prevalences of gsα, ras, p53 mutations and ret/ptc rearrangement in differentiated thyroid tumours...

Clinical Endocrinology (1998) 49, 317–323

317q 1998 Blackwell Science Ltd

Prevalences of G sa, ras, p53 mutations and ret/PTCrearrangement in differentiated thyroid tumours in aKorean population

K. Y. Park§, J. M. Koh*, Y. I. Kim*, H. J. Park§,G. Gong†, S. J. Hong‡ and I. M. Ahn**Division of Endocrinology and Metabolism, Departmentof Internal Medicine, †Department of DiagnosticPathology, ‡Department of General Surgery, AsanMedical Centre, University of Ulsan College of Medicine,and §Asan Institute for Life Sciences and Technology,Seoul, Korea

(Received 26 August 1997; returned for revision 28 October1997; finally revised 4 December 1997; accepted 6 March 1998)

Summary

OBJECTIVE In thyroid tumours, ras, G sa, p53 muta-tions and ret/PTC rearrangement have been reportedwith variable prevalences in different geographicalregions. We studied the prevalence of these muta-tions and rearrangement in thyroid tumours in aKorean population. As MDM2 and Bcl-1 protein exp-ressions have been suggested to be associated withp53 protein, we also studied possible relationshipsamong them.PATIENTS AND DESIGN Eleven cases of adenoma-tous goitre, eight cases of follicular adenoma, fivecases of follicular carcinoma and 37 cases of papillarycarcinoma were included in this study. To find muta-tions and rearrangement, RT–PCR, SSCP and/ordirect sequencing, after subcloning if necessary,were used, and immunohistochemical stainingswere performed for p53, MDM2 and Bcl-2 proteins incases of papillary carcinoma.RESULTS We could not find any rearrangement forret/PTC-1, -2, -3 and mutation of G sa. For the rasoncogene, K and H-ras mutations were not found,but N-ras mutations, point mutation of CAA to CGAin codon 61, were detected in one follicular adenoma(12·5%, 1/8) and one follicular carcinoma (33%, 1/3).p53 mutations were detected in only one case ofpapillary carcinoma (3%, 1/31: exon 8, codon 266

GGA → GAA). In 30 cases of papillary carcinoma with-out p53 mutation, the prevalences of positive immuno-histochemical staining were 13·3% for p53 protein,53·3% for MDM2 protein and 56·7% for Bcl-2 protein.While over-expression of p53 protein was not sig-nificantly related to that of MDM2 and Bcl-2 proteins,over-expression of MDM2 and Bcl-2 in papillarycarcinoma were associated.CONCLUSION ret/PTC rearrangement, G sa, ras andp53 mutations are relatively rare in differentiatedthyroid neoplasms from a Korean population, whichmay reflect genetic and environmental differencesfrom patients in countries with high prevalences.P53 protein over-expression was noted in 13·3% ofpapillary carcinoma cases without p53 mutation andwas not significantly related to MDM2 and Bcl-2expression.

Recent advances in thyroid tumorigenesis have assisted under-standing of the appearance of different functions andhistologies originated from thyroid follicular cells (Fagin,1992). TSH-receptor (R) and Gsa mutations are the causes of asubstantial proportion of autonomous functioning thyroidadenomas. Gsa and ras mutations are involved in thedevelopment of follicular adenomas and carcinomas. ret/PTCand NTRK1 rearrangements are seen uniquely in papillarycarcinomas. p53 mutations are found frequently in anaplasticcarcinomas. Genetic alterations in proto-oncogenes and tumoursuppressor genes are also likely to be related to particular stepsin tumorigenesis (Saidet al., 1994). Mutations of ras and Gsaappear to be early events for they are often seen in both benignand malignant tumours (Faridet al., 1995). Also, ret/PTCrearrangements are frequently seen in occult papillary carcino-mas (Viglietto et al., 1995). Mutations of p53 are highlyprevalent in anaplastic carcinomas and may represent thetransitional step in the development of these aggressivetumours (Itoet al. 1992). The prevalences of Gsa, ras, p53mutations and ret/PTC rearrangements have been reported tovary widely in different geographic areas.

Malignant thyroid tumours are classified according to theirhistological patterns: papillary, follicular and anaplastic, eachhaving unique biological behaviours. Within these subgroups,however, considerable variations in terms of growth and

Correspondence: Il-Min Ahn, Division of Endocrinologyand Metabolism, Department of Internal Medicine, Asan MedicalCentre, Poong-Nap Dong, Song-Pa Ku, Seoul 138-040, Korea.Fax:þ82 2 224 6962.

aggressiveness have been observed. For papillary thyroidcarcinoma, the most common thyroid carcinoma, pathologicalcriteria (Burmanet al., 1996) such as insular, tall cell, columnarcell and diffuse sclerosing variant, have been used to predictthose patients with poor prognosis, but the proportion of thevariants is too small to cover the patient group with poorprognosis. Thus, tumour markers that can predict differentclinical courses for these tumours, most of which have a goodlong-term prognosis, would be very useful in guiding futuretherapy.

In an attempt to understand the molecular basis of thyroidneoplasms in Korea, we screened for the presence of Gsa, rasand p53 mutations and ret/PTC rearrangement in variousthyroid tumours. Also, immunohistochemical stainings wereperformed for p53, MDM2 and Bcl-2 proteins in cases ofpapillary carcinoma, as MDM2 and Bcl-2 proteins have beensuggested to be associated with p53 protein (Jenningset al.,1995; Pollinaet al., 1996).

Materials and methods

All tissues were obtained during thyroid surgery from AsanMedical Center (AMC) in Seoul, Korea from October 1995 toApril 1996. Tissues were frozen immediately in liquid nitrogenand stored at¹708C until processed. All tumours wereclassified by strict pathological criteria and variants, such astall cell and insular type papillary carcinoma, were not includedin this study. The clinical staging was performed as describedby DeGrootet al. (1990). Sixty-one samples were analysed formutations of H-ras, N-ras and Gsa, 56 samples for K-ras and 59for p53 gene, including 11 cases of adenomatous goitre, eightcases of follicular adenoma (10 cases for p53), 37 cases ofpapillary carcinoma (32 cases for K-ras, 31 for p53), three casesof follicular carcinoma (5 cases for p53) and two cases of

anaplastic carcinoma (Table 1). Five normal thyroid tissueswere used as controls. ARO, WRO, NPA cell lines (kindlyprovided by Dr Nagataki, Nagasaki University, Japan) contain-ing p53 mutations of exon 6 and 8, and total DNA samples withras, Gsa mutations and p53 mutation of exon 5 and 7 wereobtained from the Division of Haematooncology, AMC aspositive controls. Twenty-four samples of papillary carcinomawere analysed for ret/PTC rearrangement by RT-PCR. TPC-1cell line (provided by Dr Nagataki, Nakasaki University, Japan)containing the ret/PTC-1 rearrangement and the plasmidscontaining the ret/PTC-2, -3 rearrangement (provided by DrJhiang, Ohio State University, USA) were used to set conditionsfor RT–PCR and served as positive controls. Thirty samples ofpapillary carcinoma without p53 mutations were immuno-histochemically stained for p53, MDM2 and Bcl-2 proteins.

DNA extraction, SSCP and sequencing for mutations ofGsa, ras oncogenes and p53 tumour suppressor gene

Genomic DNAs were extracted from frozen thyroid tissues asdescribed by Kaufmanet al. (1995). To screen for mutations ofGsa gene, primers which cover exons 8 and 9 were used. Exons1 and 2 of H-ras, N-ras, K-ras genes and exons 5–8 of p53 gene,where most mutations of p53 gene were reported, werescreened (Table 2). PCR amplification was performed underfollowing conditions: 3 min initial denaturation at 948C wasfollowed by 30 cycles of 45 s denaturation at 948C, 1 min

318 K. Y. Park et al.

q 1998 Blackwell Science Ltd,Clinical Endocrinology, 49, 317–323

Table 1 Results and number of mutation and rearrangementscreenings

MNG FA FTC PTC ATC Normal

ret/PTC [24]K-ras 11 8 3 32 0(2) 0(5)H-ras 11 8 3 1(37) 0(2) 0(5)N-ras 11 8* 3* 3(37) 0(2) 0(5)Gsa 11 8 3 1(37) 0(2) 0(5)p53 0(11) 0(10) 0(5) 9(31)* 1(2)* 0(5)

* mutation or rearrangement detected, sequencing, (SSCP), [RT–PCR];one case each.MNG, multinodular goitre; FA, follicular adenoma; FTC, follicularthyroid carcinoma; PTC, papillary; thyroid carcinoma; ATC, anaplasticthyroid carcinoma.

Table 2 Primers used in the PCR of Gsa. H, N, K-ras oncogenes andp53 tumour suppressor gene

Gsa 8, 9 50-GTGATCAAGCAGGCTGACTATGTG-30

50-GCTGCTGGCCACCACGAAGATGAT-30 (538 bp)H-ras: 1 50-CAGGCCCCTG AGGAGCGATG-30

50-TTCGTCCACAAAATGGTTCT-30 (120 bp)2 50-TCCTGCAGGATTCCTACCGG-30

50-GGTTCACCTGTACTGGTGGA-30 (194 bp)N-ras: 1 50-GACTGAGTACAAACTGGTGG-30

50-GGGCCTCACCTCTATGGTG-30 (118 bp)2 50-GGTGAAACCTGTTTGTTGGA-30

50-ATACACAGAGGAAGCCTTCG-30 (102 bp)K-ras: 1 50-GGCCTGCTGAAAATGACTGA-30

50-GTCCTGCACCAGTAATATGC-30 (162 bp)2 50-TTCCTACAGGAAGCAAGTAG-30

50-CACAAAGAAAGCCCTCCCCA-30 (128 bp)p53: 5 50-TTCCTCTTCCTGCCAGTAC-30

50-GCCACACCTAAGAGCAAT-30 (287 bp)6 50-TGCTCAGATAGCGATGGT-30

50-AGTTGCAAACCAGACCTC-30 (210 bp)7 50-GTGTTGTCTCCTAGTTG-30

50-AGTATGGAAGAAATCGGTAA-30 (210 bp)8 50-CCTATCCTGAGTAGTGGT-30

50-GTCCTGCTTGCTTACCTC-30 (158 bp)

annealing at 568C (K-ras), 588C (p53) or 608C (N, H-ras) and1 min extension at 728C, with 10 min final extension. Single-stranded conformation polymorphism (SSCP) analysis wasperformed with [a-35S]dATP as described by Oritaet al.(1989). In performing PCR–SSCP for p53, primers weredesigned with or without GC clamp, but the results were notdifferent. The reaction mixture was initially analysed bypolyacrylamide gel and MDE gel (Mutation Detection Enhance-ment) (FMC, USA) at 6–8 watts constant power for 14 h at 48Cand room temperature in the screening of p53 mutation. OnlyMDE gel was used later at room temperature for the screening ofother oncogene mutation as increased sensitivity, not specificity,was noticed in these conditions. The DNAs were sequenced bydirect sequencing technique, using Sequenase PCR productsequencing kit (USB, USA). Six of 10 PCR products for p53were subcloned into TA cloning vector (Invitrogen, USA) andthen sequenced to obtain better resolution.

RNA extraction and RT-PCR for ret/PTC rearrangement

Our preliminary study did not show any case with ret/PTCrearrangement by Southern blot analysis, suggesting that suchrearrangement in papillary carcinoma from a Korean popula-tion might be rare. We therefore performed more sensitive RT–PCR with the 5 primer sets for ret/PTC-1, -2, -3 rearrangementsdetection, designed by Sugget al. (1996) with house-keepinggenes,pgk and GAPDH as internal controls. Total RNA wasextracted from 80–100 mg of frozen thyroid tissues or TPC-1cells by the single-step guanidine isothiocyanate method. TotalRNA (1mg) was reverse transcribed by Superscript II Rnase H-Reverse Transcriptase (Gibco, BRL) using 50 ng of randomprimer. Ten per cent of cDNA (2ml) or 0·1 ng of ret/PTC-2 orret/PTC-3 was used as a template for amplification with a pairof primers. PCR was performed under the following condition:4 min initial denaturation at 958C was followed by 25 cycles of30 s denaturation at 958C, 30 s annealing at 578C and 2 minextension at 728C with 5 min final extension. PCR conditionwas set after trials of different cycles (25–35) and conditions.PCR products were analysed using 1·4% agarose gel electro-phoresis. After obtaining negative results, Southern hybridiza-tion for enhanced detection of PCR product was performed in20 of 24 cases of papillary carcinoma with the probes, fromRT–PCR with [a-32P]dCTP using TPC-1 cells or plasmidscontaining ret/PTC-2, -3 rearrangement as templates.

Immunohistochemistry for P53, MDM2 and Bcl-2proteins

Immunostaining of papillary carcinomas without p53 mutationswas performed for p53, MDM2 and Bcl-2 proteins, essentiallyas described elsewhere on formalin-fixed paraffin-embedded

tissue sections (Hsuet al., 1981). Briefly, the slides weredewaxed, rehydrated, treated with 3% hydrogen peroxide indistilled water to inhibit endogenous peroxidase activity, andthen immersed in boiling citrate buffer, pH 6, in a microwaveoven for 5 min. After extensive washings in distilled water andPBS, primary antibodies; (1) monoclonal mouse anti-humanp53 protein, clone DO-7 (DAKO, USA); (2) monoclonal mouseanti-human MDM2 protein (Novocastra, UK); (3) monoclonalmouse anti-human Bcl-2 oncoprotein, clone 124 (DAKO, USA)were applied. Reaction time was for 1 h at room temperatureand samples were stained with LSAB kit (DAKO, USA). Fivenormal tissues were immunostained at the same time asnegative controls. After the strength of immunostaining ontested samples was compared with that of normal tissues,positive immunostaining for p53 protein was defined as 30%more staining of nuclei, for MDM2 protein, 30% more stainingof nuclei and Bcl-2 protein, 25% more of cytoplasms, scored bytwo independent pathologists.

Analytical methods

Pearson correlation coefficients and Student’st-test were usedto examine associations between variables.P value less than0·05 (two-tailed) was considered to be statistically significant.

Results

Mutations of Gsa and ras oncogenes

The primers used to detect activating mutation of Gsa coveredthe two known hot spots in exon 8 (codon 201) and exon 9(codon 227). The primers employed to detect mutation of eachras were designed to cover codons 12, 13 (exon 1) and 59, 61(exon 2). SSCP was performed for the samples of papillarycarcinoma and anaplastic carcinoma. Five DNA samples ofpapillary carcinoma which showed a migration pattern differentfrom normal control were sequenced directly but did notrevealed any activating mutation. DNA samples from 11 casesof adenomatous goitre, eight cases of follicular adenoma andthree cases of follicular carcinoma were directly sequenced. Asthere was a local report which showed 90% of K-ras mutation inpapillary carcinoma by paired PCR method (Songet al., 1993),K-ras of the 32 cases were included in the direct sequencinggroup (Table 1). Only two cases of point mutations of N-raswith same substitution at codon 61, where CAA was replacedby CGA (Fig. 1), were found in one follicular adenoma (12·5%,1/8) and one follicular carcinoma (33%, 1/3).

Mutations of p53 tumour suppressor gene

As p53 mutations were reported mainly in DNA binding

Genetic abnormalities in thyroid tumours in Korea 319


regions over exons 5–8, primers were designed to cover theseexons. DNAs from 11 cases of adenomatous goitre, 10 cases offollicular adenoma, five cases of follicular carcinoma, 31 casesof papillary carcinoma and two cases of anaplastic carcinomawere analysed by SSCP. Four DNA samples which migrateddifferently from normal control were sequenced directly, andsix samples sequenced after subcloning to visualize mutation

clearly because of the compressions or high baseline activities.Two point mutations were detected from the above samples,one case of papillary carcinoma (3%, stage III, 1/31, exon 8,codon 266 GGA→ GAA) (Fig. 2) and another from anaplasticcarcinoma (50%, 1/2, exon 7, codon 239 AAC→ AGT).Although the gel run was adjusted to find mutations in exons,two samples which migrated differently from normalcontrols were noticed to have base substitutions in intronicsequences.

ret/PTC rearrangement

Our preliminary study to detect ret/PTC rearrangement bySouthern blot analysis on 37 cases of papillary carcinoma didnot reveal any case of the rearrangement (data not shown).Therefore, the more sensitive RT–PCR method was applied.TPC-1 cell line with ret/PTC-1 rearrangement and plasmids,containing ret/PTC-2, -3 rearrangements were used as positivecontrols and to set conditions of RT–PCR after sequencing. Wewere unable to identify any ret/PTC-1, -2, -3 rearrangement in24 cases of papillary carcinoma by RT-PCR using five differentsets of primers, and even after Southern hybridization forenhanced detection of PCR product.

Immunohistochemistry for P53, MDM2 and Bcl-2proteins

One case of papillary carcinoma with p53 mutation showedstrong staining for p53 only. Thirty samples of papillarycarcinoma without p53 mutation included five cases of stage I,two cases of stage II and 23 cases of stage III papillarycarcinoma. Strong immunostaining for p53 was noticed in foursamples (13·3%), in 16 samples for MDM2 (53·3%) and in 17samples for Bcl-2 (56·7%), respectively. We could find nocorrelation between staining pattern for p53, MDM2 and Bcl-2proteins with clinical staging, tumour size (<1, 1 to 4,>4 cm)and age (<45,>45). While over-expression of p53 protein wasnot significantly related to that of MDM2 and Bcl-2 proteins,over-expression of MDM2 and Bcl-2 proteins were associated(P<0·05) (Table 3).



Fig. 1 Direct sequencing of DNA amplified for N-ras oncogene froma follicular adenoma showing a point mutation of CAA for CGA atcodon 61.

Fig. 2 Sequencing of DNA after subcloning for p53 gene from apapillary carcinoma showing a point mutation, exon 8, codon 266,GGA→ GTA.

Table 3 Interrelationship of p53, MDM2 and Bcl-2 protein over-expression in papillary carcinomas without p53 mutations

Bcl-2 P53 P53MDM2 (þ) (¹) MDM2 (þ) (¹) Bcl-2 (þ) (¹)

(þ) 11 5 (þ) 3 13 (þ) 1 16(¹) 6 8 (¹) 1 13 (¹) 3 10

Discussion

With the advent of molecular biological techniques, the geneticbasis of thyroid tumorigenesis has become clearer. However,the prevalences of the genetic alterations have been reported tovary widely according to geographical areas. Studies on theprevalence of Gsa mutation in thyroid neoplasms haveestimated its occurrence to be roughly 0–15% in follicularadenoma and papillary carcinoma (Yoshimotoet al., 1993;Spambalget al., 1996) and 0–43% in follicular carcinoma(Spambalget al., 1996; Gorelovet al., 1996). For ras, a higherprevalence of mutation was noticed in iodine deficient areas,compared to iodine sufficient areas (follicular adenoma: 85%vs. 17%, follicular carcinoma: 50%vs. 10%) (Shiet al., 1991).The reports of prevalence studies of ras mutations in papillaryand anaplastic carcinomas vary widely from none to 56% (Saidet al., 1994; Shiet al., 1991). In the thyroid, p53 mutations havebeen implicated as a transitional step from well differentiatedthyroid carcinoma (WDTC) to anaplastic carcinoma (p53mutation: 20–86%) (Itoet al., 1992; Zou et al., 1993).Prevalence of p53 mutations in WDTC has ranged from none(Ito et al., 1992) to 25% (11/44) (Zouet al., 1993). Thediscoveries of ret/PTC-2 and -3 rearrangements were maderecently, and most clinical series have reported only theprevalence of the ret/PTC-1 rearrangement. When all the resultswere combined, three versions of ret/PTC rearrangement havebeen described in a variable proportion of papillary thyroidcarcinoma, ranging from 2·5 to 34·5%, depending on thegeographical areas investigated and on the sensitivity of thedetection method used (Sugget al., 1996).

Our study on Gsa, ras and p53 mutations and ret/PTCrearrangement in differentiated thyroid neoplasms from aKorean population, whose ethnic composition is uniform andliving in an iodine sufficient area, showed rare occurrences ofmutations of the oncogenes and tumour suppressor gene. No K-ras, H-ras or Gsa mutations were detected. Only two of 61 caseshad N-ras mutations (one from follicular adenoma and the otherfrom follicular carcinoma) and two of 59 cases harboured p53mutations (one from papillary carcinoma, the other fromanaplastic carcinoma) from the same samples. No ret/PTCrearrangement was seen in 24 overlapping cases of papillarycarcinoma (Table 1).

There are certain points for caution in this study. One is theinherent limitation of SSCP in detecting mutations in spite ofseveral measures such as different temperature setting, primerswith GC clamp, and choosing each samples which showed evena slightly different migration pattern, which were applied toincrease sensitivity. The other concerns p53 mutations on exons5–8, since mutations may have been clustered around thesesites just because many studies only examined these exons. Ithas been known the hot spots are related to DNA binding

regions of p53 protein (Friend, 1994; Prives, 1994), such as themutations detected in this study, which also makes them to begermline mutations a remote possibility.

This result reconfirms previous suggestion of markedgeographical variation in thyroid tumorigenesis (Faridet al.,1994). Other examples of geographical differences are seen inTSH-R and Gsa mutations. In autonomously functioningthyroid adenomas the frequencies of TSH-R mutations were63·6% in European countries compared with 0% in Japan(Takeshita et al., 1995), and those of Gsa mutation 25–40% and4% (Tanakaet al., 1996). Differences in the prevalence ofoncogenes, tumour suppressor gene mutations and ret/PTCrearrangements in differentiated thyroid neoplasms have beenattributed in part to environmental factors such as regionaldifferences in iodide supply (Krameret al., 1989) and radiationexposure in the case of ret/PTC rearrangements (Klugbaueret al., 1995; Fugazzolaet al., 1996). Given the dependency ofthyroid function on iodide trapping, it would be reasonable tospeculate that the genes which are linked to thyroid hormogen-esis should be activated in case of iodine deficiency. Radiationexposure, external beam irradiation and internal radiation,especially of radioiodine which has been estimated asresponsible for 60% of the radiation dose to the thyroid in theChernobyl accident (Beckeret al., 1996), have been linked tothe molecular mechanisms of thyroid tumorigenesis.

Our results showed 13·3% positive immunohistochemicalstaining for p53 protein in papillary carcinomas without p53mutations. It has been reported that 11–19% of papillarycarcinomas over-express p53 protein (Dobashiet al., 1993;Holm et al., 1994; Gersimovet al., 1995), where p53 mutationswere not evaluated. p53 protein over-expression without p53mutation in WDTC has been reported as a poor prognosticmarker (Hosaiet al., 1997), which we could not confirm,possibly because variants such as tall cell and insular type ofpapillary carcinoma were not included in this study. In thethyroid, over-expression of MDM2, which binds to p53 proteinand inactivates its transcription-activating function, wassuggested to participate in the pathogenesis of papillarycarcinoma with the result of co-expression of MDM2 and p53in three cases of well-differentiated papillary carcinomas from24 cases studied (Jenningset al., 1995). In our data, MDM2over-expression was seen in 16 of 30 cases (53·3%) and threecases showed co-expression with p53 protein, but there was nostatistically significant correlation (Fig. 3). This result isdifferent from a previous report (Zouet al., 1995) whereMDM2 over-expression was observed only in cases thatharboured p53 mutations (4 out of 22 cases of thyroidneoplasms). Over-expression of Bcl-2, which inhibits theapoptosis regulated by p53, was reported in low grademalignant neoplasms including WDTC, and was suggested asa good prognostic factor with inverse correlation compared to



p53 over-expression (Basoloet al., 1997). The same grouppreviously reported their observation of inverse correlation inthyroid neoplasms as a whole, but not in each histotype (Pollinaet al., 1996). In the present data, Bcl-2 over-expression wasobserved in 17 of 30 cases (56·7%), which again did not showstatistically significant correlation with expression of p53.Instead, our results showed significant association of MDM2with Bcl-2 expressions in papillary carcinomas. Although theyhave different individual modes of action, it appears thatanother molecular defect(s) may control these expressions inpapillary carcinomas from a Korean population.

In summary, the process of thyroid tumorigenesis seems tovary according to a patients genetic and environmentalbackground. It may be necessary to find the unique geneticalterations in each geographical regions to understand thispathogenesis. Further studies are needed to confirm the under-lying molecular mechanism(s) of co-expression of MDM2 andBcl-2 in papillary carcinoma.

Acknowledgements

The authors gratefully acknowledge Drs S. H. Kim and M. S.Kook for critically reviewing this manuscript. This study wassupported by a grant (no. HMP-96-M-1-1026) of the 1996 GoodHealth R&D Project, Ministry of Health & Welfare, ROK andpartly by a grant (96-151) from Asan Institute for Life Sciencesand Technology.

References

Basolo, F., Pollina, L., Fontanini, F., Fiore, L., Pacini, F. & Baldanzi, A.(1997) Apoptosis and proliferation in thyroid carcinoma: correlationwith bcl-2 and p53 protein expression.British Journal of Cancer, 75,537–541.

Becker, D.V., Robbins, J., Beebe, G.W., Bouville, A.C. & Wachholz,B.W. (1996) Childhood thyroid cancer following Cherobyl accident:a status report.Endocrinology and Metabolism Clinics of NorthAmerica, 25, 197–211.

Burman, K.D., Ringel, M.D. & Wartofsky, L. (1996) Unusual types ofthyroid neoplasms.Endocrinology and Metabolism Clinics of NorthAmerica, 26, 49–68.

DeGroot, L.J., Kaplan, E.L., McCormick, M. & Straus, F.H. (1990)Natural history, treatment, and course of papillary thyroid carcinoma.Journal of Clinical Endocrinology and Metabolism, 71, 414–424.

Dobashi, Y., Sakamoto, A., Sugimura, H., Mernua, M., Mori, M.,Oyama, T. & Machinami, R. (1993) Overexpression of p53 as apossible prognostic factor in human thyroid carcinoma.AmericanJournal of Surgical Pathology, 17, 375–381.

Fagin, J.M. (1992) Genetic basis of endocrine disease 3: moleculardefects in thyroid gland neoplasia.Journal of Clinical Endocrinologyand Metabolism, 75, 1398–1400.

Farid, N.R., Shi, Y. & Zou, M. (1994) Molecular basis of thyroidcancer.Endocrine Review, 15, 202–232.

Farid, N.R., Zou, M. & Shi, Y. (1995) Genetics of follicular thyroid

cancer.Endocrinology and Metabolism Clinics of North America, 24,865–883.

Friend, S. (1994) A glimpse at the puppet behind the shadow play.Science, 265,334–335.

Fugazzola, L., Pierotti, M.A., Vigano, E., Pacini, F., Vorontsova, T.V.& Bongarzone, I. (1996) Molecular and biochemical analysis ofRET/PTC4, a novel oncogenic rearrangement between RET andELE1 genes, in a post-Chernobyl papillary thyroid cancer.Oncogene, 13, 1093–1097.

Gersimov, G., Bronstein, M., Troshina, K., Alexandrov, G., Deirov, I.,Lesisos, T., Kalikarury, B.V., Izaquterdo, R., Bogunewicz, A.,Robbison, L., Breese, K., Ross, J.S. & Figge, J. (1995) Nuclear p53immunoreactivity in papillary thyroid cancers is associated with twoestablished indicators of poor prognosis.Experimental and Mole-cular Pathology, 62, 52–62.

Gorelov, V.N., Gyense, M., Neser, F., Roher, H.D. & Goretzki, P.E.(1996) Distribution of Gs-alpha activating mutations in humanthyroid tumors measured by subcloning.Journal of Cancer Researchand Clinical Oncology, 122,453–457.

Holm, R. & Nesiland, J.M. (1994) Retinoblastoma and p53 tumorsuppressor gene protein expression in carcinomas of the thyroidgland.Journal of Pathology, 172,267–272.

Hosai, S.A., Apel, R.L., Freeman, J.L., Azadian, A., Rosen, I.B.,Livolsi, V.A. & Asa, S.L. (1997) Immunohistochemical localizationof P53 in human thyroid neoplasms—Correlation with biologicalbehavior.Endocrine Pathology, 8, 21–28.

Hsu, S.M., Raine, L. & Fanger, H. (1981) The use of avidin-biotinperoxidase complex (ABC) in immunoperoxidase techniques; acomparison between ABC and unlabeled antibody (PAP) procedures.American Journal of Clinical Pathology, 29, 577–581.

Ito, T., Seyama, T., Mizuno, T., Tsuyama, N., Hayashi, T., Hayashi, Y.,Dohi, K., Nakamura, N. & Akiyama, M. (1992) Unique associationof p53 mutations with undifferentiated but not with differentiatedcarcinomas of the thyroid gland.Cancer Research, 52, 1369–1371.

Jennings, T., Bratsavsky, G., Gersimov, G., Troshina, K., Bronstein, M.,Dedov, I., Alexandrova, G. & Figge, J. (1995) Nuclear accumulationof MDM2 protein in well-differentiated papillary thyroid carcinomas.Experimental and Molecular Pathology, 62, 199–206.

Kaufman, P.B., Wu, W., Kim, D. & Cseke, L. (1995) Isolation andpurification of DNA. In Handbook of Molecular and CellularMethods in Biology and Medicine(eds P.B. Kaufman, W. Wu, D.Kim & L. Cseke), pp. 1–26. CRC Press, Boca Raton.

Klugbauer, S., Lengfelder, E., Deemidchik, E.P. & Rabes, H.H. (1995)High prevalence of RET rearrangement in thyroid tumors of childrenfrom Belarus after the Chernobyl reactor accident.Oncogene, 11,2459–2467.

Kramer, J.B. & Wells, Jr S.A. (1989) Thyroid carcinoma.Advances inSurgery, 32, 195–224.

Orita, M., Suzuki, Y., Seikiya, T. & Hayashi, K. (1989) Rapid andsensitive detection of point mutations and DNA polymorphismsusing polymerase chain reaction.Genomics, 5, 874–879.

Pollina, L., Pacini, F., Fontanini, G., Vignati, S., Bevilacqua, G. &Basolo, F. (1996) bcl-2, p53 and proliferating cell nuclear antigenexpression is related to the degree of defferentiation in thyroidcarcinomas.British Journal of Cancer, 73, 139–143.

Prives, C. (1994) How loops, beta sheets, and alpha helices help us tounderstand p53.Cell, 78, 543–546.

Said, S., Schlumberger, M. & Suarez, H.G. (1994) Oncogenes andanti-oncogenes in human epithelial thyroid tumors.Journal ofEndocrinological Investigation, 17, 371–379.

Shi, Y.F., Zou, M.J., Schimidt, H., Juhasz, F., Stenszky, V., Robb, D. &



Farid, N.R. (1991) High rates of ras codon 61 mutation in thyroidtumors in an iodide-deficient area.Cancer Research, 51,2690–2693.

Song, M.H. & Chang, S.K. (1993) Mutations of K-ras oncogene inthyroid tumor tissue [in Korean].Journal of the Catholic MedicalCollege, 46, 139–147.

Spambalg, D., Sharifi, N., Elisei, R., Gross, J.L., Medeiros-Neto, G. &Fagin, J.A. (1996) Structural studies of the thyrotropin receptor andGsa in human thyroid cancers: low prevalence of mutations predictsinfrequent involvement in malignant transformations.Journal ofClinical Endocrinology and Metabolism, 81, 3898–3901.

Sugg, S.L., Zheng, L., Rosen, I.B., Freeman, J.L., Ezzat, S. & Asa, S.(1996) ret/PTC-1, -2, and -3 oncogene rearrangements in humanthyroid carcinomas: implications for metastatic potential?Journal ofClinical Endocrinology and Metabolism, 81, 3360–3365.

Takeshita, A., Nagayama, Y., Yokoyama, N., Ishikawa, N., Ito, K.,Yamashita, T., Obera, T., Murakan, Y., Kuma, K. & Takamatsu, J.(1995) Rarity of oncogenic mutations in the thyrotropin receptor ofautonomously functioning thyroid nodules in Japan.Journal ofClinical Endocrinology and Metabolism, 80, 2607–2611.

Tanaka, K., Nagayama, Y., Takeshita, A., Namba, H., Yamashita, S.,Niwa, M. & Nagataki, S. (1996) Low incidence of the stimulatory Gprotein a-subunit mutation in autonomously functioning thyroidadenomas in Japan.Thyroid, 6, 195–199.

Viglietto, G., Chiapetta, G., Martinez-Tello, F.J., Fuknaga, F.H.,Tallini, G., Rigopoulou, D., Visconti, R., Mastro, A., Santoro, M.& Fusco, A. (1995) RET/PTC oncogene activation is an early eventin thyroid carcinogenesis.Oncogene, 11, 1207–1210.

Yoshimoto, K., Iwahana, H., Fukuda, A., Sano, T. & Itakura, M. (1993)Rare mutations of Gs alpha subunit gene in human endocrine tumors.Mutation detection by polymerase chain reaction-prime introducedrestriction analysis.Cancer, 72, 1386–1393.

Zou, M., Shi, Y. & Farid, N.R. (1993) p53 Mutations in all stages ofthyroid carcinomas. Journal of Clinical Endocrinology andMetabolism, 77, 1054–1058.

Zou, M., Shi, Y., Al-Sedairy, S., Hussain, S.S. & Farid, N.R. (1995) Theexpression of the MDM2 gene, a p53 bidning protein, in thyroidcarcinogenesis.Cancer, 76, 314–318.



prevalences of gsα, ras, p53 mutations and ret/ptc rearrangement in differentiated thyroid tumours...

Documents