BackgroundTransforming growth factor β (TGF-β) signaling functions as a suppressor or a promoter in tumor development, depending on the tumor stage and type. However, the role of TGF-β signaling in nonfunctioning pituitary adenomas (NFPAs) has not been explored.MethodsTGF-β1, Smad2, phospho-Smad2 (p-Smad2), Smad3, phospho-Smad3 (p-Smad3), Smad4, and Smad7 were detected in 5 cases of normal anterior pituitaries, 29 cases of invasive NFPAs, and 21 cases of noninvasive NFPAs by real-time quantitative reverse transcription polymerase chain reaction (qRT-PCR), Western blot, and immunohistochemical analysis.ResultsThe Smad3 and p-Smad3 protein levels gradually decreased from normal anterior pituitaries, noninvasive NFPAs, to invasive NFPAs. However, there were no significant differences in Smad2 (P = 0.122) and p-Smad2 protein levels (P = 0.101) or Smad2 mRNA level (P = 0.409). In addition, the TGF-β1 mRNA level gradually decreased while the Smad7 mRNA level gradually increased from normal anterior pituitaries, noninvasive NFPAs, to invasive NFPAs. Furthermore, proliferating cell nuclear antigen (PCNA) mRNA level was markedly increased in invasive NFPAs compared to noninvasive ones (P < 0.01), and its level was negatively correlated with Smad3 mRNA level (P < 0.01).ConclusionThe activity of TGF-β signaling may be restrained in NFPAs and is correlated with the development and invasion of NFPAs.

No abstract

Non-functioning pituitary adenomas (NFPAs) are usually macroadenomas and display invasion into surrounding tissues. The treatment for invasive NFPAs is still challenging. This study describes the differential patterns of gene expression between invasive and non-invasive NFPAs and identifies novel biomarkers involved in invasion of NFPAs for diagnosis and treatment. Using gene microarray technology, we examined the gene expression profile and found 1160 differentially expressed messenger RNA (mRNA) between invasive and non-invasive NFPAs. Then, we examined the protein profile by liquid chromatography tandem mass spectrometry (LC-MS/MS) and found 433 differentially expressed proteins between invasive and non-invasive NFPAs. Subsequently, we integrated the proteomics and transcriptomics datasets and identified 29 common changed molecules. Through bioinformatics analysis using Ingenuity Pathway Analysis (IPA) software, we showed that the 29 molecules were enriched in 25 canonical signaling pathways, 25 molecular and cellular functions, and 2 networks. Eight genes were identified involved in the invasion function by the molecular and cellular functions analysis, including CAT, CLU, CHGA, EZR, KRT8, LIMA1, SH3GLB2 and SLC2A1. Furthermore, we validated the decreased CHGA expression and increased CLU expression in invasive NFPAs by qRT-PCR and Western blot. Our study demonstrated that integration of proteomics and transcriptomics could prove advantageous for accelerating tumor biomarker discovery and CHGA and CLU might be important novel biomarkers and therapeutic targets for invasion of NFPAs.