Vitamin D Signaling Suppresses Early Prostate

Published OnlineFirst April 26, 2019; DOI: 10.1158/1940-6207.CAPR-18-0401

Research Article Cancer Prevention Research Vitamin D Signaling Suppresses Early Prostate

Carcinogenesis in TgAPT121 Mice James C. Fleet1,2, Pavlo L. Kovalenko1, Yan Li1, Justin Smolinski3, Colleen Spees3, Jun-Ge Yu3, Jennifer M.Thomas-Ahner3, Min Cui1, Antonio Neme4, Carsten Carlberg5, and Steven K. Clinton3,6

Abstract

We tested whether lifelong modiﬁcation of vitamin and improved bone mass but increased the inci- D signaling can alter the progression of early prostate dence of adenocarcinoma. Analysis of the VDR carcinogenesis in studies using mice that develop high- cistrome in RWPE1 prostate epithelial cells revealed grade prostatic intraepithelial neoplasia that is similar vitamin D–mediated regulation of multiple cancer- to humans. Two tissue-limited models showed that relevant pathways. Our data support the hypothesis prostate vitamin D receptor (VDR) loss increased that the loss of vitamin D signaling accelerates prostate carcinogenesis. In another study, we fed diets the early stages of prostate carcinogenesis, and our with three vitamin D3 levels (inadequate ¼ 25 IU/kg results suggest that different dietary requirements diet, adequate for bone health ¼ 150 IU/kg, or high ¼ may be needed to support prostate health or max- 1,000 IU/kg) and two calcium levels (adequate for imize bone mass. bone health ¼ 0.5% and high ¼ 1.5%). Dietary vitamin D caused a dose-dependent increase in serum Signiﬁcance: This work shows that disrupting vita- 25-hydroxyvitamin D levels and a reduction in the min D signaling through diet or genetic deletion percentage of mice with adenocarcinoma but did increases early prostate carcinogenesis through multiple not improve bone mass. In contrast, high calcium pathways. Higher-diet vitamin D levels are needed for suppressed serum 1,25-dihydroxyvitamin D levels cancer than bone.

Introduction conditional (4, 5), and negative (6, 7) effects of increased vitamin D status on prostate cancer. More recently, the Serum 25-hydroxyvitamin D (25OHD) concentrations VITAL trial showed that 2,000 IU vitamin D per day over a beyond those needed for bone health may reduce cancer 3 median follow-up of 5.3 years reduced the HR for prostate risk (1). However, the evidence relating vitamin D status to cancer to 0.88 (8). However, the intervention was short prostate cancer risk in humans is mixed. Early association relative to the process of human prostate carcinogenesis, studies with small sample size were supportive (2), but and because few cases were documented, the comparison larger population-based studies have shown protective (3), was underpowered and not significant (8). There is also evidence that high-dietary calcium (Ca) intake, a nutrient known to suppress renal production of the vitamin D 1Department of Nutrition Science, Purdue University, West Lafayette, Indiana. 2Purdue University Center for Cancer Research, Purdue University, West hormone, 1,25-dihydroxyvitamin D (1,25(OH)2D), may Lafayette, Indiana. 3Division of Medical Oncology, College of Medicine, increase prostate cancer risk (9). Thus, although the overall Columbus, Ohio. 4Instituto de Investigaciones en Matematicas Aplicadas y en body of literature suggests a protective role for vitamin D Sistemas-Merida, Universidad Nacional Autonoma de Mexico, Yucatan, Mexico. on prostate cancer risk, the exact nature of the relationship, 5School of Medicine, Institute of Biomedicine, University of Eastern Finland, Kuopio, Finland. 6The Ohio State University Comprehensive Cancer Center, the protective serum vitamin D metabolite concentrations, Columbus, Ohio. and the mechanism of protective action have been elusive. Note: Supplementary data for this article are available at Cancer Prevention Animal and cell studies have provided essential proof for Research Online (http://cancerprevres.aacrjournals.org/). a direct causal relationship between vitamin D and prostate Corresponding Author: James C. Fleet, Purdue University, 700 West State St., cancer development. In cultured prostate epithelial cells West Lafayette, IN 47907-2059. Phone: 765-494-0302; Fax: 765-494-0906; (PEC) or prostate cancer cells, 1,25(OH)2D suppresses fl E-mail: [email protected] proliferation, induces apoptosis, and promotes differenti- Cancer Prev Res 2019;12:1–56 ation (10). Consistent with these data, vitamin D deficien- doi: 10.1158/1940-6207.CAPR-18-0401 cy (11) and vitamin D receptor (VDR) gene deletion (12) 2019 American Association for Cancer Research. increase, whereas injections with 1,25(OH)2D or vitamin

www.aacrjournals.org OF1

Fleet et al.

D analogs suppress (13) prostate tumor growth in various were used for microarray analysis, whereas the dorsolateral animal models. However, experiments linking prostate prostate lobes were assessed for Vdr mRNA level by qPCR cancer development to human-relevant ranges of vitamin and VDR protein level by Western blot analysis. Prostates þ/ D status or Ca intake are limited (14–16). from a second 12-week-old cohort of 10 TgAPT121 and / Previously, we determined the vitamin D3 intake neces- 4 TgAPT121 mice were used for histology and IHC. sary to model human vitamin D status in mice (17) and fi found that intake as low as 100 IU vitamin D3/kg diet is Experiment 2: PEC-speci c Vdr gene deletion on prostate þ/ fl/fl þ/ sufficient to maintain the traditional vitamin D functions cancer. The TgAPT121 , Vdr , and PB-Cre lines were of bone growth and mineralization. Using this informa- crossed to make mice with PEC-specific Vdr gene deletion þ/ fl/fl þ/ tion, we showed that dietary vitamin D deficiency (PEC-VDR KO: TgAPT121 , Vdr , PB-Cre ; n ¼ 32) increased PEC proliferation, reduced PEC apoptosis, and and those with normal Vdr gene status (Cre-negative lit- þ/ fl/fl / ¼ increased the incidence of high-grade prostatic intraepithe- termate controls: TgAPT121 , Vdr , Pb-Cre ; n 33). lial neoplasia (HGPIN) lesions in mice (18). Here, we At 28 weeks of age, mice were sacrificed and the prostate report studies that extend our earlier work and directly was harvested and prepared for histology. address whether lifelong variation in the dietary levels of vitamin D and Ca can modify early-stage prostate cancer. In Experiment 3: Whole prostate Vdr gene deletion on prostate addition, we examine the importance of signaling through cancer. TgAPT121 mice were crossed to Vdr knockout mice the VDR during early prostate carcinogenesis, and we with intestine-specific transgenic expression of a hemag- identify potential candidate genes mediating the action of glutinin-tagged human VDR to generate mice lacking VDR þ/ in all cells in the prostate (HV2-VDR KO: TgAPT121 , vitamin D on the PEC during carcinogenesis. þ HV2 / , VDR / ; n ¼ 23) and littermate controls þ/ þ/ þ/ ¼ Materials and Methods (TgAPT121 , HV2 , VDR ; n 27). Transgenic expression of VDR in the intestine prevents abnormal Ca Animals metabolism in Vdr knockout mice (20). At 26 weeks of age, þ/ In our studies, we used TgAPT121 mice (TgAPT121 , mice were sacrificed and the prostate was harvested and C57BL/6 DBA/2 background; ref. 19) bred to male prepared for histology. B6D2F1 mice (Jackson Laboratories). This model has probasin promoter–driven, PEC-specific expression of a Experiment 4: Impact of diet on prostate tumorigenesis. Male truncated SV40 Large T antigen protein that inactivates pRb TgAPT121 transgenic mice were generated at Purdue Uni- family proteins. It has strong histologic similarities to the versity and then shipped to Ohio State University at wean- earlier stages of human prostate cancer (i.e., demonstrating ing. Mice were randomly assigned to one of six AIN-76A– hyperplasia, PIN, and adenocarcinoma over 6 months). based diets with varying levels of dietary vitamin D3 Other mouse models used include mice with a floxed exon (25, 150, and 1,000 IU/kg diet) and Ca (0.5, 1.5%) in a fl fl 2 in the Vdr gene (Vdr / ; C57BL/6; ref. 18); mice with Cre 2 3 factorial design (n ¼ 34 mice per group). Diets and recombinase expression driven by the rat probasin pro- water were fed ad libitum. Mice were sacrificed at 28 weeks þ moter (PB-Cre / ; C57BL/6; NCI Mouse Models of Human of age when blood, the right femur, and the prostate was Cancer Consortium, # 01XF5); and Vdr knockout mice harvested. Serum was prepared from blood, and a ran- with intestine-specific transgenic expression of the human domly selected subset of samples was analyzed for vitamin VDR gene (C57BL/6; ref. 20). Mice were genotyped as D metabolites (n ¼ 10 per diet group). Intact femora were previously described (18–21). Mice were housed with a prepared for DEXA analysis (n ¼ 25–27 per diet group). 12-hour light/12-hour dark cycle, in shoebox cages with Prostates were processed for histology. individual ventilation. Lights were covered with a UVB filter (Pegasus Associates Lighting). Diets and water were Experiment 5: VDR chromatin immunoprecipitation fed ad libitum. Except for Experiment 4, mice were fed sequencing analysis. RWPE1 cells were cultured as we have AIN-76A diet (1,000 IU vitamin D3/kg diet; ref. 22). All described elsewhere (23). Cells were grown to 80% con- experiments were approved by either the Purdue Univer- fluence and then treated with vehicle or 10 mmol/L 1,25 sity's or the Ohio State University's Animal Care and Use (OH)2D for 3 hours. Chromatin immunoprecipitation Committees. sequencing (ChIP-seq) and ChIP-qPCR analysis was conducted as described below. Experimental design

Experiment 1: Characterization of TgAPT121 mice. For a Methods þ/ microarray study, male TgAPT121 (n ¼ 8) and Prostate preparation, histologic grading, and scoring. The / TgAPT121 (n ¼ 16) mice were used. After an overnight bladder, prostate, and seminal vesicles were removed fast, 12-week-old mice were sacriﬁced, and prostate lobes en block and processed for histology as described previous- were dissected on ice, snap-frozen in liquid nitrogen, and ly (18). Histologic examination of the anterior prostates stored at 80C. The anterior prostates from these mice was conducted using a modiﬁcation of established

OF2 Cancer Prev Res; 12(6) June 2019 Cancer Prevention Research

Vitamin D and Carcinogenesis

guidelines (24) that better reflect the diversity of the early [12 mmol/L NaCl, 0.5% Nonidet P-40, 0.02 mmol/L lesions in the TgAPT121 mouse prostate (see Supplemen- sodium orthovanadate, 5 mmol/L Tris-HCl (pH 8.0), and tary Table S1 and Supplementary Fig. S1 for a more one Protease Inhibitor Cocktail tablet per 5 mL (Roche)]. complete description of the grading criterion). Blinded Homogenates were centrifuged at 16,300 g at 4C for 15 assessment was conducted by one reviewer and was con- minutes, and supernatants were harvested. Equal amounts firmed by a second reviewer. of protein from 8 mice of each genotype were combined to Slides were digitized and the images were saved generate two pooled samples. Samples (100 mg protein) (ScanScope CS; Aperio). For experiments 2 and 3, lesions were analyzed by Western blot analysis as previously were marked on the digital images using the ImageScope described (23) and overnight incubation at 4C with Viewer (Aperio). The area covered by each lesion and the mouse anti-human VDR antibody (1:500; D-6, sc-13133; numbers of independent foci were determined, and the Santa Cruz Biotechnology) or mouse anti–b-actin data were exported for further analysis. Total tissue area ¼ antibody (1:5,000; AC-74; Sigma-Aldrich) followed by the sum of the areas of all lesion foci plus the area covered incubation with horseradish peroxidase–conjugated by normal epithelium. Percent area ¼ the proportion of goat-anti-mouse IgG light chain antibody (1:5,000; total tissue area exhibiting the features of a lesion. Lesion Jackson ImmunoResearch Laboratories) at room temper- count ¼ the number of independent lesion foci. Incidence ature for 1 hour. ¼ the percentage of mice in a group with a particular lesion. Average lesion size ¼ the ratio of the total area for a lesion Prostate RNA preparation. Frozen anterior prostates were type over the count of the lesion type. crushed under liquid N2, and RNA was extracted with the For experiment 4, slides for the 204 mice were visually SV Total RNA Isolation kit (Promega). Genomic DNA was examined and scored for their highest histologic pheno- removed by treating RNA with RQ1 RNase-Free DNase type. Data were reported for the percentage of mice in a (Promega) prior to repurification of RNA with the RNeasy group with adenocarcinoma as their most severe lesion. Mini Kit (Qiagen). RNA was quantified, and RNA integrity was assessed visually from agarose gels. IHC analyses. p53. Unstained sections were deparaffinized and rehydrated, followed by antigen retrieval for 30 min- Analysis of VDR mRNA levels. RNA was reverse transcribed utes (Antigen Retrieval Citra Plus Solution, BioGenex). into cDNA, and qPCR was conducted using primers and Endogenous peroxidase activity was quenched (DAKO PCR conditions as we have previously reported (20). Peroxidase Blocking Reagent, DAKO). Sections were incu- bated with rabbit polyclonal anti-mouse p53 antibody Microarray analysis. RNA isolation Equal amounts of RNA þ/ (NCL-p53-CM5p, 1:500; Leica Microsystems GmbH) for from 2 mice were pooled to yield 4 pooled TgAPT121 / 1 hour at room temperature, followed by 30 minutes with and 8 pooled TgAPT121 control samples. All samples DAKO EnVisionþ Labeled Polymer-HRP anti-rabbit IgG used for microarray analysis had high RNA Integrity secondary antibody. Antibody binding was visualized Number scores upon analysis on an Agilent Bioanalyzer. fi using the DAKO EnVisionþ DABþ Chromogen Solution, Gene expression pro ling was conducted using the Gen- and slides were counterstained with hematoxylin. eChip Mouse Gene 1.0 ST Array (ThermoFisher) and standard Affymetrix protocols at the Center for Medical Apoptosis. Unstained sections were evaluated by terminal Genomics at the Indiana University School of Medicine. deoxynucleotidyl transferase–mediated dUTP nick end Microarray data were preprocessed using the Bioconduc- labeling (TUNEL) staining with the ApopTag Plus Perox- tor package "xps," and data were normalized by the Robust idase In Situ Apoptosis Detection Kit following the man- Multichip Average method. The data for this experiment ufacturer's instructions (Millipore Corp.). Slides were are available in GEO (GSE50662). All arrays passed quality counterstained with methyl green. control tests and were included in the analyses. The preprocessed data were filtered using the detection above Quantification. Three to five representative, nonoverlapping background algorithm (P 0.05: score ¼ 1; 0.05 < P images without artifacts were captured by a high-resolution 0.07: score ¼ 0.6; P > 0.07: score ¼ 0), and a probeset was digital camera using bright field microscopy (Nikon called "present" if sum score was 4.8 (n ¼ 8, control) or ¼ þ/ ECLIPSE E800), and specific outcomes were analyzed 2.4 (n 4, TgAPT121 ). Probesets "present" in at least one using Image-Pro Plus 7.0 (Media Cybernetics, Inc.). A of the two genotype groups and those with a coefficient of labeling index was calculated as the number of p53 or variation < 50 were included in subsequent steps. We TUNEL-positive nuclei divided by total number of nuclei removed probesets with no gene annotation, and only in the fields examined: (%) ¼ L/(L þ C) 100, where one probeset was used when multiple probesets were L ¼ labeled nuclei and C ¼ unlabeled nuclei. available. The final list contained 21,332 probesets. Differentially expressed transcripts were identified using Measurement of prostate VDR protein level. Frozen dorso- the Significance Analysis of Microarray v3.09 (ref. 25; lateral prostates were homogenized on ice in lysis buffer log2-transformed data, two class, unpaired analysis,

www.aacrjournals.org Cancer Prev Res; 12(6) June 2019 OF3

Fleet et al.

900 permutations). Signiﬁcant transcripts (FDR < 5%) RWPE VDR ChIP-seq peaks were compared with with fold change > 1.5 were analyzed using MetaCore 1,25(OH)2D-regulated genes from: (i) RWPE1 cells (Clarivate Analytics). (10% FDR; ref. 34), (ii) mouse prostate stem cells (10% FDR), and (iii) 587 topologically associated domains Reanalysis of data from Tomlins and colleagues (26). Three (TAD) with vitamin D–regulated genes from THP-1 mono- normal PEC samples and 13 samples with HGPIN from cytes (35). Lists of genes with VDR-binding sites were GEO entry GSE6099 were examined for differential expres- examined for function and pathway enrichment using the sion with GEO2R (15% FDR). IPA pathway analysis tool (Qiagen).

Reanalysis of Maund and colleagues (27) data. CEL files from ChIP-PCR validation. Eighty percent confluent flasks of 1,25(OH)2D-treated (100 nmol/L; 6 or 48 hours) mouse RWPE1 cells were treated with vehicle or 10 mmol/L ¼ prostate epithelial stem cells (ref. 27; GEO entry 1,25(OH)2D for 3 hours (n 3/treatment). ChIP-qPCR GSE18993) were processed using RMAexpress (http:// was conducted as described above for peaks associated rmaexpress.bmbolstad.com/). One array was abnormal with MAPK6; CTSD/SYT8; DDIT4; KLF4; CA9; ETNK2/ and was excluded from the reanalysis (GSM469894). Dif- REN; and CYP24A1 (Supplementary Fig. S2). ferential gene expression was determined with Significance Analysis for Microarray (10% FDR) within BRB ArrayTools Statistical analysis of nonmicroarray data. Statistical anal- v 4.5.1 (28) on data filtered to remove the transcripts yses were conducted using SPSS for Windows-v.19.0 (IBM whose variance was in the bottom 50th percentile. Corp.). In all experiments, differences were considered significant at P < 0.05, and a trend was defined as P < Serum vitamin D metabolite analysis. Serum 1,25(OH)2D 0.10. Values are expressed as mean SEM of the non- and 25OHD were analyzed as previously described (20). transformed data. All sample size calculations were conducted for a ¼ 0.05 and b ¼ 0.8. In the tumor endpoint Bone analyses. The right femur from each mouse was studies, 22 mice per group are necessary to detect 60% stripped of muscle and analyzed for bone mineral density difference in the proportion of mice with adenocarcinoma (BMD; g/cm2) using DEXA (20). using a x2 test. For serum vitamin D metabolites, 10 mice per group are necessary to detect a significant main effect of ChIP-seq analysis for VDR-binding sites. Sample preparation diet or genotype of 100% based on a variance ¼ 75% of the and ChIP-seq. ChIP was conducted as previously difference between means (17). For bone, 17 mice are described (29) using normal rabbit IgG polyclonal anti- necessary to detect a significant main effect of diet or body (12-370, Millipore Sigma) or anti-VDR polyclonal genotype of 4% based on a variance ¼ difference between antibody (C-20, sc-1008, Santa Cruz Biotechnology, Inc.). means (20). ChIP-PCR (29) was used to identify three replicates with The distribution of all continuous variables was exam- ligand-inducible VDR binding to the -252 to -51 bp region ined by the Shapiro–Wilk test, and normal Q–Q plots and of the CYP24A1 gene promoter and no VDR binding to the log- or square root transformations were used to achieve a GAPDH gene promoter. These were pooled by treatment to normal distribution. If transformation did not normalize generate three samples (control treated/VDR IP, 1,25 the distribution, the nonparametric Mann–Whitney U test þ (OH)2D treated/VDR IP, and a pooled control 1,25 was used. IHC and TUNEL data were analyzed by one-way (OH)2D treated/IgG IP). ChIP was repeated on these ANOVA, followed by the Holm–Sidak step-down multiple samples and sent to the Michael Smith Genome Science comparison test to detect differences among lesion types. Centre for sequencing (library construction on 100–300 bp For the data from the mouse VDR deletion experiments fragments, 36 bp single-end reads, Illumina sequencing; and for Vdr mRNA levels, a Student t test was used to assess ref. 30). The data for two lanes for each sample were pooled differences between genotype groups. For categorical vari- for analysis (GEO dataset GSE116843). ables in the cancer studies, the x2 test and Fisher exact tests were used to detect difference in frequencies. For the Bioinformatic analysis. ChIP-seq data were analyzed as noncancer endpoints in the diet study, data were analyzed reported elsewhere (31). We also reanalyzed ChIP-seq by two-way ANOVA to account for main effects and inter- data from THP-1 cells ( 100 nmol/L 1,25(OH)2D, 2 actions between the dietary interventions. or 24 hours), LPS-treated THP-1 cells (100 nmol/L 1,25(OH) D, 24 hours), LS180 cells (100 nmol/L 2 Results 1,25(OH)2D, 3 hours), and LX2 cells (100 nmol/L 1,25(OH)2D, 16 hours; ref. 32; and GEO entry GSE53041; There were no observed adverse effects of the dietary multicell analysis upon request). ChIP-seq peaks were interventions or genetic modifications on the growth, body annotated to the nearest neighbor gene, and a search region weight, or appearance of the mice. of 100 bp from the peak summit was examined for At 3 months of age, the anterior prostate of TgAPT121 transcription factor–binding motifs with HOMER (33). mice is filled with a mix of low-grade (PIN I and II) and

OF4 Cancer Prev Res; 12(6) June 2019 Cancer Prevention Research

Vitamin D and Carcinogenesis

high-grade (PIN III and IV) lesions (18). We used micro- protein p53). Increased p53 expression and apoptosis array analysis to determine whether the molecular pheno- during the development of HGPIN lesions in TgAPT121 type of the anterior prostate in TgAPT121 mice adequately mouse prostate was conﬁrmed by IHC (Fig. 1A and B). models the features of human HGPIN, the precursor lesion Collectively, these analyses support the use of the ante- to human prostate cancer (36). TgAPT121 mouse prostates rior prostate from TgAPT121 as a model of human had 3,440 upregulated and 2,087 downregulated genes HGPIN. compared with normal prostate (data available upon Vdr mRNA and protein levels were elevated in the þ/ request). There were similarities in the TgAPT121 mouse dorsolateral lobe of 12-week-old TgAPT121 mice prostate transcriptome compared with published studies (Fig. 1C and D), and the microarray analysis shows comparing normal prostate epithelium with HGPIN in that Vdr mRNA levels are also elevated in the anterior lobe humans, i.e., 17 of the 21 genes reported individually (2.1-fold higher, Supplementary Table S2). To test the in the literature, 12 of 72 genes from ref. 37, and 97 of importance of VDR signaling in the progression of early þ/ 526 genes from ref. 26 (Supplementary Table S2). Meta- stages of prostate cancer in TgAPT121 mice, we con- Core analysis identiﬁed enriched pathways that link ducted two VDR gene deletion studies. to prostate carcinogenesis within the TgAPT121 transcrip- To test the role of VDR in the PEC, we used PEC-VDR KO tome including those regulating the DNA damage mice. At 28 weeks of age, PIN III was the most prevalent response, p53 signaling, apoptosis, TNF receptor signal- lesion in the anterior prostate of TgAPT121 mice (60%– ing, TGFb signaling, and signaling through the androgen 70% of total prostate tissue area), and the next most receptor (Supplementary Table S3). Bioinformatic anal- common lesion types were PIN II and PIN IV (Fig. 2). ysis also found transcription factor networks centered on Compared with controls, the PEC-VDR KO mice had more proto-oncogenic transcription factors (e.g., Myc, Ap-1, independent adenocarcinoma foci (P ¼ 0.003, Fig. 2A), Creb1, Nfkb1, Ar, Egr1, Hnf4a)andTrp53 (encoding the the average size of the adenocarcinoma foci was larger

Figure 1. Impact of prostate phenotype progression on (A) P53 levels and (B) TUNEL staining. Bars are mean SEM (n ¼ 5–28 images/lesion type, n ¼ 4

Control, 10 TgAPT121 mice). Data were analyzed by ANOVA followed by the Holm–Sidak pairwise multiple comparisons. Bars without a common letter superscript differ signiﬁcantly (P < 0.05). C, Vdr mRNA level in the dorsolateral prostates of TgAPT121 and control mice. Bars are mean SEM. (, P < 0.01, n ¼ 8/genotype, Student t test). D, VDR protein level in pooled dorsolateral prostates (n ¼ 8/pool).

www.aacrjournals.org Cancer Prev Res; 12(6) June 2019 OF5

Fleet et al.

Figure 2. Histologic grading of the anterior prostates from 28-week-old PEC-VDR KO (n ¼ 32) and control (PB-Cre, VDRf/f; n ¼ 33) mouse prostates. A, Count of independent lesion foci per section; (B) average lesion size; (C) area of each lesion; and (D) percent area of each lesion type. Bars are mean SEM (, P < 0.05 vs. Control). Other comparisons where the difference approached signiﬁcance are reported with the measured P value.

(P ¼ 0.005, Fig. 2B), and the area covered by adenocarci- (P ¼ 0.032), microinvasion (P ¼ 0.004), and adenocarcinoma was larger (P ¼ 0.004, Fig. 2C). In addition, a higher noma (P ¼ 0.026; Fig. 3A). There was a trend toward larger percentage of PEC-VDR KO mice developed focal adeno- adenocarcinoma lesion size (P ¼ 0.062, Fig. 3B) and carcinoma than the control mice (73% vs. 38%; P ¼ 0.004). a greater area occupied by adenocarcinoma (P ¼ Finally, although the percent area of the total prostate 0.047, Fig. 3C) in the HV2-VDR KO mice. In addition, tissue occupied by adenocarcinoma was low, it was 2.8- more HV2-VDR KO mice had adenocarcinoma than the fold higher in the PEC-VDR KO mice than the control mice control mice (48% vs. 19%; P ¼ 0.027). Finally, HV2-VDR (P ¼ 0.003, Fig. 2D). KO mice had higher percent area of adenocarcinoma and Mice with intestine-speciﬁc expression of VDR in VDR microinvasion (P ¼ 0.001), lower percent area of the PIN II KO mice (HV2-VDR KO) have normal calcium homeosta- lesions (P ¼ 0.024), and a trend toward higher percent area sis but no VDR within the entire prostate microenviron- for the PIN IV lesions (P ¼ 0.08, Fig. 3D). ment. In these mice, the lesion count was higher for all of Serum 25OHD levels were increased by raising dietary the advanced lesions compared with controls: PIN IV vitamin D3 intake (P < 0.001, Fig. 4A). The levels in the

OF6 Cancer Prev Res; 12(6) June 2019 Cancer Prevention Research

Vitamin D and Carcinogenesis

Figure 3. Histologic grading of the anterior prostates from 26-week-old HV2-VDR KO (n ¼ 23) and control mice (HV2þ, VDR wild-type, n ¼ 27). A, Count of independent lesion foci per section; (B) average lesion size; (C) area of each lesion; (D) percent area of each lesion type. Bars are mean SEM. (, P < 0.05 vs. Control). Other comparisons where the difference approached signiﬁcance are reported with the measured P value.

25 IU/kg vitamin D3 group were slightly higher than we than the rodent requirement caused a uniform 8.5% previously reported due to the contribution of residual increase in adenocarcinoma incidence across the vitamin vitamin D3 from casein in the diet. Dietary Ca did not alter D diet groups (P < 0.05, Fig. 4C). serum 25OHD levels but reduced serum 1,25(OH)2D Although higher-dietary vitamin D suppressed the pros- levels by more than 57% (Fig. 4B). There was a trend tate cancer phenotype, higher vitamin D intake had no toward higher serum 1,25(OH)2D levels in mice fed the signiﬁcant impact on BMD (Fig. 4D). However, increased 1,000 IU/kg vitamin D3 diet (P ¼ 0.074). dietary Ca increased BMD regardless of the dietary vitamin Low-dietary vitamin D intake increased the percentage of D intake level (P < 0.01). mice with adenocarcinoma as their most advanced lesion, VDR ChIP-seq analysis in RWPE1 cells was used to such that mice fed the 25 IU/kg vitamin D3 diet had identify vitamin D target genes in prostate that may the highest incidence of this phenotype (93% vs. 68% contribute to the protection of PECs against prostate incidence in the 1,000 IU/kg vitamin D3 group, P < cancer. VDR ChIP-seq analysis revealed 3,762 peaks in 0.001, Fig. 4C). Increasing dietary Ca to 3 times greater control cells and 3,445 peaks in 1,25(OH)2D-treated

www.aacrjournals.org Cancer Prev Res; 12(6) June 2019 OF7

Fleet et al.

Figure 4.

The effect of dietary Ca and vitamin D on TgAPT121 mice. Diets were fed from weaning until 28 weeks of age (n ¼ 34 per diet). A, Serum 25OHD and (B) serum 1,25 (OH)2D are expressed as the mean SEM (n ¼ 10/diet group). Values with different letter superscripts are significantly different (P < 0.05). C, Incidence of prostate adenocarcinoma in the anterior prostate (n ¼ 34 per diet group). Values with different letter superscripts are significantly different between vitamin D groups (n ¼ 60/group), while signifies a significant effect of Ca (n ¼ 90/diet Ca level, x2 test, P < 0.05). D, BMD of the femur. signifies a significant effect of Ca (P < 0.01). Bone data are expressed as mean SEM (n ¼ 50–54/diet Ca level).

cells with just 175 peaks common to both groups MAPK6, CTSD/SYT8, DDIT4, KLF4, CA9,andETNK2 (Fig. 5A; Supplementary Table S4). DR3-type VDR-bind- (Supplementary Fig. S2). Note that 492 of the VDR sites ing motifs that mediate VDR binding to DNA were found in RWPE1 cells (associated with 461 genes) were found in 14.2% of the peaks, although more were found in the in ChIP-seq data from more than one cell type. Most of 1,25(OH)2 D group (20%) and the overlap group these peaks were from 1,25(OH)2D-treated cells or were (72.6%). VDR ChIP-seq sites were annotated to 3,390 peaks found in both treatment groups (n ¼ 418, 73% protein coding genes, 682 long noncoding RNAs, and with DR3 motifs). There was significant overlap between 472 miRNAs. This includes VDR-binding peaks in known the VDR ChIP-seq peaks in RWPE1 cells and other vitamin D target genes like CYP24A1 (ref. 38; 5 peaks, 4 datasets: (i) 1,573 peaks in 1,093 1,25(OH)2D-regulated with DR3) and IGFBP3 (ref. 39; 2 peaks, 1 with DR3). genes from RWPE1 cells (ref. 34; Fig. 5B), (ii) 319 peaks The bulk of the peaks were intergenic (50.7%) or intronic in 193 1,25(OH)2D-regulated genes from mouse pros- (43.8%), reflecting long-distance regulation by enhancer tate epithelial stem cells (27), (iii) 263 genes containing elements. We confirmed the ChIP-seq results for 7 VDR- a VDR peak within a TAD for vitamin D–regulated genes binding peaks by ChIP-PCR in the genes for: CYP24A1, from human THP-1 cells (40).

OF8 Cancer Prev Res; 12(6) June 2019 Cancer Prevention Research

Vitamin D and Carcinogenesis

Figure 5.

Identiﬁcation of VDR-binding sites in RWPE1 cells. A, A comparison of ethanol vehicle (EtOH)– or 1,25(OH)2D (Vitamin D)-treated cells (10 nmol/L, 3 hours). Traditional DR3-type VDR-binding motifs were identiﬁed below the summits of VDR peaks. B, Peaks were assigned to genes and then compared with transcriptome results from vitamin D–treated RWPE1 cells. Overlap among the groups and the DR3-containing VDR peaks was assessed.

The potential functional impact of vitamin D signaling growth (12, 13, 41), but definitive evidence for a beneficial on the PEC was broad. Enriched canonical pathways effect of increasing vitamin D intake on prostate cancer risk included those with cancer relevant functions (Table 1; has been lacking. Thus, specific recommendations for Supplementary Table S5). Pathways regulating gene tran- optimal vitamin D intake to reduce human prostate cancer scription included: one called "Transcriptional role for VDR risk are not yet possible. Rodent studies that manipulate in regulation of genes involved in osteoporosis" driven by dietary vitamin D intake have used either severe deficien- genes like CYP24A1 and IGFBP3; four related to NF-kB cy (11) or a large vitamin D excess (15, 16) and thus have signaling that included the genes NFKBIA and NFKBIZ; one modest translational potential. Meanwhile, human vita- for the oxidative stress response driven by the genes min D3 interventions for cancer prevention (refs. 8, 42, 43; NFE2L2, TXNRD1, FOS, and JUNB; and several that includ- ClinicalTrials ID #NCT01463813) suffer from several lim- ed transcription factors whose genes contain VDR-binding itations. First, they are typically conducted in older subjects sites, i.e., ESR1, CREB, HIF1A, and RBRJ. Note that 109 with high baseline serum 25OHD levels (>75 nmol/L) and transcription factors were vitamin D–regulated in RWPE1 indolent cancer. Second, some studies use supplement cells and contained a VDR peak. In addition, we found 259 levels that are too low for subjects who already have high interaction networks in the VDR cistrome that were cen- vitamin D status (e.g., 800–2,000 IU/d; refs. 8, 43), where- tered on transcription factors, and 44 of these networks as other studies use pharmacologic, bolus doses (e.g., were centered on transcription factors whose genes had a monthly doses at 10,000 IU; ref. 42) that cause large DR3 motif below their VDR peak (Table 2). swings in serum 25OHD and which may have adverse effects on other health outcomes (e.g., increased falls and Discussion fractures in older women; ref. 44). Finally, these studies are typically short even though targeting prostate carcinogen- Vitamin D signaling through the VDR regulates cancer- esis may require decades of intervention to capture effects relevant cellular events (10) and reduces prostate tumor of vitamin D. This may explain why the VITAL study results Table 1. Categories of enriched pathways for VDR ChIP-seq peaks from RWPE1 were promising (i.e., reduced cancer mortality only when cells the first 2 years of follow-up were excluded, nonsignificant Map category Representative genes reduction in HR for prostate cancer to 0.88) but reported as Apoptosis CALM2, FOXO1, p73, PTK2, ROCK1, SRF negative for the entire 5.3-year median follow-up AREG, HGF, HGFR, ITGB1, MYC Cancer period (8). Cell adhesion/Cytoskeleton C3G, EFNA5, EPHB1, EPHA2, MLCK, PTK2, ROCK2 Cell cycle FOS, EGR1, MAPK8, MAD1, MDM2, NEK7 With the challenges of conducting human intervention Immune response CCL16, CD40L, GNAI1, IL6, IL18RAP, IFNG, PTPRC, studies in mind, we designed mouse studies that test TNFSF5 the lifelong effects of manipulating vitamin D signaling Signaling ADCY1, 3, and 8, CAMKD, CAMKK2, MAP3K4, PIK3CA, PRKCE on early stages of prostate carcinogenesis. Many animal Transcription CBF1, FOS, MYC, CREB1, CREM, ESR1, HIF1A, models for prostate cancer exist (45), but we chose the NFKBIA, NFKBIZ, SMAD3 TgAPT121 model because it models the early steps of

www.aacrjournals.org Cancer Prev Res; 12(6) June 2019 OF9

Fleet et al.

Table 2. Enriched transcription factors with VDR ChIP-seq peaks from RWPE1 cells TF-Related gene TF interaction network Peak signala DR3 in peak VD effect in RWPE1 array Peak in multiple cell types CREB1 YVD 48 h, N EPAS1 YVD 24 h, þ N FOS YVDþ all, þ Y GLI3 YVDþ NS Y HIF1A YVDþ 6, 48 h, þ Y KLF4 YVDþ 6h,þ Y KLF9 YVDþ NS Y NFKBIA RELB VD þ 6, 24 h, þ Y RARA YVDþ NS Y SMAD2 YVDþ NS N SRF YVD 48 h, þ N TP73 YVD 46 h, N DEC1 YCþ NS N ESR1 YCþ NS N ETS1 YC 6, 48 h, Y GLI2 YC 48 h, N MYC YC 6h, N SMAD3 YC 24 h, N AHR YBþ 6h,þ N FOXO1 YBþ 6, 48 h, þ Y RBPJ (CBF1) YBþ 48 h, Y TBP YBþ 6h, Y a C, peak in ethanol-treated control; B, peak in both groups; VD, peak in 1,25(OH)2D-treated group.

prostate carcinogenesis, its prostate cancer phenotype approximately 30 days increased prostate 1,25(OH)2 D develops gradually over 6 months (19), and because our levels. They also showed that prostate 1,25(OH)2 D levels data show that the HGPIN in TgAPT121 mice has a molec- were inversely associated with proliferation, suggesting ular phenotype similar to human HGPIN (46). Our find- protection resulting from local 1,25(OH)2 D production. ings in TgAPT121 mice consistently show that disrupting Our study using higher-dietary Ca levels shows that the vitamin D signaling increases the progression of HGPIN severity of prostate lesions in mice is also increased by lesions to focal adenocarcinoma. These studies extend our suppressing serum 1,25(OH)2 D levels. However, earlier report (18), and our new findings are similar to the although high-dietary Ca also increases prostate cancer, effect of whole-body VDR gene deletion in the more two other mouse models with slow-growing lesions (16) aggressive LPB-Tag mouse (12). In addition, by using and low-serum 1,25(OH)2D levels caused by high Ca diets dietary vitamin D3 levels that allow us to model human did not accelerate prostate cancer in the more rapidly relevant ranges of serum 25OHD [i.e., from adequate for progressing LPB-Tag model (12). This suggests that the bone health (50 nmol/L) to high natural levels seen in negative impact of high-dietary Ca on prostate carcinogen- humans (125–150 nmol/L)], we can make two important esis mediated through serum 1,25(OH)2D levels may be conclusions. First, our data show that extreme deficiency limited to specific stages of cancer development; this that disrupts Ca homeostasis is not necessary to increase hypothesis has not been formally tested. Some data also cancer risk. Also, we found that raising serum 25OHD to suggest that the benefit of higher-serum 25OHD on pros- levels that are 50%–150% higher than those needed to tate cancer is modified by the level of dietary Ca (50), but protect bone can slow prostate cancer progression we did not observe a significant Ca-by-vitamin D interac- (Fig. 4). Still, before we can define the optimal vitamin tion in our study. D intake for prostate cancer prevention in humans, Our VDR ChIP-seq analysis demonstrates that activating more research is needed to determine the life stages vitamin D signaling affects multiple, complementary, where increased vitamin D exposure is effective as well cancer protective pathways in PECs, and this is consistent as to definethedose-responsecurvefortheantiprostate with our previous microarray reports (27, 34). The vitamin cancer effects. D–responsive gene signature in nontransformed PECs Although dietary vitamin D raises serum 25OHD levels, includes an antioxidant pathway signature that involves this metabolite is not a high-affinity ligand for the VDR. As activation of NFE2L2, TXNRD1, FOS, and JUNB. A similar such, the benefit of higher serum 25OHD levels is likely not vitamin D effect has been observed in other systems (e.g., direct but may be due to local conversion to 1,25(OH)2 D. diabetes; ref. 51), and in the context of cancer, activation of Consistent with this hypothesis, cell studies have shown this pathway may synergize with inhibition of genes con- that primary PECs and prostate cancer cells can synthesize trolling proliferation (e.g., TCF7L1 and KLF6 in our study), 1,25(OH)2 D from 25OHD (47, 48). In addition, Wagner promoting apoptosis, and activating DNA protective path- and colleagues (49) found that supplementing prepros- ways (e.g., GADD45A) to limit the accumulation of DNA tectomy patients with 40,000 IU vitamin D3 a day for mutations in PECs. Our VDR ChIP-seq data also suggest

OF10 Cancer Prev Res; 12(6) June 2019 Cancer Prevention Research

Vitamin D and Carcinogenesis

that vitamin D may amplify signals mediated through Disclosure of Potential Conflicts of Interest other transcription factors, e.g., by regulating genes for No potential conflicts of interest were disclosed. NFKBIA, an inhibitor of cancer-promoting NF-kB signaling (52), or FOXO1, whose loss promotes prostate carci- Authors' Contributions nogenesis when TMPRSS2-ERG is overexpressed (53). Conception and design: J.C. Fleet, P.L. Kovalenko, S.K. Clinton Finally, the ChIP-seq data revealed VDR-binding peaks Development of methodology: J.C. Fleet, P.L. Kovalenko, S.K. Clinton near genes that are involved in immune responses (e.g., Acquisition of data (provided animals, acquired and managed IL1R2, Il20RB, IRAK2, and IRAK1BP1). These peaks may be patients, provided facilities, etc.): J.C. Fleet, P.L. Kovalenko, responsible for the downregulation of immune/inflamma- J. Smolinski, C. Spees, JM. Thomas-Ahner, S.K. Clinton tory mRNAs we reported earlier (34) and make PECs less Analysis and interpretation of data (e.g., statistical analysis, responsive to proinflammatory, mitogenic signals from biostatistics, computational analysis): J.C. Fleet, P.L. Kovalenko, immune cells. Y. Li, J. Smolinski, C. Spees, A. Neme, C. Carlberg, S.K. Clinton A final interesting finding from our mouse studies is Writing, review, and/or revision of the manuscript: J.C. Fleet, fi P.L. Kovalenko, J. Smolinski, C. Spees, A. Neme, C. Carlberg, that they suggest that the bene ts of vitamin D signaling S.K. Clinton in the prostate may extend beyond the PEC. Although Administrative, technical, or material support (i.e., reporting or both prostate VDR deletion models had more advanced organizing data, constructing databases): P.L. Kovalenko, M. Cui, cancer, the impact of deletion was more dramatic in S.K. Clinton HV2-VDR KO mice that lack VDR in all prostate- Study supervision: J.C. Fleet, P.L. Kovalenko, S.K. Clinton associated cells than in the mice where VDR deletion was limited to PECs. Others have shown that prostate Acknowledgments We thank Ryan Schoch, Kimberly Carter, and Valerie DeGroff for stromal cells are vitamin D target cells (54), and we technical support, and Sanjay Bhave, PhD, and Hsueh-Li Tan, PhD, for previously showed that stromal cell proliferation is assistance with necropsy and biosample collections. reduced in the prostates of PEC VDR KO mice (18). This work was supported by NIH/NCI R01 CA10113 (to J.C. Fleet), Collectively, these observations suggest that vitamin D American Institute for Cancer Research (AICR) Award #05A131 (to S. signaling regulates multiple cells in the prostate, and it K. Clinton), and a Showalter Trust Award (to J.C. Fleet). Y. Li was may modulate complex interactions among those cells. supported in part through a Cancer Prevention Interdisciplinary This hypothesis requires further testing. Education Fellowship (NIH/NCI R25CA128770) and a Purdue Research Foundation Graduate Research Assistantship. The transgenic In summary, our data provide solid support for the proof mouse studies were made possible by technical support from the of principle that vitamin D signaling modulates progres- Transgenic Mouse Core Facility, a core service of the Purdue University sion through the early stages of prostate carcinogenesis and Center for Cancer Research (P30 CA023168). Additional support was for the idea that early, life-long dietary manipulation of provided by the Ohio State University Comprehensive Cancer Center serum vitamin D metabolites can modify the course of (OSUCCC; NIH/NCI P30 CA016058). early-stage prostate cancer. In addition, as Ames first sug- gested in his "triage theory" of nutrition and chronic The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked disease (55), our data in mice demonstrate that there can advertisement in accordance with 18 U.S.C. Section 1734 solely to be different dietary requirements for vitamin D and Ca indicate this fact. based on the health outcome examined (i.e., higher vitamin D and lower Ca for prostate protection; lower vitamin Received October 17, 2018; revised March 3, 2019; accepted April D and higher Ca for bone). 22, 2019; published first April 26, 2019.

References 1. Feldman D, Krishnan AV, Swami S, Giovannucci E, Feldman BJ. 5. Schenk JM, Till CA, Tangen CM, Goodman PJ, Song X, Torkko KC, The role of vitamin D in reducing cancer risk and progression. et al. Serum 25-hydroxyvitamin D concentrations and risk of Nat Rev Cancer 2014;14:342–57. prostate cancer: results from the Prostate Cancer Prevention Trial. 2. Corder EH, Guess HA, Hulka BS, Friedman GD, Sadler M, Vollmer Cancer Epidemiol Biomarkers Prev 2014;23:1484–93. RT, et al. Vitamin D and prostate cancer: a prediagnostic study 6. Brandstedt J, Almquist M, Ulmert D, Manjer J, Malm J. Vitamin D, with stored sera. Cancer Epidemiol Biomarkers Prev 1993;2: PTH, and calcium and tumor aggressiveness in prostate cancer: a 467–72. prospective nested case-control study. Cancer Causes Control 3. Li H, Stampfer MJ, Hollis JB, Mucci LA, Gaziano JM, Hunter D, 2016;27:69–80. et al. A prospective study of plasma vitamin D metabolites, 7. Travis RC, Perez-Cornago A, Appleby PN, Albanes D, Joshu CE, vitamin D receptor polymorphisms, and prostate cancer. Lutsey PL, et al. A collaborative analysis of individual participant PLoS Med 2007;4:e103. data from 19 prospective studies assesses circulating vitamin D 4. Kristal AR, Till C, Song X, Tangen CM, Goodman PJ, Neuhauser and prostate cancer risk. Cancer Res 2019;79:274–85. ML, et al. Plasma vitamin D and prostate cancer risk: results from 8. Manson JE, Cook NR, Lee IM, Christen W, Bassuk SS, Mora S, et al. the Selenium and Vitamin E Cancer Prevention Trial. Vitamin D supplements and prevention of cancer and cardiovas- Cancer Epidemiol Biomarkers Prev 2014;23:1494–504. cular disease. N Engl J Med 2019;380:33–44.

www.aacrjournals.org Cancer Prev Res; 12(6) June 2019 OF11

Fleet et al.

9. Giovannucci E, Rimm EB, Wolk A, Ascherio A, Stampfer MJ, 25. Tusher VG, Tibshirani R, Chu G. Signiﬁcance analysis of micro- Colditz GA, et al. Calcium and fructose intake in relation to risk arrays applied to the ionizing radiation response. Proc Natl Acad of prostate cancer. Cancer Res 1998;58:442–47. Sci U S A 2001;98:5116–21. 10. Fleet JC, DeSmet M, Johnson R, Li Y. Vitamin D and cancer: a 26. Tomlins SA, Mehra R, Rhodes DR, Cao X, Wang L, Dhanasekaran review of molecular mechanisms. Biochem J 2012;441: SM, et al. Integrative molecular concept modeling of prostate 61–76. cancer progression. Nat Genet 2007;39:41–51. 11. Ray R, Banks M, Abuzahra H, Eddy VJ, Persons KS, Lucia MS, et al. 27. Maund SL, Barclay WW, Hover LD, Axanova LS, Sui G, Hipp JD, Effect of dietary vitamin D and calcium on the growth of andro- et al. Interleukin-1a mediates the antiproliferative effects of 1,25-

gen-insensitive human prostate tumor in a murine model. dihydroxyvitamin D3 in prostate progenitor/stem cells. Anticancer Res 2012;32:727–31. Cancer Res 2011;71:5276–86. 12. Mordan-McCombs S, Brown T, Wang WL, Gaupel AC, Welsh J, 28. Simon R, Lam A, Li MC, Ngan M, Menenzes S, Zhao Y. Analysis of Tenniswood M. Tumor progression in the LPB-Tag transgenic gene expression data using BRB-ArrayTools. Cancer Inform 2007; model of prostate cancer is altered by vitamin D receptor and 3:11–7. serum testosterone status. J Steroid Biochem Mol Biol 2010;121: 29. Cui M, Zhao Y, Hance KW, Shao A, Wood RJ, Fleet JC. Effects of 368–71. MAPK signaling on 1,25-dihydroxyvitamin D-mediated CYP24 13. Lokeshwar BL, Schwartz GG, Selzer MG, Burnstein KL, gene expression in the enterocyte-like cell line, Caco-2. J Cell Zhuang SH, Block NL, et al. Inhibition of prostate cancer Physiol 2009;219:132–42. metastasis in vivo: a comparison of 1,23- dihydroxyvitamin 30. Robertson G, Hirst M, Bainbridge M, Bilenky M, Zhao Y, Zeng T, D (calcitriol) and EB1089. Cancer Epidemiol Biomarkers et al. Genome-wide profiles of STAT1 DNA association using Prev 1999;8:241–48. chromatin immunoprecipitation and massively parallel sequenc- 14. Xue L, Yang K, Newmark H, Lipkin M. Induced hyperproliferation ing. Nat Methods 2007;4:651–57. in epithelial cells of mouse prostate by a Western-style diet. 31. Neme A, Seuter S, Carlberg C. Selective regulation of biological Carcinogenesis 1997;18:995–99. processes by vitamin D based on the spatio-temporal cistrome of 15. Swami S, Krishnan AV, Wang JY, Jensen K, Horst R, Albertelli MA, its receptor. Biochim Biophys Acta 2017;1860:952–61. et al. Dietary vitamin D3 and 1,25-dihydroxyvitamin D3 (Calci- 32. Tuoresmaki P, Vaisanen S, Neme A, Heikkinen S, Carlberg C. triol) exhibit equivalent anticancer activity in mouse xenograft Patterns of genome-wide VDR locations. PLoS One 2014;9: models of breast and prostate cancer. Endocrinology 2012;153: e96105. 2576–87. 33. Heinz S, Benner C, Spann N, Bertolino E, Lin YC, Laslo P, et al. 16. Bernichtein S, Pigat N, Barry Delongchamps N, Boutillon F, Simple combinations of lineage-determining transcription fac- Verkarre V, Camparo P, et al. Vitamin D3 prevents calcium- tors prime cis-regulatory elements required for macrophage and B induced progression of early-stage prostate tumors by counter- cell identities. Mol Cell 2010;38:576–89. acting TRPC6 and calcium sensing receptor upregulation. 34. Kovalenko PL, Zhang Z, Cui M, Clinton SK, Fleet JC. 1,25 Cancer Res 2017;77:355–65. dihydroxyvitamin D-mediated orchestration of anticancer, tran- 17. Fleet JC, Gliniak C, Zhang Z, Xue Y, Smith KB, McCreedy R, et al. script-level effects in the immortalized, non-transformed prostate Serum metabolite profiles and target tissue gene expression epithelial cell line, RWPE1. BMCGenomics 2010;11:26. define the effect of cholecalciferol intake on calcium metabolism 35. Neme A, Seuter S, Carlberg C. Vitamin D-dependent chromatin in rats and mice. J Nutr 2008;138:1114–20. association of CTCF in human monocytes. Biochim Biophys Acta 18. Kovalenko PL, Zhang Z, Yu JG, Li Y, Clinton SK, Fleet JC. Dietary 2016;1859:1380–88. vitamin D and vitamin D receptor level modulate epithelial cell 36. Brawer MK. Prostatic intraepithelial neoplasia: an overview. proliferation and apoptosis in the prostate. Cancer Prev Res Rev Urol 2005;7Suppl 3:S11–8. (Phila) 2011;4:1617–25. 37. Ashida S, Nakagawa H, Katagiri T, Furihata M, Iiizumi M, Ana- 19. Hill R, Song Y, Cardiff RD, Van Dyke T. Heterogeneous tumor zawa Y, et al. Molecular features of the transition from prostatic evolution initiated by loss of pRb function in a preclinical intraepithelial neoplasia (PIN) to prostate cancer: genome-wide prostate cancer model. Cancer Res 2005;65:10243–54. gene-expression profiles of prostate cancers and PINs. Cancer Res 20. Xue YB, Fleet JC. Intestinal vitamin D receptor is required for 2004;64:5963–72. normal calcium and bone metabolism in mice. Gastroenterology 38. Meyer MB, Goetsch PD, Pike JW. A downstream intergenic cluster 2009;136:1317–27. of regulatory enhancers contributes to the induction of CYP24A1 21. Song Y, Kato S, Fleet JC. Vitamin D receptor (VDR) knockout mice expression by 1 alpha,25-dihydroxyvitamin D-3. J Biol Chem reveal VDR-independent regulation of intestinal calcium absorp- 2010;285:15599–610. tion and ECaC2 and calbindin D9k mRNA. J Nutr 2003;133: 39. Peng LH, Malloy PJ, Feldman D. Identification of a functional 374–80. vitamin D response element in the human insulin-like growth 22. AIN. Report of the American Institute of Nutrition ad hoc com- factor binding protein-3 promoter. Mol Endocrinol 2004;18: mittee on standards for nutritional studies. J Nutr 1977;107: 1109–19. 1340–48. 40. Neme A, Nurminen V, Seuter S, Carlberg C. The vitamin D-depen- 23. Zhang Z, Kovalenko P, Cui M, DeSmet M, Clinton SK, Fleet JC. dent transcriptome of human monocytes. J Steroid Biochem Mol Constitutive activation of the mitogen-activated protein kinase Biol 2016;164:180–87. pathway impairs vitamin D signaling in human prostate epithe- 41. Polek TC, Murthy S, Blutt SE, Boehm MF, Zou A, Weigel NL, et al. lial cells. J Cell Physiol 2010;224:433–42. Novel nonsecosteroidal vitamin D receptor modulator inhibits 24. Shappell SB, Thomas GV, Roberts RL, Herbert R, Ittmann MM, the growth of LNCaP xenograft tumors in athymic mice without Rubin MA, et al. Prostate pathology of genetically engineered increased serum calcium. Prostate 2001;49:224–33. mice: definitions and classification. The consensus report from 42. Scragg R, Khaw KT, Toop L, Sluyter J, Lawes CMM, Waayer D, et al. the Bar Harbor meeting of the Mouse Models of Human Cancer Monthly high-dose Vitamin D supplementation and cancer risk: a Consortium Prostate Pathology Committee. Cancer Res 2004;64: post hoc analysis of the Vitamin D assessment randomized 2270–305. clinical trial. JAMA Oncol 2018;4:e182178.

OF12 Cancer Prev Res; 12(6) June 2019 Cancer Prevention Research

Vitamin D and Carcinogenesis

43. Lappe J, Watson P, Travers-Gustafson D, Recker R, Garland C, prostatic vitamin D metabolite levels and ki67 labeling in pros- Gorham E, et al. Effect of vitamin D and calcium supplementa- tate cancer patients. J Clin Endocrinol Metab 2013;98:1498–507. tion on cancer incidence in older women: a randomized clinical 50. Steck SE, Arab L, Zhang H, Bensen JT, Fontham ET, Johnson CS, trial. JAMA 2017;317:1234–43. et al. Association between Plasma 25-Hydroxyvitamin D, Ances- 44. Sanders KM, Stuart AL, Williamson EJ, Simpson JA, Kotowicz MA, try and Aggressive Prostate Cancer among African Americans and Young D, et al. Annual high-dose oral vitamin D and falls and European Americans in PCaP. PLoS One 2015;10:e0125151. fractures in older women: a randomized controlled trial. JAMA 51. Nakai K, Fujii H, Kono K, Goto S, Kitazawa R, Kitazawa S, et al. 2010;303:1815–22. Vitamin D activates the Nrf2-Keap1 antioxidant pathway and 45. Grabowska MM, DeGraff DJ, Yu X, Jin RJ, Chen Z, Borowsky AD, ameliorates nephropathy in diabetic rats. Am J Hypertens 2014; et al. Mouse models of prostate cancer: picking the best model for 27:586–95. the question. Cancer Metastasis Rev 2014;33:377–97. 52. Jin R, Yi Y, Yull FE, Blackwell TS, Clark PE, Koyama T, et al. NF- 46. Bostwick DG, Liu L, Brawer MK, Qian J. High-grade prostatic kappaB gene signature predicts prostate cancer progression. intraepithelial neoplasia. Rev Urol 2004;6:171–9. Cancer Res 2014;74:2763–72. 47. Schwartz GG, Whitlatch LW, Chen TC, Lokeshwar BL, Holick MF. 53. Yang Y, Blee AM, Wang D, An J, Pan Y, Yan Y, et al. Loss of FOXO1

Human prostate cells synthesize 1,25-dihydroxyvitamin D3 from cooperates with TMPRSS2-ERG overexpression to promote pros- 25-hydroxyvitamin D3. Cancer Epidemiol Biomarkers Prev 1998; tate tumorigenesis and cell invasion. Cancer Res 2017;77: 7:391–95. 6524–37. 48. Whitlatch LW, Young MV, Schwartz GG, Flanagan JN, Burnstein 54. Lou YR, Miettinen S, Kagechika H, Gronemeyer H, Tuohimaa P. KL, Lokeshwar BL, et al. 25-Hydroxyvitamin D-1alpha-hydroxy- Retinoic acid via RARalpha inhibits the expression of 24-hydrox- lase activity is diminished in human prostate cancer cells and is ylase in human prostate stromal cells. Biochem Biophys Res enhanced by gene transfer. J Steroid Biochem Mol Biol 2002;81: Commun 2005;338:1973–81. 135–40. 55. Ames BN. Low micronutrient intake may accelerate the degen- 49. Wagner D, Trudel D, Van der Kwast T, Nonn L, Giangreco AA, Li erative diseases of aging through allocation of scarce micronu- D, et al. Randomized clinical trial of vitamin D3 doses on trients by triage. Proc Natl Acad Sci U S A 2006;103:17589–94.

www.aacrjournals.org Cancer Prev Res; 12(6) June 2019 OF13

Fleet et al.

OF14 Cancer Prev Res; 12(6) June 2019 Cancer Prevention Research

Vitamin D Signaling Suppresses Early Prostate Carcinogenesis in TgAPT 121 Mice

James C. Fleet, Pavlo L. Kovalenko, Yan Li, et al.

Cancer Prev Res Published OnlineFirst April 26, 2019.

Updated version Access the most recent version of this article at: doi:10.1158/1940-6207.CAPR-18-0401

Supplementary Access the most recent supplemental material at: Material http://cancerpreventionresearch.aacrjournals.org/content/suppl/2019/04/26/1940-6207.CAPR -18-0401.DC1

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerpreventionresearch.aacrjournals.org/content/early/2019/05/14/1940-6207.CAPR- 18-0401. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.