| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ARTICLES |
Centro di Endocrinologia ed Oncologia Sperimentale del CNR (F.C., R.M.M., R.V., G.S., G.D.V., G.L., G.V., M.S.), c/o Dipartimento di Biologia e Patologia Cellulare e Molecolare, Facoltá di Medicina e Chirurgia, Universitá di Napoli "Federico II", 80131 Naples, Italy; Amgen (Y.Y., S.J.), Thousand Oaks, California 91320-1789; and Dipartimento di Medicina Sperimentale e Clinica (A.F.), Facoltá di Medicina e Chirurgia di Catanzaro, Universitá di Reggio Calabria, 88100 Catanzaro, Italy
Address all correspondence and requests for reprints to: Massimo Santoro, Centro di Endocrinologia ed Oncologia Sperimentale del Consiglio Nazionale Delle Ricerche, via S. Pansini 5, 80131 Napoli, Italy. E-mail: masantor@unina.it or afusco{at}synapsis.it
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
, acts as a functional ligand of Ret and stimulates
its tyrosine kinase and biological activity. To ascertain whether the
biological effects of ret mutations are modulated by
GDNF, we have investigated the responsiveness to GDNF of
ret mutants in cell lines coexpressing GDNFR- and
MEN2A-, MEN2B-, FMTC-, or HSCR-associated ret mutants.
Here, we show that triggering of GDNF affected only
ret/MEN2B, i.e. it stimulated
ret/MEN2B mitogenic and kinase activities, as well as
its ability to phosphorylate Shc, a bona fide Ret
substrate. In contrast, ret mutants associated with
MEN2A or FMTC (carrying Cys634 or Cys620 mutations) were unresponsive
to GDNF. HSCR mutations, by affecting either the extracellular or the
intracellular Ret domain, impaired responsiveness to GDNF. These data
suggest that the phenotype of human diseases caused by
ret mutations can be differentially influenced by GDNF. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
It recently has been reported that the glial cell line-derived
neurotrophic factor (GDNF) acts as a functional ligand for Ret. GDNF
induces Ret tyrosine phosphorylation, and survival and proliferation of
Ret-expressing cells (19, 20, 21, 22). Moreover, mice with targeted disruption
of the GDNF gene show a phenotype similar to that of
ret knock-out mice, including megacolon (23, 24). The
mechanism by which GDNF stimulates Ret has been elucidated:
GDNF interacts with GDNFR-, a glycosyl phosphatidylinositol
(GPI)-linked cell surface receptor, which, in turn, mediates Ret
activation (19, 20). More recently, another GPI-linked protein,
NTNR-, has been cloned, which mediates Ret stimulation by a
GDNF-related neurotrophin named neurturin or NTN (25, 26).
So far, the effects of MEN2 and of HSCR mutations on ret
function have been tested only under ligand-free conditions. However,
in vivo, the presence of the ligand probably affects the
activity of ret mutants in tissues targeted by these
diseases. We have studied the responsiveness of HSCR-associated and of
oncogenic Ret mutants to ligand stimulation. Because Ret and GDNFR-
coexpression is required for high affinity binding of GDNF (and,
consequently, for full biological activity of the receptor), we
cotransfected NIH 3T3 cells with GDNFR- and with constructs encoding
wild-type Ret, Ret/Cys634 (more frequently associated with MEN2A),
Ret/Cys620 (more frequently associated with FMTC), Ret/Met918 (MEN2B),
and Ret/HSCR972 and Ret/HSCR32 (HSCR). GDNF stimulated tyrosine
phosphorylation of the wild-type receptor, and this was followed by a
mitogenic response. Two phenotypes were identified among the Ret
mutants analyzed: Ret/Cys634, Ret/Cys620, Ret/HSCR972 and Ret/HSCR32
were unresponsive to GDNF, whereas Ret/Met918 was responsive to GDNF
triggering.
These data demonstrate that HSCR mutants are unresponsive to the physiological Ret ligand. Moreover, our data on the activated Ret mutants show that GDNF influences the biological behavior of the Ret/Met918 mutant but not of the MEN2A- and FMTC-associated Ret mutants and (consistent with the clinical aggressiveness of the MEN2B syndrome) indicate that, when triggered by the ligand, Ret/Met918 is a very potent oncoprotein.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
(pSJA45-GDNFR-),
which carries G418-resistance (19).
Cells and transfection experiments
NIH 3T3 cells were grown in DMEM supplemented with 10% calf
serum (Gibco BRL, Life Technologies, Gaithersburg, MD). Transfections
were performed by calcium phosphate precipitation, as described
elsewhere (27). To obtain coexpression of GDNFR- (G418-R) together
with ret constructs (mycophenolic acid-R), cells were
cotransfected with 1 µg GDNFR- and 10 µg ret
constructs and were selected for plasmid expression by growth in 400
µg/ml G418 (Gibco BRL). Soft agar colony assay was performed as
reported (27); colonies were scored at 10 days. The expression of Ret
and GDNFR- in each cell line was verified by Western blotting and
RT-PCR, respectively. Briefly, total RNA was isolated from each cell
line by the acid guanidinium thiocyanate phenol method (28) and then
subjected to ribonuclease-free deoxyribonuclease digestion (Promega
Corp., Madison, WI). RNAs (1 µg) were reverse transcribed with the
Moloney murine leukemia virus reverse transcriptase (Perkin-Elmer
Cetus, Branchburg, NJ), and the complementary DNA (cDNA) products were
PCR-amplified using standard conditions (Perkin-Elmer Cetus). The
primers used for GDNFR- were the following: forward,
5'-CGGTTAACAGCAGGTTGTCAGA-3'; and reverse,
5'-GTGTGGGGATCTCATTCTCAGAC-3'.
The PCR conditions included an initial denaturation step at 94 C for 4
min, followed by 20 cycles at 94 C for 1 min, 58 C for 2 min, 72 C for
2 min, and a final extension step at 72 C for 5 min. Amplified products
(1/10 of the reaction mixture) were analyzed by electrophoresis on 1%
agarose gel and hybridized with a GDNFR- probe excised from the
pSJA45-GDNFR- plasmid. The expected size of the reaction product was
801 bp. RT-PCR reactions, performed without previous reverse
transcription, gave negative results, demonstrating that the
amplification was not caused by contaminating DNA. The same RNAs were
subjected to RT-PCR amplification using primers specific for the
glyceraldehyde-3-phosphate dehydrogenase (GAPDH) gene for messenger RNA
(mRNA) control. The amplification product for GAPDH cDNA is 174 bp. The
primer set flanks intron 3 of the gene, so that cDNA and the eventual
genomic DNA amplification products can be distinguished by size (29).
[3H]thymidine incorporation assays were performed as
described previously (30). NIH 3T3 transfectants, grown to confluence
in 24-well plates (Corning Costar Corp., Acton, MA), were serum-starved
for 24 h and then treated with GDNF (Alomone Laboratories,
Jerusalem, Israel) for 24 h in the presence of 4 µCi
[3H]thymidine/ml. Data are expressed as mitogenic index,
calculated as the fraction of stimulation obtained in the presence of
GDNF, with respect to the stimulation obtained with 1% calf serum. The
t test was used for the statistical ANOVA.
Protein studies
GDNF was purchased from Alomone Laboratories.
Immunoprecipitation and immunoblotting experiments were performed as
described elsewhere (30). Briefly, cells were lysed in a buffer
containing 50 mM HEPES (pH 7.5), 1% (vol/vol) Triton
X-100, 50 mM NaCl, 5 mM EGTA, 50 mM
NaF, 20 mM sodium pyrophosphate, 1 mM sodium
vanadate, 2 mM phenylmethylsulfonyl fluoride, and 0.2
µg/ml each of aprotinin and leupeptin. Lysates were clarified by
centrifugation at 10,000 x g for 15 min, and the
supernatant was processed for immunoblotting or for
immunoprecipitation. Protein concentration was estimated with the
Bio-Rad kit (Bio Rad Laboratories, Hercules, CA). To analyze
tyrosine-phosphorylation of Ret products, equal amounts of proteins
were immunoprecipitated and assayed for phosphotyrosine (pTyr) content.
Anti-Ret is a polyclonal antibody directed against the tyrosine-kinase
domain of Ret (30), and the anti-pTyr is a monoclonal antibody (4G10)
purchased from Upstate Biotechnology (Lake Placid, NY). To detect the
pTyr content of the Shc proteins, cells were serum-starved overnight,
stimulated or not with GDNF, and lysed as described above. Equal
amounts of total proteins were immunoprecipitated with a polyclonal
anti-Shc antibody (Upstate Biotechnology). The immunocomplexes were
divided in two aliquots and probed with either the anti-pTyr antibody
or with the anti-Shc antibody. Immunoblots were stained with
appropriate secondary antibodies and revealed with the Amersham ECL
system (Amersham Life Sciences, Buckinghamshire, UK). Treatment of the
cells with phosphatidylinositol-specific phospholipase C (PI-PLC)
(Boheringer, Mannheim, Germany) was performed as previously reported
(19). Briefly, cells were incubated with 1 U/ml PI-PLC (Boheringer,
Mannheim) at 37 C for 45 min, washed three times with serum-free
medium, and processed for Ret phosphorylation experiments.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
is a schematic
representation of the mutants assayed. To obtain a full response to
GDNF, we engineered NIH 3T3 cells coexpressing the different
ret constructs and GDNFR-. Marker (G418)-selected mass
populations were obtained. All the generated cell lines expressed
comparable levels of Ret proteins (see below). Furthermore, all the
cell lines expressed similar levels of GDNFR-, as demonstrated by a
semiquantitative RT-PCR assay (Fig. 2).
First, we evaluated the tyrosine phosphorylation of the different Ret
mutants after 5 min triggering with 50 ng/ml GDNF (Fig. 3). As reported previously (9, 11),
Ret/Cys634, Ret/Cys620, and Ret/Met918 (but not wild type Ret) proteins
showed a basal level of tyrosine phosphorylation. As expected, GDNF
induced prompt phosphorylation of wild-type Ret products. Among the
active Ret mutants, only Ret/Met918 was further stimulated by GDNF
triggering. GDNFR- is associated with the plasma membrane through a
GPI tail (19). It has been reported that treatment of cells expressing
GDNFR- with PI-PLC, which is able to digest the GPI anchor,
dramatically affects Ret response to GDNF (19). We have used this assay
to demonstrate that the observed effects were specific. Figure 3 shows
that pretreatment of the cells with PI-PLC greatly diminished
GDNF-induced tyrosine phosphorylation of both-wild type Ret and
Ret/Met918. As expected, HSCR-associated Ret mutants, Ret/HSCR972 and
Ret/HSCR32, were devoid of basal kinase activity. In addition, the
responsiveness to GDNF of HSCR-associated Ret mutants was much lower,
with respect to wild-type Ret proteins (Fig. 3).
|
|
|
(Jing et al., unpublished). Also
in this experiment, Ret/Met918, differently from the other Ret mutants,
responded to the stimulation (data not shown). In these conditions, the
addition of soluble GDNFR- was essential for efficient Ret
stimulation; indeed, a very weak response was obtained when cells were
stimulated with GDNF in the absence of soluble GDNFR- (data not
shown).
Ret activation results in proliferation of NIH 3T3 cells (2, 30). Thus,
we used a thymidine incorporation assay to determine whether
GDNF-induced Ret triggering leads to mitogenesis. GDNF dose-dependently
induced DNA synthesis in NIH 3T3 cells coexpressing Ret and GDNFR-
(Fig. 4) but not in untransfected cells
(data not shown). GDNF had a potent mitogenic effect on
Ret/Met918-expressing cells but not on Ret/Cys620- and
Ret/Cys634-expressing cells (Fig. 4). Consistent with the protein
phosphorylation data, cells expressing Ret/HSCR32 (Fig. 4) and
Ret/HSCR972 (data not shown) did not respond to GDNF. To evaluate
further these mitogenic effects, we compared the colony-forming
efficiency in soft agar of Ret/Met918-, Ret/Cys634-, and
Ret/HSCR32-expressing cells in the presence and absence of GDNF. As
reported previously (9), cells expressing Ret/Met918 and Ret/Cys634
showed a high clonogenic efficiency in soft agar. Ten days after
plating, the GDNF-stimulated Ret/Met918 colonies had increased in
number (80% vs. 45% of the plated cells) and in average
size (Fig. 5). In contrast, GDNF failed
to affect the colony-forming efficiency of Ret/Cys634-cells, in terms
of number (80% of plated cells) and size (Fig. 5).
Ret/HSCR32-expressing cells did not form colonies in soft agar, either
in the presence or in the absence of GDNF (data not shown).
|
|
, upon addition of GDNF to the cultures,
Shc phosphorylation promptly increased in the NIH 3T3-Ret cells. As
expected, basal Shc phosphorylation was higher in Ret/Met918-expressing
cells than in the wild-type Ret cells; however, this basal activation
was further increased by GDNF triggering. Also, Ret/Cys634 cells had a
basal level of Shc phosphorylation, but GDNF treatment did not increase
Shc phosphorylation (Fig. 6). Ret/Cys620- and Ret/HSCR-expressing cells
were totally unresponsive to GDNF (data not shown).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
to evaluate the responsiveness of these mutants
to GDNF. Our data demonstrate that the kinase and the biological
activities of Ret and Ret/MEN2B, although partly constitutive, are
triggered by GDNF. Indeed, GDNF treatment induced an increase of Shc
phosphorylation and DNA duplication in both Ret and Ret/MEN2B cells.
GDNF even induced the growth of Ret/MEN2B cells in soft agar. In
contrast, Ret/Cys634 and Ret/Cys620 failed to respond to GDNF.
It is well established that mutations detected in MEN2A, MEN2B, and
FMTC convert ret into a dominant oncogene by activating its
intrinsic kinase activity and transforming potential. MEN2A is
characterized by the occurrence of medullary thyroid carcinoma,
phaeochromocytoma, and hyperparathyroidism; MEN2B is an early onset and
very aggressive disease associated with medullary thyroid carcinoma,
phaeochromocytoma, ganglioneuromas of the intestinal tract, mucosal
neuromas, and skeletal abnormalities. Finally, medullary thyroid
carcinoma is the only FMTC disease phenotype (for reviews, see Refs. 37, 38). The different clinical manifestations of these syndromes have
been attributed to the different molecular mechanisms underlying
ret activation in the three conditions. In MEN2A, Ret
activation is triggered by a disulfide-bond-mediated dimerization of
the receptor that results in constitutive activation of the kinase (9, 10). A similar mechanism (i.e. dimerization) has been
proposed for cases of FMTC caused by cysteine mutations. However, there
is a strong correlation between specific Ret mutations and MEN2A and
FMTC phenotypes. In fact, mutation of codon 634 is responsible for more
than 80% of cases of MEN2A, and mutations of cysteines other than
codon 634 are more frequently associated with FMTC (8). This
correlation may be related to the observation that the FMTC-associated
mutations are less efficient than the MEN2A-associated mutation in
inducing Ret activation; a scarse supply of Ret to the plasma membrane
is associated with a lower level of dimerization and low Ret activation
(11, 12, 13). Also the MEN2B mutation, which affects the substrate
recognition pocket of the receptor, activates the Ret kinase, not by
inducing the formation of Ret dimers (9), but by altering the catalytic
specificity of the receptor (9, 14). It is likely that the different
responsiveness to ligand triggering observed here, between MEN2B and
FMTC/MEN2A Ret products, largely depends on the different mechanisms by
which these mutations activate Ret. At the present state of knowledge,
we may only speculate about why GDNF is not able to trigger activation
of MEN2A and FMTC mutants. Trupp and co-workers (21) demonstrated that
the Cys634 (MEN2A) mutation does not alter GDNF-GDNFR- binding to
Ret; and thus, the possibility of impaired binding can be excluded, at
least for this mutant. Because GDNF-GDNFR- binding has been reported
to activate Ret by inducing its dimerization (19), a more plausible
explanation is that the MEN2A and FMTC mutants, which are
constitutively dimerized, are already maximally activated and cannot be
further stimulated by GDNF. Alternatively, mutations of Cys634 and
Cys620 could alter the folding of the extracellular domain of Ret, thus
impairing the structural changes required to transmit the
GDNF-GDNFR- signal to the intracytoplasmic domain of Ret.
Whatever the mechanism of action, our data suggest that the tissue availability of the ligand may influence the expression of the biological effects of the MEN2B mutation, as well as the aggressiveness of the associated disease phenotype. When stimulated by GDNF, Ret/Met918 was a very potent oncoprotein. Thus, it is conceivable that the presence of the ligand may affect some of the disease phenotypes that characterize the MEN2B syndrome, such as the presence of gastro-intestinal neuromas, the skeletal abnormalities, and the lack of parathyroid involvement. Studies are in progress in our laboratory to ascertain whether the observations reported here are applicable to the NTN, the other functional ligand of Ret (25, 26). Other FMTC-associated mutations affect the Ret kinase domain, and not cysteine residues, thus leading to an activation of its transforming activity (39). Moreover, a novel intracytoplasmic mutation (codon 883) has been detected in some MEN2B families (40). We are currently investigating whether the ligand affects the activity of these mutants.
Two mechanisms of oncogenic activation of receptor tyrosine kinases have been proposed. In some cases, e.g. ret/PTC oncogenes, the extracellular encoding domain of the receptor is replaced by 5'-encoding sequences of heterologous genes (2). Similar to mutants of trk or met protooncogenes (for a review, see Ref. 41), these chimeric receptors cannot be modulated by their ligand. In other instances, single amino acid substitutions can activate receptor tyrosine kinases. Examples are ret mutations in MEN2 syndromes, and also met oncogene point mutations that have recently been described in both familial and sporadic tumors. In particular, the substitution of the methionine residue, corresponding to the ret/MEN2B residue (Met918), has been found in the met gene in papillary renal carcinomas (42). All these mutant receptor tyrosine kinases retain their extracellular encoding domain and, therefore, could still be able to interact with their ligand. Thus, ret/MEN2B could be a paradigmatic example of how the biological activity of these mutant receptors (and the associated disease phenotypes) might be influenced by the ligand.
It is well known that mutations of the Ret receptor responsible for HSCR are dispersed throughout the gene (15, 16). Cases of HSCR can be caused by large deletions of the gene, small intragenic deletions or insertions, splicing alterations, or nonsense mutations, which obviously cause a loss of function of Ret. Other cases are caused by more subtle missense mutations (for a review, see Ref. 43). In this study, we have analyzed two HSCR mutants belonging to the last category. Ret/HSCR32 and Ret/HSCR972 mutants showed a very low level of tyrosine phosphorylation and a scarse biological response to GDNF triggering. These results confirm, in a physiological setting (i.e., conditions under which the receptor is stimulated by its specific ligand) that HSCR mutations impair ret function by affecting its responsiveness to the ligand. Whereas the quiescence of the Ret/HSCR972 mutant is probably caused by impaired kinase function (17, 18), the lack of response to GDNF of the HSCR32 mutant is probably caused by its incorrect exposure on the outer cell surface (18). The HSCR972 mutant retained a low response to GDNF, rather than a complete knock-out, suggesting that even a partial loss of Ret function may lead to HSCR.
In conclusion, these findings indicate that the GDNF-Ret interaction
plays a role in the establishment of a MEN2B or HSCR phenotype and
suggest that GDNF and GDNFR- are possible modifier genes in the
expression of these syndromes. Moreover, the responsiveness of some of
the Ret mutants to GDNF raises the possibility of devising therapeutic
approaches that could intervene in this pathway.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 These authors contributed equally to the work.
3 Recipient of a Fogarty Scholar fellowship of the National
Institutes of Health, Bethesda, Maryland.
Received December 12, 1997.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  J. W. B. de Groot, T. P. Links, J. T. M. Plukker, C. J. M. Lips, and R. M. W. Hofstra RET as a Diagnostic and Therapeutic Target in Sporadic and Hereditary Endocrine Tumors Endocr. Rev., August 1, 2006; 27(5): 535 - 560. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. Mograbi, R. Bocciardi, I. Bourget, T. Juhel, D. Farahi-Far, G. Romeo, I. Ceccherini, and B. Rossi The Sensitivity of Activated Cys Ret Mutants to Glial Cell Line-Derived Neurotrophic Factor Is Mandatory To Rescue Neuroectodermic Cells from Apoptosis Mol. Cell. Biol., October 15, 2001; 21(20): 6719 - 6730. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Salvatore, R. M. Melillo, C. Monaco, R. Visconti, G. Fenzi, G. Vecchio, A. Fusco, and M. Santoro Increased in Vivo Phosphorylation of Ret Tyrosine 1062 Is a Potential Pathogenetic Mechanism of Multiple Endocrine Neoplasia Type 2B Cancer Res., February 1, 2001; 61(4): 1426 - 1431. [Abstract] [Full Text]
|
|
|
|
|
|
H. Le Hir, L. G. Colucci-D’Amato, N. Charlet-Berguerand, P.-F. Plouin, X. Bertagna, Vittorio de Franciscis, and C. Thermes High Levels of Tyrosine Phosphorylated Proto-Ret in Sporadic Pheochromocytomas Cancer Res., March 1, 2000; 60(5): 1365 - 1370. [Abstract] [Full Text]
|
|
|
|
|
|
R. M. Melillo, M. V. Barone, G. Lupoli, A. M. Cirafici, F. Carlomagno, R. Visconti, B. Matoskova, P. P. Di Fiore, G. Vecchio, A. Fusco, et al. Ret-mediated Mitogenesis Requires Src Kinase Activity Cancer Res., March 1, 1999; 59(5): 1120 - 1126. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Lindahl, D. Poteryaev, L. Yu, U. Arumae, T. Timmusk, I. Bongarzone, A. Aiello, M. A. Pierotti, M. S. Airaksinen, and M. Saarma Human Glial Cell Line-derived Neurotrophic Factor Receptor alpha 4 Is the Receptor for Persephin and Is Predominantly Expressed in Normal and Malignant Thyroid Medullary Cells J. Biol. Chem., March 16, 2001; 276(12): 9344 - 9351. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Qiao, T. Iwashita, T. Furukawa, M. Yamamoto, G. Sobue, and M. Takahashi Differential Effects of Leukocyte Common Antigen-related Protein on Biochemical and Biological Activities of RET-MEN2A and RET-MEN2B Mutant Proteins J. Biol. Chem., March 16, 2001; 276(12): 9460 - 9467. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
