Molecular Genetic Abnormalities in Sporadic Hyperparathyroidism Part II
(4273 total words in this text) -
read: 121 times
Tumor Suppressor Genes: MEN1
The Multiple Endocrine Neoplasia Type 1 Syndrome
The multiple endocrine neoplasia type 1 (MEN1) syndrome, an autosomal dominant inherited disease, is characterized by multiple tumors of the parathyroids, enteropancreatic tissues and anterior pituitary.
Linkage studies in families affected by MEN1 discovered that the gene responsible for MEN1 is located on chromosome 11q13 near PYGM, a muscle phosphorylase gene.
The MEN1 gene, like the genes in other previously characterized familial tumor syndromes, was thought to act as a classic tumor suppressor gene, as acquired inactivating alterations at 11q13 uniformly involved the allele inherited from the unaffected parent in MEN1-associated tumors.
These DNA alterations were seen, in many such tumors including those of the parathyroid, as loss of polymorphic DNA markers from this region of chromosome 11.
Thus, it was hypothesized that an inherited mutation in one allele in combination with an acquired deletion of the remaining, normal MEN1 allele would result in the absence of functional protein product and a selective growth advantage to the transformed cell.
The MEN1 gene was eventually identified by positional cloning and a classic tumor suppressor role for MEN1 was confirmed.
Germ-line mutations were found throughout the protein-coding region of MEN1 in patients with MEN1 syndrome.
Nonsense or frameshift mutations that result in a truncated protein make up 70% of germline mutations while missense mutations or in-frame deletions make up the remaining 30%.
No correlation between mutation and phenotype of the syndrome has been found, and the functional consequences of many of the subtler mutations have not yet been precisely elucidated.
Clues to the Action of Menin
Menin, the protein product of the 10 exon MEN1 gene, is a 610-amino acid protein with no homology to any known proteins.
Its sequence was initially reported to contain no common motifs that provide clues to its function. Subsequently, two nuclear localization signal sequences were identified at the carboxyl-terminus of the menin protein, and as expected, menin seems to be targeted to the nucleus.
Interestingly, MEN1 is expressed at the mRNA and protein level in nearly all tissue types examined, rather than being limited to the tissue types susceptible to MEN1-associated tumors. Menin levels vary as cells in culture proliferate.
Pituitary cells synchronized at the G1-S phase boundary express menin at a lower level than G0-G1-synchronized cells, and the expression of menin increases as the cell enters S phase.
Cells synchronized at the G2-M phase express lower levels of menin.
The stable overexpression of menin in NIH3T3 cells transformed with ras was able to inhibit cell growth and tumor formation by these cells in xenografts.
This finding supports the hypothesis that menin can act as a growth inhibitor, but it is still unclear how it does so.
Proteins that interact with menin have been, and continue to be, identified in hopes that this information may provide further clues to its function.
Menin was shown to interact with the activator protein 1 factor JunD in a yeast two hybrid system, and can repress JunD-activated transcription.
Mutants of JunD have been characterized that fail to interact with menin.
These studies indicate that JunD binds menin at JunD’s amino terminus in a region that shares little homology with other jun proteins.
Unlike other jun family members, JunD appears to be antimitogenic.
Therefore, it is far from clear how menin could act as a tumor suppressor by repressing JunD activity.
It also has been reported that menin can directly interact with Smad3, a downstream component of the TGF-β signaling pathway. When TGF-β receptors are activated, Smad3 becomes phosphorylated and can subsequently enter the nucleus to alter transcription.
TGF-β signals usually inhibit cell growth. However, when menin levels are decreased by antisense MEN1 gene transcripts, cultured pituitary cells are less susceptible to the growth inhibitory effects of TGF-β.
Antisense menin also inhibits the transcription of TGF-β target genes.
These results indicate that menin may play a role in the growth inhibitory actions of the TGF-β pathway, and loss of menin may decrease the growth restriction exerted by this pathway.
This explanation of menin’s role in normal cell biology seems more consistent with its tumor suppressor activity than its paradoxical effects on JunD.
However, more work needs to be done to better understand menin’s role in the TGF-β pathway, JunD transcription, as well as perhaps in other cellular processes.
A better understanding of menin’s function may also help to explain why menin is expressed in most tissues, but only a limited number of tissues are affected by tumors in the MEN1 syndrome.
Menin in Sporadic Parathyroid Adenomas
Studies of parathyroid tumors have found that allelic loss of chromosome 11 markers occurs in 25-40% of sporadic parathyroid adenomas.
This allelic loss usually includes the chromosomal region to which the MEN1 gene has been mapped.
Somatic mutation and/or deletion of MEN1 resulting in its complete (biallelic) inactivation have been documented in 12-17% of sporadic parathyroid adenomas.
This 12-17% accounts for approximately half of the 25-40% of parathyroid adenomas with losses of 11q13.
Noncoding mutations in MEN1 such as mutations within promoter or enhancer regions, and/or epigenetic inactivation by promoter hypermethylation, which would have been missed in these analyses, might explain some of this discrepancy.
However, there is also a real possibility that a second tumor suppressor gene on chromosome 11 might be important in the pathogenesis of parathyroid adenomas and is disrupted in those tumors with 11q13 loss and no MEN1 mutations.
Consistent with this possibility, studies in another endocrine tumor, follicular thyroid tumors, have also suggested that a tumor suppressor gene in addition to MEN1 may be located on chromosome 11q13.
Somatic MEN1 mutation is not limited to sporadic parathyroid tumors as it has also been reported in sporadic gastrinomas, insulinomas, lung carcinoids and angiofibromas.
A Mouse Model of MEN1
A mouse model of MEN1 has been developed.
Mouse Men1 shares 97% identity with its human counterpart at the amino acid level and, like the human version, is expressed in many tissues.
A mouse with a floxed neomycin cassette in intron 2 and a third lox P site in intron 8 on the MEN1 gene was created. Mice homozygous for this targeted disruption were not viable.
Mice heterozygous for the disrupted men1 allele developed tumors characteristic of MEN1 syndrome including pancreatic islet cell tumors, pituitary adenomas, adrenal cortical tumors, and parathyroid tumors.
Pituitary tumors that developed in the mouse secreted prolactin as is commonly seen in the pituitary tumors which arise in humans with the MEN1 syndrome.
Pancreatic hyperplasias and tumors produced insulin leading to elevated serum insulin levels that correlated in severity with the abnormalities found in the pancreas.
However, mice with parathyroid tumors did not develop elevated PTH or calcium levels.
This suggests that despite their increased cell mass, these parathyroids may retain the ability to appropriately regulate PTH production in each cell in order to maintain normal calcium and PTH levels.
One might therefore hypothesize that these cells maintain normal expression of the calcium sensing receptor.
It will be important to determine this in order to better understand the parathyroid pathophysiology in this mouse model and illuminate the basis of this interesting discrepancy in its physiology as compared with the cognate human condition.
Alternatively, unlike human tumors, the mouse parathyroid tumors may be nonfunctional.
Tumors that were analyzed for loss of heterozygosity showed acquired loss of the normal allele, indicating that tumors are developing in this mouse model in a manner similar to the way tumors develop in human MEN1.
Further studies of this model may help to understand the role of MEN1 in tumor formation, although the lack of hormonal activity of tumors in this model appears to represent a significant limitation in its potential utility for investigating some of the hormonal syndromes of MEN1.
RET and Calcium Sensing Receptor
Rare inherited conditions such as MEN1 with predisposition to a particular tumor type have yielded important information about the pathogenesis of the more common, sporadic type of the same tumor.
Several such disorders in addition to MEN1 are accompanied by altered parathyroid function.
Benign parathyroid tumors are found in 10-20% of MEN2A patients.
Inherited mutations in the RET gene, encoding a receptor tyrosine kinase, have been found in nearly all patients with MEN2A, but in contrast with MEN1 no somatic RET mutations have been observed in sporadic parathyroid adenomas.
Thus, acquired activating mutations in RET either do not occur in normal individuals’ parathyroid cells with any appreciable frequency, or such mutations do not appear to confer a clinically significant selective advantage upon a parathyroid cell when they do occur. Familial hypocalciuric hypercalcemia (FHH) can result from a germline inactivating mutation in one allele of the calcium-sensing receptor (CASR), located on chromosome 3q21.1.
Although the parathyroid cells of typical FHH patients do not properly sense the level of blood calcium, parathyroid cell proliferation is not significantly increased.
However, inheritance of two defective CASR genes results in neonatal severe hyperparathyroidism, a disease in which the parathyroids grow excessively at a very young age.
This link between inherited severe calcium sensing defects and parathyroid cell proliferation raised the possibility that acquired severe (i.e., biallelic inactivating) alterations in the CASR gene might also increase growth in parathyroid cells and result in sporadic parathyroid tumors.
However, no inactivating mutations in CASR have been discovered in sporadic tumors indicating that CASR mutations do not appear to provide a growth advantage when somatically acquired in a parathyroid cell.
It remains possible that the decreased expression of the CaSR observed in parathyroid adenomas, apparently a secondary result of distinct primary tumorigenic alterations, might still play a role in their pathogenesis.
Nonetheless, among the genes most directly responsible for the rare inherited forms of hyperparathyroidism, only MEN1 has been solidly linked to the development of the common sporadic form of the disease.
Vitamin D Receptor
Vitamin D, more specifically 1,25 dihydroxyvitamin D3, is known to inhibit parathyroid proliferation. The vitamin D receptor (VDR) gene has been investigated as a possible target for acquired inactivation in parathyroid tumors but no specific clonal mutations have been found, either in parathyroid adenomas or in severe secondary/tertiary HPT of uremia.
Thus, VDR does not function as a classic tumor suppressor gene in parathyroid tissue, and its inactivation does not appear to be a primary driving force in parathyroid tumorigenesis.
Studies have found a reduced level of VDR messenger RNA and protein levels in parathyroid adenomas and have correlated specific VDR germline polymorphisms with an increased susceptibility to parathyroid tumor formation, but the exact significance and mechanism of these associations remain to be defined.
Other Genetic Abnormalities in Sporadic Parathyroid Adenomas
Studies of sporadic parathyroid adenomas have revealed highly recurrent clonal allelic losses that may indicate the genomic locations of key parathyroid tumor suppressor genes.
Molecular allelotyping studies have shown frequent loss of heterozygosity on chromosomes 1, 6, 11 and 15 as well as less frequent losses on 9 and 13.
Studies of sporadic adenomas using comparative genomic hybridization (CGH), a molecular cytogenetic method which detects regions of chromosomal gains/losses in the tumor cell genome, have confirmed these common areas of deletion.
CGH has also identified areas of chromosomal gain on chromosome 7, 16 and that may signify the locations of new parathyroid oncogenes.
Unfortunately, traditional cytogenetic studies have not yielded significant insight into the locations of important genes involved in the pathogenesis of parathyroid tumors.
While a cytogenetic translocation between chromosomes 1 and 5 has been reported in a single parathyroid adenoma, the significance of this finding remains unclear.
Several candidate tumor suppressor genes that map to these regions of loss have been examined for mutations in parathyroid adenomas.
RAD51 and RAD54, located on 15q and 1p respectively, served as good candidate genes as they play an important role in DNA repair and recombination mechanisms that are important following radiation.
The incidence of parathyroid adenomas is increased following ionizing radiation, and possibly, a defective DNA repair machinery may contribute to the development of these tumors as well as tumors not associated with radiation.
CDK inhibitors, p15, p16 (both on 9p) and p18 (on 1p), which are involved in the negative regulation of the cell cycle, are located in known regions of chromosomal loss.
No tumor-specific mutations were found in either the CDK inhibitors or the RAD genes in parathyroid adenomas, indicating that these genes do not appear to play an important role in the development of sporadic adenomas.
No proto-oncogenes in candidate regions have been investigated for activating mutations.
However, the ras gene was examined, and no tumor-specific mutations were discovered. Knowledge of the patterns of acquired and recurrent chromosomal aberrations will hopefully prove useful for the identification of the full set of oncogenes and tumor suppressor genes that contribute to the development of parathyroid adenomas.
Microsatellite Instability
Studies of several tumor types have demonstrated that cancer cells lack regulation of genomic stability.
One type of such instability is microsatellite instability (MSI), caused by a defective mismatch repair mechanism.
Microsatellites, or short tandem repeats are composed of di-, tri-, tetra and pentanucleotide repeat sequences. MSI is defined as a change in the length of repeats within a tumor, when compared with normal tissue from the same individual.
This tumor-specific allelic change, either due to insertion or deletion of repeating units, is reflective of a defective mismatch repair (MMR) system.
The defective MMR fails to recognize and repair DNA replication errors, and thus, enhances the accumulation of single nucleotide mutations and alterations in the length of simple, repetitive microsatellite sequences that occur ubiquitously throughout the genome.
MSI is observed in most hereditary nonpolyposis colorectal cancers (HNPCC), and is associated with inherited mutations in the MMR genes hMSH2, hMLH1, hPMS1, hPMS2 and hMSH3.
In addition to HNPCC, MSI is also observed in nearly 20% of sporadic colorectal tumors, and extracolonic tumors including breast, endometrial, gastric and ovarian cancer.
It has been hypothesized that the growth rates of parathyroid tumors may be too low to account for the number of clonal mutational events that have been detected in these neoplasms, 6 although this argument may not adequately recognize the power of selection over an extended time period.
In any case, a mutational event(s) that results in an increase the rate of genomic instability could potentially be operational in the parathyroid.
Indeed, two studies have suggested that MSI may play a role in parathyroid tumorigenesis.
However, a limitation of these studies was their small sample size and a lack of systematic analysis of a genome-wide state of microsatellite instability.
Thus, the contribution of this type of instability to the genesis of parathyroid tumors needs to be examined more conclusively.
Molecular Genetics of Parathyroid Carcinoma
While parathyroid carcinomas are very rare, they are associated with significant morbidity and mortality and are important to consider in the differential diagnosis of patients with primary HPT.
Carcinomas often present in younger patients than do adenomas, and are more frequently symptomatic with more severe hypercalcemia.
Nevertheless, it remains difficult to distinguish between adenomas and carcinomas on clinical and histological grounds, in the absence of distant metastases.
Therefore, the discovery of genetic changes unique to carcinomas may aid in the discrimination between the two types of tumors.
Such molecular diagnostic information might be especially valuable in the “atypical adenoma”, which has certain histologic features that suggest, but are not specific for, an aggressive phenotype.
[Note added in proof: Mutation of the HRPT2 tumor suppressor gene has recently been identified as a major factor in the pathogenesis of parathyroid carcinoma (Shattuck et al, N Engl J Med 2003; Howell et al, J Med Genet 2003).] Observed patterns of acquired clonal chromosomal changes appear to hold promise for their identification and for diagnostic use in their own right.
Molecular allelotyping and CGH have been used to identify areas in the genome where oncogenes and tumor suppressor genes involved in parathyroid carcinomas might be located.
Losses on 1p, 3q, 4q, 13q and 21q indicate possible areas where tumor suppressor genes involved in the development of carcinomas may lie.
CGH studies have described nonrandom chromosome gains, suggesting the possible genomic locations of oncogenes involved in parathyroid carcinogenesis. However, most gains have not been seen consistently in multiple studies.
A subset of these genetic abnormalities appears to occur with greater frequency in parathyroid carcinomas than in adenomas.
These special regions may harbor genes that contribute to the invasive or metastatic behavior of the carcinomas.
Interestingly, carcinomas also tend to lack most of the chromosomal changes that are commonly present in adenomas.
This finding suggests that carcinomas do not generally originate from typical benign adenomas, but instead that carcinomas and adenomas develop along separate pathways driven by distinct genetic changes. Strong evidence indicates that one key tumor suppressor gene important for the development of malignant parathyroid tumors is located on chromosome 13.
Several groups have demonstrated that definite carcinomas and other clinically aggressive parathyroid tumors frequently have loss of heterozygosity on 13q.
The region of loss was shown to include the RB1 gene and BRCA2.
In tumors with 13q loss, protein levels of pRB, as detected by immunohistochemistry, were decreased in accord with a potential role for RB; analogous expression evidence is not available for BRCA2.
That said, there are many genes on 13q, and it is quite conceivable that the true target of these acquired deletions, i.e., a classic 13q tumor suppressor whose biallelic inactivation is a driving force in selection of parathyroid cancers, will prove to be neither RB nor BRCA2.
Resolution of this issue for any given 13q candidate gene awaits analysis of its sequence for specific internal inactivating mutations, the key evidence required to prove involvement as a parathyroid tumor suppressor gene. 13q loss has also been reported in a smaller percentage of parathyroid adenomas.
Future research will determine if the relevant 13q tumor suppressor genes in parathyroid carcinomas vs. adenomas are identical.
Loss of the p53 tumor suppressor gene has been found occasionally in parathyroid carcinomas, but direct mutations have not been described.
Therefore, p53 is unlikely to play a major role as a classic tumor suppressor in the development of parathyroid carcinomas.
Immunohistochemical studies have detected overexpression of the cyclin D1 oncogene in 50-91% of parathyroid carcinomas, apparently even higher than the 20-40% overexpression observed in parathyroid adenomas.
These findings raise the possibility that cyclin D1 may also play a major role in the development of parathyroid cancers.
In addition to its established role in adenomas, it will be important to determine the effects of cyclin D1 overexpression in the context of parathyroid malignancy, and whether cyclin D1 can act as a primary driver of parathyroid carcinomas.
Molecular Genetics of Secondary and Tertiary Hyperparathyroidism
The molecular basis of severe secondary or tertiary HPT is poorly understood. Because of multigland involvement, it was previously assumed that this condition predominantly involves polyclonal (non-neoplastic) cellular proliferations.
Although this is likely to be the case in the initial proliferative phase, it is now clear that monoclonal parathyroid expansion does occur in most patients with severe secondary or tertiary HPT.
Emergence of such monoclonal expansions may well be a major factor in the acquisition of an increasingly autonomous PTH regulation and the refractoriness to conventional medical therapy found in severe secondary or tertiary HPT.
Immunohistochemical studies have not detected overexpression of cyclin D1 in parathyroid glands from patients with uremic HPT, suggesting an infrequent role for cyclin D1 in the development of this disease.
Interestingly, acquired inactivation of the MEN1 gene, also relatively common in sporadic primary HPT, seems to play only a negligible role in the clonal expansion of uremia-associated lesions.
Only 2-3% of this form of sporadic parathyroid tumors exhibit allelic loss at 11q13, and MEN1 mutation was found in just a subset of these.
Thus, the molecular genetic basis for the development of monoclonal parathyroid tumors in uremic patients appears to differ markedly from that in primary HPT.
These findings lend further support to the hypothesis that different forms of parathyroid neoplasia develop through unique pathogenic mechanisms.
