Medical Encyclopedia

Most, and probably all, adenomas and carcinomas are monoclonal in origin, and specific clonal genetic lesions have been identified in most of these tumors.

Only two genes have been definitively proven to be important players in the pathogenesis of typical sporadic parathyroid tumors, an oncogene, cyclin D1 and a tumor suppressor gene, MEN1.

The cyclin D1 oncoprotein is overexpressed in 20-40% of parathyroid adenomas and the identification of clonal rearrangements which activate the cyclin D1 gene in a subset of tumors indicates that such activation is a primary genetic driver of parathyroid neoplasia.

Cyclin D1 plays an important role in regulation of the cell cycle and may have non-cell cycle effects which contribute to tumorigenesis as well.

The central role that cyclin D1 plays in parathyroid tumorigenesis has been confirmed in a mouse model where cyclin D1 is overexpressed specifically in the parathyroids and in which many features of human hyperparathyroidism are reproduced.

Germline mutations of MEN1 cause multiple endocrine neoplasia type 1, a genetic syndrome in which patients develop tumors of multiple endocrine (and some nonendocrine) tissues including the parathyroid glands.

Acquired (somatic) mutations in MEN1 have also been identified in 12-17% of sporadic parathyroid adenomas.

The function of MEN1 remains elusive, but the discovery of proteins that interact with the MEN1 protein product, menin, and the development of a mouse model of MEN1 syndrome may help to shed light on menin’s function. The HRPT2 gene has recently been identified as a major contributor to the development of sporadic parathyroid carcinoma.

Identification of acquired chromosomal aberrations in parathyroid adenomas and carcinomas using techniques such as molecular allelotyping and comparative genomic hybridization has highlighted several areas of the genome that may harbor other important parathyroid tumor suppressor genes and oncogenes.

The eventual identification of the full spectrum of genes involved in parathyroid tumorigenesis will be important for developing a complete understanding of the molecular basis of primary hyperparathyroidism.

Primary hyperparathyroidism (HPT), a disorder characterized by hypercalcemia due to the excessive secretion of parathyroid hormone (PTH), affects about 1:1000 people.

In these patients, the parathyroid gland(s) increases in mass, and the control of PTH secretion from the parathyroid cell by the ambient calcium concentration (represented by the calcium set-point) is reset.

One or more enlarged parathyroid gland(s) secretes excessive quantities of PTH due to a defect within the gland itself rather than resulting from other physiological disturbances as occurs in secondary hyperparathyroidism.

In the majority of cases (85%) a single, benign adenoma is identified, while multiple hypercellular glands are found in about 15%.

Malignant parathyroid carcinoma is responsible for less than 1% of cases of primary hyperparathyroidism.

The classical symptoms of primary hyperparathyroidism result from metabolic abnormalities rather than the tumor mass per se, and include kidney stones, gastrointestinal disruption, neuropsychiatric symptoms and a prototypical bone disorder osteitis fibrosa cystica.

However, in many parts of the world it is now more common for this disorder to be diagnosed in asymptomatic or minimally symptomatic patients, with abnormal calcium levels discovered in routine blood tests or blood tests for other conditions.

Concordant with this clinical observation, the rates of parathyroid tumor cell proliferation, once hyperparathyroidism is fully established, often appear to be quite low.

In addition to the mode of patient ascertainment, a number of factors may bear upon the severity of the proliferative and biochemical defects in hyperparathyroidism, including the patient’s vitamin D status.

The molecular basis of the relationship between the proliferative defect and the PTH regulatory abnormality characteristic of such tumors is a fundamental issue in endocrine neoplasia.

Parathyroid adenomas have been differentiated from primary parathyroid hyperplasia on the basis of the number of abnormal, hypercellular parathyroid glands found in the patient.

It is, however, impossible to determine histologically if a hypercellular gland is a solitary tumor or one of several.

Multiple glands may be enlarged in a nonuniform fashion with one gland being larger than the others.

Such asymmetric hyperplasia may be confused with a solitary adenoma, or with true “double adenomas”.

To better define the pathogenic mechanisms underlying the development of adenomas and hyperplasias, the clonality of these entities was examined.

A method frequently used to study the clonality of a tumor involves assessment of the X-chromosome inactivation pattern of the cells in a tumor.

According to the Lyon phenomenon, in females, one X-chromosome is randomly inactivated in each cell early in embryonic development.

A monoclonal collection of cells will therefore all have the same X-chromosome inactivated, whereas a polyclonal group of cells should have inactivated either of the X-chromosomes in a random distribution.

Initially, X-chromosome inactivation patterns of parathyroid adenomas were examined using the glucose-6-phosphate dehydrogenase (G6PD) protein polymorphism.

These studies indicated that the adenomas were polyclonal rather than monoclonal growths, suggesting that a parathyroid adenoma is actually a form of asymmetric multigland hyperplasia and that no clonal DNA changes that are characteristic of monoclonal tumors should be found in these growths.

Subsequently, X-chromosome inactivation analysis using a DNA polymorphism-based method was used to reexamine the clonality of parathyroid adenomas.

These DNA based approaches are informative in a much higher percentage of tumors and avoid many of the pitfalls of the G6PD protein polymorphism technique.

These studies concluded that many (and probably all) parathyroid adenomas are monoclonal in origin.

The findings of clonal DNA rearrangements, mutations and areas of loss of heterozygosity, which will be described below, have overwhelmingly confirmed the monoclonality of parathyroid adenomas.

Similar discoveries have established that parathyroid carcinomas are also monoclonal lesions. The implications of monoclonality in the context of modern concepts of tumor genetics are considerable.

Clonal DNA alterations in key growth-regulating genes, generally categorized as protooncogenes and tumor suppressor genes, are key contributors in the conversion of normal cells to neoplastic cells. Protooncogenes are often involved in the physiological control of cellular growth, proliferation, or differentiation.

Protooncogenes are converted into oncogenes by DNA damage of various sorts including chromosomal translocations or inversions, point mutations, proviral insertions, or gene amplifications. These result in a deregulation of the expression of the protein product or formation of an intrinsically abnormal product.

Such an “activated” oncogene then can contribute to the development or growth of the tumor. In contrast, tumor-suppressor genes normally serve to restrain cellular proliferation or activate cell death, directly or indirectly (for example through maintenance of genomic stability).

This inactivation can occur by mutation or deletion of the gene or by regulatory derangements such as abnormal methylation, which causes loss of transcription of the gene. Inactivation of both alleles of a “classic” tumor suppressor gene is necessary to completely deplete the protein product and promote neoplastic transformation.

Cells gain a selective advantage when a sufficient number of key changes occur in protooncogenes and/or tumor suppressor genes.

Progeny of these cells grow and accumulate more genetic changes in a process known as clonal evolution, ultimately forming a clinically apparent mass of cells.

As all cells in the tumor arose from the same single parent cell that had acquired the initial rare genetic changes, they comprise a clonal population.

For certain genes, the same pattern of DNA damage can be found in each cell of a clonal tumor and the damaged regions represent and can help to define the important genetic events that led to the clonal expansion.

Subpopulations of a clone may develop due to the acquisition of additional DNA damage after the initial clone is established.

Therefore, there are some genes involved in the initial clonal expansion of the tumor and others that are important for the clonal evolution of the tumors.

Those genes that are involved in each process have not been clearly defined in most tumor types, but the importance of selection as the driving process cannot be overemphasized.

It is thought that it is necessary for several different genes, within the same cell, to be damaged for the cell to become neoplastic.

While some genes are implicated in one or a few types of tumors, other genes have been implicated in many different tumor types.

The emergence of a particular tumor type, for example parathyroid adenoma vs. carcinoma, may reflect the specific biochemical pathways that are disrupted.

However, it is unlikely that a disruption of one specific gene is both a necessary and sufficient cause of a specific tumor type. Furthermore, different combinations of mutated genes may lead to similar pathological and clinical results.

More studies are necessary to better understand how specific genetic changes in a cell determine the pathological outcome.

As discussed, the early studies of parathyroid adenomas indicated that parathyroid adenomas are indeed monoclonal neoplasms, suggesting that these tumors are caused by mutations that alter the growth regulation of parathyroid cells.

Subsequent molecular analyses revealed the presence of tumor-specific DNA rearrangements in a subset of adenomas in such rearrangements the 5' regulatory region of the PTH gene became separated from its coding exons and was shown to recombine with a novel DNA locus, D11S287.

This rearrangement was both clonal, suggesting its importance in tumor cell selection, and remarkably similar to specific chromosomal translocations observed in various B-cell lymphomas, wherein the tissue-specific regulatory sequences of the immunoglobulin heavy chain gene are juxtaposed with oncogenes like BCL-2 or C-MYC, causing their overexpression in the B-lineage cells.

Thus, by analogy, it was hypothesized that DNA from the non-PTH side of the breakpoint in the parathyroid tumors would harbor an oncogene whose deregulation provided the host parathyroid cell with a selective growth advantage.

An mRNA transcript from this breakpoint-adjacent gene was identified and was found to be dramatically overexpressed in these parathyroid adenomas, providing evidence that the tissue-specific enhancer elements of the PTH gene were indeed deregulating the expression of this putative oncogene.

Subsequent cloning of this candidate oncogene led to the identification of a novel gene (PRAD1/Cyclin D1) with sequence similarities to cyclins.

The PTH gene is localized to chromosomal region 11p15, whereas the cyclin D1 oncogene maps to 11q13. Thus, the simplest explanation for this rearrangement is a pericentromeric inversion, inv(11)(p15;q13), that positions the 5' PTH regulatory region upstream of the cyclin D1 gene.

Cyclin D1 was also cloned independently by two other groups as a murine gene that was induced by growth factor exposure in a macrophage cell line and as a human cDNA that rescued yeast with mutant G1-phase cyclins.

The cyclin D1 gene encodes a 35 kDa protein that shares structural homology and some functional properties with other cyclins. The cyclin D1 protein contains the conserved ‘cyclin box’ and a retinoblastoma (pRB)-binding domain.

During the G1 phase of the cell cycle, cyclin D1 complexes with and activates its kinase partners CDK4 or CDK6. The activated kinases participate in the phosphorylation and inactivation of pRb, effecting entry into S-phase.

Thus, one plausible mechanism for the oncogenic activity of cyclin D1 is via inactivation of the growth inhibitory effects of pRb, a well-established tumor suppressor gene product. Innumerable studies have demonstrated an important role for the cyclin D1 pathway beyond its role in parathyroid neoplasia.

The cyclin D1 gene is amplified in multiple human malignancies, including breast cancers and head and neck carcinomas, and is activated by gene rearrangements (analogous to those in parathyroid tumors) in mantle cell lymphomas and multiple myeloma.

Moreover, activating mutations of CDK4 have been detected in human melanomas and CDK4 is amplified in sarcomas and gliomas. Also, the cyclin D1/cdk4-inhibitor p16 (p16INK4a) has been well established as a tumor suppressor. Thus, the cyclin D1 pathway is a central target in oncogenesis in many tissue types.

As noted above, a subset of parathyroid adenomas contains clonal rearrangements that juxtapose the PTH 5' regulatory region with the cyclin D1 gene. The 11q13 chromosomal breakpoint may be found close to cyclin D1 or may be located as much as 300 kb centromeric of cyclin D1 (Y. Hosokawa et al, unpublished data).

This variability in the location of the breakpoint complicates standard approaches for detection of the rearrangement such as Southern blotting.

Thus, there are no accurate estimates to date of the percentage of parathyroid adenomas that harbor such cyclin D1-activating rearrangements. However, as determined by immunohistochemistry, 20-40% of parathyroid adenomas overexpress cyclin D1.

It is highly plausible that in addition to the described chromosomal rearrangement, other molecular mechanism such as gene amplification, rearrangement with other enhancer/promoters active in parathyroid cells, or transcriptional activation can serve as alternative routes toward the cell’s acquisition of the selective advantage inherent in cyclin D1 overexpression.

For cyclin D1, overexpression of the normal gene product is oncogenic, as no internal activating mutations have been found in parathyroid adenomas or other human tumors.

Other than the still unproven possibility that pRb may be the tumor suppressor target of 13q deletions in parathyroid carcinomas, molecular analyses of parathyroid tumors have failed to detect mutations in other genes involved in the cyclin D1 pathway.

Specifically, inactivating mutations or homozygous deletions of the p16 and p15 cdk inhibitor genes occur uncommonly, if ever, in parathyroid adenomas. Also, it is not yet known whether activating mutations of β-catenin, which is a reported transcriptional activator of cyclin D1, play a role in parathyroid tumorigenesis. Interestingly, there are data suggesting that the existing paradigm that cyclin D1 promotes tumorigenesis only through its effects on pRb and cdks may be too simplistic.

In breast cancer cells, for example, cyclin D1 was reported to complex with the estrogen receptor and activate estrogen receptor-mediated transcription, independent of a cdk partner.

Certainly, the possibility that cyclin D1 may regulate parathyroid-cell growth via yet unknown mechanisms must be seriously considered in the future.

Our laboratory has generated a transgenic mouse model for parathyroid neoplasia, by targeting cyclin D1 overexpression to the parathyroid glands.

These mice (PTH-cyclin D1 mice) harbor a transgene in which the cyclin D1 gene is placed under the control of a 5.2 kb fragment of the PTH regulatory region, thereby mimicking the rearrangement and resultant cyclin D1 overexpression observed in the human tumors.

The resulting phenotype in these animals is remarkably similar to the abnormalities that develop in patients with primary parathyroid neoplasia. By the age of 6-10 months, PTH-cyclin D1 mice develop chronic biochemical HPT, as evidenced by increased serum calcium and PTH levels, and develop bone abnormalities characteristic of HPT.

The parathyroid glands in PTH-cyclin D1 mice are hypercellular, and the relative PTH-calcium setpoint, as estimated in vivo by the concentration of serum calcium needed to half-maximally suppress PTH levels, is increased.

Expression of the calcium sensing receptor protein (CaSR) is decreased in the parathyroid glands of the HPT animals to approximately the same extent as occurs in human parathyroid adenomas.

Furthermore, assessments of parathyroid cell proliferation in HPT animals discovered increased uptake of 5’-bromo-2-deoxyuridine and increased levels of proliferating cell nuclear antigen, as detected by immunohistochemistry.

Thus, tissue-specific overexpression of cyclin D1 does induce parathyroid cell proliferation resulting in HPT, substantiating the role of cyclin D1 as a driver of parathyroid cell growth.

The development of this animal model provides an attractive system to study parathyroid biology and endocrine neoplasia on several fronts. Parathyroid neoplasia is a complex process that involves abnormal cell proliferation coupled with an aberrant control of hormonal secretion.

This link between proliferation and hormonal production/secretion is also evident in normal parathyroid cells which respond to the stimulus of chronic hypocalcemia not only by increasing hormonal secretion but also by a secondary expansion of parathyroid cell mass.

Likewise, in addition to a proliferative defect leading to parathyroid hypercellularity, in vivo studies on parathyroid neoplasms in humans have shown an abnormality in the feedback system through which extracellular calcium regulates PTH secretion an apparent resetting of the ‘set-point’ mechanism that normally tightly couples PTH secretion with ambient calcium levels.

It has been hypothesized that most parathyroid adenomas are caused by acquired mutation(s) in the genes of the set-point pathway, with the abnormal PTH response to the ambient calcium level being the initial driving force for parathyroid tumor cell proliferation.

However, while severe germline deficiency of the CaSR can cause the parathyroid hyperproliferation of neonatal severe hyperparathyroidism, somatic mutations of the Ca++-sensing receptor gene have not been found in sporadic parathyroid tumors.

The PTH-cyclin D1 mouse model has shed light on the issue of whether abnormal cell proliferation is the result or cause of parathyroid cell hormonal setpoint dysregulation.

In wild-type mice, as in humans, the inverse relationship between PTH and serum calcium levels takes the shape of a sigmoidal curve.

In the PTH-cyclin D1 mice, this curve is shifted upward and to the right, resulting in an abnormally high relative set-point, similar to findings in patients with primary or severe secondary hyperparathyroidism.

Thus, it is now clear that the hormonal regulatory defect need not be primary, but can result secondary to primary growth-control disturbances in the gland.

Similarly, the model shows that a primary growth disturbance, at least as driven by cyclin D1, can cause a secondary diminution in expression of the parathyroid calcium-sensing receptor.

In more general terms, the PTH-cyclin D1 mouse model provides an experimental system in which the molecular mechanisms that deregulate set-point control can be dissected.

For example, the status of the vitamin D receptor, an important regulator of PTH gene expression and parathyroid cell proliferation, could be examined in a relevant in vivo context.

These studies will further elucidate the critical links between proliferation and functional abnormalities in parathyroid neoplasia.

Finally, the development of larger adenoma-like lesions in some animals suggests that they may be monoclonal expansions, and the screening of these lesions for acquired genetic alterations could aid in the identification of additional genes involved in human parathyroid neoplasia.