| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
CLINICAL CASE SEMINAR |
Divisions of Endocrinology (G.O., G.P., A.C., S.M.), Department of Medical Sciences, Med-Università di Cagliari and Medical Genetics (S.L., C.C.), University of Cagliari, 09042 Monserrato-Cagliari, Italy; Endocrinology and Experimental Oncology (V.D.F., M.S., R.M.M.), "Federico II" University, 80131 Naples, Italy; and Department of Endocrinology (R.E., C.R.), University of Pisa, 56124 Pisa, Italy
Address all correspondence and requests for reprints to: Prof. S. Mariotti, Endocrinology, Department of Medical Sciences, University of Cagliari-Policlinico Universitario, S.S. 554, Bivio per Sestu, 09042 Monserrato-Cagliari, Italy. E-mail: mariotti{at}pacs.unica.it.
| Abstract |
|---|
|
|
|---|
CAG)], serendipitously found in a 23-yr-old woman with hypothyroidism due to atrophic Hashimotos thyroiditis and primary ovarian failure, without altered calcitonin secretion. Familial history and clinical and biochemical evaluation of first-degree relatives were negative for FMTC, MEN 2A and 2B, and Hirschsprung disease. Genetic analysis showed that the mutation was inherited from the mother, who was submitted 2 yr before to thyroidectomy for goitrous Hashimotos thyroiditis. Histological revision and immunohistochemical studies documented normal C cell number and morphology. We cloned the mutation in an expression vector encoding a full-length RET protein. The construct was transiently expressed in 293T cells in parallel with a wild-type RET and a C634R MEN 2A-associated RET mutant. Proteins were harvested from transfected cells, and tyrosine phosphorylation levels were assayed. The mutation did not exert significant potentiating effects on RET kinase. A focus assay was also performed on NIH3T3 fibroblasts; the mutant did not exert significant transforming activity. In conclusion, a new RET mutation was found in two subjects without any evidence of MEN and FMTC. In keeping with clinical data, transfection studies confirmed lack of activating activity. This serendipitous discovery, apparently devoid of oncogenic potential, underscores the problems that may be encountered in genomic studies on RET.
| Introduction |
|---|
|
|
|---|
A rather restricted number of genomic RET point mutations are responsible for all cases of inherited multiple endocrine neoplasia type 2 (MEN 2A, MEN 2B) and familial medullary carcinoma (FMTC). The same somatic mutations are found in nearly half of the sporadic forms of medullary thyroid carcinoma (MTC). MEN 2A and FMTC are both characterized by MTC, but MEN 2A patients also develop pheochromocytoma and parathyroid hyperplasia (8, 9, 10), rarely associated with cutaneous lichen amyloidosis (11, 12). In addition to MTC and pheochromocytoma, phenotype MEN 2B shows multiple ganglioneuromas of the gastrointestinal tract, mucosal neuromas, and marfanoid skeletal abnormalities (13, 14). All forms are transmitted as an autosomal dominant pattern with a high degree of penetrance but variable clinical expression. Heterozygous point mutations convert RET into a dominant transforming gene with oncogenic activity (15). In contrast, inactivating RET point mutations with a loss of function effect have been identified in familial and sporadic forms of Hirschsprung disease (HSCR), a congenital absence of parasympathetic innervation in the lower intestinal tract that may occur sporadically or inherited as an incompletely penetrant, autosomal dominant trait (16, 17).
In MEN 2A and FMTC, the majority of germline RET mutations are detected in one of the cysteines encoded by exons 10 or 11; in FMTC less often do they also involve specific codons in the exons 8, 13, 14, and 15 (10, 18, 19, 20).
To evaluate the frequency of RET polymorphisms in the general population, we performed a genetic screening of exons 1016 by direct sequencing and restriction analysis on a large panel of 161 unrelated subjects comprising healthy individuals and patients with miscellaneous thyroid disorders without personal and familial history or MEN 2A, MEN 2B, and MTC. Unexpectedly, we found a novel germline missense mutation in the exon 11 of RET gene (CGG
CAG) corresponding to R694Q substitution in the juxtamembrane RET domain in a 23-yr-old woman with atrophic autoimmune thyroiditis and primary ovarian failure. This mutation was not detected in any other unrelated subjects evaluated and was associated with two already described polymorphisms in exons 11 (G691S) (21) and 15 (S904S) (22). This is the first report of a RET point mutation without clinical and biochemical evidence of MTC, MEN 2, or HSCR.
| Patients and Methods |
|---|
|
|
|---|
A total of 161 unrelated (105 females and 56 males) normal subjects or patients with miscellaneous thyroid disease without any evidence in the personal and familial clinical history of MEN 2A/2B and FMTC have been investigated for germline mutations in the RET protooncogene as detailed below. Before undergoing genetic testing, informed consent was obtained from all subjects.
Mutational analysis of the RET protooncogene
DNA extraction and PCR amplification. High molecular weight DNA was isolated from peripheral blood leukocytes by QIAamp blood kit (Qiagen, Hilden, Germany) according to the manufacturers protocol. Nucleotide sequences of exons 10 through 16, including corresponding splice junction regions of RET gene, were amplified using PCR conditions already described and primers designed from flanking intronic sequences according to the published sequences (23). Contamination problems were ruled out by including PCR control samples with no DNA as a template. In addition, all positive genetic analyses were repeated two times on samples collected on two different occasions. After PCR, the amplicon size was resolved by electrophoresis in a 2.0% agarose gel stained with ethidium bromide.
DNA sequencing. PCR products from the proband and her family members were directly sequenced in both orientations by standard protocols using the Sanger method with the same primers as for PCR. Additionally, two sets of primer sequences were designed for exons 15 and 16 (23). After purification, the reactions were run on an automated sequencer (ABI Prism 9600 DNA Analyzer; Applied Biosystems, Foster City, CA).
Restriction analysis. DNA sequence change in affected individuals was independently confirmed by restriction-enzyme analysis of exon 11 PCR product for the presence of R694Q mutation according to the manufacturers instructions. An abnormality on exon 11 was demonstrated to be absent as a common polymorphism in DNA obtained from 161 unrelated, normal individuals. The resulting fragments were separated by electrophoresis on an agarose gel of 2%, and DNA bands were visualized by ethidium bromide staining. To exclude possible PCR errors, confirmation was performed twice on PCR-amplified DNA obtained independently of those used for direct sequencing. Appropriate negative controls were included in each restriction analysis.
Protein studies. Anti-RET polyclonal rabbit antibodies, anti-RET(TK), directed against the TK domain of RET (amino acids 738-1058), were affinity-purified by sequential chromatography on RET and glutathione-S-transferase-coupled agarose columns. Their features and specificity have been previously characterized (15). Anti-RETpY1062 and anti-RETpY905 are affinity-purified polyclonal antibodies raised against RET peptides containing phosphorylated Y1062 or Y905 residues (24). Monoclonal antitubulin antibodies were obtained from Sigma Chemical Co. (St. Louis, MO). Secondary antibodies coupled to horseradish peroxidase were purchased from Amersham Pharmacia Biotech (Little Chalfont, Buckinghamshire, UK). Immunoprecipitation and immunoblotting were performed according to standard procedures, as previously described (24).
Molecular constructs. All the constructs used in this study encode the short (RET-9) RET spliced form and were cloned in pCDNA3(Myc-His) (Invitrogen, Groningen, The Netherlands). The wild-type RET, RET/C634R, and RET/C620Y constructs have been described previously (15, 25). RET/R694Q was generated by site-directed mutagenesis using the QuickChange mutagenesis kit (Stratagene, La Jolla, CA). The mutation was confirmed by DNA sequencing.
Cell culture and transfection experiments. NIH3T3 fibroblasts were grown in DMEM (Life Technologies, Inc., Paisley, PA) containing 5% calf serum (Life Technologies, Inc.) and were transfected (15). Transformed foci were scored at 3 wk. Transforming efficiency was calculated in focus-forming units per picomole of added DNA after normalization for the efficiency of colony formation in parallel dishes subjected to marker selection in G418 (neomycin). Soft agar colony assay was performed as reported (15); colonies were scored at 2 wk.
Hormonal and other serum assays
Serum calcitonin (CT) levels were measured by immunoradiometric assay, using the ELISA-hCT kit (DRG Diagnostics, Mountainside, NJ), with a normal range of 015 ng/liter. This assay uses two monoclonal antibodies directed against different regions of the monomeric CT. The pentagastrin (Pg) stimulation test consisted of an intravenous injection of 0.5 µg/kg body weight Pg (Pentagastrin Injection BP; Cambridge Laboratories, Northwich, UK). Blood samples were collected before (10, 0 min) and after (1, 2, 3, 4, 5, and 10 min) Pg infusion. The CT response was expressed as the maximal CT peak after initiation of the Pg injection and was considered normal when it was less than 50 ng/liter in women and less than 80 ng/liter in men.
Serum TSH was measured by a chemiluminescent method (Ortho-Clinical Diagnostic, Amersham Pharmacia Biotech) with normal values ranging from 0.3 to 3 mIU/liter. Free T3 and free T4 were measured by means of a chromatographic method based on separation of free T4 on Lisophase columns (Technogenetics, Milan, Italy; normal values: free T4, 8.520.6 pmol/liter; free T3, 4.38.6 pmol/liter). Antithyroperoxidase antibody by RIA (Biocode, Liège, Belgium; normal values, <20 IU/ml); thyroid microsomal antibodies and antithyroglobulin antibodies (anti-Tg) were assayed by passive agglutination (SERODIA-M and SERODIA-Tg, respectively; Fujirebio Inc. Pharmaceutical, Tokyo, Japan). Intact PTH was measured by a commercial immunoradiometric assay (DPC, Los Angeles, CA). Serum carcinoembryonic antigen (CEA) was measured by a commercial assay (CEA Elecsys; Roche Diagnostics, Mannheim, Germany); serum calcium and phosphate were measured by standard methods (Roche Diagnostics); and urinary and plasma catecholamines and their metabolites were measured by HPLC (Chromsystems & Chemicals GmBH, München, Germany).
Adrenal and ovary antibodies were kindly measured by Dr. Corrado Betterle (Division of Endocrinology, Department of Medicine, University of Padua, Padua, Italy), using indirect immunofluorescence as previously described (26).
| Results |
|---|
|
|
|---|
Direct sequencing of the PCR products of the proband revealed a new single heterozygous CGG
CAG transition located at position 2081 of the coding sequence in exon 11 of the RET gene. This missense mutation predicts that a nonconservative amino acid change at codon 694 replaces a conserved arginine with glutamine (R694Q) in the juxtamembrane RET domain (Fig. 1
). R694Q transition was absent in the DNA sequence from 161 normal individuals and remaining family members tested under the same experimental conditions. Furthermore, two previously described single-nucleotide polymorphisms of RET gene were detected at position in exon 11 (GGT
AGT) and in exon 15 (TCC
TCG) of the coding sequence, corresponding respectively to a missense mutation G691S and a silent mutation S904S. Screening of exons 10, 13, 14, 15, and 16 by direct sequencing failed to reveal further RET point mutations. In addition to R694Q mutation, the same heterozygous polymorphisms were also detected in the probands mother. Genomic DNA from the remaining three family members was sequenced, and only the wild-type sequences were detected.
|
CAG missense mutation was further confirmed by HaeIII restriction analysis, which showed a partial digestion of PCR products caused by a loss of one of five HaeIII restriction sites in the mutated sequence. In the normal control, restriction analysis revealed complete digestion of both alleles with fragments of 195-, 118-, 38-, and two lower molecular weight bands of 19- and 10-bp, indicating the homozygous state for the normal allele. In the proband, an additional undigested 156-bp band, deriving from the mutated allele, was detectable when compared with a normal sample, indicating that the mutation eliminated one of five HaeIII restriction sites from the 418-bp PCR product and was present in a heterozygous state. Clinical and biochemical evaluation of the proband and study of first-degree relatives
The index case with positive genetic screening underwent a complete clinical examination, laboratory tests, and extensive diagnostic imaging investigation. Biochemical evaluation is documented in Table 1
. She presented a primary hypothyroidism due to atrophic Hashimotos thyroiditis and primary ovarian failure with normal karyotype and no mutations at the FOXL2 gene, which has recently been found to be involved in the pathogenesis of rare forms of gonadal dysgenesis (27). Although no evidence of serum adrenal and ovary autoantibodies was found, the association of primary ovarian failure with autoimmune thyroiditis, both in the index case and in the sister (see below), suggested a diagnosis of autoimmune ovaritis in a context of a potentially incomplete autoimmune polyglandular syndrome type 2 (28). Serum CT concentration was very low and displayed minimal increase after Pg stimulation. Thyroid echography showed a small (2 ml volume) hypoechoic gland with no evidence of associated nodules. Samples subsequently obtained from her parents, one brother, and one sister demonstrated germline transmission of this mutation. Genetic analysis showed that the same abnormality was inherited from the probands mother, a 52-yr-old woman who 2 yr before was submitted to total thyroidectomy for goitrous Hashimotos thyroiditis. Histological revision of the paraffin-embedded thyroid tissue and further immunohistochemical studies documented normal C cell number and morphology (Fig. 2
). The three remaining siblings did not carry any RET variant. Although the father and the brother were apparently healthy subjects, the probands sister, a 16-yr-old woman, had clinical features very similar to the index case: subclinical hypothyroidism due to Hashimotos thyroiditis associated with premature ovarian failure developed 1 yr after menarche. No evidence of MTC (as assessed by basal serum CT), hyperparathyroidism (by serum calcium, phosphate, PTH levels, and urinary calcium levels in the normal range), pheochromocytoma (by plasma and urinary catecholamines), or other associated clinical features attributed to MEN 2A/2B or HSCR was found in any family members.
|
|
We introduced the Arg 694 to Gln (R694Q) mutation into an expression vector carrying the wild-type RET sequence. The structure of the normal RET protein and the position of the mutations tested in this study are represented in Fig. 3A
. To verify whether the R694Q mutation had a gain-of-function effect and was able to convert RET into a dominantly transforming oncogene similarly to MEN 2/FMTC-associated RET point mutations, a focus-forming assay on NIH3T3 cells was performed. Wild-type RET, RET/R694Q, RET/C634R (a strong RET oncogene associated with MEN 2A), and RET/C620Y (a less strong RET oncogene associated with FMTC) (25) were transfected, and transformed foci were calculated as the average of three independent experiments performed in duplicate. As previously reported, RET/C634R and RET/C620Y were able to induce the formation of transformed foci, with RET/C634R being significantly more active than RET/C620Y. In contrast, RET/R694Q, similar to wild-type RET, had no detectable transforming ability (Fig. 3C
).
|
The expression levels of the various RET constructs was evaluated in the selected mass populations. RET/R694Q protein products were correctly synthesized as Mr 145,000 and 160,000 isoforms. The Mr 160,000 form represents a mature glycosylated protein present on the cell surface, whereas the Mr 145,000 form is an immature precursor (29). Of note, an altered maturation of RET products has been described as the consequence of loss-of-function RET mutations found in congenital megacolon or HSCR (29).
Oncogenic activation of RET results in the constitutive activation of the kinase which, in turn, causes autophosphorylation, recruitment of intracellular substrates, and activation of diverse signaling pathways (30). Upon phosphorylation, tyrosine 1062 behaves as a multiple-effector docking site, recruiting growth factor receptor-bound protein 2-son of sevenless (Grb2-Sos) complexes leading to Ras/MAPK activation. Tyrosine 905, in the catalytic core, stabilizes the active conformation of the kinase. Thus, in vivo tyrosine phosphorylation levels of the different mutant proteins were assessed, as a "read-out" of their activation status, by immunoblot with phospho-RET antibodies specific for phosphorylated forms of Y905 and Y1062. The phosphorylation levels of RET/R694Q were very weak and indistinguishable from those of wild-type RET. Indeed, despite similar expression levels, RET/R694Q scored significantly (
7-fold) less phosphorylated than the constitutively active RET/C634R (Fig. 3B
). An in vitro immunocomplex kinase assay confirmed these findings.
Together these results demonstrated that the R694Q mutation displayed no detectable effects on the correct synthesis and maturation of the RET protein. In addition, the R694Q mutation did not confer detectable oncogenic activity to RET.
| Discussion |
|---|
|
|
|---|
In exon 11, in addition to the mutations at cysteines 630 and 634, only two insertions and two missense mutations (at codons 631 and 640) have been documented (34, 35, 36, 37).
We report here a previously undescribed germ line Arg
Gln mutation at codon 694 in the juxtamembrane domain of RET protooncogene from a patient with Hashimotos thyroiditis associated with primary ovarian failure. The study of first-degree relatives easily excluded that the new RET mutation could be responsible for the peculiar endocrine dysfunctions present in the index case. A milder, similar phenotype was in fact present in the sister who tested negative for RET mutations, whereas the mother, in whom the mutation was present, displayed only Hashimotos thyroiditis in the absence of any ovarian dysfunction. Although no further genetic studies were carried out, we believe that the frequent endocrine abnormalities encountered in this family may be explained on the basis of a potential autoimmune polyglandular syndrome type 2 (28). Furthermore, similar to the proband, the mother did not show any biochemical or clinical evidence of MTC, parathyroid disease, pheochromocytoma, or other associated clinical features attributed to MEN 2A/2B or HSCR. This nonconservative mutation has not been detected either in remaining family members or in a group of 161 control individuals, indicating that is not a common variant of the RET protooncogene.
All germline point mutations responsible for FMTC and MEN 2 syndromes have a gain-of-function effect (17). To determine whether the R694Q mutation is able to convert RET into a dominant transforming gene with oncogenic activity, transfection studies of mutated RET protein have been performed. We evaluated in vitro the transforming capacity of the mutated RET protein only at codon R694Q, comparing it with a C634R mutation, typically a MEN 2A-associated RET mutant. In keeping with clinical data, the novel mutation did not exert significant potentiating effects on RET kinase or transforming activity. Instead, the lack of pathological transformational mechanism could introduce the concept of a "reduced activating potential" already applied to the FMTC mutations affecting the RET TK domain (38) or, alternatively, suggest the hypothesis of a probable benign polymorphism present at low frequency (<1%).
Several reports documented the identical mutations in the RET protooncogene brought to different phenotypes, suggesting the possible presence of "modifier" genes that can determine the expression of the RET mutation. In the two subjects with R694S mutation, but not in the remaining family members, we found two already documented heterozygous polymorphisms in exons 11 (G691S) (21) and 15 (S904S) (22). The G691S is a conservative amino acid substitution without an oncogenic effect that occurs close to R694Q in the RET extracellular domain. In contrast, no amino acid alteration is observed in the conserved polymorphism S904S.
Recently, Gil et al. (39) observed a strong cosegregation between G691S and S904S polymorphisms and documented that G691S /S904S haplotype may influence the age of onset in MEN 2A patients. Further studies in HSCR patients (40) suggest that these polymorphisms may show a modifier effect. In our case, the observed association of G691S and S904S polymorphisms with R694Q mutation in the affected patients could suggest a possible functional interaction. New transfection studies are needed to understand whether the observed polymorphisms may influence RET expression and act as modifier genes.
In conclusion, we detected a new case of a RET mutation in two related cases without clinical and biochemical evidence of FMTC, MEN 2, and HSCR. The serendipitous discovery of this mutation underscores some important problems that may be encountered in genomic studies on RET.
| Acknowledgments |
|---|
| Footnotes |
|---|
Abbreviations: CEA, Carcinoembryonic antigen; CT, calcitonin; FMTC, familial medullary thyroid carcinoma; HSCR, Hirschprung disease; MEN 2A, multiple endocrine neoplasia type 2A; MEN 2B, MEN type 2B; Pg, pentagastrin; Tg, thyroglobulin; TK, tyrosine kinase.
Received February 27, 2004.
Accepted June 22, 2004.
| References |
|---|
|
|
|---|
, a novel receptor for GDNF. Cell 85:11131124[CrossRef][Medline]
Tyr mutation of the RET proto-oncogene in a pedigree with multiple endocrine neoplasia type 2A and localized cutaneous lichen amyloidosis. J Endocrinol Invest 17:201204[Medline]
1 and GDNF genes in Spanish families with multiple endocrine neoplasia type 2A. Int J Cancer 99:299304[CrossRef][Medline]
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |