[CANCER RESEARCH 60, 1365–1370, March 1, 2000]

High Levels of Tyrosine Phosphorylated Proto-Ret in SporadicPheochromocytomas1

Herve Le Hir, 2 Luca G. Colucci-D’Amato,2 Nicolas Charlet-Berguerand,2 Pierre-Francois Plouin, Xavier Bertagna,Vittorio de Franciscis, and Claude Thermes3

Centre de Ge´netique Moleculaire, Laboratoire associe´ a l’Universite Pierre et Marie Curie, C.N.R.S., 91190 Gif sur Yvette, France [H. L. H., N. C-B., C. T.]; Centro diEndocrinologia ed Oncologia Sperimentale del Consiglio Nazionale delle Richerche, Facolta` di Medicina e Chirurgia, 80131 Naples, Italy [L. G. C-D., V. d. F.]; Unite´d’hypertension et Laboratoire de Ge´netique Moleculaire, Hopital Broussais, 75014 Paris, France [P-F. P.]; and Groupe d’Etude en Physiopathologie Endocrinienne, Institut deGenetique Moleculaire, Universite´ ReneDescartes, 75014 Paris 7, France [X. B.]

ABSTRACT

Pheochromocytomas are tumors originating from chromaffin cells, thelarge majority of which are sporadic neoplasms. The genetic and molec-ular events determining their tumorigenesis continue to remain unknown.On the other hand, RET germ-line mutations cause the inheritance offamilial tumors in multiple endocrine neoplasia (MEN)-2 diseases, whichaccount for a minority of pheochromocytomas. We investigated the ex-pression of theRET gene in 14 sporadic tumors harboring no activatingmutations. A subset of highly RET-expressing tumors (50%) could bedistinguished. They showed RET transcript, protein amounts as well asRet-associated phosphotyrosine levels similar to those measured in MEN-2A-associated pheochromocytomas. We also determined the GDNF andGDNF family receptor a (GFRa)-1 transcript levels in tumors and innormal tissues. Whereas the GFRa-1 transcripts were detected at similarlevels in normal tissues and in tumors, GDNF was frequently foundexpressed in sporadic tumors at levels several times higher than in con-trols. These results led us to propose the existence of an autocrine orparacrine loop leading to chronic stimulation of the Ret signaling path-way, which could participate in the pathogenesis of a number of sporadicpheochromocytomas.

INTRODUCTION

Pheochromocytomas are tumors originating from chromaffin cells,80–85% of which are sporadic neoplasms (1–3) that most oftendevelop from the adrenal gland and typically follow a benign course(3). The remaining 15–20% of pheochromocytomas are of familialorigin associated with the MEN4-2 diseases, the von Hippel-Lindaudisease, or the type 1 neurofibromatosis, which present germ-linemutations in theRET, VHL, andNF-1 genes, respectively (1, 2, 4, 5).

The proto-oncogeneRET is a receptor-like protein tyrosine kinase(6). Four distinct ligands for the Ret protein have recently beenidentified. All are polypeptide growth factors belonging to the glialcell line-derived neurotrophic factor family (GDNF, neurturin, perse-phin, and artemin). Ret association to any of the ligands is mediatedby the presence, in the same molecular complex, of distinct glycosyl-phosphatidylinositol anchored proteins, the GFRa-1–4 (Ref. 7 and

references therein). Germ-line mutations ofRETcause the inheritanceof the MEN-2 syndromes (for review, see Ref. 8). Mutations incysteine residues of the extracellular domain (exons 10 and 11) are themost frequent causative genetic events of familial medullary thyroidcarcinoma and MEN-2A syndromes. A single-point mutation thatresults in a Thr-for-Met substitution at codon 918 (exon 16) within theRet catalytic domain is responsible for the MEN-2B syndrome (9).These mutations cause chronic induction of the tyrosine kinase andconvertRET into a dominant oncogene (10–12).

Contrary to familial pheochromocytomas, knowledge of the geneticand biochemical events involved in the pathogenesis of the sporadicpheochromocytomas is still lacking. Mutations in theGDNF gene donot seem to play a major role in the pathogenesis of these tumors (13,14). Moreover, sporadic pheochromocytomas are only rarely associ-ated with somatic activating mutations ofRET. Substitution of me-thionine 918 occurs in 10–15% of total sporadic pheochromocytomas,and mutations in the Ret extracellular cysteine-rich domain in lessthan 5% of these tumors (5). On the other hand, the wild-typeRETgene is frequently expressed in pheochromocytomas, but few data areavailable on the Ret protein activity in these tumors (15–17). Indeed,the presence of Ret raises the question of whether stimulation ofwild-type Ret activity in sporadic pheochromocytomas might be im-plicated in determining the neoplastic phenotype.

The causal function played byRETmutants in pheochromocytomasof familial origin has raised the hypothesis that, in addition to the Retprotein, other partners of the Ret ligand-receptor complex might beexpressed in sporadic pheochromocytomas. This might then proceedto a persistent stimulation of the Ret signaling pathway. To test thishypothesis we asked: (a) whether sporadic pheochromocytomas pre-sented enhancedRET expression levels; (b) whether Ret enzymaticactivity was stimulated; (c) whether other Ret partners,i.e., GDNFand GFRa-1 were also expressed in these tumors; and (d) whether Retexpression and activity levels were compatible with a contribution ofRetwt to a multistep process resulting in the tumorigenesis of chro-maffin cells.

Here we show the persistent stimulation of the Ret protein in anumber (50%) of sporadic adrenal pheochromocytomas harboring noactivating mutation, and we discuss the possible effects of the ob-served enhanced RET and/or GDNF transcript levels in these tumors.

MATERIALS AND METHODS

Human Tissue Samples and DNA Analysis.The adrenal RNA samplea.g.1 is a pool of total RNAs extracted from six whole adrenal gland specimens(purchased from Clontech); a frozen postmortem a.m. was obtained from Dr.P. F. Plouin (Hoˆpital Broussais, Paris, France) and two frozen postmortem s.n.were obtained from Dr. E. Hirsch (Ho¨pital Pitie Salpetriere, Paris, France).Frozen samples of human pheochromocytomas were obtained from the Re´seauComete (Hopital Broussais). Tumor samples were frozen immediately aftersurgery and kept in liquid nitrogen. All of the tumors were adrenal tumors.Malignancy was defined by histological evidence of distant metastases (3).Genomic and tumor DNA were assayed forRET-activating mutations situatedin exons 10, 11, and 16 (4, 5, 18) for all of the patients. Nineteen tumors,

Received 8/2/99; accepted 12/21/99.The costs of publication of this article were defrayed in part by the payment of page

charges. This article must therefore be hereby markedadvertisementin accordance with18 U.S.C. Section 1734 solely to indicate this fact.

1 Supported by the Centre National de la Recherche Scientifique (CNRS); the InstitutNational de la Sante´ et de la Recherche Me´dicale (INSERM); the Ligue Nationale ContreLe Cancer (LNCC); the Association Pour La Recherche Sur Le Cancer (ARC); and in partby PHRC Grant AOM95201 for the COMETE Network; the Associazione Italiana per laRicerca sul Cancro (AIRC); the Consiglio Nazionale delle Ricerche, Target Project onBiotechnology; the Fondazione Telethon Grant A.097. H. L. H. was supported by fellow-ships from LNCC and ARC; N. C. B. was supported by a fellowship from the FrenchMinistere de l’Education Nationale et de la Recherche; and V. d. F. was supported by ECGrant BIO4-CT97-5078.

2 These three authors contributed equally to this study.3 To whom requests for reprints should be addressed, at the Centre de Ge´netique

Moleculaire, CNRS, 91190, Gif sur Yvette, France. Phone: 00-33-1-69-82-38-28; Fax:00-33-1-69-82-38-77; E-mail: thermes@cgm.cnrs-gif.fr.

4 The abbreviations used are: MEN, multiple endocrine neoplasia; RPA, RNasesprotection assay; atmol, attomole; GFRa, GDNF family receptora; a.g.1, a pool of sixtotal adrenal glands; Retwt, Ret wild type; a.m., adrenal medulla; s.n., substantia nigra.

1365

numbered 2 to 20, were analyzed. Germinal mutations were found in fourcases: (a) tumor 2 (MEN-2A, C634R); (b) tumor 3 (MEN-2A, C634P); (c)tumor 4 (MEN-2A, C634R); and (d) tumor 5 (MEN-2B, M918T); one tumorDNA mutation was found for patient tumor 6 (M918T), but no germinalmutation was found in this case. All of the other tumor DNA samples (tumors7 to 20) as well as the corresponding genomic DNAs presented no activatingmutation.

RNA Preparation. For all of the samples, total cellular RNA was extractedfrom frozen tissues or tumor samples with RNAzol B (Bioprobe) according tothe manufacturer’s instructions.

In Vitro RNA Synthesis. DNA templates for transcription of labeledantisense RNA probes and unlabeled sense RNAs were obtained by PCRincluding the sequence of the T7 promoter. Antisense RNA probe and thecorresponding sense RNA corresponded to the following regions: (a) GDNF,exon 2 G378 to G545 and T352to G564; (b) GFRa-1, G1044 to G1182 andG908 to G1257; (c) RET, exon 19 T31 to G129 and exon 18 T23 to exon 20G58; (d) tyrosine hydroxylase, exon 14 C61 to G140 and exon 13 T102 to exon14 C213; and (e) b-actin, exon 5 G1 to T64 and exon 4 C342 to exon 5 C58.Labeled antisense RNA probes were synthesized in 40 mM Tris-HCl (pH 7.5),10 mM NaCl, 6 mM MgCl2, 2 mM spermidine, 6 mM DTT, 140 mM ATP, 140mM GTP, 140 mM CTP, 3.5 mM UTP, 0.5 mM [a-32P]UTP (3000 Ci/mmol) forthe RET probe, 1.05 mM [a-32P]UTP for the GDNF and GFRa-1 probes, 2pmoles of template DNA, 50 u of T7 RNApolymerase (TEBU), in a totalvolume of 10ml, incubated for 30 min at 37°C, then incubated for 30 min with10 units of DNase I (Boehringer Mannheim) and purified on 5% denaturingacrylamide gel. Sense RNA was synthesized in the same conditions with 0.6mM ATP, 0.6 mM GTP, 0.6 mM CTP, 0.5 mM UTP and 0.17mM [a-32P]UTP.RNA probes and synthetic RNAs were resuspended in H2O containing 2 ng/mltRNA (Escherichia coliextract).

RPAs. Cellular RNAs (or synthetic mRNAs) and 0.3–1 fmol of radiola-beled RNA probe were mixed, lyophilized, resuspended in 80% formamide,0.4M NaCl, 40 mM PIPES, and 1MM EDTA in a total volume of 7.5ml; heatedfor 5 min at 85°C; incubated for 12 h at 60°C; digested by the addition of 87.5ml of 300 mM NaCl, 10 mM Tris (pH 7.5), 5 mM EDTA, 50 mg/ml RNase A,and 120 units/ml RNase T1 (Sigma); incubated for 30 min at 30°C; then treatedwith 0.2 mg/ml proteinase K and 1% SDS for 30 min at 37°C; treated withphenol and ethanol; and analyzed on 7% acrylamide denaturing gels. Densi-tometric measurements were performed with a PhosphorImager. The intensi-ties of the bands were obtained after subtraction of the background. RNA-protected bands that were obtained with known amounts of synthetic RNAswere used as standards to calculate the absolute amounts of the correspondingmRNA species contained in the cellular RNA samples. In all of the experi-ments, the intensities of the bands of interest increased linearly with theamount of cellular RNA (see Fig. 1b).

Statistical Analysis. The comparison of the distributions of the RET tran-script levels of tumors to that of normal tissues (a.m. and a.g.1) was performedby the Studentt test with the following data. One mean value (of at least threemeasurements) was used for each tumor. Because a.m. is only one specimenand a.g.1 is a pool of six adrenal glands, one mean value was used for the a.m.sample and six different measurements were used for the a.g.1 sample; theresultingP-value was 0.0031.

Ret Protein Analysis. Frozen tumor fragments (0.05–0.1 g) were crushedin liquid nitrogen, resuspended in 1 ml of ice-cold lysis buffer [10 mM

Tris-HCl (pH 8), 150 mM NaCl, 0.4 mM EDTA, 1% NP40, 10 mM NaF, 10 mM

Na2H2P2O7, 2 mg/ml aprotinin, 2 mg/ml leupeptin, 100 mg/ml AEBSF, and 2mM Na3Va04], incubated for 10 min on ice, and clarified by centrifugation(19). Four mg of proteins were immunoprecipitated with 10ml of anti-Retantibody C-19 (Santa Cruz) as described previously (19) and then werefractionated on 8% SDS-PAGE gel and transferred to nitrocellulose membraneand probed for antiphosphotyrosine with antibody 4G10 (Santa Cruz; 19). Todetect the Ret protein, the membrane was incubated in 100 ml of strippingbuffer [62.5 mM Tris-HCl (pH 6.8), 2% SDS, 100 mM b-mercaptoethanol]during 30 min at 55°C, treated with T-TBS 5% nonfat milk, and probed withanti-Ret antibody C-19 (Santa Cruz) in 20 ml T-TBS for 2 h at roomtemperature. Detection was performed by an antirabbit secondary antibody(ImmunoPure) and chemiluminescence reagents (SuperSignal, Pierce).

RESULTS

We first analyzed the RET, GDNF, and GFRa-1 transcript levels innineteen adrenal pheochromocytomas. All of the tumors were char-acterized and tested for the presence ofRET-activating mutations inexons 10, 11, and 16 as described in “Materials and Methods.”Fourteen sporadic tumors presented no activating mutation 4 wereMEN-2-associated tumors and 1 was sporadic harboring aRET-activating mutation. Several normal tissues (a.g.1 and an a.m. sample)were studied as controls; two samples of s.n. were also examined asadditionalRET-expressing tissues. To measure the absolute amountsand proportions of transcripts, we performed quantitative RPAs.These were carried out with total cellular RNA samples and in parallelwith known amounts of purified RET synthetic RNAs used as stand-ards (see “Materials and Methods”).

Labeled antisense RNA probes specific for the RET, GDNF, andGFRa-1 transcripts were first tested separately and then mixed toallow more reliable measurements (see “Materials and Methods”).Invitro transcribed RNAs corresponding to these three genes were alsotested separately and mixed to be used as controls. Fig. 1a shows atypical RPA experiment, in which separated and mixed control RNAsexhibit the protected RNA bands specific for the RET, GDNF, andGFRa-1 transcripts (synthetic RNAs Lanes); a short artifactual bandmigrating slightly above the RET specific band can be identified inthese control Lanes. Fig. 1b illustrates the linearity of the measure-ments.

Expression of theRET Gene. We found that all of the tumorsexpressed RET transcripts in highly variable amounts (mean5 10.1,SD 5 9.85 atmol/mg, Fig. 2a). These RET values differed stronglyfrom those found in normal tissues (0.74, 1.42, and 0.42 atmol/mg fors.n, a.m., and a.g.1 samples, respectively; Fig. 2a). To determinewhether these differences were statistically significant, the distribu-tions of the RET tumor values were compared with the RET values forthe a.m. and a.g.1 samples with the Studentt test (see “Materials andMethods”). This showed that the two distributions were significantlydifferent (the probability that they were similar wasP 5 0.0031).Interestingly, the distribution of the RET transcript levels among thetumors presented a bimodal aspect, revealing two different subsets ofsporadic tumors (Fig. 2b). Half of the sporadic tumors expressed RETtranscripts at high levels (mean5 18.1, SD5 7.1 atmol/mg) rangingfrom 10 to 32 times the level found in the control tissues (mean5 0.9atmol/mg). On the other hand, in the remaining half of sporadictumors the RET transcript levels (mean5 2.1, SD5 1.7 atmol/mg)were dispersed around the values found in normal tissues. In addition,the three MEN-2A associated pheochromocytomas (tumors 2–4) aswell as the sporadic tumor harboring a RET mutation (tumor 6)showed RET levels (12–29 atmol/mg) similar to those found in thehighly RET-expressing sporadic tumors (Fig. 2b). We note that nocorrelation could be observed between malignancy of tumors and theRET transcript levels.

Phosphorylation of the Ret Protein. Because stimulation of theRet protein results in autocatalytic tyrosine kinase activity, we inves-tigated whether Ret products were phosphorylated on tyrosine resi-dues in sporadic tumors. We immunoprecipitated protein extracts withanti-Ret antibodies, followed by immunoblot with antiphosphoty-rosine (Fig. 3a, lower panel) and anti-Ret antibodies (Fig. 3a, upperpanel). A neuroblastoma cell line, Neuro-2A, transfected with theGFRa-1 receptor was used as control. The relative amounts of Retprotein in the tumors analyzed paralleled the amount of transcripts. Inneuroblastoma cells, immunoreactivity of Ret with the antiphospho-tyrosine monoclonal antibodies depended as expected on stimulationby the GDNF ligand (Fig. 3a, Lanes 1and2).

In agreement with the RPA data, Ret was present in pheochromo-1366

RET PHOSPHORYLATION IN SPORADIC PHEOCHROMOCYTOMAS

cytoma 3 and strongly reacted with antiphosphotyrosine antibodies,consistently with the fact that it is a MEN-2-associated tumor. Intumors 18 and 7, Ret was expressed at high levels and also stronglyreacted with antiphosphotyrosine antibodies (Fig. 3a). This is a sur-prising result because these tumors did not harbor any known Retactivating mutation. Interestingly, the amount of tyrosine-phosphory-lated Ret in tumors 18 and 7 was comparable to that found in tumor3, raising the possibility that, in these tumors, Ret may also play a rolein tumorigenesis (Fig. 3a). More generally, in all of the other sporadictumors analyzed that presented high RET transcript levels (tumors 8,10, and 14) Ret reacted with antiphosphotyrosine antibodies to levelssimilar to that of tumor 3 (Fig. 3b), raising the possibility that, in thesetumors, Ret may also play a role in tumorigenesis. In low-Ret-expressing tumors 20 and 17, the amount of phosphorylated Ret wasmuch less than that found in the MEN-2A tumor 3.

Finally, we investigated the expression of the tyrosine hydroxylasegene. This gene is a well- established marker of differentiated chro-maffin cells. We found as expected that, in most of the tumorsanalyzed,TH was expressed at higher levels than in normal tissues(Fig. 4). On the other hand, we could not observe any correlationbetween RET and TH transcript levels.

Expression of the GDNF and GFRa-1 Genes.The finding oftyrosine-phosphorylated Retwt in sporadic tumors suggested that ei-ther an autocrine or a paracrine stimulation could take place in thesetumors. Because Ret activity is physiologically stimulated by growthfactors of the GDNF family and requires the presence of the GFRas(1–4) as components of the same cell surface complex (7), we asked

whether Ret-persistent-tyrosine phosphorylation might be induced bythe presence of GDNF and GFRa-1 in the same tumor samples. We,thus, examined the GDNF and GFRa-1 absolute transcript levels insporadic pheochromocytomas.

The GDNF transcripts were not detected in the a.m. sample butwere detected in the a.g.1 to a level of 0.03 atmol/mg (Fig. 5). All ofthe benign (tumors 7 to 17) and two of three malignant tumorsexpressed GDNF. Most of them presented enhanced GDNF levels, upto 14 times larger than that found in the a.g.1 tissue sample, but nostrong correlation was observed between the GDNF and RET levels.Two malignant tumors (tumors 18 and 19) showed high RET levelsand in one of them (tumor 18) no GDNF transcript could be detected.Three MEN-2 associated pheochromocytomas (3 to 5) as well as thesporadic tumor 6 (harboring a RET mutation) showed enhancedGDNF levels.

The GFRa-1 transcript levels among the sporadic tumors did notpresent strong variations except for tumor 18 (this tumor showed nodetectable GDNF transcript). The GFRa-1 mean value for sporadictumors, 0.08 atmol/mg, was similar to the values observed in controltissues.

Whether the levels of GDNF and GFRa-1 transcripts reflect thecorresponding protein levels remains to be investigated, but this is sug-gested by the analysis of several poly(A)1 tumor RNA fractions, whichshowed GDNF and GFRa-1 levels similar to those found in the totalRNA samples (data not shown). The overexpression of GDNF observedin most tumors, thus, supports the hypothesis that a paracrine or autocrinestimulation may occur in the process of Ret activation.

Fig. 1. a, RNase mapping of RET, GDNF, andGFRa-1 transcripts.In vitro synthesized RNAs ortotal RNAs extracted from tumor tissues were hy-bridized with labeled RNA probes in excess, di-gested with RNases, and analyzed on polyacryl-amide gel (see “Materials and Methods”). In all ofthe assays, the RET, GDNF, and GFRa-1 probeswere mixed and hybridized simultaneously to eachRNA sample to improve comparative measurements.M, marker; P, non-digested probe mixture;C, di-gested probe mixture;synthetic RNAs, the assayswere performed separately with 5 atmoles of theindicated RET, GFRa-1, and GDNF syntheticRNAs; mix, 1, 3, and 9 atmoles of an equimolarmixture of RET, GFRa-1, and GDNF syntheticRNAs. The numbers refer to tissue samples:1, 5 mgof a.m. RNA;4–18, 5 mg of the corresponding tumorRNA samples;b, linearity of the transcript levelmeasurements. RNase mapping assays were per-formed as ina with 2, 4, and 8mg of the RNAsample from tumor 3; absolute transcript amountswere calculated as described in “Materials and Meth-ods.”E, GDNF;F, RET;‚, GFRa-1.

1367

RET PHOSPHORYLATION IN SPORADIC PHEOCHROMOCYTOMAS

DISCUSSION

Ret-activating mutations cause different types of tumors, includingmedullary thyroid carcinoma and pheochromocytoma, of either spo-

radic or familial origin. Here we report that the normal Ret productsare expressed at high levels and also are phosphorylated on tyrosineresidues in a number of sporadic adrenal pheochromocytomas. More-over, we find that the amount of Retwt phosphorylation in sporadicpheochromocytomas is similar to that found in a tumor where theRETgene is mutated and, therefore, constitutively active. Furthermore, thepresence in these tumors of GDNF and/or GFRa-1 transcripts sug-gests the existence of an autocrine or paracrine stimulation involving

Fig. 3. a, Ret phosphorylation in human pheochromocytomas. One familial tumor(tumor 3), five sporadic tumors with high RET transcript levels (tumors 7, 8, 10, 14, and18), and two tumors with low RET levels (tumors 17 and 20) were studied. Total cellularproteins were analyzed by immunoblotting with anti-Ret (upper panel) and antiphospho-tyrosine (lower panel) antibodies; thenumbers, the tumor samples. Neuro-2A culture cellstransfected with GFRa-1 were stimulated when indicated with GDNF.b, relative amountof Ret phosphotyrosine in sporadic tumors. Theordinate, the amount of phosphotyrosinein sporadic tumors measured by densitometry from immunoblots, divided by the amountof phosphotyrosine observed in the MEN-2A tumor 3.

Fig. 2. a, RET andb-actin absolute transcript amounts. The measurements wereobtained by reference to assays performed with defined amounts ofin vitro synthesizedRET andb-actin mRNA fragments (see “Materials and Methods”). Thenumbers alongthe ordinateand above the bars, RNA amounts in atmol permg of total RNA; 3, notdetermined. The indicationsalong the abscissarefer to normal tissues (s.n., the meanobtained with the two s.n. samples) or to tumor samples:2–4, MEN-2A; 5, MEN-2B; 6,sporadic tumor RET mutation;7–20, sporadic adrenal tumors without RET-activatingmutation; B, benign;M, malignant.Arrow, the mean value (0.9 atmol/mg) of controltissues a.m. and a.g.1. Transcript levels were measured for theb-actin gene as a markerof the total RNA content and did not show significant variations among the samples.b,histogram of the RET transcript amounts. On theabscissa, theposition of the columns, theamounts of RET transcripts (given ina); on theordinate, the number of tumors presentingthe indicated amount of RET transcripts;open columns, sporadic pheochromocytomaswithout RET-activating mutation;cross-hatched columns, MEN-2 pheochromocytomasand tumor harboring a RET mutation ( tumor 6);arrow, the position corresponding to thenormal tissues. - - -, two subsets of tumors: one presents low RET values (,5 atmol/mg);the other presents high RET values (10–30 atmol/mg).

1368

RET PHOSPHORYLATION IN SPORADIC PHEOCHROMOCYTOMAS

the Ret signaling pathway that may contribute to the maintenanceand/or development of a large subset of sporadic pheochromocyto-mas.

The finding that in a number of sporadic pheochromocytomas,Retwt is highly expressed and phosphorylated raises the question ofits implication in the determination of the final neoplastic phenotypeof these tumors. Indeed, enhanced expression of receptor tyrosinekinases and polypeptide growth factors are common features of manytumor and tumor-derived cell lines that are likely to contribute toproliferation or invasion (20–22). However, because of the causal roleplayed by the Ret signaling in MEN-2 pheochromocytomas, thefinding of enhanced expression of Retwt (in sporadic pheochromocy-tomas) assumes a more important significance. Indeed, the MEN-2A-like RETmutations induce chronic activation of Ret tyrosine kinaseby forming stable disulphide bonds between Ret monomers, thusmimicking ligand-induced dimerization. On the other hand, amongthe sporadic pheochromocytomas with no activatingRETmutations,50% of tumors exhibited Ret protein expression and phosphorylationat levels close to those found in MEN-2A tumors. Thus, it seemsreasonable to infer that the overall signaling triggered by Ret in thesesporadic tumors may be quantitatively, and likely qualitatively, com-parable to that triggered by Ret-2A in the familial ones. In this respect,Ret would cause similar biological effects in both tumor types(Ret-2A in familial versusRetwt in sporadic tumors). A main differ-ence between sporadic and familial tumors may consist in the devel-opmental stage at which Ret becomes chronically stimulated. In thecase of the inherited mutated allele, we may assume that Ret becomesactive as soon as expressed. However, in the case of Retwt in sporadictumors, we have no indication about the stage at which its expressionbecomes deregulated and, thus, whether its activation is implicated inthe initial steps of tumor progression. We cannot exclude the possi-bility that distinct genetic events determine the tumor formation,which in turn would induce the expression of the Ret ligand/receptorcomplexes.

As previously reported, in the a.m., Ret either is undetectableduring rat embryogenesis or is expressed in only a small number ofchromaffin cells in the adult tissue (23). Thus the high Ret levelsobserved in sporadic adrenal tumors may be explained by a clonalexpansion of the highly Ret-expressing cells of the original tissue.However, the small number of Ret-positive cells detected in thenormal tissue cannot lead to the observed large proportion (50%) ofRet-overexpressing tumors. A more likely possibility is that the Ret-positive chromaffin cells have a growth advantage; some of these cellswould then be preferentially selected to give rise to tumor formation.This scenario is in good agreement with our hypothesis that the Retactivity plays an essential role in the neoplastic processes leading tothe Ret-overexpressing sporadic pheochromocytomas.

Until now, four different ligands for Ret have been isolated, each ofwhich activates the Ret tyrosine kinase in association with a member

of the membrane-bound receptor family (GFRa-1–4). In this study weanalyzed the expression of three of the components of the Ret ligand/receptor complex,i.e., Ret, GFRa-1, and GDNF. GFRa-1 was presentat similar levels in all of the sporadic tumors and normal tissuesanalyzed, thus providing the necessary receptor for GDNF signaling.GDNF was expressed in all of the tumors but one, and frequently itsexpression was highly enhanced as compared with control tissues. Thepresence of GFRa-1 and GDNF strongly indicates that both partici-pate in stimulating Ret tyrosine kinase activity in these tumors by anautocrine mechanism. The only exception was tumor 18; in thissample, Retwt was highly phosphorylated in tyrosine residues, even inthe absence of GDNF. A likely possibility is that other ligands, eitherdescribed or still unidentified, may replace GDNF to stimulate Ret inthese tumors, or, alternatively, ligands may be provided by surround-ing cells, thus stimulating Ret by a paracrine mechanism. Further-more, even though unlikely, we cannot exclude the possibility thatligand-independent dimerization of Ret may take place in some ofthese tumors.

An alternative interpretation of these results is that Retwt signalingmay contribute to the maintenance of the differentiated phenotyperather than to the neoplastic progression of these tumors. Indeed, theeffects of Ret activity in neuroendocrine cells still remains controver-sial. In vitro experiments indicate that Ret causes differentiation (insome cases, even terminal differentiation) of neuroectodermal cells,including PC12 (24–26), neuroblastoma (27), and primary culturesfrom human pheochromocytomas (28). This is true either if the Retkinase is activated by an oncogenic mutation or if it is induced byligand stimulation (29). However, thesein vitro observations seem atleast in part, difficult to reconcile with thein vivo data. In fact: (a) the

Fig. 5. GDNF and GFRa-1 absolute transcript amounts. The measurements wereobtained by reference to assays performed with defined amounts ofin vitro synthesizedGDNF and GFRa-1 mRNA fragments (see “Materials and Methods”); indications onabscissa as in Fig. 2a.

Fig. 4. TH absolute transcript amounts. The measurements were obtained by referenceto assays performed with defined amounts ofin vitro synthesized TH mRNA fragments(see “Materials and Methods”); indications on abscissa as in Fig. 2a.

1369

RET PHOSPHORYLATION IN SPORADIC PHEOCHROMOCYTOMAS

Ret oncogene causes tumor formation in MEN-2 syndromes; (b)transgenic expression of theRetoncogene in C-cells of mice thyroidcauses proliferation and tumors (30); (c) cells from either sporadic orMEN-2A tumors showed mitotic activityin vivo. However, whengrown as primary culturesin vitro, they lack the ability to incorporatebromodeoxy-uridine, observed in the tumors from which they origi-nate; moreover, in these cells, GDNF stimulation induces neuriteoutgrowth but not bromodeoxy-uridine incorporation (28); and (d) inour experiments, neither the expression of theRETgene nor the Retprotein activity correlate with the expression levels of theTH gene, amarker of chromaffin cells; the TH transcript levels appeared to besimilar in all of the tumors analyzed, either familial or sporadic.

High expression levels ofGDNF and RET genes were also ob-served in the MEN-2 tumors. Although it is conceivable that enhancedexpression of mutatedRETalleles would confer a growth advantageto MEN-2-associated pheochromocytomas, the presence of high lev-els of GDNF in these tumors was quite unexpected and difficult toreconcile with the ligand-independent activation of the RetC634 mu-tants. In tumors with a RetM918T mutation (tumors 5 and 6), it isconceivable that the ligand availability can further stimulate thereceptor kinase activity (31). In contrast, mutations of the Ret cysteineC634 are believed to be sufficient to induce complete activation of theRet receptor (32), and it seems unlikely that it can be further stimu-lated by binding of the ligand. However, our findings are consistentwith the recently describedin vitro effects of GDNF (28). Indeed,primary cultures from human MEN-2-associated pheochromocytomaswith a Ret C634 mutant protein still respond to GDNF, which likelystimulates the wild-type Ret encoded by the nonmutated allele (28).Whether in these MEN-2A tumors, the acute stimulation of thewild-type Ret can participate to define the final phenotype remains tobe determined.

In conclusion, we report an analysis of the expression levels of thegenes that code for three elements of the Ret receptor complex in thenormal adrenal tissues as well as in adrenal pheochromocytomas.High levels of phosphotyrosine-containing Ret molecules in one-halfof sporadic pheochromocytomas indicate that a persistent stimulationof the Ret activity takes place in these tumors. This suggests that theRet signaling pathway may be implicated in the pathogenesis ofsporadic tumors which represent the large majority of pheochromo-cytomas. A more extensive analysis of sporadic pheochromocytomasis in progress to determine whether the presence of highRETexpres-sion may be relevant to further define a subset of these tumors.

ACKNOWLEDGMENTS

We thank Anne Julien for technical assistance in DNA mutation analysis.We thank Prof. G. Vecchio for helpful discussions.

REFERENCES

1. Ball, D. W. Clinical manifestations of multiple endocrine neoplasia type 2.In: B. D.Nelkin (ed.), Genetic Mechanisms in Multiple Endocrine Neoplasia Type 2, pp. 2–20.Landes Company, 1996.

2. Eng, C. Seminars in medicine of the Beth Israel Hospital, Boston. TheRETproto-oncogene in multiple endocrine neoplasia type 2 and Hirschsprung’s disease. N. Engl.J. Med.,335: 943–951, 1996.

3. Plouin, P. F., Chatellier, G., Fofol, I., and Corvol, P. Tumor recurrence and hyper-tension persistence after successful pheochromocytoma operation. Hypertension,29:1133–1139, 1997.

4. Smith, D. P., Eng, C., and Ponder, B. A. J. Mutations of theRETproto-oncogene inthe multiple endocrine neoplasia type 2 syndromes and Hirschsprung disease. J. CellSci. 18 (Suppl.): 43–49, 1994.

5. Ponder, B. A. J., and Pierotti, M. A. Mutations in Ret in MEN 2.In: B. D. Nelkin(ed.), Genetic Mechanisms in Multiple Endocrine Neoplasia Type 2, pp. 21–35.Landes Company, 1996.

6. Takahashi, M., Buma, Y., Iwamoto, T., Inaguma, Y., Ikeda, H., and Hiai, H. Cloningand expression of theRET proto-oncogene encoding a tyrosine kinase with twopotential transmembrane domains. Oncogene,3: 571–578, 1988.

7. Rosenthal, A. The GDNF protein family: gene ablation studies reveal what they reallydo and how. Neuron,22: 201–203, 1999.

8. Goodfellow, P. J. Inherited cancers associated with theRET proto-oncogene. Curr.Opin. Genet. Dev.,4: 446–452, 1994.

9. Edery, P., Eng, C., Munnich, A., and Lyonnet, S. RET in human development andoncogenesis. BioEssays,19: 389–395, 1997.

10. Asai, N., Iwashita, T., Matsuyama, M., and Takahashi, M. Mechanism of activationof theRETproto-oncogene by multiple endocrine neoplasia 2A mutations. Mol. Cell.Biol., 15: 1613–1619, 1995.

11. Borrello, M. G., Smith, D. P., Pasini, B., Bongarzone, I., Greco, A., Lorenzo, M. J.,Arighi, E., Miranda, C., Eng, C., Alberti, L.,et al. RET activation by germlineMEN2A and MEN2B mutations. Oncogene,11: 2419–2427, 1995.

12. Santoro, M., Carlomagno, F., Romano, A., Bottaro, D. P., Dathan, N. A., Grieco, M.,Fusco, A., Vecchio, G., Matoskova, B., Kraus, M. H.,et al. Activation of RET asdominant transforming gene by germline mutations of MEN 2A and MEN 2B.Science (Washington DC),267: 381–383, 1995.

13. Dahia, P. L., Toledo, S. P, Mulligan, L. M., Maher, E. R., Grossman, A. B., and Eng,C. Mutation analysis of glial cell line-derived neurotrophic factor (GDNF), a ligandfor the RET/GDNF receptora complex, in sporadic phaeochromocytomas. CancerRes.,57: 310–313, 1997.

14. Woodward, E. R., Eng, C., McMahon, R., Voutilainen, R., Affara, N. A., Ponder,B. A., and Maher, E. R. Genetic predisposition to pheochromocytoma: analysis ofcandidate genesGDNF, RET, andVHL. Hum. Mol. Genet.,6: 1051–1056, 1997.

15. Santoro, M., Rosati, R., Grieco, M., Berlingieri, M. T., Colucci D’Amato, G. L., deFranciscis, V., and Fusco, A. Theret proto-oncogene is consistently expressed inhuman pheochromocytomas and thyroid medullary carcinomas. Oncogene,5: 1595–1598, 1990.

16. Matias-Guiu, X., Colomer, A., Mato, E., Cuatrecasas, M., Komminoth, P., Prat, J.,and Wolfe, H. Expression of theret proto-oncogene in pheochromocytoma. Anin situhybridization and Northern blot study. J. Pathol.,176: 63–68, 1995.

17. Takaya, K., Yoshimasa, T., Arai, H., Tamura, N., Miyamoto, Y., Itoh, H., and Nakao,K. Expression of theRETproto-oncogene in normal human tissues, pheochromocy-tomas, and other tumors of neural crest origin. J. Mol. Med.,74: 617–621, 1996.

18. Rodien, P., Jeunemaitre, X., Dumon, C., Beldjord, C., and Plouin, P. F. Geneticalterations of theRETproto-oncogene in familial and sporadic pheochromocytomas.Horm. Res. (Basel),47: 263–268, 1997.

19. Colucci-D’Amato, G. L., D’Alessio, A., Filliatreau, G., Florio, T., Di Giamberardino,L., Chiappetta, G., Vecchio, G., Fusco, A., Santoro, M., and de Franciscis, V.Presence of physiologically stimulatedRET in adult rat brain: induction ofRETexpression during nerve regeneration. Cell Growth Diff.,7: 1081–1086, 1996.

20. Ciardiello, F., Valverius, E. M., Colucci-D’Amato, G. L., Kim, N., Bassin, R. H., andSalomon, D. S. Differential growth factors expression in transformed mouse NIH-3T3cells. J. Cell. Biochem.,42: 45–57, 1990.

21. Jiang, W. G., and Hiscox, S. Hepatocyte growth factor/scatter factor, a cytokineplaying multiple and converse roles. Histol. Histopathol.,12: 537–555, 1997.

22. Porter, A. C., and Vaillancourt, R. R. Tyrosine kinase receptor-activated signaltransduction pathways which lead to oncogenesis. Oncogene,17: 1343–1352, 1998.

23. Tsuzuki, T., Takahashi, M., Asai, N., Iwashita, T., Matsuyama, M., and Asai, J.Spatial and temporal expression of theret proto-oncogene product in embryonic,infant, and adult rat tissues. Oncogene,10: 191–198, 1995.

24. Borrello, M. G., Smith, D. P., Pasini, B., Bongarzone, I., Greco, A., Lorenzo, M. J.,Arighi, E., Miranda, C., Eng, C., Alberti, L., Bocciardi, R., Mondellini, P., Scopsi, L.,Romeo, G., Ponder, B. A. J., and Pierotti, M. A. RET activation by germline MEN2Aand MEN2B mutations. Oncogene,11: 2419–2417, 1995.

25. Rossel, M., Pasini, A., Chappuis, S., Geneste, O., Fournier, L., Schuffenecker, I.,Takahashi, M., van Grunsven, L. A., Urdiales, J. L., Rudkin, B. B., Lenoir, G. M., andBillaud M. Distinct biological properties of two RET isoforms activated by MEN 2Aand MEN 2B mutations. Oncogene,14: 265–275, 1997.

26. Califano, D., D’Alessio, A., Colucci-D’Amato, G. L., De Vita, G., Monaco, C.,Santelli, G., Di Fiore, P. P., Vecchio, G., Fusco, A., Santoro, M., and de Franciscis,V. A potential pathogenetic mechanism for multiple endocrine neoplasia type 2syndromes involves ret-induced impairment of terminal differentiation of neuroepi-thelial cells. Proc. Natl. Acad. Sci. USA,93: 7933–7937, 1996.

27. D’Alessio, A., De Vita, G., Calı`, G., Nitsch, L., Fusco, A., Vecchio, G., Santelli, G.,Santoro, M., and de Franciscis, V. Expression of theREToncogene induces differ-entiation of SK-N-BE neuroblastoma cells. Cell Growth Differ.,6: 1387–1394, 1995.

28. Powers, J. F., Tsokas, P., and Tischler, A. S. The ret-activating ligand GDNF isdifferentiative and not mitogenic for normal and neoplastic human chromaffin cellsinvitro. Endocr. Pathol.,9: 325–331, 1998.

29. Maxwell, G. D., Reid, K., Elefanty, A., Bartlett, P. F., and Murphy, M. Glial cellline-derived neurotrophic factor promotes the development of adrenergic neurons inmouse neural crest cultures. Proc. Natl. Acad. Sci. USA,93: 13274–13279, 1996.

30. Michielis, F. M., Chappuis, S., Callou, B., Pasini, A., Talbot, M., Monier, R., Lenoir,G. M., Feunteun, J., and Billaud, M. Development of medullary thyroid carcinoma intransgenic mice expressing theRETprotooncogene altered by a multiple endocrineneoplasia type 2A mutation. Proc. Natl. Acad. Sci. USA.,94: 3330–3335, 1997.

31. Rizzo, C., Califano, D., Colucci-D’Amato, G. L., De Vita, G., D’Alessio, A., Dathan,N. A., Fusco, A., Monaco, C., Santelli, G., Vecchio, G., Santoro, M., and deFranciscis, V. Ligand stimulation of a RET chimeric receptor carrying the activatingmutation responsible for the multiple endocrine neoplasia type 2B. J. Biol. Chem.,271: 29497–29501, 1996.

32. Carlomagno, F., Melillo, R. M., Visconti, R., Salvatore, G., De Vita, G., Lupoli, G.,Yu, Y., Jing, S., Vecchio, G., Fusco, A., and Santoro., M. Glial cell line-derivedneurotrophic factor differentially stimulates ret mutants associated with the multipleendocrine neoplasia type 2 syndromes and Hirschsprung’s disease. Endocrinology,139: 3613–3619, 1998.

1370

RET PHOSPHORYLATION IN SPORADIC PHEOCHROMOCYTOMAS