Overexpression of Gs alpha subunit in thyroid tumors bearing a mutated Gs alpha gene

Structural and Functional Implication of Natural Variants of Gαs

Heterotrimeric guanine nucleotide-binding proteins (G proteins) are among the most important cellular signaling components, especially G protein-coupled receptors (GPCRs). G proteins comprise three subunits, Gα, Gβ, and Gγ. Gα is the key subunit, and its structural state regulates the active status of G proteins. Interaction of guanosine diphosphate (GDP) or guanosine triphosphate (GTP) with Gα switches G protein into basal or active states, respectively. Genetic alteration in Gα could be responsible for the development of various diseases due to its critical role in cell signaling. Specifically, loss-of-function mutations of Gαs are associated with parathyroid hormone-resistant syndrome such as inactivating parathyroid hormone/parathyroid hormone-related peptide (PTH/PTHrP) signaling disorders (iPPSDs), whereas gain-of-function mutations of Gαs are associated with McCune-Albright syndrome and tumor development. In the present study, we analyzed the structural and functional implications of natural variants of the Gαs subtype observed in iPPSDs. Although a few tested natural variants did not alter the structure and function of Gαs, others induced drastic conformational changes in Gαs, resulting in improper folding and aggregation of the proteins. Other natural variants induced only mild conformational changes but altered the GDP/GTP exchange kinetics. Therefore, the results shed light on the relationship between natural variants of Gα and iPPSDs.

Identification of novel GNAS mutations in intramuscular myxoma using next-generation sequencing with single-molecule tagged molecular inversion probes

Background

Intramuscular myxoma (IM) is a hypocellular benign soft tissue neoplasm characterized by abundant myxoid stroma and occasional hypercellular areas. These tumors can, especially on biopsy material, be difficult to distinguish from low-grade fibromyxoid sarcoma or low-grade myxofibrosarcoma. GNAS mutations are frequently involved in IM, in contrast to these other malignant tumors. Therefore, sensitive molecular techniques for detection of GNAS aberrations in IM, which frequently yield low amounts of DNA due to poor cellularity, will be beneficial for differential diagnosis.

Methods

In our study, a total of 34 IM samples from 33 patients were analyzed for the presence of GNAS mutations, of which 29 samples were analyzed using a gene-specific TaqMan genotyping assay for the detection of GNAS hotspot mutations c.601C > T and c602G > A in IM, and 32 samples using a novel next generation sequencing (NGS)-based approach employing single-molecule tagged molecular inversion probes (smMIP) to identify mutations in exon 8 and 9 of GNAS. Results between the two assays were compared for their ability to detect GNAS mutations with high confidence.

Results

In total, 23 of 34 samples were successfully analyzed with both techniques showing GNAS mutations in 12 out of 23 (52%) samples. The remaining 11 samples were analyzed with either TaqMan assay or smMIP assay only. The TaqMan assay revealed GNAS mutations in 16 out of 29 samples (55%), with six samples c.601C > T (p.R201C; 38%) and ten samples c.602G > A (p.R201H; 62%) missense mutations. The smMIP assay identified mutations in 16 out of 28 samples (57%), with five samples c.601C > T (p.R201C; 31%) and seven samples c.602G > A (p.R201H; 44%) missense mutations. In addition, four samples (25%) revealed novel IM-associated mutations, including c.601C > A (p.R201S), c.602G > T (p.R201L), c.602G > C (p.R201P) and c.680A > G (p.Q227R). Combining the results of both tests, 23 out of 34 sporadic IM samples (68%) showed a GNAS mutation.

Conclusions

Both the TaqMan and the smMIP assay a show a high degree of concordance in detecting GNAS hotspot mutations in IM with comparable sensitivity. However, since the NGS-based smMIP assay permits mutation detection in whole exons of GNAS, a broader range of GNAS mutations can be identified by the smMIP approach.

Frequency of GNAS R201H substitution mutation in polyostotic fibrous dysplasia: Pyrosequencing analysis in tissue samples with or without decalcification

Guanine nucleotide-binding protein/α-subunit (GNAS) mutations are involved in fibrous dysplasia (FD) pathogenesis. Here, we analyzed GNAS mutations in FD which were performed by pyrosequencing DNA isolated from formalin-fixed paraffin-embedded (FFPE) tissue. The mutation detection rate was determined in FD specimens with and without decalcification. GNAS mutation was identified in 28 cases out of 87 FDs (32.18%) [p.R201C (N = 14) and p.R201H (N = 14)]. GNAS mutation was more likely to occur in polyostotic FD (7/28, 25.0%); FD without GNAS mutation was mostly monostotic form (56/59, 94.9%, P = 0.011). The G > A (R201H) mutation was more frequent in polyostotic FD (6/14 patients, 42.9%) than the C > T (R201C) mutation (1/14, 7.1%) (P = 0.077). We divided the FD cases into two subgroups: tissue specimens that were not decalcified (N = 35, 40.2%), and tissue specimens that were decalcified (N = 52, 59.8%). GNAS mutation was more frequently identified in FD specimens that were not subjected to decalcification (23/35, 65.7%) than in FD specimens that were decalcified (5/52, 9.6%) (P = 0.001). In conclusion, mutation analysis of GNAS by pyrosequencing has diagnostic value in FFPE tissue of patients with FD, especially in specimens that were not decalcified. The R201H substitution mutation of GNAS may be involved in the pathogenesis of polyostotic FD.

GNAS mutations are not detected in parosteal and low-grade central osteosarcomas

Parosteal osteosarcoma, low-grade central osteosarcoma, and fibrous dysplasia share similar histological features that may pose a diagnostic challenge. The detection of GNAS mutations in primary bone tumors has been useful in clinical practice for diagnosing fibrous dysplasia. However, the recent report of GNAS mutations being detected in a significant proportion of parosteal osteosarcoma challenges the specificity of this mutation. As the number of cases reported in this study was small we set out to determine if these results could be reproduced. We studied 97 formalin-fixed paraffin-embedded low-grade osteosarcomas from 90 patients including 62 parosteal osteosarcomas, of which MDM2 amplification was detected in 79%, 11 periosteal osteosarcomas and 24 low-grade central osteosarcoma samples. The mutational status of GNAS was analyzed in codons p.R201, p.Q227, and other less common GNAS alterations by bidirectional Sanger sequencing and/or next generation sequencing using the Life Technologies Ion Torrent platform. GNAS mutations were not detected in any of the low-grade osteosarcomas from which informative DNA was extracted. Our findings therefore support prior observations that GNAS mutations are highly specific for fibrous dysplasia and occur rarely, if ever, in parosteal and other low-grade osteosarcomas.

Diagnostic value of investigating GNAS mutations in fibro-osseous lesions: a retrospective study of 91 cases of fibrous dysplasia and 40 other fibro-osseous lesions

GNAS (guanine nucleotide-binding protein/α-subunit) mutations that induce the activation of G-protein α-subunit participate in the pathogenesis of fibrous dysplasia. The aim of this study was to evaluate the sensitivity and specificity of GNAS mutations in fibrous dysplasia and other fibro-osseous lesions, to assess the value of investigating this mutation in the diagnosis of fibro-osseous lesions. We studied 91 cases of fibrous dysplasia. The quality and/or quantity of genomic DNA were suitable for molecular analysis for 51 cases of fibrous dysplasia. GNAS mutations were investigated by three techniques: high-resolution melting (exon 8), allele-specific PCR (exons 8 and 9) and/or direct DNA sequencing (exons 8 and 9). Fibrous dysplasia samples were classified blind to the GNAS mutation status into six histological subtypes as conventional, fibro-involutive, osteosclerosing, cementifying, osteocartilaginous and with prominent aneurysmal cystic changes. We also studied 14 cases of low-grade osteosarcoma, 21 cases of ossifying fibroma, 3 cases of osteofibrous dysplasia, 1 case of osseous dysplasia of the jawbone and 1 post-traumatic lesion of the ribs. Twenty-three cases of fibrous dysplasia (45%) showed mutations of codon 201 (exon 8, p.R201H or p.R201C). No mutation was found on codon 227 (exon 9). GNAS mutations in conventional fibrous dysplasia were detected in the same proportion (47%) as in the other histological subtypes (47%, P=0.96), regardless of sex (P=0.44), age (P=0.90) and location (P=1). GNAS mutations were not detected in any other fibro-osseous lesions. The GNAS mutation was thus specific to fibrous dysplasia in the context of fibro-osseous lesions. The particular mosaicism of mutant and non-mutant cells within the lesion or the existence of other mutations not already described could explain the lack of GNAS mutation in cases of fibrous dysplasia. Investigating this mutation may constitute a valuable complementary diagnostic tool, despite its low sensitivity, particularly in unconventional morphologically different subtypes of fibrous dysplasia.

A sensitive mutation-specific screening technique for GNAS1 mutations in cases of fibrous dysplasia: the first report of a codon 227 mutation in bone

AIMS

To report on the mutation-specific restriction enzyme digest (MSRED) method using paraffin-embedded tissue as a means of detecting GNAS1 mutations in fibrous dysplasia (FD), and to determine if any of the reported GNAS1 mutations in endocrine neoplasms, not previously documented in FD, can be found in FD.

METHODS AND RESULTS

Sixty-seven cases of extragnathic FD were analysed as two groups, 1997-2002 and 2003-06, chosen because tissue fixation and decalcification methods were more accurately recorded in the latter. MSRED revealed that between 2003 and 2006, 93% of 28 'in house' extragnathic cases harboured a GNAS1 mutation, compared with 75% of 32 cases before 2003. Fixation times of no more than 48 h and decalcification in ethylenediamine tetraacetic acid gave the best results. Of the 56 mutations detected (five gnathic, 51 extragnathic), 32 (57%) were R201H, 21 (38%) were R201C and three (5%) were Q227L. Two Q227L extragnathic cases had unusual clinical/radiological findings. No mutations were detected in osteofibrous dysplasia.

CONCLUSION

Detection of GNAS1 mutations by MSRED is a valuable adjunct to the histopathological diagnosis of FD. This is the first report of a Q227L mutation in FD, although it has been previously documented in pituitary adenoma.

Mutations of the GNAS1 gene, stromal cell dysfunction, and osteomalacic changes in non-McCune-Albright fibrous dysplasia of bone

Activating missense mutations of the GNAS1 gene, encoding the alpha subunit of the stimulatory G protein (Gs), have been identified in patients with the McCune-Albright syndrome (MAS; characterized by polyostotic fibrous dysplasia, café au lait skin pigmentation, and endocrine disorders). Because fibrous dysplasia (FD) of bone also commonly occurs outside of the context of typical MAS, we asked whether the same mutations could be identified routinely in non-MAS FD lesions. We analyzed a series of 8 randomly obtained, consecutive cases of non-MAS FD and identified R201 mutations in the GNAS1 gene in all of them by sequencing cDNA generated by amplification of genomic DNA using a standard primer set and by using a novel, highly sensitive method that uses a protein nucleic acid (PNA) primer to block amplification of the normal allele. Histologic findings were not distinguishable from those observed in MAS-related FD and included subtle changes in cell shape and collagen texture putatively ascribed to excess endogenous cyclic adenosine monophosphate (cAMP). Osteomalacic changes (unmineralized osteoid) were prominent in lesional FD bone. In an in vivo transplantation assay, stromal cells isolated from FD failed to recapitulate a normal ossicle; instead, they generated a miniature replica of fibrous dysplasia. These data provide evidence that occurrence of GNAS1 mutations, previously noted in individual cases of FD, is a common and perhaps constant finding in non-MAS FD. These findings support the view that FD, MAS, and nonskeletal isolated endocrine lesions associated with GNAS1 mutations represent a spectrum of phenotypic expressions (likely reflecting different patterns of somatic mosaicism) of the same basic disorder. We conclude that mechanisms underlying the development of the FD lesions, and hopefully mechanism-targeted therapeutic approaches to be developed, must also be the same in MAS and non-MAS FD.

Gs alpha mutations in hyperfunctioning thyroid adenomas

Hyperfunctioning thyroid adenomas are benign tumors characterized by their autonomous growth and functional activity, which frequently cause clinical hyperthyroidism and show a predominant radioactive iodine uptake in the nodule. Activating mutations in the gene encoding the alpha subunit of the stimulatory G protein (Gs alpha), as well as activating mutations in the gene encoding thyrotropin receptor in hyperfunctioning thyroid adenomas, have been reported. The mutations in Gs alpha involved the replacement of either arginine 201 with cysteine or histidine, or glutamine 227 with arginine or leucine. These residues are involved in GDP/GTP binding of Gs alpha and these mutations inhibit intrinsic GTPase activity that results in constitutive activation of adenylyl cyclase. The pathophysiological roles of these mutations in the formation of hyperfunctioning thyroid adenoma have been suggested.

Activating mutations of the Gs alpha gene are associated with low levels of Gs alpha protein in growth hormone-secreting tumors

Evidence suggests the existence of a direct relationship between cellular Gs alpha content and activation of the adenylyl cyclase system. Data on Gs alpha levels in endocrine tumors that depend on cAMP for growth, particularly pituitary adenomas, are still limited. The levels of Gs alpha protein were evaluated in 11 GH-secreting adenomas with Gs alpha mutations (gsp+) and 15 without (gsp). Complementary DNAs from gsp+ tumors contained very low amounts of wild-type Gs alpha sequences, indicating a preponderance of the mutant Gs alpha transcripts in these tumors. Immunoblotting of Gs alpha protein showed that the two isoforms were present at high levels in all gsp-, but were undetectable or barely detectable in gsp+. The low Gs alpha content in gsp+ tumors was not due to a reduction in ribonucleic acid synthesis or stability, as Gs alpha messenger ribonucleic acid levels were similar in wild-type and mutant tissues. Treatment of gsp- cells with cholera toxin caused a marked reduction of Gs alpha levels. As in other cell systems cholera toxin increases Gs alpha degradation, our data are consistent with an accelerated removal of mutant Gs alpha. This may represent an additional mechanism of feedback response to the constitutive activation of cAMP signaling in pituitary tumors with mutations in the Gs alpha gene.

Expression of G protein alpha subunits in normal rat colon and in azoxymethane-induced colonic neoplasms

BACKGROUND & AIMS

Heterotrimeric G proteins are important in growth-regulating signal transduction. The aim of this study was to characterize the relative expression of G protein alpha subunits in rat colonocytes, colonocyte antipodal plasma membranes, and colonic neoplasms.

METHODS

Antipodal plasma membranes were prepared from isolated colonocytes. Azoxymethane was administered to rats to induce colonic neoplasms. K-ras mutations in the neoplasms were determined by oligonucleotide hybridization and confirmed by primer mediated-restriction fragment length polymorphism. Colonocyte and tumor homogenates or membranes were probed for Galpha subunits by Western blotting with isoform-specific antibodies.

RESULTS

The expressions of Galphai2, alphai3, and alphaq/11 were significantly enriched in the basolateral compared with brush border fraction of colonic antipodal plasma membranes. In neoplasms without K-ras mutations, the expression of Galphai2 increased 4-fold, Galphas(long) increased 2.5-fold, and Galphai3 increased 1.5-2-fold. Expression did not differ among tumor grades. K-ras mutations were associated with lowered expression of G proteins, especially Galphao.

CONCLUSIONS

In colonocytes, Galpha subunits are localized primarily in basolateral plasma membranes. The increased expressions of Galphai2 and, to a lesser degree, Galphai3 and Galphas(long) in tumors was independent of tumor grade but was modulated by the presence of K-ras mutations.

Distribution of Gs-alpha activating mutations in human thyroid tumors measured by subcloning

In the search for the prevalence and distribution pattern of Gs-alpha gene mutations in differentiated thyroid tumors we examined 66 tumor tissue samples for the presence of mutations at "hot-spot" codons 201 and 227 using methods based on the polymerase chain reaction, subcloning and sequencing. The prevailing type of single-base substitution at codon 201 (71.4%) corresponded to the replacement of the wild-type sequence CGT (Arg) with TGT (Cys). The fragments of the Gs-alpha gene, including codon 201 or 227 from five follicular carcinomas and one follicular adenoma, were subcloned in Escherichia coli and it was found that the proportion of alleles with mutated codon 201 varied from 3.2% to 43%. Sequencing of the corresponding region has confirmed preliminary data indicating that the single-base changes CGT (Arg) to TGT (Cys) or CGT to CAT (His) occurred. There was only a weak correlation between the prevalence of cells bearing a mutation in the Gs-alpha gene and the level of Gs-alpha protein expression in the corresponding thyroid tumors.