OR26-2 Increased expression of miR-34a explains the low AIP protein level present in 50% of sporadic somatotropinomas

Program: Abstracts - Orals, Featured Poster Presentations, and Posters
Session: OR26-Pituitary Tumorigenesis: Advances in Mechanism & Treatment
Translational
Sunday, June 16, 2013: 11:15 AM-12:45 PM
Presentation Start Time: 11:30 AM
Room 130 (Moscone Center)
Giampaolo Trivellin*1, Leandro Kasuki2, Judit Denes3, Leandro Machado Colli4, Margaret De Castro4, Monica Gadelha2 and Marta Korbonits3
1Barts and The London School of Medicine, Queen Mary University of London, London, United Kingdom, 2Federal University of Rio de Janeiro, Rio de Janeiro, Brazil, 3Barts and the London School of Medicine, London, United Kingdom, 4School of Medicine of Ribeirao Preto-University of Sao Paulo, Ribeirao Preto-SP, Brazil
Half of sporadic somatotropinomas show low AIP protein expression (1,2). These tumors usually exhibit an invasive phenotype and show decreased response to somatostatin analogue (SSA) treatment, corresponding to similar behaviour of somatotropinomas in patients with germline AIP mutations. To study the mechanism of low AIP protein expression in sporadic somatotropinomas we studied AIP mRNA expression and the expression of microRNAs (miRs) predicted in silico to bind to the 3’ untranslated region (3’UTR) of AIP.

Thirty one sporadic somatotropinomas with no AIP mutations with low (n=13) or high (n=18) AIP protein expression were analyzed for expression of AIP mRNA and 13 miRs predicted to bind the 3’UTR of AIP using RT-qPCR. A luciferase reporter assay was used to study the effect of selected miRs on 3’UTR of AIP in GH3 cells. miRs with significant effect were further studied following introduction of mutations at the predicted miR binding sites.

There was no significant difference in AIP mRNA expression between tumors with low or high AIP expression. This suggests that the expression of AIP protein might be regulated post-transcriptionally. One of the possible mechanisms is via miRs, which can repress protein expression by inhibiting translation. Out of the strictly selected predicted miRs, (let-7b, let-7a, miR-202, -22, -31, -324, -34a, -34c, -449, -510, -612, -639, -671) the expression of miR-22 and miR-34a was higher in tumors with low AIP expression than in tumors with high AIP expression (2.5x for miR-22, p=0.046 and 3x for miR-34a, p=0.022). Co-transfection of miR-34a precursor and the luciferase-WT-AIP-3’UTR plasmid construct showed a significant negative effect on the luciferase expression (31±4% decrease, p<0.001), suggesting that miR-34a binds to the AIP-3’UTR, while miR-22 showed no inhibition. miR-34a is predicted in silico to interact with 3 different specific target sequences located within the 3’-UTR of AIP. By using deletion mutants of each of these 3 sites, we have identified the c.*6-30 target site to be involved in miR-34a’s activity, as mutation of this binding site completely abolished the inhibitory effect of miR-34a on luciferase activity.

We have identified and proved that miR-34a is a negative regulator of AIP protein expression and hypothesized that miR-34a could be responsible for the low AIP expression observed in half of somatotropinomas which can lead to invasive phenotype and resistance to SSA.

(1) Jaffrain-Rea et al., Endocr Relat Cancer 2009, 16(3):1029-43. (2) Kasuki et al., Endocr Relat Cancer 2012, 19(3):L25-9

Nothing to Disclose: GT, LK, JD, LMC, MD, MG, MK

*Please take note of The Endocrine Society's News Embargo Policy at http://www.endo-society.org/endo2013/media.cfm

Sources of Research Support: Leandro Kasuki has received grants from Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq) -Brazil