| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Studies |
Dipartimento di Biochimica e Biotecnologie Mediche and CEINGE Centro di Ingegneria Genetica (L.Q., O.F., M.P.C., V.C.), and Dipartimento di Patologia e Biologia Cellulare e Molecolare, Facoltà di Medicina e Chirurgia, Università degli Studi di Napoli Federico II (A.P.), 80131 Naples; Divisione di Endocrinologia Ospedale Niguarda (U.V., A.L.), 20100 Milan; and Facoltà di Scienze Matematiche, Fisiche e Naturali, Università del Sannio (L.Q., V.C.), 82100 Benevento, Italy
Address all correspondence and requests for reprints to: Dr. Loredana Quadro, Facoltà di Scienze Matematiche, Fisiche e Naturali, Università del Sannio, 82100 Benevento, Italy. E-mail: quadro{at}dbbm unina.it; or Dr. Vittorio Colantuoni, Dipartimento di
| Abstract |
|---|
|
|
|---|
| Introduction |
|---|
|
|
|---|
The gene responsible for these tumors is the RET protooncogene (2, 3, 4, 5, 6), coding for a membrane tyrosine kinase receptor expressed in neural crest-derived cells (7, 8, 9). Ninety-five percent of MEN 2A cases bear point mutations of the RET gene (10) mainly in exons 10 and 11 (2, 3, 5, 6). Thirty to 50% of sporadic MTCs cases bear a RET mutation in exon 16 (4, 11, 12, 13, 14, 15); less frequent are mutations in other exons (11, 12, 15, 16).
Duplication or deletion of segments of various lengths of the RET gene take place only occasionally; 9- or 12 bp duplications in exon 11 have been reported in a few MEN 2A cases (17, 18). A 27-bp deletion in exon 10 (19) and a deletion-insertion in exon 11 have been reported in sporadic MTCs (20).
In other hereditary tumors, a germline mutation in a predisposing gene is usually followed by functional loss of the homologous allele in the tumor tissue (21). In these cases the affected gene acts as a tumor suppressor, and the disease is established only when both copies of the gene are altered (22). In MEN 2 tumor DNA, no gross abnormalities are found at the RET locus, nor are other markers in proximity of the gene affected. In a percentage of MEN 2-related tumors, loss of heterozygosity (LOH) on chromosomes 1, 3, 11, 17, and 22 has been described and related to loss of function of putative suppressor genes (23, 24, 25, 26, 27, 28, 29). However, the role of these LOH in the tumor progression process is not known yet.
Here we report a MEN 2A case affected by a typical heterozygous germline mutation (Cys634Arg) in exon 11 and by an additional somatic deletion of the wild-type allele. The deletion spans exon 4 to exon 16, does not include extragenic sequences, and is not associated with base pair changes in any other mutation-bearing RET exons. This is the first report of a MEN 2A case with loss of heterozygosity of the RET gene. This genetic profile may be related to the early onset, the clinical course with local and distant metastases, and the very poor outcome.
| Materials and Methods |
|---|
|
|
|---|
DNA was extracted from the primary tumor, a metastatic node, and peripheral blood lymphocytes and used as a template for PCR reactions performed as previously described (30). RET exons 10, 11, 13, 14, 15, and 16 were amplified using as primers the oligonucleotides reported previously (4, 5, 31, 32, 33, 34). The purified PCR products were sequenced on both strands with the Sanger method, using the Sequenase PCR product sequencing kit (U.S. Biochemical Corp., Cleveland, OH) and the primers used for the amplification. Sequence reactions were then run on 6% denaturing polyacrylamide gels and autoradiographed overnight. In some cases, 100200 ng of the PCR products of exons 2, 11, 13, 14, and 16 amplicons were digested with 25 U EagI, BanI and HhaI, AluI, NlaIII, or FokI restriction enzymes, respectively (31, 32, 35, 36). Samples were then run on 2% or 3% agarose gels, and DNA bands were visualized after ethidium bromide staining. The D10S94 locus was amplified using the primers described previously (37), and 100200 ng of the purified PCR product were digested with 25 U PvuII restriction enzyme. Genomic DNA was also analyzed by Southern blot.
Quantitative PCR procedure
The quantitative PCR analysis was performed as previously described (38). The duplex PCR reaction for all RET exons was performed by amplifying as internal control exon 10 of the piruvate kinase (PK) gene (39). The primers used for the PK gene were: 10F, 5'-CTCGTTCACCACTTTCTTGC-3'; and 10R, 5'-GGGAAGCTGGGTTGGGGGC-3', both of which were derived from the intronic sequence. The primers used to amplify the RET exons 10, 11, 13, 14, 15, and 16 were reported previously (4, 5, 31, 32, 33, 34); the others were designed according to the gene sequence (1). The duplex PCR reactions were performed in a 50-µL final volume, containing 100200 ng genomic DNA, 50 mmol/L KCl, 10 mmol/L Tris-HCl (pH 8.3), 1.5 mmol/L MgCl2, 5 mmol/L NH4Cl, 200 µmol/L of each deoxy-NTP, 30 pmol of each primer, 4 µCi (3000 Ci mM) of 32P-labeled deoxy-CTP, and 2.5 U Taq polymerase (AmpliTaq, Perkin-Elmer Corp., Norwalk, CT). The PCR cycles were as follows: denaturation step for 7 min at 94 C; 22 cycles of denaturation, annealing, and extension performed at 94 C for 1 min, 52 C for 2 min, and 72 C for 5 min; and a final elongation step at 72 C for 7 min. The samples were stored at 4 C until required. Four microliters of each PCR product were pipetted from each tube and mixed with 1 µL loading dye. The samples were electrophoresed on a 5% polyacrylamide gel (39:1 acrylamide/bisacrylamide ratio) in 1 x Tris Borate EDTA buffer for 1 h at 1800 V. The gel was dried and exposed on the GS-250 Molecular Imager (Bio-Rad Laboratories, Inc., Hercules, CA) for about 1518 h.
| Results |
|---|
|
|
|---|
In 1986, a 18-yr-old woman was referred for evaluation to the Niguarda Hospital and underwent fine needle aspiration biopsy of a lateral cervical lymph node that had rapidly enlarged. Histological examination revealed MTC, as confirmed by elevated serum calcitonin (314 µg/ml) and carcinoembryonic antigen (44.5 ng/mL) levels. Total thyroidectomy with central compartment resection was immediately performed. The thyroid was bilaterally affected by the tumor that had invaded the neighboring vessels. Local nodes were metastasized. No clinical signs or symptoms of a pheochromocytoma were present. After treatment and a period of well-being, the patient complained of general discomfort due in part to enlargement of the cervical nodes that compressed the adjacent veins. They were removed, but later the patient was again hospitalized for respiratory problems due to several metastatic infiltrates in both lungs, as revealed by computed tomography. The patients general condition progressively worsened, and in February 1995 she died at the age of 27 yr. Her family history was negative for relevant thyroid pathology.
Identification of RET proto-oncogene mutations
DNA from peripheral blood lymphocytes, primary (1986 surgery) and
metastatic (1991 surgery) tumors were analyzed for RET
protooncogene mutations. Exon 11 from peripheral blood DNA was PCR
amplified and sequenced. A heterozygous TGC to CGC mutation was
detected at codon 634 causing a Cys to Arg substitution (Fig. 1
, A and B, patient blood). The
heterozygous mutation of germline DNA was also found in the DNA from
the primary tumor (data not shown). In contrast, DNA from the
metastatic node revealed only the sequence corresponding to the mutated
allele (Fig. 1
, A and B, patient metastatic tumor). The specific
mutation was confirmed by HhaI restriction enzyme digestion
(data not shown). A new site for this enzyme is, in fact, generated by
the Cys634Arg mutation (36).
|
We performed phosphorimager-based quantitative PCR analysis to
establish whether the genomic profile detected in the metastatic DNA
was due to RET LOH (38). This assay entails the
quantitative analysis of two 32P-labeled PCR
products, a possibly deleted exon vs. a nondeleted exon.
Each RET exon from germline, primary tumor, and metastatic
DNA was coamplified with exon 10 of the PK gene (39) as an
internal control. This gene was chosen because it maps on the long arm
of chromosome 1, and it should not be involved in any mutation
affecting the RET gene located in the centromeric region of
chromosome 10 (1). Each experiment was carried out on DNA
from the patient and from a normal individual. The radioactive
amplified products were quantified by phosphorimager scanning, and the
ratio of the intensity of the RET exon vs. the
internal control was referred to as R. The ratio of the
R value of the patient vs. that of a normal
individual was the diagnostic index (DI). If the DI ranges between
0.81, the tested exon is present in two copies; if the DI ranges
between 0.40.6, the exon is present in a single copy. All
RET exons amplified from germline and primary tumor DNA gave
a DI between 0.81, clearly indicating that they were present in two
copies, as in normal DNA. In contrast, the DI for exons 416 amplified
from the metastatic DNA ranged between 0.40.6, indicating that they
were in a single copy. RET exons 13 and 1721 had normal
values, indicating that they were preserved on both alleles. Figure 2
illustrates the results obtained with
some representative exons, i.e. 3, 11, 16, and 17. The
histogram shows the DI obtained with normal DNA and with constitutive,
primary, and metastatic tumor DNA from the proband.
|
To further confirm the results of the quantitative PCR, two
RET intragenic heterozygous polymorphisms were analyzed. One
maps at the Ala45 codon in exon 2, and the second
maps at the Gly691 codon in exon 11
(35). The former polymorphism eliminates the restriction
site for the enzyme EagI, the latter eliminates that for
BanI. Patient DNA from peripheral blood and metastatic tumor
were heterozygous for exon 2 polymorphism (Fig. 3A
). Exon 2 PCR-amplified products
digested with EagI produced, in both cases, a band of 265
bp, corresponding to the undigested allele and two bands of 203 and 62
bp derived from the digested allele. In contrast, a different profile
was detected on DNA from peripheral blood and metastatic tissue
analyzed for the polymorphism at exon 11 (Fig. 3B
). BanI
digestion of the exon 11 PCR product from peripheral blood yielded a
band of 338 bp, corresponding to the undigested allele and two bands of
269 and 69 bp generated from the digested allele. In metastatic tumor
DNA, the BanI digestion profile gave only a single high
molecular weight band, corresponding to the undigested allele,
indicating that the polymorphism was not retained. In both digestions,
DNA from a normal individual showing the same polymorphism in
heterozygosity was used as a positive control.
|
Southern blot analysis confirms RET LOH
The probands genomic DNA from peripheral blood and metastatic
tumor tissue was analyzed by Southern blot. Figure 3C
shows the results
of EcoRI and EcoRI/SalI digestions of
patients DNAs hybridized with the XhoI/BglII
RET complementary DNA (cDNA) fragment, spanning from exon 14
to the 3'-end of the cDNA (1). Figure 3D
shows the results
of the hybridization with the full-length RET cDNA of the
same DNA sample digested with EcoRI (1). The
pattern observed corresponds to that expected, except for a 2.8-kb band
present in the germline DNA and absent in the metastatic tumor DNA. The
2.8-kb band was also observed on the DNA from two independent normal
subjects digested with the same enzymes. The disappearance of this band
in the tumor DNA could be due to a polymorphic EcoRI site
present in the wild-type allele, downstream from exon 14. This confirms
the RET LOH, because the site is no longer retained in the
metastatic DNA as a consequence of the interstitial deletion on the
wild-type allele. However, from the data obtained, we cannot establish
the type of rearrangement that occurred, i.e. translocation
of part of the gene to another chromosome or rejoining of the two
portions of the gene.
| Discussion |
|---|
|
|
|---|
The RET mutations underlying all inherited tumor syndromes cause thyroid C cell hyperplasia, which does not fully explain either tumor progression or the phenotypic variations observed not only among families with the same mutations, but also among individuals in the same family bearing the same mutation (41). Accumulation of genetic alterations, functional loss of tumor suppressor genes, and activation of other oncogenes could contribute to tumor development. In fact, LOH on chromosomes 1, 3, 11, 17, and 22 has been reported in MEN 2 cases (23, 24, 25, 26, 27, 28, 29).
Despite the consolidated idea that RET acts as a dominant oncogene due to single mutations, other alterations of this gene have recently been reported in MTC. Simultaneous germline mutations at exon 10 or 11 and somatic mutations at codon 918 in exon 16 have been described (34, 42). Miyauchi et al. (43) also found a germline mutation at codon 768 in exon 13 and a concomitant somatic mutation at codon 919 in exon 16 in a familial MTC. These reports, however, did not show on which allele the additional somatic mutations occurred. More interestingly, Uchino et al. (44) reported two sporadic MTCs with point mutations at codons 918 and 919 of exon 16 and at codon 778 of exon 13, respectively. In both cases, the wild-type allele was absent.
The case analyzed here bears a typical RET MEN/2A germline mutation in exon 11 (Cys634Arg) and loss of the wild-type allele in the tumor. In particular, LOH was detected only in the metastasis and not in the primary tumor, suggesting that this second mutational event had occurred somatically and that it probably could have played a role in tumor progression. Our hypothesis is that this LOH may be related to the clinical course observed. No additional mutations were found in other RET exons analyzed (data not shown). This represents the first report of a MEN 2A case characterized by a germline mutation and a somatic loss of the wild-type allele.
The alteration found was a large intragenic deletion of about 20 kb encompassing exons 416 of the RET locus. These results ruled out a gene conversion event that should transform the wild-type to the mutated allele. In fact, gene conversion usually involves larger chromosomal regions and is detected for genes present in clusters and among nonallelic copies (45). This was not the case, as the deletion did not extend to extragenic sequences at the 3'-end of the RET gene and was present only in the metastatic DNA. In addition, the loss of genetic material at the RET locus was a random event and was not due to patients genome instability. LOH at other loci distributed either on chromosome 10 or on other chromosomes was tested, but no instability was detected (data not shown).
We analyzed the rearrangement that occurred at the wild-type locus by quantitative PCR, Southern blot, and RFLPs of chromosome 10. Furthermore, the expression study of the metastatic tissue ruled out that the rearranged wild-type allele was still transcribed, indicating that the deletion had irreversibly compromised its architecture and function. In fact, in addition to the transcripts derived from the mutated allele, no bands corresponding to shorter or longer wild-type RET messenger ribonucleic acid, derived from the rearranged copy, were detected by Northern blot or ribonuclease protection (data not shown).
LOH of tumor suppressor genes has been described in inherited cancers, classically represented by retinoblastoma. The genes responsible are inactivated by a germline mutation in one allele and by somatic loss of the second allele. The tumorigenic phenotype is manifested only when both copies are altered, according to Knudsons two-hit model (22). The case we report is unusual and novel, because two mutations were detected in a dominant oncogene. This event does not appear to be redundant. We hypothesize that the somatic LOH could add to the effects of the germline mutation, resulting in an aggressive phenotype, and could represent the second or one of the subsequent steps in tumor progression. It was, in fact, found only in two distinct consecutive metastases at different locations.
We propose a model in which the wild-type RET form influences the activity of mutated RET species, functioning as a mitigating factor in the transformation process. In classical heterozygous MEN 2A, the mutant RET may form either homodimers, whose activation is ligand independent or heterodimeric complexes with wild-type RET, whose activation may still be ligand dependent. In contrast, in our case the absence of the wild-type species could allow the mutant RET to form only constitutively activated homodimers, thus enhancing the transduction of mitogenic signals. In fact, wild-type RET can form complexes with mutant RET, influencing the amount of active heterodimers exposed on the cell surface (46). This mechanism could explain the rapid tumor growth, its wide dissemination, the distant and very invasive metastases, and the poor outcome of the disease. Experiments to test this hypothesis in vitro are underway. Moreover, a larger series of patients needs to be analyzed to make a correlation between severe symptomatic MTC and the pathogenetic mechanism we propose.
| Acknowledgments |
|---|
| Footnotes |
|---|
Received December 29, 1999.
Revised May 23, 2000.
Revised August 10, 2000.
Accepted October 1, 2000.
| References |
|---|
|
|
|---|
-like globin
gene family. Hum Genet. 104:117125.[CrossRef][Medline]
This article has been cited by other articles:
![]() |
D. Cordella, M. Muzza, L. Alberti, P. Colombo, P. Travaglini, P. Beck-Peccoz, L. Fugazzola, and L. Persani An in-frame complex germline mutation in the juxtamembrane intracellular domain causing RET activation in familial medullary thyroid carcinoma. Endocr. Relat. Cancer, September 1, 2006; 13(3): 945 - 953. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. J. Marx and W. F. Simonds Hereditary Hormone Excess: Genes, Molecular Pathways, and Syndromes Endocr. Rev., August 1, 2005; 26(5): 615 - 661. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. Elisei, B. Cosci, C. Romei, L. Agate, P. Piampiani, P. Miccoli, P. Berti, F. Basolo, C. Ugolini, R. Ciampi, et al. Identification of a Novel Point Mutation in the RET Gene (Ala883Thr), Which Is Associated with Medullary Thyroid Carcinoma Phenotype Only in Homozygous Condition J. Clin. Endocrinol. Metab., November 1, 2004; 89(11): 5823 - 5827. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. J. Marsh, G. Theodosopoulos, K. Martin-Schulte, A.-L. Richardson, J. Philips, H.-D. Roher, L. Delbridge, and B. G. Robinson Genome-Wide Copy Number Imbalances Identified in Familial and Sporadic Medullary Thyroid Carcinoma J. Clin. Endocrinol. Metab., April 1, 2003; 88(4): 1866 - 1872. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |