help button home button Endocrine Society JCEM
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Submit a related Letter to the Editor
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Copyright Permission
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Ivan, M.
Right arrow Articles by Wynford-Thomas, D.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Ivan, M.
Right arrow Articles by Wynford-Thomas, D.
The Journal of Clinical Endocrinology & Metabolism Vol. 82, No. 8 2702-2709
Copyright © 1997 by The Endocrine Society

Experimental Studies

An Amphotropic Retroviral Vector Expressing a Mutant gsp Oncogene: Effects on Human Thyroid Cells in Vitro1

M. Ivan, M. Ludgate, V. Gire, J. A. Bond and D. Wynford-Thomas

Clinical Research Center Thyroid Tumor Biology Research Group, University of Wales College of Medicine, Cardiff, United Kingdom CF4 4XN

Address all correspondence and requests for reprints to: Prof. D. Wynford-Thomas, Clinical Research Center Thyroid Tumor Biology Research Group, University of Wales College of Medicine, Heath Park, Cardiff, United Kingdom CF4 4XN.


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
Point mutations of the gsp protooncogene (encoding the {alpha}-subunit of the Gs protein) that constitutively activate the cAMP signaling pathway are a common feature of and a plausible causative mechanism for thyroid hyperfunctioning adenomas (hot nodules). To investigate the extent to which mutant gsp acting alone can induce proliferation of thyroid follicular cells, we generated an amphotropic retroviral vector (based on the pBABE-neo plasmid and psi-CRIP packaging line) to permit stable introduction of a hemagglutinin-tagged Gln227->Leu mutant gsp gene into normal human thyrocytes in vitro. The biological activity of the vector was confirmed by detection of HA-tagged Gsp protein expression and induction of cAMP synthesis in selected target cells. Normal human thyroid follicular cells in primary monolayer culture were infected with the gsp retroviral vector or with corresponding vectors expressing mutant H-ras or neo only as positive and negative controls, respectively. Although, as before, mutant ras generated 10–20 well differentiated epithelial colonies/dish of 105 infected cells, with an average lifespan of 15–20 population doublings, only small groups of no more than 15–50 differentiated thyrocytes were observed with the gsp vector. In addition to standard conditions (10% FCS), infections were performed in reduced serum (1% FCS, TSH, and insulin), in the presence of isobutylylmethylxanthine, or in the presence of agents capable of closing gap junctions, with no significant difference in outcome. Although little or no proliferative response was observed regardless of the conditions, there was clear evidence of morphological response (rearrangement of the actin cytoskeleton and increased cell size). The results suggest that gsp mutation may not be a sufficient proliferogenic stimulus by itself to account for hot nodule formation.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
SOMATIC point mutations at codons 201 and 227 of the gsp protooncogene (coding for the {alpha}-subunit of the hetero-trimeric Gs protein) cause constitutive activation of the adenylate cyclase pathway and are commonly associated with GH-secreting human pituitary adenomas (1), thyroid hyperfunctioning adenomas (hot nodules), and a low proportion of well differentiated thyroid carcinomas (2, 3, 4). Mutations in the TSH receptor gene, resulting in a similar overactivity of the cAMP cascade, account for an even greater proportion of thyroid toxic nodules (4, 5, 6).

In human thyrocytes in vitro, agents stimulating cAMP production have been shown to have mitogenic effects (7, 8, 9). Likewise, transgenic mice selectively expressing an A2 adenosine receptor in thyroid (which constitutively activates the adenylyl cyclase pathway) show early and homogeneous follicular cell hyperplasia (10). A causal role of gsp mutation (or, alternatively, TSH receptor mutations) has, therefore, been proposed.

However, other lines of evidence suggest that elevation of cAMP may not be sufficient to induce sustained thyrocyte proliferation without additional abnormalities. For example, transgenic mice expressing the Arg201->His Gs{alpha} subunit under the control of the bovine Tg promoter developed only late and sporadic thyroid nodules without generalized hyperplasia (11). Similarly, patients with McCune-Albright syndrome who are mosaic for activating gsp mutations (12), usually exhibit a thyroid gland of normal or marginally increased size, with a much smaller number of nodules detected by ultrasonography (13) than might be predicted from the proportion of thyrocytes expected to possess the genetic abnormality. Furthermore, germ line-activating mutations in the TSH receptor gene usually result in a predominantly multinodular goiter (14, 15). (Although there is a background homogeneous hyperplasia, if this involves all the cells, it must amount to no more than a few population doublings.) Indeed, a recently discovered germ-line TSH receptor mutation induced severe hyperthyroidism, yet the glands of the two affected patients were entirely normal in size (16).

The extent to which an activated gsp oncogene acting alone can stimulate cell proliferation has been tested experimentally in several cell line models. In Swiss 3T3 cells, for example, the Gln227->Leu gsp mutant stimulated proliferation only in the presence of forskolin and a phosphodiesterase inhibitor (17). In the rat thyroid FRTL5 line, expression of the Gln227->Leu mutant conferred TSH-independent growth (18), but required phosphodiesterase inhibitors for optimal response (19).

The major drawback of all of these approaches stems from the artificiality of the models employed. The tissue culture data have all been based on cell lines rather than normal cells, raising the possibility of interaction with unknown preexisting abnormalities. Conversely, transgenic mice suffer from the fact that the gsp oncogene is expressed from very early in development and in all thyrocytes, providing, therefore, a potentially unrepresentative model for tumors arising from a sporadic somatic mutation in a single adult cell. Finally, both fail to take account of the possibility of more stringent controls on human compared to rodent cell proliferation that have been frequently noted in other cell types (reviewed in 20 .

Here we have attempted to overcome some of these limitations by exploiting the ability of amphotropic retroviral vectors to stably express mutant oncogenes in normal adult human thyrocytes in monolayer culture, an approach that has been extensively validated in our laboratory in relation to two other putative initiators of thyroid tumorigenesis, ras and ret (21, 22). Our results using such a vector encoding mutant gsp do not support the ability of this gene to initiate sustained clonal proliferation of thyrocytes in the absence of cooperating abnormalities.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Cells and culture conditions

Monolayer cultures were prepared from surgical samples of normal human thyroid tissue (from 12 glands) by protease digestion and mechanical disaggregation, as described previously (23). Cultures were maintained in a 2:1:1 mixture of DMEM, Ham’s F-12, and MCDB 104 (21) supplemented with 1% or 10% FCS (Life Technologies, Paisley, UK). In the experiments using 1% FCS, a mixture of 10 µg/mL insulin, 5 µg/mL transferrin, and 40 µg/mL ascorbic acid (all from Calbiochem-Novabiochem, Nottingham, UK) with or without 300 µU/mL TSH (Sigma Chemical Co., Poole, UK) was added. The A431 and Chinese hamster ovary (CHO) cell lines were cultured in RPMI 1640 and Ham’s F-12, respectively, supplemented with 10% FCS.

Retroviral vectors

A pcDNA Id plasmid with a full-length complementary DNA coding a hemagglutinin-tagged, Gln227->Leu mutant of the rat Gs{alpha} was kindly donated by Dr. Henry Bourne (University of California, San Francisco, CA). [Rat and human Gs{alpha} differ at just one amino acid (codon 139) (24).] After double digestion with BstXI and XhoI, the gsp complementary DNA insert was subcloned into the BstXI and SalI sites of plasmid pBABE-neo. The resulting pBABE-gsp plasmid was used to transfect the Omega E ecotropic packaging line, viral supernatant from which was, in turn, used to infect the amphotropic producer cell line psi-CRIP (25). Resulting clones were selected using G418 viral titer assessed by stable transduction of G418 resistance to a human epithelial target cell line (A431) (21).

The producer cell lines, psi-CRIP-neo (containing the parent vector pBABE-neo only) and psi-CRIP-DOEJ, expressing a mutant H-ras (Val12) gene, have been described previously (21).

Retroviral gene transfer

Primary cultures were plated at 5 x 105/60-mm dish and infected 2 days later with retrovirus-containing medium from near-confluent producer cells (containing 8 µg/mL polybrene) (21). Four days later, cells were passaged with or without G418 selection. For cultures performed in 1% FCS, TSH, insulin, transferrin, and ascorbic acid were added on day 1, infection was performed on day 4, and TSH was withdrawn 2 days later. Infection of cell lines (A431 and CHO) was carried out as previously described (26).

Reverse transcription-PCR (RT-PCR)

Total ribonucleic acid (RNA) was extracted using an RNAzol B kit (Biotecx, Bournemouth, UK). One microgram was subjected to RT-PCR using primers designed to amplify selectively the mutant rat gsp sequence and not the endogenous wild-type human (upstream primer, 5'-CCAAACTTTGACTTCCCACCT-3'; downstream primer, 5'-CGCAGGCCGCCCACATCG-3') selected from 2 regions of maximum divergence in nucleotide (but not amino acid) sequence. RNA was reverse transcribed for 15 min at 42 C using the downstream primer, followed by inactivation of the reverse transcriptase at 95 C for 5 min. The upstream primer was subsequently added, and 35 cycles of amplification were carried out (1 min at 95 C, 1 min at 56 C, 1 min at 72 C). PCR products were examined by agarose gel electrophoresis.

Immunocytochemistry

A431 cells were fixed in 4% formaldehyde (10 min); permeabilized in 0.2% Tween-20 (30 min); blocked in a cocktail of 10% dried milk powder, 1% BSA, and 0.2% Tween-20 in phosphate-buffered saline; and incubated for 1 h with 10 µg/mL 12CA5 monoclonal antibody (Boehringer Mannheim, Mannheim, Germany) against the HA tag. After extensive washes in phosphate-buffered saline-Tween, the second antibody (horseradish peroxidase-coupled rabbit antimouse Ig) was added for 45 min, and binding sites were visualized with diaminobenzidene. A similar protocol, but with longer fixation (30 min) and shorter permeabilization (5 min), was applied for gsp-infected thyroid cells.

Western analysis

Cells (5 x 106) were lysed in 100 µL NET buffer (150 mmol/L NaCl, 50 mmol/L Tris, 1% Nonidet P-40, and 10 µmol/L aprotinin) and centrifuged at 20,000 x g for 30 min. Twenty microliters of rabbit polyclonal IgG anti-Gs{alpha}/Golf (Santa Cruz Biotechnology, Santa Cruz, CA) was used to immunoprecipitate both the wild-type and mutant Gs{alpha} from infected cells (overnight at 4 C). Immunocomplexes were extracted with protein G-Sepharose (Pharmacia, Piscataway, NJ), separated on a 12% SDS-polyacrylamide gel alongside mol wt markers (Bio-Rad Laboratories, Richmond, CA), electroblotted onto polyvinylidene difluoride membrane, and probed with 1 µg/mL 12CA5 monoclonal anti-HA tag antibody. Binding sites were visualized using the ECL system (Amersham, Amersham, UK).

cAMP assay

CHO cells were infected with psi-CRIP-gsp or the control psi-CRIP-neo vectors using the same protocol as that for normal thyroid cells (21). The resulting G418-resistant clones were pooled and subjected to cAMP assay (Amersham). Briefly, CHO gsp or CHO neo cells were seeded in 24-microtiter plates at a density of 3 x 105 cells/well in Ham’s F-12 medium. The next day, all cells were incubated for 3 h in medium containing 2 mmol/L isobutylmethylxanthine (IBMX; Calbiochem, La Jolla, CA) in the presence and absence of 10-6 mol/L forskolin (Sigma). Cells were then lysed in 500 µL 0.1 mmol/L HCl, and the suspension was evaporated to dryness and resuspended in 120 µL ethylenediamine tetraacetate buffer. The cAMP assay was then performed according to the manufacturer’s protocol. All determinations were performed in triplicate.

Microinjection

Gap junction-mediated communication between human thyrocytes was assayed by cell to cell transfer of the low molecular weight fluorescent dye Lucifer Yellow (27). Medium was replaced with Leibowitz’s L-15 supplemented with 10% FCS, and 5% (wt/vol) Lucifer Yellow (Sigma) in 0.3 mol/L LiCl was microinjected into the cytoplasm of the thyroid cells under nitrogen pressure applied for 0.1–0.3 s using an Eppendorf system (Micromanipulator 5170 and Microinjector 5242, Carl Zeiss, Oberkochen, Germany). After allowing 10 min for diffusion, cells were fixed in 4% paraformaldehyde (30 min at room temperature), and the extent of Lucifer Yellow transfer was assessed by counting the number of labeled cells adjacent to each microinjected cell by fluorescence microscopy.

Proliferation assay

Seven, 14, and 21 days after starting G418 selection, [3H]thymidine ([3H]TdR; 86 µCi/mmol; New England Nuclear-DuPont, Brussels, Belgium) was added to cultures for 24 h to a final concentration of 10 µCi/mL. Dishes were then washed, fixed in methanol-acetic acid (3:1, vol/vol), and processed for emulsion autoradiography as described previously (23).


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
Validation of retroviral vector encoding mutant gsp

All five clones of psi-CRIP-gsp examined gave a titer of above 105 colony-forming units/mL, as assessed by transduction of G418 resistance to A431 cells, similar to that of our neo only and mutant ras producer cells. Clone 17, which gave the highest titer (5 x 105 colony-forming units/mL), was chosen for the experiments described below.

Transcription of a stably transduced mutant gsp gene was demonstrated in A431 target cells by RT-PCR analysis of total RNA using primers specific for the exogenous (rat) gsp sequence. The expected 276-bp fragment was observed in cells infected with psi-CRIP-gsp, but not in controls infected with psi-CRIP-neo alone (Fig. 1Go).



View larger version (65K):
[in this window]
[in a new window]
  		Figure 1. RT-PCR with primers specific for the exogenous (rat) sequence demonstrates the presence of the expected 276-bp transcript fragment in A431 cells infected with the gsp retrovirus produced by different producer clones (lanes 2–6), but not in controls infected with psi-CRIP-neo alone (lane 1). Molecular mass markers, pBR322 digested with TaqI, are shown (lane 7).

 
Translation of the viral transcript was demonstrated by immunocytochemistry, making use of the HA-tag to specifically detect the vector-encoded Gsp protein. Immunostaining with anti-HA antibody was detected in the cytoplasm of A431 cells infected with the gsp vector, but not in those infected with the neo control (Fig. 2Go, a and b). This result was confirmed by immunoprecipitation/Western analysis, which revealed the expected 47-kDa HA-tagged protein only in gsp-infected cells (Fig. 2cGo).



View larger version (49K):
[in this window]
[in a new window]
  		Figure 2. Immunostaining with anti HA antibody shows cytoplasmic expression of the tagged mutant Gs{alpha} in A431 cells infected with psi-CRIP-gsp (a) which is absent in control cells infected with psi-CRIP-neo alone (b). The apparent presence of staining in the latter is due to retention of optical contrast under the conditions of mounting used. Magnification, x100. Immunoprecipitation/Western analysis (c) confirmed the expression of a 47-kDa Gsp protein in gsp-infected A431 cells (lane 1), but not in the neo control (lane 2).

 
Finally, to demonstrate that the protein produced from the vector was functional, cAMP assay was performed on CHO target cells (chosen for their low basal level). Whereas with this tritium-based assay, levels were below detection in control (neo only) CHO cells, infection with psi-CRIP-gsp resulted in readily detectable levels (24 ± 0.6 pmol/106 cells), which were even higher than those produced by stimulation for 3 h with 10-6 mol/L forskolin (14.8 ± 0.25 pmol/106 cells; Fig. 3Go).



View larger version (14K):
[in this window]
[in a new window]
  		Figure 3. CHO cells infected with psi-CRIP-gsp contain readily detectable levels of cAMP comparable to that in forskolin-treated control. The level in control neo-infected CHO cells was below the detection limit of this tritium-based assay (means of three replicate assays are shown ± SEM).

 
Taken together, these data confirm the capacity of the psi-CRIP-gsp retroviral vector to transduce a fully functional mutant Gs{alpha} protein.

Retroviral gene transfer into primary thyroid cells

After infection of primary monolayers in 10% FCS with the psi-CRIP-neo (negative control) vector and G418 selection, no colonies of well differentiated thyrocytes were obtained, consistent with the known very limited proliferative capacity of these cells. Conversely, infection with psi-CRIP-DOEJ (positive control) vector led to the outgrowth of multiple colonies (~20/dish of 105 cells infected) of rapidly dividing epithelial cells that retained thyroid-specific differentiation and continued to proliferate for, on the average, 15–20 population doublings, as described previously (21, 28). In contrast, infection with psi-CRIP-gsp (which displayed a similar titer on A431 cells) did not generate any colonies of differentiated thyrocytes from any of 12 different sources of thyroid cells infected. The outcome was identical regardless of whether cells were passaged after infection.

In neo infections, a variable number of colonies were observed consisting of a variant epithelial cell type that we have previously identified (29) to be present at low frequency in normal thyroid primary cultures (Fig. 4aGo). These cells show a more mesenchymal morphology and growth pattern, although initially expressing some thyroid markers (30) and, importantly, have a much greater proliferative capacity than the well differentiated majority.



View larger version (145K):
[in this window]
[in a new window]
  		Figure 4. a, Phase contrast photomicrograph showing a variant colony of thyroid epithelial cells with pseudomesenchymal morphology and high proliferative capacity in 10% FCS, obtained after infection of primary cultures with the psi-CRIP-neo vector. b, Increase in cell size and reorganized cytoskeleton in variant cells obtained after infection with psi-CRIP-gsp (note the same magnification as in a). Immunofluorescence with an antiactin antibody shows diffuse cytoplasmic globular actin in neo variants (c). By contrast, bundles of fibrilar actin were detected in the cells infected with psi-CRIP-gsp (d). One or 2 weeks of treatment with forskolin induced a similar fibrilar pattern, but without cell hypertrophy, in variant cells (e). Magnification: a, b, and e, x100; c and d, x200.

 
These colonies were observed after gsp infection and G418 selection at approximately the same frequency as in neo only controls. However, a large proportion of the gsp-expressing cells, beginning at the periphery of colonies, rapidly underwent a morphological alteration characterized by increase in cell size and cytoskeletal reorganization (Fig. 4bGo). Immunofluorescence using a rabbit polyclonal antibody (Sigma) demonstrated that the cytoskeletal fibrillar structures were derived from bundles of actin filaments (Fig. 4Go, c and d). Interestingly, a very similar phenotype (increased size and similar filaments) was observed in FRTL5 infected with mutant gsp or with a retroviral vector expressing a mutant TSH receptor (data not shown). Stimulation of the variant colonies obtained after neo-only infection with 10-6 mol/L forskolin for 1–2 weeks partially reproduced this phenotypic change (Fig. 4eGo), although no increase in cell size was observed.

The greater proliferative capacity of the variant cells compared to differentiated thyrocytes permitted analysis of the effect of mutant gsp on their growth rate. Pulse labeling with [3H]TdR showed no significant effect; nuclear labeling indexes for neo vs. gsp colonies were 31 ± 0.7% vs. 35.5 ± 0.9% at 7 days (mean ± SEM), 19.6 ± 1.2% vs. 18.5 ± 1.1% at 14 days, and 12.2 ± 0.9% vs. 9.2 ± 0.6% at 21 days after the beginning of G418 selection.

Infection in reduced serum concentration

It has been shown that 10% FCS (and epidermal growth factor) abolish the proliferative response of human thyrocytes to TSH (31) or, in other experiments (32), that the mitogenic effect of TSH disappears after 5 days in 10% FCS. As retroviral vectors only target cells that are actively replicating at the time of infection (33), the use of 10% FCS may, therefore, favor targeting of less differentiated cells, which may be intrinsically unresponsive to cAMP and also may inhibit differentiated cells from responding even if they take up the gsp vector. To correct this potential bias, we repeated the experiments maintaining cells before infection in a reduced serum medium containing TSH (0.3 mU/mL), insulin (10 µg/mL), and transferrin (5 µg/mL), which should maintain the differentiated state of the follicular cell while still stimulating proliferation (7, 9). Consistent with this, TSH-stimulated thyroid cells at the time of infection (days 4–5) showed an up to 6-fold increase in the nuclear 24-h bromouridine labeling index compared to nontreated cells (data not shown). TSH was withdrawn 2 days after infection.

Under these conditions, infection with psi-CRIP-neo again failed to generate colonies of well differentiated thyrocytes. Variant colonies (Fig. 5aGo) were obtained, but, as predicted, their growth rate was slowed compared to that in 10% FCS.



View larger version (149K):
[in this window]
[in a new window]
  		Figure 5. Results of retroviral gene transfer in reduced serum conditions. a, Colony of variant cells obtained after psi-CRIP-neo infection as described in Fig. 4aGo, but using 1% FCS. b, Morphologically well differentiated epithelial colony induced by mutant ras. c, Small colonies of G418-resistant, well differentiated cells, following infection with psi-CRIP-gsp. d, Spontaneous increase in cell size and reorganization of the cytoskeleton observed after 10 days in a previously morphologically well differentiated small colony induced by gsp. e, Immunocytochemistry with an anti-HA monoclonal antibody demonstrates expression of the retrovirally transduced Gsp protein in large flattened cells from a colony similar to that shown in d. a–d: phase contrast, x100; e: immunostaining, x100, not counterstained.

 
Mutant ras still induced the formation of approximately 10 morphologically well differentiated colonies/105 infected cells, the phenotypic features of which were identical to those obtained in high serum conditions (Fig. 5bGo).

Infection with psi-CRIP-gsp, however, in contrast to the negative result in 10% FCS, led to the emergence, within a few weeks after G418 selection, of small colonies of cells with closely similar morphology to that of normal thyrocytes, with a very variable yield of less than 20/dish of 105 infected cells. These ceased proliferating at an average size of 10–20 cells and never exceeded 50 cells (Fig. 5cGo). Within 1–2 weeks, most colonies underwent a change in morphology very similar to that seen in variant colonies, i.e. a dramatic increase in cell size and reorganization of the cytoskeleton (Fig. 5dGo). Immunocytochemistry with the anti-HA antibody demonstrates the expression of the vector-encoded Gsp protein in the cytoplasm of these flattened cells (Fig. 5eGo).

As with the neo control, small colonies of variant cells were obtained after gsp infection, which continued to proliferate up to an average of several hundred cells. Again, [3H]TdR labeling showed that gsp expression did not significantly influence the proliferation rate compared to the effect of neo infection alone; labeling indexes in gsp vs. neo colonies were 16 ± 1.2% vs. 14.2 ± 0.9% at 7 days (mean ± SEM), 14.4 ± 0.4% vs. 11.5 ± 0.7% at 14 days, and 4 ± 0.3% vs. 6.3 ± 0.5% at 21 days after beginning G418 selection.

Modification of gap junctions

We were concerned that, given its low molecular mass (330 Da), cAMP generated in an individual gsp-infected cell could diffuse through gap junctions (34, 35) into neighboring thyrocytes, thereby diluting a potential growth stimulatory signal, an effect that might be exaggerated in gsp-infected cells given the reported stimulation by cAMP of gap junctional communication (27).

Microinjection of Lucifer Yellow into normal follicular cells revealed a variable extent of cell-cell communication; the number of adjacent cells labeled in 10 min varied from 0–30 (mean ± SEM, 8 ± 0.8). Addition of a known inhibitor of gap junctions, 18{alpha}-glycyrrhetinic acid (36, 37) at 10 µmol/L 30 min before microinjection of Lucifer Yellow resulted in 100% inhibition of intercellular communication, as revealed by the total block of transfer of dye into surrounding cells, without any observable toxic effect on either normal cells or ras-induced colonies. Treatment with 18{alpha}-glycyrrhetinic acid at this concentration from the second day after infection, however, failed to produce any significant effect on the response to psi-CRIP-gsp.

Other potential modifiers

Given the well recognized importance of blocking phosphodiesterase in maximizing cAMP responses in other experimental models (16, 19), we also examined the influence of the phosphodiesterase inhibitor IBMX. Maintenance of IBMX at 10-4 mol/L in the medium from the time of gsp or neo infection had no effect on the yield or behavior of the variant colonies, but, interestingly, prevented the induction by gsp of the morphologically well differentiated colonies in 1% FCS.


   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
We have successfully constructed a high titer amphotropic retroviral vector psi-CRIP-gsp clone 17, which, on the basis of validation in several cell line models, is able to stably transduce expression of a biologically active mutant gsp oncogene. Although accurate validation was not possible in thyroid cells due to the lack of proliferation, this vector reproducibly induced a phenotypic change consisting of hypertrophy and cytoskeletal reorganization in colonies derived from both variant and (in low serum) morphologically well differentiated thyrocytes. Furthermore, immunostaining confirmed the presence of HA-tagged Gsp protein in the latter. Similar cytoskeletal changes could be observed following treatment with known cAMP-elevating agents.

Taken together, therefore, this provides compelling evidence that our gsp vector is successfully targeting normal thyrocytes in monolayer culture at a frequency not significantly different from that obtained previously with mutant ras and ret vectors with a similar titer (22). Despite this, however, in standard culture conditions (10% FCS), no colonies of differentiated thyrocytes were ever obtained, and in the modified (1% FCS) medium, only abortive groups of no more than 50 such cells were observed.

Although the acute mitogenic effect on normal thyroid cells of agents capable of activating the cAMP pathway (TSH, forskolin, cholera toxin, 8-bromo-cAMP) is well established (7, 8, 9, 23), longer term experiments show that the final increase in cell number (DNA content) nearly always represents little more than a few extra divisions (population doubling) (8, 38). Likewise, there is also no evidence that cAMP elevation can induce the formation of colonies of well differentiated thyrocytes in sparsely seeded primary cultures. It could be argued, therefore, that the lack of sustained proliferation in response to mutant gsp expression was a predictable outcome.

There are, however, several potentially significant differences between our retroviral vector-based system and the experiments described above. First, in our model, as would be the case in naturally occurring tumors, only rare individual cells are targeted by the cAMP-elevating stimulus, as opposed to the entire population of cells in experiments using TSH or pharmacological agonists. There may be important differences between these two scenarios in the extent to which potential thyrocyte growth is influenced by paracrine controls.

Second, parallel work with our gsp vector on the GH3 rat pituitary cell line indicates that mutant Gs{alpha} may not be functionally identical to exogenous agonists of the cAMP pathway. For example, activated gsp induced a 10-fold increase in GH and PRL secretion, whereas forskolin and dibutyryl cAMP produced only a 1.5- to 2-fold increase despite inducing a higher level of cAMP (39). Finally, even cholera toxin, which would be expected to mimic mutant gsp most closely because it ADP-ribosylates one of the residues affected by some gsp mutations (201), has now been reported to also act on other cellular proteins (40, 41, 42, 43), raising the possibility of additional effects. Even this agent, therefore, may not be exactly equivalent to mutant Gs{alpha}.

Several explanations for our largely negative findings need to be considered. First, as retroviral vectors can only target dividing cells, it is theoretically possible that psi-CRIP-gsp is not targeting the relevant subpopulation of thyrocytes. Evidence for at least functional heterogeneity in thyroid is well documented (44, 45). Against this, however, gsp still failed to induce significant proliferation in experiments in which TSH and low serum levels were used, so as to maximize the chance that the subpopulation of cells in cycle at the time of infection would respond to cAMP.

Second, the human thyrocyte in primary culture may have undergone sufficient dedifferentiation to uncouple its proliferative response to cAMP. It is of course well recognized that culture of such polarized epithelial cells in monolayer on plastic distorts their normal architecture, with the basal surface which in vivo is normally exposed to growth factors (including TSH) becoming inaccessible due to attachment to the plastic. This problem can be potentially overcome using a two-compartment culture chamber approach (reviewed in 46 . However, in our hands, the use of such a system did not permit any increase in the proliferative response of human thyroid cells to TSH, for example (Wynford-Thomas, D., unpublished observations). Most importantly, rodent experiments have clearly shown that even in the intact tissue, long term exposure to circulating levels of TSH, sufficient to induce sustained stimulation of thyroid function, induces only very limited thyrocyte proliferation (averaging approximately three population doublings) (47). In conclusion, there seems to be little to support the idea that thyrocytes in monolayer culture should exhibit a significantly different proliferative response to cAMP compared to that of the intact gland, although we recognize that this possibility cannot be formally excluded.

Taken together, therefore, we suggest that the most likely explanation for our data is that they do indeed reflect the in vivo situation and that additional genetic or epigenetic events are required for the generation of a hyperfunctioning adenoma. The number of cell divisions induced by mutant gsp (no more than six) is far too low to account for the formation of a tumor and is insufficient for a second event to occur with a reasonable probability. The gsp mutation, therefore, may not be the first in the sequence of events that leads to a hot nodule, but may follow clonal expansion initiated by some other unknown event.


   Acknowledgments
 
We are grateful to M. F. Haughton for preparation of the normal thyroid cells.


   Footnotes
 
1 This work was supported by grants from the U.K. Cancer Research Campaign and Medical Research Council (to M.L.). Back

Received January 3, 1997.

Revised March 10, 1997.

Accepted April 24, 1997.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Landis CA, Masters SB, Spada A, Pace AM, Bourne HR, Vallar L. 1989 GTP-ase inhibiting mutations activate the {alpha} chain of Gs and stimulate the adenylyl cyclase in human pituitary tumours. Nature. 340:692–696.[CrossRef][Medline]
  2. O’Sullivan C, Barton CM, Staddon SL, Brown CL, Lemoine NL. 1991 Activating point mutations of the gsp oncogene in human thyroid adenomas. Mol Carcinogen. 4:345–349.[Medline]
  3. Suarez HG, du Villard JA, Caillou B, Schlumberger M, Parmentier C, Monier R. 1991 gsp mutations in human thyroid tumours. Oncogene. 6:677–679.[Medline]
  4. Russo D, Arturi F, Wicker R, et al. 1995 Genetic alterations in thyroid hyperfunctioning adenomas. J Clin Endocrinol Metab. 80:1347–1351.[Abstract]
  5. Parma J, Duprez L, Van Sande, et al. 1993 Somatic mutations in the thyrotropin receptor gene cause hyperfunctioning thyroid adenomas. Nature. 365:649–651.[CrossRef][Medline]
  6. Russo D, Arturi F, Suarez HG, et al. 1996 Thyrotropin receptor gene alterations in thyroid hyperfunctioning adenomas. J Clin Endocrinol Metab. 81:1548–1551.[Abstract]
  7. Roger PP, Dumont JE. 1987 Thyrotropin is a potent growth factor for normal human thyroid cells in primary culture. Biochem Biophys Res Commun. 149:707–711.[CrossRef][Medline]
  8. Williams DW, Wynford-Thomas D, Williams ED. 1987 Control of human thyroid follicular cell proliferation in suspension and monolayer culture. Mol Cell Endocrinol. 51:33–40.[CrossRef][Medline]
  9. Roger PP, Taton M, Van Sande J, Dumont JE. 1988 Mitogenic effects of thyrotropin and adenosine 3',5'-monophosphate in differentiated normal human thyroid cells in vitro. J Clin Endocrinol Metab. 66:1158–1165.[Abstract/Free Full Text]
  10. Ledent C, Dumont JE, Vassart G, Parmentier M. 1992 Thyroid expression of an A2 adenosine receptor transgene induces thyroid hyperplasia and hyperthyroidism. EMBO J. 11:537–542.[Medline]
  11. Michiels FM, Caillou B, Talbot M, et al. 1994 Oncogenic potential of guanine nucleotide stimulatory factor {alpha} subunit in thyroid glands of transgenic mice. Proc Natl Acad Sci USA. 91:10488–10492.[Abstract/Free Full Text]
  12. Weinstein L, Shenker A, Gejman PV, Friedman E, Spiegel AM. 1991 Activating mutations of the stimulatory G protein in the McCune-Albright syndrome. N Engl J Med. 325:1688–1695.[Abstract]
  13. Feuillan PP, Shawker T, Rose SR, Jones J, Jeevanram RK, Nisula B. 1990 Thyroid abnormalities in the McCune-Albright syndrome: ultrasonography and hormonal studies. J Clin Endocrinol Metab. 71:1596–1601.[Abstract/Free Full Text]
  14. Duprez L, ParmaJ, Van Sande J, et al. 1994 Germline mutations in the thyrotropin receptor gene cause non autoimmune autosomal dominant hyperthyroidism. Nat Genet. 7:396–401.[CrossRef][Medline]
  15. Kopp P, Van Sande J, Parma J, et al. 1995 Congenital hyperthyroidism caused by a mutation in the thyrotropin-receptor gene. N Engl J Med. 332:150–154.[Free Full Text]
  16. Derwahl M. 1996 Editorial: TSH receptor and Gs-{alpha} gene mutations in the pathogenesis of toxic thyroid adenomas: a note of caution. J Clin Endocrinol Metab. 81:2783–2785.[Abstract/Free Full Text]
  17. Zachary I, Masters SB, Bourne HR. 1990 Increased mitogenic responsiveness of Swiss 3T3 cells expressing costitutively active Gs{alpha}. Biochem Biophys Res Commun. 168:1184–1193.[CrossRef][Medline]
  18. Muca C, Vallar L. 1994 Expression of mutationally activated G{alpha}s stimulates growth and differentiation of FRTL5 cells. Oncogene. 9:3647–3653.[Medline]
  19. Nemoz G, Sette C, Hess M, Muca C, Vallar L, Conti M. 1995 Activation of cyclic nucleotide phosphodiesterases in FRTL5 thyroid cells expressing a constitutively active Gs{alpha}. Mol Endocrinol. 9:1279–1287.[Abstract/Free Full Text]
  20. Wynford-Thomas D. 1993 In vitro models of thyroid cancer. Cancer Surv. 16:115–134.[Medline]
  21. Bond J, Dawson T, Lemoine N, Wynford-Thomas D. 1992 Effect of serum growth factors and phorbol ester on growth and survival of human thyroid epithelial cells expressing mutant ras. Mol Carcinogen. 5:129–135.[Medline]
  22. Bond J, Wyllie FS, Rowson J, Radulescu A, Wynford-Thomas D. 1994 In vitro reconstruction of tumour initiation in a human epithelium. Oncogene. 9:281–290.[Medline]
  23. Williams DW, Williams ED, Wynord-Thomas D. 1988 Loss of dependence in IGF-1 for proliferation of human thyroid adenoma cells. Br J Cancer. 57:535–539.[Medline]
  24. Kosaza T, Itoh H, Tsukamoto T, Kaziro Y. 1988 Isolation and characterization of the human Gs{alpha} gene. Proc Natl Acad Sci USA. 85:2081–2085.[Abstract/Free Full Text]
  25. Danos O, Mulligan RC. 1988 Safe and efficient generation of recombinant retroviruses with amphotropic and ecotropic host ranges. Proc Natl Acad Sci USA. 85:6460–6464.[Abstract/Free Full Text]
  26. Wyllie FS, Lemoine NR, Barton CM, Dawson TP, Bond JA, Wynford-Thomas D. 1993 Direct growth stimulation of normal epithelial cells by mutant p53. Mol Carcinogen. 7:83–88.[Medline]
  27. Munari-Silem Y, Audebet C, Rousset B. 1991 Hormonal control of cell-to-cell communication: regulation by thyrotropin of the gap junction-mediated dye transfer between thyroid cells. Endocrinology. 128:3299–3309.[Abstract/Free Full Text]
  28. Lemoine NR, Staddon S, Bond J, Wyllie FS, Shaw JJ, Wynford-Thomas D. 1990 Partial transformation of human thyroid epithelial cells by mutant Ha-ras oncogene. Oncogene. 5:1833–1837.[Medline]
  29. Bond JA, Wyllie FS, Ivan M, Dawson T, Wynford-Thomas D. 1993 A variant epithelial sub-population in normal thyroid with high proliferative capacity in vitro. Mol Cell Endocrinol. 93:175–183.[CrossRef][Medline]
  30. Bond JA, Ness GO, Rowson J, Ivan M, White D, Wynford-Thomas D. 1996 Spontaneous de-differentiation correlates with extended lifespan in transformed thyroid epithelial cells: an epigenetic mechanism of tumour progression? Int J Cancer. 67:563–572.[CrossRef][Medline]
  31. Roger PP, Baptist M, Dumont JE. 1992 A mechanism generating heterogeneity in thyroid epithelial cells: suppression of the thyrotropin/cAMP-depndent mitogenic pathway after cell division induced by cAMP-independent factors. J Cell Biol. 117:383–393.[Abstract/Free Full Text]
  32. Kraiem Z, Sadeh O, Sobel E. 1990 Thyrotropin, acting at least partially via adenosine 3',5'-monophosphate, exerts both mitogenic and antimitogenic effects in cultured human thyroid cells. J Clin Endocrinol Metab. 70:497–502.[Abstract/Free Full Text]
  33. Miller DG, Adam MA, Miller D. 1990 Gene transfer by retrovirus vectors occurs only in cells that are actively replicating at the time of infection. Mol Cell Biol. 10:4239–4242.[Abstract/Free Full Text]
  34. Laurence TS, Beers WH, Gilula NB. 1978 Transmission of hormonal stimulation by cell-to-cell communication. Nature. 272:501–506.[CrossRef][Medline]
  35. Murray SA, Fletcher WH. 1984 Hormone-induced intercellular signal transfer dissociates cyclic AMP-dependent protein kinase. J Cell Biol. 98:1710–1719.[Abstract/Free Full Text]
  36. Davidson JS, Baumgarten IM, Harley EH. 1986 Reversible inhibition of intercellular junctional communication by glycyrrhetinic acid. Biochem Biophys Res Commun. 134:29–36.[CrossRef][Medline]
  37. Davidson JS, Baumgarten IM. 1988 Glycyrrhetinic acid derivatives: a novel class of inhibitors of gap-junctional intercellular communication. Structure-activity relationship. J Pharmacol Exp Ther. 246:1104–1107.[Abstract/Free Full Text]
  38. Maenhaut C, Lefort A, Libert F, et al. 1990 Function, proliferation and differentiation of the dog and human thyrocyte. Horm Metab Res. 23:51–61.
  39. Ham J, Ivan M, Wynford-Thomas D, Scanlon MF. 1997 GH3 cells expressing constitutively active Gs{alpha} (Q227L) show enhanced hormone secretion and proliferation. Mol Cell Endocrinol. 127:41–47.[CrossRef][Medline]
  40. Watkins PA, Moss J, Vaughan M. 1980 Effects of GTP on choleragen-catalyzed ADP ribosylation of membrane and soluble proteins. J Biol Chem. 255:3959–3963.[Abstract/Free Full Text]
  41. Iiri T, Ohoka Y, Ui M, Katada T. 1992 Modification of the function of pertussis toxin substrate GTP-binding protein by cholera toxin-catalyzed ADP-ribosylation. J Biol Chem. 267;1020–1026.
  42. Glineur C, Locht C. 1994 Importance of ADP-ribosylation in the morphological changes of PC12 cells induced by cholera toxin. Infect Immun. 62:4176–4185.[Abstract/Free Full Text]
  43. Sheffler LA, Wink DA, Melillo G, Cox GW. 1995 Characterisation of oxide-stimulated ADP-ribosylation of various proteins from the mouse macrophage cell line ANA-1 using sodium nitroprusside and the novel nitric oxide-donating compound diethylamine dinitric oxide. J Leukocyte Biol. 57:152–159.[Abstract]
  44. Studer H, Peter HJ, Gerber H. 1989 Natural heterogeneity of thyroid cells: the basis for understanding thyroid function and nodular goiter growth. Endocr Rev. 10:125–135.[Abstract/Free Full Text]
  45. Studer H, Derwahl M. 1995 Mechanism of nonneoplastic endocrine hyperplasia- a changing concept: a review focused on the thyroid gland. Endocr Rev. 16:411–426.[Abstract/Free Full Text]
  46. Bidey SP, Tomlinson S. 1988 The regulation and integration of thyroid follicular differentiation and function. Clin Endocrinol (Oxf). 28:423–444.[Medline]
  47. Wynford-Thomas D, Stringer BMJ, Williams ED. 1982 Dissociation of growth and function in the rat thyroid during prolonged goitrogen administration. Acta Endocrinol (Copenh). 101:210–216.[Abstract/Free Full Text]



This article has been cited by other articles:


Home page
 J EndocrinolHome page
L. Zhang, C. Paddon, M. D Lewis, F. Grennan-Jones, and M. Ludgate
Gs{alpha} signalling suppresses PPAR{gamma}2 generation and inhibits 3T3L1 adipogenesis
J. Endocrinol., August 1, 2009; 202(2): 207 - 215.
[Abstract] [Full Text] [PDF]

Home page
 J. Clin. Endocrinol. Metab.Home page
M. Derwahl, M. Broecker, and Z. Kraiem
Thyrotropin May Not Be the Dominant Growth Factor in Benign and Malignant Thyroid Tumors
J. Clin. Endocrinol. Metab., March 1, 1999; 84(3): 829 - 834.
[Full Text]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Submit a related Letter to the Editor
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Copyright Permission
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Ivan, M.
Right arrow Articles by Wynford-Thomas, D.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Ivan, M.
Right arrow Articles by Wynford-Thomas, D.

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