Next-generation sequencing of microRNAs reveals a unique expression pattern in different types of pituitary adenomas

Micromanaging the neuroendocrine system - A review on miR-7 and the other physiologically relevant miRNAs in the hypothalamic-pituitary axis

The hypothalamic-pituitary axis is central to the functioning of the neuroendocrine system and essential for regulating physiological and behavioral homeostasis and coordinating fundamental body functions. The expanding line of evidence shows the indispensable role of the microRNA pathway in regulating the gene expression profile in the developing and adult hypothalamus and pituitary gland. Experiments provoking a depletion of miRNA maturation in the context of the hypothalamic-pituitary axis brought into focus a prominent involvement of miRNAs in neuroendocrine functions. There are also a few individual miRNAs and miRNA families that have been studied in depth revealing their crucial role in mediating the regulation of fundamental processes such as temporal precision of puberty timing, hormone production, fertility and reproduction capacity, and energy balance. Among these miRNAs, miR-7 was shown to be hypothalamus-enriched and the top one highly expressed in the pituitary gland, where it has a profound impact on gene expression regulation. Here, we review miRNA profiles, knockout phenotypes, and miRNA interaction (targets) in the hypothalamic-pituitary axis that advance our understanding of the roles of miRNAs in mammalian neurosecretion and related physiology.

The molecular biology of sporadic acromegaly

GH-secreting tumors represent 15 % to 20 % of all pituitary neuroendocrine tumors (pitNETs), of which 95 % occur in a sporadic context, without an identifiable inherited cause. Recent multi-omic approaches have characterized the epigenomic, genomic, transcriptomic, proteomic and kynomic landscape of pituitary tumors. Transcriptomic analysis has allowed us to discover specific transcription factors driving the differentiation of pituitary tumors and gene expression patterns. GH-secreting, along with PRL- and TSH-secreting pitNETs are driven by POU1F1; ACTH-secreting tumors are determined by TBX19; and non-functioning tumors, which are predominantly of gonadotrope differentiation are conditioned by NR5A1. Upregulation of certain miRNAs, such as miR-107, is associated with tumor progression, while downregulation of others, like miR-15a and miR-16-1, correlates with tumor size reduction. Additionally, miRNA expression profiles are linked to treatment resistance and clinical outcomes, providing insights into potential therapeutic targets. Specific somatic mutations in GNAS, PTTG1, GIPR, HGMA2, MAST and somatic variants associated with cAMP, calcium signaling, and ATP pathways have also been associated with the development of acromegaly. This review focuses on the oncogenic mechanisms by which sporadic acromegaly can develop, covering a complex series of molecular alterations that ultimately alter the balance between proliferation and apoptosis, and dysregulated hormonal secretion.

[Plasma miRNA expression in patients with genetically confirmed multiple endocrine neoplasia type 1 syndrome and its phenocopies]

BACKGROUND

MEN-1 is a rare autosomal dominant disease caused by mutations in MEN1 gene encoding the menin protein. This syndrome is characterized by the occurrence of parathyroid tumors, gastroenteropancreatic neuroendocrine tumors, pituitary adenomas, as well as other endocrine and non-endocrine tumors. If a patient with the MEN-1 phenotype carry no mutations in the MEN1 gene, the condition considers a phenocopy of syndrome (phMEN1). The possible cause of this changes could be changes in epigenetic regulation, particularly in microRNA expression that might affect menin signaling pathways.

AIM

to identify differently expressed circulating miRNAs in plasma in patients with genetically confirmed MEN-1 syndrome, its phenocopies and healthy controls.

MATERIALS AND METHODS

single-center, case-control study was conducted. We assessed plasma microRNA expression in patients with genetically confirmed MEN-1 (gMEN1), phMEN1 and healthy controls. Morning plasma samples were collected from fasting patients and stored at -80°C. Total RNA isolation was performed using miRNeasy Mini Kit with QIAcube. The libraries were prepared by the QIAseq miRNA Library Kit following the manufacturer. Circulating miRNA sequencing was done on Illumina NextSeq 500 (Illumina). Subsequent data processing was performed using the DESeq2 bioinformatics algorithm.

RESULTS

we enrolled 21 consecutive patients with gMEN1 and 11 patients with phMEN1, along with 12 gender matched controls. Median age of gMEN1 was 38,0 [34,0; 41,0]; in phMEN1 - 59,0 [51,0; 60,0]; control - 59,5 [51,5; 62,5]. The gMEN1 group differed in age (p<0.01) but not gender (р=0.739) or BMI (р=0.116) compared to phMEN1 and controls group, the last two groups did not differ by these parameters (p>0.05). 25 microRNA were differently expressed in groups gMEN1 and phMEN1 (21 upregulated microRNAs, 4 - downregulated). Comparison of samples from the phMEN-1 group and relatively healthy controls revealed 10 differently expressed microRNAs: 5 - upregulated; 5 - downregulated. In the gMEN-1 and control groups, 26 differently expressed microRNAs were found: 24 - upregulated; 2 - downregulated. The miRNAs most differing in expression among the groups were selected for further validation by RT-qPCR (in the groups of gMEN1 vs phMEN1 - miR-3613-5p, miR-335-5p, miR-32-5p, miR-425-3p, miR-25-5p, miR-576-5p, miR-215-5p, miR-30a-3p, miR-141-3p, miR-760, miR-501-3p; gMEN1 vs control - miR-1976, miR-144-5p miR-532-3p, miR-375; as well as in phMEN1 vs control - miR-944, miR-191-5p, miR-98-5p).

CONCLUSION

In a pilot study, we detected microRNAs that may be expressed differently between patients with gMEN-1 and phMEN-1. The results need to be validated using different measurement method with larger sample size.

Emerging Roles of miRNA, lncRNA, circRNA, and Their Cross-Talk in Pituitary Adenoma

Pituitary adenoma (PA) is a common intracranial tumor without specific biomarkers for diagnosis and treatment. Non-coding RNAs (ncRNAs), including microRNAs (miRNA), long non-coding RNA (lncRNA), and circular RNA (circRNA), regulate a variety of cellular processes, such as cell proliferation, differentiation, and apoptosis. Increasing studies have shown that the dysregulation of ncRNAs, especially the cross-talk between lncRNA/circRNA and miRNA, is related to the pathogenesis, diagnosis, and prognosis of PA. Therefore, ncRNAs can be considered as promising biomarkers for PA. In this review, we summarize the roles of ncRNAs from different specimens (i.e., tissues, biofluids, cells, and exosomes) in multiple subtypes of PA and highlight important advances in understanding the contribution of the cross-talk between ncRNAs (e.g., competing endogenous RNAs) to PA disease.

Genome wide analysis of circulating miRNAs in growth hormone secreting pituitary neuroendocrine tumor patients’ plasma

Background

Circulating plasma miRNAs have been increasingly studied in the field of pituitary neuroendocrine tumor (PitNET) research. Our aim was to discover circulating plasma miRNAs species associated with growth hormone (GH) secreting PitNETs versus assess how the plasma levels of discovered miRNA candidates are impacted by SSA therapy and whether there is a difference in their levels between GH secreting PitNETs versus other PitNET types and healthy individuals.

Design

We compared plasma miRNA content and levels before and after surgery focusing on GH secreting PitNET patients. Selected miRNA candidates from our data and literature were then tested in a longitudinal manner in somatostatin analogues (SSA) treatment group. Additionally, we validated selected targets in an independent GH secreting PitNET group.

Methods

miRNA candidates were discovered using the whole miRNA sequencing approach and differential expression analysis. Selected miRNAs were then analyzed using real-time polymerase chain reaction (qPCR).

Results

Whole miRNA sequencing discovered a total of 16 differentially expressed miRNAs (DEMs) in GH secreting PitNET patients’ plasma 24 hours after surgery and 19 DEMs between GH secreting PitNET patients’ plasma and non-functioning (NF) PitNET patients’ plasma. Seven miRNAs were selected for further testing of which miR-625-5p, miR-503-5p miR-181a-2-3p and miR-130b-3p showed a significant downregulation in plasma after 1 month of SSA treatment. mir-625-5p was found to be significantly downregulated in plasma of GH secreting PitNET patients vs. NF PitNET patients. miR-625-5p alongside miR-130b-3p were also found to be downregulated in GH PitNETs compared to healthy individuals.

Conclusions

Our study suggests that expression of plasma miRNAs miR-625-5p, miR-503-5p miR-181a-2-3p and miR-130b-3p in GH secreting PitNETs is affected by SSA treatment. Additionally, miR-625-5p can distinguish GH secreting PitNETs from other PitNET types and healthy controls warranting further research on these miRNAs for treatment efficacy.

Genetic and Epigenetic Pathogenesis of Acromegaly

Simple Summary

Various genetic and epigenetic factors are involved in the pathogenesis of somatotroph tumors. Although GNAS mutations are the most prevalent cause of somatotroph tumors, the cause of half of all pathogenesis occurrences remains unclarified. However, recent findings including the pangenomic analysis, such as genome, transcriptome, and methylome approaches, and histological characteristics of pituitary tumors, the involvement of AIP and GPR101, the mechanisms of genomic instability, and possible involvement of miRNAs have gradually unveiled the whole landscape of underlying mechanisms of somatotroph tumors. In this review, we will focus on the recent advances in the pathogenesis of somatotroph tumors.

Abstract

Acromegaly is caused by excessive secretion of GH and IGF-I mostly from somatotroph tumors. Various genetic and epigenetic factors are involved in the pathogenesis of somatotroph tumors. While somatic mutations of GNAS are the most prevalent cause of somatotroph tumors, germline mutations in various genes (AIP, PRKAR1A, GPR101, GNAS, MEN1, CDKN1B, SDHx, MAX) are also known as the cause of somatotroph tumors. Moreover, recent findings based on multiple perspectives of the pangenomic approach including genome, transcriptome, and methylome analyses, histological characterization, genomic instability, and possible involvement of miRNAs have gradually unveiled the whole landscape of the underlying mechanisms of somatotroph tumors. In this review, we will focus on the recent advances in genetic and epigenetic pathogenesis of somatotroph tumors.

The cAMP-signaling cancers: Clinically-divergent disorders with a common central pathway

The cAMP-signaling cancers, which are defined by functionally-significant somatic mutations in one or more elements of the cAMP signaling pathway, have an unexpectedly wide range of cell origins, clinical manifestations, and potential therapeutic options. Mutations in at least 9 cAMP signaling pathway genes (TSHR, GPR101, GNAS, PDE8B, PDE11A, PRKARA1, PRKACA, PRKACB, and CREB) have been identified as driver mutations in human cancer. Although all cAMP-signaling pathway cancers are driven by mutation(s) that impinge on a single signaling pathway, the ultimate tumor phenotype reflects interactions between five critical variables: (1) the precise gene(s) that undergo mutation in each specific tumor type; (2) the effects of specific allele(s) in any given gene; (3) mutations in modifier genes (mutational "context"); (4) the tissue-specific expression of various cAMP signaling pathway elements in the tumor stem cell; and (5) and the precise biochemical regulation of the pathway components in tumor cells. These varying oncogenic mechanisms reveal novel and important targets for drug discovery. There is considerable diversity in the "druggability" of cAMP-signaling components, with some elements (GPCRs, cAMP-specific phosphodiesterases and kinases) appearing to be prime drug candidates, while other elements (transcription factors, protein-protein interactions) are currently refractory to robust drug-development efforts. Further refinement of the precise driver mutations in individual tumors will be essential for directing priorities in drug discovery efforts that target these mutations.

MicroRNAs as Potential Biomarkers in Pituitary Adenomas

Pituitary adenomas (PAs) are one of the most common lesions of intracranial neoplasms, occurring in approximately 15% of the general population. They are typically benign, although some adenomas show aggressive behavior, exhibiting rapid growth, drug resistance, and invasion of surrounding tissues. Despite ongoing improvements in diagnostic and therapeutic strategies, late first diagnosis is common, and patients with PAs are prone to relapse. Therefore, earlier diagnosis and prevention of recurrence are of importance to improve patient care. MicroRNAs (miRNAs) are short non-coding single stranded RNAs that regulate gene expression at the post-transcriptional level. An increasing number of studies indicate that a deregulation of their expression patterns is related with pituitary tumorigenesis, suggesting that these small molecules could play a critical role in contributing to tumorigenesis and the onset of these tumors by acting either as oncosuppressors or as oncogenes, depending on the biological context. This paper provides an overview of miRNAs involved in PA tumorigenesis, which might serve as novel potential diagnostic and prognostic non-invasive biomarkers, and for the future development of miRNA-based therapeutic strategies for PAs.

Dual inhibition of cMET and EGFR by microRNA-338-5p suppresses metastasis of esophageal squamous cell carcinoma

MicroRNAs, as a group of post-transcriptional regulators, regulate multiple pathological processes including metastasis during tumor development. Here, we demonstrated the metastasis-suppressive function of microRNA (miR)-338-5p in esophageal squamous cell carcinoma (ESCC). Overexpression of miR-338-5p had inhibitory effect on invasive ability of ESCC cells and extracellular matrix degradation, whereas silencing miR-338-5p had opposite effects. Mechanistically, miR-338-5p directly targeted the 3' untranslated regions of hepatocellular growth factor receptor cMet (cMET) and epidermal growth factor receptor (EGFR). As a result, miR-338-5p inhibited the downstream signaling cascades of cMET and EGFR and repressed cMET- and EGFR-mediated ESCC cell invasion. Re-expression of cMET or EGFR in miR-338-5p overexpressing ESCC cells was sufficient to derepress ESCC cell invasion both in vitro and in vivo. We further showed that such manipulation downregulated the expression and secretion of matrix metalloproteinases 2 and 9, which resulted in impaired extracellular matrix degradation and cell invasion. Most importantly, systemic delivery of miR-338-5p mimic significantly inhibited metastasis of ESCC cells in nude mice. Taken together, our results uncovered a previously unknown mechanism through which miR-338-5p suppresses ESCC invasion and metastasis by regulating cMET/EGFR-matrix metalloproteinase 2/9 axis and highlighted the potential significance of miR-338-5p-based therapy in treating patients with metastatic ESCC.

Essential Role of the 14q32 Encoded miRNAs in Endocrine Tumors

BACKGROUND

The 14q32 cluster is among the largest polycistronic miRNA clusters. miRNAs encoded here have been implicated in tumorigenesis of multiple organs including endocrine glands.

METHODS

Critical review of miRNA studies performed in endocrine tumors have been performed. The potential relevance of 14q32 miRNAs through investigating their targets, and integrating the knowledge provided by literature data and bioinformatics predictions have been indicated.

RESULTS

Pituitary adenoma, papillary thyroid cancer and a particular subset of pheochromocytoma and adrenocortical cancer are characterized by the downregulation of miRNAs encoded by the 14q32 cluster. Pancreas neuroendocrine tumors, most of the adrenocortical cancer and medullary thyroid cancer are particularly distinct, as 14q32 miRNAs were overexpressed. In pheochromocytoma and growth-hormone producing pituitary adenoma, however, both increased and decreased expression of 14q32 miRNAs cluster members were observed. In the background of this phenomenon methodological, technical and biological factors are hypothesized and discussed. The functions of 14q32 miRNAs were also revealed by bioinformatics and literature data mining.

CONCLUSIONS

14q32 miRNAs have a significant role in the tumorigenesis of endocrine organs. Regarding their stable expression in the circulation of healthy individuals, further investigation of 14q32 miRNAs could provide a potential for use as biomarkers (diagnostic or prognostic) in endocrine neoplasms.

Large Scale Molecular Studies of Pituitary Neuroendocrine Tumors: Novel Markers, Mechanisms and Translational Perspectives

Simple Summary

Pituitary neuroendocrine tumors are non-cancerous tumors of the pituitary gland, that may overproduce hormones leading to serious health conditions or due to tumor size cause chronic headache, vertigo or visual impairment. In recent years pituitary neuroendocrine tumors are studied with the latest molecular biology methods that simultaneously investigate a large number of factors to understand the mechanisms of how these tumors develop and how they could be diagnosed or treated. In this review article, we have studied literature reports, compiled information and described molecular factors that could affect the development and clinical characteristics of pituitary neuroendocrine tumors, discovered factors that overlap between several studies using large scale molecular analysis and interpreted the potential involvement of these factors in pituitary tumor development. Overall, this study provides a valuable resource for understanding the biology of pituitary neuroendocrine tumors.

Abstract

Pituitary neuroendocrine tumors (PitNETs) are non-metastatic neoplasms of the pituitary, which overproduce hormones leading to systemic disorders, or tumor mass effects causing headaches, vertigo or visual impairment. Recently, PitNETs have been investigated in large scale (exome and genome) molecular analyses (transcriptome microarrays and sequencing), to uncover novel markers. We performed a literature analysis on these studies to summarize the research data and extrapolate overlapping gene candidates, biomarkers, and molecular mechanisms. We observed a tendency in samples with driver mutations (GNAS, USP8) to have a smaller overall mutational rate, suggesting driver-promoted tumorigenesis, potentially changing transcriptome profiles in tumors. However, direct links from drivers to signaling pathways altered in PitNETs (Notch, Wnt, TGF-β, and cell cycle regulators) require further investigation. Modern technologies have also identified circulating nucleic acids, and pinpointed these as novel PitNET markers, i.e., miR-143-3p, miR-16-5p, miR-145-5p, and let-7g-5p, therefore these molecules must be investigated in the future translational studies. Overall, large-scale molecular studies have provided key insight into the molecular mechanisms behind PitNET pathogenesis, highlighting previously reported molecular markers, bringing new candidates into the research field, and reapplying traditional perspectives to newly discovered molecular mechanisms.

Circulating microRNA Expression in Cushing's Syndrome

Context

Cushing's syndrome (CS) is a rare disease of endogenous hypercortisolism associated with high morbidity and mortality. Diagnosis and classification of CS is still challenging.

Objective

Circulating microRNAs (miRNAs) are minimally invasive diagnostic markers. Our aim was to characterize the circulating miRNA profiles of CS patients and to identify distinct profiles between the two major CS subtypes.

Methods

We included three groups of patients from the German Cushing's registry: ACTH-independent CS (Cortisol-Producing-Adenoma; CPA), ACTH-dependent pituitary CS (Cushing's Disease; CD), and patients in whom CS had been ruled out (controls). Profiling of miRNAs was performed by next-generation-sequencing (NGS) in serum samples of 15 CS patients (each before and after curative surgery) and 10 controls. Significant miRNAs were first validated by qPCR in the discovery cohort and then in an independent validation cohort of 20 CS patients and 11 controls.

Results

NGS identified 411 circulating miRNAs. Differential expression of 14 miRNAs were found in the pre- and postoperative groups. qPCR in the discovery cohort validated 5 of the significant miRNAs from the preoperative group analyses. Only, miR-182-5p was found to be significantly upregulated in the CD group of the validation cohort. Comparing all CS samples as a group with the controls did not reveal any significant differences in expression.

Outcome

In conclusion, our study identified miR-182-5p as a possible biomarker for CD, which has to be validated in a prospective cohort. Furthermore, our results suggest that presence or absence of ACTH might be at least as relevant for miRNA expression as hypercortisolism itself.

MicroRNA-7a2 Regulates Prolactin in Developing Lactotrophs and Prolactinoma Cells

Prolactin production is controlled by a complex and temporally dynamic network of factors. Despite this tightly coordinated system, pathological hyperprolactinemia is a common endocrine disorder that is often not understood, thereby highlighting the need to expand our molecular understanding of lactotroph cell regulation. MicroRNA-7 (miR-7) is the most highly expressed miRNA family in the pituitary gland and the loss of the miR-7 family member, miR-7a2, is sufficient to reduce prolactin gene expression in mice. Here, we used conditional loss-of-function and gain-of-function mouse models to characterize the function of miR-7a2 in lactotroph cells. We found that pituitary miR-7a2 expression undergoes developmental and sex hormone-dependent regulation. Unexpectedly, the loss of mir-7a2 induces a premature increase in prolactin expression and lactotroph abundance during embryonic development, followed by a gradual loss of prolactin into adulthood. On the other hand, lactotroph development is delayed in mice overexpressing miR-7a2. This regulation of lactotroph function by miR-7a2 involves complementary mechanisms in multiple cell populations. In mouse pituitary and rat prolactinoma cells, miR-7a2 represses its target Raf1, which promotes prolactin gene expression. These findings shed light on the complex regulation of prolactin production and may have implications for the physiological and pathological mechanisms underlying hyperprolactinemia.

GH and IGF System: The Regulatory Role of miRNAs and lncRNAs in Cancer

Growth hormone (GH) and the insulin-like growth factor (IGF) system are involved in many biological processes and have growth-promoting actions regulating cell proliferation, differentiation, apoptosis and angiogenesis. A recent chapter in epigenetics is represented by microRNAs (miRNAs) and long non-coding RNAs (lncRNAs) which regulate gene expression. Dysregulated miRNAs and lncRNAs have been associated with several diseases including cancer. Herein we report the most recent findings concerning miRNAs and lncRNAs regulating GH and the IGF system in the context of pituitary adenomas, osteosarcoma and colorectal cancer, shedding light on new possible therapeutic targets. Pituitary adenomas are increasingly common intracranial tumors and somatotroph adenomas determine supra-physiological GH secretion and cause acromegaly. Osteosarcoma is the most frequent bone tumor in children and adolescents and was reported in adults who were treated with GH in childhood. Colorectal cancer is the third cancer in the world and has a higher prevalence in acromegalic patients.

Novel Insights into Pituitary Tumorigenesis: Genetic and Epigenetic Mechanisms

Substantial advances have been made recently in the pathobiology of pituitary tumors. Similar to many other endocrine tumors, over the last few years we have recognized the role of germline and somatic mutations in a number of syndromic or nonsyndromic conditions with pituitary tumor predisposition. These include the identification of novel germline variants in patients with familial or simplex pituitary tumors and establishment of novel somatic variants identified through next generation sequencing. Advanced techniques have allowed the exploration of epigenetic mechanisms mediated through DNA methylation, histone modifications and noncoding RNAs, such as microRNA, long noncoding RNAs and circular RNAs. These mechanisms can influence tumor formation, growth, and invasion. While genetic and epigenetic mechanisms often disrupt similar pathways, such as cell cycle regulation, in pituitary tumors there is little overlap between genes altered by germline, somatic, and epigenetic mechanisms. The interplay between these complex mechanisms driving tumorigenesis are best studied in the emerging multiomics studies. Here, we summarize insights from the recent developments in the regulation of pituitary tumorigenesis.

Multiple Endocrine Neoplasia Type 1: The Potential Role of microRNAs in the Management of the Syndrome

Multiple endocrine neoplasia type 1 (MEN1) is a rare inherited tumor syndrome, characterized by the development of multiple neuroendocrine tumors (NETs) in a single patient. Major manifestations include primary hyperparathyroidism, gastro-entero-pancreatic neuroendocrine tumors, and pituitary adenomas. In addition to these main NETs, various combinations of more than 20 endocrine and non-endocrine tumors have been described in MEN1 patients. Despite advances in diagnostic techniques and treatment options, which are generally similar to those of sporadic tumors, patients with MEN1 have a poor life expectancy, and the need for targeted therapies is strongly felt. MEN1 is caused by germline heterozygous inactivating mutations of the MEN1 gene, which encodes menin, a tumor suppressor protein. The lack of a direct genotype–phenotype correlation does not permit the determination of the exact clinical course of the syndrome. One of the possible causes of this lack of association could be ascribed to epigenetic factors, including microRNAs (miRNAs), single-stranded non-coding small RNAs that negatively regulate post-transcriptional gene expression. Some miRNAs, and their deregulation, have been associated with MEN1 tumorigenesis. Recently, an extracellular class of miRNAs has also been identified (c-miRNAs); variations in their levels showed association with various human diseases, including tumors. The aim of this review is to provide a general overview on the involvement of miRNAs in MEN1 tumor development, to be used as possible targets for novel molecular therapies. The potential role of c-miRNAs as future non-invasive diagnostic and prognostic biomarkers of MEN1 will be discussed as well.

MicroRNAs expression in pituitary tumors: differences related to functional status, pathological features, and clinical behavior

BACKGROUND

MicroRNAs (miRNAs) are small non-coding RNA molecules that regulate gene expression at post-transcriptional level, having a role in many biological processes, such as control of cell proliferation, cell cycle, and cell death. Altered miRNA expression has been reported in many neoplasms, including pituitary adenomas (PAs).

PURPOSE

In this study, we aimed to evaluate the expression of 20 miRNAs involved in pathways relevant to pituitary pathophysiology, in PAs and normal pituitary tissue and to correlate their expression profile with clinical and pathological features.

METHODS

Pituitary tumor samples were obtained during transphenoidal surgery from patients with non-functioning (NFPA, n = 12) and functioning (n = 11, 5 GH-, 3 ACTH-, 3 PRL-omas) PAs. The expression of selected miRNAs in PAs and in normal pituitary was analyzed by RT-qPCR. miRNAs expression was correlated with demographic, clinical, and neuroradiological data and with histopathological features including pituitary hormones immunostaining, Ki-67 proliferation index, and p53 immunohistochemistry evaluation.

RESULTS

All evaluated miRNAs except miR-711 were expressed in both normal and tumor pituitary tissue. Seventeen miRNAs were significantly down-regulated in pituitary tumors compared to normal pituitary. miRNAs were differentially expressed in functioning PAs or in NFPAs, as in the latter group miR-149-3p (p = 0.036), miR-130a-3p (p = 0.014), and miR-370-3p (p = 0.026) were significantly under expressed as compared to functioning tumors. Point-biserial correlation analysis demonstrated a negative correlation between miR-26b-5p and Ki-67 (p = 0.031) and between miR-30a-5p and 'atypical' morphological features (p = 0.038) or cavernous sinus invasion (p = 0.049), while 508-5p was inversely correlated with clinical aggressiveness (p = 0.043).

CONCLUSIONS

In this study, we found a significant down-regulation of 17 miRNAs in PAs vs normal pituitary, with differential expression profile related to functional status and tumor aggressiveness.

DNA Methylation Influences miRNA Expression in Gonadotroph Pituitary Tumors

microRNAs are involved in pathogenesis of cancer. DNA methylation plays a role in transcription of miRNA-encoding genes and may contribute to changed miRNA expression in tumors. This issue was not investigated in pituitary neuroendocrine tumors (PitNETs) previously. DNA methylation patterns, assessed with HumanMethylation450K arrays in 34 PitNETs and five normal pituitaries, were used to determine differentially methylated CpGs located at miRNA genes. It showed aberrant methylation in regions encoding for 131 miRNAs. DNA methylation data and matched miRNA expression profiles, determined with next-generation sequencing (NGS) of small RNAs, were correlated in 15 PitNETs. This showed relationship between methylation and expression levels for 12 miRNAs. DNA methylation and expression levels of three of them (MIR145, MIR21, and MIR184) were determined in the independent group of 80 tumors with pyrosequencing and qRT-PCR and results confirmed both aberrant methylation in PitNETs and correlation between methylation and expression. Additionally, in silico target prediction was combined with analysis of established miRNA profiles and matched mRNA expression pattern, assessed with amplicon-based NGS to indicate putative target genes of epigenetically deregulated miRNAs. This study reveals aberrant DNA methylation in miRNA-encoding genes in gonadotroph PitNETs. Methylation changes affect expression level of miRNAs that regulate putative target genes with tumorigenesis-relevant functions.

The Search of miRNA Related to Invasive Growth of Nonfunctioning Gonadotropic Pituitary Tumors

PURPOSE

Nonfunctioning gonadotropic pituitary neuroendocrine tumors (PitNETs) are among the most frequent neoplasms of pituitary gland. Although PitNETs are commonly considered benign, a notable part of patients suffer from tumor recurrence after treatment. Invasive growth of pituitary tumor is among the most important prognostic factors. Since molecular features of invasiveness are of potential clinical usefulness, this study was aimed to verify whether invasive and noninvasive nonfunctioning gonadotropic PitNETs differ in the miRNA expression profile and whether the differences could provide a possible molecular classifier.

METHODS

miRNA profiles were determined in 20 patients (11 invasive and 9 noninvasive tumors) using next-generation sequencing. The expression of selected miRNAs was assessed in the independent cohort of 80 patients with qRT-PCR.

RESULTS

When miRNA profiles of invasive and noninvasive tumors were compared, 29 miRNAs were found differentially expressed. Hsa-miR-184, hsa-miR-181a-2-3p, hsa-miR-93-3p, hsa-miR-574-5p, hsa-miR-185-5p, and hsa-miR-3200-5p showed a potential clinical value according to ROC curve analysis. Unfortunately, differential expression of only hsa-miR-185-5p was confirmed in the validation cohort, with AUG at 0.654.

CONCLUSION

Differences in miRNAs expression profiles in invasive and noninvasive gonadotropic PitNETs are slight and the level of miRNA expression seems not to be applicable as useful classifier of tumor invasiveness.