The Genomic Landscape of Corticotroph Tumors: From Silent Adenomas to ACTH-Secreting Carcinomas

Posted on May 21, 2022 by MaryO

Abstract

Corticotroph cells give rise to aggressive and rare pituitary neoplasms comprising ACTH-producing adenomas resulting in Cushing disease (CD), clinically silent ACTH adenomas (SCA), Crooke cell adenomas (CCA) and ACTH-producing carcinomas (CA). The molecular pathogenesis of these tumors is still poorly understood. To better understand the genomic landscape of all the lesions of the corticotroph lineage, we sequenced the whole exome of three SCA, one CCA, four ACTH-secreting PA causing CD, one corticotrophinoma occurring in a CD patient who developed Nelson syndrome after adrenalectomy and one patient with an ACTH-producing CA. The ACTH-producing CA was the lesion with the highest number of single nucleotide variants (SNV) in genes such as USP8, TP53, AURKA, EGFR, HSD3B1 and CDKN1A. The USP8 variant was found only in the ACTH-CA and in the corticotrophinoma occurring in a patient with Nelson syndrome. In CCA, SNV in TP53, EGFR, HSD3B1 and CDKN1A SNV were present. HSD3B1 and CDKN1A SNVs were present in all three SCA, whereas in two of these tumors SNV in TP53, AURKA and EGFR were found. None of the analyzed tumors showed SNV in USP48, BRAF, BRG1 or CABLES1. The amplification of 17q12 was found in all tumors, except for the ACTH-producing carcinoma. The four clinically functioning ACTH adenomas and the ACTH-CA shared the amplification of 10q11.22 and showed more copy-number variation (CNV) gains and single-nucleotide variations than the nonfunctioning tumors.

1. Introduction

The pathological spectrum of the corticotroph includes ACTH (adrenocorticotropic hormone)-secreting pituitary adenomas (PA), causing Cushing disease (CD), silent corticotroph adenomas (SCA), Crooke cell adenomas (CCA) and the rare ACTH-secreting carcinoma (ACTH-CA). Pituitary carcinomas account for 0.1 to 0.2% of all pituitary tumors and are defined by the presence of craniospinal or distant metastasis [1,2,3]. Most pituitary carcinomas are of corticotroph or lactotrope differentiation [3]. Although a few cases present initially as CA, the majority develop over the course of several months or years from apparently benign lesions [3,4]. CCA are characterized by the presence of hyaline material in more than 50% of the cells of the lesion, and most of them arise from silent corticotroph adenomas (SCA) or CD-provoking ACTH-secreting adenomas [5]. SCA are pituitary tumors with positive immunostaining for ACTH but are not associated with clinical or biochemical evidence of cortisol excess; they are frequently invasive lesions and represent up to 19% of clinically non-functioning pituitary adenomas (NFPA) [6]. ACTH-secreting PA represents up to 6% of all pituitary tumors and causes eloquent Cushing disease (CD), which is characterized by symptoms and signs of cortisol hypersecretion, including a two- to fivefold increase in mortality [7,8]. The 2017 World Health Organization (WHO) classification of PA considers not only the hormones these tumors synthesize but also the transcription factors that determine their cell lineage [9]. TBX19 is the transcription factor responsible for the terminal differentiation of corticotrophs [9]. All tumor lesions of corticotroph differentiation are positive for both ACTH and TBX19.
ACTH-secreting PA causing CD are among the best genetically characterized pituitary tumors, with USP8 somatic variants occurring in up to 25–35% of sporadic cases [9]. Yet, information regarding the molecular pathogenesis of the lesions conforming to the whole pathological spectrum of the corticotroph is scarce. The aim of the present study is to characterize the genomic landscape of pituitary tumors of corticotroph lineage. For this purpose, we performed whole exome sequencing to uncover the mutational burden (single-nucleotide variants, SNV) and copy-number variations (CNVs) of these lesions.

2. Results

2.1. Clinical and Demographic Characteristics of the Patients

A total of 10 tumor samples from 10 patients were evaluated: 4 ACTH-secreting adenomas causing clinically evident CD, three non-functioning adenomas that proved to be SCA upon immunohistochemistry (IHC), one ACTH-secreting CA with a prepontine metastasis, one rapidly growing ACTH-secreting adenoma after bilateral adrenalectomy (Nelson syndrome) in a patient with CD and one non-functioning, ACTH-producing CCA (Table 1). All except one patient were female; the mean age was 38.8 ± 16.5 years (range 17–61) (Table 1). They all harbored macroadenomas with a mean maximum diameter of 31.9 ± 13 mm (range 18–51). Cavernous sinus invasion was evident on MRI in all but one of the patients (Table 1). Homonymous hemianopia was present in seven patients, whereas right optic nerve atrophy and amaurosis were evident in patient with the ACTH-CA, and in patient with CD and pituitary apoplexy (Table 1). Detailed clinical data are included in Supplementary Table S1. Death was documented in only the patient with pituitary apoplexy, and one patient was lost during follow-up, as of October 2018.
Table 1. Clinical features of the tumors analyzed and SNV present in each tumor.
Table

2.2. General Genomic Characteristics of Neoplasms of Corticotrophic Lineage

Overall, approximately 18,000 variants were found, including missense, nonsense and splice-site variants as well as frameshift insertions and deletions. Of these alterations, the majority corresponded to single-nucleotide variants, followed by insertions and deletions. The three most common base changes were transitions C > T, T > C and C > G; most of the genetic changes were base transitions rather than transversions (Figure 1). There were several genes across the whole genome affected in more than one way, meaning that the same gene presented missense and nonsense variants, insertions, deletions and splice-site variants (Figure 2). Many of these variants are of unknown pathogenicity and require further investigation. Gains in genetic material were found in 44 cytogenetic regions, whereas 72 cytogenetic regions showed loss of genetic material in all corticotroph tumors.
Ijms 23 04861 g001 550
Figure 1. Panel (A) shows the gadolinium-enhanced magnetic resonance imaging of the patient with ACTH-CA, highlighting in red the metastatic lesion in the prepontine area. Panel (B) shows the hematoxylin and eosin staining displaying the hyaline structures in the perinuclear areas denoting a Crooke cell adenoma. Panel (C,D) depict a representative corticotroph tumor with positive ACTH and TBX19 immunohistochemistry, respectively. Panel (E) shows four graphics: variant classification, variant type, SNV class and transition (ti) or transversion (tv) describing the general results of exome sequencing of the corticotroph tumors.
Ijms 23 04861 g002 550
Figure 2. Representative rainfall plots showing the SNV alterations throughout the whole genome of corticotroph tumors (A) CCA, (B) SCA, (C) CD and (D) ACTH-CA, displaying all base changes, including transversions and transitions. No kataegis events were found. Alterations across the genome were seen in all corticotroph tumors.

2.3. ACTH-Secreting Carcinoma (Tumor 1)

SNV missense variants were found in the genes encoding TP53 (c.215G > C [rs1042522], p.Pro72Arg); AURKA (c.91T > A [rs2273535], p.Phe31Ile); EGFR (epidermal growth factor receptor, c.1562G > A [rs2227983], p.Arg521Lys); HSD3B1 (3-ß-hydroxisteroid dehydrogenase, c.1100C > A [rs1047303], p.Thr367Asn); CDKN1A (cyclin-dependent kinase inhibitor 1A or p21, c.93C > A [rs1801270], p.Ser31Arg); and USP8 (c.2159C > G [rs672601311], p.Pro720Arg). Interestingly, the previously reported USP48, BRAF, BRG1 and CABLES1 variants in pituitary CA cases were not found in this patient’s tumor (Figure 3). All SNV detected in WES experiments were validated by Sanger sequencing. The variants described were selected due to their potential pathogenic participation in other tumors and the allelic-risk association with tumorigenesis. Hereafter, all the mentioned variants in other corticotroph tumors are referred to by these aforementioned variants. Even though these same genes presented other variants, currently the significance of those variants is unknown.
Ijms 23 04861 g003 550
Figure 3. Panel (A) shows the oncoplot from the missense variants of the selected genes and their clinical–pathological features. Panels (B–G) depict USP8, EGFR, TP53, AURKA, CDKN1A and HSD3B1 proteins, respectively, with the changes found in DNA impacting aminoacidic changes.
In general, the pituitary CA presented more CNV alterations than the benign tumors, with 27 and 32 cytogenetic regions showing gains and losses of genetic material, respectively. The cytogenetic regions showing gains were 10q11.22, 15q11.2, 16p12.3, 1p13.2 and 20p, where genes SYT15, POTEB, ARL6IP1, HIPK1 and CJD6 are coded, respectively. By contrast, 8p21.2 was the cytogenetic region showing loss of genetic material. The previously reported amplification of 1p13.2 was also detected in this tumor (Figure 4) [10].
Ijms 23 04861 g004 550
Figure 4. Hierarchical clustering of corticotroph tumors according to their gains and losses across the whole genome (somatic chromosomes only). High contrast was used to enhance potential CNV alterations; nevertheless, there were only 44 unique cytogenetic regions that showed gains in genetic material with statistical significance, whereas only 72 unique cytogenetic regions showed loss of genetic material with statistical significance.

2.4. Crooke Cell Adenoma (Tumor 2)

The CCA showed SNV in the genes encoding TP53, EGFR, HSD3B1 and CDKN1A. However, neither the genes encoding AURKA and USP8 nor those encoding USP48, BRAF, BRG1 and CABLES were affected in this tumor. In CCA, only two and fifteen gains and losses were observed in copy-number variation, respectively. CNVs only showed gains in cytogenetic regions 17q12 and 10q11.22, harboring genes CCL3L1 and NPY4R, respectively, whereas losses were found in cytogenetic regions 18q21.1, 15q12 and 2q11.2, harboring genes KATNAL2, TUBGCP5 and ANKRD36.

2.5. Silent Corticotroph Adenomas (Tumors 3–5)

The three SCA shared SNVs in the genes encoding HSD3B1 and CDKN1A. SCA 4 and 5 showed SNV in the genes encoding EGFR, whereas SNV in the genes encoding AURKA and TP53 were present in SCA 3 and 5. None of the SCA were found to have SNV in the genes encoding USP8, USP48, BRAF, BRG1 or CABLES1.
The SCA presented only two and eighteen gains and losses (CNV), respectively. In regard to CNV, the these clinically silent tumors presented gains of genetic material in cytogenetic regions 17q22 and 10q11.22, which harbor genes encoding CCL3L1 and NPY4R. Eighteen losses were found distributed in cytogenetic regions 18q21.1, 15q12 and 2q11.2, encompassing the genes encoding KATNAL2, TUBGCP5 and ANKRD36. This CNV pattern closely resembles the one found in the CCA, which is somewhat expected if we consider that both neoplasms are clinically non-functioning

2.6. ACTH-Secreting Adenomas (Cushing Disease) (Tumors 6–9)

SNV of the genes encoding TP53 and HSD3B1 were present in tumor samples from all four CD patients, whereas none of these patients harbored adenomas with SNV in the genes encoding USP8 or CDKN1A. An SNV in the gene encoding AURKA was identified in only one of these tumors (tumor 8). EGFR SNV were found in tumors 7 and 9. None of the CD-causing ACTH-secreting adenomas showed the previously reported SNV in the genes encoding USP48, BRAF, BRG1 and CABLES1.
CNV analysis in this group of eloquent-area corticotroph tumors revealed 25 gains and 55 losses of genetic material. The gains occurred in cytogenetic regions 17q12, 2p12, 9p24 and 10q11.22, where genes CCL3L1, CTNNA2, FOXD4 and NPY4R are coded, respectively. The losses were localized in cytogenetic regions 21p12, 15q11.2, and 8p23, harboring genes USP16, KLF13 and DEF130A, respectively. We also detected the previously reported 20p13 amplification [10].

2.7. ACTH-Secreting Adenoma Causing Nelson Syndrome (Tumor 10)

This patient’s tumor showed SNV in the genes encoding USP8, TP53, HSD3B1 and CDKN1A but no alterations were found in the genes encoding EGFR and AURKA. This tumor and the ACTH-CA were the only two neoplasms that harbored a USP8 variant. No SNV were identified in the genes encoding USP48, BRAF, BRG1 and CABLES1. Interestingly, CNV analysis revealed the same gains and losses of genetic material found in tumors from other patients with CD.

2.8. Tumor Phylogenic Analysis

We performed a phylogenetic inference analysis to unravel a hypothetical sequential step transformation from an SCA to a functioning ACTH-secreting adenoma and finally to an ACTH-CA. The theoretical evolutive development of the ACTH CA, departing from the SCA, shows two main clades, with the smallest one comprising two of the three SCA and two of the five ACTH-adenomas causing CD. Since these four tumors have the same SNV profile, we can assume that they harbor the genes that must be altered to make possible the transition from a silent to a clinically eloquent adenoma; the gene encoding ATF7IP (c.1589A > G [rs3213764], p.K529R) characterizes this clade. The second and largest clade includes the CCA, the ACTH-CA, one of the three SCA and three of the five most aggressive ACTH adenomas causing CD, including the adenoma of the patient with Nelson syndrome. This clade represents the molecular alterations required to evolve from a CD-causing ACTH-adenoma to a more aggressive tumor, or even to a CA and is characterized by the gene encoding MSH3 (c.235A > G [rs1650697], p.I79V) (Figure 5).
Ijms 23 04861 g005 550
Figure 5. Phylogenetic analysis of the corticotroph tumors. The theoretical evolutive development of the ACTH-CA, departing from the SCA shows two main clades. The first clade, characterized by ATF7IP gene, comprises 2 of the 3 SCA and 2 of the 5 ACTH-adenomas causing CD. The second clade is characterized by the gene encoding MSH3 and includes the CCA, the ACTH-CA, one of the 3 SCA and 3 of the 5 most aggressive ACTH adenomas causing CD, including the adenoma of the patient with Nelson syndrome. Red dots represent the Cushing Disease provoking adenomas, green dots represent the silent corticotroph tumors, brown dot represent the Crooke cell adenoma and the blue dot represent the corticotroph carcinoma.

2.9. Correlation between Gene Variants and Clinicopathological Features

The USP8 variant positively correlated with increased tumor mass (p = 0.019). The CDKN1A variant was significantly associated with silent tumors (p = 0.036). The rest of the genetic variants did not correlate with any of the clinicopathological features tested. The presence of the EGFR variant was not distinctly associated with any of the clinical parameters and was equally present in functional as well as non-functional tumors (p = 0.392). AURKA SNV did not correlate with any of the features, including recurrence (p = 0.524). Detailed statistical results are presented in Supplementary Table S2.

3. Discussion

Corticotrophs are highly specialized cells of the anterior pituitary that synthesize and secrete hormones that are essential for the maintenance of homeostasis. In this study, we sequenced the exome of 10 corticotroph tumors, including three SCA, four ACTH adenomas causing CD, an ACTH adenoma in a patient with Nelson syndrome, a CCA and an ACTH-CA in total, representing the broad pathological spectrum of this cell. Our results portray the genomic landscape of all the neoplasms that are known to affect the corticotroph.
The neoplasm with the highest number of genomic abnormalities, including SNV and CNV, was the ACTH-CA, followed by the CCA and the CD tissues. Of all the genes harboring SNVs, six were found to be present in at least two of our tumor samples: HSD3B1, TP53, CDKN1A, EGFR, AURKA and USP8.
The HSD3B1 gene encodes a rate-limiting enzyme required for all pathways of dihydrotestosterone synthesis and is abundantly expressed in adrenal tumors. Gain of function of this HSD3B1 variant, which has a global allelic prevalence of 0.69678 [11], results in resistance to proteasomal degradation with the consequent accumulation of the enzyme and has been associated with a poor prognosis in patients with prostate cancer [12]. Nine of the ten corticotroph tumors in our cohort harbored an SNV of the tumor suppressor gene TP53. The TP53 variant described in our cohort has been reported to be present in 80% of non-functioning pituitary adenomas and is apparently associated with a younger age at presentation and with cavernous sinus invasion [13]. Furthermore, this TP53 variant results in a reduced expression of CDKN1A and an increased expression of vascular endothelial growth factor (VEGF) as well as an increased cellular proliferation rate [13]. CDKN1A (also known as p21) is a cyclin-dependent kinase inhibitor regulating cell cycle progression. The SNV described in our study was reported to alter DNA binding ability and expression and has a global allelic frequency of 0.086945 [14]. This cyclin-dependent kinase inhibitor SNV was found to be associated with breast carcinoma [15] and lung cancer [16]. The presence of this SNV has not been previously explored in pituitary adenomas, although CDKN1A is downregulated in clinically non-functioning pituitary adenomas of gonadotrophic lineage but not in hormone-secreting tumors [17]. EGFR encodes a transmembrane tyrosine kinase receptor, activation of which leads to mitogenic signaling [18]. This gene is upregulated in several cancers and represents a target for molecular therapies [19]. The EGFR SNV described in our corticotroph tumor series was found to be associated with the response to neoadjuvant chemotherapy in patients with breast and lung cancer [18]. EGFR is normally expressed in corticotrophs, where it participates in the regulation of POMC (proopiomelanocortin) gene transcription and cellular proliferation [20]. The EGFR rs2227983 has a 0.264334 global allelic frequency [21]. AURKA is a cell-cycle regulatory serine/threonine kinase that promotes cell cycle progression by the establishment of the mitotic spindle and centrosome separation [22]. Alterations of these gene are related to centrosomal amplification, dysfunction of cytokinesis and aneuploidy [22]; it has a global allelic frequency of 0.18078 [23]. This same SNV has been associated with overall cancer risk, particularly breast, gastric, colorectal, liver and endometrial carcinomas, but it has never been formally studied in pituitary tumors [22]. Activating somatic variants of the gene encoding USP8 were recently found in 25–40% of ACTH-secreting adenomas causing CD [24,25]. Patients harboring these variants are usually younger, more frequently females and were found to have higher long-term recurrence rates in some but not all studies [26,27]. USP8 mediates the deubiquitination of EGFR by inhibiting its interaction with protein 14-3-3, which in turn prevents its proteosomal degradation. Signaling through the recycled deubiquitinated EGFR is increased, leading to increased POMC transcription and cellular proliferation. Most activating USP8 variants are located within its 14-3-3 binding motif [24,25]. Recently, USP8 and TP53 SNV were described in corticotroph tumors as drivers of aggressive lesions [28]. To our knowledge, USP8 variants have not been evaluated in patients with pituitary carcinomas, and none of the previously mentioned studies have included patients with Nelson syndrome. In our cohort, neither the CCA nor the SCA showed variants in USP8, in concordance with previously published studies [25,29], or in the genes USP48, BRAF, BRG1 and CABLES1 [9], and none of them were present in our cohort.
Genetic structural variations in the human genome can be present in many forms, from SNV to large chromosomal aberrance [30]. CNV are structurally variant regions, including unbalanced deletions, duplications and amplifications of DNA segments ranging from a dozen to several hundred base pairs, in which copy-number differences have been observed between two or more genomes [31,32]. CNV are involved in the development and progression of many tumors and occur frequently in PA [30,33]. Hormone-secreting pituitary tumors show more CNV than non-functioning tumors [34]. Accordingly, our non-functioning SCA and CCA had considerably fewer chromosomal gains and losses than the CD-causing adenomas and the ACTH-CA. Expectedly, the ACTH-CA had significantly more cytogenetic abnormalities than any other tumor in our series. Interestingly, the ACTH-adenomas causing CD, the SCA and the CCA shared the gain of genetic material in 17q12, highlighting their benign nature. The 17q12 amplification has been described in gastric neoplasms [35]. The only cytogenetic abnormality shared by all types of corticotroph tumors was the gain of genetic material in 10q11.22. Amplification of 10q11.22 was previously described in Li–Fraumeni cancer predisposition syndrome [36]. The ACTH-CA, the CCA and one SCA clustered together showing a related CNV pattern; this CNV profile could be reflective of the aggressive nature of these neoplasms, since both CCA and SCA can follow a clinically aggressive course [5,6].
Our results show that all lesions conforming to the pathological spectrum of the corticotroph share some of the SNV and CNV profiles. These genomic changes are consistent with the potential existence of a continuum, whereby silent tumors can transform into a clinically eloquent tumor and finally to carcinoma, or at least a more aggressive tumor. It can also be interpreted as the common SNV shared by aggressive tumors. It is known that silent corticotroph adenomas may switch into a hormone-secreting tumor [37] and are considered a marker for aggressiveness and a risk factor for malignancy since most of the carcinomas are derived from functioning hormone-secreting adenomas. Our phylogenetic inference analysis showed that the genes ATF7IP and MSH3 could participate in a tumor transition ending in aggressive entities or even carcinomas. ATF7IP is a multifunctional nuclear protein mediating heterochromatin formation and gene regulation in several contexts [38], while MSH3 is a mismatch-repair gene [39]. Events related to heterochromatin remodeling and maintenance have been related to aggressive pituitary adenomas and carcinomas [40]. Additionally, alterations in mismatch-repair genes are related to pituitary tumor aggressiveness and resistance to pharmacologic treatment [41,42]. The variants described in ATF7IP and MSH3 are related to prostate and colorectal cancer, respectively [43,44]. There is evidence suggesting that the ATF7IP variant could be deleterious because it leads to a negative regulation of transcription [45]. Thus, these events could be biologically relevant to corticotroph tumorigenesis, although more research is needed.

4. Conclusions

We have shown genomic evidence that within the tumoral spectrum of the corticotroph, functioning ACTH-secreting lesions harbor more SNV and CNV than non-functioning ACTH adenomas. The ACTH-secreting CA shows more genomic abnormalities than the other lesions, underscoring its more aggressive biological behavior. Phylogenetic inference analysis of our data reveals that silent corticotroph lesions may transform into functioning tumors, or at least potentially, into more aggressive lesions. Alterations in genes ATF7IP and MSH3, related to heterochromatin formation and mismatch repair, could be important in corticotroph tumorigenesis. The main drawback of our study is the limited sample size. We are currently increasing the number of samples to corroborate our findings and to be able to perform a more comprehensive complementary phylogenetic analysis of our data. Finally, further research is needed to uncover the roles of these variants in corticotroph tumorigenesis.

5. Materials and Methods

5.1. Patients and Tumor Tissue Samples

Ten pituitary tissues were collected: one ACTH-CA, one CCA, three SCA, and five ACTH-secreting PA causing CD, including the tumor of a patient who developed Nelson syndrome after bilateral adrenalectomy. All tumors included in the study were sporadic and were collected from patients diagnosed, treated and followed at the Endocrinology Service and the Neurosurgical department of Hospital de Especialidades, Centro Médico Nacional Siglo XXI of the Instituto Mexicano del Seguro Social, Hospital General de Mexico “Dr. Eduardo Liceaga” and Instituto Nacional de Neurologia y Neurocirugia “Manuel Velazquez”. All participating patients were recruited with signed informed consent and ethical approval from the Comisión Nacional de Ética e Investigación Científica of the Instituto Mexicano del Seguro Social, in accordance with the Helsinki declaration.
CD was diagnosed according to our standard protocol. Briefly, the presence of hypercortisolism was documented based on two screening tests, namely a 24 h urinary free-cortisol level above 130 µg and the lack of suppression of morning (7:00–8:00) cortisol after administration of 1 mg dexamethasone the night before (23:00) to less than 1.8 µg/dL, followed by a normal or elevated plasma ACTH to ascertain ACTH-dependence. Finally, an overnight, high-dose (8 mg) dexamethasone test, considered indicative of a pituitary source, and a cortisol suppression > 69%, provided that a pituitary adenoma was clearly present on magnetic resonance imaging (MRI) of the sellar region. In none of the 10 patients included in the study was inferior petrosal venous sampling necessary to confirm the pituitary origin of the ACTH excess. Invasiveness was defined by the presence of tumor within the cavernous sinuses (CS).
DNA was extracted from paraffin-embedded tumor tissues using the QIAamp DNA FFPE tissue kit. From frozen tumors, DNA was obtained using the Proteinase K-ammonium acetate protocol.

5.2. Construction and Sequencing of Whole Exome Libraries

Exome libraries were prepared according to the Agilent SureSelect XT HS Human All exon v7 instructions. Briefly, 200 ng of DNA was enzymatically fragmented with Agilent SureSelect Enzymatic Fragmentation Kit. Fragmented DNA was end-repaired and dA-tail was added at DNA ends; then, molecular barcode adaptors were added, followed by AMPure XP bead purification. The adaptor-ligated library was amplified by PCR and purified by AMPure XP beads. DNA libraries were hybridized with targeting exon probes and purified with streptavidin-coated magnetic beads. The retrieved libraries were amplified by PCR and purified by AMPure XP beads and pooled for sequencing in NextSeq 500 using Illumina flow cell High Output 300 cycles chemistry. All quality controls of the libraries were carried out using Screen tape assays and quantified by Qubit fluorometer. Quality parameters included a DNA integrity number above 8 and a 100X sequencing depth aimed with at least 85% of coverage.

5.3. Bioinformatics Analysis

The fastq files were subjected to quality control using FastQC v0.11.9, the adapters were removed using Cutadapt v3.4, the alignment was carried out with Burrows–Wheeler Alignment Tool v0.7.17 with the -M option to ensure compatibility with Picard and GRCh38 as a reference genome. The marking of duplicates as well as the sorting was carried out with Picard v2.26.4 with the AddOrReplaceReadGroups programs with the option SORT_ORDER = coordinate and MarkDuplicates, respectively. Variant calling was carried out using Genomic Analysis Toolkit (GATK) v4.2.2.0 following the Best Practices guide (available at https://gatk.broadinstitute.org/) [46] and with the parameters used by Genomic Data Commons (GDC), available at https://docs.gdc.cancer.gov/ [47]. The GATK tools used were CollectSequencingArtifactMetrics, GetPileupSummaries, CalculateContamination and Mutect2. Mutect2 was run with the latest filtering recommendations, including a Panel of Normal and a Germline Reference from the GATK database. Filtering was performed with the CalculateContamination, LearnReadOrientationModel and FilterMutectCalls tools with the default parameters. For the calculation of CNV GISTIC v2.0.23 was used with the parameters used by GDC. Catalog of Somatic Mutation in Cancer (COSMIC) was used to uncover pathogenic variants. For the analysis of variants and CNV, the maftool v2.10.0 and ComplexHeatmap 2.10.0 packages were used. All analyses were carried out on the GNU/Linux operating system under Ubuntu v20.01.3 or using the R v4.0.2 language in Rstudio v2021.09.0+351. A second bioinformatics pipeline was also used, SureCall software (Agilent) with the default parameters used for SNV variant calling. The variants found by both algorithms were taken as reliable SNV. Data were deposited in Sequence Read Archive hosted by National Center for Biotechnology Information under accession number PRJNA806516.
Phylogenetic tree inference (PTI) was run by means of the default parameters using matrices for each sample. These matrices contain an identifier for each variant, mutant read counts, counts of reference reads and the gene associated with the variant. The only PTI parameter was Allele Frequency of Mutation and was used to improve the speed of the algorithm. Briefly, PTI uses an iterative process on the variants shared between the samples. First, it builds the base of the tree using the variants shared by all the samples; second, it eliminates these variants and establishes a split node; and third, it eliminates the variants of the sample that produced the division (split). PTI iteratively performs these three steps for all division possibilities. Each tree is given a score based on an aggregated variant count, and the tree with the highest score is chosen as the optimal tree.

5.4. Sanger Sequencing forConfirmation of Exome Findings

Exome variant findings in exome sequencing were validated by Sanger sequencing using BigDye Terminator v3.1 Cycle Sequencing kit (ThermoFischer) in a 3500 Genetic Analyzer. Primers used for USP8 [48], TP53 [49], EGFR [50], AURKA [51], CDKN1A [52,53] and HSD3B1 sequencing have been previously reported.

5.5. Hormone and Transcription Factor Immunohistochemistry

Paraffin-embedded, formalin-fixed tissue blocks were stained with hematoxylin–eosin and reviewed by a pathologist. Tumors were represented with a 2-fold redundancy. Sections (3 μm) were cut and placed onto coated slides. Immunostaining was performed by means of the HiDef detection HRP polymer system (Cell Marque, CA, USA), using specific antibodies against each pituitary hormone (TSH, GH, PRL, FSH, LH and ACTH) and the lineage-specific transcription factors TBX19, POU1F1 and NR5A1, as previously described [54]. Two independent observers performed assessment of hormones and transcription factors expression at different times.

5.6. Statistical Analysis

Two-tailed Fisher exact tests and Student’s t tests were used to evaluate the relationship between the identified gene variants and clinicopathological features. A p value of <0.05 was considered statistically significant. Statistical software consisted of SPSS v28.0.1

Supplementary Materials

The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/ijms23094861/s1.

Author Contributions

D.M.-R., K.T.-P. and M.M. conceived, designed and coordinated the project, performed experiments, analyzed, discussed data and prepared the manuscript. S.A.-E., G.S.-R., E.P.-M., S.V.-P., R.S., L.B.-A., C.G.-T., J.G.-C. and J.T.A.-S. performed DNA purification, library preparation, sequencing experiments, bioinformatics analysis and wrote the manuscript. A.-L.E.-d.-l.-M., I.R.-S., E.G.-A., L.A.P.-O., G.G., S.M.-J., L.C.-M., B.L.-F. and A.B.-L. provided biological samples and detailed patient information. All authors have read and agreed to the published version of the manuscript.

Funding

This work was partially supported by grants 289499 from Fondos Sectoriales Consejo Nacional de Ciencia y Tecnologia, Mexico, and R-2015-785-015 from Instituto Mexicano del Seguro Social (MM).

Institutional Review Board Statement

Protocol approved by the Comisión Nacional de Ética e Investigación Científica of the Instituto Mexicano del Seguro Social, in accordance with the Helsinki declaration (R-2019-785-052).

Informed Consent Statement

Informed consent was obtained from all subjects involved in the study.

Data Availability Statement

Data were deposited in Sequence Read Archive hosted by National Center for Biotechnology Information under accession number PRJNA806516.

Acknowledgments

Sergio Andonegui-Elguera is a doctoral student from Programa de Doctorado en Ciencias Biomédicas, Universidad Nacional Autónoma de México (UNAM) and received fellowship 921084 from CONACYT. KTP is a recipient of Consejo Nacional de Ciencia y Tecnología (CONACyT) fellowship “Estáncias posdoctorales por Mexico 2021” program. DMR is a recipient of the National Council for Science and Technology Fellowship “Catedra CONACyT” program.

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Heaney, A. Clinical review: Pituitary carcinoma: Difficult diagnosis and treatment. J. Clin. Endocrinol. Metab. 2011, 96, 3649–3660. [Google Scholar] [CrossRef] [PubMed]
  2. Raverot, G.; Burman, P.; McCormack, A.; Heaney, A.; Petersenn, S.; Popovic, V.; Trouillas, J.; Dekkers, O.M.; The European Society of Endocrinology. European Society of Endocrinology Clinical Practice Guidelines for the management of aggressive pituitary tumours and carcinomas. Eur. J. Endocrinol. 2018, 178, G1–G24. [Google Scholar] [CrossRef]
  3. Todeschini, A.B.; Beer-Furlan, A.; Montaser, A.S.; Jamshidi, A.O.; Ghalib, L.G.; Chavez, J.A.; Lehman, N.L.; Prevedello, D.M. Pituitary carcinomas: Review of the current literature and report of atypical case. Br. J. Neurosurg. 2020, 34, 528–533. [Google Scholar] [CrossRef]
  4. Dudziak, K.; Honegger, J.; Bornemann, A.; Horger, M.; Müssig, K. Pituitary carcinoma with malignant growth from first presentation and fulminant clinical course-case report and review of the literature. J. Clin. Endocrinol. Metab. 2011, 96, 2665–2669. [Google Scholar] [CrossRef] [PubMed]
  5. Di Ieva, A.; Davidson, J.; Syro, L.; Rotondo, F.; Montoya, J.; Horvath, E.; Cusimano, M.D.; Kovacs, K. Crooke’s cell tumors of the pituitary. Neurosurgery 2015, 76, 616–622. [Google Scholar] [CrossRef] [PubMed]
  6. Fountas, A.; Lavrentaki, A.; Subramanian, A.; Toulis, K.; Nirantharakumar k Karavitaki, N. Recurrence in silent corticotroph adenomas after primary treatment: A systematic review and meta-analysis. J. Clin. Endocrinol. Metab. 2019, 104, 1039–1048. [Google Scholar] [CrossRef] [PubMed]
  7. Molitch, M. Diagnosis and Treatment of Pituitary Adenomas: A Review. JAMA 2017, 317, 516–524. [Google Scholar] [CrossRef]
  8. Melmed, S. Pituitary-Tumor Endocrinopathies. N. Engl. J. Med. 2020, 382, 937–950. [Google Scholar] [CrossRef] [PubMed]
  9. Lopes, M.B.S. The 2017 World Health Organization classification of tumors of the pituitary gland: A sumary. Acta Neuropathol. 2017, 134, 521–535. [Google Scholar] [CrossRef] [PubMed]
  10. Song, Z.-J.; Reitman, Z.; Ma, Z.-Y.; Chen, J.-H.; Zhang, Q.-L.; Shou, X.-F.; Huang, C.X.; Wang, Y.F.; Li, S.Q.; Mao, Y.; et al. The genome-wide mutational landscape of pituitary adenomas. Cell Res. 2016, 26, 1255–1259. [Google Scholar] [CrossRef]
  11. NCBI. Available online: https://www.ncbi.nlm.nih.gov/snp/rs1047303#frequency_tab (accessed on 23 February 2022).
  12. Shiota, M.; Narita, S.; Akamatsu, S.; Fujimoto, N.; Sumiyoshi, T.; Fujiwara, M.; Uchiumi, T.; Habuchi, T.; Ogawa, O.; Eto, M. Association of Missense Polymorphism in HSD3B1 With Outcomes Among Men With Prostate Cancer Treated With Androgen-Deprivation Therapy or Abiraterone. JAMA Netw. Open 2019, 2, e190115. [Google Scholar] [CrossRef]
  13. Yagnik, G.; Jahangiri, A.; Chen, R.; Wagner, J.; Aghi, M. Role of a p53 polymorphism in the development of nonfunctional pituitary adenomas. Mol. Cell Endocrinol. 2017, 446, 81–90. [Google Scholar] [CrossRef]
  14. Heidari, Z.; Harati-Sadegh, M.; Arian, A.; Maruei-Milan, R.; Salimi, S. The effect of TP53 and P21 gene polymorphisms on papillary thyroid carcinoma susceptibility and clinical/pathological features. IUBMB Life 2020, 72, 922–930. [Google Scholar] [CrossRef]
  15. Akhter, N.; Dar, S.; Haque, S.; Wahid, M.; Jawed, A.; Akhtar, M.S.; A Alharbi, R.; A A Sindi, A.; Alruwetei, A.; Choudhry, H.M.Z.; et al. Crosstalk of Cyclin-dependent kinase inhibitor 1A (CDKN1A) gene polymorphism with p53 and CCND1 polymorphism in breast cancer. Eur. Rev. Med. Pharmacol. Sci. 2021, 25, 4258–4273. [Google Scholar]
  16. Wang, C.; Nie, H.; Li, Y.; Liu, G.; Wang, X.; Xing, S.; Zhang, L.; Chen, X.; Chen, Y.; Li, Y. The study of the relation of DNA repair pathway genes SNPs and the sensitivity to radiotherapy and chemotherapy of NSCLC. Sci. Rep. 2016, 6, 26526. [Google Scholar] [CrossRef] [PubMed]
  17. Taniguchi-Ponciano, K.; Portocarrero-Ortiz, L.A.; Guinto, G.; Moreno-Jimenez, S.; Gomez-Apo, E.; Chavez-Macias, L.; Peña-Martínez, E.; Silva-Román, G.; Vela-Patiño, S.; Ordoñez-García, J.; et al. The kinome, cyclins and cyclin-dependent kinases of pituitary adenomas, a look into the gene expression rofile among tumors different lineages. BMC Med. Genom. 2022, 15, 52. [Google Scholar] [CrossRef]
  18. Sobral-Leite, M.; Lips, E.; Vieira-Monteiro, H.; Giacomin, L.; Freitas-Alves, D.; Cornelissen, S.; Mulder, L.; Wesseling, J.; Schmidt, M.K.; Vianna-Jorge, R. Evaluation of the EGFR polymorphism R497K in two cohorts of neoadjuvantly treated breast cancer patients. PLoS ONE 2017, 12, e0189750. [Google Scholar] [CrossRef]
  19. Zhang, H.; Berezov, A.; Wang, Q.; Zhang, G.; Drebin, J.; Murali, R.; Greene, M. ErbB receptors: From oncogenes to targeted cancer therapies. J. Clin. Investig. 2007, 117, 2051–2058. [Google Scholar] [CrossRef] [PubMed]
  20. Liu, X.; Feng, M.; Dai, C.; Bao, X.; Deng, K.; Yao, Y.; Wang, R. Expression of EGFR in Pituitary Corticotroph Adenomas and Its Relationship With Tumor Behavior. Front. Endocrinol. 2019, 10, 785. [Google Scholar] [CrossRef]
  21. NCBI. Available online: https://www.ncbi.nlm.nih.gov/snp/rs2227983#frequency_tab (accessed on 23 February 2022).
  22. Wang, S.; Qi, J.; Zhu, M.; Wang, M.; Nie, J. AURKA rs2273535 T>A Polymorphism Associated With Cancer Risk: A Systematic Review With Meta-Analysis. Front. Oncol. 2020, 10, 1040. [Google Scholar] [CrossRef] [PubMed]
  23. NCBI. Available online: https://www.ncbi.nlm.nih.gov/snp/rs2273535#frequency_tab (accessed on 23 February 2022).
  24. Reincke, M.; Sbiera, S.; Hayakawa, A.; Theodoropoulou, M.; Osswald, A.; Beuschlein, F.; Meitinger, T.; Mizuno-Yamasaki, E.; Kawaguchi, K.; Saeki, Y.; et al. Mutations in the deubiquitinase gene USP8 cause Cushing’s disease. Nat. Genet. 2015, 47, 31–38. [Google Scholar]
  25. Perez-Rivas, L.; Theodoropoulou, M.; Ferraù, F.; Nusser, C.; Kawaguchi, K.; Faucz, F.; Nusser, C.; Kawaguchi, K.; Stratakis, C.A.; Faucz, F.R.; et al. The Gene of the Ubiquitin-Specific Protease 8 Is Frequently Mutated in Adenomas Causing Cushing’s Disease. J. Clin. Endocrinol. Metab 2015, 100, E997–E1004. [Google Scholar] [CrossRef] [PubMed]
  26. Albani, A.; Pérez-Rivas, L.G.; Dimopoulou, C.; Zopp, S.; Colón-Bolea, P.; Roeber, S.; Honegger, J.; Flitsch, J.; Rachinger, W.; Buchfelder, M.; et al. The USP8 mutational status may predict long-term remission in patients with Cushing’s disease. Clin. Endocrinol. 2018, 89, 454–458. [Google Scholar] [CrossRef] [PubMed]
  27. Wanichi, I.Q.; de Paula Mariani, B.M.; Frassetto, F.P.; Siqueira, S.A.C.; de Castro Musolino, N.R.; Cunha-Neto, M.B.C.; Ochman, G.; Cescato, V.A.S.; Machado, M.C.; Trarbach, E.B.; et al. Cushing’s disease due to somatic USP8 mutations: A systematic review and meta-analysis. Pituitary 2019, 22, 435–442. [Google Scholar] [CrossRef]
  28. Uzilov, A.; Taik, P.; Cheesman, K.; Javanmard, P.; Ying, K.; Roehnelt, A.; Wang, H.; Fink, M.Y.; Lau, C.Y.; Moe, A.S.; et al. USP8 and TP53 Drivers are Associated with CNV in a Corticotroph Adenoma Cohort Enriched for Aggressive Tumors. J. Clin. Endocrinol. Metab. 2021, 106, 826–842. [Google Scholar] [CrossRef]
  29. Hayashi, K.; Inoshita, N.; Kawaguchi, K.; Ibrahim, A.; Suzuki, H.; Fukuhara, N.; Okada, M.; Nishioka, H.; Takeuchi, Y.; Komada, M.; et al. The USP8 mutational status may predict drug susceptibility in corticotroph adenomas of Cushing’s disease. Eur. J. Endocrinol. 2016, 174, 213–226. [Google Scholar] [CrossRef] [PubMed]
  30. Shao, X.; Lv, N.; Liao, J.; Long, J.; Xue, R.; Ai, N.; Xu, D.; Fan, X. Copy number variation is highly correlated with differential gene expression: A pan-cancer study. BMC Med Genet. 2019, 20, 175. [Google Scholar] [CrossRef] [PubMed]
  31. Shlien, A.; Malkin, D. Copy number variations and cancer. Genome Med. 2009, 1, 62. [Google Scholar] [CrossRef]
  32. Pös, O.; Radvanszky, J.; Styk, J.; Pös, Z.; Buglyó, G.; Kajsik, M.; Budis, J.; Nagy, B.; Szemes, T. Copy Number Variation: Methods and Clinical Applications. Appl. Sci. 2021, 11, 819. [Google Scholar] [CrossRef]
  33. Cui, Y.; Li, C.; Jiang, Z.; Zhang, S.; Li, Q.; Liu, X.; Zhou, Y.; Li, R.; Wei, L.; Li, L.; et al. Single-cell transcriptome and genome analyses of pituitary neuroendocrine tumors. Neuro-Oncol. 2021, 23, 1859–1871. [Google Scholar] [CrossRef]
  34. Neou, M.; Villa, C.; Armignacco, R.; Jouinot, A.; Raffin-Sanson, M.-L.; Septier, A.; Letourneur, F.; Diry, S.; Diedisheim, M.; Izac, B.; et al. Pangenomic Classification of Pituitary Neuroendocrine Tumors. Cancer Cell 2020, 37, 123–134.e5. [Google Scholar] [CrossRef] [PubMed]
  35. Varis, A.; Wolf, M.; Monni, O.; Vakkari, M.-L.; Kokkola, A.; Moskaluk, C.; Frierson, H.; Powell, S.M.; Knuutila, S.; Kallioniemi, A.; et al. Targets of gene amplification and overexpression at 17q in gastric cancer. Cancer Res. 2002, 62, 2625–2629. [Google Scholar] [PubMed]
  36. Shlien, A.; Tabori, U.; Marshall, C.; Pienkowska, M.; Feuk, L.; Novokmet, A.; Nanda, S.; Druker, H.; Scherer, S.W.; Malkin, D. Excessive genomic DNA copy number variation in the Li-Fraumeni cancer predisposition syndrome. Proc. Natl. Acad. Sci. USA 2008, 105, 11264–11269. [Google Scholar] [CrossRef] [PubMed]
  37. McCormack, A.; Dekkers, O.; Petersenn, S.; Popovic, V.; Trouillas, J.; Raverot, G.; Burman, P. Treatment of aggressive pituitary tumours and carcinomas: Results of a European Society of Endocrinology (ESE) survey 2016. Eur. J. Endocrinol. 2018, 178, 265–276. [Google Scholar] [CrossRef]
  38. Trouillas, J.; Jaffrain-Rea, M.; Vasiljevic, A.; Raverot, G.; Roncaroli, F.; Villa, C. How to Classify the Pituitary Neuroendocrine Tumors (PitNET)s in 2020. Cancers 2020, 12, 514. [Google Scholar] [CrossRef]
  39. Hu, H.; Khodadadi-Jamayran, A.; Dolgalev, I.; Cho, H.; Badri, S.; Chiriboga, L.A.; Zeck, B.; Gregorio, M.L.D.R.; Dowling, C.M.; Labbe, K.; et al. Targeting the Atf7ip-Setdb1 Complex Augments Antitumor Immunity by Boosting Tumor Immunogenicity. Cancer Immunol. Res. 2021, 9, 1298–1315. [Google Scholar] [CrossRef]
  40. Park, J.; Huang, S.; Tougeron, D.; Sinicrope, F. MSH3 mismatch repair protein regulates sensitivity to cytotoxic drugs and a histone deacetylase inhibitor in human colon carcinoma cells. PLoS ONE 2013, 8, e65369. [Google Scholar] [CrossRef]
  41. Raverot, G.; Ilie, M.; Lasolle, H.; Amodru, V.; Trouillas, J.; Castinetti, F.; Brue, T. Aggressive pituitary tumours and pituitary carcinomas. Nat. Rev. Endocrinol. 2021, 17, 671–684. [Google Scholar] [CrossRef]
  42. Syro, L.; Rotondo, F.; Camargo, M.; Ortiz, L.; Serna, C.; Kovacs, K. Temozolomide and Pituitary Tumors: Current Understanding, Unresolved Issues, and Future Directions. Front. Endocrinol. 2018, 9, 318. [Google Scholar] [CrossRef]
  43. Mamidi, T.K.K.; Wu, J.; Hicks, C. Integrating germline and somatic variation information using genomic data for the discovery of biomarkers in prostate cancer. BMC Cancer 2019, 19, 229. [Google Scholar] [CrossRef]
  44. Caja, F.; Vodickova, L.; Kral, J.; Vymetalkova, V.; Naccarati, A.; Vodicka, P. Mismatch repair gene variant in sporadic solid cancers. Int. J. Mol. Sci. 2020, 21, 5561. [Google Scholar] [CrossRef]
  45. Song, G.G.; Kim, J.H.; Lee, H. Genome-wide pathway analysis in major depresive disorder. J. Mol. Neurosci. 2013, 51, 428–436. [Google Scholar] [CrossRef] [PubMed]
  46. GATK. Available online: https://gatk.broadinstitute.org/ (accessed on 6 October 2021).
  47. GDC. Available online: https://docs.gdc.cancer.gov/ (accessed on 6 October 2021).
  48. Chang, M.; Yang, C.; Bao, X.; Wang, R. Genetic and Epigenetic Causes of Pituitary Adenomas. Front. Endocrinol. 2021, 11, 596554. [Google Scholar] [CrossRef] [PubMed]
  49. Ballmann, C.; Thiel, A.; Korah, H.E.; Reis, A.C.; Saeger, W.; Stepanow, S.; Köhrer, K.; Reifenberger, G.; Knobbe-Thomsen, C.B.; Knappe, U.J.; et al. USP8 Mutations in Pituitary Cushing Adenomas-Targeted Analysis by Next-Generation Sequencing. J. Endocr. Soc. 2018, 2, 266–278. [Google Scholar] [CrossRef] [PubMed]
  50. Naidoo, P.; Naidoo, R.; Ramkaran, P.; Chuturgoon, A. Effect of maternal HIV infection, BMI and NOx air pollution exposure on birth outcomes in South African pregnant women genotyped for the p53 Pro72Arg (rs1042522). Int. J. Immunogenet. 2020, 47, 414–429. [Google Scholar] [CrossRef] [PubMed]
  51. Leite, M.; Giacomin, L.; Piranda, D.; Festa-Vasconcellos, J.; Indio-do-Brasil, V.; Koifman, S.; de Moura-Neto, R.S.; de Carvalho, M.A.; Vianna-Jorge, R. Epidermal growth factor receptor gene polymorphisms are associated with prognostic features of breast cancer. BMC Cancer 2014, 14, 190. [Google Scholar] [CrossRef]
  52. Baumann, A.; Buchberger, A.; Piontek, G.; Schüttler, D.; Rudelius, M.; Reiter, R.; Gebel, L.; Piendl, G.; Brockhoff, G.; Pickhard, A. The Aurora-Kinase A Phe31-Ile polymorphism as possible predictor of response to treatment in head and neck squamous cell carcinoma. Oncotarget 2018, 9, 12769–12780. [Google Scholar] [CrossRef]
  53. Vargas-Torres, S.L.; Portari, E.A.; Silva, A.L.; Klumb, E.M.; da Rocha Guillobel, H.C.; de Camargo, M.J.; Santos-Rebouças, C.B.; Russomano, F.B.; Macedo, J.M.B. Roles of CDKN1A gene polymorphisms (rs1801270 and rs1059234) in the development of cervical neoplasia. Tumour Biol. 2016, 37, 10469–10478. [Google Scholar] [CrossRef]
  54. Taniguchi-Ponciano, K.; Andonegui-Elguera, S.; Peña-Martínez, E.; Silva-Román, G.; Vela-Patiño, S.; Gomez-Apo, E.; Chavez-Macias, L.; Vargas-Ortega, G.; Espinosa-de-Los-Monteros, L.; Gonzalez-Virla, B.; et al. Transcriptome and methylome analysis reveals three cellular origins of pituitary tumors. Sci. Rep. 2020, 10, 19373. [Google Scholar] [CrossRef]
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Filed under: Cushing's, pituitary | Tagged: , , , | Leave a comment »

Possible Good News! Effects of Tubastatin A on Adrenocorticotropic Hormone Synthesis and Proliferation of Att-20 Corticotroph Tumor Cells

Posted on March 31, 2022 by MaryO
  • Rie HagiwaraDepartment of Endocrinology and Metabolism, Hirosaki University Graduate School of Medicine, Hirosaki 036-8562, Japan
  • Kazunori KageyamaDepartment of Endocrinology and Metabolism, Hirosaki University Graduate School of Medicine, Hirosaki 036-8562, Japan
  • Yasumasa IwasakiSuzuka University of Medical Science, Suzuka 510-0293, Japan
  • Kanako NiiokaDepartment of Endocrinology and Metabolism, Hirosaki University Graduate School of Medicine, Hirosaki 036-8562, Japan
  • Makoto DaimonDepartment of Endocrinology and Metabolism, Hirosaki University Graduate School of Medicine, Hirosaki 036-8562, Japan
Abstract

Cushing’s disease is an endocrine disorder characterized by hypercortisolism, mainly caused by autonomous production of ACTH from pituitary adenomas. Autonomous ACTH secretion results in excess cortisol production from the adrenal glands, and corticotroph adenoma cells disrupt the normal cortisol feedback mechanism. Pan-histone deacetylase (HDAC) inhibitors inhibit cell proliferation and ACTH production in AtT-20 corticotroph tumor cells. A selective HDAC6 inhibitor has been known to exert antitumor effects and reduce adverse effects related to the inhibition of other HDACs. The current study demonstrated that the potent and selective HDAC6 inhibitor tubastatin A has inhibitory effects on proopiomelanocortin (Pomc) and pituitary tumor-transforming gene 1 (Pttg1) mRNA expression, involved in cell proliferation. The phosphorylated Akt/Akt protein levels were increased after treatment with tubastatin A. Therefore, the proliferation of corticotroph cells may be regulated through the Akt-Pttg1 pathway. Dexamethasone treatment also decreased the Pomc mRNA level. Combined tubastatin A and dexamethasone treatment showed additive effects on the Pomc mRNA level. Thus, tubastatin A may have applications in the treatment of Cushing’s disease.

Access the PDF at https://www.jstage.jst.go.jp/article/endocrj/advpub/0/advpub_EJ21-0778/_pdf/-char/en

 

Filed under: Cushing's, pituitary, Rare Diseases, Treatments | Tagged: , , , , | Leave a comment »

Pregnancy Doesn’t Boost Cushing Disease Recurrences

Posted on March 25, 2022 by MaryO

Researchers published the study covered in this summary on Research Square as a preprint that has not yet been peer reviewed.

Key Takeaways

  • Among women who underwent pituitary surgery to treat Cushing disease subsequent pregnancy had no apparent effect on Cushing disease recurrence, in a single-center review of 113 women treated over a 30-year period.

Why This Matters

  • No single factor predicts the recurrence of Cushing disease during long-term follow-up of patients who have undergone pituitary surgery.
  • This is the first study to assess the effect of pregnancy on Cushing disease recurrence in a group of reproductive-age women who initially showed post-surgical remission.

Study Design

  • Retrospective study of 355 patients with confirmed Cushing disease who were admitted to a single tertiary hospital in Brazil between 1990 and 2020. All patients had transsphenoidal surgery, with a minimum follow-up of 6 months and median follow-up of 83 months. Remission occurred in 246 of these patients.
  • The current analysis focused on 113 of the patients who achieved remission, were women, were 45 years old or younger at time of surgery (median 32 years old), and had information available on their obstetric history.
  • Ninety-one of these women (81%) did not become pregnant after their surgery, and 22 (19%) became pregnant after surgery.

Key Results

  • Among the 113 women in the main analysis 43 (38%) had a Cushing disease recurrence, a median of 48 months after their pituitary surgery.
  • Following surgery, 11 women in each of the two subgroups (recurrence, no recurrence) became pregnant.
  • Although the subgroup with recurrence had a higher incidence of pregnancy (11/43; 26%) compared with those with no recurrence (11/70; 16%) Kaplan-Meier analysis showed that survival free of Cushing disease recurrence was similar and not significantly different in the women with a postsurgical pregnancy and those who did not become pregnant (P=.531).
  • The review also showed that, of the women who became pregnant, several obstetrical measures were similar between patients who had a recurrence and those who remained in remission, including number of pregnancies per patient, maternal weight gain, type of delivery (normal or cesarean), delivery time (term or premature), neonatal weight, and neonatal size. The review also showed roughly similar rates of maternal and fetal complications in these two subgroups of women who became pregnant.

Limitations

  • The study was retrospective and included a relatively small number of patients.
  • The authors collected information on obstetric history for some patients by telephone or email contacts.

Disclosures

  • The study received no commercial funding.
  • None of the authors had disclosures.

This is a summary of a preprint research study , ” Pregnancy After Pituitary Surgery Does Not Influence the Recurrence of Cushing‘ s Disease, “ written by researchers at the Sao Paulo (Brazil) University Faculty of Medicine on Research Square provided to you by Medscape. This study has not yet been peer reviewed. The full text of the study can be found on researchsquare.com.

Filed under: Cushing's, pituitary | Tagged: , , , | Leave a comment »

Difference in miRNA Expression in Functioning and Silent Corticotroph Pituitary Adenomas Indicates the Role of miRNA in the Regulation of Corticosteroid Receptors

Posted on March 19, 2022 by MaryO

Abstract

Corticotroph pituitary adenomas commonly cause Cushing’s disease (CD), but some of them are clinically silent. The reason why they do not cause endocrinological symptoms remains unclear. We used data from small RNA sequencing in adenomas causing CD (n = 28) and silent ones (n = 20) to explore the role of miRNA in hormone secretion and clinical status of the tumors. By comparing miRNA profiles, we identified 19 miRNAs differentially expressed in clinically functioning and silent corticotroph adenomas. The analysis of their putative target genes indicates a role of miRNAs in regulation of the corticosteroid receptors expression. Adenomas causing CD have higher expression of hsa-miR-124-3p and hsa-miR-135-5p and lower expression of their target genes NR3C1 and NR3C2. The role of hsa-miR-124-3p in the regulation of NR3C1 was further validated in vitro using AtT-20/D16v-F2 cells. The cells transfected with miR-124-3p mimics showed lower levels of glucocorticoid receptor expression than control cells while the interaction between miR-124-3p and NR3C1 3′ UTR was confirmed using luciferase reporter assay. The results indicate a relatively small difference in miRNA expression between clinically functioning and silent corticotroph pituitary adenomas. High expression of hsa-miR-124-3p in adenomas causing CD plays a role in the regulation of glucocorticoid receptor level and probably in reducing the effect of negative feedback mediated by corticosteroids.

1. Introduction

Pituitary adenomas (also referred to as pituitary neuroendocrine tumors, PitNETs) represent about 10–20% of intracranial neoplasms in adults. They may originate from different kinds of secretory pituitary cells including corticotroph ACTH-secreting cells. Corticotroph adenomas commonly cause ACTH-dependent Cushing’s disease, but a significant proportion of these tumors are endocrinologically non-functioning and classified as subclinical/silent corticotroph adenomas (SCAs) [1].
CD-causing ACTH tumors are commonly small microadenomas with approximately 50% being smaller than 5 mm, which is challenging for MRI diagnostics [2]. In contrary, SCAs are commonly diagnosed due to neurological symptoms related to tumor mass at the stage of large macroadenomas. Frequently they show invasive growth and increased proliferation index [1]. According to current recommendations, SCAs are now referred to as “high-risk” pituitary adenomas which refers to their fast and invasive growth, high risk of recurrence and resistance to medical therapy [3,4]. They are recognized to be more aggressive than other clinically nonfunctioning pituitary tumors such as those of gonadotroph origin or null-cell adenomas [5].
The mechanism underlying the difference in secretory activity of CD-causing and subclinical tumors is unclear and only a few studies focused on this issue were published. The results indicated a role of the expression levels of particular genes/proteins involved in the regulation of POMC expression and pro-hormone conversion into ACTH as well as genes involved in pituitary differentiation [6,7,8,9,10,11,12,13]. However, it also appears that both active and silent corticotroph adenomas share a similar overall gene expression profile [14,15].
The aim of this study was to compare the profiles of microRNA (miRNA) expression in clinically functioning and silent corticotroph adenomas and to identify miRNAs that play a role in different ACTH secretory activity.

2. Results

2.1. Patients Characteristics

The study included 28 patients with CD and 20 patients suffering from SCA. All patients with CD had clear clinical signs and symptoms of hypercortisolism verified according to biochemical criteria including elevated midnight cortisol levels and 24 h urinary free cortisol (UFC). Patients with SCA had no clinical or biochemical signs of hypercortisolism and showed normal levels of midnight cortisol and 24 h UFC. Patients with CD had significantly higher morning serum cortisol levels than patients with SCAs (p = 0.0002) while no significant difference was observed in the morning serum ACTH levels. No difference in cortisol/ACTH ratio was observed between CD and SCA patients.
All the adenoma samples were ACTH-positive upon immunohistochemical staining against pituitary hormones (ACTH, GH, TSH, FSH, LH, α-subunit) and had characteristic ultrastructural features of corticotroph adenoma. Forty-one adenomas were positive only for ACTH, while seven ACTH-positive adenomas showed additional moderate/weak immunoreactivity for α-subunit. Increased proliferation assessed by Ki67 index ≥ 3% was observed in a similar proportion of CD and SCA patients, seven tumors causing CD and five SCAs. A higher proportion of sparsely vs. densely granulated adenomas was observed in SCAs than in CD-related adenomas, but the difference did not cross a significance threshold (p = 0.0787). No difference in the proportion of invasive/noninvasive adenomas was observed in clinically functioning and silent corticotroph adenomas.
All SCAs were macroadenomas, while tumors causing CD included 17 macroadenomas and 11 microadenomas. No significant differences in preoperative clinical parameters, including 24 h UFC, morning serum ACTH level, morning and midnight serum cortisol level, cortisol/ACTH ratio, were observed between CD patients with micro- and macroadenomas. Irrespectively, a correlation between tumors size and ACTH level (Spearman R= 0.4678; p = 0.0121) and a negative correlation between cortisol/ACTH ratio (Spearman R= −0.4015; p = 0.0342) was observed in CD patients.
No correlation was found between the remaining biochemical parameters and tumor size. Overall, the patients’ characteristics are presented in Table 1, while details including both the clinical and histopathological data are shown in Supplementary Table S1.
Table 1. Summary of clinical features of patients with Cushing’s disease and silent corticotroph adenomas.
Table

2.2. Identification of miRNAs Differentially Expressed in Corticotroph Adenomas Causing CD and Subclinical Cortiotroph Adenomas

NGS data on miRNA expression of 48 corticotroph adenomas from previous investigation were used to compare miRNA expression levels between adenomas causing CD (n = 24) and subclinical corticotroph adenomas (n = 20). Sequencing of small RNA libraries produced approximately 2,497,367 reads per sample, which were mapped to the human genome (hg19) and used for quantification of expression levels of known miRNAs, according to miRBase 22 release. Sequencing reads were annotated to 1917 miRNAs. Measurements of 1902 mature miRNAs expression were included in the analysis, after filtering out the miRNAs with low expression.
When miRNA profiles of adenomas causing CD and SCAs were compared, a total of 19 differentially expressed miRNAs were found that met the criteria of adjusted p-value < 0.05. This set included 16 miRNAs with higher expression in tumors causing CD: hsa-miR-129-2-3p, hsa-miR-129-5p, hsa-miR-124-3p, hsa-miR-132-5p, hsa-miR-129-1-3p, hsa-miR-135b-5p, hsa-miR-27a-3p, hsa-miR-10b-5p, hsa-miR-9-3p, hsa-miR-6506-3p, hsa-miR-6864-5p, hsa-let-7b-5p, hsa-miR-670-3p, hsa-miR-22-5p, hsa-miR-346 and hsa-miR-9-5p, Three miRNAs with lower expression in CD patients were found: hsa-miR-1909-3p, hsa-miR-4319 and hsa-miR-181b-3p. Details are presented in Table 2 and Figure 1A,B.
Ijms 23 02867 g001 550
Figure 1. MiRNA expression profiling in corticotroph adenomas. (A). Difference in miRNA expression between functioning and silent corticotroph adenomas. Volcano plot showing differentially expressed miRNAs. Significance and fold change thresholds are marked with dashed lines. (B). Heat map representing the expression of differentially expressed miRNAs and clustering the samples of adenomas causing Cushing’s disease (CD) and silent corticotroph adenomas (SCA). (C). The correlation between the expression levels of differentially expressed miRNAs and POMC expression or hormonal laboratory measurements in patients: morning plasma ACTH level, morning and midnight plasma cortisol levels and 24 h urinary free cortisol; * indicate p-value < 0.05; ** indicate p-value < 0.01; *** indicate p-value < 0.001
Table 2. The list of miRNAs differentially expressed in corticotroph pituitary adenomas causing CD and silent corticotroph adenomas.
Table

2.3. The Correlation of miRNA Expression and Patients’ Clinical Data

Since the clustering of the tumors based on the expression of differentially expressed miRNAs did not clearly separate functioning and silent adenomas, we determined whether the expression of the identified differentially expressed miRNAs is directly related to the results of patients’ laboratory tests as well as POMC expression, measured in tumor samples with qRT-PCR. For this purpose, Spearman’s correlation was applied to calculate a correlation matrix. We observed a significant positive correlation between 13 miRNAs out of 19 differentially expressed miRNAs and at least one of clinical laboratory parameters: serum ACTH, morning cortisol level, midnight cortisol level or 24 h UFC. For 11 miRNAs, with higher expression in patients with CD a positive correlation was observed, while a negative correlation was observed for 3 miRNAs that have lower expression in patients with CD. Four of the differentially expressed miRNAs, hsa-miR-9-3p, hsa-miR-9-5p, hsa-miR-27a-3p and hsa-miR-6506-3p, are correlated with POMC expression level in tumor tissue. The absolute value of correlation coefficient ranged between 0.31 and 0.55 which indicates a weak/moderate relationship. Details are presented in Figure 1C.

2.4. Funtional Enrichment Analysis of Differentially Expressed miRNAs

To investigate the possible functional role of the identified miRNAs with different expression levels in CD tumors and SCAs, we used the information on experimentally validated miRNA targets gathered in the miRtarbase release 8.0 database. High confidence known miRNA targets that were validated with luciferase reporter assay, reported in miRtarbase, were included in the analysis. The enrichment of the genes reported as miRNA targets of our 19 miRNAs of interest was determined with gene set over-representation analysis (GSOA) based on Gene Ontology (GO) Molecular Function and GO Biological Processes. The list of all the genes reported in miRTarbase as validated with reporter gene assay was used as reference. As a result, we found 30 GO Molecular Function terms and 293 GO Biological Processes terms as significantly enriched with genes that are targets of the 19 differentially expressed miRNAs. Top 10 enriched terms were related mainly to steroid hormone activity, regulation of transcription and regulation of stem cell differentiation, as shown in Figure 2. Details are presented in Supplementary Table S2. We paid special attention to the terms that refer to steroid hormone action, i.e., steroid hormone receptor activity (GO:0003707), nuclear receptor activity (GO:0004879), ligand-activated transcription factor activity (GO:0098531), as well as steroid hormone-mediated signaling pathway (GO:0043401) and hormone-mediated signaling pathway (GO:0009755). Importantly, the miRNA target genes that were overrepresented in these terms included NR3C1 and NR3C2 that encode for adrenal hormones glucocorticoid receptor (GR) and mineralocorticoid receptor (MR), respectively. According to the miRtarbase 9.0 database, hsa-miR-124-3p is a negative regulator of NR3C1 gene [16] while both hsa-miR-124-3p and hsa-miR-135b-5p downregulate MR [17].
Ijms 23 02867 g002 550
Figure 2. Gene set over-representation analysis of putative target genes of miRNAs differentially expressed in clinically functioning and silent corticotroph adenomas.
Using the PubMed search, we found additional evidence strongly supporting the role of hsa-miR-124-3p in the regulation of NR3C1 [18,19,20,21] as well as the role of hsa-miR-135b-5p in downregulating NR3C2 [22,23].

2.5. Comparison of the Expression of NR3C1 and NR3C2 in Corticotroph Adenomas Causing CD and Silent Adenomas

We determined the expression levels of NR3C1 and NR3C2 in corticotroph adenomas with qRT-PCR. We observed a significantly lower expression of both genes in samples from CD patients (n = 24) as compared to SCAs (n = 24); fold change (FC) 0.49 p = 0.0166 and FC 0.37 p = 0.0132, for NR3C1 and NR3C2, respectively. However, the observed difference is rather slight and a notable dispersion of the results was observed (Figure 3). The differences in NR3C1 and NR3C2 expression correspond to the differences in hsa-miR-124-3p and hsa-miR-135b-5p levels. Patients with CD have higher levels of both miRNAs and lower levels of NR3C1 and NR3C2 mRNA (Figure 3). Unfortunately, we did not find a direct correlation between the expression levels of hsa-miR-124-3p and NR3C1 or hsa-miR-135b-5p and NR3C2.
Ijms 23 02867 g003 550
Figure 3. The expression levels of NR3C1 and NR3C2 measured with qRT-PCR as well as hsa-miR-124-3p and hsa-miR-135b-5p measured with small RNA sequencing in tumor samples from CD patients and silent corticotroph adenomas; * indicate p-value < 0.05

2.6. Investigtion of miRNA-Related Regulation of NR3C1 In Vitro

Transfecting the cultured cells with miRNA mimics is the commonly used approach of in vitro validation of specific miRNA–mRNA interaction. We used mice corticotroph tumor AtT-20/D16v-F2 cells for in vitro experiment and initially verified whether these cells do express Nr3c1 and Nr3c2 genes using deposited RNAseq data from a previous experiment on AtT-20 cells (GSE132324; Gene Expression Omnibus) and qRT-PCR. This showed that the AtT-20/D16v-F2 have relatively high expression of Nr3c1 but do not express Nr3c2. Thus, we focused on the regulatory role of miR-124-3p on Nr3c1 expression. We used miRBase [24] and Targetscan [25] to determine whether miR-124-3p is evolutionarily conserved in humans and mice and whether it targets NR3C1 in both species. It confirmed that miR-124-3p is broadly conserved and it shares the same sequence of mature miRNA in humans and mice. Importantly, GR is among highly rated miR-124-3p predicted targets in both humans and mice and two highly conserved miR-124-3p binding motifs in 3′UTR of this gene were identified in these two species (Figure 4A).
Ijms 23 02867 g004 550
Figure 4. Role of mir-124-3p in regulation of glucocorticoid receptor gene. (A). Putative hsa-mir-124-3p target sites in 3′UTR of NR3C1. (B). Reduced expression of Nr3c1 gene expression and glucocorticoid receptor (GR) protein level in AtT-20/D16v-F2 cells treated with hsa-miR-124-3p mimics. (C). Results of luciferase reporter gene assay, showing the interaction between Nr3c1 3′UTR site 2 and mir-124-3p; * indicate p-value < 0.05; ns—not significant.
When we transfected AtT-20/D16v-F2 cells with miR-124-3p miRNA mimic and unspecific negative control miRNA mimic, we observed a significant decrease in Nr3c1 expression in cells treated with miR-124-3p miRNA mimic (Figure 4B). It was significantly lower than in cells treated with unspecific miRNA mimic. This difference was also clearly visible at the protein level. The GR level was reduced in cells treated with miR-124-3p miRNA mimic as compared to control (Figure 4B).
Two fragments of Nr3c1 3′UTR including each of putative miR-124-3p binding motifs were cloned in plasmid vector into 3′ region of the firefly luciferase gene. AtT-20/D16v-F2 cells were transfected with empty vector, vector with miR-124-3p binding site 1 and vector miR-124-3p binding site 2. Each of the three variants of the cells were cotransfected with miR-124-3p miRNA mimic or unspecific miRNA mimic that served as a negative control. Luminescence was developed 48 h after transfection and detected with microplate reader. As a result, we observed a significant decrease in luminescence in the cells with introduced plasmid with miR-124-3p binding site 2 treated with miR-124-3p mimic as compared to the cells transfected with the same plasmid construct but with control miRNA mimic. This observation confirms the interaction between miR-124-3p and 3′ UTR of Nr3c1 at putative binding site 2 (Figure 4C). The experiment did not confirm an interaction between miR-124-3p and 3′ UTR of Nr3c1 at binding site 1 since no significant difference of luminescence was found in cells transfected with plasmid vector harboring this binding motif treated with miR-124-3p mimic and the same cells treated with negative miRNA mimic (Figure 4C).

3. Discussion

Based on the clinical manifestation and biochemical tests results, pituitary corticotroph adenomas can be divided into functioning adenomas causing Cushing’s disease and SCAs. These two subtypes of tumors also differ in terms of some characteristics in MRI [2,26] and pathological features [27]. In contrast to CD-causing adenomas which are commonly small microadenomas, SCAs are diagnosed as macroadenomas due to neurological symptoms related to tumor mass. They are characterized by invasive growth, high risk of recurrence and resistance to medical therapy and are therefore referred to as “high-risk” pituitary adenomas according to current classification [3,4]. In our study, the SCAs were larger than functioning counterparts, as expected. A clear prevalence of women is observed among CD patients according to literature data [28], while it is not observed in patients suffering from SCAs. Our SCA group contained near equal representation of women and men as in previous reports [29,30]; however, some studies indicated female prevalence in SCAs [31].
Comparing functioning and silent corticotroph adenomas, we did not observe difference in patients’ age as well as differences in invasive growth status, ratio of adenomas with increased proliferation index and proportions of sparsely and densely granulated adenomas that may suggest the lack of difference in the tumors’ “aggressiveness”. Importantly, limitations for generalization of our results should be noted. The number of patients included in the analysis is relatively low and the group is not representative of the general population, especially in the case of patients suffering from Cushing’s disease. Since the main goal of our study was a molecular profiling of tumor tissue, we intentionally preselected large adenomas, which allowed us to have enough tissue for DNA/RNA isolation and successful molecular procedures.
In our investigation, we observed a negative correlation between cortisol/ACTH ratio and tumor volume in functioning corticotroph adenomas as described previously [32]. However, we did not observe any difference between micro- and macroadenomas causing CD as compared to SCAs (data not shown) as was found in previous studies [12].
The reason why some of corticotroph adenomas exhibit excessive hormone secretion and the others remain clinically silent is unclear and only few attempts have been made to determine the possible molecular mechanism underlying this difference in secretory activity. They were mainly focused on investigating the expression of the selected genes or proteins by comparing subclinical and functioning corticotroph adenomas. These studies indicated different expression levels of prohormone convertase 1/3 POMC, genes encoding somatostatin receptors, corticotropin releasing hormone receptor 1, vasopressin receptor (V1BR), corticosteroid 11-beta-dehydrogenase as well as NEUROD1 and TPIT [6,7,8,9,10,11,12,13]. However, whole transcriptome studies indicated that adenomas causing CD and subclinical corticotroph adenomas share a very common gene expression profile and a very low number of differentially expressed genes can be found by comparing transcriptome of silent and CD-causing ACTH tumors [14,15].
In our study, we determined the miRNA expression profile of 28 clinically functioning adenomas and 20 SCAs with next-generation sequencing of small RNA fraction. This allowed for the quantification of over 1900 miRNA annotated to current version of miRbase database and comparing their expression in two groups of tumor samples. We found a significant difference only in the expression levels of 19 miRNAs, that represent less than 1% of the miRNAs included in the analysis. This result resembles the observation from previous comparison of whole transcriptome profiles in functioning adenomas and SCAs where only 34 differentially expressed genes were found. Generally, both observations indicate a very common molecular profile of corticotroph adenomas, regardless of the functional status.
In our study, the expression levels of 13 out of 19 identified differentially expressed miRNAs were also correlated with peripheral ACTH/cortisol levels, further supporting the role of these miRNAs in secretory activity of corticotroph adenomas.
The possible role of miRNA in subclinical nature of SCAs was addressed in only one previous study by García-Martínez A et al. [33]. The authors compared the expression of 5 miRNAs in 24 functioning and 23 silent adenomas and observed a difference in hsa-miR-200a and hsa-miR-103 levels [33]. Their results were not confirmed by our investigation since these two miRNAs were not found among differentially expressed miRNAs. In our data, very a similar expression level of hsa-miR-200a was observed in clinically functioning and silent adenomas. In turn, a slightly higher expression of hsa-miR-103a-3p was observed in SCAs as previously reported, but the difference did not cross the significance threshold level. We should note that different methods were used for these two studies and technical and analytical differences could result in this discrepancy.
Since miRNAs play a role in gene regulation, their effect should be investigated in the context of the function of targeted genes. The interaction between miRNA and its target mRNA 3′UTR can be predicted with in silico tools. Unfortunately, prediction results can be very difficult to interpret since a huge number of predicted interactions can be found for some miRNAs. For example, when using the Targetescan (http://www.targetscan.org; accessed on 28 February 2022) prediction tool [25], over 4000 target genes were predicted for each hsa-miR-9-3p, hsa-miR-1909-3p, hsa-miR-22-5p and hsa-miR-181b-3p that we found as differentially expressed in CD and SCA. Therefore, to investigate a possible functional relevance of differentially expressed miRNAs we used a database of experimentally validated miRNA targets [34]. Gene set over-representation analysis of miRNA target genes indicated their enrichment in the pathways of steroid hormone nuclear receptors functioning. This result indicates that miRNAs that have different expression levels in CD and SCAs play a role in the regulation of expression of genes involved in steroid hormone signaling at hormone receptor level. It is especially interesting since this group of compounds includes adrenal hormones that play a role in the regulation of the hypothalamic–pituitary–adrenal (HPA) axis.
The particular enriched miRNA target genes included NR3C1 and NR3C2 that encode for corticosteroid hormone receptors (GR and MR, respectively). Both receptors are located in the cytoplasm where they bind glucocorticoids. Upon ligand binding, they are translocated to nucleus where they form dimers on DNA at glucocorticoid response elements (GREs). Glucocorticoid and mineralocorticoid receptors directly regulate the expression of target genes and/or influence the expression indirectly through the interaction with other transcription factors [35].
Glucocorticoids play a role in the basic mechanism of negative feedback of HPA axis. They act on hypothalamus, where high cortisol levels reduce secretion of corticotropin-releasing hormone (CRH), thus they directly reduce stimulation of ACTH secretion by anterior pituitary lobe. Glucocorticoids also inhibit the activity of pituitary cells indirectly. Corticotroph cells express GRs and their activation results in the reduction of POMC expression and secretion of ACTH [36,37]. In pituitary corticotroph adenomas, NR3C1 point mutations and loss of heterozygosity in NR3C1 locus were identified [38]. These mutations seem to affect the secretory activity and result in tumor resistance to corticosteroids [39]. Reduced expression of corticosteroid receptors in corticotroph adenomas has been reported in patients with resistance to high doses of dexamethasone [40]. These data indicate a role of GR in secretory activity of clinically functioning corticotroph adenomas. The expression of corticosteroid genes was previously investigated in CD-causing tumors and SCAs and no significant differences were found. However, it is worth noting that a low number of SCA patients was included in these studies: n = 9 [13], n = 8 [11] and n = 2 [41].
According to previously published results, hsa-miR-124-3p is a negative regulator of NR3C1 [16,18,19,20,21]. This was observed in acute lymphoblastic leukemia [19], adipocytes [20] and human embryonic kidney cells [21], where the reduced expression of NR3C1 upon an increase in hsa-miR-124-3p as well as a direct interaction between this miRNA and 3′UTR of GR gene were observed. Some additional clinical observations also suggest the role of hsa-miR-124-3p in the regulation of the response to cortiosteroids in patients with acute-on-chronic liver failure [18] and lymphoblastic leukemia [19]. Hsa-miRNA-124 also mediates corticosteroid resistance in T-cells of sepsis patients through the downregulation of GR [42].
Our analysis of the expression level of NR3C1 in corticotroph adenomas showed that tumors causing CD have lower gene expression and accordingly they exhibit higher levels of hsa-miR-124-3p. Subsequently, the role of hsa-miR-124-3p in NR3C1 downregulation was confirmed in mice AtT-20/D16v-F2 corticotroph cells using miRNA mimics and reporter gene assay. Transfection of AtT-20/D16v-F2 cells with hsa-miR-124-3p mimics resulted in reduced NR3C1 mRNA expression and GR protein level. We also confirmed the interaction between hsa-miR-124-3p and one of two predicted binding motifs in 3′UTR of NR3C1 with luciferase reporter gene assay. Since sequences of hsa-miR-124-3p and target sequence in 3′UTR of NR3C1 mRNA are the same in mice and in humans, we believe that results showing the regulation of the GR-encoding gene in mice AtT-20/D16v-F2 cells are also relevant to humans. Together, the available data indicate that in pituitary corticotrophs, hsa-miR-124-3p downregulates the expression of the GR gene. Since this receptor mediates the response of pituitary cells to cortisol, the expression of hsa-miR-124-3p appears to be an important element in the regulation of secretory activity of corticotroph cells. Based on these results, we can hypothesize that in CD, a high level of hsa-miR-124-3p contributes to lowering of GR expression and in consequence it plays a role in lowering the effect of glucocorticoid feedback on the activity of corticotroph adenoma. Hsa-miR-124-3p and hsa-miR-135b-5p can downregulate the expression level of MR, as proven in model HeLa cells [17]. Expression of both miRNAs is higher in corticotroph adenomas causing CD which corresponds to the lower expression of the NR3C2 gene in these tumors as compared to SCAs. Since the role of the MR receptor expression in pituitary cells is poorly understood, the functional implication of this observation is much less clear than in the case of GR downregulation. MR and GR have similar amino acid sequences, especially in DNA-binding domain, but they differ in affinity to corticosteroids. MR is specific for both mineralocorticoids and glucocorticoids while GR is specific predominantly for glucocorticoids. MRs have much higher affinity for glucocorticoids than GRs and are activated at basal glucocorticoid conditions, while GR occupancy is increased when glucocorticoid levels rise during the circadian peak or stress. Due to these differences, these two receptors play slightly different roles, despite the fact that they share a number of target genes [43]. MR expression is considered more tissue-specific than GR and was reported to be the most prevalent in kidney and adipose tissue but also in the hippocampus and hypothalamus [44]. However, the available databases of human expression pattern such as the Genotype-Tissue Expression project (https://gtexportal.org; accessed on 10 December 2021) or Protein atlas (https://www.proteinatlas.org; accessed on 10 December 2021) indicate that MR is widely expressed in multiple human tissues and organs including the pituitary gland. Unfortunately, a role of MR receptor in pathogenesis of pituitary tumors remains unknown.
AtT-20 cells, which are the only available cell line model of corticotroph adenoma, do not express MR receptor, thus the procedure of experimental validation of the role of miRNA in NR3C2 silencing is not applicable. With a lack of experimental data on the exact role of MR, we can only hypothesize that miRNA-mediated silencing of NR3C2 may have the similar effect on HPA axis feedback as silencing of NR3C1. It may enhance ACTH secretion by reducing the direct inhibitory effect of glucocorticoids on neoplastic pituitary corticotrophs.
The difference in expression of hsa-miR-124-3p and hsa-miR-135b-5p between subclinical and CD-causing adenomas is not big, thus we suppose that high expression of these miRNAs is not the only cause of difference in ACTH secretion. Presumably this is one of the mechanisms in the regulation of corticotrophs’ secretory activity. The model of miRNA-based corticosteroid receptor regulation does not undermine the role of previously described differences in the expression of convertase 1/3, POMC, somatostatin receptors or corticotropin releasing hormone receptor 1 or genes involved in differentiation of pituitary cells [6,7,8,9,10,11,12,13]. When considering the complex nature of the regulation of ACTH secretion, it can be assumed that multiple mechanisms may be involved in the silent character of subclinical adenomas. The low number of identified differentially expressed miRNAs or genes in silent and clinically functioning adenomas probably results from the intertumoral molecular heterogeneity of SCAs. This is also in line with clinical evidence indicating that some silent corticotroph adenomas can transform into clinically functioning ones while the others remain silent [1].
The misregulation of GR expression or NR3C1 mutation may have important therapeutical implications in CD patients. Non-selective GR antagonist Mifepristone was officially approved for treatment in patients with Cushing’s syndrome [45] while another new GR inhibitor, Relacorilant (CORT125134), is under clinical investigation for its use in this group of patients [46]. The further studies will be required to assess the role of GR abnormalities in response to GR-targeting treatment in CD.
In our study, we focused mainly on the role of hsa-miR-124-3p and hsa-miR-135b-5p in the regulation of corticosteroid receptors, but the role of other differentially expressed miRNAs can also be elucidated, based on the function of putative target genes. In the pathways enrichment analysis of the putative targets, molecular functions related to transcriptional regulation were found among the top processes. Interestingly, five miRNAs, i.e., hsa-miR-132-5p, hsa-miR-135b-5p, hsa-miR-27a-3p, hsa-miR-9-3p and hsa-miR-9-5p, were previously reported to downregulate the expression of FOXO1 transcription factor [47,48,49,50,51]. FOXO1 plays an important role in the differentiation of pituitary cells [52] and secretion of gonadotropic hormones [53,54] and prolactin [55]. The role of FOXO1 in pituitary corticotroph cells was not investigated but it was shown to regulate POMC expression in POMC hypothalamic neurons [56]. In POMC, neurons of arcuate nucleus FOXO1 directly suppresses POMC expression. A similar mechanism was also observed in prolactin pituitary adenomas where FOXO1 suppresses the promoter of PRL gene [55]. It is possible that high expression of hsa-miR-132-5p, hsa-miR-135b-5p, hsa-miR-27a-3p, hsa-miR-9-3p and hsa-miR-9-5p in pituitary corticotroph adenomas reduces the level of FOXO1 and eventually contributes to the upregulation of POMC expression. In our data from corticotroph adenomas, we observed the correlation between levels of hsa-miR-9-3p/hsa-miR-9-5 and POMC expression, which also supports this concept, but the exact role of miRNAs in possible FOXO1-related regulation of secretory activity of corticotroph cells requires further functional investigation.

4. Materials and Methods

4.1. Patients and Tissue Samples

Pituitary tumor samples from 48 patients were collected during transsphenoidal surgery. Formalin-fixed and paraffin-embedded (FFPE) tissue samples, including 28 samples from patients with Cushing’s disease and 20 samples of SCA were used for the study. Diagnosis of hypercortisolism was based on standard hormonal criteria: increased UFC in three 24 h urine collections, disturbances of cortisol circadian rhythm, increased serum cortisol levels accompanied by increased or not suppressed plasma ACTH levels at 8.00 and a lack of suppression of serum cortisol levels to <1.8 µg/dL during an overnight dexamethasone suppression test (1 mg at midnight). The pituitary etiology of Cushing’s disease was confirmed based on the serum cortisol levels or UFC suppression < 50% with a high-dose dexamethasone suppression test (2 mg q.i.d. for 48 h) or a positive result of a corticotrophin-releasing hormone stimulation test (100 mg i.v.) and positive pituitary magnetic resonance imaging.
ACTH levels were assessed using IRMA (ELSA-ACTH, CIS Bio International, Gif-sur-Yvette Cedex, France). The analytical sensitivity was 2 pg/mL (reference range: 10–60 pg/mL). Serum cortisol concentrations were determined by the Elecsys 2010 electrochemiluminescence immunoassay (Roche Diagnostics, Mannheim, Germany). Sensitivity of the assay was 0.02 μg/dL (reference range: 6.2–19.4 μg/dL). UFC was determined after extraction (liquid/liquid with dichloromethane) by electrochemiluminescence immunoassay (Elecsys 2010, Roche Diagnostics)—reference range: 4.3–176 μg/24 h.
All the tumors underwent detailed histopathological diagnosis including immunohistochemical staining with antibodies against particular pituitary hormones (ACTH, GH, TSH, FSH, LH, α-subunit) and Ki67 as well as ultrastructural analysis with electron microscopy.
The SCAs were characterized by the following clinicopathological criteria: positive immunohistochemical staining for ACTH, lack of signs and symptoms of hypercortisolism (Cushing’s syndrome), negative hormonal evaluation and non-compliance with diagnostic criteria of the CD.
Macroadenoma was defined as an adenoma with at least one diameter exceeding 10 mm, and the tumor volume was assessed with the diChiro Nelson formula (height × length × width × π/6). Invasive growth of the tumors was evaluated using Knosp grading [57]. Adenomas with Knosp grades 0, 1 and 2 were considered non-invasive, while those with Knosp 3 and 4 were considered invasive.
Forty-three patients had a clear history of not using any drugs that control the overproduction of the cortisol or ACTH (ketoconazole, mitotane, metyrapone, osilodrostat, mifepristone, pasireotide) before surgical treatment. The information on preoperative pharmacological treatment was not available for 5 patients.
Tumor tissue content of each FFPE sample ranged between 80 and 100% (median 99%), as assessed with histopathological examination. Patients’ characteristics are presented in Table 1 and details on each patient’s data are available in Supplementary Table S1.
The study was approved by the local Ethics Committee of Maria Sklodowska-Curie National Research Institute of Oncology in Warsaw, Poland. Each patient provided informed consent for the use of tissue samples for scientific purposes.
Total RNA from FFPE samples was purified with RecoverAll™ Total Nucleic Acid Isolation Kit for FFPE tissue (Thermo Fisher Scientific, Waltham, MA, USA) and measured using NanoDrop 2000 (Thermo Fisher Scientific). RNA was stored at −70 °C.

4.2. Micro RNA Expression Profiling

For comparing the miRNA expression profiles in CD-causing and clinically silent adenomas, NGS data from our previous investigation of miRNA expression in corticotroph adenomas were used. The dataset is available at Gene Expression Omnibus, accession no GSE166279. Sequencing of small RNA fraction was performed in 48 tumor samples (28 CD patients and 20 SCA patients) with ion semiconductor sequencing technology, as described previously [58]. Briefly, Ion Total RNA-Seq Kit v2 (Thermo Fisher Scientific) was used for sequencing library construction, Ion Xpress™ RNA-Seq Barcode Kit was used for hybridization and ligation of RNA adapters. RNA reverse transcription and subsequent cDNA purification and library size selection were performed using Nucleic Acid Binding Beads and verified using Bioanalyzer 2100 with High Sensitivity DNA Kit (Agilent, Santa Clara, CA, USA). Ion Chef instrument, with Ion PI™ Hi-Q™ Chef Kit (Thermo Fisher Scientific) and Ion Proton sequencer (Thermo Fisher Scientific) were used for library preparation and sequencing, respectively.
BamToFastq package was applied for converting unmapped bam files into fastq files. miRDeep2 was applied for read mapping to known human miRNAs (according to miRBase release 22) and reads quantification. Data normalization and differential expression analysis were performed using DESeq2. Filtration for low-expression miRNAs was applied as described previously. FC of expression calculated as the ratio of the normalized read-count value in CD-causing and silent adenomas was used as a measure of expression difference. Adjusted p-value < 0.05 was used as significance threshold. MiRtarbase release 9.0 database [34] was used to identify known miRNA target genes. PANTHER (http://pantherdb.org; accessed on 10 December 2021) [59] was used for gene set over-representation analysis.

4.3. qRT-PCR gene Expression Analysis

One microgram of RNA was subjected to reverse transcription with Transcriptor First Strand cDNA Synthesis Kit (Roche Diagnostics). qRT-PCR reaction was carried out in 384-well format using 7900HT Fast Real-Time PCR System (Applied Biosystems, Foster City, CA, USA) and Power SYBR Green PCR Master Mix (Thermo Fisher Scientific) in a volume of 5 μL, containing 2.25 pmol of each primer. The samples were amplified in triplicates. GAPDH was used as reference gene. Delta Ct method was used to calculate the relative expression level. PCR primers’ sequences are presented in Supplementary Table S3.

4.4. Cell Line Culture and miRNA Mimic Transfection

AtT-20/D16v-F2 cells were purchased from ATCC collection and cultured in DMEM medium supplemented with 10% FBS, as recommended. MiRCURY LNA miRNA Mimics including hsa-miR-124-3p mimic (YM00471256, Qiagen, Hilden, Germany) and negative control mimic (YM00479902-ADB, Qiagen) were used. AtT-20/D16v-F2 cells were seeded at 5 × 104 per well of a 24-well plate in culture medium and transfected with 50 nM miRNA with 1% (v/v) HiPerFect Transfection Reagent (Qiagen), according to the manufacturer’s instructions. The next day, the culture medium was changed. In total, 48 h after transfection the cells were harvested and subjected to isolation of total RNA with RNeasy Mini Kit (Qiagen). The expression of the putative hsa-miR-124-3p target gene was determined with qRT-PCR.

4.5. Luciferase Reporter Gene Assay

Hsa-miR-124-3p target sites in 3′UTR of NR3C1 were determined with Targetscan [25]. Each of two predicted hsa-miR-124-3p target sites were cloned into pmirGLO Dual-Luciferase miRNA Target Expression Vector (Promega, Madison, WI, USA). AtT-20/D16v-F2 cells (2 × 104/well) were seeded onto a 96-well plate in 100 µL culture medium. The next day, the cells were transfected with 100 ng of each plasmid vector, independently using 0.25% (v/v) lipofectamine 3000 (Invitrogen, Carlsbad, CA, USA) in 10 µL of DMEM. The cells were subsequently transfected with either hsa-miR-124-3p mimic (YM00471256, Qiagen) or negative control mimic (YM00479902-ADB, Qiagen) in a final concentration of 50 nM using HiPerfectReagent (Qiagen). Culture medium was changed on the next day. Luciferase activity was measured with One-Glo Luciferase Assay System (Promega) 48 h after transfection.

4.6. Western Blotting

Cells were lysed in ice cold RIPA buffer, incubated for 30 min in 4 °C and centrifuged at 12,500× g rpm for 20 min at 4 °C. Samples were resolved using SDS-PAGE and electrotransferred to polyvinylidene fluoride membranes (PVDF) (Thermo Fisher). GR protein was detected with monoclonal anti-Glucocorticoid Receptor antibody (ab183127, Abcam, Cambridge, UK), and secondary anti-rabbit antibody conjugated to HRP (#7074, Cell Signaling, Beverly, MA, USA). Glyceraldehyde-3-Phosphate Dehydrogenase (#MAB374, Millipore, Bedford, MA, USA) detected with mouse HRP-conjugated antibody (#7076 Cell Signaling) served as control. Visualization was performed with SuperSignal West Pico Chemiluminescent Substrate (Thermo Fisher Scientific) and CCD digital imaging system Alliance Mini HD4 (UVItec Limited, Cambridge, UK).

4.7. Statistical Analysis

A two-sided Mann–Whitney U-test was used for analysis of continuous variables. The Spearman correlation method was used for correlation analysis. Significance threshold of α = 0.05 was adopted. Data were analyzed using GraphPad Prism 6.07 (GraphPad Software, La Jolla, CA, USA). Hierarchical clustering analysis was carried out with Cluster 3.0, and the results were visualized using TreeView 1.6 software (Stanford University School of Medicine, Stanford, CA, USA).

Supplementary Materials

The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/ijms23052867/s1.

Author Contributions

Conceptualization, M.M. and M.B.; Methodology, M.B. and B.J.M.; Software, J.B.; Formal analysis, P.K., B.J.M. and M.B.; Investigation, B.J.M., P.K., N.R., M.B. and M.P.; Resources, J.K., G.Z., A.S. and T.M.; Data curation, J.B., B.J.M. and M.B.; Writing—original draft preparation, M.B., P.K. and B.J.M.; Writing—review and editing, all the authors; Visualization, M.B. and B.J.M.; Supervision, M.M.; Project administration M.B.; Funding acquisition, M.M. All authors have read and agreed to the published version of the manuscript.

Funding

This research was funded by National Science Centre, Poland, grant number 2021/05/X/NZ5/01874.

Institutional Review Board Statement

The study was conducted in accordance with the Declaration of Helsinki, and approved by the local Ethics Committee of Maria Sklodowska-Curie Institute—Oncology Center in Warsaw, Poland; approval no. number 44/2018, date of approval 26 July 2018.

Informed Consent Statement

Informed consent was obtained from all subjects involved in the study.

Data Availability Statement

Data from next-generation sequencing of small RNA fraction of 48 corticotroph adenoma samples are available at Gene Expression Omnibus, accession no GSE166279.

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Ben-Shlomo, A.; Cooper, O. Silent Corticotroph Adenomas. Pituitary 2018, 21, 183–193. [Google Scholar] [CrossRef] [PubMed]
  2. Vitale, G.; Tortora, F.; Baldelli, R.; Cocchiara, F.; Paragliola, R.M.; Sbardella, E.; Simeoli, C.; Caranci, F.; Pivonello, R.; Colao, A. Pituitary Magnetic Resonance Imaging in Cushing’s Disease. Endocrine 2017, 55, 691–696. [Google Scholar] [CrossRef] [PubMed]
  3. Kontogeorgos, G.; Thodou, E.; Osamura, R.Y.; Lloyd, R.V. High-Risk Pituitary Adenomas and Strategies for Predicting Response to Treatment. Hormones 2022, 1, 3. [Google Scholar] [CrossRef] [PubMed]
  4. Osamura, R.Y.; Grossman, A.; Korbonits, M.; Kovacs, K.; Lopes, M.B.S.; Matsuno, A.; Trouillas, J. WHO Classification of Tumours of Endocrine Organs, 4th ed.; Lloyd, R.V., Osamura, R.Y., Rosari, J., Eds.; IARC Press: Lyon, France, 2017. [Google Scholar]
  5. Jiang, S.; Chen, X.; Wu, Y.; Wang, R.; Bao, X. An Update on Silent Corticotroph Adenomas: Diagnosis, Mechanisms, Clinical Features, and Management. Cancers 2021, 13, 6134. [Google Scholar] [CrossRef]
  6. Tateno, T.; Kato, M.; Tani, Y.; Oyama, K.; Yamada, S.; Hirata, Y. Differential Expression of Somatostatin and Dopamine Receptor Subtype Genes in Adrenocorticotropin (ACTH)- Secreting Pituitary Tumors and Silent Corticotroph Adenomas. Endocr. J. 2009, 56, 579–584. [Google Scholar] [CrossRef]
  7. Gabalec, F.; Beranek, M.; Netuka, D.; Masopust, V.; Nahlovsky, J.; Cesak, T.; Marek, J.; Cap, J. Dopamine 2 Receptor Expression in Various Pathological Types of Clinically Non-Functioning Pituitary Adenomas. Pituitary 2012, 15, 222–226. [Google Scholar] [CrossRef]
  8. Righi, A.; Faustini-Fustini, M.; Morandi, L.; Monti, V.; Asioli, S.; Mazzatenta, D.; Bacci, A.; Foschini, M.P. The Changing Faces of Corticotroph Cell Adenomas: The Role of Prohormone Convertase 1/3. Endocrine 2017, 56, 286–297. [Google Scholar] [CrossRef]
  9. Ohta, S.; Nishizawa, S.; Oki, Y.; Yokoyama, T.; Namba, H. Significance of Absent Prohormone Convertase 1/3 in Inducing Clinically Silent Corticotroph Pituitary Adenoma of Subtype I—Immunohistochemical Study. Pituitary 2002, 5, 221–223. [Google Scholar] [CrossRef]
  10. Tateno, T.; Izumiyama, H.; Doi, M.; Akashi, T.; Ohno, K.; Hirata, Y. Defective Expression of Prohormone Convertase 1/3 in Silent Corticotroph Adenoma. Endocr. J. 2007, 54, 777–782. [Google Scholar] [CrossRef]
  11. Tateno, T.; Izumiyama, H.; Doi, M.; Yoshimoto, T.; Shichiri, M.; Inoshita, N.; Oyama, K.; Yamada, S.; Hirata, Y. Differential Gene Expression in ACTH -Secreting and Non-Functioning Pituitary Tumors. Eur. J. Endocrinol. 2007, 157, 717–724. [Google Scholar] [CrossRef]
  12. Nagaya, T.; Seo, H.; Kuwayama, A.; Sakurai, T.; Tsukamoto, N.; Nakane, T.; Sugita, K.; Matsui, N. Pro-Opiomelanocortin Gene Expression in Silent Corticotroph-Cell Adenoma and Cushing’s Disease. J. Neurosurg. 1990, 72, 262–267. [Google Scholar] [CrossRef] [PubMed]
  13. Raverot, G.; Wierinckx, A.; Jouanneau, E.; Auger, C.; Borson-Chazot, F.; Lachuer, J.; Pugeat, M.; Trouillas, J. Clinical, Hormonal and Molecular Characterization of Pituitary ACTH Adenomas without (Silent Corticotroph Adenomas) and with Cushing’s Disease. Eur. J. Endocrinol. 2010, 163, 35–43. [Google Scholar] [CrossRef] [PubMed]
  14. Neou, M.; Villa, C.; Armignacco, R.; Jouinot, A.; Raffin-Sanson, M.L.; Septier, A.; Letourneur, F.; Diry, S.; Diedisheim, M.; Izac, B.; et al. Pangenomic Classification of Pituitary Neuroendocrine Tumors. Cancer Cell 2020, 37, 123–134.e5. [Google Scholar] [CrossRef] [PubMed]
  15. Bujko, M.; Kober, P.; Boresowicz, J.; Rusetska, N.; Paziewska, A.; Dabrowska, M.; Piaścik, A.; Pȩkul, M.; Zieliński, G.; Kunicki, J.; et al. USP8 Mutations in Corticotroph Adenomas Determine a Distinct Gene Expression Profile Irrespective of Functional Tumour Status. Eur. J. Endocrinol. 2019, 181, 615–627. [Google Scholar] [CrossRef]
  16. Vreugdenhil, E.; Verissimo, C.S.L.; Mariman, R.; Kamphorst, J.T.; Barbosa, J.S.; Zweers, T.; Champagne, D.L.; Schouten, T.; Meijer, O.C.; De Ron Kloet, E.; et al. MicroRNA 18 and 124a Down-Regulate the Glucocorticoid Receptor: Implications for Glucocorticoid Responsiveness in the Brain. Endocrinology 2009, 150, 2220–2228. [Google Scholar] [CrossRef]
  17. Sõber, S.; Laan, M.; Annilo, T. MicroRNAs MiR-124 and MiR-135a Are Potential Regulators of the Mineralocorticoid Receptor Gene (NR3C2) Expression. Biochem. Biophys. Res. Commun. 2010, 391, 727–732. [Google Scholar] [CrossRef]
  18. Wang, X.; Xu, H.; Wang, Y.; Shen, C.; Ma, L.; Zhao, C. MicroRNA-124a Contributes to Glucocorticoid Resistance in Acute-on-Chronic Liver Failure by Negatively Regulating Glucocorticoid Receptor Alpha. Ann. Hepatol. 2020, 19, 214–221. [Google Scholar] [CrossRef]
  19. Liang, Y.N.; Tang, Y.L.; Ke, Z.Y.; Chen, Y.Q.; Luo, X.Q.; Zhang, H.; Huang, L. Bin MiR-124 Contributes to Glucocorticoid Resistance in Acute Lymphoblastic Leukemia by Promoting Proliferation, Inhibiting Apoptosis and Targeting the Glucocorticoid Receptor. J. Steroid Biochem. Mol. Biol. 2017, 172, 62–68. [Google Scholar] [CrossRef]
  20. Liu, K.; Zhang, X.; Wei, W.; Liu, X.; Tian, Y.; Han, H.; Zhang, L.; Wu, W.; Chen, J. Myostatin/Smad4 Signaling-Mediated Regulation of Mir-124-3p Represses Glucocorticoid Receptor Expression and Inhibits Adipocyte Differentiation. Am. J. Physiol. Endocrinol. Metab. 2019, 316, E635–E645. [Google Scholar] [CrossRef]
  21. Roy, B.; Dunbar, M.; Shelton, R.C.; Dwivedi, Y. Identification of MicroRNA-124-3p as a Putative Epigenetic Signature of Major Depressive Disorder. Neuropsychopharmacology 2017, 42, 864–875. [Google Scholar] [CrossRef]
  22. Zhang, Z.; Che, X.; Yang, N.; Bai, Z.; Wu, Y.; Zhao, L.; Pei, H. MiR-135b-5p Promotes Migration, Invasion and EMT of Pancreatic Cancer Cells by Targeting NR3C2. Biomed. Pharmacother. 2017, 96, 1341–1348. [Google Scholar] [CrossRef] [PubMed]
  23. Huang, Y.; Wang, Y.; Ouyang, Y. Elevated MicroRNA-135b-5p Relieves Neuronal Injury and Inflammation in Post-Stroke Cognitive Impairment by Targeting NR3C2. Int. J. Neurosci. 2021, 132, 58–66. [Google Scholar] [CrossRef] [PubMed]
  24. Kozomara, A.; Birgaoanu, M.; Griffiths-Jones, S. MiRBase: From MicroRNA Sequences to Function. Nucleic Acids Res. 2019, 47, D155–D162. [Google Scholar] [CrossRef] [PubMed]
  25. McGeary, S.E.; Lin, K.S.; Shi, C.Y.; Pham, T.M.; Bisaria, N.; Kelley, G.M.; Bartel, D.P. The Biochemical Basis of MicroRNA Targeting Efficacy. Science 2019, 366, eaav1741. [Google Scholar] [CrossRef] [PubMed]
  26. Kasuki, L.; Antunes, X.; Coelho, M.C.A.; Lamback, E.B.; Galvão, S.; Silva Camacho, A.H.; Chimelli, L.; Ventura, N.; Gadelha, M.R. Accuracy of Microcystic Aspect on T2-Weighted MRI for the Diagnosis of Silent Corticotroph Adenomas. Clin. Endocrinol. 2020, 92, 145–149. [Google Scholar] [CrossRef] [PubMed]
  27. Thodou, E.; Argyrakos, T.; Kontogeorgos, G. Galectin-3 as a Marker Distinguishing Functioning from Silent Corticotroph Adenomas. Hormones 2007, 6, 227–232. [Google Scholar]
  28. Valassi, E.; Santos, A.; Yaneva, M.; Tóth, M.; Strasburger, C.J.; Chanson, P.; Wass, J.A.H.; Chabre, O.; Pfeifer, M.; Feelders, R.A.; et al. The European Registry on Cushing’s Syndrome: 2-Year Experience. Baseline Demographic and Clinical Characteristics. Eur. J. Endocrinol. 2011, 165, 383–392. [Google Scholar] [CrossRef]
  29. Langlois, F.; Shao, D.; Lim, T.; Yedinak, C.G.; Cetas, I.; Mccartney, S.; Cetas, J.; Dogan, A.; Fleseriu, M. Predictors of Silent Corticotroph Adenoma Recurrence; a Large Retrospective Single Center Study and Systematic Literature Review. Pituitary 2018, 21, 32–40. [Google Scholar] [CrossRef]
  30. Jahangiri, A.; Wagner, J.R.; Pekmezci, M.; Hiniker, A.; Chang, E.F.; Kunwar, S.; Blevins, L.; Aghi, M.K. A Comprehensive Long-Term Retrospective Analysis of Silent Corticotrophic Adenomas vs Hormone-Negative Adenomas. Neurosurgery 2013, 73, 8–17. [Google Scholar] [CrossRef]
  31. Strickland, B.A.; Shahrestani, S.; Briggs, R.G.; Jackanich, A.; Tavakol, S.; Hurth, K.; Shiroishi, M.S.; Liu, C.-S.J.; Carmichael, J.D.; Weiss, M.; et al. Silent Corticotroph Pituitary Adenomas: Clinical Characteristics, Long-Term Outcomes, and Management of Disease Recurrence. J. Neurosurg. 2021, 135, 1–8. [Google Scholar] [CrossRef]
  32. Machado, M.C.; Alcantara, A.E.E.; Pereira, A.C.L.; Cescato, V.A.S.; Castro Musolino, N.R.; de Mendonça, B.B.; Bronstein, M.D.; Fragoso, M.C.B.V. Negative Correlation between Tumour Size and Cortisol/ ACTH Ratios in Patients with Cushing’s Disease Harbouring Microadenomas or Macroadenomas. J. Endocrinol. Invest. 2016, 39, 1401–1409. [Google Scholar] [CrossRef] [PubMed]
  33. García-Martínez, A.; Fuentes-Fayos, A.C.; Fajardo, C.; Lamas, C.; Cámara, R.; López-Muñoz, B.; Aranda, I.; Luque, R.M.; Picó, A. Differential Expression of MicroRNAs in Silent and Functioning Corticotroph Tumors. J. Clin. Med. 2020, 9, 1838. [Google Scholar] [CrossRef] [PubMed]
  34. Huang, H.Y.; Lin, Y.C.D.; Li, J.; Huang, K.Y.; Shrestha, S.; Hong, H.C.; Tang, Y.; Chen, Y.G.; Jin, C.N.; Yu, Y.; et al. MiRTarBase 2020: Updates to the Experimentally Validated MicroRNA-Target Interaction Database. Nucleic Acids Res. 2020, 48, D148–D154. [Google Scholar] [CrossRef] [PubMed]
  35. Koning, A.S.C.A.M.; Buurstede, J.C.; van Weert, L.T.C.M.; Meijer, O.C. Glucocorticoid and Mineralocorticoid Receptors in the Brain: A Transcriptional Perspective. J. Endocr. Soc. 2019, 3, 1917–1930. [Google Scholar] [CrossRef]
  36. Nakai, Y.; Usui, T.; Tsukada, T.; Takahashi, H.; Fukata, J.; Fukushima, M.; Senoo, K.; Imura, H. Molecular Mechanisms of Glucocorticoid Inhibition of Human Proopiomelanocortin Gene Transcription. J. Steroid Biochem. Mol. Biol. 1991, 40, 301–306. [Google Scholar] [CrossRef]
  37. Drouin, J.; Charron, J.; Gagner, J.-P.; Jeannotte, L.; Nemer, M.; Plante, R.K.; Wrange, Ö. Pro-opiomelanocortin Gene: A Model for Negative Regulation of Transcription by Glucocorticoids. J. Cell. Biochem. 1987, 35, 293–304. [Google Scholar] [CrossRef]
  38. Huizenga, N.A.T.M.; De Lange, P.; Koper, J.W.; Clayton, R.N.; Farrell, W.E.; Van Der Lely, A.J.; Brinkmann, A.O.; De Jong, F.H.; Lamberts, S.W.J. Human Adrenocorticotropin-Secreting Pituitary Adenomas Show Frequent Loss of Heterozygosity at the Glucocorticoid Receptor Gene Locus. J. Clin. Endocrinol. Metab. 1998, 83, 917–921. [Google Scholar] [CrossRef]
  39. Miao, H.; Liu, Y.; Lu, L.; Gong, F.; Wang, L.; Duan, L.; Yao, Y.; Wang, R.; Chen, S.; Mao, X.; et al. Effect of 3 NR3C1 Mutations in the Pathogenesis of Pituitary ACTH Adenoma. Endocrinology 2021, 162, bqab167. [Google Scholar] [CrossRef]
  40. Mu, Y.M.; Takayanagi, R.; Imasaki, K.; Ohe, K.; Ikuyama, S.; Yanase, T.; Nawata, H. Low Level of Glucocorticoid Receptor Messenger Ribonucleic Acid in Pituitary Adenomas Manifesting Cushing’s Disease with Resistance to a High Dose-Dexamethasone Suppression Test. Clin. Endocrinol. 1998, 49, 301–306. [Google Scholar] [CrossRef]
  41. Ebisawa, T.; Tojo, K.; Tajima, N.; Kamio, M.; Oki, Y.; Ono, K.; Sasano, H. Immunohistochemical Analysis of 11-β-Hydroxysteroid Dehydrogenase Type 2 and Glucocorticoid Receptor in Subclinical Cushing’s Disease Due to Pituitary Macroadenoma. Endocr. Pathol. 2008, 19, 252–260. [Google Scholar] [CrossRef]
  42. Ledderose, C.; Möhnle, P.; Limbeck, E.; Schütz, S.; Weis, F.; Rink, J.; Briegel, J.; Kreth, S. Corticosteroid Resistance in Sepsis Is Influenced by MicroRNA-124-Induced Downregulation of Glucocorticoid Receptor-α. Crit. Care Med. 2012, 40, 2745–2753. [Google Scholar] [CrossRef] [PubMed]
  43. Spencer, R.L.; Deak, T. A Users Guide to HPA Axis Research. Physiol. Behav. 2017, 178, 43–65. [Google Scholar] [CrossRef] [PubMed]
  44. Han, F.; Ozawa, H.; Matsuda, K.I.; Nishi, M.; Kawata, M. Colocalization of Mineralocorticoid Receptor and Glucocorticoid Receptor in the Hippocampus and Hypothalamus. Neurosci. Res. 2005, 51, 371–381. [Google Scholar] [CrossRef]
  45. Castinetti, F.; Fassnacht, M.; Johanssen, S.; Terzolo, M.; Bouchard, P.; Chanson, P.; Cao, D.; Morange, I.; Picó, A.; Ouzounian, S.; et al. Merits and Pitfalls of Mifepristone in Cushing’s Syndrome. Eur. J. Endocrinol. 2009, 160, 1003–1010. [Google Scholar] [CrossRef] [PubMed]
  46. Fleseriu, M.; Laws, E.R.; Witek, P.; Pivonello, R.; Ferrigno, R.; de Martino, M.C.; Simeoli, C.; di Paola, N.; Pivonello, C.; Barba, L.; et al. Medical Treatment of Cushing’s Disease: An Overview of the Current and Recent Clinical Trials. Front. Endocrinol. 2020, 11, 648. [Google Scholar] [CrossRef]
  47. Senyuk, V.; Zhang, Y.; Liu, Y.; Ming, M.; Premanand, K.; Zhou, L.; Chen, P.; Chen, J.; Rowley, J.D.; Nucifora, G.; et al. Critical Role of MiR-9 in Myelopoiesis and EVI1-Induced Leukemogenesis. Proc. Natl. Acad. Sci. USA 2013, 110, 5594–5599. [Google Scholar] [CrossRef] [PubMed]
  48. Liu, D.Z.; Chang, B.; Li, X.D.; Zhang, Q.H.; Zou, Y.H. MicroRNA-9 Promotes the Proliferation, Migration, and Invasion of Breast Cancer Cells via down-Regulating FOXO1. Clin. Transl. Oncol. 2017, 19, 1133–1140. [Google Scholar] [CrossRef] [PubMed]
  49. Guttilla, I.K.; White, B.A. Coordinate Regulation of FOXO1 by MiR-27a, MiR-96, and MiR-182 in Breast Cancer Cells. J. Biol. Chem. 2009, 284. [Google Scholar] [CrossRef]
  50. Xu, Y.; Zhao, S.; Cui, M.; Wang, Q. Down-Regulation of MicroRNA-135b Inhibited Growth of Cervical Cancer Cells by Targeting FOXO1. Int. J. Clin. Exp. Pathol. 2015, 8, 10294–10304. [Google Scholar]
  51. Lau, P.; Bossers, K.; Janky, R.; Salta, E.; Frigerio, C.S.; Barbash, S.; Rothman, R.; Sierksma, A.S.R.; Thathiah, A.; Greenberg, D.; et al. Alteration of the MicroRNA Network during the Progression of Alzheimer’s Disease. EMBO Mol. Med. 2013, 5, 1613–1634. [Google Scholar] [CrossRef]
  52. Kapali, J.; Kabat, B.E.; Schmidt, K.L.; Stallings, C.E.; Tippy, M.; Jung, D.O.; Edwards, B.S.; Nantie, L.B.; Raeztman, L.T.; Navratil, A.M.; et al. Foxo1 Is Required for Normal Somatotrope Differentiation. Endocrinology 2016, 157, 4351–4363. [Google Scholar] [CrossRef] [PubMed]
  53. Garrel, G.; Denoyelle, C.; L’Hôte, D.; Picard, J.Y.; Teixeira, J.; Kaiser, U.B.; Laverrière, J.N.; Cohen-Tannoudji, J. GnRH Transactivates Human AMH Receptor Gene via Egr1 and FOXO1 in Gonadotrope Cells. Neuroendocrinology 2019, 108, 65–83. [Google Scholar] [CrossRef] [PubMed]
  54. Skarra, D.V.; Arriola, D.J.; Benson, C.A.; Thackray, V.G. Forkhead Box O1 Is a Repressor of Basal and GnRH-Induced Fshb Transcription in Gonadotropes. Mol. Endocrinol. 2013, 27, 1825–1839. [Google Scholar] [CrossRef] [PubMed]
  55. Xiao, Z.; Wang, Z.; Hu, B.; Mao, Z.; Zhu, D.; Feng, Y.; Zhu, Y. MiR-1299 Promotes the Synthesis and Secretion of Prolactin by Inhibiting FOXO1 Expression in Drug-Resistant Prolactinomas. Biochem. Biophys. Res. Commun. 2019, 520, 79–85. [Google Scholar] [CrossRef]
  56. Benite-Ribeiro, S.A.; de Lima Rodrigues, V.A.; Machado, M.R.F. Food Intake in Early Life and Epigenetic Modifications of Pro-Opiomelanocortin Expression in Arcuate Nucleus. Mol. Biol. Rep. 2021, 48, 3773–3784. [Google Scholar] [CrossRef]
  57. Knosp, E.; Steiner, E.; Kitz, K.; Matula, C.; Parent, A.D.; Laws, E.R.; Ciric, I. Pituitary Adenomas with Invasion of the Cavernous Sinus Space: A Magnetic Resonance Imaging Classification Compared with Surgical Findings. Neurosurgery 1993, 33, 610–618. [Google Scholar] [CrossRef]
  58. Bujko, M.; Kober, P.; Boresowicz, J.; Rusetska, N.; Zeber-Lubecka, N.; Paziewska, A.; Pekul, M.; Zielinski, G.; Styk, A.; Kunicki, J.; et al. Differential MicroRNA Expression in USP8-Mutated and Wild-Type Corticotroph Pituitary Tumors Reflect the Difference in Protein Ubiquitination Processes. J. Clin. Med. 2021, 10, 375. [Google Scholar] [CrossRef]
  59. Mi, H.; Ebert, D.; Muruganujan, A.; Mills, C.; Albou, L.P.; Mushayamaha, T.; Thomas, P.D. PANTHER Version 16: A Revised Family Classification, Tree-Based Classification Tool, Enhancer Regions and Extensive API. Nucleic Acids Res. 2021, 49, D394–D403. [Google Scholar] [CrossRef]
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Filed under: Cushing's, pituitary, Rare Diseases | Tagged: , , , | Leave a comment »

Covid-19 and Cushing’s Disease in a Patient with ACTH-secreting Pituitary Carcinoma

Posted on March 12, 2022 by MaryO

Abstract

Summary

The pandemic caused by severe acute respiratory syndrome coronavirus 2 is of an unprecedented magnitude and has made it challenging to properly treat patients with urgent or rare endocrine disorders. Little is known about the risk of coronavirus disease 2019 (COVID-19) in patients with rare endocrine malignancies, such as pituitary carcinoma. We describe the case of a 43-year-old patient with adrenocorticotrophic hormone-secreting pituitary carcinoma who developed a severe COVID-19 infection. He had stabilized Cushing’s disease after multiple lines of treatment and was currently receiving maintenance immunotherapy with nivolumab (240 mg every 2 weeks) and steroidogenesis inhibition with ketoconazole (800 mg daily). On admission, he was urgently intubated for respiratory exhaustion. Supplementation of corticosteroid requirements consisted of high-dose dexamethasone, in analogy with the RECOVERY trial, followed by the reintroduction of ketoconazole under the coverage of a hydrocortisone stress regimen, which was continued at a dose depending on the current level of stress. He had a prolonged and complicated stay at the intensive care unit but was eventually discharged and able to continue his rehabilitation. The case points out that multiple risk factors for severe COVID-19 are present in patients with Cushing’s syndrome. ‘Block-replacement’ therapy with suppression of endogenous steroidogenesis and supplementation of corticosteroid requirements might be preferred in this patient population.

Learning points

  • Comorbidities for severe coronavirus disease 2019 (COVID-19) are frequently present in patients with Cushing’s syndrome.
  • ‘Block-replacement’ with suppression of endogenous steroidogenesis and supplementation of corticosteroid requirements might be preferred to reduce the need for biochemical monitoring and avoid adrenal insufficiency.
  • The optimal corticosteroid dose/choice for COVID-19 is unclear, especially in patients with endogenous glucocorticoid excess.
  • First-line surgery vs initial disease control with steroidogenesis inhibitors for Cushing’s disease should be discussed depending on the current healthcare situation.

Background

The pandemic caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has had a significant impact on the health care systems to date. The clinical presentation of coronavirus disease 2019 (COVID-19) is diverse, ranging from asymptomatic illness to respiratory failure requiring admission to the intensive care unit (ICU). Risk factors for severe course include old age, male gender, comorbidities such as arterial hypertension, diabetes mellitus, chronic lung-, heart-, liver- and kidney disease, malignancy, immunodeficiency and pregnancy (1). Little is known about the risk of COVID-19 in patients with rare endocrine malignancies, such as pituitary carcinoma.

Case presentation

This case concerns a 43-year-old man with adrenocorticotrophic hormone (ACTH)-secreting pituitary carcinoma (with cerebellar and cervical drop metastases) with a severe COVID-19 infection. He had previously received multiple treatment modalities including surgery, radiotherapy, ketoconazole, pasireotide, cabergoline, bilateral (subtotal) adrenalectomy and temozolomide chemotherapy as described elsewhere (2). His most recent therapy was a combination of immune checkpoint inhibitors consisting of ipilimumab (3 mg/kg) and nivolumab (1 mg/kg) (anti-CTLA-4 and anti-PD-1, respectively) every 3 weeks for four cycles, after which maintenance therapy with nivolumab (240 mg) every 2 weeks was continued. Residual endogenous cortisol production was inhibited with ketoconazole 800 mg daily. He had stabilized disease with a decrease in plasma ACTH, urinary free cortisol and stable radiological findings (2). Surgical resection of the left adrenal remnant was planned but was not carried out due to the development of a COVID-19 infection.

In March 2021, he consulted our emergency department for severe respiratory complaints. He had been suffering from upper respiratory tract symptoms for one week, with progressive dyspnoea in the last three days. He tested positive for SARS-CoV-2 the day before admission. On examination, his O2 saturation was 72%, with tachypnoea (40/min) and bilateral pulmonary crepitations. His temperature was 37.2°C, blood pressure 124/86 mmHg and pulse rate 112 bpm. High-flow oxygen therapy was initiated but yielded insufficient improvement (O2 saturation of 89% and tachypnoea 35/min). He was urgently intubated for respiratory exhaustion.

Investigation

Initial investigations showed type 1 respiratory insufficiency with PaO2 of 52.5 mmHg (normal 75–90), PaCO2 of 33.0 mmHg (normal 36–44), pH of 7.47 (normal 7.35–7.45) and a P/F ratio of 65.7 (normal >300). His inflammatory parameters were elevated with C-reactive protein level of 275.7 mg/L (normal <5·0) and white blood cell count of 7.1 × 10⁹ per L with 72.3% neutrophils. His most recent morning plasma ACTH-cortisol level (measured using the Elecsys electrochemiluminescence immunoassays on a Cobas 8000 immunoanalyzer [Roche Diagnostics]) before his admission was 213 ng/L (normal 7.2–63) and 195 µg/L (normal 62–180) respectively, while a repeat measurement 3 weeks after his admission demonstrated increased cortisol levels of 547 µg/L (possibly iatrogenic due to treatment with high-dose hydrocortisone) and a decreased ACTH of 130 ng/L.

Treatment

On admission, he was started on high-dose dexamethasone therapy for 10 days together with broad-spectrum antibiotics for positive sputum cultures containing Serratia, methicillin-susceptible Staphylococcus aureus and Haemophilus influenzae. Thromboprophylaxis with an intermediate dose of low molecular weight heparin (tinzaparin 14 000 units daily for a body weight of 119 kg) was initiated. A ‘block-replacement’ regimen was adopted with the continuation of ketoconazole (restarted on day 11) in view of his endocrine treatment and the supplementation of hydrocortisone at a dose depending on the current level of stress. The consecutive daily dose of hydrocortisone and ketoconazole is shown in Fig. 1.

Figure 1View Full Size
Figure 1
‘Block-replacement’ therapy with ketoconazole and hydrocortisone/dexamethasone. Dexamethasone 10 mg daily was initially started as COVID-19 treatment, followed by hydrocortisone at a dose consistent with current levels of stress. Ketoconazole was restarted on day 11 and titrated to a dose of 800 mg daily to suppress endogenous glucocorticoid production.

Citation: Endocrinology, Diabetes & Metabolism Case Reports 2022, 1; 10.1530/EDM-21-0182

Outcome and follow-up

He developed multiple organ involvement, including metabolic acidosis, acute renal failure requiring continuous venovenous hemofiltration, acute coronary syndrome type 2, septic thrombophlebitis of the right jugular vein, and critical illness polyneuropathy. He was readmitted twice to the ICU, for ventilator-associated pneumonia and central line-associated bloodstream infection respectively. He eventually recovered and was discharged from the hospital to continue his rehabilitation.

Discussion

We describe the case of a patient with severe COVID-19 infection with active Cushing’s disease due to pituitary carcinoma, who was treated with high-dose dexamethasone followed by ‘block-replacement’ therapy with hydrocortisone in combination with off-label use of ketoconazole as a steroidogenesis inhibitor. His hospitalization was prolonged by multiple readmissions to the ICU for infectious causes. Our case illustrates the presence of multiple comorbidities for a severe and complicated course of COVID-19 in a patient with active Cushing’s disease.

Dexamethasone was initially chosen as the preferred corticosteroid therapy, in analogy with the RECOVERY trial, in which dexamethasone at a dose of 6mg once daily (oral or i.v.) resulted in lower 28-day mortality in hospitalized patients with COVID-19 requiring oxygen therapy or invasive mechanical ventilation (3). However, the optimal dose/choice of corticosteroid therapy is unclear, especially in a patient population with pre-existing hypercortisolaemia. A similar survival benefit for hydrocortisone compared to dexamethasone has yet to be convincingly demonstrated. This may be explained by differences in anti-inflammatory activity but could also be due to the fact that recent studies with hydrocortisone were stopped early and were underpowered (4, 5).

Multiple risk factors for a complicated course of COVID-19 are present in patients with Cushing’s syndrome and might increase morbidity and mortality (6, 7). These include a history of obesity, arterial hypertension and impaired glucose metabolism. Prevention and treatment of these pre-existing comorbidities are essential.

Patients with Cushing’s syndrome also have an increased thromboembolic risk, which is further accentuated by the development of severe COVID-19 infection (6, 7). Thromboprophylaxis with low molecular weight heparin is associated with lower mortality in COVID-19 patients with high sepsis‐induced coagulopathy score or high D-dimer levels (8) and is presently widely used in the treatment of severe COVID-19 disease (9). Subsequently, this treatment is indicated in hospitalized COVID-19 patients with Cushing’s syndrome. It is unclear whether therapeutic anticoagulation dosing could provide additional benefits (6, 7). An algorithm based on the International Society on Thrombosis and Hemostasis-Disseminated Intravascular Coagulation score was proposed to evaluate the ideal anticoagulation therapy in severe/critical COVID-19 patients, with an indication for therapeutic low molecular weight heparin dose at a score ≥5 (9).

Furthermore, the chronic cortisol excess induces suppression of the innate and adaptive immune response. Patients with Cushing’s syndrome, especially when severe and active, should be considered immunocompromised and have increased susceptibility for viral and other (hospital-acquired) infections. Prophylaxis for Pneumocystis jirovecii with trimethoprim/sulfamethoxazole should therefore be considered (6, 7).

Additionally, there is a particular link between the pathophysiology of COVID-19 and Cushing’s syndrome. The SARS-CoV-2 virus (as well as other coronaviruses) enter human cells by binding the ACE2 receptor. The transmembrane serine protease 2 (TMPRSS2), expressed by endothelial cells, is additionally required for the priming of the spike-protein of SARS-CoV-2, leading to viral entry. TMPRSS2 was studied in prostate cancer and found to be regulated by androgen signalling. Consequently, the androgen excess frequently associated with Cushing’s syndrome might be an additional risk factor for contracting COVID-19 via higher TMPRSS2 expression (10), especially in women, in whom the effect of excess androgen would be more noticeable compared to male patients with Cushing’s syndrome.

Treating Cushing’s syndrome with a ‘block-replacement’ approach, with suppression of endogenous steroidogenesis and supplementation of corticosteroid requirements, is an approach that should be considered, especially in severe or cyclic disease. The use of this method might decrease the need for monitoring and reduce the occurrence of adrenal insufficiency (7). Our patient was on treatment with ketoconazole, which was interrupted at initial presentation and then restarted under the coverage of a hydrocortisone stress regimen. Ketoconazole was chosen because of its availability. Advantages of ketoconazole over metyrapone include its antifungal activity with the potential for prevention of invasive pulmonary fungal infections, as well as its antiandrogen action (especially in female patients) and subsequent inhibition of TMPRSS2 expression (10). Regular monitoring of the liver function (every month for the first 3 months, at therapy initiation or dose increase) is necessary. Caution is needed due to its inhibition of multiple cytochrome P450 enzymes (including CYP3A4) and subsequently greater risk of drug-drug interactions vs metyrapone (7, 10). Another disadvantage of ketoconazole is the need for oral administration. In our patient, ketoconazole was delivered through a nasogastric tube. i.v. etomidate is an alternative in case of an unavailable enteral route.

Finally, as a general point, the first-line treatment of a patient with a novel diagnosis of Cushing’s disease is transsphenoidal surgery. Recent endocrine recommendations pointed out the possibility of initial disease control with steroidogenesis inhibitors in patients without an indication for urgent intervention during a high prevalence of COVID-19 (7). This would allow the optimalization of metabolic parameters; emphasizing that the short-to mid-term prognosis is related to the cortisol excess and not its cause. Surgery could then be postponed until the health situation allows for safe elective surgery (7). This decision depends of course on the evolution of COVID-19 and the healthcare system in each country and should be closely monitored by policymakers and physicians.

Declaration of interest

The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the research reported.

Funding

This work did not receive any specific grant from any funding agency in the public, commercial, or not-for-profit sector.

Patient consent

Written informed consent for publication of their clinical details and/or clinical images was obtained from the patient.

Author contribution statement

J M K de Filette is an endocrinologist-in-training and was the main author. All authors were involved in the clinical care of the patient. All authors contributed to the reviewing and editing process and approved the final version of the manuscript.

References

Filed under: Cushing's, pituitary, Rare Diseases, Treatments | Tagged: , , , , | Leave a comment »

« Previous Page — Next Page »