Genetics and Molecular Biology
Sociedade Brasileira de Genética
image
Transcriptional regulators and regulatory pathways involved in prostate gland adaptation to a hypoandrogen environment
Volume: 42, Issue: 4
DOI 10.1590/1678-4685-GMB-2018-0362
  • PDF   
  • XML   
  •       
Abstract

Anti-androgen therapies, including orchiectomy, are effective at promoting prostate cancer remission, but are followed by progression to the more aggressive castration-resistant prostate cancer (CRPC). Castration promotes gland and tumor shrinkage. However, prostate adaptation to androgen deprivation involves striking parallel events, all requiring changes in gene expression. We hypothesized that transcription factors (TF) and other transcription-related genes are needed to orchestrate those changes. In this work, downstream analysis using bioinformatic tools and published microarray data allowed us to identify sixty transcriptional regulators (including 10 TF) and to integrate their function in physiologically relevant networks. Functional associations revealed a connection between Arnt, Bhlhe41 and Dbp circadian rhythm genes with the Ar circuitry and a small gene network centered in Pex14, which might indicate a previously unanticipated metabolic shift. We have also identified human homologs and mapped the corresponding genes to human chromosome regions commonly affected in prostate cancer, with particular attention to the PTEN/HHEX/MXI1 cluster at 10q23-25 (frequently deleted in PCa) and to MAPK1 at 22q11.21 (delete in intermediate risk but not in high risk PCa). Twenty genes were found mutated or with copy number alterations in at least five percent of three cancer cohorts and six of them (PHOX2A, NFYC, EST2, EIF2S1, SSRP1 and PARP1) associated with impacted patient survival. These changes are specific to the adaptation to the hypoandrogen environment and seem important for the progression to CRPC when mutated.

Keywords
Nishan, da Rosa-Ribeiro, Damas-Souza, Barbosa, and Carvalho: Transcriptional regulators and regulatory pathways involved in prostate gland adaptation to a hypoandrogen environment

Introduction

Prostate diseases in general and prostate cancer (PC) in particular are major concerns of public health care. One eighth to one sixth of males will develop PC and experience the risk of prostate cancer progression if not properly diagnosed, monitored and treated. Molecular markers for diagnosis and disease progression risk assessment are extremely necessary. One of the palliative treatments for advanced prostate cancer is androgen blockade, achieved by chemical or surgical castration. Besides the psychological and physiological side effects, the risk of androgen blockade resides in the common progression to the highly malignant and life-threatening form of castration-resistant prostate cancer (CRPC). Several mutations and chromosomal rearrangements have been associated with PC and CRPC. Recently, a series of chromosomal translocations including frequent bridging and rearrangements was described and demonstrated to occur in a few steps and to affect a series of important tumor suppressors (Baca et al., 2013). Nonetheless, the linear progression among progressive stages has been questioned, as metastatic lesions are clonally derived, show signatures particular to each affected individual and might result from less advanced local primary lesions (Holcomb et al., 2009, Haffner et al., 2013).

Androgens, acting through the androgen receptor (AR), are required for prostate development and normal function (Roy et al., 1998). Hence, surgical or chemical castration defines a hypoandrogenic state in which a series of events sum up to promote organ (and tumor) shrinkage; androgen deprivation/blocking is the first line therapy for advanced prostate cancer. Gain-of-function mutations enabling the AR to recover activity in the hypoandrogen environment have been associated with the progression to CRPC (Feldman and Feldman, 2001), and include mutations, deletions and inversions at the ligand binding site leading to ligand-independent activation (Nyquist et al., 2013), and gene amplifications. In spite of these AR-centered modifications, a series of genes has been implicated in prostate cancer progression, including overall changes in gene expression (Abate-Shen and Shen, 2000, Tomlins et al., 2007, Shen and Abate-Shen, 2010) and complex chromosomal rearrangements frequently involving PTEN, NKX3.1, SPOP, CHD1, TP53, MAP3K7, FOXP1 and the T2-ERG fusion (Baca et al., 2013).

Furthermore, we consider that other physiological aspects of the adaptation to the hypoandrogen environment might be corrupted in cancer cells, and cooperate in establishing the selective pressure that contributes to the clonality of cells harboring chromosomal changes in general and AR modifications in particular.

Although epithelial cell apoptosis is a major event in prostate regression occurring in response to castration, it is not the sole event. For instance, remarkable reprogramming of immune system cells (Desai et al., 2004) and smooth muscle cells (Antonioli et al., 2004, 2007) as well as reorganization of the extracellular matrix (Vilamaior et al., 2000) have been described and associated with a re-defined functional state and immune barrier system. Additionally, we have reported the occurrence of desquamation as an additional phenomenon contributing to epithelial cell deletion (Rosa-Ribeiro et al., 2014a), and a relevant role for two macrophage subpopulations in both (a) the induction of epithelial cell death (Barbosa et al., 2019) and (b) the clearance of cell corpses and maintenance of the non-inflammatory status (Silva et al., 2018).

However, little has been studied beyond the induction of apoptosis in epithelial cells. Progress has been made in terms of showing that epithelial expression of the AR is not necessary for epithelial cell death (Kurita et al., 2001), while the remaining cells develop resistance to androgen deprivation and preserve a differentiation-immature signature (Rosa-Ribeiro et al., 2014b).

The rat ventral prostate has been a valuable and robust in vivo model system to explore androgen regulation of gene expression (Wang et al., 1997, Kwong et al., 1999, Desai et al., 2004). The ventral prostate responds to androgen withdrawal with increased epithelial cell apoptosis, whereas the dorsolateral lobes show negligible cell death (Kwong et al., 1999). The responsiveness of the ventral prostate to androgens is also characterized by the number of differentially expressed genes after castration as compared to the dorsolateral lobe (1496 vs. 256 genes, respectively) (Desai et al., 2004). Most of these changes take place in the hypoandrogen environment, and are triggered by disengaging AR signaling.

We hypothesized that other transcription factors (TF) and transcriptional regulators (TR) are co-opted to coordinate the sequential changes observed in the gland after castration. In this work, we explore this idea in an attempt to find new genes, unforeseen metabolic process and chromosomal hotspots that might reveal possible connections between gland physiology under androgen deprivation promoted by castration and progression to CRPC after hormone therapies.

We have used bioinformatics to (a) select class-specific genes from a published list of genes and ESTs differentially expressed in response to castration and androgen supplementation after DNA microarray analysis, (b) to identify the regulatory networks in which the selected genes are involved, (c) to map the homologs of rat genes to human chromosomes, and (d) to find mutations and/or copy number alterations and changes in patient survival.

Accordingly, this study has unveiled a list of TF and TR genes and a series of unexplored physiological pathways, such as circadian rhythms (genes Arnt/Bhlhe41/Dpb) and peroxisome biogenesis (Pex14), hitherto neglected pathways in prostate biology. We also correlated the selected genes with chromosomal regions commonly deleted in prostate cancer, such as 10q23, which contains the PTEN/HHEX/MXI1 gene cluster, and 22q11.21, harboring the MAPK1 gene, found 20 genes mutated in at least 5% of three patient cohorts and six genes affecting patient survival when mutated.

Materials and Methods

The microarray data from Desai et al. (2004) reported 1496 genes/ESTs differentially expressed in response to castration and testosterone supplementation. The list of gene bank accession IDs for all genes and ESTs was loaded into DAVID v6.7. Gene IDs and biological annotations are highly redundant within the vast array of public databases. The DAVID knowledge base collects and integrates various gene identifiers as well as more than 40 well-known publicly annotation categories, which are then centralized by the internal DAVID identifier in a non-redundant manner. A significant portion of input gene IDs failed to be mapped and were then processed using the gene ID conversion tool. All the identified IDs/gene names were listed by the gene name batch viewer. We further processed the identified IDs for the identification of functional annotations centered on TFs and TRs, and the identified genes were further studied to find their functional annotation clustering and possible integration in known biological functions.

The TFs/TRs were also studied for possible functional associations using the Ingenuity Pathway Analysis (IPA) software, with filtering for information in the rat, and choosing only direct interactions.

The human homologs to the rat genes were searched manually using the NCBI database, and their chromosomal location was used to map them to the human ideogram.

Finally, we assessed the cBioPortal (cbioportal.org) and checked three cohorts of prostate adenocarcinomas for the existence of mutations and/or copy number alterations and possible effect on patient survival (Armenia et al., 2018, Liu et al., 2018, Abida et al., 2019).

A limit of 5% mutations and a log rank test p-value smaller than 0.05% were set for each analysis, respectively.

Results

Data processing

Using the gene accession conversion tool of DAVID v6.7, the program managed to convert 468 IDs from the list of 1477 total unique user IDs. The number of genes identified was similar to that obtained in the original work (Desai et al., 2004). Out of the 468 IDs, DAVID identified 60 TFs/TRs. Table 1 lists the detailed annotation and functional enrichment information that was retrieved using the terms TFs/TRs. The chromosomal location for each gene was determined using the NCBI databank. Twenty-two genes were identified as transcription factors (bold in Table 1).

Table 1
Genes with accession numbers, names and functions and chromosomal location in the rat chromosomes as well as the chromosomal location of the human homologs. Known and putative transcription factors are bold-faced.
S.NoGene Accession NumberGene NameAbbreviationRat Chromosome locationLocation of the human homologGene Function
1NM_017259B-cell translocation gene 2, anti-proliferativeBtg213q13-q311q32Regulation of transcription
2NM_057109BarH-like homeobox 1Barhl13p129q34Regulation of transcription, transcription factor activity
3NM_013154CCAAT/enhancer binding protein (C/EBP), deltaCebpd118p11.2-p11.1Regulation of transcription from RNA polymerase II promoter, transcription factor activity
4NM_012543D site of albumin promoter (albumin D-box) binding proteinDbp1q2219q13.3Transcription factor activity, Basic-leucine zipper (bZIP) transcription factor
5NM_020083GTPase activating Rap/RanGAP domain-like 1Ralgapa16q2314q13.2Regulation of transcription
6NM_012855Janus kinase 3Jak316p1419p13.1Regulation of transcription, transcription factor binding
7NM_013160MAX interactor 1Mxi11q5510DNA binding, transcription repressor activity, transcription regulator activity
8NM_022856Ngfi-A binding protein 1Nab19q222q32.3-q33Transcription repressor activity, transcription regulator activity
9NM_019275SMAD family member 4Smad418q12.318q21.1Transcription factor complex, transcription activator activity, transcription regulator activity
10NM_017359RAB10, member RAS oncogene familyRab106q122p23.3Regulation of transcription, transcription factor binding
11NM_021693SNF1-like kinaseSik120p1221q22.3Transcription repressor activity, transcription regulator activity
12NM_012903Acidic (leucine-rich) nuclear phosphoprotein 32 family, member AAnp32a8q2415q23Regulation of transcription
13NM_031018Activating transcription factor 2Atf23q232q32Transcription factor activity, transcription activator, Basic-leucine zipper (bZIP) transcription factor, Cyclic AMP-dependent transcription factor ATF-2, bZIP transcription factor
14M64780AgrinAgrn5q361p36.33Regulation of transcription, regulation of transcription from RNA polymerase II promoter
15NM_012907Apolipoprotein B mRNA editing enzyme, catalytic polypeptide 1Apobec14q4212p13.1Posttranscriptional regulation of gene expression
16AF015953Aryl hydrocarbon receptor nuclear translocator-likeArntl1q3411p15Positive regulation of transcription, transcription factor activity
17AF009329Basic helix-loop-helix family, member e41Bhlhe414q4312p12.1Transcription regulator activity
18NM_017338Calcitonin/calcitonin-related polypeptide, alphaCalca1q3411p15.2Regulation of transcription
19X83579Cyclin-dependent kinase 7Cdk72q125q12.1Transcription factor complex, transcription regulation
20NM_012698Dystrophin, muscular dystrophyDmdXq22Xp21.2Regulation of transcription
21NM_012754Estrogen receptor 2 (ER beta)Esr26q2414q23.2Transcription factor activity
22NM_031041General transcription factor IIBGtf2b2q441p22-p21transcription initiation, transcription factor complex, Transcription factor TFIIB related
23J05181Glutamate-cysteine ligase, catalytic subunitGclc8q316p12Regulation of transcription
24NM_021592Heart and neural crest derivatives expressed 1Hand110q225q33DNA binding, transcription factor activity, transcription cofactor activity, transcription coactivator activity, transcription factor binding, enzyme binding, transcription regulator activity, bHLH transcription factor binding
25NM_024385Hematopoietically expressed homeoboxHhex1q5310q23.33DNA binding, transcription factor activity, eukaryotic initiation factor 4E binding, general transcriptional repressor activity, transcription regulator activity, translation initiation factor binding, sequence-specific DNA binding,
26NM_032070High mobility group AT-hook 2Hmga27q2212q15Regulation of transcription
27NM_031787Homeodomain interacting protein kinase 3Hipk33q3211p13Regulation of transcription
28NM_019356Eukaryotic Translation initiation factor 2, subunit 1 alphaEif2s16q2414q23.3Posttranscriptional regulation of gene expression
29NM_013060Inhibitor of DNA binding 2Id26q162p25Regulation of transcription factor activity,
30NM_053355Inhibitor of kappa light polypeptide gene enhancer in B-cells, kinase betaIkbkb16q12.58p11.2Regulation of transcription factor activity, positive regulation of NF-kappaB transcription factor activity
31NM_012591Interferon regulatory factor 1Irf110q225q31.1DNA binding, transcription factor activity, sequence-specific DNA binding
32NM_053842Mitogen activated protein kinase 1Mapk111q2322q11.21Transcription factor binding, positive regulation of transcription
33NM_017322Mitogen-activated protein kinase 9Mapk910q225q35Regulation of transcription
34NM_053718myeloid/lymphoid or mixed-lineage leukemia (trithorax homolog, Drosophila); translocated to, 3Mllt35q329p22Regulation of transcription
35U68726Neogenin homolog 1 (chicken)Neo18q2415q22.3-q23Transcription regulator activity
36NM_012866Nuclear transcription factor-Y gammaNfyc5q361p32DNA binding, transcription factor activity, sequence-specific DNA binding,
37NM_053869Paired-like homeobox 2aPhox2a1q3211q13.2DNA binding, transcription factor activity, sequence-specific DNA binding
38AB017544Peroxisomal biogenesis factor 14Pex145q361p36.22Transcription cofactor activity, transcription corepressor activity, transcription factor binding, transcription repressor activity,
39NM_031606Phosphatase and tensin homologPten1q41-q4310q23.3Posttranscriptional regulation of gene expression,
40NM_013063poly (ADP-ribose) polymerase 1Parp113q261q41-q42Transcription regulation
41NM_053949potassium voltage-gated channel, subfamily H (eag-related), member 2Kcnh24q117q36.1Transcription regulation
42NM_019243prostaglandin F2 receptor negative regulatorPtgfrn2q341p13.1Posttranscriptional regulation of gene expression
43NM_031149proteasome (prosome, macropain) 26S subunit, ATPase, 5Psmc510q32.117q23.3Regulation of transcription, transcription factor binding
44NM_031528Retinoic acid receptor, alphaRara10q3117q21Transcription factor activity, transcription cofactor activity, transcription coactivator activity, Transcription, transcription
45AJ223083Retinoid X receptor gammaRxrg13q241q22-q23Transcription factor activity, transcription regulator activity
46L29259Similar to transcription elongation factor B (SIII), polypeptide 1; transcription elongation factor B (SIII), polypeptide 1Tceb15q118q21.11RNA polymerase II transcription factor activity, transcription elongation regulator activity, regulation of transcription
47NM_030835Stress-associated endoplasmic reticulum protein 1Serp12q313q25.1Posttranscriptional regulation of gene expression,
48L08814Structure specific recognition protein 1Ssrp13q2411q12DNA binding, transcription regulator activity
49NM_053800Thioredoxin 1Txn15q249q31Regulation of transcription
50U30789Thioredoxin interacting proteinTxnip2q341q21.1Regulation of transcription
51NM_012887ThymopoietinTmpo7q1312q22Regulation of transcription
52J03819Thyroid hormone receptor betaThrb15p163p24.2DNA binding, double-stranded DNA binding, transcription factor activity
53U54632Transmembrane protein 215; similar to Ubiquitin-conjugating enzyme E2 I (Ubiquitin-protein ligase I) (Ubiquitin carrier protein I) (SUMO-1-protein ligase)Ube2i10q1216p13.3transcription factor binding, specific transcriptional repressor activity, transcription regulator activity, bHLH transcription factor binding
54NM_013091Tumor necrosis factor receptor superfamily, member 1aTnfrsf1a4q4212p13.2Regulation of transcription
55NM_053928Ubiquitin-conjugating enzyme E2N; similar to ubiquitin-conjugating enzyme E2N (homologous to yeast UBC13)Ube2n7q13Xq27Regulation of transcription factor activity, regulation of transcription
56NM_012555v-ets erythroblastosis virus E26 oncogene homolog 1 (avian)Ets18q2111q23.3Transcription factor activity, transcription regulator activity, sequence-specific DNA binding
57NM_017058Vitamin D (1, 25- dihydroxyvitamin D3) receptorVdr7q3612q13.11DNA binding, transcription factor activity, transcription factor binding
58X52590Zinc finger protein 36, C3H type-like 1Zfp36l16q2414q22-q24Posttranscriptional regulation of gene expression
59AF072439Zinc finger protein 37Zfp375q249q32Regulation of transcription
60AF052042Zinc finger protein 394Zfp39412p117q22.1Transcription factor activity, Regulation of transcription

Roles of transcription factors and functional associations among the selected genes

The selected TFs and TRs function in 17 important cellular pathways identified by DAVID (Table 2). Some of the ontogenies were very general (such as “Pathways in cancer” or “Prostate cancer” or “Type II diabetes mellitus”); all but 3 (14/17) included Mapk1, and about half (9/17) contained Mapk9, related to areas of strong research or hubs in central signaling pathways. The ontogenies also pointed to TGF-β, Toll-like receptors and T-cell receptor signaling pathways. Novel ontogenies implicated the genes Arnt and Bhlhe41 in “Circadian rhythms” and Ets-1 (E26 oncogene homolog 1) in “Dorso-ventral axis formation”.

Table 2
Regulatory pathways involving the selected genes identified by DAVID.
S.NoPathwayGene nameGenes count%P-ValueBenjamini’s false discovery rate
1Pathways in cancerSMAD family member 4 B-cells, kinase beta Mitogen activated protein kinase 1 Mitogen-activated protein kinase 9 Phosphatase and tensin homolog Retinoic acid receptor, alpha Retinoid X receptor gamma transcription elongation factor B (SIII)813.34.5E-43.3E-2
2Adipocytokine signaling pathwayB-cells, kinase beta Mitogen-activated protein kinase 9 Retinoid X receptor gamma Superfamily, member 1a46.73.5E-31.2E-1
3Pancreatic cancerSMAD family member 4 B-cells, kinase beta Mitogen activated protein kinase 1 Mitogen-activated protein kinase 946.73.8E-38.9E-2
4Type II diabetes mellitusB-cells, kinase beta Mitogen activated protein kinase 1 Mitogen-activated protein kinase 935.02.1E-23.3E-1
5Acute myeloid leukemiaB-cells, kinase beta Mitogen activated protein kinase 1 Retinoic acid receptor, alpha35.02.7E-23.3E-1
6MAPK signaling pathwayActivating transcription factor 2 B-cells, kinase beta Mitogen activated protein kinase 1 Mitogen-activated protein kinase 9 Superfamily, member 1a58.33.3E-23.3E-1
7NOD-like receptor signaling pathwayB-cells, kinase beta Mitogen activated protein kinase 1 Mitogen-activated protein kinase 935.03.3E-23.0E-1
8Renal cell carcinomaMitogen activated protein kinase 1 Transcription elongation factor B (SIII) E26 oncogene homolog 1 (avian)35.04.0E-23.1E-1
9Chronic myeloid leukemiaSMAD family member 4 B-cells, kinase beta Mitogen activated protein kinase 135.04.7E-23.2E-1
10Colorectal cancerSMAD family member 4 Mitogen activated protein kinase 1 Mitogen-activated protein kinase 935.05.4E-23.3E-1
11Small cell lung cancerB-cells, kinase beta Phosphatase and tensin homolog Retinoid X receptor gamma35.05.6E-23.2E-1
12Circadian rhythmAryl hydrocarbon receptor Basic helix-loop-helix family, member e4123.35.9E-23.1E-1
13TGF-beta signaling pathwaySMAD family member 4 Inhibitor of DNA binding 2 Mitogen activated protein kinase 135.06.0E-22.9E-1
14Prostate cancerB-cells, kinase beta Mitogen activated protein kinase 1 Phosphatase and tensin homolog35.06.5E-23.0E-1
15Toll-like receptor signaling pathwayB-cells, kinase beta Mitogen activated protein kinase 1 Mitogen-activated protein kinase 935.06.5E-23.0E-1
16T cell receptor signaling pathwayB-cells, kinase beta Mitogen activated protein kinase 1 Mitogen-activated protein kinase 935.09.0E-23.7E-1
17Dorso-ventral axis formationMitogen activated protein kinase 1 E26 Oncogene homolog 1 (avian)23.39.8E-23.7E-1

Upon further inspection, using IPA to set the interactions among the 60 genes, we found eight networks corresponding to known pathways (Figure 1; Table S1). They vary in terms of the number of individual nodes, but reveal interesting aspects of the yet-to-be proven physiology of the prostate gland in the hypoandrogen environment. Perhaps not surprisingly, they are ascribed to gene expression regulation, cell death and survival, and also to nucleic acid and carbohydrate metabolism and cancer. They also implicate particular pathways such as estrogen receptor, retinoic acid receptor, thyroid hormone receptor, NFκB signaling, TGF-β and establishing connections with the newly identified genes. It is interesting to note that Arntl and Bhlhe41 , both involved in circadian rhythms (Pathway 12, in Table 2), appeared together in network number 2 (Figure 1). Arntl connected to the AR via either p300/EP300 or CREBBP acetyl transferases, and directly do Dbp, another circadian rhythm gene. Additionally, IPA retrieved one particularly interesting pathway, peroxisomal biogenesis and function, referenced by network number 8, which is centered on the gene Pex14.

Functional associations among the 60 TF/TR in eight networks,
according to IPA. Functional descriptors are presented in
Table
S1. The genes shown in gray are
those identified in this work.
Figure 1
Functional associations among the 60 TF/TR in eight networks, according to IPA. Functional descriptors are presented in Table S1. The genes shown in gray are those identified in this work.

Chromosomal mapping

We identified the human homolog for each gene (Table 1) and determined their location in the human chromosomes ideogram, also using the NCBI databank. This data set was used to localize each gene in the human ideogram (Figure 2). Chromosome 1 contained eight genes (AGRN, PEX14, NFYC, GTF2B, PTGFRN, TXNIP, RXRG, BTG2, PARP1) and small clusters were observed in 2q (SERP1, ATF2, NAB1), 5q (IRF1, HAND1, MAPK9), 9q (TXN1, ZPF37, BARHL1), 10q (PTEN, HHEX, MXI1), 12p (TNFRSF1A, APOBEC1, BHLHE41) and 14q (ZFP36L1, ESR2, EIF2S1). The 10q23 region included Pten, Hhex and Mxi1 . In contrast, not a single gene among the selected 60 mapped to chromosomes 4, 13, 16, 20 and Y. Areas of frequent variation (i.e. gains or deletions) (Iafrate et al., 2004) were included for the determination of proximity to the set of the human homologs of the selected genes. The location of the regions frequently affected by gains/losses in healthy individuals (Iafrate et al., 2004) revealed almost no association with the selected genes (Figure 2). On the other hand, half of the selected genes were mapped to chromosomal regions found to be amplified or deleted in prostatic diseases (Figure 3). Associations were found with metastatic cancer (12/6; gains/losses), localized high (1/4; gain/losses) and low risk (1 loss; RXRG ), prostate intraepithelial neoplasia (PIN) (2/1; gains/loss) (Kim et al., 2007), and with the intermediate risk prostate cancer (6 losses) (Ishkanian et al., 2009). Remarkably, ETS1 and IRF1 appeared in regions of gains in PIN but not in other disease states, and MAPK1 and SIK1 were located in (or nearby) regions deleted in intermediate-risk cancer.

Mapping of the human orthologs of 60 TF/TR rat genes to the human
ideogram. Indicated are the regions of copy number gain (blue) and
losses (red) reported for healthy human individuals, according to Iafrate et al.
(2004).
Figure 2
Mapping of the human orthologs of 60 TF/TR rat genes to the human ideogram. Indicated are the regions of copy number gain (blue) and losses (red) reported for healthy human individuals, according to Iafrate et al. (2004).
Association of frequently amplified (red), deleted (green) or both
(hatched) chromosomal regions harboring 30 of the selected TF/TR in
metastatic prostate cancer (MET), prostate cancer (PCA), high grade PCA
(H-PCA), low grade PCA (L-PCA), prostate intraepithelial neoplasia
(PIN), and intermediate risk PCA (I-PCA). Based on Kim et al. (2007) and Ishkamian et al.
(2009).
Figure 3
Association of frequently amplified (red), deleted (green) or both (hatched) chromosomal regions harboring 30 of the selected TF/TR in metastatic prostate cancer (MET), prostate cancer (PCA), high grade PCA (H-PCA), low grade PCA (L-PCA), prostate intraepithelial neoplasia (PIN), and intermediate risk PCA (I-PCA). Based on Kim et al. (2007) and Ishkamian et al. (2009).

Mutation rates and effect on patient survival

PTEN is a tumor suppressor frequently associated with prostate cancer (Abate-Shen and Shen, 2000, Tomlins et al., 2007, Shen and Abate-Shen, 2010, Baca et al., 2013). We found PTEN mutated in 16%, 21% and 33% of the patients for the three cohorts studied. Beside PTEN, only IKBKB was found mutated in at least 5% of the patients in the three cohorts. CEBPD and DMD showed mutations in more than 10% of the patients in one cohort. BTG2, SERP1, CKNH2, RXRG, TXN1P, UBE2I, ZFP37, BARHL1, RALGAPA1, CDK7, PARP1, PTEFGRN, TXN1, TMPO, TNFRSF1A and AGRN were mutated in at least 5% of the patients in at least one cohort. However, more than 50% of the studied genes showed deletions in the three cohorts.

Survival curves existed for two of the three cohorts studied. PHOX2a and NFYC were associated with significant impact on patient survival (P < 0.05) in the first cohort and EST2, EIF251, SSRP1 and PARP1, in the second cohort.

Discussion

Sixty differentially expressed TF and TR were retrieved from the microarray data published by Desai et al. (Desai et al., 2004). These genes were assorted into 17 pathways by the DAVID knowledge base and into eight functional networks by Ingenuity Pathway Analysis (IPA). Though most these pathways were too general, and led to central metabolic hubs, such as PTEN (which was further validated in the parent study) and nuclear receptors pathways, some revealed particularly interesting and unforeseen aspects of prostate biology, such as circadian rhythms and peroxisome biogenesis. The selected genes were also mapped to chromosome regions frequently affected in prostate cancers, as their identification might serve as risk factors or therapeutic targets relevant to progression to CRPC. Additionally, 20 genes were found mutated in at least 5% of prostate adenocarcinoma patients.

We used the DAVID module Gene ID Conversion Tool (Huang et al., 2007a, 2007b) to identify gene IDs from the initial gene list. The number of genes was roughly the same as those uncovered in the parent study, perhaps indicating a potential limit for retrieving information from the early commercially available microarray chips. In order to further advance our understanding of the physiology and endocrinology of the VP gland and to facilitate the biological interpretation of prostate biology in a broad range of biological processes, the 468 genes were further processed to track down their functional classification and resulted in 60 different TF or TR, which were then assigned to known regulatory pathways using DAVID and to functional interaction networks using IPA, resulting in the identification of 22 transcription factors.

Prostate cancer will affect one eighth to one sixth of men worldwide. In spite of enormous progress in the understanding of many facets of the disease, new concepts are still emerging. One of these is the clonal origin of metastases (Haffner et al., 2013), punctuated rather than gradual progression (Baca et al., 2013) and the non-linear relationship of on-site and distant metastases, meaning that metastases might be generated from less advanced local tumor foci (Haffner et al., 2013). Accordingly, it is not completely understood as to which foci in the advanced stage of the disease will progress to CRPC. We followed the idea that the physiological adaptation of the gland to the hypoandrogenic castration-induced environment involves regulatory pathways to maintain the gland in a regressed, low proliferative and less functional (meaning less differentiated) state, and that these pathways in cancer cells might be corrupted and thereby contribute to the progression to CRPC. In this scenario, molecules with enhanced expression after castration might be found to be “tumor suppressors”, particularly if they function as hubs in regulatory networks that are defective in cancer cells. Smad4, Ikbkb, Rara, ets1, Bhlhe41, Id2, Tnfrsf1a, Mxi1 and Dbp are new candidates, together with the well-known tumor suppressor Pten. As a matter of fact, these genes (except ID2 and Mxi1) were found deleted in prostate adenocarcinomas and advanced metastatic prostate cancers.

This also raises the question of whether Pten-defective tumors should be submitted to androgen deprivation or blockade, as a major aspect of the prostate response to falling androgen levels relies on this phosphatase.

The functional characterization of the 60 genes revealed interesting attributes. DAVID retrieved 17 pathways, most of them centered on either Mapk1 or Mapk9 , which is perhaps too general to indicate new physiological functions. It is worth noticing that MAPK pathway has been implicated in increased survival of castrate-resistant prostate cancer patients (Mukherjee et al., 2011).

Next, we uncovered circadian rhythms as a relevant pathway, centered on the genes Arntl and Bhlhe41. These genes were included in network number 2 retrieved by IPA, which also included the Dbp gene, also implicated in circadian regulation. Arntl is directly linked to Dbp and indirectly to Ar via the p300 and CREBBP acetyl transferases. Dbp has been reported to have peak expression at 8 h within the light portion of the 12h:12h light/dark cycle (Zeitegeber, ZT 8) in the rat prostate gland (Qi et al., 2009, Sunkel and Wang, 2014) in a similar fashion to other core clock genes in the mouse prostate (Bebas et al., 2009). Moreover, ARNTL polymorphisms have been significantly associated with susceptibility to prostate cancer (Zhu et al., 2009). This evidence raises the possibility that AR-dependent and AR-independent circadian functions contribute to the prostate gland physiology, by opening a new connection to environmental factors, knowingly significant in prostate cancer risk and incidence.

The peroxisome biogenesis pathway, represented by the sole gene Pex14, is also another connection to the environment, as peroxisome proliferation and activity are related to several environmental (and dietary) factors, adding further complexity to the peculiar metabolic adaptations of the gland given its function in accumulating citrate in secretions (Singh et al., 2006).

We also found that some of the genes identified in the present investigation map to regions commonly deleted in prostate cancer. In particular, we refer to the Pten/Hhex/Mxi1 cluster at 10q23, which was characterized in detail before (Hermans et al., 2004). It will be interesting to investigate whether the differently sized deletions in this region might affect the behavior of prostate cancer cells, as MXI1 is usually lacking or inactive in prostate cancer (Eagle et al., 1995, Prochownik et al., 1998); its function is to suppress proliferation by antagonizing Myc (Taj et al., 2001), which in turn is commonly amplified in prostate cancer (Ishkanian et al., 2009). It is important to mention that Mxi1 expression is suggested to decay after castration, according to the parent study (Desai et al., 2004). In contrast, Hhex expression is increased, which functions as a coordinator of hematopoiesis and the development of endoderm-derived organs such as the liver and thyroid (Martinez Barbera et al., 2000).

Additionaly, we found the Mapk1 gene, whose homolog MAPK1 maps to 22q11.21, in a region between the 22q11.21 and 22q12.1 segments deleted in 29% and 33% of intermediate risk tumors, respectively, but not frequently observed in high risk cancers (Ishkanian et al., 2009). The MAPK1 gene product is better known as ERK-2 (or p42 MAPK) and funnels down a variety of extracellular signals to control several functions, particularly the G1-S transition within the cell cycle (Meloche and Pouysségur, 2007). The importance of MAPK1 is highlighted by the fact that it was enlisted as a node in 14 of the 17 pathways identified by DAVID, including prostate cancer among others, and the recent demonstration of the existence of identified mutations in members of the MAPK signaling pathway in the serum of 96% tested individuals harboring different tumors (Bettegowda et al., 2014). Given the particular association of deletions in this region and the intermediate but not high risk of cancer, MAPK1 might be a protooncogene contributing to PCa progression, metastasis and/or transition to CRPC, and its deletion might represent a lower risk of disease progression.

Finally, we found 20 genes mutated in at least 5% of patients. In contrast to PTEN, mutations in these genes are secondary. However, it has been noted that prostate cancer is commonly associated with diverse low frequency mutations (Armenia et al., 2018). Nonetheless, six of the identified genes (PHOX2a, NFYC, EST2, EIF251, SSRP1 and PARP1) were found associated with significant impact on patient survival.

The present analysis cannot distinguish between epithelial and stromal contributions to gene expression. Therefore, it is possible that some of the genes studied are expressed in the stroma. As a matter of fact, a previous approach from our laboratory has identified stromal and epithelial subsets of transcription factors (Nishan et al., 2019).

In conclusion, this work provides insights into the vastness of physiological pathways involving multiple regulatory interactions among genes needed to adjust prostate biology to the reduced androgen levels achieved by surgical or chemical castration. These results are expected to help us understand the idiosyncrasies of prostate cancer.

Acknowledgments

UN was recipient of a TWAS/CNPq fellowship. This work was funded by a grant from FAPESP (Nr. 2009/16150-6). The authors thank the Laboratory for Bioinformatics at LNBio/CNPEM for assistance with the use of IPA.

Notes

Associate Editor: Anamaria Camargo

Conflict of interest

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

Authors contributions

UN conducted the experiments, analyzed the data and wrote the manuscript. RRR, DMD and GOB conducted the experiments and analyzed the data. HFC conceived the study, analyzed the data and wrote the manuscript. All authors approved the final version.

References

1 

    Abate-Shen C, Shen MM 2000. . Molecular genetics of prostate cancer. Genes Dev 14: , pp.2410-34
Abate-Shen C and Shen MM (2000) Molecular genetics of prostate cancer. Genes Dev 14: 2410-34.

2 

    Abida W, Cyrta J, Heller G, Prandi D, Armenia J, Coleman I, Cieslik M, Benelli M, Robinson D, Van Allen EM, 2019. . Genomic correlates of clinical outcome in advanced prostate cancer. Proc Natl Acad Sci U S A 116: , pp.11428-36
Abida W, Cyrta J, Heller G, Prandi D, Armenia J, Coleman I, Cieslik M, Benelli M, Robinson D, Van Allen EM et al. (2019) Genomic correlates of clinical outcome in advanced prostate cancer. Proc Natl Acad Sci U S A 116:11428-36.

3 

    Antonioli E, Cardoso AB, Carvalho HF 2007. . Effects of long-term castration on the smooth muscle cell phenotype of the rat ventral prostate. J. Androl 28: , pp.777-783
Antonioli E, Cardoso AB and Carvalho HF (2007) Effects of long-term castration on the smooth muscle cell phenotype of the rat ventral prostate. J. Androl 28:777-783.

4 

    Antonioli E, Della-Colleta HHM, Carvalho HF 2004. . Smooth muscle cell behavior in the ventral prostate of castrated rats. J Androl 25: , pp.50-56
Antonioli E, Della-Colleta HHM and Carvalho HF (2004) Smooth muscle cell behavior in the ventral prostate of castrated rats. J Androl 25:50-56.

5 

    Armenia J, Wankowicz SAM, Liu D, Gao J, Kundra R, Reznik E, Chatila WK, Chakravarty D, Han GC, Coleman I, 2018. . The long tail of oncogenic drivers in prostate cancer. Nat Genet 51: , pp.1194
Armenia J, Wankowicz SAM, Liu D, Gao J, Kundra R, Reznik E, Chatila WK, Chakravarty D, Han GC, Coleman I et al. (2018) The long tail of oncogenic drivers in prostate cancer. Nat Genet 51:1194.

6 

    Baca SC, Prandi D, Lawrence MS, Mosquera JM, Romanel A, Drier Y, Park K, Kitabayashi N, MacDonald TY, Ghandi M, 2013. . Punctuated evolution of prostate cancer genomes. Cell 153: , pp.666-677
Baca SC, Prandi D, Lawrence MS, Mosquera JM, Romanel A, Drier Y, Park K, Kitabayashi N, MacDonald TY, Ghandi M et al. (2013) Punctuated evolution of prostate cancer genomes. Cell 153:666–677.

7 

    Barbosa GO, Silva JAF, Siqueira-Berti A, Nishan U, Rosa-Ribeiro R, Oliveira SBP, Baratti MO, Ferrucci D, Santana JCO, Damas-Souza DM, 2019. . Castration-induced prostate epithelial cell death apoptosis results from targeted oxidative stress attack of M1 142-macrophages. J Cell Physiol 234: , pp.19048-19058
Barbosa GO, Silva JAF, Siqueira-Berti A, Nishan U, Rosa-Ribeiro R, Oliveira SBP, Baratti MO, Ferrucci D, Santana JCO, Damas-Souza DM et al. (2019) Castration-induced prostate epithelial cell death apoptosis results from targeted oxidative stress attack of M1 142-macrophages. J Cell Physiol 234:19048-19058.

8 

    Bebas P, Goodall CP, Majewska M, Neumann A, Giebultowicz JM, Chappell PE 2009. . Circadian clock and output genes are rhythmically expressed in extratesticular ducts and accessory organs of mice. FASEB J 23: , pp.523-533
Bebas P, Goodall CP, Majewska M, Neumann A, Giebultowicz JM and Chappell PE (2009) Circadian clock and output genes are rhythmically expressed in extratesticular ducts and accessory organs of mice. FASEB J 23:523-533.

9 

    Bettegowda C, Sausen M, Leary RJ, Kinde I, Wang Y, Agrawal N, Bartlett BR, Wang H, Luber B, Alani RM, 2014. . Detection of circulating tumor DNA in early- and late-stage human malignancies. Sci Transl Med 6: , pp.224ra24
Bettegowda C, Sausen M, Leary RJ, Kinde I, Wang Y, Agrawal N, Bartlett BR, Wang H, Luber B, Alani RM et al. (2014) Detection of circulating tumor DNA in early- and late-stage human malignancies. Sci Transl Med 6:224ra24.

10 

    Desai KV, Michalowska AM, Kondaiah P, Ward JM, Shih JH, Green JE 2004. . Gene expression profiling identifies a unique androgen-mediated inflammatory/immune signature and a PTEN (phosphatase and tensin homolog deleted on chromosome 10)-mediated apoptotic response specific to the rat ventral prostate. Mol Endocrinol 18: , pp.2895-28907
Desai KV, Michalowska AM, Kondaiah P, Ward JM, Shih JH and Green JE (2004) Gene expression profiling identifies a unique androgen-mediated inflammatory/immune signature and a PTEN (phosphatase and tensin homolog deleted on chromosome 10)-mediated apoptotic response specific to the rat ventral prostate. Mol Endocrinol 18:2895-28907.

11 

    Eagle LR, Yin X, Brothman AR, Williams BJ, Atkin NB, Prochownik EV 1995. . Mutation of the MXI1 gene in prostate cancer. Nat Genet 9: , pp.249-255
Eagle LR, Yin X, Brothman AR, Williams BJ, Atkin NB and Prochownik EV (1995) Mutation of the MXI1 gene in prostate cancer. Nat Genet 9:249-255 .

12 

    Feldman BJ, Feldman D 2001. . The development of androgen-independent prostate cancer. Nat Rev Cancer 1: , pp.34-45
Feldman BJ and Feldman D (2001) The development of androgen-independent prostate cancer. Nat Rev Cancer 1:34-45.

13 

    Haffner MC, Mosbruger T, Esopi DM, Fedor H, Heaphy CM, Walker DA, Adejola N, Gürel M, Hicks J, Meeker AK, 2013. . Tracking the clonal origin of lethal prostate cancer. J Clin Invest 123: , pp.4918-4922
Haffner MC, Mosbruger T, Esopi DM, Fedor H, Heaphy CM, Walker DA, Adejola N, Gürel M, Hicks J, Meeker AK et al. (2013) Tracking the clonal origin of lethal prostate cancer. J Clin Invest 123:4918-4922.

14 

    Hermans KG, Van Alewijk DC, Veltman JA, Van Weerden W, Van Kessel AG, Trapman J 2004. . Loss of a small region around the PTEN locus is a major chromosome 10 alteration in prostate cancer xenografts and cell lines. Genes Chromosom Cancer 39: , pp.171-184
Hermans KG, Van Alewijk DC, Veltman JA, Van Weerden W, Van Kessel AG, and Trapman J (2004) Loss of a small region around the PTEN locus is a major chromosome 10 alteration in prostate cancer xenografts and cell lines. Genes Chromosom Cancer 39:171-184.

15 

    Holcomb IN, Young JM, Coleman IM, Salari K, Grove DI, Li H, True LD, Roudier MP, Morrissey CM, Higano CS, 2009. . Comparative analyses of chromosome alterations in soft-tissue metastases within and across patients with castration-resistant prostate cancer. Cancer Res 69: , pp.7793-7802
Holcomb IN, Young JM, Coleman IM, Salari K, Grove DI, Li H, True LD, Roudier MP, Morrissey CM, Higano CS et al. (2009) Comparative analyses of chromosome alterations in soft-tissue metastases within and across patients with castration-resistant prostate cancer. Cancer Res 69:7793-7802.

16 

    Huang DW, Sherman BT, Tan Q, Collins JR, Alvord WG, Roayaei J, Stephens R, Baseler MW, Lane HC, Lempicki RA 2007. . The DAVID Gene Functional Classification Tool: A novel biological module-centric algorithm to functionally analyze large gene lists. Genome Biol 8: , pp.R183
Huang DW, Sherman BT, Tan Q, Collins JR, Alvord WG, Roayaei J, Stephens R, Baseler MW, Lane HC and Lempicki RA (2007a) The DAVID Gene Functional Classification Tool: A novel biological module-centric algorithm to functionally analyze large gene lists. Genome Biol 8:R183.

17 

    Huang DW, Sherman BT, Tan Q, Kir J, Liu D, Bryant D, Guo Y, Stephens R, Baseler MW, Lane HC, 2007. . DAVID Bioinformatics Resources: Expanded annotation database and novel algorithms to better extract biology from large gene lists. Nucleic Acids Res 35: Suppl 2, pp.W169-W175
Huang DW, Sherman BT, Tan Q, Kir J, Liu D, Bryant D, Guo Y, Stephens R, Baseler MW, Lane HC et al. (2007b) DAVID Bioinformatics Resources: Expanded annotation database and novel algorithms to better extract biology from large gene lists. Nucleic Acids Res 35 Suppl 2:W169–W175.

18 

    Iafrate AJ, Feuk L, Rivera MN, Listewnik ML, Donahoe PK, Qi Y, Scherer SW, Lee C 2004. . Detection of large-scale variation in the human genome. Nat Genet 36: , pp.949-951
Iafrate AJ, Feuk L, Rivera MN, Listewnik ML, Donahoe PK, Qi Y, Scherer SW and Lee C (2004) Detection of large-scale variation in the human genome. Nat Genet. 36:949-951.

19 

    Ishkanian AS, Mallof CA, Ho J, Meng A, Albert M, Syed A, Van Der Kwast T, Milosevic M, Yoshimoto M, Squire JA, 2009. . High-resolution array CGH identifies novel regions of genomic alteration in intermediate-risk prostate cancer. Prostate 69: , pp.1091-1100
Ishkanian AS, Mallof CA, Ho J, Meng A, Albert M, Syed A, Van Der Kwast T, Milosevic M, Yoshimoto M, Squire JA et al. (2009) High-resolution array CGH identifies novel regions of genomic alteration in intermediate-risk prostate cancer. Prostate 69:1091-1100.

20 

    Kim JH, Dhanasekaran SM, Mehra R, Tomlins SA, Gu W, Yu J, Kumar-Sinha C, Cao X, Dash A, Wang L, 2007. . Integrative analysis of genomic aberrations associated with prostate cancer progression. Cancer Res 67: , pp.8229-8239
Kim JH, Dhanasekaran SM, Mehra R, Tomlins SA, Gu W, Yu J, Kumar-Sinha C, Cao X, Dash A, Wang L et al. (2007) Integrative analysis of genomic aberrations associated with prostate cancer progression. Cancer Res. 67:8229-8239.

21 

    Kurita T, Wang YZ, Donjacour AA, Zhao C, Lydon JP, O’Malley BW, Isaacs JT, Dahiya R, Cunha GR 2001. . Paracrine regulation of apoptosis by steroid hormones in the male and female reproductive system. Cell Death Differ 8: , pp.192-200
Kurita T, Wang YZ, Donjacour AA, Zhao C, Lydon JP, O’Malley BW, Isaacs JT, Dahiya R, and Cunha GR (2001) Paracrine regulation of apoptosis by steroid hormones in the male and female reproductive system. Cell Death Differ. 8:192-200.

22 

    Kwong J, Choi HL, Huang Y, Chan FL 1999. . Ultrastructural and biochemical observations on the early changes in apoptotic epithelial cells of the rat prostate induced by castration. Cell Tissue Res 298: , pp.123-136
Kwong J, Choi HL, Huang Y and Chan FL (1999) Ultrastructural and biochemical observations on the early changes in apoptotic epithelial cells of the rat prostate induced by castration. Cell Tissue Res 298:123-136.

23 

    Liu J, Lichtenberg T, Hoadley KA, Poisson LM, Lazar AJ, Cherniack AD, Kovatich AJ, Benz CC, Levine DA, Lee AV, 2018. . An integrated PCGA pan-cancer clinical data resource to drive high quality survival outcome analytics. Cell 173: , pp.400-16
Liu J, Lichtenberg T, Hoadley KA, Poisson LM, Lazar AJ, Cherniack AD, Kovatich AJ, Benz CC, Levine DA, Lee AV et al. (2018) An integrated PCGA pan-cancer clinical data resource to drive high quality survival outcome analytics. Cell 173:400-16.

24 

    Martinez Barbera JP, Clements M, Thomas P, Rodriguez T, Meloy D, Kioussis D, Beddington RSP 2000. . The homeobox gene Hex is required in definitive endodermal tissues for normal forebrain, liver and thyroid formation. Development 127: , pp.2433-2445
Martinez Barbera JP, Clements M, Thomas P, Rodriguez T, Meloy D, Kioussis D and Beddington RSP (2000) The homeobox gene Hex is required in definitive endodermal tissues for normal forebrain, liver and thyroid formation. Development 127:2433-2445.

25 

    Meloche S, Pouysségur J 2007. . The ERK1/2 mitogen-activated protein kinase pathway as a master regulator of the G1- to S-phase transition. Oncogene 26: , pp.3227-3239
Meloche S and Pouysségur J (2007) The ERK1/2 mitogen-activated protein kinase pathway as a master regulator of the G1- to S-phase transition. Oncogene 26:3227-3239.

26 

    Mukherjee R, McGuinness DH, McCall P, Underwood MA, Seywright M, Orange C, Edwards J 2011. . Upregulation of MAPK pathway is associated with survival in castrate-resistant prostate cancer. Br J Cancer 104: , pp.1920-1928
Mukherjee R, McGuinness DH, McCall P, Underwood MA, Seywright M, Orange C and Edwards J. (2011). Upregulation of MAPK pathway is associated with survival in castrate-resistant prostate cancer. Br J Cancer 104:1920-1928.

27 

    Nishan U, Rosa-Ribeiro R, Cesar CL, Carvalho HF 2019. . Transcription regulators are transiently expressed during the prostate gland adaptation to the hypoandrogenic enviroment. Histol Histopathol 34: , pp.1025-1036
Nishan U, Rosa-Ribeiro R, Cesar CL and Carvalho HF. (2019) Transcription regulators are transiently expressed during the prostate gland adaptation to the hypoandrogenic enviroment. Histol Histopathol 34:1025-1036.

28 

    Nyquist MD, Li Y, Hwang TH, Manlove LS, Vessella RL, Silverstein KAT, Voytas DF, Dehm SM 2013. . TALEN-engineered AR gene rearrangements reveal endocrine uncoupling of androgen receptor in prostate cancer. Proc Natl Acad Sci 110: , pp.17492-17497
Nyquist MD, Li Y, Hwang TH, Manlove LS, Vessella RL, Silverstein KAT, Voytas DF and Dehm SM (2013) TALEN-engineered AR gene rearrangements reveal endocrine uncoupling of androgen receptor in prostate cancer. Proc Natl Acad Sci 110:17492-17497.

29 

    Prochownik EV, Grove LE, Deubler D, Zhu XL, Stephenson RA, Rohr LR, Yin X, Brothman AR 1998. . Commonly occurring loss and mutation of the MXII gene in prostate cancer. Genes Chromosomes Cancer 22: , pp.295-304
Prochownik EV, Grove LE, Deubler D, Zhu XL, Stephenson RA, Rohr LR, Yin X and Brothman AR (1998) Commonly occurring loss and mutation of the MXII gene in prostate cancer. Genes Chromosomes Cancer 22:295-304.

30 

    Qi C, Gery S, Dashti A, Dong Y, Yan Z, Jiang G, Koeffler HP 2009. . A role for the clock gene Per1 in prostate cancer. Cancer Res 69: , pp.7619-7625
Qi C, Gery S, Dashti A, Dong Y, Yan Z, Jiang G and Koeffler HP (2009) A role for the clock gene Per1 in prostate cancer. Cancer Res 69:7619-7625.

31 

    Rosa-Ribeiro R, Barbosa GO, Kühne F, Carvalho HF 2014. . Desquamation is a novel phenomenon for collective prostate epithelial cell deletion after castration. Histochem. Cell Biol. 141: , pp.213-220
Rosa-Ribeiro R, Barbosa GO, Kühne F and Carvalho HF (2014a) Desquamation is a novel phenomenon for collective prostate epithelial cell deletion after castration. Histochem. Cell Biol. 141:213-220.

32 

    Rosa-Ribeiro R, Nishan U, Vidal RO, Barbosa GO, Reis LO, Cesar CL, Carvalho HF 2014. . Transcription factors involved in prostate gland adaptation to androgen deprivation. PLoS One 9: , pp.e97080
Rosa-Ribeiro R, Nishan U, Vidal RO, Barbosa GO, Reis LO, Cesar CL and Carvalho HF (2014b) Transcription factors involved in prostate gland adaptation to androgen deprivation. PLoS One 9:e97080.

33 

    Roy AK, Lavrovsky Y, Song CS, Chen S, Jung MH, Velu NK, Bi BY, Chatterjee B 1998. . Regulation of androgen action. Vitam Horm 55: , pp.309-332
Roy AK, Lavrovsky Y, Song CS, Chen S, Jung MH, Velu NK, Bi BY and Chatterjee B (1998) Regulation of androgen action. Vitam Horm 55:309-332.

34 

    Shen MM, Abate-Shen C 2010. . Molecular genetics of prostate cancer: new prospects for old challenges. Genes Dev 24: , pp.1967-2000
Shen MM and Abate-Shen C (2010) Molecular genetics of prostate cancer: new prospects for old challenges. Genes Dev. 24:1967-2000.

35 

    Silva JAF, Bruni-Cardoso A, Augusto TM, Damas-SOuza DM, Barbosa GO, Felisbino SL, Stach-Machado DR, Carvalho HF 2018. . Macrophage roles in the clearance of apoptotic cells and control of inflammation in the prostate gland after castration. Prostate 78: , pp.95-103
Silva JAF, Bruni-Cardoso A, Augusto TM, Damas-SOuza DM, Barbosa GO, Felisbino SL, Stach-Machado DR and Carvalho HF. (2018) Macrophage roles in the clearance of apoptotic cells and control of inflammation in the prostate gland after castration. Prostate 78:95-103.

36 

    Singh KK, Desouki MM, Franklin RB, Costello LC 2006. . Mitochondrial aconitase and citrate metabolism in malignant and nonmalignant human prostate tissues. Mol Cancer 5: , pp.14
Singh KK, Desouki MM, Franklin RB and Costello LC (2006) Mitochondrial aconitase and citrate metabolism in malignant and nonmalignant human prostate tissues. Mol Cancer 5:14.

37 

    Sunkel B, Wang Q 2014. . Looking beyond androgen receptor signaling in the treatment of advanced prostate cancer. Adv Androl 2014: , pp.1-9
Sunkel B and Wang Q (2014) Looking beyond androgen receptor signaling in the treatment of advanced prostate cancer. Adv Androl 2014:1-9.

38 

    Taj MM, Tawil RJ, Engstrom LD, Zeng Z, Hwang C, Sanda MG, Wechsler DS 2001. . Mxi1, a Myc antagonist, suppresses proliferation of DU145 human prostate cells. Prostate 47: , pp.194-204
Taj MM, Tawil RJ, Engstrom LD, Zeng Z, Hwang C, Sanda MG and Wechsler DS (2001) Mxi1, a Myc antagonist, suppresses proliferation of DU145 human prostate cells. Prostate 47:194-204.

39 

    Tomlins SA, Mehra R, Rhodes DR, Cao X, Wang L, Dhanasekaran SM, Kalyana-Sundaram S, Wei JT, Rubin MA, Pienta KJ, 2007. . Integrative molecular concept modeling of prostate cancer progression. Nat Genet 39: , pp.41-51
Tomlins SA, Mehra R, Rhodes DR, Cao X, Wang L, Dhanasekaran SM, Kalyana-Sundaram S, Wei JT, Rubin MA, Pienta KJ et al. (2007). Integrative molecular concept modeling of prostate cancer progression. Nat Genet 39:41-51.

40 

    Vilamaior PSL, Felisbino SL, Taboga SR, Carvalho HF 2000. . Collagen fiber reorganization in the rat ventral prostate following androgen deprivation: A possible role for smooth muscle cells. Prostate 45: , pp.253-258
Vilamaior PSL, Felisbino SL, Taboga SR and Carvalho HF (2000) Collagen fiber reorganization in the rat ventral prostate following androgen deprivation: A possible role for smooth muscle cells. Prostate 45:253-258.

41 

    Wang Z, Tufts R, Haleem R, Cai X 1997. . Genes regulated by androgen in the rat ventral prostate. Proc Natl Acad Sci U S A 94: , pp.12999-13004
Wang Z, Tufts R, Haleem R and Cai X (1997) Genes regulated by androgen in the rat ventral prostate. Proc Natl Acad Sci U S A. 94:12999-13004.

42 

    Zhu Y, Stevens RG, Hoffman AE, FitzGerald LM, Kwon EM, Ostrander EA, Davis S, Zheng T, Stanford JL 2009. . Testing the circadian gene hypothesis in prostate cancer: A population-based case-control study. Cancer Res 69: , pp.9315-9322
Zhu Y, Stevens RG, Hoffman AE, FitzGerald LM, Kwon EM, Ostrander EA, Davis S, Zheng T and Stanford JL (2009) Testing the circadian gene hypothesis in prostate cancer: A population-based case-control study. Cancer Res 69:9315-9322.

Appendices

Supplementary material

The following online material is available for this article:

 - Functional associateions among the selected genes

https://www.researchpad.co/tools/openurl?pubtype=article&doi=10.1590/1678-4685-GMB-2018-0362&title=Transcriptional regulators and regulatory pathways involved in prostate gland adaptation to a hypoandrogen environment&author=Umar Nishan,Rafaela da Rosa-Ribeiro,Danilo Marchete Damas-Souza,Guilherme Oliveira Barbosa,Hernandes F. Carvalho,&keyword=Androgens,Arnt,castration,prostate,transcription factors,&subject=Genomics and Bioinformatics,