- Email: [email protected]
tertiary hyperparathyroidism
http://www.kidney-international.org
original article
& 2008 International Society of Nephrology
Mutational analysis of CDKN1B, a candidate tumor-suppressor gene, in refractory secondary/tertiary hyperparathyroidism KB Lauter1 and A Arnold1 1
Center for Molecular Medicine, Division of Endocrinology and Metabolism, University of Connecticut School of Medicine, Farmington, Connecticut, USA
Most patients with refractory secondary/tertiary hyperparathyroidism have monoclonal parathyroid tumors. Inactivating mutations of CDKN1B, encoding the p27 cyclin-dependent kinase inhibitor, were reported to cause hyperparathyroidism in a multiple endocrine neoplasia type 1-like syndrome. Further, there was decreased expression of CDKN1B in parathyroid tumors of patients with chronic kidney disease. We sequenced the entire coding region and splice sites of CDKN1B in 50 parathyroid tumors from 35 patients to see if inactivating mutations could cause monoclonal tumorigenesis in refractory secondary/tertiary hyperparathyroidism. No frameshift, nonsense, or other clearly inactivating mutations were found, nor was there evidence of homozygous deletion or loss of heterozygosity. The absence of clonal inactivating mutations suggests that CDKN1B is not a classical tumor-suppressor gene in secondary/tertiary parathyroid tumors. Kidney International (2008) 73, 1137–1140; doi:10.1038/ki.2008.28; published online 20 February 2008 KEYWORDS: p27; p27Kip1; cyclin-dependent kinase inhibitor; parathyroid neoplasia; parathyroid hormone; secondary/tertiary hyperparathyroidism
Correspondence: A Arnold, Center for Molecular Medicine, University of Connecticut Health Center, 263 Farmington Avenue, Farmington, Connecticut 06030-3101, USA. E-mail: [email protected] Received 15 August 2007; revised 7 November 2007; accepted 13 December 2007; published online 20 February 2008 Kidney International (2008) 73, 1137–1140
In secondary hyperparathyroidism, chronic stimulation of the parathyroid glands typically results in multi-gland hyperplasia. Some patients, usually after years of dialysis, develop refractory hyperparathyroidism in which excessive parathyroid hormone secretion no longer responds to physiologic influences or standard medical treatments.1–3 Moreover, some patients develop irreversible hypercalcemia in the absence of calcium or vitamin D supplementation, and are said to have tertiary hyperparathyroidism. Medically refractory secondary or tertiary hyperparathyroidism, which require parathyroidectomy, are thus very different entities from typical secondary hyperparathyroidism. Importantly, development of monoclonal parathyroid expansions is common in the majority of patients undergoing parathyroidectomy for medically refractory secondary or tertiary hyperparathyroidism,1,4 and such monoclonal neoplasia may be a key determinant of the medically refractory state. However, little is known about the specific acquired genetic abnormalities that are responsible for the selective advantage underlying these clonal expansions. An excellent candidate gene for which acquired mutation could drive monoclonal hyperparathyroidism in chronic kidney disease (CKD) patients is CDKN1B, which encodes the p27 (or p27Kip1) cyclin-dependent kinase inhibitor. p27 serves to repress the function of cyclin-dependent kinases that are necessary for cellular progression through G1 and into the S phase of the cell cycle.5,6 CDKN1B contains three exons, which form a 2.4-kb transcript that is translated to a 198-amino-acid protein. p27 protein expression has been shown to be decreased in various tumor types, including colon cancer.7 Recently, mutation in CDKN1B was found to cause hyperparathyroidism in the context of MENX, a novel multiple endocrine neoplasia syndrome observed in rats.8,9 Furthermore, CDKN1B mutation was linked to parathyroid disease in humans, specifically in a family with a proband who presented with an MEN1-like phenotype of pituitary adenoma and primary hyperparathyroidism, with negative testing for MEN1 gene mutation.8 It is plausible that the effects of p27 deficiency could be similar to MEN1 inactivation, since the MEN1 gene product menin has been reported to complex with histone H3 lysine 4, and act as a 1137
original article
methylation complex necessary for proper p27 activity.10 Additionally, parathyroid glands from patients with secondary hyperplasia show less CDKN1B expression than normal controls;11 moreover, those glands showing nodular hyperplasia displayed significantly less p27 staining than those with diffuse secondary hyperplasia.11–13 The involvement of p27 in endocrine neoplasia, specifically in parathyroid disease, and its decreased expression in renal failure-associated parathyroid tumors, raises the hypothesis that CDKN1B may be subject to acquired somatic mutation with loss of function, thus serving as a tumor-suppressor gene in monoclonal secondary or tertiary hyperparathyroidism. While the term ‘tumor suppressor’ is often loosely used, the fundamental and most rigorous definition is that of ‘genes that sustain loss-of-function mutations’ in the development of tumors.14 This crucial importance of demonstrating mutations, as opposed to expression abnormalities, derives from their clonality, which serves as functional evidence that the mutation conferred a selective advantage in an original tumor progenitor cell. Thus, while genes can be under-expressed in tumors for many reasons, for example, as a result of rapid cellular growth or downstream of true drivers of tumorigenesis, only recurrent inactivating mutations provide the crucial evidence for a gene’s primary driving role in tumor development and thereby define it as a tumor suppressor. In this study, we sought such mutational evidence for CDKN1B. RESULTS
We analyzed DNA from the parathyroid tumors of patients with severe secondary or tertiary hyperparathyroidism for intragenic mutations within the coding region of CDKN1B, because clonal inactivating mutations provide the most crucial and definitive evidence that a gene serves as a tumor suppressor. No point mutations, insertions, or deletions that would be expected to inactivate the p27 gene product were identified. Furthermore, the two coding exons of CDKN1B were readily amplified by polymerase chain reaction (PCR) in all samples, indicating that homozygous deletion of the gene had not occurred. Heterozygosity for the common V109G polymorphism was present in 7 of the 35 patients, which is approximately in agreement with the predicted heterozygote frequency of 0.3 (http://www.hapmap.org), providing evidence that loss of heterozygosity was not a major feature at this locus. This observation is reinforced by the absence of copy number alterations found on the 12p chromosomal arm (the location of CDKN1B) in previous studies using microsatellite analysis and comparative genomic hybridization in 33 of these 50 tumors.4 CDKN1B contains many known polymorphisms, three of which were present in our set of samples15–18 (http:// genome.ucsc.edu). The V109G polymorphism, noted above, was present in tumors from 10 patients, with homozygosity for the variant present in samples from three patients, consistent with its reported high prevalence in the population.15–17 Next, one patient’s samples had a T-to-C transition 1138
KB Lauter and A Arnold: CDKN1B in secondary/tertiary hyperparathyroidism
at codon 119 in exon 1, changing the encoded amino acid from Ile to Thr. This patient had two glands analyzed and the change was found on one allele in each gland, with the second allele retaining the wild-type sequence. Matched normal control DNA from this individual was unavailable for sequencing, but the presence of precisely the same change in two independent tumors strongly suggests that it was a germline alteration in agreement with its reported prevalence of 0.6% as a rare polymorphism in a population with hereditary prostate cancer.18 The final CDKN1B alteration uncovered was a G-to-A change at the third position of codon 142, conserving the encoded amino acid Thr. This germline heterozygous change was found in one patient’s tumor and leukocyte DNA. Interestingly, the identical germline change was found in 1 of 34 patients with familial hyperparathyroidism and an MEN1like phenotype without MEN1 mutation,19 but has otherwise only been reported in 1 of 36 breast carcinomas.17 This alteration was not found in numerous other studies, which collectively sequenced CDKN1B in over 450 samples,15,16,18,20 raising the possibility that it might be a rare hyperparathyroid-predisposition allele enriched in the disease population. Aside from these alterations, no other variations from the published normal CDKN1B sequence were present in any of the samples studied. DISCUSSION
CDKN1B, the gene encoding the cyclin-dependent kinase inhibitor p27Kip1, has recently been identified as a tumorsuppressor gene whose mutational inactivation caused hyperparathyroidism in the context of a multiple endocrine neoplasia disorder.8 This finding, in conjunction with decreased p27 expression in secondary parathyroid tumors has established CDKN1B as an important candidate gene, which could contribute to parathyroid tumorigenesis through inactivating somatic mutation.11–13 We have sequenced the entire coding region of CDKN1B, plus splice sites, and found no clonal somatic mutations. Since inactivating mutations are the central and crucial element in defining classical tumor-suppressor genes, we have used direct sequencing to address the central criterion that would allow CDKN1B to be classified as a tumor suppressor gene in these parathyroid tumors. Our sequencing approach would not have detected mutations in introns, the promoter region, or the untranslated regions, but on the basis of the knowledge of other classical tumor suppressors, it would be unprecedented if inactivating mutations were to occur solely in these locations to the complete exclusion of coding mutations. Thus, the absence of detectable somatic mutations in our study makes it exceedingly unlikely that acquired CDKN1Binactivating mutation is a primary contributor to monoclonal growth, or to decreased p27 protein expression, in severe secondary and tertiary parathyroid tumors. The three alterations we observed in CDKN1B are previously reported germline polymorphisms.15–18 We found the rare T142T polymorphism in one patient in the study Kidney International (2008) 73, 1137–1140
original article
KB Lauter and A Arnold: CDKN1B in secondary/tertiary hyperparathyroidism
1AF
1B 1CF
2F
Exon 1
Exon 2
1AR 436 bp
1AF 1AF
1BR
203 bp 1BF
1BR
2R
2F
311 bp
2R
1AR
336 bp 1CF
1CR
Exon 3
1BR 444 bp
1CR
Figure 1 | CDKN1B amplification and sequencing strategy. (a) Schematic diagram of the PCR amplification of the coding regions and splice sites of CDKN1B. The untranslated region is represented by hashed markings. (b) Fragments produced are depicted with their primers and corresponding lengths. Annealing temperatures are as follows for each primer pair: 1AF–1BR, 61 1C; 1AF–1AR, 551C; 1BF–1BR, 55 1C; 1CF–1CR, 61 1C; and 2F–2R, 55 1C.
population, and it was also found recently in a patient with multiple parathyroid tumors and familial disease, which resembled MEN1.19 While it is unlikely that this alteration would lead to a decrease in cellular p27 protein, the possibility that it could affect gene expression and predispose to hyperparathyroidism cannot be excluded. While our observations strongly suggest that CDKN1B inactivation is not an important primary driver of clonal parathyroid growth in CKD patients, one cannot rule out the possibility that the decreased p27 staining previously found in many parathyroid tumors might participate in parathyroid tumorigenesis in a secondary manner. Additionally, 1,25dihydroxyvitamin D3 (1,25(OH2)D3) has been shown to decrease the activity of cyclin E–CDK2 proteins, as well as decrease amounts of Skp2, which act to target p27 for ubiquitination and degradation.21 Therefore, increased 1,25(OH2)D3 action on the cell could augment p27 levels and suppress cellular growth. Conversely, the low levels of 1,25(OH2)D3 characteristic of severe CKD could lead to increased p27 degradation and lower p27 levels. An association between decreased p27 levels and decreased vitamin D receptor levels has also been illustrated in nodular parathyroid tumors from uremic patients.12,13 The mechanism underlying decreased vitamin D receptor quantity, which may contribute to decreased p27 levels, has not yet been elucidated; mutational analysis of vitamin D receptor in secondary parathyroid tumors has revealed no mutations.22 Thus, further studies are needed to clarify the mechanisms by which clonal growth occurs in severe secondary and tertiary hyperparathyroidism. MATERIALS AND METHODS Patients and samples We studied 50 resected parathyroid tumors from 35 patients with CKD and refractory secondary or tertiary hyperparathyroidism. The mean age of these patients (data available for 33/35 patients) was 44 years, (range 16–68). All patients were undergoing treatment with maintenance hemodialysis. Parathyroidectomy was indicated because of severe secondary hyperparathyroidism associated with pruritus, osteitis fibrosa, soft tissue calcification, hyperphosphatemia, Kidney International (2008) 73, 1137–1140
Table 1 | Primers used for both PCR amplification and sequencing of the coding regions and splice sites of CDKN1B Primer name
Sequence
1AF 1AR 1BF 1BR 1CF 1CR 2F 2R
50 -GGCGCTTTGTTTTGTTCGGTT-30 50 -GTCCCGGGTTAACTCTTCGTG-30 50 -CCCTAGCCTGGAGCGGATGG-30 50 -GGAGCCCCAATTAAAGG-30 50 -ATTTTCAGAATCACAAACCC-30 50 -AAGCAGTGGGCCAGGTAGCAC-30 50 -GTTTTTCATCCCCTGACTAT-30 50 -AATTTGCCAGCAACCAGTAAG-30
PCR, polymerase chain reaction.
and/or other symptoms and signs, which were resistant to medical treatment. It is crucial to emphasize that the patients in our study with ‘secondary’ hyperparathyroidism were not representative of common secondary hyperparathyroidism, which would be expected to exhibit a generalized polyclonal hyperplasia of the parathyroids; instead, the fact that our patients required parathyroidectomy puts them in the special category of hyperparathyroidism refractory to medical therapies. In no case was CKD a consequence of primary hyperparathyroidism. The mean serum calcium level was 10.6 mg per 100 ml (data available for 31/35 patients) and serum parathyroid hormone levels were markedly elevated (average 17-fold above the upper limit of normal) (data available for 30/35 patients). No patient had a family history of hyperparathyroidism or multiple endocrine neoplasia, or a history of head and neck irradiation. In all patients, multiple hypercellular parathyroid tumors were identified and resected. The number of glands excised by the surgeon for each patient was not always known, but the surgeons’ usual practice was to perform a X3.5 gland resection; in general, when portions of only one or two of the resected glands were made available to us for study, these tended to be from the largest gland(s). In other words, this represented selection by the pathologist rather than a limited resection by the surgeon. Importantly, glands obtained in exactly the same manner as these have been frequently demonstrated to be monoclonal, including fully 19 of the glands in this study which were part of a previous clonal analysis by X-chromosome inactivation.1 One parathyroid gland was available for study from 24 patients; two glands were available from each of eight patients; three glands were available from one patient; and four glands were available from each of two patients. Peripheral blood leukocytes, 1139
original article
providing matched control non-tumor DNA, were available from 25 patients. All tissue and blood specimens were obtained in accordance with the University of Connecticut institutional review board-approved protocol no. 00-089. Surgical samples were grossly dissected and quickly frozen in liquid nitrogen. Parathyroid specimens were frozen in liquid nitrogen after surgical removal and stored at 80 1C. Highmolecular-weight genomic DNA was extracted from frozen tissue using standard procedures, including proteinase K digestion, phenol–chloroform extraction, and ethanol precipitation. Leukocyte DNA was extracted using standard procedures.23 CDKN1B sequencing The entire coding region, consisting of two exons and splice sites of the CDKN1B gene, was amplified by PCR from the DNA of 50 resected tumors (Figure 1) and, as described below, in one patient’s matched leukocyte DNA sample to evaluate an apparent polymorphism. Exon 1 was amplified as four overlapping fragments, with primers listed in Table 1. The following primers were paired together: 1AF and 1BR; 1CF and 1CR. Some samples required that the first fragment be split further into two fragments to achieve adequate PCR amplification. Primer 1AF was paired with 1AR and 1BF was paired with 1BR. CDKN1B exon 2 was amplified to form one single PCR product with primers 2F and 2R. PCR reactions were performed in 20-ml reaction volumes containing 25 ng genomic DNA, 20 pmol of each primer, 200 mM of each dNTP, 1.25 U of Amplitaq Gold DNA Polymerase (Applied Biosystems, Foster City, CA, USA), and 2 mM magnesium chloride. Thermocycling consisted of an initial denaturation step of 95 1C for 10 min; 35 cycles of 95 1C for 30 s, the optimized Ta (Figure 1) for 30 s, 72 1C for 30 s; and a final extension step at 72 1C for 10 min. PCR products were purified using EXOSAP 10 U exonuclease I and 1 U Shrimp alkaline phosphatase (Amersham Pharmacia Biotech, Piscataway, NJ, USA). Purified PCR fragments were then sequenced in both forward and reverse directions using the Dye Terminator Cycle Sequencing Quick Start kit (Beckman-Coulter, Fullerton, CA, USA) with the same primers used for PCR, under conditions recommended by the manufacturer. The resulting sequence fragments were purified through Sephadex G-50 columns (Sigma Aldrich, St Louis, MO, USA) and electrophoresed on a CEQ 8800 Genetic Analysis System (Beckman-Coulter). Some samples, particularly those of exon 1, were confirmed by sequencing performed by Genewiz Inc (North Brunswick, NJ, USA) using Applied Biosystems Big Dye version 3.1. The reactions were then run on the Applied Biosystems 3730xl DNA Analyzer. Resulting sequence data were analyzed and compared with the published sequence (RefSeq ID: NM_000315) using Sequencher software (GeneCodes Corporation, Ann Arbor, MI, USA).
KB Lauter and A Arnold: CDKN1B in secondary/tertiary hyperparathyroidism
REFERENCES 1.
2.
3.
4.
5. 6.
7.
8.
9. 10.
11.
12.
13.
14. 15.
16.
17. 18.
19.
20.
21.
DISCLOSURE
The authors have no financial interests or relationships to disclose. ACKNOWLEDGMENTS
We thank Kristin Corrado for expert technical assistance. This work was supported in part by the Medical Scientist Training Program T32-GM008607 from the National Institutes of Health, by the Murray-Heilig Fund in Molecular Medicine, and by the Student Scholar Grant from the American Society of Nephrology.
1140
22.
23.
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: 2047–2053. Drueke TB. Hyperparathyroidism in chronic kidney disease. In: www.ENDOTEXT.org In: De Groot L (ed). MDTEXT.com Inc: S. Dartmouth, MA, 2007. Chudek J, Nagy A, Kokot F et al. Phosphatemia is related to chromosomal aberrations of parathyroid glands in patients with hyperparathyroidism. J Nephrol 2007; 20: 164–172. Imanishi Y, Tahara H, Palanisamy N et al. Clonal chromosomal defects in the molecular pathogenesis of refractory hyperparathyroidism of uremia. J Am Soc Nephrol 2002; 13: 1490–1498. Clurman BE, Porter P. New insights into the tumor suppression function of P27(kip1). Proc Natl Acad Sci USA 1998; 95: 15158–15160. Sgambato A, Cittadini A, Faraglia B et al. Multiple functions of p27(Kip1) and its alterations in tumor cells: a review. J Cell Physiol 2000; 183: 18–27. Ogino S, Kawasaki T, Kirkner GJ et al. Loss of nuclear p27 (CDKN1B/KIP1) in colorectal cancer is correlated with microsatellite instability and CIMP. Mod Pathol 2007; 20: 15–22. Pellegata NS, Quintanilla-Martinez L, Siggelkow H et al. Germ-line mutations in p27Kip1 cause a multiple endocrine neoplasia syndrome in rats and humans. Proc Natl Acad Sci USA 2006; 103: 15558–15563. Polyak K. The p27Kip1 tumor suppressor gene: still a suspect or proven guilty? Cancer Cell 2006; 10: 352–354. Hughes CM, Rozenblatt-Rosen O, Milne TA et al. Menin associates with a trithorax family histone methyltransferase complex and with the hoxc8 locus. Mol Cell 2004; 13: 587–597. Buchwald PC, Akerstrom G, Westin G. Reduced p18INK4c, p21CIP1/WAF1 and p27KIP1 mRNA levels in tumours of primary and secondary hyperparathyroidism. Clin Endocrinol (Oxford) 2004; 60: 389–393. Tokumoto M, Tsuruya K, Fukuda K et al. Reduced p21, p27 and vitamin D receptor in the nodular hyperplasia in patients with advanced secondary hyperparathyroidism. Kidney Int 2002; 62: 1196–1207. Tokumoto M, Tsuruya K, Fukuda K et al. Parathyroid cell growth in patients with advanced secondary hyperparathyroidism: vitamin D receptor and cyclin-dependent kinase inhibitors, p21 and p27. Nephrol Dial Transplant 2003; 18(Suppl 3): iii9–iii12. Haber D, Harlow E. Tumour-suppressor genes: evolving definitions in the genomic age. Nat Genet 1997; 16: 320–322. Morosetti R, Kawamata N, Gombart AF et al. Alterations of the p27KIP1 gene in non-Hodgkin’s lymphomas and adult T-cell leukemia/lymphoma. Blood 1995; 86: 1924–1930. Kawamata N, Morosetti R, Miller CW et al. Molecular analysis of the cyclindependent kinase inhibitor gene p27/Kip1 in human malignancies. Cancer Res 1995; 55: 2266–2269. Spirin KS, Simpson JF, Takeuchi S et al. p27/Kip1 mutation found in breast cancer. Cancer Res 1996; 56: 2400–2404. Chang BL, Zheng SL, Isaacs SD et al. A polymorphism in the CDKN1B gene is associated with increased risk of hereditary prostate cancer. Cancer Res 2004; 64: 1997–1999. Ozawa A, Agarwal SK, Mateo CM et al. The parathyroid/pituitary variant of MEN1 usually has causes other than p27Kip1 mutations. J Clin Endocrinol Metab 2007; 92: 1948–1951. Gonzalez P, Diez-Juan A, Coto E et al. A single-nucleotide polymorphism in the human p27kip1 gene (838C4A) affects basal promoter activity and the risk of myocardial infarction. BMC Biol 2004; 2: 5. Li P, Li C, Zhao X et al. p27(Kip1) stabilization and G(1) arrest by 1,25-dihydroxyvitamin D(3) in ovarian cancer cells mediated through downregulation of cyclin E/cyclin-dependent kinase 2 and Skp1–Cullin–F-box protein/Skp2 ubiquitin ligase. J Biol Chem 2004; 279: 25260–25267. 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: 868–872. Sambrook JFE, Maniatis T. Molecular Cloning: a Laboratory Manual, 2nd edn. Cold Spring Harbor Laboratory: Cold Spring Harbor, NY, 1989.
Kidney International (2008) 73, 1137–1140
original article
& 2008 International Society of Nephrology
Mutational analysis of CDKN1B, a candidate tumor-suppressor gene, in refractory secondary/tertiary hyperparathyroidism KB Lauter1 and A Arnold1 1
Center for Molecular Medicine, Division of Endocrinology and Metabolism, University of Connecticut School of Medicine, Farmington, Connecticut, USA
Most patients with refractory secondary/tertiary hyperparathyroidism have monoclonal parathyroid tumors. Inactivating mutations of CDKN1B, encoding the p27 cyclin-dependent kinase inhibitor, were reported to cause hyperparathyroidism in a multiple endocrine neoplasia type 1-like syndrome. Further, there was decreased expression of CDKN1B in parathyroid tumors of patients with chronic kidney disease. We sequenced the entire coding region and splice sites of CDKN1B in 50 parathyroid tumors from 35 patients to see if inactivating mutations could cause monoclonal tumorigenesis in refractory secondary/tertiary hyperparathyroidism. No frameshift, nonsense, or other clearly inactivating mutations were found, nor was there evidence of homozygous deletion or loss of heterozygosity. The absence of clonal inactivating mutations suggests that CDKN1B is not a classical tumor-suppressor gene in secondary/tertiary parathyroid tumors. Kidney International (2008) 73, 1137–1140; doi:10.1038/ki.2008.28; published online 20 February 2008 KEYWORDS: p27; p27Kip1; cyclin-dependent kinase inhibitor; parathyroid neoplasia; parathyroid hormone; secondary/tertiary hyperparathyroidism
Correspondence: A Arnold, Center for Molecular Medicine, University of Connecticut Health Center, 263 Farmington Avenue, Farmington, Connecticut 06030-3101, USA. E-mail: [email protected] Received 15 August 2007; revised 7 November 2007; accepted 13 December 2007; published online 20 February 2008 Kidney International (2008) 73, 1137–1140
In secondary hyperparathyroidism, chronic stimulation of the parathyroid glands typically results in multi-gland hyperplasia. Some patients, usually after years of dialysis, develop refractory hyperparathyroidism in which excessive parathyroid hormone secretion no longer responds to physiologic influences or standard medical treatments.1–3 Moreover, some patients develop irreversible hypercalcemia in the absence of calcium or vitamin D supplementation, and are said to have tertiary hyperparathyroidism. Medically refractory secondary or tertiary hyperparathyroidism, which require parathyroidectomy, are thus very different entities from typical secondary hyperparathyroidism. Importantly, development of monoclonal parathyroid expansions is common in the majority of patients undergoing parathyroidectomy for medically refractory secondary or tertiary hyperparathyroidism,1,4 and such monoclonal neoplasia may be a key determinant of the medically refractory state. However, little is known about the specific acquired genetic abnormalities that are responsible for the selective advantage underlying these clonal expansions. An excellent candidate gene for which acquired mutation could drive monoclonal hyperparathyroidism in chronic kidney disease (CKD) patients is CDKN1B, which encodes the p27 (or p27Kip1) cyclin-dependent kinase inhibitor. p27 serves to repress the function of cyclin-dependent kinases that are necessary for cellular progression through G1 and into the S phase of the cell cycle.5,6 CDKN1B contains three exons, which form a 2.4-kb transcript that is translated to a 198-amino-acid protein. p27 protein expression has been shown to be decreased in various tumor types, including colon cancer.7 Recently, mutation in CDKN1B was found to cause hyperparathyroidism in the context of MENX, a novel multiple endocrine neoplasia syndrome observed in rats.8,9 Furthermore, CDKN1B mutation was linked to parathyroid disease in humans, specifically in a family with a proband who presented with an MEN1-like phenotype of pituitary adenoma and primary hyperparathyroidism, with negative testing for MEN1 gene mutation.8 It is plausible that the effects of p27 deficiency could be similar to MEN1 inactivation, since the MEN1 gene product menin has been reported to complex with histone H3 lysine 4, and act as a 1137
original article
methylation complex necessary for proper p27 activity.10 Additionally, parathyroid glands from patients with secondary hyperplasia show less CDKN1B expression than normal controls;11 moreover, those glands showing nodular hyperplasia displayed significantly less p27 staining than those with diffuse secondary hyperplasia.11–13 The involvement of p27 in endocrine neoplasia, specifically in parathyroid disease, and its decreased expression in renal failure-associated parathyroid tumors, raises the hypothesis that CDKN1B may be subject to acquired somatic mutation with loss of function, thus serving as a tumor-suppressor gene in monoclonal secondary or tertiary hyperparathyroidism. While the term ‘tumor suppressor’ is often loosely used, the fundamental and most rigorous definition is that of ‘genes that sustain loss-of-function mutations’ in the development of tumors.14 This crucial importance of demonstrating mutations, as opposed to expression abnormalities, derives from their clonality, which serves as functional evidence that the mutation conferred a selective advantage in an original tumor progenitor cell. Thus, while genes can be under-expressed in tumors for many reasons, for example, as a result of rapid cellular growth or downstream of true drivers of tumorigenesis, only recurrent inactivating mutations provide the crucial evidence for a gene’s primary driving role in tumor development and thereby define it as a tumor suppressor. In this study, we sought such mutational evidence for CDKN1B. RESULTS
We analyzed DNA from the parathyroid tumors of patients with severe secondary or tertiary hyperparathyroidism for intragenic mutations within the coding region of CDKN1B, because clonal inactivating mutations provide the most crucial and definitive evidence that a gene serves as a tumor suppressor. No point mutations, insertions, or deletions that would be expected to inactivate the p27 gene product were identified. Furthermore, the two coding exons of CDKN1B were readily amplified by polymerase chain reaction (PCR) in all samples, indicating that homozygous deletion of the gene had not occurred. Heterozygosity for the common V109G polymorphism was present in 7 of the 35 patients, which is approximately in agreement with the predicted heterozygote frequency of 0.3 (http://www.hapmap.org), providing evidence that loss of heterozygosity was not a major feature at this locus. This observation is reinforced by the absence of copy number alterations found on the 12p chromosomal arm (the location of CDKN1B) in previous studies using microsatellite analysis and comparative genomic hybridization in 33 of these 50 tumors.4 CDKN1B contains many known polymorphisms, three of which were present in our set of samples15–18 (http:// genome.ucsc.edu). The V109G polymorphism, noted above, was present in tumors from 10 patients, with homozygosity for the variant present in samples from three patients, consistent with its reported high prevalence in the population.15–17 Next, one patient’s samples had a T-to-C transition 1138
KB Lauter and A Arnold: CDKN1B in secondary/tertiary hyperparathyroidism
at codon 119 in exon 1, changing the encoded amino acid from Ile to Thr. This patient had two glands analyzed and the change was found on one allele in each gland, with the second allele retaining the wild-type sequence. Matched normal control DNA from this individual was unavailable for sequencing, but the presence of precisely the same change in two independent tumors strongly suggests that it was a germline alteration in agreement with its reported prevalence of 0.6% as a rare polymorphism in a population with hereditary prostate cancer.18 The final CDKN1B alteration uncovered was a G-to-A change at the third position of codon 142, conserving the encoded amino acid Thr. This germline heterozygous change was found in one patient’s tumor and leukocyte DNA. Interestingly, the identical germline change was found in 1 of 34 patients with familial hyperparathyroidism and an MEN1like phenotype without MEN1 mutation,19 but has otherwise only been reported in 1 of 36 breast carcinomas.17 This alteration was not found in numerous other studies, which collectively sequenced CDKN1B in over 450 samples,15,16,18,20 raising the possibility that it might be a rare hyperparathyroid-predisposition allele enriched in the disease population. Aside from these alterations, no other variations from the published normal CDKN1B sequence were present in any of the samples studied. DISCUSSION
CDKN1B, the gene encoding the cyclin-dependent kinase inhibitor p27Kip1, has recently been identified as a tumorsuppressor gene whose mutational inactivation caused hyperparathyroidism in the context of a multiple endocrine neoplasia disorder.8 This finding, in conjunction with decreased p27 expression in secondary parathyroid tumors has established CDKN1B as an important candidate gene, which could contribute to parathyroid tumorigenesis through inactivating somatic mutation.11–13 We have sequenced the entire coding region of CDKN1B, plus splice sites, and found no clonal somatic mutations. Since inactivating mutations are the central and crucial element in defining classical tumor-suppressor genes, we have used direct sequencing to address the central criterion that would allow CDKN1B to be classified as a tumor suppressor gene in these parathyroid tumors. Our sequencing approach would not have detected mutations in introns, the promoter region, or the untranslated regions, but on the basis of the knowledge of other classical tumor suppressors, it would be unprecedented if inactivating mutations were to occur solely in these locations to the complete exclusion of coding mutations. Thus, the absence of detectable somatic mutations in our study makes it exceedingly unlikely that acquired CDKN1Binactivating mutation is a primary contributor to monoclonal growth, or to decreased p27 protein expression, in severe secondary and tertiary parathyroid tumors. The three alterations we observed in CDKN1B are previously reported germline polymorphisms.15–18 We found the rare T142T polymorphism in one patient in the study Kidney International (2008) 73, 1137–1140
original article
KB Lauter and A Arnold: CDKN1B in secondary/tertiary hyperparathyroidism
1AF
1B 1CF
2F
Exon 1
Exon 2
1AR 436 bp
1AF 1AF
1BR
203 bp 1BF
1BR
2R
2F
311 bp
2R
1AR
336 bp 1CF
1CR
Exon 3
1BR 444 bp
1CR
Figure 1 | CDKN1B amplification and sequencing strategy. (a) Schematic diagram of the PCR amplification of the coding regions and splice sites of CDKN1B. The untranslated region is represented by hashed markings. (b) Fragments produced are depicted with their primers and corresponding lengths. Annealing temperatures are as follows for each primer pair: 1AF–1BR, 61 1C; 1AF–1AR, 551C; 1BF–1BR, 55 1C; 1CF–1CR, 61 1C; and 2F–2R, 55 1C.
population, and it was also found recently in a patient with multiple parathyroid tumors and familial disease, which resembled MEN1.19 While it is unlikely that this alteration would lead to a decrease in cellular p27 protein, the possibility that it could affect gene expression and predispose to hyperparathyroidism cannot be excluded. While our observations strongly suggest that CDKN1B inactivation is not an important primary driver of clonal parathyroid growth in CKD patients, one cannot rule out the possibility that the decreased p27 staining previously found in many parathyroid tumors might participate in parathyroid tumorigenesis in a secondary manner. Additionally, 1,25dihydroxyvitamin D3 (1,25(OH2)D3) has been shown to decrease the activity of cyclin E–CDK2 proteins, as well as decrease amounts of Skp2, which act to target p27 for ubiquitination and degradation.21 Therefore, increased 1,25(OH2)D3 action on the cell could augment p27 levels and suppress cellular growth. Conversely, the low levels of 1,25(OH2)D3 characteristic of severe CKD could lead to increased p27 degradation and lower p27 levels. An association between decreased p27 levels and decreased vitamin D receptor levels has also been illustrated in nodular parathyroid tumors from uremic patients.12,13 The mechanism underlying decreased vitamin D receptor quantity, which may contribute to decreased p27 levels, has not yet been elucidated; mutational analysis of vitamin D receptor in secondary parathyroid tumors has revealed no mutations.22 Thus, further studies are needed to clarify the mechanisms by which clonal growth occurs in severe secondary and tertiary hyperparathyroidism. MATERIALS AND METHODS Patients and samples We studied 50 resected parathyroid tumors from 35 patients with CKD and refractory secondary or tertiary hyperparathyroidism. The mean age of these patients (data available for 33/35 patients) was 44 years, (range 16–68). All patients were undergoing treatment with maintenance hemodialysis. Parathyroidectomy was indicated because of severe secondary hyperparathyroidism associated with pruritus, osteitis fibrosa, soft tissue calcification, hyperphosphatemia, Kidney International (2008) 73, 1137–1140
Table 1 | Primers used for both PCR amplification and sequencing of the coding regions and splice sites of CDKN1B Primer name
Sequence
1AF 1AR 1BF 1BR 1CF 1CR 2F 2R
50 -GGCGCTTTGTTTTGTTCGGTT-30 50 -GTCCCGGGTTAACTCTTCGTG-30 50 -CCCTAGCCTGGAGCGGATGG-30 50 -GGAGCCCCAATTAAAGG-30 50 -ATTTTCAGAATCACAAACCC-30 50 -AAGCAGTGGGCCAGGTAGCAC-30 50 -GTTTTTCATCCCCTGACTAT-30 50 -AATTTGCCAGCAACCAGTAAG-30
PCR, polymerase chain reaction.
and/or other symptoms and signs, which were resistant to medical treatment. It is crucial to emphasize that the patients in our study with ‘secondary’ hyperparathyroidism were not representative of common secondary hyperparathyroidism, which would be expected to exhibit a generalized polyclonal hyperplasia of the parathyroids; instead, the fact that our patients required parathyroidectomy puts them in the special category of hyperparathyroidism refractory to medical therapies. In no case was CKD a consequence of primary hyperparathyroidism. The mean serum calcium level was 10.6 mg per 100 ml (data available for 31/35 patients) and serum parathyroid hormone levels were markedly elevated (average 17-fold above the upper limit of normal) (data available for 30/35 patients). No patient had a family history of hyperparathyroidism or multiple endocrine neoplasia, or a history of head and neck irradiation. In all patients, multiple hypercellular parathyroid tumors were identified and resected. The number of glands excised by the surgeon for each patient was not always known, but the surgeons’ usual practice was to perform a X3.5 gland resection; in general, when portions of only one or two of the resected glands were made available to us for study, these tended to be from the largest gland(s). In other words, this represented selection by the pathologist rather than a limited resection by the surgeon. Importantly, glands obtained in exactly the same manner as these have been frequently demonstrated to be monoclonal, including fully 19 of the glands in this study which were part of a previous clonal analysis by X-chromosome inactivation.1 One parathyroid gland was available for study from 24 patients; two glands were available from each of eight patients; three glands were available from one patient; and four glands were available from each of two patients. Peripheral blood leukocytes, 1139
original article
providing matched control non-tumor DNA, were available from 25 patients. All tissue and blood specimens were obtained in accordance with the University of Connecticut institutional review board-approved protocol no. 00-089. Surgical samples were grossly dissected and quickly frozen in liquid nitrogen. Parathyroid specimens were frozen in liquid nitrogen after surgical removal and stored at 80 1C. Highmolecular-weight genomic DNA was extracted from frozen tissue using standard procedures, including proteinase K digestion, phenol–chloroform extraction, and ethanol precipitation. Leukocyte DNA was extracted using standard procedures.23 CDKN1B sequencing The entire coding region, consisting of two exons and splice sites of the CDKN1B gene, was amplified by PCR from the DNA of 50 resected tumors (Figure 1) and, as described below, in one patient’s matched leukocyte DNA sample to evaluate an apparent polymorphism. Exon 1 was amplified as four overlapping fragments, with primers listed in Table 1. The following primers were paired together: 1AF and 1BR; 1CF and 1CR. Some samples required that the first fragment be split further into two fragments to achieve adequate PCR amplification. Primer 1AF was paired with 1AR and 1BF was paired with 1BR. CDKN1B exon 2 was amplified to form one single PCR product with primers 2F and 2R. PCR reactions were performed in 20-ml reaction volumes containing 25 ng genomic DNA, 20 pmol of each primer, 200 mM of each dNTP, 1.25 U of Amplitaq Gold DNA Polymerase (Applied Biosystems, Foster City, CA, USA), and 2 mM magnesium chloride. Thermocycling consisted of an initial denaturation step of 95 1C for 10 min; 35 cycles of 95 1C for 30 s, the optimized Ta (Figure 1) for 30 s, 72 1C for 30 s; and a final extension step at 72 1C for 10 min. PCR products were purified using EXOSAP 10 U exonuclease I and 1 U Shrimp alkaline phosphatase (Amersham Pharmacia Biotech, Piscataway, NJ, USA). Purified PCR fragments were then sequenced in both forward and reverse directions using the Dye Terminator Cycle Sequencing Quick Start kit (Beckman-Coulter, Fullerton, CA, USA) with the same primers used for PCR, under conditions recommended by the manufacturer. The resulting sequence fragments were purified through Sephadex G-50 columns (Sigma Aldrich, St Louis, MO, USA) and electrophoresed on a CEQ 8800 Genetic Analysis System (Beckman-Coulter). Some samples, particularly those of exon 1, were confirmed by sequencing performed by Genewiz Inc (North Brunswick, NJ, USA) using Applied Biosystems Big Dye version 3.1. The reactions were then run on the Applied Biosystems 3730xl DNA Analyzer. Resulting sequence data were analyzed and compared with the published sequence (RefSeq ID: NM_000315) using Sequencher software (GeneCodes Corporation, Ann Arbor, MI, USA).
KB Lauter and A Arnold: CDKN1B in secondary/tertiary hyperparathyroidism
REFERENCES 1.
2.
3.
4.
5. 6.
7.
8.
9. 10.
11.
12.
13.
14. 15.
16.
17. 18.
19.
20.
21.
DISCLOSURE
The authors have no financial interests or relationships to disclose. ACKNOWLEDGMENTS
We thank Kristin Corrado for expert technical assistance. This work was supported in part by the Medical Scientist Training Program T32-GM008607 from the National Institutes of Health, by the Murray-Heilig Fund in Molecular Medicine, and by the Student Scholar Grant from the American Society of Nephrology.
1140
22.
23.
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: 2047–2053. Drueke TB. Hyperparathyroidism in chronic kidney disease. In: www.ENDOTEXT.org In: De Groot L (ed). MDTEXT.com Inc: S. Dartmouth, MA, 2007. Chudek J, Nagy A, Kokot F et al. Phosphatemia is related to chromosomal aberrations of parathyroid glands in patients with hyperparathyroidism. J Nephrol 2007; 20: 164–172. Imanishi Y, Tahara H, Palanisamy N et al. Clonal chromosomal defects in the molecular pathogenesis of refractory hyperparathyroidism of uremia. J Am Soc Nephrol 2002; 13: 1490–1498. Clurman BE, Porter P. New insights into the tumor suppression function of P27(kip1). Proc Natl Acad Sci USA 1998; 95: 15158–15160. Sgambato A, Cittadini A, Faraglia B et al. Multiple functions of p27(Kip1) and its alterations in tumor cells: a review. J Cell Physiol 2000; 183: 18–27. Ogino S, Kawasaki T, Kirkner GJ et al. Loss of nuclear p27 (CDKN1B/KIP1) in colorectal cancer is correlated with microsatellite instability and CIMP. Mod Pathol 2007; 20: 15–22. Pellegata NS, Quintanilla-Martinez L, Siggelkow H et al. Germ-line mutations in p27Kip1 cause a multiple endocrine neoplasia syndrome in rats and humans. Proc Natl Acad Sci USA 2006; 103: 15558–15563. Polyak K. The p27Kip1 tumor suppressor gene: still a suspect or proven guilty? Cancer Cell 2006; 10: 352–354. Hughes CM, Rozenblatt-Rosen O, Milne TA et al. Menin associates with a trithorax family histone methyltransferase complex and with the hoxc8 locus. Mol Cell 2004; 13: 587–597. Buchwald PC, Akerstrom G, Westin G. Reduced p18INK4c, p21CIP1/WAF1 and p27KIP1 mRNA levels in tumours of primary and secondary hyperparathyroidism. Clin Endocrinol (Oxford) 2004; 60: 389–393. Tokumoto M, Tsuruya K, Fukuda K et al. Reduced p21, p27 and vitamin D receptor in the nodular hyperplasia in patients with advanced secondary hyperparathyroidism. Kidney Int 2002; 62: 1196–1207. Tokumoto M, Tsuruya K, Fukuda K et al. Parathyroid cell growth in patients with advanced secondary hyperparathyroidism: vitamin D receptor and cyclin-dependent kinase inhibitors, p21 and p27. Nephrol Dial Transplant 2003; 18(Suppl 3): iii9–iii12. Haber D, Harlow E. Tumour-suppressor genes: evolving definitions in the genomic age. Nat Genet 1997; 16: 320–322. Morosetti R, Kawamata N, Gombart AF et al. Alterations of the p27KIP1 gene in non-Hodgkin’s lymphomas and adult T-cell leukemia/lymphoma. Blood 1995; 86: 1924–1930. Kawamata N, Morosetti R, Miller CW et al. Molecular analysis of the cyclindependent kinase inhibitor gene p27/Kip1 in human malignancies. Cancer Res 1995; 55: 2266–2269. Spirin KS, Simpson JF, Takeuchi S et al. p27/Kip1 mutation found in breast cancer. Cancer Res 1996; 56: 2400–2404. Chang BL, Zheng SL, Isaacs SD et al. A polymorphism in the CDKN1B gene is associated with increased risk of hereditary prostate cancer. Cancer Res 2004; 64: 1997–1999. Ozawa A, Agarwal SK, Mateo CM et al. The parathyroid/pituitary variant of MEN1 usually has causes other than p27Kip1 mutations. J Clin Endocrinol Metab 2007; 92: 1948–1951. Gonzalez P, Diez-Juan A, Coto E et al. A single-nucleotide polymorphism in the human p27kip1 gene (838C4A) affects basal promoter activity and the risk of myocardial infarction. BMC Biol 2004; 2: 5. Li P, Li C, Zhao X et al. p27(Kip1) stabilization and G(1) arrest by 1,25-dihydroxyvitamin D(3) in ovarian cancer cells mediated through downregulation of cyclin E/cyclin-dependent kinase 2 and Skp1–Cullin–F-box protein/Skp2 ubiquitin ligase. J Biol Chem 2004; 279: 25260–25267. 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: 868–872. Sambrook JFE, Maniatis T. Molecular Cloning: a Laboratory Manual, 2nd edn. Cold Spring Harbor Laboratory: Cold Spring Harbor, NY, 1989.
Kidney International (2008) 73, 1137–1140