Article Text


Original article
Analysis of GNAQ mutations, proliferation and MAPK pathway activation in uveal melanomas
  1. Helena Pópulo1,2,
  2. João Vinagre1,2,3,
  3. José Manuel Lopes1,2,4,
  4. Paula Soares1,2
  1. 1Institute of Molecular Pathology and Immunology, University of Porto (IPATIMUP), Porto, Portugal
  2. 2Medical Faculty, University of Porto, Porto, Portugal
  3. 3Institute of Biomedical Sciences Abel Salazar (ICBAS), University of Porto, Porto, Portugal
  4. 4Department of Pathology, Hospital São João, Porto, Portugal
  1. Correspondence to Dr Paula Soares, IPATIMUP, Rua Dr Roberto Frias s/n, 4200-465 Porto, Portugal; psoares{at}


Aim To study the GNAQ mutational status in a series of uveal melanomas and evaluate possible associations with mitogen-activated protein kinase (MAPK) pathway protein expression and tumour proliferation markers.

Methods Mutational analysis was performed by PCR/sequencing of exon 5 of the GNAQ gene in a series of 22 uveal melanomas in which total and phosphorylated extracellular signal-regulated kinase (ERK) 1/2 overexpression without coexistent BRAF and NRAS mutations had previously been observed. Expression of the cell cycle markers (Ki-67, cyclin D1 and p27) was evaluated by immunohistochemistry. The association between GNAQ mutational status, ERK1/2, phospho-ERK1/2, Ki-67, cyclin D1 and p27 expression levels and the clinicopathological prognostic parameters of uveal melanomas was also assessed.

Results GNAQ mutations were found in 36% of uveal melanomas. No associations were found between the GNAQ mutational status and prognostic parameters, the expression of ERK1/2, pERK1/2 and cell cycle markers.

Conclusion The results of this study suggest that GNAQ mutated uveal melanomas do not exhibit a higher deregulation of proliferation or higher activation of the MAPK signalling pathway than uveal melanomas without GNAQ overactivation.

  • Choroids
  • genetics
  • GNAQ
  • MAPK pathway
  • neoplasia
  • pathology
  • uveal melanoma

Statistics from

Uveal melanoma is the most common primary intraocular tumour in adults, the majority of which occur in the choroid and a few arise in the ciliary body and iris.1 Choroid and ciliary body melanomas are associated with a poor prognosis, whereas iris melanomas comprise approximately 3% of uveal melanomas and are generally associated with a good prognosis.2

The prognostic markers of outcome in patients with uveal melanoma include the location, tumour cell type, mitotic rate, largest tumour dimension and scleral invasion.3

Uveal and cutaneous melanocytes have a similar embryological origin in the neural crest, but the tumours arising in these two localisations display different biological behaviours.4 Uveal melanomas show overexpression and phosphorylation of the mitogen-activated protein kinase (MAPK) pathway,5 in the absence of mutations of BRAF and NRAS,6 which are commonly found in cutaneous melanomas.7 The GNAQ gene encodes the G-protein αq-subunit, which is part of a complex membranous signalling network that mediates the transfer of information from cell surface receptors to a variety of effector molecules. The G-proteins are ubiquitous in eukaryotic cells, and control metabolic, neural and developmental cell functions.8

Heterotrimeric G-proteins consist of α, β and γ subunits that form a stable complex in an inactive guanosine diphosphate-bound state. Upon a physical interaction between a G-protein and an agonist-occupied receptor, the exchange of guanosine diphosphate for guanosine triphosphate (GTP) occurs on the α-subunit provoking its release from the β and γ subunits complex.9 Gαq subunit activation mediates intracellular signals, which can activate the phospholipase C and MAPK pathways, promoting proliferation and cell survival in several cell types (for a review, see Hubbard and Hepler),10 including melanoma cell lines.11 Activating mutations in the GNAQ gene were reported to be responsible for diffuse skin hyperpigmentation in mice by the induction of melanocyte proliferation.12 Subsequently, GNAQ gene mutations were found in nearly 50% of uveal melanomas. It has been shown, in experimental models, that the depletion of GNAQ from uveal melanoma cells harbouring the mutation causes a reduction in extracellular signal-regulated kinase (ERK) 1/2 phosphorylation, but thus far it has not been shown whether GNAQ gene mutations relate to a higher activation of the MAPK pathway in human tumours.11 13 The GNAQ mutations detected in uveal melanoma cluster at a single codon (209) located in the catalytic domain of the protein. The mutation inactivates the catalytic domain, locking GNAQ in its active GTP-bound state.14

In a previous study, we found MAPK activation in a series of uveal melanomas, and none of the cases showed BRAF or NRAS mutations,15 as was also reported by Zuidevaart et al.5

To determine the possible involvement of GNAQ mutations in MAPK pathway activation and proliferation in a series of 22 uveal melanomas, PCR and sequencing analyses were performed to evaluate the prevalence of GNAQ mutations. The expression of total and phosphorylated ERK1/2, proliferation marker Ki-67 and two cell cycle regulators, cyclin D1 and p27 (both regulated by the MAPK pathway),16 17 was evaluated by immunohistochemistry.

Materials and methods

Patients and clinicopathological features

The patient series used in this study has previously been described.15 Briefly, 22 enucleated uveal melanomas (19 located in the choroid and three in the ciliary body) and the clinicopathological data were retrieved from the Department of Pathology and the Oncology Registry of Hospital S João (HSJ), Porto, Portugal. The procedures were in accordance with institutional ethical standards.

The median age, cytological type (epithelioid, spindle or mixed) of the tumours, pT staging (according to AJCC)18 mitotic rate, median/range of the largest diameter of the tumour and scleral extension of the cases are summarised in table 1. None of the cases had clinicopathological evidence of lymph node involvement and/or distant metastasis at diagnosis.

Table 1

Clinicopathological features of uveal melanomas

GNAQ mutation analysis

Extraction of tumour DNA was performed as previously described.15

Briefly, extraction of tumour DNA from manually dissected 10 μm whole sections of paraffin-embedded material was performed using the Invisorb spin tissue mini kit (Invitek, Berlin). For tumours smaller than 5 mm, laser microdissection of 3 μm sections of paraffin-embedded tissue was performed using the PALM MicroLaser System (PALM, Germany) and DNA was extracted using the Quiamp DNA micro kit (Quiagen, Hilden) according to the manufacturer's instructions.

Fragments encompassing GNAQ exon 5 were amplified by PCR of the tumour samples with the primers 5′–TTTTCCCTAAGTTTGTAAGTAGTGC and 3′–CCTCATTGTCTGACTCCACG. Genomic DNA (25–100 ng) was amplified by PCR using the following cycling conditions: 30 s at 95°C, 30 s at 58°C and 45 s at 72°C for 40 cycles. All PCR products were purified and directly sequenced on an ABI Prism 3130 xl Automatic sequencer (Perkin-Elmer, Foster City, California, USA) using the ABI Prism Dye Terminator Cycle sequencing kit (Perkin-Elmer). The sequencing reaction was performed in the forward direction, and an independent PCR amplification, both in the forward and reverse directions, was performed in samples that carried mutations.


Taking tumour heterogeneity into account, representative tumour areas were selected and marked on haematoxylin and eosin-stained slides. Duplicated tissue cores were punched from each donor block and transferred to a tissue microarray (TMA) recipient block using a tissue-array instrument (Beecher Instruments, Silver Spring, Maryland, USA). Multiple 3 μm sections were cut from the TMA blocks and transferred to ultrafrost slides.15

The Envision G/2 System/AP (K5355; Dako, Glostrup, Denmark) was used to stain the TMA slides according to the manufacturer's instructions. Deparaffinised and rehydrated sections were subjected to microwave treatment in 10 mM sodium citrate buffer, pH 6.0 for antigen retrieval. The sections were incubated in a humidified chamber with the following primary antibodies: cyclin D1 (Thermo Scientific (Massachusetts, USA), monoclonal, rabbit, 1:100, 30 min incubation at room temperature); p27 (Santa Cruz Biotechnology (California, USA), polyclonal, rabbit, 1:200, overnight incubation at 4°C); and Ki-67 (Thermo Scientific, monoclonal, rabbit, 1:300, 2 h incubation at room temperature). Incubation with the ERK1/2 (Cell Signaling Technology (Massachusetts, USA), polyclonal, rabbit, 1:100, citrate, overnight at 4°C) and phospho-ERK1/2 Thr202/Tyr204 (Cell Signaling Technology, monoclonal, mouse, 1:75, citrate, overnight at 4°C) antibodies was previously described.15

The alkaline phosphatase antiphosphatase method was used for detection, and the samples were developed with permanent red chromogen to avoid interference of the melanin pigmentation with the immunohistochemical analysis. The slides were mounted using a water-miscible mounting medium, after counterstaining with haematoxylin.

Negative and positive controls were used simultaneously to ensure the specificity and reliability of the staining process. Previously tested positive cases of parathyroid adenoma and appendix were used as positive controls for cyclin D1 and for p27 and Ki-67, respectively. Omission of the primary antibody incubation was used as a negative control.

Two observers (JML and HP) independently evaluated tumour cell immunoreactivity, without knowledge of any clinical information about the cases, and in discrepant cases a consensus was reached.

The evaluation of the percentage of immunoreactive cells for cyclin D1, p27 and Ki-67 was carried out by counting, at ×400 magnification, a minimum of 200 tumour cells. Cells with unequivocal nuclear reaction were considered positive.19 20

For ERK1/2 and pERK1/2 Thr202/Tyr204 immunoreactivity was evaluated using a score as previously described.15 Briefly, the intensity was graded as 0 when no staining was observed, 1+ for weak, 2+ for intermediate, or 3+ for strong staining. The percentage of immunoreactive cells was graded as either focal (1–≤30%) or diffuse (2–>30%). For each case and antibody, an immunoreactivity score was estimated by multiplying the intensity grade by the grade of the immunoreactive cell percentage. Immunoreactivity scores were classified as follows: negative (0), low (1), moderate (2, 3 and 4), or high (6).

Statistical analysis

The statistical analysis was performed using STAT VIEW-J 5.0. The relationship between the GNAQ mutation status and the clinicopathological parameters, including age, gender, cytological type, pT group, mitotic rate, largest tumour diameter and scleral extension were evaluated by Fisher's exact test. The Pearson correlation coefficient, with the Fisher's exact test, was used to evaluate a possible correlation between cyclin D1, p27 and Ki-67 protein expression. The relationship between the expression level of the immunohistochemical markers and the mutational status of GNAQ and the clinicopathological parameters were evaluated by ANOVA. A p value of less than 0.05 was considered statistically significant.


We found that 36% of the cases had a GNAQ gene mutation, 22.7% of which were GNAQQ209P and 13.6% were GNAQQ209L (figure 1).

Figure 1

Representative electropherograms of GNAQ gene sequencing from a case with wild-type Q209 sequence, a case with a heterozygous Q209P mutation and a case with a heterozygous Q209L mutation.

To evaluate whether the GNAQ mutational status was associated with the prognostic parameters of the cases, the relationship between the presence or absence of the GNAQ mutation and the clinicopathological features of the cases, including age (median), gender, cytological type (epithelioid, spindle and mixed), pT group (pT1/T2 and pT3/T4), mitotic rate (≤1 mitoses/10 high-power fields and >1 mitoses/10 high-power fields), largest tumour diameter (≤10 mm and >10 mm) and sclera extension (present and absent), were analysed (table 2). There was no significant correlation between the GNAQ mutational status and any clinicopathological feature.

Table 2

Clinicopathological features and GNAQ mutational status in uveal melanomas

To evaluate whether expression of the MAPK pathway effectors was related to the mutational status of the GNAQ gene, the calculated average expression levels of ERK1/2 and pERK1/2 Thr202/Tyr20415 were compared in cases with and without GNAQ mutations (table 3). Six mutations occurred in pERK1/2 Thr202/Tyr204-positive cases and two mutations occurred in cases without pERK1/2 Thr202/Tyr204 expression. No significant association was found between the expression of any marker and the GNAQ mutational status of uveal melanomas, although a trend towards a higher average of the total ERK1/2 expression level was observed in uveal melanomas that carried GNAQ mutations (table 4).

Table 3

GNAQ, BRAF and NRAS gene status results, and the immunohistochemical expression of pERK1/2 Thr202/Tyr204 and proliferation markers in 22 uveal melanomas

Table 4

Summary of the average comparative immunohistochemical expression of the MAPK effector molecules and proliferation markers in uveal melanomas

To evaluate a possible association between the GNAQ mutation status and uveal melanoma proliferation, the expression of Ki-67, a proliferation marker, and cyclin D1 and p27, a positive and a negative cell cycle regulator, respectively, was determined (table 3). The mean expression of Ki-67 was 2.1% (SD±1.9, range 0–7.1%), whereas cyclin D1-positive cells were observed in all tumours with a mean expression of 13.6% (SD±9.8, range 1.2–37.2%); the mean expression of p27 was 2.7% (SD±4.0, range 0–14.7%). As expected, there was a significant (p=0.02) positive correlation between cyclin D1 and Ki-67 expression. The correlation between p27 and Ki-67 expression, although inverse as expected, was not significant. No significant association was found between the expression of any cell cycle marker and the mutational status of the GNAQ gene (table 4), although six out of eight cases harbouring GNAQ mutations were negative for p27 expression.

No further differences were observed when we separately compared the two GNAQ mutations found in our study.


GNAQ gene mutations have recently been reported in up to 50% of uveal melanomas.11 13 In our study, GNAQ mutations involving the hot spot codon 209 were found in 36% of cases, indicating a significant role for mutations of the GNAQ gene in uveal melanoma development. Similar to the study by Onken et al,13 we did not find an association between the GNAQ mutation and clinicopathological prognostic parameters, such as age, gender, cytological type, pT group, mitotic rate, largest tumour diameter and scleral extension, reinforcing the idea that the GNAQ mutation may be an early event in uveal melanoma development. In agreement with the possibility of an early event, it has already been shown that GNAQ mutations did not correlate with disease-free survival in uveal melanoma and cannot be used to predict the clinical outcome of patients.21

Although an association between the GNAQ mutational status and proliferation was demonstrated in mouse melanocytes,12 in our study no significant differences were observed in the expression of cell cycle markers (Ki-67, cyclin D1 and p27) in uveal melanoma tumours with or without the GNAQ mutation. Noteworthy, the expression of the negative cell cycle regulator p27 was absent in 75% of the cases harbouring a GNAQ mutation.

The GNAQ mutations detected in our study affected the glutamine encoded by codon 209 and led to substitution by a proline or a leucine. Both mutations induce a conformational change in the catalytic domain of the Gαq protein, inhibiting its GTPase hydrolytic activity. The Gαq subunit remains in an active GTP-bound state that prevents reassembling of the α, β and γ complex. According to De Vivo et al,22 the constitutive activation of the Gαq subunit induced by these activating mutations can mediate an enhanced phospholipase C stimulation. As a result of this phospholipase C stimulation, protein kinase C is activated and can trigger the RAF/MEK/ERK and other intracellular pathways.23 Such a mechanism that mimics growth factor signalling stimulation would provide an alternative route for activation of the MAPK pathway in the absence of BRAF and NRAS mutations.

We have previously observed MAPK pathway upregulation in uveal melanomas lacking BRAF or NRAS mutations.15 GNAQQ209L was shown to induce MAPK pathway activation in cell lines.11 However, although we observed high total ERK1/2 expression in cases harbouring a GNAQ mutation, no significant association was found between the GNAQ mutational status and the expression of phosphorylated ERK1/2. Moreover, GNAQ mutations were found in cases without pERK1/2 Thr202/Tyr204 expression, whereas cases without a mutation did express pERK1/2 Thr202/Tyr204. As suggested by Houben et al,24 who did not find a correlation between the presence of BRAF mutations and the ERK1/2 phosphorylation status in cutaneous melanoma, we can also speculate that alternative mechanisms may play a role in the upregulation of MAPK pathway activation in uveal melanomas without the GNAQ mutation and therefore, at the level of ERK1/2 phosphorylation, the two tumour sets behave identically.

We did not find any GNAQ mutation in conjunctival melanomas (data not shown) in which BRAF mutations have already been described by others, contrary to the results seen in uveal melanomas.25 These findings reinforce the concept that diverse aetiopathogenic mechanisms may operate in the development of conjunctival and uveal melanomas.

In conclusion, our study concurs with the previously reported high frequency of GNAQ mutations in uveal melanomas. No association was found between the presence of the GNAQ mutation and any clinicopathological prognostic parameters of uveal melanomas. The cases harbouring the GNAQ mutation did not differ significantly in the expression of ERK1/2 and pERK1/2 or in the expression of proliferation markers from cases without the mutation. Further studies of larger series are needed to clarify fully the role of GNAQ activation in uveal melanomas.


View Abstract


  • Linked articles 182097.

  • Funding This study was supported by the Portuguese Foundation for Science and Technology through a PhD grant to HP (ref SFRH/BD/31369/2006) and project PTDC/SAU-OBD/69787/2006 (project 3599/PPCDT).

  • Competing interests None.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Linked Articles