CHD1 deletion stabilizes HIF1α to promote angiogenesis and glycolysis in prostate cancer

Abstract

Chromodomain-helicase-DNA-binding protein 1 (CHD1) deletion is among the most common mutations in prostate cancer (PCa), but its role remains unclear. In this study, RNA sequencing was conducted in PCa cells after clustered regularly interspaced palindromic repeat (CRISPR)/CRISPR-associated protein 9 (Cas9)-based CHD1 knockout. Gene set enrichment analysis (GSEA) indicated upregulation of hypoxia-related pathways. A subsequent study confirmed that CHD1 deletion significantly upregulated hypoxia-inducible factor 1α (HIF1α) expression. Mechanistic investigation revealed that CHD1 deletion upregulated HIF1α by transcriptionally downregulating prolyl hydroxylase domain protein 2 (PHD2), a prolyl hydroxylase catalyzing the hydroxylation of HIF1α and thus promoting its degradation by the E3 ligase von Hippel–Lindau tumor suppressor (VHL). Functional analysis showed that CHD1 deletion promoted angiogenesis and glycolysis, possibly through HIF1α target genes. Taken together, these findings indicate that CHD1 deletion enhances HIF1α expression through PHD2 downregulation and therefore promotes angiogenesis and metabolic reprogramming in PCa.

INTRODUCTION

Prostate cancer (PCa) is the most common malignancy in men, with more than 1.2 million new cases and more than 300 000 deaths occurring each year.¹ Since at early diagnosis, PCa is strongly dependent on the androgen receptor (AR), inhibition of AR signaling by androgen deprivation therapy (ADT) is the main therapeutic approach for metastatic lesions. However, most tumors eventually develop into castration-resistant PCa (CRPC).² Thus, PCa is a refractory disease, and it is necessary to explore the mechanism of its initiation and progression.

Chromodomain-helicase-DNA-binding protein 1 (CHD1) is a chromatin remodeling factor belonging to the adenosine triphosphate (ATP)-dependent CHD family, and regulates gene transcription by altering nucleosome structure.³ The CHD1 locus is on chromosome 5q21, and the encoded protein changes the accessibility of genes by unwinding supercoiled DNA, thereby performing its functions, which include changing nucleosome positioning and regulating DNA transcription, replication, and damage repair.⁴ A somatic mutation, CHD1 deletion, occurs in 8%–15% of PCa cases.^5,6 However, its role in PCa remains controversial.⁷-¹⁵ For example, coordinated loss of mitogen-activated protein kinase kinase kinase 7 (MAP3K7) and CHD1 has been reported to promote aggressive behavior,⁷ but another study indicated that CHD1 is a putative synthetic essential gene in phosphatase and tensin homolog (PTEN)-deficient PCa.⁸

In this study, we explored the role of CHD1 deletion in PCa cells using RNA sequencing (RNA-seq) and found that CHD1 knockout significantly stabilized hypoxia-inducible factor 1α (HIF1α) by inhibiting the transcription of prolyl hydroxylase domain protein 2 (PHD2) and consequently promoted angiogenesis and glycolysis in PCa cells.

MATERIALS AND METHODS

Cell culture and reagents

The human PCa cell line ARCaP_E was kindly provided by Prof. Leland W.K. Chung (Cedars inai Medical Center, Los Angeles, CA, USA). The human PCa cell lines DU145 and PC3 and human umbilical vein endothelial cells (HUVECs) were obtained from the American Type Culture Collection (ATCC; Manassas, VA, USA). ARCaP_E and PC3 cells were routinely propagated in Roswell Park Memorial Institute-1640 medium (Merck KGaA, Darmstadt, Germany), while DU145 cells and HUVECs were routinely propagated in Dulbecco’s modified Eagle’s medium (DMEM; Gibco, Thermo Fisher Scientific, Inc., Waltham, MA, USA) supplemented with 5% fetal bovine serum (FBS; Gibco, Thermo Fisher Scientific, Inc.) at 37°C in a humidified incubator with 5% CO₂ in air. Antibodies against β-actin (#20536-1-AP), PHD1 (#12984-1-AP), PHD2 (#20368-1-AP), and PHD3 (#18325-1-AP) were purchased from Proteintech (Wuhan, China). Antibodies against CHD1 (#4351), HIF1α (#14179), von Hippel–Lindau tumor suppressor (VHL; #68547), and hydroxy-HIF1α (#3434) were purchased from Cell Signaling Technology (Danvers, MA, USA). An antibody against eukaryotic translation initiation factor 2 subunit alpha (eIF2α; Sc133132) was purchased from Santa Cruz Biotechnology (Dallas, TX, USA). An antibody against phosphorylated eIF2α (Ser52; ab227593) was purchased from Abcam (Cambridge, MA, USA). A Lactate Colorimetric/Fluorometric Assay Kit was purchased from BioVision (Milpitas, CA, USA) and was used according to the manufacturer’s instructions.

Clustered regularly interspaced palindromic repeat (CRISPR)/CRISPR-associated protein 9 (Cas9)-based CHD1 knockout

CRISPR/Cas9-based CHD1 knockout was conducted with a Cas9 lentiviral vector (#71489, Addgene, Watertown, MA, USA) and a guide RNA (gRNA) vector (#84752, Addgene). The gRNA vector was engineered to contain a gRNA sequence targeting CHD1 (TTCCGATGACTCATCAAGTG). PCa cells transduced with the Cas9 lentiviral vector and gRNA vector were screened using blasticidin (20 μg ml⁻¹; A1113903, Thermo Fisher Scientific, Inc.) and puromycin (2 μg ml⁻¹; A1113803, Gibco), respectively.

Real-time polymerase chain reaction (RT-PCR) analysis

After quantification, 1000 ng of total RNA was reverse transcribed using 4 μl of reverse transcription mix (RR037A, Takara, Beijing, China) for 15 min at 37°C. Quantitative PCR was performed using SYBR Green (RR091A, Takara) with a Bio-Rad CFX96 real-time fluorescence quantitative PCR instrument (Bio-Rad, Hercules, CA, USA). The amounts of cDNA and primers and the thermal cycling program were determined in accordance with the instructions of the PCR mix (RR066A, Takara). mRNA expression levels were normalized to that of human actin beta (ACTB). The primer sequences are listed in Table 1.

Table 1: Primers of polymerase chain reaction

Western blot analysis

Cells washed with PBS were lysed using radioimmunoprecipitation assay (RIPA) buffer (P0013B, Beyotime, Shanghai, China), and the cell lysate was then centrifuged (3K30, Sigma, Darmstadt, Germany) at 20 000g for 15 min. The supernatant was taken for protein quantification by the bicinchoninic acid (BCA) method, and sodium dodecyl sulfate (SDS) loading buffer (P0015L, Beyotime) was then added for incubation in a boiling water bath for 10 min. Proteins were separated on SDS–polyacrylamide gel electrophoresis (PAGE) gels at 90–120 V and transferred to nitrocellulose filter membranes (Thermo Fisher Scientific, Inc.) at 110 V. Nonspecific binding was blocked with skim milk for 1 h. The membranes were incubated with both primary and secondary antibodies after washes using tris-buffered saline with Tween® 20 (TBST). Fluorescence was visualized using a Bio-Rad imaging system (Bio-Rad).

Luciferase reporter assay

Cells in the logarithmic growth phase were used to prepare a cell suspension, counted, and seeded in 24-well culture plates. When the cells were approximately 60% confluent, a fluorescent reporter plasmid (containing a hypoxia response element [HRE] sequence or an oxygen-dependent degradation [ODD] sequence) and an internal reference fluorescent plasmid were cotransfected into the cells using polyethylenimine (PEI; Thermo Fisher Scientific, Inc.). After incubation for 24 h, the cells were lysed, luciferase substrates were added, and the fluorescence intensity was measured with a fluorometer (BD FACSCalibur, Becton Dickinson, San Jose, CA, USA). The relative fluorescence intensity was determined by comparing the reporter fluorescence intensity with the internal reference fluorescence intensity.

RNA-seq and data processing

Target RNA was extracted by adding 1 ml of TRIzol (15596026, Invitrogen, Waltham, MA, USA). A next-generation sequencing library was constructed according to the manufacturer’s protocol (NEBNext^® Ultra™ RNA Library Prep Kit for Illumina^®, New England Biolabs, Ipswich, MA, USA). The library was processed and analyzed by GENEWIZ, Inc. (Hangzhou, China). Fragments per kilobase per million (FKPM) reads values were used for statistical analysis. P < 0.05 by Student’s t-test and a mean fold change of >1.5 were considered to indicate statistically significant differential expression. The RNA-seq data were deduplicated, normalized, and subjected to differential expression analysis using the edgeR package according to the instructions. Pathway enrichment analysis of the differentially expressed genes (DEGs) was performed using gene set enrichment analysis (GSEA; https://www.gsea-msigdb.org/gsea).

HUVEC tube formation assay

Tumor cells were washed with PBS and cultured in serum-free DMEM for 24 h. The cell culture medium was then collected, and the supernatant obtained after centrifugation was used as the conditioned medium. Matrigel (#356230, Corning, Inc., Corning, NY, USA) was added to a 96-well plate (100 μl per well) and incubated at 37°C for 1 h. Then, 80 000 HUVECs were added to each well and incubated with a conditioned medium at 37°C for 6 h. The number of tubules per randomly selected field was determined.

HUVEC migration assay

Tumor cells were washed with PBS, resuspended in serum-free medium, and seeded in a 24-well plate (20 000 cells per well). After the tumor cells adhered, a Transwell chamber (#3464, Corning, Inc.) was placed in a 24-well plate. HUVECs were harvested by digestion, centrifuged, and resuspended in a serum-free medium. After counting, cells (4000 cells per well) were added to the upper compartment of the Transwell chamber. After 24 h, the cells in the chambers were fixed with 4% paraformaldehyde and stained with 0.1% crystal violet. The positively stained cells were counted under a microscope (Nikon ECLIPSE Ts2R, Nikon, Tokyo, Japan). The number of migrated cells in the vector group was set to “1”.

Immunoprecipitation

When the cells in a 10-cm dish were 60% confluent, a specific amount of the target plasmid was transfected, and the medium was changed after 8–12 h. After 24 h, 20 μg ml⁻¹ MG132 (T2154, TargetMol, Wellesley Hills, MA, USA) was added for 6 h. After cell lysate was centrifuged and quantified, the input lysate was directly supplemented with SDS loading buffer and incubated in a boiling water bath, and the immunoprecipitant (IP) lysate was incubated at 4°C with magnetic beads (88802, Thermo Fisher Scientific, Inc.) coated with specific antibodies. After 4–8 h, the magnetic beads were washed three times, and SDS loading buffer was added to the samples, which were then boiled in a water bath.

Statistical analyses

For each assay, the experiments were performed in triplicate, and the results are presented as means with standard errors. Statistical analysis of differences between experimental groups was performed using Student’s t-test. P < 0.05 was considered to be statistically significant. All statistical analyses were performed using GraphPad Prism 8 (GraphPad Software, San Diego, CA, USA).

RESULTS

CHD1 deletion activates the HIF1α pathway

To explore the role of CHD1 in PCa, CRISPR/Cas9-based CHD1 knockout was performed in ARCaP_E cells (Figure 1a), and RNA-seq was then conducted. GSEA of the DEGs revealed that the hypoxia-related pathway was activated after CHD1 knockout (Figure 1b and 1c). Since HIF1α is the most well-known transcription factor related to hypoxia, we measured the expression of HIF1α in ARCaP_E cells with CHD1 knockout by western blot analysis. CHD1 knockout in ARCaP_E cells significantly upregulated HIF1α expression (Figure 1d). Similar results were observed in DU145 and PC3 cells (Figure 1d). HIF1α functions mainly as a transcription factor that forms heterodimers with HIF1β, thereby entering the nucleus, binding to gene promoter regions with HREs, promoting the assembly of the transcription complex, and initiating the transcription of target genes.^16–18 To detect the transcriptional activity of HIF1α, we constructed a fluorescent reporter plasmid containing the HRE motif and applied a luciferase reporter assay to detect the transcriptional activity of HIF1α after CHD1 knockout. CHD1 knockout significantly enhanced the transcriptional activity of HIF1α (P < 0.05; Figure 1e).

Figure 1: CHD1 deletion activates the HIF1α pathway. (a) Expression of CHD1 in ARCaP_E cells after CRISPR/Cas9-based CHD1 knockout. (b) Pathways upregulated in ARCaP_E cells after CHD1 knockout. (c) The hypoxia pathway was enriched in ARCaP_E cells after CHD1 knockout. (d) Expression of HIF1α in ARCaP_E, DU145, and PC3 cells after CHD1 knockout. (e) Luciferase reporter assay of HIF1α transcriptional activity in ARCaP_E, DU145, and PC3 cells after CHD1 knockout. *P < 0.05. CHD1: chromodomain-helicase-DNA-binding protein 1; CRISPR: clustered regularly interspaced palindromic repeat; Cas9: CRISPR-associated protein 9; HIF1α: hypoxia-inducible factor 1α; KRAS: KRAS proto-oncogene, GTPase; IL6: interleukin 6; STAT3: signal transducer and activator of transcription 3; NF-κB: nuclear factor-kappa B; UV: ultra violet; DN: DNA damage; UP: up-regulated; TNFA: tumor necrosis factor α; NOM: normalized; KO: knockout.

CHD1 deletion inhibits HIF1α degradation

To explore the mechanism of CHD1 deletion-mediated HIF1α upregulation, real-time PCR was performed to determine whether HIF1α is regulated at the transcriptional level in ARCaP_E, DU145, and PC3 cells after CHD1 knockout. However, the results indicated that CHD1 did not affect the mRNA expression of HIF1α (Figure 2a; all P > 0.05). An immunoprecipitation-based ubiquitination assay was then conducted to determine whether the HIF1α level is regulated by protein degradation. CHD1 knockout significantly enhanced the ubiquitination of HIF1α (Figure 2b). Thus, these findings indicated that CHD1 deletion in PCa cells might inhibit HIF1α degradation.

Figure 2: CHD1 deletion inhibits HIF1α degradation. (a) HIF1α mRNA expression in ARCaP_E, DU145, and PC3 cells after CHD1 knockout (all P > 0.05). (b) Decreased ubiquitination of HIF1α in ARCaP_E and DU145 cells after CHD1 knockout. HIF1α: hypoxia-inducible factor 1α; CHD1: chromodomain-helicase-DNA-binding protein 1; KO: knockout; IB: immunoblotting; IP: immunoprecipitation; His-Ub: his-tagged ubiquitin.

CHD1 deletion stabilizes HIF1α via PHD2

Degradation mediated by the E3 ligase VHL is the most well-known mechanism of HIF1α degradation. HIF1α is hydroxylated by PHD at Pro402 and Pro564 in the oxygen-dependent degradation (ODD) domain, and hydroxylated HIF1α is then recognized by VHL and consequently degraded by the ubiquitin–proteasome system.^19,20 The PHD family comprises PHD1, PHD2, and PHD3.^21,22 Thus, we measured the expression levels of VHL, PHD1, PHD2, and PHD3 in PCa cells after CHD1 knockout. CHD1 knockout downregulated the expression of PHD2, but not PHD1, PHD3, or VHL in ARCaP_E and DU145 cells (Figure 3a). In addition, it has been reported that phosphorylation of eIF2α is a master mediator of proteasome inhibitor-induced HIF1α expression.²³ However, in our study, CHD1 knockout did not affect the level of eIF2α expression or phosphorylation (Figure 3a). Real-time PCR analysis also indicated downregulation of PHD2 mRNA expression after CHD1 knockout (P < 0.05; Figure 3b). Since HIF1α degradation by VHL is dependent on PHD-mediated hydroxylation at Pro402 and Pro564, HIF1α hydroxylation was detected by an immunoprecipitation-based assay, and the results indicated inhibition of HIF1α hydroxylation after CHD1 knockout (Figure 3c). Then, a luciferase reporter vector containing the ODD domain of HIF1α was constructed, and CHD1 knockout was found to inhibit the degradation of the ODD domain, which contains the sites of PHD-mediated hydroxylation (P < 0.05; Figure 3d). Finally, we generated a CRISPR-resistant CHD1 overexpression system (Figure 3e). The results indicated that the reexpression of CHD1 in ARCaP_E and DU145 cells with CHD1 knockout upregulated PHD2 expression but downregulated HIF1α expression (Figure 3f). Thus, we concluded that CHD1 deletion stabilized HIF1α via PHD2.

Figure 3: CHD1 deletion stabilizes HIF1α via PHD2. (a) Expression of key proteins involved in HIF1α degradation in ARCaP_E and DU145 cells after CHD1 knockout. (b) PHD2 mRNA levels in ARCaP_E and DU145 cells after CHD1 knockout. *P < 0.05. (c) Hydroxylation of HIF1α in ARCaP_E and DU145 cells after CHD1 knockout. (d) Luciferase reporter assay of ODD domain degradation in ARCaP_E and DU145 cells after CHD1 knockout. *P < 0.05. (e) Diagram of the CRISPR-resistant CHD1 overexpression system. The CRISPR-targeted CHD1 sequence was mutated to escape recognition by the gRNA, but the amino acid sequence was not changed. The mutated CHD1 cDNA was cloned into a lentiviral expression vector. (f) Expression of PHD2 and HIF1α after overexpression of CRISPR-resistant CHD1 in ARCaP_E and DU145 cells with CHD1 knockout. HIF1α: hypoxia-inducible factor 1α; CHD1: chromodomain-helicase-DNA-binding protein 1; PHD2: prolyl hydroxylase domain protein 2; ODD: oxygen-dependent degradation; CRISPR: clustered regularly interspaced palindromic repeat; Cas9: CRISPR-associated protein 9; gRNA: guide RNA; cDNA: complementary DNA; KO: knockout; p-eIF2α: phosphorylated eukaryotic translation initiation factor 2 subunit alpha; VHL: von Hippel–Lindau tumor suppressor; IB: immunoblotting; IP: immunoprecipitation.

CHD1 deletion promotes angiogenesis and glycolysis in PCa

HIF1α functions as a mediator of several pathways in tumors, such as angiogenesis and metabolic reprogramming.²⁴ Among the DEGs (P <0.05, fold change >1.5) in the hypoxia pathway, several angiogenesis- or glycolysis-related genes were upregulated after CHD1 knockout (Figure 4a). Angiogenesis mainly constitutes the formation of new blood vessels via increases in the sprouting and migration of vascular endothelial cells. We used immortalized HUVECs as a model to detect the impact of tumor cells on vascular endothelial cells. The results of the tube formation assay showed that HUVECs cocultured with CHD1-deficient PCa cells had stronger tubulogenic activity than those cocultured with wild-type PCa cells (P < 0.05; Figure 4b). Transwell assays also showed that CHD1 knockout significantly enhanced HUVEC recruitment (P < 0.05; Figure 4c and 4d). In addition, quantitative analysis of lactate, a downstream metabolite of glycolysis, indicated that CHD1 knockout increased lactate production (P < 0.05; Figure 4e).

Figure 4: CHD1 deletion promotes angiogenesis and glycolysis in PCa. (a) DEGs related to angiogenesis (green) or glycolysis (red) in ARCaP_E cells after CHD1 knockout. *P < 0.05 and fold change >1.5. (b) Tube formation assay of HUVECs cocultured with ARCaP_E and DU145 cells after CHD1 knockout. *P < 0.05. (c) Representative images of HUVEC migration induced by ARCaP_E and DU145 cells after CHD1 knockout. (d) Statistic data of HUVEC migration induced by ARCaP_E and DU145 cells after CHD1 knockout. *P < 0.05. (e) Lactate production in ARCaP_E and DU145 cells after CHD1 knockout. *P < 0.05. DEGs: differentially expressed genes; CHD1: chromodomain-helicase-DNA-binding protein 1. HUVECs: human umbilical vein endothelial cells; KO: knockout; TGFB1: transforming growth factor-beta 1; ALDOA: aldolase, fructose-bisphosphate A; ALDOC: aldolase, fructose-bisphosphate C; HK1: hexokinase 1; HK2: hexokinase 2; ENO2: enolase 2; NDRG1: N-myc downstream-regulated 1; ANGPTL4: angiopoietin-like 4.

DISCUSSION

Hypoxia is present in 90% of solid tumors and is closely related to tumor proliferation, differentiation, angiogenesis, energy metabolism, drug resistance of cancer cells, and poor patient prognosis.^16,25 Hypoxia stabilizes the α subunit of HIFs (HIF1α and HIF2α).¹⁷ The α subunit and the constitutively expressed β subunit together form a heterodimeric protein, which enters the nucleus to function as a transcription factor.¹⁸ HIF1α increases the expression of key proteins in cellular biological processes such as glycolysis and angiogenesis by binding to HREs (5-RCGTG-3, where R=A or G) in the regulatory regions of the corresponding target genes.²⁴ The expression of HIF1α is a clinically significant factor supporting PCa progression,²⁶ but the reason that HIF1α is highly expressed in PCa has not been conclusively determined.^27,28

In this study, we found that CHD1 knockout transcriptionally inhibited PHD2 expression and HIF1α hydroxylation and consequently inhibited HIF1α degradation by the ubiquitin–proteasome system. This novel mechanism of HIF1α regulation could partially explain why HIF1α is overexpressed in PCa. As a member of the chromatin remodeling factor family, CHD1 changes gene accessibility by binding to different forms of histones; for example, by binding to the transcriptional activation marker, trimethylated lysine 4 on histone H3 (H3K4me3), CHD1 recruits other proteins to form transcription initiation complexes to promote the transcription of the corresponding genes.^29–31 Our findings indicated that CHD1 does not regulate HIF1α expression at the transcriptional level but indirectly stabilizes HIF1α via PHD2, a hydroxylase of HIF1α. However, whether CHD1 can directly regulate PHD2 transcription remains unclear, and further studies are needed.

HIF1α is a hypoxia-related transcription factor and plays multiple roles during the initiation and progression of solid tumors. Angiogenesis and metabolic reprogramming are its classical effects. Several HIF1α downstream genes, such as vascular endothelial growth factor (VEGF), N-myc downstream-regulated 1 (NDRG1), and angiopoietin-like 4 (ANGPTL4), are key proangiogenic factors in human cancers. As a hallmark of cancer, metabolic reprogramming, especially glycolytic reprogramming, plays vital roles in carcinogenesis and cancer progression. More importantly, most glycolysis-related genes are regulated by HIF1α. In this study, we observed that CHD1 knockout upregulated several glycolysis-related genes, such as aldolase, fructose-bisphosphate A (ALDOA), aldolase, fructose-bisphosphate C (ALDOC), hexokinase 1 (HK1), hexokinase 2 (HK2), and enolase 2 (ENO2). Functional studies showed that CHD1 knockout promoted angiogenesis and lactate production in vitro. Since angiogenesis and glycolysis are crucial for the malignant behavior of PCa, we concluded that CHD1 might accelerate PCa development via angiogenesis and glycolysis.

Overall, our study provides a molecular mechanism by which CHD1 deletion affects metabolic reprogramming and angiogenesis in PCa. We also revealed the relationship between CHD1 deletion and the hypoxia pathway and revealed the causes underlying the dysregulation of genes that may be involved in PCa progression.

AUTHOR CONTRIBUTIONS

YLC and LL conceived and designed the experiments. YLC and YZW drafted the manuscript. YLC and YZW performed the experiments and analyzed the RNA-seq data. YCQ, LHY, WJY, GDG, MXH, and WL collected and prepared the samples. XYF and KW provided technical assistance with immunoprecipitation and western blot. ZY and YG provided suggestions regarding data analysis. All authors read and approved the final manuscript.

COMPETING INTERESTS

All authors declare no competing interests.