Analysis of pituitary adenoma expression patterns suggests a potential role for the NeuroD1 transcription factor in neuroendocrine tumor-targeting therapies

NeuroD1’s roles in the pathogenesis of pituitary adenomas and in the biology of the normal adult pituitary gland have been insufficiently researched. Much of the work investigating its expression patterns has yielded contradictory results. Objective: morphological study of NeuroD1 transcription factor expression in different types of pituitary adenomas and in normal adult human pituitary glands. Materials and methods: This study analyzed 48 pituitary adenomas and nine normal pituitary glands. In all cases, immunohistochemical study was performed with antibodies to NeuroD1, 6 hormones of adenohypophysis, Ki-67, and CK7. We used confocal laser scanning microscopy, electron microscopy and electron immunocytochemistry. Results: NeuroD1 expression was detected in all cases of plurihormonal adenomas, mammosomatotropinomas, corticotropinomas, prolactinomas, gonadotropinomas, null-cell pituitary adenomas, and in normal pituitary glands. The average numbers of NeuroD1 expressing cells in normal adenohypophysis specimens were significantly lower than in the adenomas overall (p=0.006). NeuroD1 expression was confirmed by several methods (in prolactinomas, by double stain immunohistochemistry; in mammosomatotropinomas, by double stain immunohistochemistry, confocal laser scanning microscopy, and electron immunocytochemistry; and in somatotropinomas, by electron immunocytochemistry). Conclusion: Immunohistochemistry, confocal microscopy, and double label electron immunocytochemistry confirmed NeuroD1’s key role in the pathogenesis of pituitary tumors, regardless of their hormonal state. Its expression level in pituitary adenomas is significantly higher than in the normal pituitary gland and has no reliable correlation with any studied hormones or Ki-67. These findings suggest that NeuroD1 should be investigated further as a potential molecular target in tumor-targeting therapies.


Confocal laser scanning microscopy
The study was done on deparaffinized and rehydrated sections ranging from 4 to 10 μm in thickness. Heat-induced epitope retrieval was performed with 0.01 M citrate buffer (pН 6.0) under pressure. PBS buffer with Tween 20 was used as a wash buffer. Then, sections were incubated for 30 minutes in blocking solution (PBS containing 2% BSA) at room temperature. After washing, the first primary NeuroD1 antibodies were applied for 1 hour at room temperature. Then, after additional washing, the sections were incubated with an additional primary antibody (GH or prolactin) for 1 hour at room temperature. Alexa Fluor 647 ® -labeled anti-mouse secondary (Abcam, UK) and Alexa Fluor 488 ® -labeled anti-rabbit secondary antibodies (Abcam, UK) were used for visualization.
After washing, nuclei in the sections were stained with DAPI (AppliChem). Dako Mounting Medium (DAKO, Denmark) was used for mounting all tissue specimens. As a result, NeuroD1 signal was seen as red fluorescence, and GH or prolactin signals were seen as green fluorescence; contrasted nuclei were seen as blue fluorescence. Preparations were analyzed using the Olympus FV1000D confocal laser scanning microscope. We evaluated the intensity and colocalization of NeuroD1 expression and DAPI fluorescence.
Micrographs of 4 mammosomatotropinomas (5 fields per each sample, at 400x magnification), were analyzed for double staining patterns (GH/NeuroD1 or prolactin/NeuroD1). The co-expression coefficient of hormones and NeuroD1 in the same cells was defined as the ratio of the double-stained cells to the total number of cells, expressed as a percentage. The NeuroD1 expression coefficient was calculated as the ratio of NeuroD1positive nuclei to DAPI stained nuclei (NeuroD1/DAPI, as a percent). The NeuroD1 expression coefficient and coexpression coefficients (of hormones and NeuroD1) were determined using image analysis software (Image Scope Color M, Russia).

Electron immunocytochemistry
Small (1-2 mm) fragments of tumors were first fixed in PBS-buffered 4% paraformaldehyde solution, containing 0.2% glutaraldehyde, for 1 hr and post-fixed in 1% PBS-buffered OsO 4 solution for 1 hour. Next, specimens were dehydrated, using a series of increasing ethanol concentrations, and embedded in LR White resin (Sigma-Aldrich Inc., St.-Louis, Missouri, USA). The resin was polymerized in tightly-capped gelatin capsules at +52 °C. Ultrathin sectioning (60-80 nm) of embedded specimens was performed using an EM UC7 ultramicrotome (Leica, Germany). Ultrathin sections of samples were collected on nickel electron microscopy grids. In order to prevent non-specific binding of primary antibodies, the sections on grids were incubated in PBS containing 1% bovine serum albumin (BSA-PBS, Sigma-Aldrich Inc.) at room temperature for 15 min.
In the ultrathin sections, an indirect immunogold labeling procedure was used for the detection of NeuroD1 in mammosomatotropinoma specimens and for the detection of NeuroD1 and GH in somatotropinoma specimens. Mouse monoclonal anti-NeuroD1 antibody (clone ab60704, Abcam, United Kingdom), diluted 1:400, was used for NeuroD1 detection. Rabbit polyclonal anti-GH antibody (BioGenex, USA), diluted 1:100, was used for GH detection. As secondary antibodies, we used goatanti mouse antibody conjugated to 10nm colloidal gold (Sigma-Aldrich Inc., St.-Louis, Missouri, USA) and goatanti rabbit antibody conjugated to 5nm colloidal gold (Sigma-Aldrich Inc., St.-Louis, Missouri, USA), both diluted 1:100.
Sections on grids were incubated in phosphate buffer containing primary antibodies for 1 hour, and subsequently washed in PBS buffer containing 0.05% TWEEN 20. Next, grids were incubated with secondary antibodies for 1 hour and washed, in the same manner, with PBST. Finally, sections on grids were contrasted in aqueous solutions of uranyl acetate, followed by lead citrate. Electron-microscopic examination was carried out using a JEM 1011 transmission electronic microscope (JEOL, Tokyo, Japan) equipped with a high-resolution digital camera (Morada, Olympus, Japan).