N-type calcium channel v2.2 is a target of TCF21 in adrenocortical carcinomas

Barbara dos Santos PASSAIA’, Jean Lucas KREMER1, Maria Candida Villares FRAGOSO2, Claudimara Ferini Pacicco LOTFI1,*

‘Department of Anatomy, Institute of Biomedical Science, University of São Paulo, São Paulo, Brazil; 2Adrenal Unit, Hormone and Molecular Genetic Laboratory/LIM42, Hospital of Clinics, School of Medicine; University of São Paulo, São Paulo, Brazil

*Correspondence: clotfi@usp.br

Received January 6, 2022 / Accepted May 12, 2022

Transcription factor 21 (TCF21) directly binds and regulates SF1 mRNA expression in tumor and normal adrenocortical cells, and both are involved in the development and steroidogenesis of the adrenal cortex. TCF21 is a tumor suppressor gene and its expression is reduced in malignant tumors. In adrenocortical tumors, it is less expressed in adrenocortical carcinomas (ACC) than in adrenocortical adenomas (ACA) and normal tissues. However, a comprehensive analysis to identify TCF21 targets has not yet been conducted in any type of cancer. In this study, we performed Chromatin Immuno- precipitation and Sequencing (ChIP-Seq) in an adrenocortical carcinoma cell line (NCI-H295R) overexpressing TCF21, with the aim of identifying TCF21 new targets. The five most frequently identified sequences corresponded to the PRDM7, CNTNAP2, CACNA1B, PTPRN2, and KCNE1B genes. Validation experiments showed that, in NCI-H295R cells, TCF21 negatively regulates the expression of the CACNA1B gene. Recently, it was observed that the N-type calcium channel v2.2 (Cav2.2) encoded by the CACNA1B gene is important in Angiotensin II signal transduction for corticosteroid biosynthesis in NCI-H295R adrenocortical carcinoma cells. Indeed, TCF21 inhibits CACNA1B and Cav2.2 expression in NCI-H295R. In addition, in a cohort of 55 adult patients with adrenocortical tumors, CACNA1B expression was higher in ACC than ACA and was related to poor disease-free survival in ACC patients. These results suggest a mechanism of steroidogenesis control by TCF21 in adrenocortical tumor cells, in addition to the control observed through SF1 inhibition. Importantly, steroid production could impair tumor immunogenicity, contributing to the immune resistance described in adrenal cancer.

Key words: TCF21, adrenal cancer, adrenocortical carcinoma, N-type calcium channel, adrenocortical cell cultures

Transcription factor 21 (TCF21) is considered a tumor suppressor in several types of cancer because of its low expres- sion in hepatocellular carcinoma, melanoma, lung, and head and neck carcinomas, and kidney cancer [1-4]. In breast cancer cells, TCF21 reduces migration and invasion capacity, proliferation and angiogenesis, and promotes apoptosis [5]. Similar effects have been observed in colorectal cancer cell lines and gastric cancer cells [6, 7]. In bladder cancer, TCF21 is more highly expressed in primary cancer samples than in metastatic samples [8]. In adrenocortical tumors, TCF21 shows lower expression in adrenocortical carcinomas (ACC) than in adrenocortical adenomas (ACA) and normal adrenal glands [9, 10]. TCF21 gene expression levels are extremely low in ACCs [11]. Adult ACCs are rare but aggressive cancers, whereas ACAs are more common and with favor- able outcomes [12, 13]. In children, adrenocortical tumors (ACT) generally show better evolution, especially when diagnosed early [14]. Thus, ACTs have a different prognosis in children and adults. In addition, the combined expres- sion of TCF21 and BUB1B (BUB1 mitotic checkpoint serine/

threonine kinase B) was a good prognostic marker for adult patients with ACC, distinguishing two subgroups of patients with different survival times [11]. In pediatric patients with ACT, the combined expression of TCF21 and SF1/NR5A1 (steroidogenic factor 1/nuclear receptor subfamily 5, group A, member 1) presented diagnostic value for pediatric patients aged <5 years [11]. TCF21 binds directly to the SF1 promoter, inhibiting its activity and steroidogenic acute regulatory protein (StAR) expression in adrenocortical carci- noma cell lines [10]. However, the mechanisms of action of TCF21 are barely known and its targets are still emerging. In this study, 70 target sequences regulated by TCF21 in H295R cells modified to overexpress TCF21 (H295R/ TCF21) were identified through Chromatin Immunoprecipi- tation and Sequencing (ChIP-Seq). The five most frequently identified sequences corresponding to PRDM7 (PR/SET domain 7), CNTNAP2 (contactin associated protein 2), CACNA1B (calcium voltage-gated channel subunit alpha 1B), PTPRN2 (protein tyrosine phosphatase receptor type N2), and KCNE1B (potassium voltage-gated channel

subfamily E regulatory subunit 1B) genes. The CACNA1B gene and its product Cav2.2 were analyzed in H295R/TCF21 cells since their expression plays a significant role in the synthesis of aldosterone and cortisol in adrenocortical cells [15]. In H295R/TCF21 cells, we observed the inhibition of CACNA1B and Cav2.2 expression and in samples from adult patients with ACT, CACNA1B was more highly expressed in ACC than in ACA. In addition, ACC patients expressing high CACNA1B levels showed poor disease-free survival. Our results showed that CACNA1B is a target of TCF21 in ACC and suggest a mechanism of steroidogenesis control by TCF21 through the inhibition of N-type Cav2.2.

Patients and methods

Cell cultures. The human adrenocortical carcinoma cell line (NCI-H295R) [16] was obtained from ATCC (The ATCC Cell Biology Collection). NCI-H295R cells were cultured in an RPMI medium with 2% fetal bovine serum (FBS) and 1% insulin transferrin selenium (ITS, Gibco, MA, USA). NCI-H295R cells were transiently transfected with pCMVMycPod1, hereafter called H295R/TCF21, or empty vector pCMVMyc, hereafter called H295R empty, as described in Passaia et al. [11]. The human hepatocellular carcinoma cell line (HepG2) [17] was cultured in a DMEM medium with 10% FBS (Gibco, Grand Island, NY, USA). Patient secondary cell cultures from aldosterone-producing adult adenoma, ACA-T23 cells, non-functioning adult carci- noma, ACC-T227 cells, pediatric carcinoma, ACC-PedT218 cells, and a pediatric adenoma, ACA-PedT7 cells were utilized and showed in Supplementary Table S1. Patient cell cultures were obtained as described in Almeida et al. [18]. These cell cultures are called secondary cell cultures because they were utilized until the fourth or sixth passage and they were not immortalized. Mycoplasma contamina- tion was excluded by using the qPCR Mycoplasma test kit (PanReac AppliChem, ITW Reagents, Chicago, IL, USA). The cultures belong to the biorepository approved by the Human Research Ethics Committee of the Institute of Biomedical Sciences, University of São Paulo (USP) (no. 156.12). Patient cell cultures were authenticated by STR DNA profiling analysis. All cultures were kept at 37°℃ in a humidified atmosphere of 95% air/5% CO2.

Patients. Samples of ACT were obtained from 70 individ- uals, 55 adult (19.0-72.4 years old) and 15 pediatric (0.9-17.7 years old) patients. The clinical and histological charac- teristics of the patients are summarized in Table 1. These data were obtained from Hospital das Clinicas at School of Medicine of the University of São Paulo, responsible for the evaluation of the patients. The human normal adrenal gland used as an immunoblotting control was obtained from one adult individual without disease related to the adrenal gland, and the adrenal cortex was dissected and lysed as described for immunoblotting. This study was approved by the Human Research Ethics Committees of Hospital das Clinicas and

Table 1. Clinical characteristics of patients.
Adults	Characteristics	n=55		%
	Mean age (y)		41.4±14.5	–
	Range (y)		19-72	–
	Sex	Female	41	74.5
		Male	14	25.5
	Diagnosis	ACA	29	52.7
		ACC	26	46.3
	Weiss score	<3	29	52.7
		≥3	26	47.3
	Metastasis		25	45.5
	CRD		17	30.9
	Follow-up (mo)		62.7±71.9	–
Pediatric		n=15
	Mean age (y)		4.1±5.2	–
	Range (y)		0.9-17.7	–
		Female	12	80.0
	Sex	Male	3	20.0
	Diagnosis	ACA	9	60.0
		ACC	6	40.0
	Weiss score	<3	4	26.7
		≥3	11	73.3
	Metastasis		5	33.3
	CRD		2	–
	Follow-up (mo)		80.9±63.7	–

Abbreviations: y-years; mo-months; CRD-cancer-related death; ACA- adenomas; ACC-carcinomas

Institute of Biomedical Sciences (no. 822/2016), São Paulo, Brazil. Written informed consent was obtained from all patients and/or their legal guardians. The Cancer Genome Atlas (TCGA) program data were obtained from the cBio Portal for Cancer Genomics (https://www.cbioportal.org/, accessed on November 22, 2020).

CRISPR/dCas9 activation system. The lentiviruses were produced in HEK-293FT cells using the plasmids sgRNA (MS2) (#61427; Addgene, Cambridge, MA, USA), dCas9-VP64 (#61425; Addgene, Cambridge, MA, USA) and MS2-P65-HSF1 (#61426; Addgene, Cambridge, MA, USA). HepG2 CRISPR/dCas9/TCF21 cells were transduced with the constructs sgRNA TCF21 (sequence AAGGGGTCTA- AGCGCTTTGC) and sgRNA scramble (SCR) (sequence GCACTACCAGAGCTAACTCA), as described in Koner- mann et al. [19]. Transduction and cell selection were performed as described in Passaia et al. [11].

ChIP-Seq. Approximately 9.0×106 cells were plated for 24 h before transfection with plasmid pCMVMycPOD1 and maintained for 48 h after transfection. Chromatin immuno- precipitation was performed with 100 ul of fragmented chromatin, protein G magnetic beads, and 5 µg of anti-TCF21 antibody ChIP Grade (ab32981, Abcam, Cambridge, UK) using the ChIP-IT Express kit (Active Motif, Carlsbad, CA, USA) following the manufacturer’s instructions. Chromatin was fragmented by sonication with 6 pulses of 10 s each at 25%

amplitude using a VCX130PB ultrasonic processor (Sonics & Materials, Newtown, CT, USA). Libraries were prepared using the TruSeq® ChIP Sample Preparation kit (Illumina, San Diego, CA, USA), starting with 7 ng of enriched ChIP DNA and selecting 250 bp fragments on average. Sequencing was performed on the NextSeq 500 System (Illumina, San Diego, CA, USA) with 75 bp single-ended reads. Fragmented DNA without immunoprecipitation was used as a positive control, and two replicates were sequenced. Reads were mapped using GRCh38 homo sapiens (human) genome assembly.

The Phred Quality Score (Q>30) of the bases was verified, low-quality and duplicate readings were removed, and Peak calling was performed using MACS2 (v2.1.1) (p<0.01) wrapped in the AQUAS ChIP-Seq pipeline (https://github. com/NHLBI-BCB/TF_chipseq_pipeline, accessed on November 22, 2020) [20]. The IDR <0.05 (Irreproducible Discovery Rate) was calculated to identify consistent peaks, which guarantees the reproducibility of the experiment [21]. All readings considered as noise by the blacklist of the Encyclopedia of DNA Elements (ENCODE) were excluded [22]. ChIP-Seq data are available in the database Gene Expression Omnibus (GSE179308).

qPCR and immunoblotting. Total RNA was extracted from 9.0×105 cells using TRIzol reagent (Invitrogen, Carlsbad, CA, USA) 48 h after plating. The RNA quality was assessed by agarose gel electrophoresis (2%). RNA quantification and sample contaminants were determined by spectrometry (NanoDrop 2000c, Thermo Fisher Scien- tific, Waltham, MA, USA). For cDNA synthesis, 4 µg of total RNA and M-MLV reverse transcriptase (Invitrogen, Carlsbad, CA, USA) were used. qPCR experiments were conducted in a 7500 Real-Time PCR System (Applied Biosystems, Foster City, CA, USA) using SybrGreen reagents (Invitrogen, Carlsbad, CA, USA); the primers are shown in Supplementary Table S2. A total RNA commercial normal adrenal pool from 5 different donors (BioChain, Newark, CA, USA) was used for qPCR control. The relative expression was calculated by the mean of two endogenous genes: beta-actin (ACTB) and beta-glucuronidase (GUSB). The analysis of gene expression was in triplicate, with the empty vector represented in the figures with the value of 1. For immunoblotting, 9.0×105 cells were plated and, after 48 h, the cells were lysed in radioimmunoprecipitation assay buffer (RIPA buffer) containing 50 mM Tris-HCl, 150 mM NaCl, 1% Nodidet (NP40), 0.5% sodium deoxycholate, and 0.1% sodium dodecyl sulfate at pH7.5, plus a cocktail of protease and phosphatase inhibitors (Sigma-Aldrich, Saint Louis, MO, USA). After pelleting at 14,000xg for 15 min at 4℃, the samples were quantified using the Bradford assay. 15 µg of protein were resolved in 15% SDS-PAGE gels or 4-15% SDS-PAGE gradient gels for TCF21 and Cav2.2, respectively. Nitrocellulose membranes were used for protein transfer, monitored by Ponceau staining. The membranes were blocked with 5% fat-free milk and Tris-

buffered saline (TBS) solution containing 1% Tween 20 (TBST) for 2 h and then incubated with anti-ß actin (1:2000, Santa Cruz Biotechnology, Dallas, TX, USA) or anti-TCF21 (1:1000, sc-377225, Santa Cruz Biotechnology, Dallas, TX, USA), or anti-CACNA1B (1:200, ACC-002, Alomone Labs, Jerusalem, Israel). Antibodies were detected with horse- radish peroxidase-conjugated secondary antibody and chemiluminescence using ECL reagents in an Imager 600 (GE Healthcare Life Sciences, Marlborough, MA, USA). Immunoblots were quantified by densitometry using ImageJ software [23] from three different experiments.

Viability assay. 2.0-4.0×104 cells/well were plated into 96-well plates and, after 24 h, were treated with 100 nM MVIIA w-Conotoxin, as described by Bleackman et al. [24]. @-Conotoxin MVIIA is a specific Cav2.2 blocker [25]. A blockade occurs after 2 min of treatment with 100 nM w-Conotoxin and is not reversible [15, 24, 26]. After 3 min of w-Conotoxin treatment, 20 ul of tetrazolium compound (Cell Titer MTS reagent) (Promega, Madison, WI, USA) were added to 100 ul of culture medium for 4 h at 37°℃ in a humidified atmosphere of 95% air/5% CO2. The optical density was read at 490 nm using an ELISA plate reader (Bio-Tek Instruments, Winooski, VT, USA).

Statistical analysis. Statistical significance was deter- mined using the GraphPad Prism 9 software. Non-parametric Kruskal-Wallis, Mann-Whitney test, and Spearman’s rank correlation coefficient were used to analyze patient samples since these groups do not have a normal distribution. Statistical analyzes of cell culture data that have a normal distribution were used parametric one-way ANOVA and two-sample Student’s t-test. One-sample Student’s t-test was used to analyze fold-change expression after TCF21 overexpression compared to the empty control set to 1.0. Gene expression of the patients was calculated relative to a commercial normal adrenal pool (BioChain Institute, Newark, CA, USA). Results were considered statistically significant when p<0.05.

Results

Seventy TCF21 target sequences were identified through ChIP-Seq. The ChIP-Seq assay was performed using H295R cells overexpressing TCF21 with a coverage of 25±16% of the entire human genome. Seventy TCF21 target sequences were identified, corresponding to 49 different ensembled genes for which the reproducibility information of the duplicates was checked using IDR <0.05 [21]. The gene sequences identified were ranked by a peak score. The peak score was obtained by the frequency that the peaks were called (Supplemen- tary Table S3). Each peak was identified from the transcrip- tion start site (TSS), corresponding to one sequence with its respective Ensembl code. Ten different functions of the TCF21 target genes were identified according to the Gene Ontology Resource (http://geneontology.org/, accessed on November 22, 2020), as shown in Supplementary Table S4.

Validation of the five most frequent targets of TCF21. The five most frequent genes identified as targets of TCF21 were PRDM7, CNTNAP2, CACNA1B, PTPRN2, and KCNE1B (Supplementary Table S3). The expression of these genes was analyzed in H295R cells overexpressing TCF21 (H295R/TCF21) in comparison with H295R cells transfected with the empty vector using qRT-PCR. TCF21 cells regulated gene expression positively in H295R cells by 1.89±0.23-fold for PRDM7 (p=0.0180, T=3.871, DF=4), 5.33+1.5-fold for CNTNAP2 (p=0.0686, T=2.787, DF=3), and 1.58±0.18-fold for KCNE1B (p=0.0511, T=3.155, DF=3), however, only the

expression of the PRDM7 gene was statistically significant (Figures 1A-1C). There was no difference in the expression of the PTPRN2 gene between H295R/TCF21 and control cells (Figure 1D). In contrast, TCF21 transfection signifi- cantly inhibited CACNA1B gene expression to 0.36±0.04- fold (p=0.0037, T=16.39, DF=2, Figure 1E). The analysis of N-type calcium channel v2.2 protein (Cav2.2) expression, which is coded by the CACNA1B gene, was also inhibited in H295R cells overexpressing TCF21 (Figure 2).

Cav2.2 expression is higher in adult and pediatric adrenocortical carcinoma cell cultures. Since Giordano

Figure 1. Validation of the five most frequent targets of TCF21 in H295R pCMVMycTCF21 cells. A) mRNA levels of PRDM7 gene, n=5; B) CNTNAP2 gene, n=4; C) KCNE1B gene, n=4; D) PTPRN2 gene, n=3; E) CACNA1B gene, n=3. The results are shown as the mean of 3-5 independent experiments. The p values are indicated in the graph as *p<0.05 and ** p<0.01. Statistical significance was assessed by paired t-tests.

CNTNAP2 mRNA relative expression

PRDM7 mRNA relative expression

empty

TCF21

empty

TCF21

KCNE1B mRNA relative expression

2.5

PTPRN2 mRNA relative expression

1.5

2.0

1.5

1.0

0.5

0.0

empty

TCF21

0.0

empty

TCF21

CACNA1B mRNA relative expression

1.5

1.0

0.5

0.0

empty

TCF21

et al. [9] described that the CACNA1B gene showed lower expression in samples of normal adrenal cortex than in ACC samples, we analyzed the level of Cav2.2 expression by immunoblotting in a normal adrenal cortex and in adreno- cortical adenoma carcinoma cell cultures. In relation to the adult normal adrenal cortex, H295R and ACC-T227 cells showed significantly increased Cav2.2 expression. Compar- ison of Cav2.2 expression of adenoma ACA-T23 cells and normal adrenal cortex showed no statistically significant difference. In pediatric carcinoma cell culture, ACCPed- T218 cells, Cav2.2 was more highly expressed compared to adenoma cell culture, ACAPed-T7 cells (Figure 3, p<0.0001, F (5, 12)=44.90).

Cav2.2 blockade does not alter the viability of adreno- cortical carcinoma cells. H295R and ACC-T227 cells were treated with @-Conotoxin MVIIA, a specific and irrevers- ible blocker for N-type Cav2.2 [25]. In both H295R and ACC-T227 cells, the blockage of N-type voltage-gated Cav2.2 showed no difference in cell viability of treated cells compared with the control cells (Figures 4A, 4B).

Figure 2. TCF21 expression decreased Cav2.2 expression in H295R pC- MVMycTCF21 cells. A) Western blot analysis of TCF21 expression in H295R pCMVMycTCF21 cells (TCF21) and in H295R pCMVMyc con- trol cells (empty). p=0.002, T=6.788, DF=4. Points represent relative TCF21/ß-actin expression among samples; B) Western blot analysis of relative Cav2.2 expression in H295R pCMVMycTCF21 cells (TCF21) and in H295R control cells (empty), n=3. p=0.0009, T=8.811, DF=4. Points represent relative Cav2.2/ß-actin expression among samples. The results are shown as three independent experiments and their mean. The p val- ues are indicated in the graph as ** p<0.01 and *** p<0.001. Statistical sig- nificance was assessed by the unpaired t-test.

2.5

TCF21/B actin protein expression

2.0

23 kDa

TCF21

1.5

45 kDa

ß actin

empty

TCF21

1.0

0.5

0.0

empty

TCF21

Ca v2.2 /ß actin protein expression

2.5

2.0

240 kDa

Ca v2.2

1.5

45 kDa

ß actin

empty

TCF21

1.0

0.5

0.0

empty

TCF21

PRDM7 and CACNA1B genes are more expressed in adult ACC than in adult ACA. As PRDM7 and CACNA1B genes are significantly altered in H295R/TCF21 cells, the expression of these genes was analyzed in a cohort of adult and pediatric adrenocortical tumors (Figure 5). Analysis of PRDM7 and CACNA1B genes showed significantly higher expression of both genes in adult ACC than in ACA (Figures 5A, 5B, p=0.0223 and p=0.0027, respectively). In pediatric patients with ACA and ACC, no differences were observed between the expression of PRDM7 and CACNA1B (Figures 5C, 5D). Although pediatric patients with carci- nomas present two subgroups of expression of the PRDM7 and CACNA1B genes, no clinical characteristic analyzed defines this division.

CACNA1B expression correlates positively with tumor size in pediatric ACC. In pediatric carcinomas, CACNA1B mRNA expression was positively correlated (r=0.98; p=0.0333) with the tumor size of pediatric carcinomas (Figure 6A), whereas in adult ACC the correlation was negative and moderate (r =- 0.48; p=0.0327) (Figure 6B).

Figure 3. Cav2.2 expression in adrenocortical tumor cell cultures. West- ern blot analysis for Cav2.2 in total extracts from one adult normal ad- renal cortex (Normal Adrenal), the NCI-H295R cell line, adult adreno- cortical carcinoma cells (ACC-T227), adult adrenocortical adenoma cells (ACA-T23), pediatric adrenocortical carcinoma cells (ACCPed-T218), and pediatric adrenocortical adenoma cells (ACAPed-T7). Points repre- sent relative Cav2.2/ß-actin expression among samples. The results are shown as three independent experiments and their mean. The p-values are indicated in the graph as *p<0.05, ** p<0.01, and *** p<0.001. Sta- tistical significance was assessed by Tukey's multiple comparisons test. NS=not significant in relation to the normal adrenal cortex.

240 kDa

Ca v2.2

45 kDa

ß-actin

N. Adrenal

H295R

ACC-T227

ACA-T23

ACC Ped-T218

ACA-Ped-T7

1.5

Ca v2.2 / B-actin expression

1.0

0.5

0.0

H295R

ACC-T227

ACA-T23

ACCPed-T218

ACAPed-T7

Figure 4. Cav2.2 blockade and viability of adrenocortical carcinoma cells. Dose-response cytotoxicity was analyzed through the methyl thiazoletet- razolium (MTT) assay, measured by optical density (OD) at 490 nm for five days: A) NCI-H295R and B) ACC-T227 cells treated with 100 nM w-Conotoxin MVIIA (w-Cono) or untreated (control). The values are ex- pressed as the percentage of viable cells for each condition relative to un- treated controls. The results are shown as the mean of three independent experiments. Statistical significance was assessed by analysis of variance (ANOVA).

H295R cells

control

OD 490nm

@-Cono

0.1

days

ACC-T227 cells

OD 490nm

0.1

days

High CACNA1B expression was related to poor disease- free survival in adult ACC. Analysis of CACNA1B expres- sion levels in adult patients with ACC showed that higher CACNA1B expression was related to poor disease-free survival compared with patients with ACC with reduced CACNA1B expression (Figure 7). Patients with ACC with high CACNA1B expression showed metastasis 8.5 months after diagnosis on average. This result suggests that CACNA1B may be related to tumor aggressiveness.

Discussion

We identified seventy TCF21 target sequences belonging to 49 different genes using the ChIP-Seq assay in an adreno- cortical carcinoma cell line overexpressing TCF21. The five

most frequently identified sequences corresponded to the PRDM7, CNTNAP2, CACNA1B, PTPRN2, and KCNE1B genes. Validation experiments showed that PRDM7 expres- sion was positively regulated in adrenocortical carcinoma cells expressing TCF21 and was more highly expressed in adult ACC than in ACA. The PRDM family of proteins is defined based on the conserved N-terminal PR domain, which is closely related to the domains of histone methyl- transferases [27]. Although PRDM genes play diverse roles in cell cycle regulation, differentiation, and meiotic recom- bination, these genes present opposite roles in carcinomas [27, 28]. In ACC, the positive relation between TCF21 and PRDM7 expression has not been described so far and, to the best of our knowledge, there are no data regarding the role of this gene in the pathogenesis of adrenocortical tumors. Therefore, the understanding of the potential and differential regulation of PRDM7 by TCF21 in ACA and ACC should be more comprehensively addressed.

The CACNA1B gene, the third most frequent target of TCF21 in the H295R cell line, encodes an N-type voltage- dependent calcium channel, the Cav2.2 protein. CACNA1B gene and Cav2.2 protein were negatively regulated by TCF21 in the adrenocortical cell line and more highly expressed in adult ACC than in ACA. These results are in agreement with the findings of Giordano et al. [9], who showed that the CACNA1B gene showed lower expression in samples of normal adrenal cortex than in ACC samples. As TCF21 is expressed to a lesser degree in adult ACC than in ACA and normal adrenals [9-11], these results suggest that TCF21 levels are negatively related to CACNA1B in adult ACC. Moreover, high CACNA1B gene expression in carcinomas was associated with poor disease-free survival, although Cav2.2 channel blockade alone did not affect cell viability in vitro. Even though no difference was observed between pediatric ACA and ACC survival time, CACNA1B expression was positively correlated with tumor size in these patients. As small tumor size is one of the characteristics associated with favorable disease prognosis in this age group [29], this obser- vation may be relevant. The CACNA1B gene encoding Cav2.2 is located on chromosome 9q34.3 (http://www.ensembl.org/, accessed on November 22, 2020). The Cav2.2 expression was higher in pediatric carcinoma cell culture ACC-Ped218 than in adenoma culture, ACA-T7Ped cells. In contrast, the aldosterone-producing adenoma cell culture, ACA-T23, expressed lower Cav2.2 levels than H295R cells. Juhlin et al. [30] described that in aldosterone-producing adenomas ion channels are frequently mutated, and CACNA1B mutations showed a prevalence of 43%. However, according to the Catalog of Somatic Mutations in Cancer (COSMIC), the c.2359C>A (p.L787M) pathogenic mutation in CACNA1B was described only in one ACC sample [31]. Interesting, in 90% of pediatric adrenocortical tumors, there is an amplifi- cation of chromosome 9q in a region that includes NOTCH and SF1 genes [32]. Since Cav2.2 was more highly expressed in ACCPed-T218 carcinoma cells than in ACAPed-T7

Figure 5. Analysis of CACNA1B and PRDM7 expression in a cohort of adult and pediatric adrenocortical tumors. A) Relative expression of CACNA1B in adult ACA and ACC; B) Relative expression of PRDM7 in adult ACA and ACC; C) Relative expression of CACNA1B in pediatric ACA and ACC and D) Relative expression of PRDM7 in pediatric ACA and ACC, by quantitative real-time PCR. Adult ACA=26; Adult ACC=25; pediatric PedACA=9 and pediatric PedACC=6. The Y-axis shows the fold increase in gene expression relative to the endogenous expression relative to a commercial mRNA pool of normal adrenals. The Mann-Whitney test was used to assess statistical significance (p<0.05).

215

CACNA1B mRNA relative expression

PRDM7 mRNA relative expression

210

2º

2-5

2-10

ACA

ACC

2-10

ACA

ACC

215

CACNA1B mRNA relative expression

210

2º

2-5

2-10

ACA Ped ACC Ped

2-10

ACA Ped ACC Ped

adenoma cells, CACNA1B gene expression may be important in pediatric adrenocortical tumors. In addition, it is largely accepted that children have a better outcome than adoles- cents. In a study by Wieneke and collaborators [33] using the pediatric ACT, biphasic age distribution was described with a poor clinical outcome in the group aged >5 years. In another study by Cecchetto et al. [34], patients aged ≤4 years had a better outcome than older patients. There is epidemiological and molecular evidence suggesting that pediatric adrenocor- tical tumors represent a distinct disease compared to adult tumors regarding their origin, molecular alterations, clinical characteristics, and prognostic evolution [35]. Therefore, these differences may explain the observed inverse correla- tion between CACNA1B expression and adult and pediatric tumor size.

Although widely expressed in the nervous system, N-type calcium channels are also expressed in the pancreas and adrenal gland [36]. The presence and functional role of N-type Cav2.2 were described in H295R cells by Aritomi et al. [15]. In this study, patch-clamp analysis indicated that the N-type Cav2.2 inhibitor @-Conotoxin significantly reduced the transient calcium signaling induced by angio- tensin II (Ang II) and prevented Ang II-induced aldoste- rone and cortisol production. Therefore, N-type calcium channels play a significant role in transducing the Ang II signal for steroid biosynthesis. We found that TCF21 binds directly to the SF1 promoter, inhibiting its activity and StAR expression in human adrenocortical carcinoma cells and in rat primary adrenocortical cell cultures, suggesting that TCF21 regulates adrenocortical steroidogenesis [10,

Figure 6. CACNA1B expression correlates with tumor size in pediatric adrenocortical carcinomas. Correlation analysis of CACNA1B expres- sion with tumor size in A) 5 pediatric patients with ACC and B) 21 adult patients with ACC. The r and p-values are indicated - Spearman's rank correlation coefficient.

210

CACNA1B mRNA expression

r= + 0.98

P= 0.0333

2º

2-5

2º

210

tumor size (cm)

210

CACNA1B mRNA expression

r= - 0.48

P= 0.0327

2º

2-5

2º

210

tumor size (cm)

37, 38]. Taken together, these results suggest that TCF21 is part of the control mechanism for the steroidogenic process controlling SF1 and N-type Cav2.2 expression. Under- standing the different mechanisms related to the control of steroid production in adrenocortical tumors may be partic- ularly important since cortisol-secreting ACCs are associ- ated with poor overall survival [39]. In addition, increased steroid production in ACC could impair tumor immunoge- nicity, contributing to the immune resistance described in this type of cancer [40].

In summary, validation of the most frequent gene sequences identified showed that the regulation of N-type Cav2.2 by TCF21 may be important in cortical hormone- secreting ACC and in the control of steroidogenesis in adrenocortical tumors. Therefore, this mechanism must

Figure 7. High CACNA1B expression negatively impacts disease-free survival in adrenocortical patients. Kaplan-Meier analysis of disease-free survival according to CACNA1B mRNA levels using an adult ACC co- hort. Patients were dichotomized according to median gene expression for survival analysis. Low CACNA1B expression <6.0 (n=12) and high CACNA1B expression >6.0 (n=10). The p-values are indicated by the log- rank test.

100

disease- free time (%)

- Low CACNA1B

-.. High CACNA1B

P=0.017

100

200

300

400

500

Time (months)

be taken into consideration in future immunotherapy approaches for adrenocortical carcinomas.

Supplementary information is available in the online version of the paper.

Acknowledgments: We are grateful to Cintia Fridman, Department of Legal Medicine and Medical Ethics, School of Medicine, University of São Paulo, for the STR DNA profiling analysis in cell cultures. The authors would like to acknowledge all research participants contrib- uting to The Cancer Genome Atlas (TCGA) resource for providing high-quality data for analysis. BSP and JLK are the recipient of a schol- arship from FAPESP (no. 2016/12381-7; 2016/17285-6), the São Paulo State Research Foundation (FAPESP); CFPL received funding from FAPESP (no. 2015/014199-9; 2018/19035-2) and from CNPq. This study was financed in part by the Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - Brasil (CAPES) - Finance Code 001.

References

[1] LU W, YANG C, DU P, ZHANG JL, ZHANG JC. Expression tendency and prognostic value of TCF21 in hepatocellular carcinoma. Artif Cells Nanomed Biotechnol 2019; 47: 1466- 1470. https://doi.org/10.1080/21691401.2019.1601102

[2] ARAB K, SMITH LT, GAST A, WEICHENHAN D, HUANG JP et al. Epigenetic deregulation of TCF21 inhibits metastasis suppressor KISS1 in metastatic melanoma. Carcinogenesis 2011; 32:1467-1473. https://doi.org/10.1093/carcin/bgr138

[3] TESSEMA M, BELINSKY SA. Mining the epigenome for methylated genes in lung cancer. Proc Am Thorac Soc 2008; 5: 806-810. https://doi.org/10.1513/pats.200805-045TH

[4] GOOSKENS SL, KLASSON TD, GREMMELS H, LOGIS- TER I, PIETERS R et al. TCF21 hypermethylation regulates renal tumor cell clonogenic proliferation and migration. Mol Oncol 2018; 12: 166-179. https://doi.org/10.1002/1878- 0261.12149

[5] LI PW, WANG HY, XU LL. [Effect of TCF21 on proliferation, migration and vascular production of breast cancer cells]. Zhonghua Bing Li Xue Za Zhi 2019; 48: 307-311. https://doi. org/10.3760/cma.j.issn.0529-5807.2019.04.009

[6] DAI Y, DUAN H, DUAN C, ZHU H, ZHOU R et al. TCF21 functions as a tumor suppressor in colorectal cancer through inactivation of PI3K/AKT signaling. Onco Targets Ther 2017; 14: 1603-1611. https://doi.org/10.2147/OTT.S118151

[7] YANG Z, JIANG X, LI D, DONG Q, ZHAO H et al. TCF21 inhibits proliferation and chemoresistance through the AKT pathway in human gastric cancer. Gene 2019; 682: 42-49. https://doi.org/10.1016/j.gene.2018.10.011

[8] MOKKAPATI S, PORTEN SP, NARAYAN VM, LIM AH, JAYARATNA IS et al. TCF21 Promotes Luminal-Like Differ- entiation and Suppresses Metastasis in Bladder Cancer. Mol Cancer Res 2020; 18: 811-821. https://doi.org/10.1158/1541- 7786.MCR-19-0766

[9] GIORDANO TJ, KUICK R, ELSE T, GAUGER PG, VINCO M et al. Molecular classification and prognostication of ad- renocortical tumors by transcriptome profiling. Clin Cancer Res 2009; 15: 668-676. https://doi.org/10.1158/1078-0432. CCR-08-1067

[10] FRANCA MM, FERRAZ-DE-SOUZA B, SANTOS MG, LE- RARIO AM, FRAGOSO MC et al. POD-1 binding to the E- box sequence inhibits SF-1 and StAR expression in human adrenocortical tumor cells. Mol Cell Endocrinol 2013; 371: 140-147. https://doi.org/10.1016/j.mce.2012.12.029

[11] PASSAIA BDS, DIAS MH, KREMER JL, ANTONINI SRR, DE ALMEIDA MQ et al. TCF21/POD-1, a Transcritional Regulator of SF-1/NR5A1, as a Potential Prognosis Marker in Adult and Pediatric Adrenocortical Tumors. Front En- docrinol (Lausanne) 2018; 22: 38. https://doi.org/10.3389/ fendo.2018.00038

[12] AUFFORTH RD, NILUBOL N. Emerging therapy for adre- nocortical carcinoma. Int J Endocr Oncol 2014; 1: 173-182. https://doi.org/10.2217/ije.14.13

[13] TORTI JF, CORREA R (Eds.). Adrenal Cancer. In: StatPearls. Treasure Island (FL): StatPearls Publishing 2020. https:// www.ncbi.nlm.nih.gov/books/NBK546580/

[14] MICHALKIEWICZ E, SANDRINI R, FIGUEIREDO B, MIRANDA ECM, CARAN E et al. Clinical and outcome characteristics of children with adrenocortical tumors: a report from the International Pediatric Adrenocortical Tu- mor Registry. J Clin Oncol 2004; 22: 838-845. https://doi. org/10.1200/JCO.2004.08.085

[15] ARITOMI S, WAGATSUMA H, NUMATA T, URIU Y, NOGI Y et al. Expression of N-type calcium channels in hu- man adrenocortical cells and their contribution to cortico- steroid synthesis. Hypertens Res 2011; 34: 193-201. https:// doi.org/10.1038/hr.2010.191

[16] GAZDAR AF, OIE HK, SHACKLETON CH, CHEN TR, TRICHE TJ et al. Establishment and characterization of a human adrenocortical carcinoma cell line that expresses multiple pathways of steroid biosynthesis. Cancer Res 1990; 50: 5488-5496.

[17] KNOWLES BB, HOWE CC, ADEN DP. Human hepatocel- lular carcinoma cell lines secrete the major plasma proteins and hepatitis B surface antigen. Science 1980; 209: 497-499. https://doi.org/10.1126/science.6248960

[18] ALMEIDA MQ, FRAGOSO MC, LOTFI CF, SANTOS MG, NISHI MY et al. Expression of insulin-like growth factor-II and its receptor in pediatric and adult adrenocortical tu- mors. J Clin Endocrinol Metab 2008; 93: 3524-3531. https:// doi.org/10.1210/jc.2008-0065

[19] KONERMANN S, BRIGHAM MD, TREVINO AE, JOUNG J, ABUDAYYEH OO et al. Genome-scale transcriptional activation by an engineered CRISPR-Cas9 complex. Nature 2015; 517: 583-588. https://doi.org/10.1038/nature14136

[20] ZHANG Y, LIU T, MEYER CA, EECKHOUTE J, JOHNSON DS et al. Model-based analysis of ChIP-Seq (MACS). Ge- nome Biol 2008; 9: R137. https://doi.org/10.1186/gb-2008-9- 9-r137

[21] LI Q, BROWN JB, HUANG H, BICKEL PJ. Measur- ing reproducibility of high-throughput experiments. An- nals of Applied Statistics 2011; 5: 1752-1779. https://doi. org/10.1214/11-AOAS466

[22] AMEMIYA HM, KUNDAJE A, BOYLE AP. The ENCODE Blacklist: Identification of Problematic Regions of the Ge- nome. Sci Rep 2019; 9: 9354. https://doi.org/10.1038/s41598- 019-45839-z

[23] SCHNEIDER CA, RASBAND WS, ELICEIRI KW. NIH Im- age to ImageJ: 25 years of image analysis. Nat Methods 2012; 9: 671-675. https://doi.org/10.1038/nmeth.2089

[24] BLEAKMAN D, BOWMAN D, BATH CP, BRUST PF, JOHNSON EC et al. Characteristics of a human N-type calcium channel expressed in HEK293 cells. Neurophar- macology 1995; 34: 753-765. https://doi.org/10.1016/0028- 3908(95)00078-k

[25] OLIVERA BM, CRUZ LJ, DE SANTOS V, LECHEMINANT GW, GRIFFIN D et al. Neuronal calcium channel antago- nists. Discrimination between calcium channel subtypes using omega-conotoxin from Conus magus venom. Bio- chemistry 1987; 26: 2086-2090. https://doi.org/10.1021/ bi00382a004

[26] WILLIAMS ME, FELDMAN DH, MCCUE AF, BRENNER R, VELICELEBI G et al. Structure and functional expres- sion of alpha 1, alpha 2, and beta subunits of a novel human neuronal calcium channel subtype. Neuron 1992; 8: 71-84. https://doi.org/10.1016/0896-6273(92)90109-q

[27] HOHENAUER T, MOORE AW. The Prdm family: expand- ing roles in stem cells and development. Development 2012; 139: 2267-2282. https://doi.org/10.1242/dev.070110

[28] MZOUGHI S, TAN YX, LOW D, GUCCIONE E. The role of PRDMs in cancer: one family, two sides. Curr Opin Genet Dev 2016; 36: 83-91. https://doi.org/10.1016/j. gde.2016.03.009

[29] PINTO EM, ZAMBETTI GP, RODRIGUEZ-GALINDO C. Pediatric adrenocortical tumours. Best Pract Res Clin En- docrinol Metab 2020; 34: 101448. https://doi.org/10.1016/j. beem.2020.101448

[30] JUHLIN CC, BERTHERAT J, GIORDANO TJ, HAMMER GD, SASANO H et al. What Did We Learn from the Mo- lecular Biology of Adrenal Cortical Neoplasia? From Histo- pathology to Translational Genomics. Endocr Pathol 2021; 32: 102-133. https://doi.org/10.1007/s12022-021-09667-0

[31] ASSIÉ G, LETOUZÉ E, FASSNACHT M, JOUINOT A, LUSCAP W, BARREAU O et al. Integrated genomic charac- terization of adrenocortical carcinoma. Nat Genet 2014; 46: 607-612. https://doi.org/10.1038/ng.2953

[32] PINTO EM, CHEN X, EASTON J, FINKELSTEIN D, LIU Z et al. Genomic landscape of paediatric adrenocortical tu- mours. Nat Commun 2015; 6: 6302. https://doi.org/10.1038/ ncomms7302

[33] WIENEKE JA, THOMPSON LD, HEFFESS CS. Adrenal cortical neoplasms in the pediatric population: a clini- copathologic and immunophenotypic analysis of 83 pa- tients. Am J Surg Pathol 2003; 27: 867-881. https://doi. org/10.1097/00000478-200307000-00001

[34] CECCHETTO G, GANARIN A, BIEN E, VORWERK P, BI- SOGNO G et al. Outcome and prognostic factors in high- risk childhood adrenocortical carcinomas: A report from the European Cooperative Study Group on Pediatric Rare Tumors (EXPERT). Pediatr Blood Cancer 2017; 64. https:// doi.org/10.1002/pbc.26368

[35] LALLI E, FIGUEIREDO BC. Pediatric adrenocortical tu- mors: what they can tell us on adrenal development and com- parison with adult adrenal tumors. Front Endocrinol (Laus- anne) 2015; 6: 23. https://doi.org/10.3389/fendo.2015.00023

[36] MILLER RJ. Voltage-sensitive Ca2+ channels. J Biol Chem 1992; 267: 1403-1406.

[37] FRANCA MM, FERRAZ-DE-SOUZA B, LERARIO AM, FRAGOSO MC, LOTFI CF. POD-1/TCF21 Reduces SHP Expression, Affecting LRH-1 Regulation and Cell Cycle Balance in Adrenocortical and Hepatocarcinoma Tumor Cells. Biomed Res Int 2015; 2015: 841784. https://doi. org/10.1155/2015/841784

[38] FRANCA MM, LERARIO AM, FRAGOSO M, LOTFI CFP. New evidences on the regulation of SF-1 expression by POD1/TCF21 in adrenocortical tumor cells. Clinics (Sao Paulo) 2017; 72: 391-394. https://doi.org/10.6061/clin- ics/2017(06)10

[39] VANBRABANT T, FASSNACHT M, ASSIE G, DEKKERS OM. Influence of hormonal functional status on survival in adrenocortical carcinoma: systematic review and meta- analysis. Eur J Endocrinol 2018; 179: 429-436. https://doi. org/10.1530/EJE-18-0450

[40] COSENTINI D, GRISANTI S, DALLA VOLTA A, LAGANA M, FIORENTINI C et al. Immunotherapy failure in adreno- cortical cancer: where next? Endocr Connect 2018; 7: E5-E8. https://doi.org/10.1530/EC-18-0398

https://doi.org/10.4149/neo_2022_220106N25