Raf/MEK/ERK and phosphatidylinositol 3-kinase (PI3K)/Akt/mammalian target of rapamycin (mTOR) cascades are key signalling pathways interacting with each other to regulate cell growth and tumourigenesis. We have previously shown B-Raf and Akt overexpression and/or overactivation in pituitary adenomas. The aim of this study is to assess the expression of their downstream components (MEK1/2, ERK1/2, mTOR, TSC2, p70S6K) and effectors (c-MYC and CYCLIN D1). We studied tissue from 16 non-functioning pituitary adenomas (NFPAs), six GH-omas, six prolactinomas and six ACTH-omas, all collected at transsphenoidal surgery; 16 normal autopsy pituitaries were used as controls. The expression of phospho and total protein was assessed with western immunoblotting, and the mRNA expression with quantitative RT-PCR. The expression of pSer217/221 MEK1/2 and pThr183 ERK1/2 (but not total MEK1/2 or ERK1/2) was significantly higher in all tumour subtypes in comparison to normal pituitaries. There was no difference in the expression of phosphorylated/total mTOR, TSC2 or p70S6K between pituitary adenomas and controls. Neither c-MYC phosphorylation at Ser 62 nor total c-MYC was changed in the tumours. However, c-MYC phosphorylation at Thr58/Ser62 (a response target for Akt) was decreased in all tumour types. CYCLIN D1 expression was higher only in NFPAs. The mRNA expression of MEK1, MEK2, ERK1, ERK2, c-MYC and CCND1 was similar in all groups. Our data indicate that in pituitary adenomas both the Raf/MEK/ERK and PI3K/Akt/mTOR pathways are upregulated in their initial cascade, implicating a pro-proliferative signal derangement upstream to their point of convergence. However, we speculate that other processes, such as senescence, attenuate the changes downstream in these benign tumours.
Pituitary tumours are common neoplasms, with a reported overall prevalence of 16.7% in the general population, 14.4% in autopsy studies and 22.5% in radiological studies (Ezzat et al. 2004). These tumours are mostly benign, but cause significant morbidity through mass effects and/or the inappropriate secretion of pituitary hormones (Ezzat et al. 2004). Despite intensive study, the molecular pathogenesis of the majority of sporadic pituitary adenomas remains unclear (Asa & Ezzat 2002, Melmed 2008). There are several genetic syndromes associated with pituitary tumours, but abnormalities in the genes mutated in these syndromes are rarely abnormal in sporadic tumours (Beckers & Daly 2007, Boikos & Stratakis 2007, Karhu & Aaltonen 2007, Horvath & Stratakis 2008, Leontiou et al. 2008). There are also a number of tumour suppressor genes which are inactivated in sporadic tumours, often by promoter methylation, but no single causal agent or mutation has been identified (Dworakowska & Grossman 2009b).
Two major signalling pathways responsible for regulating cell growth and proliferation, downstream of growth factor receptors, are the phosphatidylinositol 3-kinase (PI3K)/Akt and Raf/mitogen-activated protein kinase (MAPK) pathways (McCubrey et al. 2007). These have been shown to be overactive in many human tumours (Adjei & Hidalgo 2005, Roberts & Der 2007). It has been reported that Akt is overexpressed and overactivated in non-functioning pituitary adenomas (NFPAs; Musat et al. 2005), and there is recent evidence that B-Raf, an upstream regulator of the MAPK cascade, is also overexpressed predominantly in NFPAs (Ewing et al. 2007). Such activation may account for the changes in cell-cycle components which are a frequent characteristic of pituitary tumours, especially p27 (Lloyd et al. 1997, Lidhar et al. 1999). In addition, we have shown that somatostatin analogues, which cause tumour regression in certain types of pituitary adenoma, downregulate the MAPK pathway in pituitary tumours (Hubina et al. 2006). However, it has been unclear as to whether the changed expression of the upstream components of these two converging pathways is associated with downstream modulation of their subsequent effectors.
Raf/MEK/ERK is a hierarchical cascade, originating at the cell membrane with receptors for mitogens or growth factors, which recruit, via adapter proteins and exchange factors, the small guanosine triphosphatase Ras (Adjei & Hidalgo 2005). Ras then activates Raf (MAPK kinase kinase, MAPKKK), which in turn activates MAPK kinase (MAPKK; MEK1/2): MEK1/2 proteins are dual-specificity kinases containing two consensus kinase motifs, one motif being involved in phosphorylation of serine/threonine residues while the other is involved in phosphorylation of tyrosine residues. Two MEK homologues, MEK1 and MEK2, are activated through phosphorylation of two serine residues at positions 217 and 221 by Raf-like molecules, and then MEK1/2 sequentially phosphorylate ERK1 and ERK2 at two sites, Tyr185 followed by Thr183 (Haystead et al. 1992). ERK1/2 in turn phosphorylates and activates ribosomal S6 kinase and transcription factors such as c-MYC, Elk1, c-Fos or CYCLIN D1 (Terada et al. 1999a,b), resulting in the modulation of genes associated with proliferation (Guan 1994) and leading to cell transformation (Joneson & Bar-Sagi 1997). ERK1 and 2 activate CYCLIN D1 promoter activity and CYCLIN D1 expression, whereas p38MAPK works in an opposing manner, inhibiting CYCLIN D1 transcription and expression (Lavoie et al. 1996).
In terms of the PI3K/Akt/mammalian target of rapamycin (mTOR) pathway, PI3K is activated as a result of the ligand-dependent activation of tyrosine kinase receptors, G-protein-coupled receptors, or integrins. Receptor-independent activation of PI3K can also occur in cells expressing constitutively-active Ras protein (Kauffmann-Zeh et al. 1997, Rodriguez-Viciana et al. 1997). The best-characterised phosphorylation target of PI3K is Akt (protein kinase B), resulting in the phosphorylation of a host of other proteins that affect cell growth, cell cycle entry, and cell survival. Akt phosphorylation at Thr308 and Ser473 activates a serine–threonine kinase mTOR at Ser2448, which further activates 40S ribosomal protein S6 kinase (p70S6K) at Thr229 and Thr389 (Volarevic & Thomas 2001) and inactivates 4E-binding protein (4E-BP1). The tuberous sclerosis complex (TSC) – which mediates between PI3K/Akt and mTOR – is germline mutated in tuberous sclerosis, and there is evidence that neuroendocrine and pituitary tumours are a feature of this disease (Dworakowska & Grossman 2009a). Mitogenic stimuli activating Akt phosphorylate TSC2 at Ser939 and Thr1462. This phosphorylation destabilises TSC2, inhibits the formation of the TSC1/2 complex, and leads to an increase in mTOR activity (Inoki et al. 2002). Changes in 4E-BP1 lead to the initiation of translation of mRNA for c-MYC or CYCLIN D1 through its association with eIF-4E (Rosenwald et al. 1995, Sonenberg & Gingras 1998), while Akt also activates CYCLIN D1 indirectly via glycogen synthase kinase-3 (GSK-3β; Diehl et al. 1998, Onofri et al. 2006). There is also further evidence that these two pathways interact, e.g. ERK1/2 indirectly activates mTOR by phosphorylation of TSC2 at Ser669. Thus, both pathways are involved in the regulation of c-MYC and CYCLIN D1 transcription. ERK1/2 also directly activates phosphorylation of c-MYC protein at Ser62 leading to its accumulation, whereas Akt decreases phosphorylation (via GSK-3β) at Thr58 which diminishes c-MYC degradation (Sears et al. 2000).
The aim of this study was to investigate in more detail the status of the intervening components of Raf/MEK/ERK and PI3K/Akt/mTOR pathways as well as their two principal targets – c-MYC and CYCLIN D1. In different types of human pituitary adenomas and normal pituitary we assessed the protein expression of phosphorylated/total MEK1/2, ERK1/2, p38MAPK, TSC2, mTOR, p70S6K, c-MYC and CYCLIN D1. In addition, we performed a quantitative analysis of mRNA expression of all of the components which had show alterations at the protein level (MEK1, MEK2, ERK1, ERK2, c-MYC and CCND1).
Materials and methods
We studied tissue from 50 human pituitaries (Table 1) including 16 NFPAs, six somatotrophinomas (GH-omas), six prolactinomas (PRL-omas) and six ACTH-oma, removed at transsphenoidal surgery and immediately ‘flash-frozen’ and stored at −80 °C. A total of 16 normal pituitary samples were used as a control group. These were collected at autopsy (which was performed within 24 h of death from non-endocrine disease), ‘flash-frozen’ and stored at −80 °C. The tumour study group consisted of 18 men (53%) and 16 women (47%), with a mean age of 58.3 years (age range 19–77 years). The controls consisted of 13 men (81%) and 3 women (19%), with a mean age of 64.7 years (age range 42–77 years). All patients gave written informed consent for the protocol, which was approved by the local Institutional Review Board (IRB). The autopsy collection was also approved by the relevant local IRB.
Clinical details of patients included in the study
|Clinical diagnosis||Samples studied (tumour size)||Patient gender||Patient age (mean+range)|
|NFPA||n=16 (15 macro, 1 micro)||4 F, 12 M||56 years (28–75)|
|GH-oma||n=6 (1 macro, 5 micro)||4 F, 2 M||48 years (31–64)|
|PRL-oma||n=6 (2 macro, 4 micro)||3 F, 3 M||33 years (19–46)|
|ACTH-oma||n=6 (2 macro, 4 micro)||5 F, 1 M||48.3 years (24–77)|
|All adenomas||n=34||16 F, 18 M||58.3 years (19–77)|
|Normal pituitary||n=16||3 F, 13 M||64.7 years (42–77)|
Protein extraction and western blot analysis
Whole lysates were prepared from 50 to 250 μg frozen fresh human pituitary tissue by homogenisation in 500 μl ice-cold cell lysis buffer with added 1 mM phenylmethylsulphonyl fluoride, followed by a brief sonication and centrifugation at 12 000 g for 20 min at 4 °C. The supernatant was saved and stored at −80 °C. Protein yield was quantified using the BCA protein assay kit (Pierce–Perbio, Cramlington, UK). Fifty microgram of total protein lysate were denatured in SDS sample buffer, separated by electrophoresis on a SDS-PAGE gel (10%) and transferred to a nitrocellulose membrane (Immobilon-P; pore size 0.45 mm; Millipore, Watford, Herts, UK). The membrane was subsequently incubated with the primary antibody in 0.1% or 5% BSA, or in 5% non-fat dry milk in Tris-buffered saline/Tween 20 overnight at 4 °C (Table 2). After six washes in Tris-buffered saline/Tween 20, the membranes were incubated with the secondary antibody for 90 min. Immunodetection was performed using the Odyssey infrared imaging system (LI-COR Biosciences, Cambridge, UK). Each membrane was incubated and double-stained with two primary antibodies (phospho-protein and β-actin or total protein and β-actin). Cross-detection between analysed protein and β-actin was avoided by using different species-raised against primary antibodies. To assess expression of each protein we have constructed a set consisting of three membranes (15 wells), including all sample types (four normal pituitaries, two PRL-omas, two ACTH-omas, two GH-omas and four NFPAs). One well was reserved for the size marker. The optical density of the appropriately sized bands was measured using the Odyssey molecular imaging software (LI-COR Biosciences).
Antibodies used for western blotting in the current study
|Anti-phospho-MEK1/2 (Ser217/221)||Rabbit||1:1000||Cell Signalling Technology, Hitchin, UK||Primary||Monoclonal|
|Anti-MEK1/2 (total)||Rabbit||1:1000||Cell Signalling Technology||Primary||Monoclonal|
|Anti-phospho-ERK1/2 (Thr183)||Rabbit||1:4000||Promega Corporation||Primary||Polyclonal|
|Anti-phospho-c-myc (Ser62)||Mouse||1:200||Spring Bioscience, Fremont, CA, USA||Primary||Monoclonal|
|Anti-phospho-c-myc (Thr58/Ser62)||Rabbit||1:1000||Cell Signalling Technology||Primary||Polyclonal|
|Anti-c-myc (total)||Rabbit||1:1000||Cell Signalling Technology||Primary||Polyclonal|
|Anti-cyclin D1||Rabbit||1:1000||Cell Signalling Technology||Primary||Monoclonal|
|Anti-phospho-p38MAPK (Thr180/Tyr182)||Rabbit||1:1000||Cell Signalling Technology||Primary||Monoclonal|
|Anti-p38MAPK (total)||Rabbit||1:1000||Cell Signalling Technology||Primary||Monoclonal|
|Anti-phospho-mTOR (Ser2448)||Rabbit||1:1000||Cell Signalling Technology||Primary||Polyclonal|
|Anti-mTOR (total)||Rabbit||1:2000||Cell Signalling Technology||Primary||Polyclonal|
|Anti-phospho-tuberin (Thr1462)||Rabbit||1:1000||Cell Signalling Technology||Primary||Polyclonal|
|Anti-phospho-tuberin (Ser939)||Rabbit||1:1000||Cell Signalling Technology||Primary||Polyclonal|
|Anti-tuberin (total)||Rabbit||1:1000||Cell Signalling Technology||Primary||Polyclonal|
|Anti-phospho-S6K (Thr389)||Rabbit||1:1000||Cell Signalling Technology||Primary||Polyclonal|
|Anti-S6K (total)||Rabbit||1:1000||Cell Signalling Technology||Primary||Polyclonal|
|Anti-β-actin||Mouse||1:5000||Abcam, Cambridge, UK||Primary||Monoclonal|
|Anti-rabbit IRDye_ 800||Goat||1:2000||LI-COR Biosciences, Cambridge, UK||Secondary||–|
|Anti-mouse IRDye_ 680||Goat||1:10 000||LI-COR Biosciences||Secondary||–|
The expression of phospho-mTOR, phospho-TSC2, phospho-p70S6K, phospho-ERK1/2, phospho-MEK1/2, phospho-c-MYC and phospho-p38MAPK was normalised to β-actin from the corresponding membrane and was calculated as a ratio of phospho-peptide/β-actin. The expression of total-mTOR, total-TSC2, total-p70S6K, total-ERK1/2, total-MEK1/2, total-c-MYC, total-p38MAPK and CYCLIN D1 was normalised to β-actin from the corresponding membrane and was calculated as a ratio of total/β-actin. The relative expression of each phospho-protein was calculated as well as the proportion of phospho/total protein. All results normalised to β-actin were compared between subgroups of pituitary tumours, and are shown as % of control.
mRNA extraction and ‘real-time’ PCR
Total RNA from tissue was extracted using an RNeasy Mini kit (Qiagen) and the protocol was followed according to the manufacturers' instructions. A DNase treatment step was performed to limit contamination of the total RNA with genomic DNA. One microgram of total RNA was reverse-transcribed (RT) in a 25 μl reaction using 5 μl 5× first strand buffer (Invitrogen), 100 mM dithiothreitol (Invitrogen), 20 mM of deoxynucleotide triphosphate (Promega), 20 μg/μl random hexamer primers (Promega), MMW-RT (Maloney murine leukemia virus transcriptase; Invitrogen) and 40 U/μl RNase inhibitor (Promega). Thermal cycling conditions were as follows: 26 °C for 10 min, 37 °C for 60 min and 92 °C for 10 min. cDNA was amplified by real-time (RT) PCR using TaqMan Universal Master Mix (Applied Biosystems, Warrington, UK), 20× assay primers and their respective assay ID for human MEK1 Hs00983256_g1, MEK2 Hs00360961_m1, MAPK1 (encoding for ERK2) Hs00177066+m1, MAPK3 (encoding for ERK1) Hs00385075_m1, CCND1 Hs00277039_m1, c-MYC Hs0015348_m1 and the β-actin Vic-MGB probe 4326315E. cDNA was amplified in a 7900HT Taqman machine (Applied Biosystems). All data were expressed in relation to β-actin and then calculated relative to the percentage of control.
Statistical analysis was performed with StatsDirect software (Buchan I; Addison Wesley Longman, Cambridge, UK). The data were noted to be non-normally distributed. Statistical analysis was performed with the use of the Kruskal–Wallis test followd by the Conover–Inman comparison, the Mann–Whitney test and the Spearman's correlation test. Statistical significance was defined at P<0.05 and all data are shown as means±s.e.m.
The expression of pSer217/221 MEK1/2 (Fig. 1A), but not total MEK1/2 (Fig. 1B), was significantly higher in PRL-omas (277±27%, P<0.001), ACTH-omas (287±14%, P<0.001), GH-omas (175±39%, P=0.009) and NFPAs (281±15%, P<0.001) in comparison to normal pituitaries. Furthermore, the ratio between pSer217/221 MEK1/2 and total MEK1/2 was significantly higher in all groups of pituitary tumours in comparison to normal pituitaries (271±13% for PRL-omas, P=0.001; 227±30% for ACTH-omas, P<0.001; 187±15% for GH-omas; P=0.01 and 371±45% for NFPAs; P<0.001; Fig. 1C).
The expression of pThr183 ERK1/2 (Fig. 2A), but not total ERK1/2 (Fig. 2B), was significantly higher in PRL-omas (174±16%, P=0.01), ACTH-omas (231±46%, P<0.001), GH-omas (184±38%, P=0.01) and NFPAs (317±38%, P<0.001) in comparison to normal pituitaries. Furthermore, the ratio between pThr183 ERK1/2 and total ERK1/2 was also significantly higher in all groups of pituitary tumours in comparison to normal pituitaries (274±13% for PRL-omas, P<0.001; 290±30% for ACTH-omas, P<0.001; 198±15% for GH-omas; P=0.01 and 214±12% for NFPAs; P<0.001; Fig. 2C).
There was no difference in either total mTOR (P=0.46) or its phospho-derivative Ser2448 (P=0.25); in total TSC2 (P=0.78) or its derivatives Ser939-TSC2 (P=0.88) or Thr1462-TSC2 (P=0.84); or in total p70S6K (P=0.72) or its phospho-derivative Thr389 (P=0.47) between any type of pituitary adenoma and normal pituitary.
There was no difference in the expression of Ser62 c-MYC between pituitary adenomas and normal pituitary (P=0.75). However, the expression of c-MYC doubly-phosphorylated at Ser62/Thr58 (Fig. 3A), but not total c-MYC (Fig. 3B), was significantly decreased in all types of adenoma in comparison to normal pituitary (45±4% for PRL-omas, P=0.01; 40±6% for ACTH-omas, P=0.01; 42±6% for GH-omas, P=0.04 and 54±9% for NFPAs, P=0.06). Furthermore, the ratio between c-MYC doubly-phosphorylated at Ser62/Thr58 and total c-MYC was also significantly lower in all secretory groups of pituitary tumours in comparison to normal pituitaries, and tending towards significance in the NFPAs (37±9% for PRL-omas, P=0.01; 33±12% for ACTH-omas, P=0.01; 35±11% for GH-omas; P=0.01 and 56±26% for NFPAs, P=0.09; Fig. 3C).
A significant increase in CYCLIN D1 expression was observed in NFPAs compared to normal controls (136%±10, P=0.01), but not in any of the other adenoma groups (Fig. 4). There was no difference in the amount of pThr180/Tyr182 p38MAPK between the normal or abnormal tissue (P=0.17; data not shown).
Representative western blots are shown for pMEK1/2, tMEK1/2 and β-actin (Fig. 5A), pERK1/2, tERK1/2 and β-actin (Fig. 5B), doubly-phosphorylated c-MYC, total c-MYC and β-actin (Fig. 5C), and CYCLIN D1 and β-actin expression (Fig. 5D), in normal pituitaries, PRL-omas, ACTH-omas, GH-omas and NFPAs in Fig. 5.
Quantification of mRNA showed similar expression for MEK1, MEK2, ERK1, ERK2, c-MYC and CCND1 in all tissue groups (P=0.78 for MEK1; P=0.46 for MEK2; P=0.73 for ERK1; P=0.23 for ERK2; P=0.59 for c-MYC and P=0.62 for CCND1, respectively). Representative results for mRNA of analysed genes normalized to β-actin in controls and NFPAs are shown in Fig. 6. There was no correlation between the amount of protein product assessed by western blotting and mRNA assessed by RT-PCR for any of the molecular species (P=0.10 for ERK1; P=0.41 for ERK2; P=0.54 for c-MYC and P=0.84 for CCND1 respectively).
The Raf/MEK/ERK and PI3K/PTEN/Akt pathways diverge from common growth factor receptors on the cell surface, modulating phosphorylation of many downstream targets involved in cell survival, proliferation, and in some cases in malignant transformation. In this study of human pituitary adenomas, we demonstrate that our previously-reported overexpression of B-Raf (MAPKKK) was associated with increased phosphorylation, and hence activation, of its immediate downstream targets MEK1/2 (MAPKK) and ERK1/2 (MAPK) in all adenoma subtypes. However, we were only able to show a small increase in CYCLIN D1 expression, in NFPAs alone, and we did not find the increase in pSer62 c-MYC expression that would be expected with activation of the MAPK pathway. In terms of the PI3K/Akt pathway, we did not see any increased phosphorylation of TSC2 by Akt nor the anticipated changes secondary to Akt in mTOR or p70S6K. Two N-terminal phosphorylation sites in c-MYC, Thr58 and Ser62, are known to be regulated by mitogen stimulation. Phosphorylation of Ser62 mediated by ERK1/2 is required for Ras-induced stabilisation of c-MYC, but conversely, phosphorylation of Thr58, mediated by GSK-3β (DePaoli-Roach 1984) but dependent on the prior phosphorylation of Ser62, is associated with degradation of c-MYC (Cross et al. 1995, Sears et al. 2000). In this study we found decreased c-MYC phosphorylation at Thr62/Ser58 concordant with previously described Akt over-activation in pituitary adenomas, an effect thought to occur via Akt-mediated phosphorylation of GSK-3β. We were unable to check the status of TSC2 phosphorylated at the site activated by ERK1/2, Ser669, since there is no commercially available antibody to detect this site.
In a very recent animal study performed in a mouse model, which spontaneously develops TSH-secreting pituitary adenomas (Lu et al. 2008), Akt and its downstream effectors mTOR and p70S6K were activated and contributed to increased cell proliferation and pituitary growth. In addition, activation of Akt led to decreased apoptosis by inhibiting pro-apoptotic activity of Bcl-2-associated death promoter, further contributing to the aberrant cell proliferation (Lu et al. 2008). In our study we did not show any alteration in mTOR and p70S6K status in different types of pituitary adenomas, although the very rare human TSH-omas were not included in this analysis.
It is recognised that ligand-dependent or mutational activation of receptor tyrosine kinase, or mutational activation of pathway intermediates such as RAS and BRAF, lead to activation of ERK signalling with demonstrably high intracellular levels of pERK (Cobb & Goldsmith 1995, Lewis et al. 1998). However, it has been suggested that expression of pERK might sometimes be a poor measure of activation of ERK pathway signalling and not necessarily reflect substrate flux through this pathway, although this is a common assumption (Pratilas et al. 2009). Nevertheless, the present data, together with previous studies, do suggest that the two major signalling pathways are likely to be over-active in pituitary tumours, and that therefore the fundamental change in pituitary adenoma cell status occurs either at, or upstream to, the point of convergence of these two pathways; this would be at the level of growth factor receptor(s), or their related control mechanisms. Indeed, a recent study has demonstrated that the membrane anchor E-cadherin is detached from the membrane of adenoma cells compared to normal pituitary (Elston et al. 2009), suggesting that membrane changes and/or matrix-related events are a major feature of such tumours. Our results are compatible with the concept that they may be driving the process rather than occurring in response to cytoplasmic or nuclear signals. Furthermore, this may be therapeutically important as growth factor receptors signal through tyrosine receptor kinases (TRKs), and such TRKs may be blocked by small-molecule TRK inhibitors (TKIs). It is possible that the TKIs may be of value in the treatment of recurrent or resistant pituitary tumours in the future.
Our findings on CYCLIN D1, which is regulated by both ERK1/2 and Akt pathways, are particularly interesting. CYCLIN D1 is a proto-oncogene (11q13) which acts in the pRb/p16/CDK4 pathway on G1-S progression of the cell cycle. CCND1 gene amplification (Hibberts et al. 1999) and polymorphisms (Gazioglu et al. 2007), and enhanced nuclear (Jordan et al. 2000, Gazioglu et al. 2007) and/or cytoplasmic (Hibberts et al. 1999, Jordan et al. 2000, Gazioglu et al. 2007) accumulation, have been described in sporadic pituitary adenomas, especially in NFPAs (Hibberts et al. 1999, Jordan et al. 2000). Over-expression of CYCLIN D1 in NFPAs in comparison to other tumour types, as observed in our study, is essentially in agreement with the study by Hibberts et al. (1999) where CYCLIN D1 was found to be specifically over-expressed in NFPAs in comparison to somatotrophinomas. However, the fact that in some cases of pituitary adenomas CYCLIN D1 over-expression occurred in the absence of CCND1 allelic imbalance suggested that additional mechanisms responsible for deregulated CYCLIN D1 expression are involved in human pituitary tumourigenesis (Hibberts et al. 1999). The mitogen-dependent induction of CYCLIN D1 expression is one of the earliest cell cycle-related events to occur during the G0/G1 to S-phase transition (Lavoie et al. 1996). Activation of the Raf/MEK/ERK cascade has been shown to be sufficient to induce CYCLIN D1 promoter activity and CYCLIN D1 expression, even in the absence of growth factors. In contrast, the p38MAPK cascade showed an opposite effect on the regulation of CYCLIN D1 expression (Lavoie et al. 1996). Based on western blot analysis, we confirmed previous reports of CYCLIN D1 over-expression in NFPAs, but the increase was slight, was not seen in other tumour subtypes, and was not associated with any change in CCND1 mRNA. We speculated that the lack of change in transcription might be secondary to changes in p38MAPK, which decreases CYCLIN D1 transcription, but we did not see any changes in either p38MAPK activation or in the total amount of this protein.
Our results therefore demonstrate that while there is indeed evidence for over-activation of both signalling pathways, their effects are to a major extent attenuated downstream. This suggests that, in these benign tumours, overdrive of the signalling pathways is blunted by protective anti-tumour mechanisms (Aoki et al. 2007, Chesnokova et al. 2008). One such mechanism is the recently – described process of senescence, particularly oncogene-induced senescence, whereby early neoplastic cells are eliminated from the proliferative pool. High pituitary p21 levels have been shown to promote senescence and restrain pituitary tumour growth, particularly in somatotroph tumours, and may underlie the failure of invariably benign pituitary tumours to progress to true malignancy (Chesnokova et al. 2008). In addition, the pleiotropic cytokine interleukin-6 (IL-6) has been demonstrated to mediate senescence in a cell-autonomous, i.e. non-paracrine, mode, and this is compatible with previous data that pituitary adenomas specifically immunostain for IL-6 (Tsagarakis et al. 1992, Jones et al. 1994).
In conclusion, in this study we describe over-activation of the downstream kinases to B-Raf in pituitary adenomas, as well as a slight increase in CYCLIN D1 in non-functioning tumours, but no change in c-MYC phosphorylation and hence stabilisation by ERK1/2. In contrast, the PI3K/Akt pathway shows no increased activity in its major downstream TSC/mTOR components, but there was decreased phosphorylation and thus increased stabilisation of c-MYC by this pathway. We speculate that the causal agent(s) in sporadic pituitary tumourigenesis are upstream to the convergence of these two Ras-dependent pathways, and that induced senescence, possibly involving p21 and/or IL-6, is responsible for the benign nature of the great majority of such tumours.
Declaration of interest
We declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the research reported.
Studies were supported by the CRC of St Bartholomew's Hospital and by OTKA K68660.
Dorota Dworakowska would like to thank the Polish Science Foundation for a post-doctoral fellowship to Barts and The London School of Medicine, London, England. Mehtap Cakir was awarded a grant from The Society of Endocrinology and Metabolism of Turkey to study at the Centre for Endocrinology, Barts and The London School of Medicine and Dentistry, London.
AokiHTakadaYKondoSSawayaRAggarwalBBKondoY2007Evidence that curcumin suppresses the growth of malignant gliomas in vitro and in vivo through induction of autophagy: role of Akt and extracellular signal-regulated kinase signaling pathways. Molecular Pharmacology7229–39.
BeckersADalyAF2007The clinical, pathological, and genetic features of familial isolated pituitary adenomas. European Journal of Endocrinology157371–382.
BoikosSAStratakisCA2007Molecular genetics of the cAMP-dependent protein kinase pathway and of sporadic pituitary tumorigenesis. Human Molecular Genetics16R80–R87.
CrossDAAlessiDRCohenPAndjelkovichMHemmingsBA1995Inhibition of glycogen synthase kinase-3 by insulin mediated by protein kinase B. Nature378785–789.
DePaoli-RoachAA1984Synergistic phosphorylation and activation of ATP-Mg-dependent phosphoprotein phosphatase by F A/GSK-3 and casein kinase II (PC0.7). Journal of Biological Chemistry25912144–12152.
DiehlJAChengMRousselMFSherrCJ1998Glycogen synthase kinase-3beta regulates cyclin D1 proteolysis and subcellular localization. Genes and Development123499–3511.
ElstonMSGillAJConaglenJVClarksonACookRJLittleNSRobinsonBGClifton-BlighRJMcDonaldKL2009Nuclear accumulation of e-cadherin correlates with loss of cytoplasmic membrane staining and invasion in pituitary adenomas. Journal of Clinical Endocrinology and Metabolism941436–1442.
EwingIPedder-SmithSFranchiGRuscicaMEmeryMVaxVGarciaECzirjakSHanzelyZKolaB2007A mutation and expression analysis of the oncogene BRAF in pituitary adenomas. Clinical Endocrinology66348–352.
EzzatSAsaSLCouldwellWTBarrCEDodgeWEVanceMLMcCutcheonIE2004The prevalence of pituitary adenomas: a systematic review. Cancer101613–619.
GaziogluNMErensoyNKadiogluPSayitogluMAErsoyIHHatirnazOKisacikBOzBSarMOzbekU2007Altered cyclin D1 genotype distribution in human sporadic pituitary adenomas. Medical Science Monitor13CR457–CR463.
GuanKL1994The mitogen activated protein kinase signal transduction pathway: from the cell surface to the nucleus. Cellular Signalling6581–589.
HibbertsNASimpsonDJBicknellJEBroomeJCHobanPRClaytonRNFarrellWE1999Analysis of cyclin D1 (CCND1) allelic imbalance and overexpression in sporadic human pituitary tumors. Clinical Cancer Research52133–2139.
HorvathAStratakisCA2008Clinical and molecular genetics of acromegaly: MEN1, Carney complex, McCune–Albright syndrome, familial acromegaly and genetic defects in sporadic tumors. Reviews in Endocrine and Metabolic Disorders91–11.
HubinaENanzerAMHansonMRCiccarelliELosaMGaiaDPapottiMTerreniMRKhalafSJordanS2006Somatostatin analogues stimulate p27 expression and inhibit the MAP kinase pathway in pituitary tumours. European Journal of Endocrinology155371–379.
JonesTHDanielsMJamesRAJusticeSKMcCorkleRPriceAKendall-TaylorPWeetmanAP1994Production of bioactive and immunoreactive interleukin-6 (IL-6) and expression of IL-6 messenger ribonucleic acid by human pituitary adenomas. Journal of Clinical Endocrinology and Metabolism78180–187.
JordanSLidharKKorbonitsMLoweDGGrossmanAB2000Cyclin D and cyclin E expression in normal and adenomatous pituitary. European Journal of Endocrinology143R1–R6.
KarhuAAaltonenLA2007Susceptibility to pituitary neoplasia related to MEN-1, CDKN1B and AIP mutations: an update. Human Molecular Genetics16R73–R79.
Kauffmann-ZehARodriguez-VicianaPUlrichEGilbertCCofferPDownwardJEvanG1997Suppression of c-Myc-induced apoptosis by Ras signalling through PI(3)K and PKB. Nature385544–548.
LavoieJNL'AllemainGBrunetAMullerRPouyssegurJ1996Cyclin D1 expression is regulated positively by the p42/p44MAPK and negatively by the p38/HOGMAPK pathway. Journal of Biological Chemistry27120608–20616.
LeontiouCAGueorguievMvan der SpuyJQuintonRLolliFHassanSChahalHSIgrejaSCJordanSRoweJ2008The role of the aryl hydrocarbon receptor-interacting protein gene in familial and sporadic pituitary adenomas. Journal of Clinical Endocrinology and Metabolism932390–2401.
LidharKKorbonitsMJordanSKhalimovaZKaltsasGLuXClaytonRNJenkinsPJMonsonJPBesserGM1999Low expression of the cell cycle inhibitor p27Kip1 in normal corticotroph cells, corticotroph tumors, and malignant pituitary tumors. Journal of Clinical Endocrinology and Metabolism843823–3830.
LuCWillinghamMCFuruyaFChengSY2008Activation of phosphatidylinositol 3-kinase signaling promotes aberrant pituitary growth in a mouse model of thyroid-stimulating hormone-secreting pituitary tumors. Endocrinology1493339–3345.
McCubreyJASteelmanLSChappellWHAbramsSLWongEWChangFLehmannBTerrianDMMilellaMTafuriA2007Roles of the Raf/MEK/ERK pathway in cell growth, malignant transformation and drug resistance. Biochimica et Biophysica Acta17731263–1284.
MusatMKorbonitsMKolaBBorboliNHansonMRNanzerAMGrigsonJJordanSMorrisDGGueorguievM2005Enhanced protein kinase B/Akt signalling in pituitary tumours. Endocrine-Related Cancer12423–433.
OnofriCTheodoropoulouMLosaMUhlELangeMArztEStallaGKRennerU2006Localization of vascular endothelial growth factor (VEGF) receptors in normal and adenomatous pituitaries: detection of a non-endothelial function of VEGF in pituitary tumours. Journal of Endocrinology191249–261.
PratilasCATaylorBSYeQVialeASanderCSolitDBRosenN2009(V600E)BRAF is associated with disabled feedback inhibition of RAF–MEK signaling and elevated transcriptional output of the pathway. PNAS1064519–4524.
RobertsPJDerCJ2007Targeting the Raf–MEK–ERK mitogen-activated protein kinase cascade for the treatment of cancer. Oncogene263291–3310.
Rodriguez-VicianaPWarnePHKhwajaAMarteBMPappinDDasPWaterfieldMDRidleyADownwardJ1997Role of phosphoinositide 3-OH kinase in cell transformation and control of the actin cytoskeleton by Ras. Cell89457–467.
RosenwaldIBKasparRRousseauDGehrkeLLeboulchPChenJJSchmidtEVSonenbergNLondonIM1995Eukaryotic translation initiation factor 4E regulates expression of cyclin D1 at transcriptional and post-transcriptional levels. Journal of Biological Chemistry27021176–21180.
SearsRNuckollsFHauraETayaYTamaiKNevinsJR2000Multiple Ras-dependent phosphorylation pathways regulate Myc protein stability. Genes and Development142501–2514.
TeradaYInoshitaSNakashimaOKuwaharaMSasakiSMarumoF1999aRegulation of cyclin D1 expression and cell cycle progression by mitogen-activated protein kinase cascade. Kidney International561258–1261.
TeradaYNakashimaOInoshitaSKuwaharaMSasakiSMarumoF1999bMitogen-activated protein kinase cascade and transcription factors: the opposite role of MKK3/6-p38K and MKK1-MAPK. Nephrology Dialysis Transplantation1445–47.
TsagarakisSKontogeorgosGGiannouPThalassinosNWoolleyJBesserGMGrossmanA1992Interleukin-6, a growth promoting cytokine, is present in human pituitary adenomas: an immunocytochemical study. Clinical Endocrinology37163–167.