Long Noncoding RNA Expression in Adrenal Cortical Neoplasms
Oyewale Shiyanbola 1 . Heather Hardin 1 . Rong Hu1 . Jens C. Eickhoff2 . Ricardo V. Lloyd1
Check for updates
Accepted: 22 July 2020
C Springer Science+Business Media, LLC, part of Springer Nature 2020
Abstract
Long noncoding RNAs (lncRNAs) consist of nucleic acid molecules that are greater than 200 nucleotides in length and they do not code for specific proteins. A growing body of evidence indicates that these lncRNAs have important roles in tumorigenesis. Separating adrenal cortical adenomas from carcinomas is often a difficult problem for the surgical pathologist. This is especially true when only small needle biopsies are available for examination. We used in situ hybridization (ISH) analysis to study normal adrenal cortical tissues and adrenal cortical tumors to determine the role of specific lncRNAs in tumor development and classification. The lncRNAS studied included metastasis-associated lung adenocarcinoma transcript 1 (MALAT1), psoriasis susceptibility-related RNA gene induced by stress (PRINS), and HOX antisense intergenic RNA myeloid 1 (HAM1). We constructed a tissue microarray (TMA) for the studies and also analyzed a subset of cases by quantitative reverse transcriptase polymerase chain reaction (qRT-PCR). Two 1-mm duplicate cores of normal adrenal cortex (NAC) (n =23), adrenal cortical adenomas (ACAs) (n =95), and adrenal cortical carcinomas (ACCs), (n =20) were used on the TMA. The results of ISH were analyzed by image analysis. ISH showed predominantly nuclear expression of lncRNAs in adrenal cortical tissues. MALAT1 showed more expression in ACCs than in NAC and ACA (p <0.05). PRINS had higher expression in NACs and ACAs than in ACCs. The lncRNAs MALAT1, PRINS, and HAM1 are all expressed in normal and neoplastic adrenal cortical tissues. MALAT1 had the highest expression in ACC compared to ACAs and may have a role in ACC development.
Keywords Adrenal cortical carcinoma . Long noncoding RNA · In situ hybridization . Immunohistochemistry
Introduction
Adrenal cortical carcinomas (ACCs) are rare steroid- producing endocrine cancers which are very aggressive ma- lignancies. Patients with advanced ACCs such as those with invasive or metastatic disease usually have poor prognosis with 5-year survival rates of 6 to 13% [1, 2]. Recent studies of The Cancer Genome Atlas (TCGA) and others collected the clinical and pathologic features as well as genomic alterations, DNA-methylation profiles and RNA, and proteomic signa- tures of a large group of ACCs [3, 4]. These studies increase the available information about known ACC driver genes. They reported that these genes included PRKAR1A, NF1,
and several others and suggested that disease progression was associated with whole genome doubling [3, 4].
Some reported analyses have shown that long noncod- ing RNA may have some roles in tumor development [5-7]. IncRNAs are RNA transcripts longer than 200 nu- cleotides in length. Many studies have shown that lncRNAs do not code for proteins, so their products can- not be detected by immunohistochemistry, unlike the pro- tein products from messenger RNAs. They may have mul- tiple roles in tissues such as in the regulation of transcrip- tion, in gene silencing, and as possible oncogenes and tumor suppressor genes [5-7]. Recent studies of lncRNAs in ACCs have described specific lncRNA signa- tures in these malignancies. Glover et al. [8] reported that 956 lncRNAs were differentially expressed between ACCs and ACAs, including MALAT1, PRINS, GAS5, and H19. They also reported that PRINS was one of sixty-six lncRNAs associated with ACC recurrence [8]. Other stud- ies of lncRNAs in ACCs have characterized other lncRNAs as having roles in the growth and regulation of ACCs [9-11].
Adrenocortical neoplasms often represent diagnostic prob- lems in surgical pathology, especially in small biopsy
☒ Ricardo V. Lloyd rvlloyd@wisc.edu
1 Department of Pathology and Laboratory Medicine, University of Wisconsin School of Medicine and Public Health, Madison, WI, USA
2 Department of Biostatistics and Medical Informatics, University of Wisconsin School of Medicine and Public Health, Madison, WI, USA
specimen when the surgical pathologist may not be able to diagnose the biopsy as being benign or malignant. We ana- lyzed several lncRNAs in adrenal cortical neoplasm to deter- mine the role of specific lncRNAs in tumor classification.
Materials and Methods
Tissue Microarrays (TMA)
TMAs were made using formalin-fixed paraffin-embedded tissues (FFPE) that were stored in the pathology department [9]. The samples included normal adrenals (NAC, n =23), adrenal cortical adenomas (ACA, n =95), and adrenal cortical carcinomas (ACC, n =20). Two 1.0-mm cores were used for each sample. We received approval from the Institutional Review Board at the University of Wisconsin-Madison for the study.
Immunohistochemistry
Immunohistochemical analyses used an automated stainer (Ventana BenchMark Ultra system, Ventana Medical Systems, Inc., Tucson, AZ). Primary antibodies utilized in- cluded Ki-67 (1:50 dilution with Van Gogh Yellow; Biocare, Pacheco, CA), Calretinin, inhibin alpha, and p53 (from Ventana Medical Systems, Inc., Tucson, AZ) and used prediluted. Steroidogenic Factor-1 (Cell Marque, Rocklin, CA) was used undiluted. Diaminobenzidine (DAB) was used to detect the antigen-antibody reactions. Positive and negative controls were pared with each analysis. Staining was scored on the TMA. The intensity of staining was graded as previ- ously reported [12]. For Ki-67 analysis, a minimum of 500 cells were enumerated in “hot spot” areas, and the proliferative index was calculated for each sample.
In Situ Hybridization (ISH)
Three probes including lncRNAS, metastasis-associated lung adenocarcinoma transcript 1 (MALAT1), psoriasis susceptibility-related RNA gene induced by stress (PRINS), and HOX antisense intergenic RNA myeloid 1 (HAM1) were utilized. The ISH procedure was previously described [12, 13]. Two of the lncRNAs (MALAT1 and PRINS) were select- ed because there were previously described in ACCs recently 8 while HAM1 was used because it was recently shown to be highly expressed in aggressive thyroid carcinomas [13]. A positive control probe, hs-PPIB, and negative control probe, dapB, were used with each experiment. (Advanced Cell Diagnostics). Probe expression levels were visualized with diaminobenzidine (DAB). ☒
Automated Image Acquisition and Analysis
The stained TMA slides were visualized and analyzed with the Vectra slide scanner and associated software (PerkinElmer, Waltham, MA) as previous described [12, 13]. Images were acquired, and then the InForm 1.4.0 software was used to segment tissue compartments separating epithelial and non-epithelial elements as well as the nucleus and the cytoplasm. MALAT1, PRINS, and HAM1 expression levels from each sample was quanti- tated. We used only nuclear signals for the analyses. The lncRNA expression level from the duplicate cores was used to analyze the various cores and for statistical comparisons.
Real-Time PCR
Paraffin blocks were used to make cores, and RNA was ex- tracted from between 5 and 10 samples from each group using TRIzol reagent (ThermoFisher Scientific, Waltham, MA) fol- lowing the manufacturer’s instructions. We measured the op- tical density and RNA quality, and concentrations were assessed with a NanoDrop 1000 spectrophotometer (ThermoFisher Scientific, Waltham, MA) to assess RNA quality and amount of each RNA sample. Several samples were not reactive during the rt-PCR analysis, and the final number of samples available for analysis included 4 NAC, 5 ACA, and 5 ACC. RT-PCR analysis was performed as previ- ously reported [12-14].
Statistics
The analysis of variance (ANOVA) with post-hoc pairwise comparisons were used to compare Ki-67 levels between groups. Analogously, lncRNA expression levels from data collected from RNA ISH and for RT-qPCR, respectively, were compared between groups using ANOVA with pairwise post-hoc comparisons. All reported p values were two-sided and p <0.05 was used to define statistical significance. Quantitative image analysis was performed using Vectra and InForm software. Data are expressed as means ± standard error of the mean (SEM). For the adrenal cortical carcinoma subset, we further assessed the correlation between lncRNA expression and the Ki-67 proliferation index, presence of ab- errant p53 expression, and tumor functional status. Specifically, we utilized the Spearman’s correlation coeffi- cient in assessing the correlation between lncRNA expression and Ki-67 proliferation index, both continuous variables. The point-serial (Pearson) coefficient was utilized in evaluating the correlation between lncRNA expressions and the categorical variables aberrant p53 expression and tumor functional status.
Results
Clinicopathologic Findings
There were equal numbers of male and female patients with ACCs while there was a higher percentage of females (69%) with ACAs (Table 1). Both ACCs and ACA were more com- mon on the left side. Among patients with ACCs, 9 had func- tional tumors, 6 had non-functional tumors, and 5 cases had an unknown tumor functionality. The majority of the ACCs were more aggressive; 6 of the 20 (30%) patients presented with lymph node metastasis, 12/20 (60%) patients had distant me- tastasis, and 8/20 (40%) patients died of ACC while 8 patients were lost to follow-up (Table 1).
Immunohistochemical Analysis
Calretinin and inhibin alpha were expressed in higher percent- ages of ACAs and NAC compared to ACCs (Table 2). Over 90% of NAC and ACAs expressed these two markers, while only 70 to 75% of ACC were positive for inhibin alpha and calretinin, respectively. p53 positive staining was present in 30% of ACC, and in none of the NAC tissues and ACAs. Steroidogenic factor-1 was more highly expressed in NAC
| Carcinoma (n=20) | Adenoma (n=95) | |
|---|---|---|
| Gender, N (%) | ||
| Male | 10 (50.0) | 30 (31.6) |
| Female | 10 (50.0) | 65 (68.4) |
| Age at diagnosis | ||
| Mean (year) | 52.7 | 49.2 |
| Laterality, N (%) | ||
| Left | 12 (60.0) | 56 (59.0) |
| Right | 8 (40.0) | 39 (41.0) |
| Lymph node involvement | ||
| Yes | 6 (30.0) | N/A |
| No | 6 (30.0) | N/A |
| Lymph nodes not submitted | 8 (30.0) | N/A |
| Metastasis | ||
| Yes | 12 (60.0) | N/A |
| No | 6 (30.0) | N/A |
| Unknown | 2 (10.0) | N/A |
| Status, N (%) | ||
| Alive with no disease | 4 (20.0) | 90 (94.7) |
| Alive with disease | 0 | 0 |
| Dead with no disease | 0 | 5 (5.3) |
| Dead with disease | 8 (40.0) | 0 |
| Lost to follow-up | 8 (40.0) | 0 |
| Diagnosis | Biomarkers | ||||
|---|---|---|---|---|---|
| Calretinin % | Inhibin % | p53% | SF- 1% | Ki-67 ( (PI) | |
| Normal adrenal | |||||
| Cortex (N=23) | 90.0 | 91.3 | 0 | 100 | 1.8 +0.17 |
| Adrenal cortical | |||||
| Adenoma (N=95) | 91.3 | 90.0 | 0 | 99.1 | 2.4 +0.22* |
| Adrenal cortical | |||||
| Carcinoma (N=20) | 75.0 | 70.0 | 30 | 70.0 | 21.3 +3.7 *** |
SF steroidogenic factor, PI, proliferative index. Ki-67 was determined by counting at least 500 cells in “hot areas” of the TMA cores. Statistical analysis was done with Student’s t test. * p<0.05, *** p<0.001. N equals number of cases analyzed
and ACAs compared to ACCs (Table 2). The Ki-67 prolifer- ative index was significantly higher in ACCs compared to ACAs and NAC (p <0.001) while the Ki-67 proliferative index in ACAs was significantly higher than NAC (p <0.05, Table 2).
In Situ Hybridization
In situ hybridization (ISH) analysis showed predominantly nuclear expression of MALAT1, PRINS, and HAM1 in the adrenal gland tissues (Fig. 1). The positive control probe PPIB showed mainly diffuse nuclear localization with weaker cytoplasmic staining. The negative control probe resulted in no staining of the tissues (data not shown). Staining in the duplicate cores was relatively consistent. Quantitative image analysis showed significantly higher levels of MALAT1 ex- pression in ACCs compared to ACAs (p <0.05) (Fig. 2a). There was decreased expression of PRINS in ACCs compared to ACAs and NACs, although these differences were not sta- tistically significant (Fig. 2b). Analysis of HAM1 showed slightly increased expression levels in ACCs compared to ACA and NA, and these differences were not statistically significant (Fig. 2c).
qRT-PCR Analysis
qRT-PCR analysis utilizing RNAs extracted from FFPE tis- sues showed the highest level of expression of MALAT1 in ACCs and the lowest level of MALAT1 expression in ACAs (Fig. 3a). Expression of PRINS was decreased in adenomas and carcinomas compared to the normal adrenal cortical tis- sues, but the differences were not statistically significant (Fig.
H&E
MALAT1
PRINS
HAM1
NORMAL
a
b
C
d
ADENOMA
e
f
g
h
.
CARCINOMA
i
j
k
l
3b). qRT-PCR analysis did not show any significant differ- ences in HAM1 expression between the three groups (Fig. 3c). qRT-PCR was used for validation of the ISH method, but the limited number of samples and the technical difficulties of performing qRT-PCR with older paraffin-embedded samples were some of the limitations of this approach.
Correlation Analyses
Though not statistically significant, Ki-67 proliferation index was negatively correlated with MALATI and PRINS (-0.013, p=0.96 and - 0.117; p = 0.62, respectively) and was positive- ly correlated with MALAT1 (0.25; p =0.29) among patients
MALAT1
00
PRINS
4
HAM1
00
8
·
+
10.
0
MALAT1
€?
PRINS
HAMI
4.
+
~
~
2.
-
0
0
0
a
Normal
Adenoma
Carcinoma
b
Normal
Adenoma
Carcinoma
c
Normal
Adenoma
Carcinoma
MALAT-1
Relative Normalized Expression
15
10
5
T
a
0
Normal
Adenoma
Carcinoma
12
PRINS
Relative Normalized Expression
8
6
T
4
2
b
0
Normal
Adenoma
Carcinoma
140
HAM1
Relative Normalized Expression
120
100
80
60
40
20
c
0
Normal
Adenoma
Carcinoma
with ACC using the Spearman’s correlation coefficient. Additionally MALAT1, PRINS, and HAM1 expression in ACCs demonstrated positive correlation with p53 (0.238, 0.117, and 0.003 respectively) and the functional status of the tumors (0,486, 0.499, and 0.316 respectively) using the point serial coefficient.
Discussion
This is the first study to analyze expression of specific lncRNAs in adrenal.
cortical carcinomas by ISH. ISH is a very useful technique in diagnostic pathology, since it is relatively easy to correlate molecular biological findings with morphology which makes it a very powerful tool for surgical pathologist [14]. In addi- tion, ISH can also pinpoint geographical/topographical differ- ences in lncRNA expression across a tumor sample. Our find- ings showed that all three lncRNAs examined were primarily expressed in the nucleus of normal and neoplastic adrenal cortical cells. In this study and in a prior study with thyroid carcinomas [12], PPIB was localized predominantly in the cell nucleus, although in other studies, PPIB was noted predomi- nantly in the cell cytoplasm [15]. These different results sug- gest that PPIB localization may vary with tissue type. Glover et al. [8] first reported that specific lncRNAs including MALAT1, PRINS, GAS5, and H19 were differentially expressed between ACCs and ACA. They used lncRNA ex- pression profiling with lncRNA microarrays to identify lncRNAs that were differentially expressed in ACCs com- pared to ACAs and normal adrenal cortical tissues. They sug- gested that specific lncRNAs could have possible diagnostic uses in the treatment of patients with adrenal cortical tumors. Our findings using ISH support their observations with the lncRNA MALAT1, since expression of this particular lncRNA was significantly increased in ACCs compared to ACA by ISH.
High expression levels of MALAT1 have been reported to be a predictor of metastasis and invasion, and have been
HAM1, but the differences were not significant. PRINS showed decreased levels of expression in ACA compared to NAC and the lowest levels of expression was in ACCs
correlated with poor outcomes and prognosis in various can- cers [16-20]. Although MALATI functions mainly as an on- cogene, recent studies have also shown that it may serve as a tumor suppressor gene in breast and other carcinomas [21, 22]. In addition to its roles as an oncogene and tumor suppres- sor gene, other studies have shown that MALAT1 is also asso- ciated with oncogenic gene fusions in some malignancies. Fusion of MALAT1 gene to the TFEB gene has been reported in subsets of renal cell carcinomas [23] while a MALAT1- GLI1 fusion has been reported in some subsets of plexiform fibromyxomas [24]. MALAT1 locates to chromosome 11q13. Interestingly, loss of heterozygosity in 11q13 has been report- ed to occur frequently in ACCs, but not in ACAs [25, 26] and was not associated with MEN1 mutations [25]. Several genes from the 11q13 region have been reported to be downregulat- ed in ACCs, but not in ACAs suggesting that there may be a tumor suppressor function for some genes in the 11q13 region in adrenal cortical tumors. Accumulating evidence in the lit- erature indicates that MALAT1 may function both as an onco- gene or a tumor suppressor gene in different tissues [16, 18, 22, 22, 27].
PRINS was decreased in ACCs compared to ACAs in our study, but this difference was not significant. PRINS was one of the earlier described lncRNA associated with human dis- eases. Specifically, it was found to be associated with psoriatic skin disease [29]. PRINS is over 2 kb bases long and consists of two exons. It is located on chromosome 10. PRINS is polyadenylated and contains a high density of stop codons, but lacks an extensive open reading frame [28]. PRINS has been shown to regulate G1P3 in keratinocytes of the skin [29]. Szendik et al. [29] showed that in Hela cells, downregulation of the PRINS gene was associated with changes in cell mor- phology and upregulation of multiple protein-coding genes, including some which form part of the Wnt/b-catenin signal- ing pathway. This finding is of interest, because abnormal activation of the Wnt pathway has also been associated with poor outcome in adrenocortical carcinomas. Further research is needed to clarify the role of PRINS and other lncRNAs in adrenocortical carcinomas. PRINS has been studied
extensively in keratinocytes and in inflammatory diseases, but not in many neoplastic diseases [8, 28-31]. Exosomal PRINS may have a possible diagnostic role in monoclonal gammopathy patients [30]. PRINS may also have a regulatory role in the resistance of colorectal carcinoma cells to apoptosis [31].
HAM1 was expressed in NAC, ACA, and ACC by ISH and qRT-PCR in this study. However, there were no significant differences in the expression of HAM1 between the various groups. HAM1 has been reported to be associated with various solid malignancies including pancreatic ductal adenocarci- nomas [32], glioblastoma multiforme [33], and ovarian carci- nomas [34]. In glioblastomas, HAM1 regulated SP1 by spong- ing miR-137 [34], while in ovarian carcinomas HAM1 sup- pressed cell proliferation and invasion by sponging miR- 106a-5p [34]. Our previous studies of thyroid carcinoma also show increased expression of HAM1 in anaplastic thyroid car- cinomas and papillary thyroid carcinomas, compared to other thyroid carcinomas and normal thyroid tissues [13].
Immunohistochemical analysis of the TMA showed signif- icantly higher levels of Ki-67 labeling in ACCs compared to ACAs and NAC. This agrees with recent findings in which higher levels of expression of Ki-67 were observed in ACCs compared to ACAs [35-39]. However, there was some vari- ability in the Ki-67 labeling, since two cases of ACAs from the TMA cores were completely negative for Ki-67. This was probably related in part to the small area (1 mm diameter) of ACC tissues on the TMA. Our findings corroborate earlier reports suggesting that the use of Ki-67 to determine prolifer- ative index can be a useful marker to separating ACCs from ACAs. P53 immunostaining was present only in a subset of ACCs in this study. Other reports have also indicated that subsets of ACCs express p53 and this finding is seldom seen in ACA or NAC. Calretinin was expressed by 75% to 91.7% of adrenal cortical tissues. Earlier studies have found that calretinin is a highly sensitive marker for adrenal cortical tis- sues, although it is not very specific [40]. SF-1 was expressed in a lower percentage of ACCs compared to ACAs. Although recent studies and reviews indicated that SF-1 was an excel- lent marker for ACCs [39-41], and that 98% of ACCs were positive for this biomarker in one study [41], our results with only 70% of ACCs on the TMA staining positively for SF-1 most likely reflects the small portion of tissues from the TMA that was analyzed. In an earlier report, staining of larger tissue sections resulted in a higher percentage of positive staining for SF-1 [41]. In addition, tumors derived from the zona reticularis usually stain more weakly than those from the zona glomerulosa and zona fasciculata [41]. Analysis of ACCs on the TMA with other markers such as Ki-67 and p53 was also limited by the heterogeneity of the carcinomas and the limited amount of tissues available on the TMAs.
There were several limitations to this study. Because only TMAs were used in the analyses of lncRNA expression, we
could not determine if there were regional differences in high- ly heterogeneous tumors such as ACCs; thus, the interpreta- tion of the ISH could vary from the TMA results compared to analysis of whole sections. Another limitation was the rela- tively small number of ACCs available for analysis in this study.
In summary, our study showed that all three lncRNAs ex- amined on the TMA by ISH were expressed in normal and neoplastic adrenal cortical tissues with predominantly nuclear expression. MALAT1 was more highly expressed in ACCs compared to ACAs. PRINs and HAM1 were both expressed in all three groups of adrenal cortical tissues. Although there was decreased expression of PRINS in ACCs compared to ACA and NAC, the differences were not significant. These findings suggest that MALAT1 may have a role in adrenal cortical tumor development.
Acknowledgments The authors thank the University of Wisconsin Translational Research Initiatives in Pathology laboratory (TRIP). Supported by the UW Department of Pathology and Laboratory Medicine, University of Wisconsin Carbone Cancer Center (P30 CA014520) and the Office of The Director-NIH (S10OD023526) for use of its facilities and services. Dr. Shiyanbola received a research grant from the Department of Pathology and Laboratory Medicine at the University of Wisconsin School of Medicine and Public Health.
Compliance with Ethical Standards
The study received clinical approval from the local Institutional Review Board.
Conflict of Interest The authors declare that they have no conflict of interest.
References
1. Else T, Kim AC, Saboloch A, Raymond VM, Kandathil A, et al. Adrenocortical carcinoma. Endocr Rev 35:282-326, 2014.
2. Fassnacht M, Kroiss M and Aliolio B. Update in adrenocortical carcinoma. J Clin Endocrinol Metab 98:4551-4564, 2013.
3. Zheng S, Cherniak AD, Dewal N, et al. Comprehensive pangenomic characterization of adrenocortical carcinoma. Cancer Cell 29:723-736, 2016.
4. Giordano TJ, Kuick R, Else T, et al. Molecular classification and prognostication of adrenocortical tumors by transcription profiling. Clin Cancer Res 15:668-676, 2009
5. Morris KV, Mattick JS. The rise of regulatory RNA. Nature Reviews Cancer 15:423- 437, 2014.
6. Prensner MR, Chinnayan AM. The emergence of lncRNAs in can- cer biology. Cancer Discovery 1:391-407, 2011.
7. Zhang R, Hardin H, Chen J, Guo Z, Lloyd RV. Non-coding RNAs in thyroid cancer. Endocr. Pathol 27:12-20, 2016.
8. Glover AR, Zhao JT, Ip JC, et al. Long noncoding RNA profiles of adrenocortical cancer can be used to predict recurrence. Endocrine- Related Cancer 22: 99-109, 2015.
9. Yan Z-C, He L, Qui J-H, et al. LncRNA HOTAIR participates in the development and progression of adrenocortical carcinoma via reg- ulating cell cycle. European Review for Medical and Pharmacological Sciences 22:6640-6649, 2018.
10. Guo N, Sun Q, Fu D, Zhang Y. Long non-coding RNA UCA1 promoted the growth of adrenocortical cancer cells via modulating the miR-298-CDK6 axis. Gene 703: 26-34, 2019.
11. Buishand FO, Liu-Chittenden Y, Fan Y, et al. Adrenocortical tu- mors have a distinct long non-coding RNA expression profile and LINC00271 is downregulated in malignancy. Surgery 167: 224- 232, 2020.
12. Zhang R, Hardin H, Huang W, Buehler D, Lloyd RV. Long Non- coding RNA Linc ROR is upregulated in papillary thyroid carcino- ma. Endocr Pathol 29:1-8, 2018.
13. Wang Y, Hardin H, Chu Y-H, Esbona K, Zhang R, Lloyd RV. Long Non-coding RNA expression in anaplastic thyroid carcino- mas. Endocr. Pathol 30:262-269, 2019.
14. Chu YH, Hardin H, Zhang R, Guo Z, Lloyd RV. In situ hybridiza- tion. Introduction to techniques, applications and pitfalls in the per- formance and interpretation of assays. Semin Diagn Pathol 36: 336- 341, 2019.
15. Didiot M-C, Ferguson CM, Ly S et al. Nuclear localization of Huntington mRNA is specific to cells of neuronal origin. Cell Rep 24, 2553-2560, 2018.
16. Zhao M, Wang S, Li Q, Qing J, Guo P, Liu X. MALAT1: A long non-coding RNA highly associated with human cancers. Oncol Lett 16: 19-26, 2018.
17. Sun Y, Ma L. New insights into long non-coding RNA MALAT1 in cancer and metastasis. Cancers (Basel) 11:216, 2019. https://doi. org/10.3390/cancers 11020216.
18. Hirata H, Hinoda Y, Shahryari V, et al. Long noncoding RNA MALAT1 promotes aggressive renal cell carcinoma through Ezh2 and interacts with miR-205. Cancer Res 75: 1322-1331, 2015.
19. Zhang R, Hardin H, Huang W, et al. MALAT1 long non-coding RNA expression in thyroid tissues: Analysis by in situ hybridization and real-time PCR Endocr Pathol 28: 7-12, 2017.
20. Chu YH, Hardin H, Schneider DF, et al. MicroRNA-21 and long non-coding RNA MALAT1 are overexpressed markers in medullary thyroid carcinoma. Exp Mol Pathol 103:229-236, 2017.
21. Kim J, Piao HL, Kim BJ, et al. Long noncoding RNA MALAT1 suppresses breast cancer metastasis. Nat Genet 50:1705-1715, 2018.
22. Kwok ZH, Roche V, Chew XH, et al. A non-canonical tumor sup- pressive role for the long non-coding RNA MALAT1 in colon and breast cancers. Int J Cancer 143(3):668-678, 2018.
23. Argani P, Yonescu R, Morsberger L, et al. Molecular confirmation of t(6;11)(9p21;q12) carcinoma in archival paraffin-embedded ma- terial using a break-apart TFEB FISH assay expands its clinicopath- ologic spectrum. Am J Surg Pathol 36:1516-1526, 2012.
24. Spans L, Fletchem CD, Antonescu CR, et al. Recurrent MALAT1- GLI1 oncogenic fusion and GLI1 up-regulation define a subset of plexiform fibromyxoma. J Pathol 239:335-343, 2016.
25. Kjellman M, Roshani L, Tea BT. Genotype of adrenocortical tu- mors: very frequent deletions of the MEN1 locus in 11q13 and of a 1-centomorgan region in 2p16. J Clin Endocrinol Metab 84: 730- 735, 1999.
26. Fernandez-Ranvier GG, Wey J, Yeh R-F. Candidate diagnostic markers and tumor suppressor genes for adrenocortical carcinoma by expression profile of genes on chromosome 11q13. World J Surg 32:873-881, 2008.
27. Chen Q, Zhu C, Jin Y. The oncogenic and tumor suppressive func- tions of the long noncoding RNA MALAT1: an emerging
controversy. Frontiers in Genetics. 11:93, 2020.doi: https://doi. org/10.3389/fgene.2020.00093.
28. Sonkoly E, Bata-Csorgo Z, Pivarcsi A, et al. Identification and characterization of a Novel psoriasis susceptibility-related noncod- ing RNA gene, PRINS. J Biol Chem 280:24159-24167, 2005.
29. Szendik M, Sonkoly E, Nagy N, et al. The anti-apoptotic protein G1P3 is overexpressed in psoriasis and regulated by non-coding RNA, PRINS. Exp Dermatol 19:269-278, 2010.
30. Sediarikova L, Bollova B, Radova L, et al. Circulating exosomal long noncoding RNA PRINS- first findings in monoclonal gammopathies Hematol Oncol 36: 786-791, 2018.
31. Hanisch C, Sharbati J, Kutz-Lohroff B. TTFE3-dependent resis- tance of colorectal adenocarcinoma cells HT-29/B6 to apoptosis is mediated by miR-491-5p regulation of lncRNA PRINS. Cell Death Discov 3:16106, 2017 doi:https://doi.org/10.1038/cddiscovery. 2016.eCollection.
32. Luo Y, He Y, Ye X, et al. High expression of long noncoding RNA HOTAIRM1 is associated with the proliferation ad migration in pancreatic ductal adenocarcinoma. Pathol Oncol Res 25:1567- 1577, 2019.
33. Hao Y, Li X, Chen H, et al. Over-expression of long noncoding RNA HOTAIRM1 promotes cell proliferation and invasion in hu- man glioblastoma by up-regulating SP1 via sponging miR-137. Neuroreport 31:109-117, 2020.
34. Chao H, Zhang M, Hou H, et al. HOTAIRM1 suppresses cell pro- liferation and invasion in ovarian cancer through facilitation ARHGAP24 expression by sponging miR 106a-5p. Life Sci 243: 117296. https://doi.org/10.1016/j.lfs.2020117296.Epub 2020.
35. Choi YM, Min HK, Sung T-Y, et al. Clinicopathological features associated with the prognosis of patients with adrenal cortical car- cinoma: Usefulness of the Ki-67 index. Medicine (Baltimore) 95: e3736, 2016.
36. Fassnacht M, Dekkers O, Else T, et al. European Society of Endocrinology Clinical Practice. Guidelines on the management of adrenocortical carcinoma in adults, in collaboration with the European Network for the study of adrenal tumors. Eur J Endocrinol 179 G1-G46, 2018.
37. Papathomas TG, Pucci E, Giordano TJ, et al. An international Ki67 reproducibility study in adrenal cortical carcinoma. Am J Surg Pathol 40:569-576, 2016.
38. Nakamura Y, Yamazaki Y, Feliziola SJ, et al. Adrenocortical car- cinoma. Endocrinol Metab Clinics 44: 399-410, 2015.
39. Mete O, Asa SL, Giordano TJ, et al. Immunohistochemical bio- markers of adrenal cortical neoplasms Endocr Pathol 29: 137-149, 2018.
40. Sangoi AR, Fujiwara M, West RB, et al. Immunohistochemical distinction of primary adrenal cortical lesions from metastatic clear cell renal cell carcinoma: A study of 248 cases. Am J Surg Pathol 35: 578-586, 2011.
41. Sbiera S, Schmull S, Hans-Ullrich Voelker A et al. High diagnostic and prognostic value of steroidogenic factor-1 expression in adrenal tumors. J Clin Endocrinol Metab 95: E161-171, 2010.
Publisher’s Note Springer Nature remains neutral with regard to jurisdic- tional claims in published maps and institutional affiliations.