Cusimano, Michael D.

Pituitary tumor transforming gene (PTTG) was initially found to be overexpressed in GH4 rat pituitary adenomas (PA). PTTG overexpression has been correlated with patient survival, tumor grade, and/or invasiveness in several tumors including those of thyroid, breast, lung, and colon. Studies of PTTG as a prognostic indicator in PAs are limited and show conflicting results. We investigated PTTG immunostaining in 79 PAs of different subtypes (growth hormone (GH), 12; prolactin (PRL), 6, adrenocorticotropic hormone (ACTH), 11, gonadotroph, 21; null cell, 29) in relation to clinico‐pathological variables including patient age and sex, as well as tumor size, and invasiveness. The streptavidin‐biotin‐peroxidase method was used, with a monoclonal PTTG antibody (Abcam, Cambridge, UK). PTTG expression was found to be cytoplasmic in all PAs, with highest levels being present in GH adenomas and lowest in the PRL ones. No correlation was found with patient age, sex, tumor size or invasiveness. Thus, it appears that PTTG cannot serve as a prognostic marker in patients with PAs. PTTG has been implicated in several tumorigenic processes including angiogenesis, DNA damage repair and invasiveness. PTTG overexpression in PA warrants further research into its role in PA initiation and progression.

Estrogen is involved in pituitary tumorigenesis, acting through estrogen receptors alpha (ERα) and beta (ERβ). Our study investigated the prognostic value of ERα/ERβ as markers in pituitary adenomas. Tissue samples and clinico‐pathological data (patient age and sex, tumor size, invasiveness and recurrence) were obtained from 78 patients with pituitary adenomas. Immunohistochemistry was performed using the streptavidin‐biotin‐peroxidase complex method. A monoclonal and polyclonal antibody were used for ERα and ERβ expression respectively. Intensity of ERα/ERβ expression was evaluated using a scale of 0 to 3. Percentage of positive cells was evaluated on a scale of 0 to 4. ERα immunopositivity was evident in 40 of 78 cases (51%), while ERβ stained positive in 49 of 78 cases (63%). No correlation was found between ERβ expression and pituitary tumor cell types or clinico‐pathological factors. A significant difference was present in ERα expression between different tumor types. ERα positivity was present to a greater extent in macroadenomas. This finding supports the role of ERα in tumor growth and progression. Elevated ERα positivity and intensity were also found in non‐invasive adenomas. It appears that ERα may be a valuable biological marker for tumor size and invasiveness, while ERβ is a poor marker. The role of ERα expression in pituitary tumor initiation and progression warrants further investigation.

The aim of our work was to investigate immunohistochemical expression of nestin, a member of the intermediate filament family, in the adenohypophysial vasculature during the development and progression of pituitary infarction. Non‐tumorous adenohypopheses (45 cases) and pituitary adenomas (34 cases) of various types, all exhibiting acute or healing infarcts were examined. The streptavidin‐biotin‐peroxidase complex method was applied. In areas of adenohypophyses and pituitary adenomas with no infarction, nestin was expressed in scattered capillaries and endothelial cells. In acute infarcts, no nestin was demonstrable in the necrotic capillaries (50 cases). Where newly formed capillaries spread into necrotic zones, nestin expression was noted in all capillaries (25 cases). In the hypocellular, fibrotic scar phase, only few vessels were apparent and immunoreactivity was focal and mild (4 cases). In conclusion, nestin is strongly expressed in newly formed capillaries and is downregulated when infarcts transform to fibrous tissue. Nestin expression may provide valuable insight into the process of pituitary angiogenesis.

Pituitary tumor transforming gene (PTTG) plays a role in many cellular processes. PTTG overexpression is seen in several tumor types and correlates with survival time/recurrence. We evaluated PTTG expression in various types of brain tumors (n=94). Immunohistochemistry was performed using a monoclonal PTTG antibody (DCS‐280, Abcam;Cambridge, MA) and the streptavidin‐biotin‐peroxidase complex method. The intensity of PTTG immunoreactivity was evaluated semi‐quantitatively on a 0‐4 scale. PTTG expression was evident in most tumor cells and was predominantly nuclear. In glial tumors, PTTG immunoreactivity was higher in glioblastomas (IV), anaplastic oligoastrocytomas (III), anaplastic oligodendrogliomas (III), oligoastrocytomas (II), oligodendrogliomas (II), and pilocytic astrocytomas (I) (range: 3.1–3.5), whereas notably lower PTTG was seen in myxopapillary ependymomas (I) and ependymomas (II) (1.5 & 1.6). In non‐glial tumors, hemangiopericytomas and schwannomas had higher PTTG score (3.4), than meningiomas (2.3). Thus, it appears that PTTG expression is not associated with tumor grade, but rather with tumor type, the most striking difference being between ependymomas and other glial tumors. PTTG may be a valuable therapy target in some brain tumors. Acknowledgments Authors thank the Jarislowsky Foundation and the Lloyd‐Carr‐Harris Foundation for their generous support.

Ghrelin, a 28 amino acid‐containing orexigenic peptide, is an endogenous ligand of the GH secretagogue receptor. It was extracted from the rat stomach, but is present in other organs. The aim of this work was to localize ghrelin by immunohistochemistry in autopsy‐derived non‐tumorous human pituitaries. Immunohistochemistry was performed using the streptavidin‐biotin‐peroxidase complex method and the results were analyzed independently by 3 of the authors. Double Immunostaining was also undertaken to determine which cell type was expressing both the adenohypophysial hormone and ghrelin. Results showed that ghrelin can be localized in the adenohypophysis. It's immunoexpression being cytoplasmic, weak to moderate, and present in only a subset of cells. The neurohypophysis is immunonegative. Double immunostaining showed that ghrelin is co‐localized with GH, ACTH and LH and is present to 51–90% of such cells. PRL, TSH and FSH cells are consistently non‐reactive. The role of ghrelin in adenohypophysial endocrine activity is not clear. It may well be that ghrelin exerts an autocrine‐paracrine effect and can modulate hormone synthesis and release.

Ghrelin, an orexigenic hormone occurs mainly in stomach, and in several tumor types. We investigated ghrelin immunopositivity using the streptavidin‐biotin‐peroxidase complex method in the full spectrum of pituitary adenoma subtypes. Quantification was undertaken by analyzing both the percentage of tumor cells stained for ghrelin and the staining intensity. Cytoplasmic ghrelin immunopositivity was identified in several adenoma subtypes. Not every cell was immunopositive; the intensity of staining differed among cells within the given lesion and among the various tumors. The highest scores were noted in GH adenomas exposed to somatostatin analogues. In decreasing order, lower scores were encountered in ACTH‐producing, silent subtype 1 and 2 adenomas in Cushing disease, silent subtype 3 adenomas, untreated GH adenomas, dopamine agonist‐treated PRL adenomas, gonadotroph adenomas, and ACTH adenomas in Nelson syndrome. No immunoreactivity was found in TSH, untreated PRL‐producing, and null cell adenomas. The high immunoexpression of ghrelin in GH adenomas exposed to somatostatin analogues, may be due to increased ghrelin synthesis or to suppression of its release.

Previous studies showed that in pregnancy the pituitary enlarges and increases in weight due to lactotroph hyperplasia. We demonstrated that in pituitaries of pregnant women, microvessel density is significantly increased compared to those of age– matched non pregnant women. Vascular endothelial growth factor (VEGF) stimulates new vessel formation in several tissues and neoplasms. The aim of the present study was to reveal whether VEGF is involved in the causation of increased vascularity in the pituitaries of pregnant women. We assessed VEGF immunoexpression in 20 autopsy obtained pituitaries of pregnant women and compared the results with 20 controls. Slides were immunostained for VEGF and pituitary hormones using the streptavidin‐biotin‐peroxidase complex method. VEGF immunoexpression was evaluated in 10 random high‐power fields using an image analyzer and the results compared by chi‐square or Fisher exact tests. Immunoexpression of VEGF was not increased in the adenohypophyses of pregnant women compared with the controls indicating that VEGF plays no major role in the causation of increased vascularity. The cause remains to be investigated.

Pituitary enlargement, development of pregnancy cells and prolactin cell hyperplasia are known to occur in pregnancy. In the present work, we assessed microvessel densities in the 20 pituitaries of pregnant women who passed away. Twenty pituitaries of age matched non‐pregnant women who died from non‐endocrine diseases were used as controls. The autopsy obtained pituitaries were formalin fixed and paraffin embedded. Slides were immunostained for adenohypophysial hormones and for CD‐34, an endothelial cell biomarker using the streptavidin‐biotin‐ peroxidase complex method. Microvessel densities were evaluated quantitatively in 10 random high‐ power fields. For statistical analysis, the chi‐square and Fisher exact tests were applied. The results confirmed that the pituitaries of pregnant women increased in weight and contained more PRL immunopositive cells compared to the controls. The formation of pregnancy cells was also evident. In the pituitaries of pregnant women compared to the control, microvessel densities were significantly higher. The cause of increased vascularity remains to be evaluated. This work was supported by the Jarislowsky and Lloyd Carr‐Harris Foundations.

Angiogenesis is a crucial event in tumor growth, invasion and development of metastasis. Endoglin (CD‐105), a disulfide‐linked homodimeric cell membrane glycoprotein of 180 kDa, located on chromosome 9q and binds to several components of the TGF beta family. In contrast to other endothelial cell markers, endoglin can be demonstrated only in proliferating endothelial cells. We investigated endoglin immunoexpression in various surgically removed human pituitary tumor types and compared the results with CD‐34 immunoexpression. For immunohistochemistry, the streptavidin‐biotin‐peroxidase complex method was used and the number of vessels were counted in 10 low power fields. Results show that endoglin immunostains fewer vessels than CD‐34 in every pituitary adenoma type. The number of endoglin immunopositive vessels is highest in untreated prolactinomas and subtype 3 silent pituitary adenomas and lowest in subtype 1 silent corticotroph adenomas and GH adenomas exposed to somatostatin analogs and prolactinomas exposed to dopamine agonists. Our findings are consistent with the view that endoglin is upregulated in proliferating endothelial cells and its expression can be used as an indicator of vascular neoformation in pituitary tumors. The low levels of endoglin expression in pituitary tumors exposed to somatostatin or dopamine agonists suggests that these therapeutic drugs can inhibit angiogenesis.

Following traumatic brain injury (TBI) approximately 25% of patients develop partial or complete hypopituitarism (HP). In our previous histologic study we found no changes in the pituitaries of patients who died instantly after TBI (Group I). In 43% of patients who survived 1‐7 days acute pituitary infarcts were apparent (Group II). In the present work, 3 group I and 7 group II pituitaries with infarcts were selected for histological and immunohistochemistry (IH) evaluation. IH was performed for adenohypophyseal hormones by the streptavidin‐biotin‐peroxidase method. In group I pituitaries, no histological changes were found and IH showed normal distribution and intensity for GH, PRL, ACTH, TSH, FSH, LH and α‐subunit (α‐SU). In group II pituitaries acute infarct (size range: 10% to 80%) was evident. IH in the non‐necrotic areas revealed no significant change. In the area of early necrosis the dying cells were still immunopositive (IP) for GH, PRL, ACTH, TSH, FSH, LS, and α‐SU. Even in the necrotic cells with loss of cell membrane, nuclear pyknosis, rhexis and lysis IP cytoplasms were still apparent indicating that hormone release is affected by ischemia. Subsequently the intensity of IP cells gradually decreased and IP granules leaked out from the necrotic cells and were seen extracellularly. In conclusion infarction and loss of adenohypophyseal cells may contribute to HP but other factors should also be considered.