pHPT characteristics. Subsequently this polymorphism is not a sufficient marker for pHPT or any clinical characteristic of the disease. The molecular mechanisms of how silent VDR polymorphisms affect pHPT, most likely by directing expression levels of proteins important for regulation of parathyroid functions, must be sought elsewhere.

32

Vitamin D receptor exon 1f transcripts in primary hyperparathyroidism – Paper III Both parathyroid tumors of primary and secondary origin have been associated with reduced VDR mRNA and protein expression (Fukuda et al., 1993, Sudhaker Rao et al., 2000, Carling et al., 2000). Reduced VDR expression is thought to interfere with the inhibitory activity of 1,25(OH)2D3 on PTH transcription and parathyroid cell proliferation (Haussler et al., 1997, Hellman et

1 1 2

1 Figure 9. Structure of 5’- region in the human VDR gene. Transcripts 1-5 (from top) originate from exon 1a. Transcript 1 corresponds to the 1 published cDNA. Transcripts 6-10 originate form exon 1d and transcripts 11-14 originate from exon 1f.

1

1

al., 1999, Jones et al., 1999a, Silver et al., 1999). Hormone activity requires the

VDR receptor and the growth-regulating effects of 1,25(OH)2D3 have been related to changed CKI p21 expression in various tissues (Liu et al., 1996, Kawa et al., 1997, Hershberger et al., 1999). Recently Miyamato et al (1997) and Crofts et al (1988) discovered novel VDR upstream exons giving raise to 14 different transcripts (Figure 9). The 1d transcripts seemed to be the most abundant ones in VDR target tissues but since the 1f transcripts appeared to be tissue- specifically expressed in target tissues for the calcitrophic effects of vitamin D, we hypothesized that these transcripts may be particularly involved in parathyroid tumorigenesis. In order to determine the contributions of 5’ terminal exon variant

33 VDR gene transcripts to the previously observed underexpression of VDR mRNA (exon 9) in lesions of primary and secondary HPT we designed probes and primers specific for the VDR exons 1f, 1a, 1d and 9. We calculated the relative VDR/GAPDH mRNA levels for each exon by real-time quantitative RT-PCR in 5 normal glands, 15 parathyroid adenomas and 10 hyperplastic glands of sHPT. In agreement with results using RNase protection assay VDR exon 9 transcripts were significantly reduced in parathyroid adenomas (p<0.04) and hyperplasias of sHPT (p<0.001). Concerning the other exons only the most distal exon 1f was significantly decreased in parathyroid adenomas (p<0.001) while exons 1f, 1a and 1d were all significantly reduced (p<0.001, p<0.04, and p<0.04 respectively) in hyperplasias of sHPT. To investigate whether VDR and its ligand exerted its effects via CKI p21 we measured the relative p21 expression levels in the above described tumors but revealed no significant difference (p>0.05) (Figure 10). We also exposed human adenoma cells harbouring reduced VDR exon 1f mRNA levels from two different individuals to 1,25(OH)2D3 or vehicle (ethanol). This resulted in a significant reduction of the p21/GAPDH ratio in 1,25(OH)2D3 treated adenoma cells from one individual but not from the other (Figure 10). Furthermore the p21/GAPDH mRNA ratio was significantly lower in non-cultured cells compared to the cultured cells.

Figure 10. p21/GAPDH relative expression ratio was determined by real-time quantitative RT-PCR in A) 5 normal parathyroid glands, 17 parathyroid adenomas and 10 hyperplastic glands of sHPT B) parathyroid adenoma cells, from one individual, exposed to 10-8 M 1,25(OH)2D3 or vehicle (ethanol) or non-cultured. p-values were calculated using ANOVA.

This study further emphasizes a distinct pathogenic pathway for primary and

34

secondary parathyroid tumors. The observed underexpression of exon 1f transcripts may be due to mutations in promoter elements, or indirectly through inactivating mutations or aberrant expression of parathyroid transcription factor gene(s). Most likely the non exon specific down-regulation of VDR transcripts in sHPT reflects physiological features such as hypocalcemia and hyperphosphatemia. The unaltered p21 expression levels in parathyroid adenomas and hyperplasias of sHPT might reflect steady state levels. The preliminary cell culture experiments indicate that p21 can be modulated by

1,25(OH)2D3 despite reduced VDR expression and this may also contribute to the normal p21 mRNA steady state levels observed in parathyroid tumors. Further studies are warranted in parathyroid tissues and other tissues to reveal the precise role of the different VDR transcripts and potentially identify a novel promoter.

25-hydroxyvitamin D3 1α-hydroxylase, 25- hydroxyvitamin D3 24- hydroxylase and vitamin D3 25- hydroxylase in parathyroid tumors– Paper IV-V Nykjaer et al (1999) identified megalin/LRP-2 as an endocytic receptor for

25(OH)D3 and we subsequently speculated that the parathyroid glands expressed 1α-hydroxylase. This was verified by RT-PCR and immunohistochemistry. It seems like renal and parathyroid 1α-hydroxylase activity is due to a single gene product since the DNA sequence of the parathyroid PCR product was identical to the published renal one (Monkawa et al., 2000). The enzyme was present in the parathyroid chief cells and the 1α- hydroxylase expression was lower in the parathyroid glands compared to kidney and as expected hepatocytes did not express 1α-hydroxylase at all. We examined the parathyroid glands for 25- and 24-hydroxylase, to evaluate whether the parathyroid glands possess the ability to 25-hydroxylate vitamin D3 and initially degrade 1,25(OH)2D3 by 24-hydroxylation. 24-hydroxylase has previously been reported in parathyroid glands at protein level (Jones et al., 1999a) and we

35 confirmed this result by RT-PCR analysis. Using RT-PCR and polyclonal antibodies (kindly provided by Dr John Russell) (Cali et al., 1991) we were able to identify 25-hydroxylase (CYP27A) transcripts as well as protein (Figure 11) in

Figure 11. Immunohistochemical detection of 25-hydroxylase protein expression and control in frozen parathyroid adenoma tissue sections.

parathyroid adenomas. To evaluate if 1α-hydroxylase, 24- or 25-hydoxylase are involved in parathyroid tumorigenesis we measured mRNA levels in normal parathyroid glands, parathyroid adenomas, hyperplastic glands of sHPT and parathyroid carcinomas by real-time quantitative RT-PCR. For 1α-hydroxylase the material consisted of 5 normal parathyroid gland, 15 parathyroid adenomas, 10 hyperplastic glands of sHPT and 5 parathyroid carcinomas. The same material apart from one parathyroid adenoma and one hyperplastic gland of sHPT were analyzed for 24- and 25-hydroxylase. mRNA-specific primers revealed variably increased 1α-hydroxylase/GAPDH ratio in the lesions from both primary (80.2 ± 18.0, p=0.03) and secondary (49.3 ± 12.8, p=0.03) HPT compared to normal parathyroid tissues (5.1 ± 1.9). The parathyroid carcinomas exhibited lower 1α- hydroxylase expression than normal glands (0.4 ± 0.11, p=0.04). For 24- hydroxylase we detected a downregulation, compared to normal parathyroid glands with 58% in the adenomas (p=0.01), 89% in sHPT (p=0.0001) and 96% in carcinomas (p=0.0003). 25-hydroxylase mRNA expression was reduced by 79% in adenomas (p=0.001), 88% in sHPT (p=0.002) and 96% in carcinomas (p=0.01). No correlation between 24-hydroxylase expression level and s-calcium, s-PTH, gland weight at time of surgery or s-creatinine were found. 61% of the examined tumors demonstrated both upregulation of 1α-hydroxylase and downregulation of 24-hydroxylase compared to normal glands. A few adenomas

36

expressed normal levels of 24-hydroxylase (3 adenomas) and 25-hydroxylase (1 adenoma). All these four tumors showed an increase of 1α-hydroxylase expression but unchanged 24- and 25-hydroxylase respectively. The overexpression of 1α-hydroxylase in a majority of parathyroid adenomas of primary HPT and in hyperplastic glands of sHPT, compared to normal parathyroid glands, indicates an increased or at least maintained local production of 1,25(OH)2D3. This findings suggest that 1,25(OH)2D3, by acting like an autocrine or paracrine factor, executes growth-controlling and possibly differentiating effects on parathyroid cells in both pHPT and sHPT. This theory is further supported by the reduced 1α-hydroxylase levels in parathyroid carcinomas and the decreased 24-hydroxylase levels in the lesions of pHPT and sHPT. Possibly, the decreased VDR levels in parathyroid tumors may trigger these events. Recent studies have confirmed an autocrine/paracrine role of 1α- hydroxylase in certain cells such as activated macrophages and keratinocytes (Zehnder et al., 1999) in contrast to the distinct endocrine actions of the enzyme in kidney tubule cells. Colon adenocarcinoma (Cross et al., 1997, Tangpricha et al., 2001, Oglunkolade et al., 2002), non-small cell lung cancer (Jones et al., 1999b), prostate cancer (Zhao et al., 2001) and cervix cancer (Friedrich et al., 2002) also have capacity to express 1α-hydroxylase suggesting that they do not only rely on systemic sources but can produce 1,25(OH)2D3 themselves. In colon carcinomas 1α-hydroxylase expression increases in the early phase of cancerogenesis whereas poorly differentiate late stage carcinomas show low levels (Cross et al., 2001), consistent with our observations in parathyroid tumors. Possibly the PTH gene may be one of the stimulators of parathyroid 1α- hydroxylase gene induction and parathyroid 24-hydroxylase inhibition concordant with PTH function in the kidney. Megalin/LRP-2, the suggested receptor of

25(OH)D3 uptake in the parathyroid glands, is reduced on both mRNA and protein levels in parathyroid adenomas and in hyperplastic glands of sHPT. The decreased 25-hydroxylase expression may reflect sufficient transmembrane transportation of 25(OH)D3 or that another enzyme is responsible for 25- hydroxylation.

37 In conclusion, parathyroid tumors seem to demonstrate aberrant expression of 1α-hydroxylase and 24-hydroxylase possibly causing increased levels of

1,25(OH)2D3. Since active vitamin D has well-characterized antiproliferative and differentiating features in multiple cell types, this may contribute to the benign features of parathyroid tumors of primary and secondary origin.

38

General discussion ______

Traditionally it is believed that certain features are crucial for tumorigenesis namely disrespect of antiproliferating and differentiating signals, sustained proliferation, evading apoptosis and angiogenesis. Parathyroid tumors have particularly interesting features since they most commonly are of benign nature and contain key characteristics for benign disease that may be of great importance to compare to truly malignant states. The genetic pattern demonstrated in parathyroid tumors is probably not random. The specific associations of events in individual tumors presumably reflect the evolution of the tumors along specific pathways. These pathways might be useful tools for genetic profiling which may provide information of clinical value. Furthermore finding genetic homogenous tumors may be valuable for environmental predisposition, which may pinpoint why pHPT has specially high frequency in postmenopausal women. This work demonstrated chromosome 1p and 11q13 to be associated with a more benign pHPT disease and further studies will likely provide better genetic guidance. Application of 1,25(OH)2D3 in treatment of pathological states such as myeolid leukemias and psoriasis has gained great interest. This thesis has proved that enzymes and receptor involved in the vitamin D metabolism conduct special significance in parathyroid tumorigenesis. The well-established effects of vitamin D on PTH transcription and parathyroid cell proliferation are probably just the tip of the iceberg on how vitamin D affect parathyroid cells. Multiple functions of vitamin D have been described in various tissues involving not only prodifferentiating and antiproliferative effects but also immunomodulation and enhancement of peptides and proteins. Several epidemiologic studies have indicated that diet and exposure of UV light are related to reduced risk of developing several types of malignancies for example colon, breast and prostate. Most likely vitamin D promotes angiogenesis, apoptosis and antimetastatic/invasive effects in several tissues including the parathyroids. Treatment or prevention of various cancers will undoubtedly depend of the

39 development of synthetic analogues of vitamin D. Vitamin D analogues are already in use for sHPT but will hopefully be usable for mild pHPT as well. New analogues will need higher efficacy and organ selectivity without inducing hypercalcemia and hypercalcuria. Certainly the vitamin D autocrine/paracrine system in normal and pathological parathyroid glands needs further investigation.

40

Acknowledgements ______

I am truly grateful and owe my sincerest thanks to:

My supervisors Gunnar Westin my source of wisdom, guiding star and scientific tutor and Göran Åkerström for inspiration and outstanding mentorship.

Daniel Lindberg my “brother” at the lab for discussions, arguments, laughs and mischief.

Tobias Carling and Jonas Rastad for giving me a kick start, Per Hellman for fruitful suggestions.

To the co-authors.

Colleagues and friends in my research group Ulrika Segersten, Eva Szabo, Birgitta Bondeson, Peter Lillhager, Ola Hessman, Peyman Björklund and Anders Knutson.

Helena Brändström, Katarina Lindroos and Thomas Lorant for brightening up the days at the research department.

To my best friends Gabriel and Anna G.

To my parents Fernando and Eva-Gun and twinsister Rebeca for continuous love and support.

….und schliesslich an Fredrik, da du mir bei Erfolg und Misserfolg zur Seite stehst.

41

References ______

Abdullah H, Bliss R, Guinea AI et al. Pathology and outcome of surgical treatment for lithium-associated hyperparathyroidism. Br J Surg 1999;86:91-3

Abbona GC, Papotti M, Gasparri G et al. Proliferative activity in parathyroid tumors as detected by Ki-67 immunostaining. Hum Pathol 1995;26:135-8

Agarwal SK, Schröck E, Kester MB et al. Comparative genomic hybridization analysis of human parathyroid tumors. Cancer Genet Cytogenet 1998;106:30-6

Agarwal SK, Guru SC, Heppner C et al. Menin interacts with the AP1 transcription factor JunD and represses JunD-activated transcription. Cell 1999;96:143-52

Arai H, Miyamoto KI, Taketani Y et al. A vitamin D receptor gene polymorphism in the translation initiation codon: effect on protein activity and relation to bone mineral density in Japanese women. J Bone Miner Res 1997;12:915-21

Arnold A, Staunton CE, Kim HG et al. Monoclonality and abnormal parathyroid hormone gene in parathyroid adenomas. N Engl J Med 1988;318:658-62

Arnold A, Brown MF, Urena P et al. Monoclonality of parathyroid tumors in chronic renal failure and in primary parathyroid hyperplasia. J Clin Invest 1995;95:2047-53

Axén E, Postlind H, Wikvall K. Effects on CYP27 mRNA expression in rat kidney and liver by 1α,25-dihydroxyvitamin D3, a suppressor of renal 25-hydroxyvitamin D3 1α- hydroxylase activity. Biochem Biophys Res Commun 1995;215:136-41

Baker AR, McDonnell DP, Hughes M et al. Cloning and expression of full-length cDNA encoding human vitamin D receptor. Proc Natl Acad Sci USA 1988;85:3294-8

Ban Y, Taniyama M, Ban Y. Vitamin D receptor gene polymorphism is associated with Graves’ disease in the Japanese population. J Clin Endocrinol Metab 2000;85:4639-43

Baylind SB, Herman JG. DNA methylation in tumorigenesis. Trends Genet 2000;16:168- 74

Bikle DD, Nemanic MK, Whitney JO et al. Neonatal human foreskin keratinocytes produce 1,25-dihydroxyvitamin D3. Biochemistry 1986;25:1545-8

Brown EM, Gamba G, Riccardi D et al. Cloning and characterization of an extracellular Ca2+-sensing receptor from bovine parathyroid. Nature 1993;366:575-80

Brown EM. Calcium receptor and regulation of parathyroid hormone secretion. Rev Endocr Metab Disord 2000;1:307-15

Brown SB, Brierley TT, Palanisamy N et al. Vitamin D receptor as a candidate tumor- suppressor gene in severe hyperparathyroidism of uremia. J Clin Endocrinol Metab 2000;85:868-72

Byrne IM, Flanagan L, Tenniswood MP et al. Identification of a hormone-responsive promoter immediately upstream of exon 1c in the human vitamin D receptor gene. Endocrinology 2000;141:2829-36

42

Cali JJ, Hsieh CL, Francke U et al. Mutations in the bile acid biosynthetic enzyme sterol 27-hydroxylase underlie cerebrotendinous xanthomatosis. J Biol Chem 1991;266:7779- 83

Carling T, Kindmark A, Hellman P et al. Vitamin D receptor alleles b, a and T: risk factors for sporadic primary hyperparathyroidism (HPT) but not HPT of uremia or MEN1. Biochem Biophys Res Comm 1997a;231:329-32

Carling T, Ridefelt P, Hellman P et al. Vitamin D receptor polymorphisms correlate to parathyroid cell function in primary hyperparathyroidism. J Clin Endocrinol Metab 1997b;82:1772-5

Carling T, Correa P, Hessman O et al. Parathyroid MEN1 gene mutations in relation to clinical characteristics of nonfamilial primary hyperparathyroidism. J Clin Endocrinol Metab 1998a;83:2960-3

Carling T, Åkerström G, Rastad J et al. Vitamin D receptor (VDR) and parathyroid hormone messenger ribonucleic acid levels correspond to polymorphic VDR alleles in human parathyroid tumors. J Clin Endocrinol Metab 1998b;83:2255-9

Carling T, Imanishi Y, Gaz RD et al. Analysis of the RAD54 gene on chromosome 1p as a potential tumor-suppressor gene in parathyroid adenomas. Int J Cancer 1999a;83:80-2

Carling T, Imanishi Y, Gaz RD et al. RAD51 as a candidate parathyroid tumor suppressor gene on chromosome 15q: absence of somatic mutations. Clin Endocrinol 1999b;51:403- 7

Carling T, Rastad J, Szabo E et al. Reduced parathyroid vitamin D receptor messenger ribonucleic acid levels in primary and secondary hyperparathyroidism. J Clin Endocrinol Metab 2000;85:2000-3

Cetani F, Pinchera A, Pardi E et al. No evidence for mutations in the calcium-sensing receptor gene in sporadic parathyroid adenomas. J Bone Miner Res 1999;14:878-82

Chandrasekharappa SK, Guru SC, Manickham P et al. Positional cloning of the gene for multiple neoplasia type 1. Science 1997;276:404-7

Chudek J, Ritz E, Kovacs G. Genetic abnormalities in parathyroid nodules of uremic patients. Clin Cancer Res 1998;4:211-4

Cole DEC, Peltekova VD, Rubin LA et al. A986S polymorphism of the calcium-sensing receptor and circulating calcium concentrations. Lancet 1999;353:112-5

Correa P, Åkerström G, Westin G. Underexpression of Gcm2, a master regulatory gene of parathyroid gland development, in adenomas of primary hyperparathyroidism. Clin Endocrinol 2002a;in press

Correa P, Lundgren E, Rastad J et al. The MEN1 gene polymorphism D418D is associated with sporadic primary hyperparathyroidism. Surgery 2002b;in press

Crabtree JS, Scacheri PC, Ward JM et al. A mouse model of multiple endocrine neoplasia, type 1, develops multiple endocrine tumors. Proc Natl Acad Sci USA 2001;98:1118-23

Crofts LA, Hancock MS, Morrison NA et al. Multiple promoters direct the tissue-specific expression of novel N-terminal variant human vitamin D receptor gene transcripts. Proc 43

Natl Acad Sci USA 1998;95:10529-34

Cross HS, Peterlik M, Reddy GS et al. Vitamin D metabolism in human colon- adenocarcinoma derived Caco-2 cells. Expression of 25-hydroxyvitamin D3 -1α- hydroxylase activity and regulation of side-chain metabolism. J Steroid Biochem Mol Biol 1997;62:21-8

Cross HS, Bareis P, Hofer H et al. 25-hydroxyvitamin D3-1α-hydroxylase and vitamin D receptor gene expression in human colonic mucosa is elevated during early cancerogenesis. Steroids 2001;66:287-92

Cryns VL, Thor A, Xu HJ et al. Loss of the retinoblastoma tumor-suppressor gene in parathyroid carcinoma. N Engl J Med 1994;330:757-61

Degenhardt S, Toell A, Weidemann W et al. Point mutations of the human parathyroid calcium receptor gene are not responsible for non-suppressible renal hyperparathyroidism. Kidney Int 1998;53:556-61

Delvin EE, Arabian A. Kinetics and regulation of 25-hydroxycholecalciferol 1-α- hydroxylase from cells isolated from human term decidua. Eur J Biochem 1987;163:659- 62

Diaz L, Sanchez I, Avila E et al. Identification of 25-hydroxyvitamin D3 1α-hydroxylase gene transcription product in cultures of human syncytiotrophoblast cells. J Clin Endocrinol Metab 2000;85:2543-9

Dotzenrath C, Teh BT, Farnebo F et al. Allelic loss of the retinoblastoma tumor suppressor gene: a marker for aggressive parathyroid tumors? J Clin Endocrinol Metab 1996;81:3194-6

Farnebo F, Teh BT, Dotzenrath C et al. Differential loss of heterozygosity in familial, sporadic, and uremic hyperparathyroidism. Hum Genet 1997a;99:342-9

Farnebo F, Enberg U, Grimelius L et al. Tumor-specific decreased expression of calcium sensing receptor messenger ribonucleic acid in sporadic primary hyperparathyroidism. J Clin Endocrinol Metab 1997b;82:3481-6

Farnebo F, Höög A, Sandelin K et al. Decreased expression of calcium-sensing messenger ribonucleic acids in parathyroid adenomas. Surgery 1998a;124:1094-8

Farnebo F, Teh BT, Kytölä S et al. Alterations of the MEN1 gene in sporadic parathyroid tumors. J Clin Endocrinol Metab 1998b;83:2627-30

Farnebo F, Auer G, Farnebo LO et al. Evaluation of retionoblastoma and Ki-67 immunostaining as diagnostic markers of benign and malignant parathyroid disease. World J Surg 1999;23:68-74

Forsberg M, Westin G. Enhancer activation by a single type of transcription factor shows cell type dependence. EMBO J 1991;10:2543-51

Franklin DS, Godfrey VL, O’Brien DA et al. Functional collaboration between different cyclin-dependent kinase inhibitors suppresses tumor growth with distinct tissue specificity. Mol Cell Biol 2000;20:6147-58

Friedman E, Sakaguchi K, Bale AE et al. Clonality of parathyroid tumors in familial multiple endocrine neoplasia type 1. N Engl J Med 1989;321:213-8 44

Friedman E, Bale AE, Marx SJ et al. Genetic abnormalities in sporadic parathyroid adenomas. J Clin Endocrinol Metab 1990;71:293-7

Friedrich M, Villena-Heinsen C, Axt-Fliedner R et al. Analysis of 25-hydroxyvitamin D3- 1α-hydroxylase in cervical tissue. Anticancer Res 2002;22:183-6

Fu GK, Lin D, Zhang MYH et al. Cloning of human 25-hydroxyvitamin D-1α-hydroxylase and mutations causing vitamin D-dependent rickets type 1. Mol Endocrinol 1997;11:1961- 70

Fukuda N, Tanaka H, Tominaga Y et al. Decreased 1,25-dihydroxyvitamin D3 receptor density is associated with a more severe form of parathyroid hyperplasia in chronic uremic patients. J Clin Invest 1993;92:1436-43

Gobl AE, Berg M, Lopes-Egido JR et al. Menin represses JunD-activated transcription by a histone deacetylase-dependent mechanism. Biochim Biophys Acta 1999;1447:51-6

Gogusev J, Duchambon P, Hory B et al. Depressed expression of calcium receptor in parathyroid gland tissue of patients with hyperparathyroidism. Kidney Int 1997;51:328-36

Gross C, Eccleshall TR, Malloy PJ et al. The presence of a polymorphism at the translation initiation site of the vitamin D receptor gene is associated with low bone mineral density in postmenopausal Mexican-American women. J Bone Miner Res 1996;11:1850-5

Gross C, Krishnan AV, Malloy PJ et al. The vitamin D receptor gene start codon polymorphism: a functional analysis of FokI variants. J Bone Miner Res 1998;13:1691-9

Günther T, Chen ZF, Kim J et al. Genetic ablation of parathyroid glands reveals another source of parathyroid hormone. Nature 2000;406:199-203

Hakim JP, Levine MA. Absence of p53 point mutations in parathyroid adenoma and carcinoma. J Clin Endocrinol Metab 1994;78:103-6

Harris SS, Eccleshall TR, Gross C et al. The vitamin D receptor start codon polymorphism (FokI) and bone mineral density in premenopausal American black and white women. J Bone Miner Res 1997:12:1043-8

Haussler MR, Haussler CA, Jurutka PW et al. The vitamin D hormone and its : molecular actions and disease states. J Endocrinol 1997;154 Suppl:S57-73

Heid CA, Stevens J, Livak KJ et al. Real time quantitative PCR. Genome Res 1996;6:986-94

Hellman P, Liu W, Westin G et al. Vitamin D and retinoids in parathyroid glands (Review). Int J Mol Med 1999;3:355-61

Heppner C, Kester MB, Agarwal SK et al. Somatic mutation of the MEN1 gene in parathyroid tumors. Nat Genet 1997;16:375-8

Hershberger PA, Modezlewski RA, Shurin ZR et al. 1,25-dihydroxycholecalciferol (1,25- D3) inhibits growth of squamous cell carcinoma and down-modulates p21Waf1/Cip1 in vitro and in vivo. Cancer Res 1999;59:2644-9

Hosokowa Y, Tu T, Tahara H et al. Absence of cyclin D1/PRAD1 point mutations in 45

human breast cancers and parathyroid adenomas and identification of a new cyclin D1 gene polymorphism. Cancer Lett 1995;93:165-70

Howard GA, Turner RT, Sherrard DJ et al. Human bone cells in culture metabolize 25- hydroxyvitamin D3 to 1,25-dihydroxyvitamin D3and 24,25-didhydroxyvitamin D3. J Biol Chem 1981;256:7738-40

Hsi ED, Zuckerberg LR, Yang WI et al. Cyclin D1/PRAD1 expression in parathyroid adenomas: an immunohistochemical study. J Clin Endocrinol Metab 1996;81:1736-9

Hunter T, Pines J. Cyclins and cancer II: Cyclin D and CDK inhibitors come of age. Cell 1994;79:573-82

Hutchinson PE, Osborne JE, Lear JT et al. Vitamin D receptor polymorphisms are associated with altered prognosis in patients with malignant melanoma. Clin Cancer Res 2000;6:498-504

Ichikawa F, Sato K, Nanjo M et al. Mouse primary osteoblasts express vitamin D3 25- hydroxylase mRNA and convert 1α-hydroxyvitamin D3 into 1α, 25 dihydroxyvitamin D3. Bone 1995;16:129-35

Imanishi Y, Hosokawa Y, Yoshimoto K et al. Primary hyperparathyroidism caused by parathyroid-targeted overexpression of cyclin D1 in transgenic mice. J Clin Invest 2001;107:1093-102

Jones G, Strugnell SA, DeLuca HF. Current understanding of the molecular actions of vitamin D. Physiol Rev 1999a;78:1193-231

Jones G, Ramshaw H, Zhang A et al. Expression and activity of vitamin D-metabolizing cytochrome P450s (CYP1α and CYP24) in human nonsmall cell lung carcinomas. Endocrinology 1999b;140:3303-10

Juhlin C, Holmdahl R, Johansson H et al. Monoclonal antibodies with exclusive reactivity against parathyroid cells and tubule cells of the kidney. Proc Natl Acad Sci USA 1987;84:2990-4

Juhlin C, Åkerström G, Klareskog L et al. Monoclonal antiparathyroid antibodies revealing defect expression of a calcium receptor mechanism in hyperparathyroidism. World J Surg 1988a;12:552-8

Juhlin C, Klareskog L, Nygren P et al. Hyperparathyroidism is associated with reduced expression of a parathyroid calcium receptor mechanism defined by monoclonal antiparathyroid antibodies. Endocrinology 1988b;122:2999-01

Kawa S, Nikaido T, Aoki Y et al. Vitamin D analogues up-regulate p21 and p27 during growth inhibition of pancreatic cancer cell lines. Br J Cancer 1997;76:884-9

Kaji H, Canaff L, Lebrun JJ et al. Inactivation of menin, a Smad3-interacting protein, blocks transforming growth factor type β signaling. Proc Natl Acad Sci USA 2001;98:3837-42

Kinzler KW, Vogelstein B. Cancer-susceptibility genes: gatekeepers and caretakers. Nature 1997;386:761-3

Kytölä S, Farnebo F, Obara T et al. Patterns of chromosomal imbalances in parathyroid carcinomas. Am J Pathol 2000;157:579-86 46

Lemmens IH, Forsberg L, Pannett AAJ et al. Menin interacts directly with the - containing protein Pem. Biochem Biophys Res Commun 2001;286:426-31

Liu M, Lee MH, Cohen M et al. Transcriptional activation of the Cdk inhibitor p21 by vitamin D3 leads to the induced differentiation of the myelomonocytic cell line U937. Genes Dev 1996;10:142-53

Lopez-Egido JR, Cunningham J, Berg M et al. Menin’s interaction with glial fibrillary acidic protein and vimentin suggests a role for the intermediate filament network in regulating menin activity. Exp Cell Res 2002;278:175-83

Lundgren E, Rastad J, Thurfjell E et al. Population-based screening for primary hyperparathyroidism with serum calcium and parathyroid hormone values in menopausal women. Surgery 1997;121:287-94

Lundgren E, Ljunghall S, Åkerström G et al. Case-control study on symptoms and signs of ”asymptomatic” primary hyperparathyroidism. Surgery 1998;124:980-5

MacDonald PN, Ritter C, Brown AJ et al. Retinoic acid suppresses parathyroid hormone (PTH) secretion and preproPTH mRNA levels in bovine cell culture. J Clin Invest 1994;93:725-30

Manley NR, Capecchi MR. Hox group 3 paralogs regulate the development and migration of the thymus, thyroid, and parathyroid glands. Dev Biol 1998;195:1-15

Mallette LE. The functional and pathologic spectrum of parathyroid abnormalities in hyperparathyroidism. In: The parathyroids, Bilezikian JP, Marcus R, Levine MA (eds). New York, Raven Press, 1994;423-57

Miyamato K, Kesterson RA, Yamamoto H et al. Structural organization of the human vitamin D receptor chromosomal gene and its promoter. Mol Endocrinol 1997;11:1165-79

Monkawa T, Yoshida T, Hayashi M et al. Identification of 25-hydroxyvitamin D3 1α- hydroxylase gene expression in macrophages. Kidney Int 2000;58:559-68

Morrison NA, Qi JC, Tokita A et al. Prediction of bone density from vitamin D receptor alleles. Nature 1994;367:284-7

Motokura T, Bloom T, Kim HG et al. A novel cyclin encoded by a bcl1-linked candidate oncogene. Nature 1991;350:512-5

Mundy GR, Guise TA. Hormonal control of calcium homeostasis. Clin Chem 1999;45:1347-52

Nagaba Y, Heishi M, Tazawa H et al. Vitamin D receptor gene polymorphisms affect secondary hyperparathyroidism in hemodialyzed patients. Am J Kidney Dis 1998;32:464- 9

Nagy A, Chudek J, Kovacs G. Accumulation of allelic changes at 7p, 18q and 2 in parathyroid lesions of uremic patients. Lab Invest 2001;81:527-33

Nygren P, Larsson R, Johansson H et al. 1,25(OH) 2D3 inhibits hormone secretion and proliferation but not functional dedifferentiation of cultured bovine parathyroid cells. Calcif Tissue Int 1988;43:213-8

47

Nykjaer A, Dragun D, Walther D et al. An endocytic pathway essential for renal uptake and activation of the steroid 25-(OH) vitamin D3. Cell 1999;96:507-15

Ogunkolade BW, Boucher BJ, Fairclough PD et al. Expression of 25-hydroxyvitamin D-1- α-hydroxylase mRNA in individuals with colorectal cancer. Lancet 2002;359:1831-2

Ohkura N, Kishi M, Tsukada T et al. Menin, a gene product responsible for multiple endocrine neoplasia type 1, interacts with the putative tumor metastasis suppressor nm23. Biochem Biophys Res Comm 2001;282:1206-10

Ohyama Y, Ozono K, Uchida M et al. Identification of a vitamin D-responsive element in the 5’ flanking region of the rat 25-hydroxyvitamin D3 24-hydroxylas gene. J Biol Chem 1994;269:10545-50

Overbergh L, Decallonne B, Valckx D et al. Identification and immune regulation of 25- hydroxyvitamin D-1-α-hydroxylase in murine macrophages. Clin Exp Immunol 2000;120:139-46

Padberg BC, Schröder S, Jochum W et al. Absence of RET proto-oncogene point mutations in sporadic hyperplastic and neoplastic lesions of the parathyroid gland. Am J Pathol 1995;147:1600-7

Palanisamy N, Imanishi Y, Rao PH et al. Novel chromosomal abnormalities identified by comparative genomic hybridization in parathyroid adenomas. J Clin Endocrinol Metab 1998;83:1766-70

Pollak MR, Brown EM, Wu Chou YH et al. Mutations in the human Ca2+-sensing receptor gene cause familial hypocalcuric hypercalcemia and neonatal severe hyperparathyroidism. Cell 1993;75:1297-303

Rastad J, Lundgren E, Ljunghall S. Clinical presentation of primary hyperparathyroidism. In: The parathyroids, Bilezikian JP, Marcus R, Levine MA (eds). New York, Raven Press, 2001;361-74

Reiss AB, Martin KO, Rojer DE et al. Sterol 27-hydroxylase: expression in human arterial endothelium. J Lipid Res 1997;38:1254-60

Renan MJ. How many mutations are required for tumorigenesis? Implications from human cancer data. Mol Carcinog 1993;7:139-46

Rudberg C, Grimelius L, Johansson H et al. Alterations in density, morphology and parathyroid hormone release of dispersed parathyroid cells from patients with hyperparathyroidism. Acta Path Microbiol Immunol Scand Sect A 1986;94:253-61

Schwartz GG, Whitlatch LW, Chen TC et al. Human prostate cells synthesize 1,25- dihydroxyvitamin D3 from 25-hydroxyvitamin D3. Cancer Epidemiol Biomarkers Prev 1998;7:391-5

Schwarz P, Sorensen H, McNair P et al. Cica-clamp technique: a method for quantifying parathyroid hormone secretion: a sequential citrate and calcium clamp study. Eur J Clin Invest 1993;23:546-53

Shane E. Parathyroid carcinoma. J Clin Endocrinol Metab 2001;86,485-93

Sherr CJ, Roberts JM. Inhibitors of mammalian G1 cyclin-dependent kinases. Genes Dev 1995;9:1149-63 48

Silver J, Yalcindag C, Sela-Brown A et al. Regulation of the parathyroid hormone gene by vitamin D, calcium and phosphate. Kidney Int 1999;73 Suppl:S2-7

Simmons JD, Mullighan C, Welsh K et al. Vitamin D receptor gene polymorphism: association with Crohn’s disease susceptibility. Gut 2000;2:211-4

Socialstyrelsen State of the art: Primär hyperparathyroidism www.sos.se/2001-123-3

Sosa M, Torres A, Martin N et al. The distribution of two different vitamin D receptor polymorphisms (BsmI and start codon) in primary hyperparathyroidism. J Intern Med 2000;247:124-30

Speer G, Dworak O, Cseh K et al. Vitamin D receptor gene BsmI polymorphism correlates with erb-2/HER-2 expression in human rectal cancer. Oncology 2000;58:242-7

Sudhaker Rao D, Han Z-H, Philips ER et al. Reduced vitamin D receptor expression in parathyroid adenomas: implications for pathogenesis. Clin Endocrinol 2000;53:373-81

Szabo A, Merke J, Beier E et al. 1,25(OH) 2D3 inhibits parathyroid cell proliferation in experimental uremia. Kidney Int 1989;35:1049-56

Tahara H, Smith AP, Gaz RD et al. Genomic localization of novel candidate tumor suppressor gene loci in human parathyroid adenomas. Cancer Res 1996a;56:599-605

Tahara H, Smith AP, Gaz RD et al. Loss of chromosome 9p DNA and analysis of the p16 and p15 cyclin-dependent kinase inhibitor genes in human parathyroid adenomas. J Clin Endocrinol Metab 1996b;81:3663-7

Tahara H, Smith AP, Gaz RD et al. Parathyroid tumor suppressor on 1p: analysis of the p18 cyclin-dependant kinase inhibitor gene as a candidate. J Bone Miner Res 1997;12:1330-4

Tahara H, Imanishi Y, Yamada T et al. Rare somatic inactivation of the multiple endocrine neoplasia type 1 gene in secondary hyperparathyroidism of uremia. J Clin Endocrinol 2000;85:4113-7

Tangpricha V, Flanagan JN, Whitlach LW et al. 25-hydroxyvitamin D-1α-hydroxylase in normal and malignant colon tissue. Lancet 2001;357:1673-4

Tisell LE, Carlsson S, Fjälling M et al. Hyperparathyroidism subsequent to neck irradiation. Risk factors. Cancer 1985;56:1529-33

Tominaga Y. Mechanism of parathyroid tumorigenesis in uraemia. Nephrol Dial Transplant 1999a;14 Suppl 1:63-5

Tominaga Y, Tsuzuki T, Uchida K et al. Expression of PRAD1/cyclin D1, retinoblastoma gene products, and Ki67 in parathyroid hyperplasia caused by chronic renal failure versus primary adenomas. Kidney Int 1999b;55:1375-83

Uitterlinden AG, Burger H, Huang Q et al. Vitamin D receptor genotype is associated with radiographic osteoarthritis at the knee. J Clin Invest 1997;100:259-63

Vasef MA, Brynes RK, Sturm M et al. Expression of cyclin D1 in parathyroid carcinomas, adenomas, and hyperplasias: a paraffin immunohistochemical study. Mod Pathol 1999;12:412-6 49

Zehnder D, Bland R, Walker EA et al. Expression of 25-hydroxyvitamin D3-1α- hydroxylase in the human kidney. J Am Soc Nephrol 1999;10:2465-73

Zehnder D, Bland R, Williams MC et al. Extrarenal expression of 25-hydroxyvitamin D3- 1α-hydroxylase. J Clin Endocrinol Metab 2001;86:888-94

Zhao XY, Feldman D. The role of vitamin D in prostate cancer. Steroids 2001;66:293-300

Zierold C, Mings JA, DeLuca HF. Parathyroid hormone regulates 25-hydroxyvitamin D3- 24-hydroxylase mRNA by altering its stability. Proc Natl Acad Sci USA 2001;98:13572-6

50