ISSN 1470-3947 (print) | ISSN 1479-6848 (online)

Endocrine Abstracts (2008) 16 P454

Aryl hydrocarbon receptor interacting protein (AIP) expression in human pituitary adenomas

Marie-Lise Jaffrain-Rea1, Cristina Spilioti1, Jean-François Vanbellinghen3, Adrian Daly4, Donatella Gargano1, Emanuela D’Innocenzo1, Felice Giangaspero2, Maria Tichomirowa4, Vincenzo Esposito2, Luca Ventura5, Luciana Naves6, Edoardo Alesse1 & Albert Beckers4

1Department of Experimental Medicine, University of L’Aquila, L’Aquila, Italy; 2Neuromed Institute, IRCCS, Pozzilli, Italy; 3Genetics, University of Liège, Liège, Belgium; 4Endocrinology, University of Liège, Liège, Belgium; 5Pathology, S. Salvatore Hospital, L’Aquila, Italy; 6Endocrinology, University of Brasilia, Brasilia, Brazil.

Background: Germline AIP mutations confer a predisposition to pituitary adenomas (PA), usually in the setting of familial isolated pituitary adenomas (FIPA); AIP mutations account for 50% of familial acromegaly. AIP-related PA are GH, PRL-secreting or non-secreting. Little is known about AIP expression in PA. Although the prevalence is low, identifying AIP mutations in apparently sporadic PA is important for studying at-risk relatives. Pre-screening criteria would help select patients for mutational analysis.

Material and methods: AIP expression was studied by Real Time RT-PCR in 45 PA, including 3 FIPA acromegaly cases with AIP mutations (AIPmut), and by immunohistochemistry (IHC) in 56 PA, including 9 AIPmut cases (7 GH, 1 PRL and 1 non-secreting PA). AIP immunostaining was scored semi-quantitatively. All PA phenotypes were represented and normal pituitaries (NP) were used as controls.

Results: AIP transcripts were detected in all cases. Mean transcript levels in PA were similar those observed in NP, but variations were observed, with 20% with moderate overexpression (mainly non-secreting and somatotropinomas) and 10% with underexpression, including the 3 AIPmut cases. AIP protein was detected in most PA (49/56), with immunostaining generally less intense than in NP (P<0.001). The strongest immunostaining was observed in somatotropinomas, though AIPmut somatotropinomas had a lower AIP score than other somatotrotropinomas (P=0.023). Only 2/9 PA from AIPmut patients (1 GH-, 1-PRL-secreting) showed a complete loss of immunostaining.

Conclusion: AIP down-regulation is frequently observed in PA, mostly at a protein level. AIPmut somatotropinomas show AIP down-regulation at both mRNA and protein levels, but AIP immunostaining is totally abolished in only a minority of cases. Though promising, further experience is needed to integrate IHC as a pre-screening tool for AIP mutational studies in patients with PA.