Acetylated KHSRP impairs DNA‐damage‐response‐related mRNA decay and facilitates prostate cancer tumorigenesis

Abstract

Androgen‐regulated DNA damage response (DDR) is one of the essential mechanisms in prostate cancer (PCa), a hormone‐sensitive disease. The heterogeneous nuclear ribonucleoprotein K (hnRNPK)‐homology splicing regulatory protein known as far upstream element‐binding protein 2 (KHSRP) is an RNA‐binding protein that can attach to AU‐rich elements in the 3′ untranslated region (3′‐UTR) of messenger RNAs (mRNAs) to mediate mRNA decay and emerges as a critical regulator in the DDR to preserve genome integrity. Nevertheless, how KHSRP responds to androgen‐regulated DDR in PCa development remains unclear. This study found that androgen can significantly induce acetylation of KHSRP, which intrinsically drives tumor growth in xenografted mice. Moreover, enhanced KHSRP acetylation upon androgen stimuli impedes KHSRP‐regulated DDR gene expression, as seen by analyzing RNA sequencing (RNA‐seq) and Gene Set Enrichment Analysis (GSEA) datasets. Additionally, NAD‐dependent protein deacetylase sirtuin‐7 (SIRT7) is a promising deacetylase of KHSRP, and androgen stimuli impairs its interaction with KHSRP to sustain the increased KHSRP acetylation level in PCa. We first report the acetylation of KHSRP induced by androgen, which interrupts the KHSRP‐regulated mRNA decay of the DDR‐related genes to promote the tumorigenesis of PCa. This study provides insight into KHSRP biology and potential therapeutic strategies for PCa treatment, particularly that of castration‐resistant PCa.

Abbreviations

• 3′-UTR
• 3′-untranslated region
• ADT
• androgen deprivation therapy
• AR
• androgen receptor
• AREs
• adenine uridine (AU)-rich elements
• ATM
• Ataxia-telangiectasia mutated kinase
• BRCA
• breast cancer susceptibility gene
• DDR
• DNA damage response
• DHT
• dihydrotestosterone
• GSEA
• Gene Set Enrichment Analysis
• IHC
• immunohistochemistry
• KHSRP
• hnRNP K homology (KH)-type splicing regulatory protein
• mRNAs
• messenger RNAs
• NAM
• nicotinamide
• PARP
• poly(ADP-ribose) polymerase
• PCa
• prostate cancer
• PTMs
• post-translational modifications
• RBPs
• RNA-binding proteins
• RNA-seq
• RNA sequencing
• SIRT7
• sirtuin-7
• SRB
• Sulforhodamine B
• TL
• terminal loop
• TSA
• trichostatin A

Introduction

Prostate cancer (PCa) is the second diagnosed malignant tumor in men and the sixth in mortality in all tumors. Finding more effective means to diagnose and treat PCa has been urgent and essential [1]. Androgen receptor (AR) plays a significant role in the growth and progression of PCa in all stages and is a crucial therapeutic target in clinical [2,3]. About 80~90% of PCa remains sensitive to androgen at the initial diagnosis, and androgen deprivation therapy (ADT), which aims to reduce serum androgen and inhibit the function of androgen receptors, is currently administered as primary systemic standard treatment for regional or advanced PCa [4,5]. In the early stage of therapy, ADT can often effectively inhibit the growth of PCa and delay the progression of the disease. However, with the duration of anti-androgen therapy extending, the AR gradually obtains androgen-independent activation, resulting in PCa from castration-sensitive to castration-resistant [6–8], making patients face an increased risk of metastasis and eventual death [9].

In castration-resistant PCa, AR is continuously activated to drive tumor progression. It remains clinically valuable to dig therapeutic targets in the AR signaling pathway. Recently, a large subset of DDR gene transcripts in PCa has been reported and functions in enhancing the capacity of DNA damage repair and promoting radioresistance [10,11], which opened up new horizons for castration-resistant PCa treatment. For example, the FDA-approved Olaparib, an inhibitor of pan-Poly(ADP-ribose) Polymerase (PARP) in the DDR pathway, has been clinically used for treating Breast Cancer Gene 1/2 (BRCA1/2)-deficient PCa [12], making research significant on AR downstream signaling, particularly in the DDR pathway.

The KH-type splicing regulatory protein, KHSRP, was reportedly targeted for phosphorylation by ATM in response to DNA damage, which is essential for pri-miRNA processing [13,14]. Being an AU-RBP with a central region of four KH domains (KH1-4), KHSRP comprises two main functions: miRNA maturation and mRNA decay [15]. In the miRNA processing progress, KHSRP is a critical component of either the Drosha complex that cleaves pri-miRNAs into pre-miRNAs featured with stem-loop structure in the nuclear or the Dicer complex that processes pre-miRNAs into mature miRNA duplexes in the cytoplasm through binding to the terminal loop of regulated miRNAs [16]. Within all the functional sequences in the 3′-UTR of mRNA, AREs characterized by AUUUA pentamer in the AU context have been widely known for the mRNA decay function [17]. Once KHSRP is combined with the AREs of target mRNAs, the third and fourth KH domains of KHSRP quickly start to complex with mRNA decay enzymes, including the poly (A) RNase PARN, the exosome components, and the decapping enzyme DCP2, to realize its mRNA decay function [18–20]. Significantly, the KHSRP function can be influenced by various critical post-translational modifications (PTMs) [13,21,22]. For instance, KHSRP phosphorylation mediated by MAPK/p38 affects the interaction between KHSRP and the AREs of target transcripts and inhibits its promotion of myogenic mRNA degradation [23]. While KHSRP SUMOylation can suppress PCa growth by preventing TL-G-Rich miRNA biogenesis [22]. Despite this, whether KHSRP-regulated mRNA decay in response to androgen stimuli or AR activity in the PCa process is still in the puzzle.

Here, we report that KHSRP acetylation, a new post-translational modification of KHSRP, is tightly associated with AR activity and DDR regulation, which serves tumor growth and malignancy by impairing DDR-related mRNA decay in PCa. The biological relevancy between KHSRP acetylation and DDR was addressed further to understand the AR-regulated DDR in this cancer disease.

Materials and methods

Cell culture and cell lines

Human embryonic kidney 293T (HEK-293T, RRID: CVCL_0063), LNCaP (RRID: CVCL_0395), 22RV1 (RRID: CVCL_1045), and DU145 (RRID: CVCL_0105) cell lines were purchased from the Cell Bank of Type Culture Collection of the Chinese Academy of Sciences (Shanghai, China). LNCaP and 22RV1 cell lines were cultured in Roswell Park Memorial Institute (RPMI) 1640 medium (Hyclone, Logan, UT, USA) with 10% fetal bovine serum (BioSun, Shanghai, China), 100 mg·mL⁻¹ streptomycin, 100 U·mL⁻¹ penicillin (Hyclone), and 1% GlutaMAX (Gibco, Grand Island, NY, USA). HEK-293T and DU145 cell lines were cultured in high-glucose Dulbecco's Modified Eagle Medium (DMEM) (Hyclone) with the abovementioned supplements. All cell lines were cultured at 37 °C in a 5% CO₂ humidified incubator. Cells were authenticated by STR profiling and tested for mycoplasma contamination.

Antibodies and reagents

ACE-lysine antibody (ab21623), anti-KHSRP antibody (ab150393), anti-SIRT7 antibody (ab259968), anti-γH2AX antibody (ab81299), anti-Ki67 antibody (ab15580), and anti-Histone H3 (acetyl K18) antibody (ab40888) were purchased from Abcam (Cambridge, MA, USA). Antibodies against mouse M2 Flag-tag antibody (F1804), dihydrotestosterone (A8380-1G), and Actinomycin D(A9415) were purchased from Sigma-Aldrich (St. Louis, MO, USA), Trichostatin A (TSA) (#S1045), Nicotinamide (NAM) (#S1899), and protease inhibitor cocktail (EDTA Free, 100× in DMSO), ARN-509 (#S2840) were purchased from Selleck (Houston, TX, USA). Pierce™ Protein A/G magnetic beads were purchased from ThermoFisher Scientific (Waltham, MA, USA). HA-Tag Rabbit (C29F4, #3724) was purchased from Cell Signaling Technology (Danvers, MA, USA). IgG fraction monoclonal mouse anti-rabbit IgG (light chain-specific) was purchased from Jackson ImmunoResearch (West Grove, PA, USA). Etoposide (HY-13629) was purchased from MedChemExpress (Monmouth Junction, NJ, USA).

Plasmids, site-directed mutagenesis, transfection, and lentivirus infection

The construction of HA-KHSRP and Flag-KHSRP plasmid has been described previously [22]. Mutations of KHSRP were generated using the KOD-plus-mutagenesis kit (TOYOBO, Osaka, Japan) and the Exnase II enzyme system (Vazyme, Nanjing, China) according to the manufacturer's instructions. Cell transfection was performed using Hiff-Trans™ Liposomal Transfection Reagent (YEASEN, Shanghai, China). To silence the endogenous KHSRP in LNCaP cells and re-applied HA-tagged KHSRP in LNCaP cells, we used the homemade lentiviral vector pGreenPuro-Dual for the recombinant constructs, which can silence endogenous genes and express exogenous genes with a single lentiviral infection as described previously [24]. All the primers are listed in Table S1.

Tissue microarray IHC

IHC was performed on the human PCa tissue microarrays obtained from Shanghai Outdo Biotech Co., Ltd (Shanghai, China). Samples were stained with an anti-KHSRP antibody and homemade KHSRP-acetyl-K205 specific antibody overnight at 4 °C. Subsequently, the secondary antibodies conjugated with horseradish peroxidase were used for 1 h incubation at 37 °C. Finally, a total of 90 pairs of cancer and non-cancer tissue samples were stained, imaged, and then scored by three independent pathologists. After removing incomplete or exfoliated samples, 69 pairs of samples were retained and included in statistics. IHC data are blindly analyzed and scored by three independent pathologists. Staining intensity: 0 = no staining, 1 = weak staining, 2 = moderate staining, 3 = strong staining; the frequency of positive cells: 0 = < 10%, 1 = 10–25%, 2 = 25–50%, 3 = 50–75%, 4 = > 75%. The final IHC score was summed with the staining intensity times the frequency of positive cells.

Nuclear/cytosol fractionation assay

The Nuclear/Cytosol fractionation kit (Beyotime Biotechnology, Shanghai, China) was used to extract nuclear and cytosolic fractions according to the protocol. Briefly, a total of 3 × 10⁶ cells were harvested for Nuclear/Cytosol fractionation and subjected to immunoprecipitation and western blotting assays with the indicated antibodies.

Immunoprecipitation

Cells were harvested and lysed with the RIPA lysis buffer (150 mm NaCl, 50 mm Tris–HCL, pH 7.4, 1% NP-40, 0.01% SDS, and a complete protease inhibitor cocktail). Cell lysates were incubated at 4 °C overnight with Pierce Protein A/G mix magnetic beads and specific antibodies. After washing with the lysis buffer three times, the immunoprecipitants were boiled for SDS/PAGE resolving and immunoblotting.

Acetylation assays

HEK-293T cells were transfected with HA-tagged KHSRP and treated with 2 μm TSA for 16 h and 5 mm NAM for 2 h before harvesting. Cell lysates in RIPA lysis buffer (150 mm NaCl, 50 mm Tris–HCL, pH 7.4, 1% NP-40, and a complete protease inhibitor cocktail) were incubated with magnetic beads and specific antibodies at 4 °C overnight, followed by immunoblotting to determine the acetylation level of KHSRP.

Cell proliferation assay

The SRB colorimetric assay was used to determine cell proliferation. Briefly, cells were seeded in quintuplicate into 96-well plates (3000 cells per well). Cells were fixed with 100 μL 25% trichloroacetic acid (Sigma, T6399) at 4 °C for 1 h. After rinsing the plate with pure water and air drying, cells were stained with 100 μL of 0.05% SRB (S1402, Sigma-Aldrich) in 1% acetic acid for 30 min. The plates were washed with 1% acetic acid to remove the unbound dye. The cells were dried before being added with 200 μL of 10 mm Tris base solution (pH 10.5) to solubilize the dye. Finally, the absorbance was measured at 510 nm to quantify the cell viability.

Animal xenograft

Twenty-four 4-week-old male NOD-SCID mice (NOD C.B7-Prkdcscid/NcrCrl) were purchased from Weitong Lihua Laboratory Animal Technology (Beijing, China) and were kept in an isolated SPF (specific pathogen-free) environment with temperature (25~26 °C), humidity (60~70%), ammonia concentration (less than 14 mg·m⁻³), light intensity (15~20 lux), noise (less than 60 dB), air change (10~15 times·h⁻¹), and 12 h day-night cycle. The mice were randomly divided into four required groups with similar weight, size, and healthy state, and at least five mice in each group were guaranteed after potential attritions. 100 μL of PBS containing 1 × 10⁷ indicated cells were mixed with an equal amount of Matrigel (Corning, 354248, Corning, NY, USA) for subcutaneous injection. Mice were sacrificed at 4 weeks post-injection, and the tumor size and weight were measured to analyze and generate the curve chart. Mice that died or were injured due to biting were excluded from the statistical analysis; however, no mouse died or was injured in this animal trial. All animal operations including breeding, welfare, execution, and protocols, were conducted with the approval and guidance of the Ethical Animal Care and Use Committee of Shanghai Jiao Tong University School of Medicine Affiliated Shanghai Ninth People's Hospital (Document No. HKDL[2018]217).

RNA sequencing and data analysis

The total RNA was extracted using the VAHTS Universal V6 RNA seq Library Prep Kit (Vazyme, NR604-02) for Illumina to construct the cDNA library. The complementary DNA library was sequenced using Illumina Nova seq6000 with 2 × 150 running circles. The unique original Fastq readings were mapped to the human transcriptome using the STAR procedure. The human transcriptome annotation file was retrieved and downloaded from the ensembl genome browser 111 (GRCh38.p14, Cambridge, Cambridgeshire, UK). The human genome was retrieved and downloaded from the ucsc Genome Browser (version hg38, Santa Cruz, CA, USA). Only genes with readings were analyzed using deseq2 to screen differentially expressed genes. The differential transcripts between the two groups were filtered for comparing the treated samples with the control samples according to the differential expression range satisfying |log₂Fc| ≥ 0.2 and P-value ≤ 0.05. The RNA-seq raw data were deposited into the NCBI database (GEO number: GSE206486).

qRT-PCR

Total RNA was extracted from the LNCaP stable cell lines using TRIzol (Invitrogen, Carlsbad, CA, USA). One microgram of total RNA was reverse-transcribed using a PrimeScript™ RT reagent Kit with gDNA Eraser (Takara, Kusatsu, Shiga, Japan). RT-qPCR was performed using SYBR Green (Takara). Primers are listed in Table S1. The gene expression levels of target genes were normalized to that of GAPDH.

Statistical analysis

Statistical analysis was performed using graphpad prism 8 (GraphPad Software, San Diego, CA, USA). Representative data are shown as the Mean ± SD from at least three independent experiments for SRB assay, 3D cell culture, animal xenograft, and qPCR. Paired or unpaired t-test was used for the significance analysis. P < 0.05 was considered as statistically significant (*), < 0.01 very significant (**), or < 0.001 extremely significant (***). Statistical analysis Each presented cell experiment was set in duplicate or triplicate and performed at least three times for the power analysis.

Results

KHSRP can be acetylated mainly at the lysine 205

To detect whether KHSRP can be acetylated, Flag-tagged KHSRP was transfected into the HEK-293T cells. The acetylation of exogenous and endogenous KHSRP was analyzed and distinguishably observed by the immunoprecipitation assay (Fig. 1A,B). KHSRP acetylation was enhanced following the expression of a general acetylase p300 (Fig. 1C). Mass spectrometric analysis of the immunoprecipitated HA-tagged KHSRP confirmed that nine lysine (K) residues in KHSRP were acetylated including K-87, -109, -177, -205, -257, -266, -291, -354, and -628 (Fig. S1). Next, a series of point mutations of K to arginine (R) revealed that K205 significantly reduced the KHSRP acetylation level according to the immunoprecipitation assay (Fig. 1D). Moreover, HEK-293T cells or LNCaP cells harboring HA-KHSRP-WT, -K205R, or -K205Q were extracted for immunoprecipitation, respectively. Western blotting analysis showed that either K205R or K205Q mutation contributed to noticeable reductions in KHSRP acetylation (Fig. 1E,F).

[IMAGE OMITTED. SEE PDF]

To further confirm the acetylation of KHSRP, we manufactured an antibody specifically against acetylated K205. A dot-blot assay verified the specificity of the antibody, and we found that the KHSRP-acetyl-K205 antibody preferentially detected the acetylated but not the unmodified peptide (Fig. S2). Differential acetylation levels between ectopically expressed HA-KHSRP-WT and -K205R in HEK-293T cells were also successfully detected by the KHSRP-acetyl-K205 antibody (Fig. 1G). Together, these results demonstrated that KHSRP is majorly acetylated at K205, and the KHSRP-acetyl-K205 antibody can be used in further analysis.

SIRT7 is a specific deacetylase of KHSRP

Since p300 can acetylate KHSRP, we wonder which deacetylase responds to removing the acetyl group in KHSRP. To identify the deacetylase of KHSRP, we treated HEK-293T cells with the HDAC family inhibitor trichostatin A (TSA) or the Sirtuin family inhibitor NAM, respectively. We found that NAM but not TSA could enhance the KHSRP acetylation level (Fig. 2A,B), suggesting that Sirtuin family members are responsible for KHSRP deacetylation. To determine the subcellular distribution of acetylated KHSRP, we separated the LNCaP cell lysis into cytosolic and nucleoplasmic fractions, and further analysis indicated that KHSRP acetylation both occurs in the cytoplasm and nucleus (Fig. 2C). Considering that KHSRP expression in the cytoplasm is much lower than that in the nucleus, a higher proportion of KHSRP is acetylated in the cytoplasm and may perform essential functions. Flag-SIRT2, -SIRT7, and -HDAC2 deacetylases, all found to exist in both the nucleus and cytoplasm, were logically transfected into HEK-293T cells. The cell lysates were immunoprecipitated for immunoblotting analysis, and the result indicated that SIRT7 could significantly weaken the KHSRP acetylation level (Fig. 2D). Moreover, both Co-IP and reciprocal Co-IP assays revealed that KHSRP could interact with SIRT7 (Fig. 2E,F). The endogenous interaction between KHSRP and SIRT7 in LNCaP cells confirmed the same outcome (Fig. 2G). Additionally, silencing SIRT7 could significantly increase endogenous KHSRP acetylation level (Fig. 2H). Therefore, these results indicated that SIRT7 is responsible for the deacetylation of KHSRP.

[IMAGE OMITTED. SEE PDF]

KHSRP acetylation intrinsically drives tumor growth in PCa

To observe the biological function of KHSRP acetylation in PCa, we generated a series of LNCaP stable cell lines using the homemade lentiviral system, as we previously reported [24]. The expression levels of silenced KHSRP, re-applied HA-tagged KHSRP-WT, or KHSRP-K205R were detected by immunoblotting with anti-KHSRP or anti-HA antibody (Fig. S3A). According to the SRB assay, the KHSRP-silenced cell group exhibited significantly higher proliferation capabilities than the control group. However, re-applied KHSRP-WT diminished the growth advantage induced by KHSRP silencing. In contrast, the growth rate of cells harboring KHSRP-K205R was even lower than that of cells with KHSRP-WT (Fig. 3A). In 3D culture assay and the xenograft model, KHSRP acetylation also showed a similar biological role in vivo and in vitro (Fig. 3B,C). Meanwhile, the harvested tumors were subjected to IHC analysis with Ki67 antibody, and the result showed a higher intensity of Ki67 staining in PCa cells with KHSRP-WT than in those with KHSRP-K205R (Fig. 3D). In addition, we verified the effectiveness of the homemade KHSRP-acetyl-K205 antibody by comparing the staining intense between KHSRP-silenced tumors and negative control tumors as well as the distinguished KHSRP acetylation level between KHSRP-WT tumors and KHSRP-K205R tumors (Fig. 3D). In contrast, the total KHSRP levels did not significantly change in the tumors with HA-KHSRP-WT or -K205R compared to negative control using the HA or KHSRP antibody (Fig. 3D). These data revealed that acetylated KHSRP in PCa cells mainly contributes to tumor growth.

[IMAGE OMITTED. SEE PDF]

KHSRP acetylation is mainly related to the PCa tumor progress in the clinical

More intriguingly, the KHSRP mRNA expression was related to neither the clinical tumor stages of PCa nor the patient's overall survival (Fig. 4A,B). In addition, there was no difference in the expression of KHSRP between patients with low and high Gleason grade PCa (Fig. 4C). Since KHSRP acetylation significantly affects the tumor growth ability of prostate cancer cells, we tried to verify whether KHSRP acetylation is variated in PCa tissue samples. The immunohistochemistry analysis with KHSRP antibody or homemade KHSRP-acetyl-K205 antibody revealed that KHSRP acetylation in tissues with Gleason score ≥ 7 was higher than that in tissues with Gleason score < 7 (Fig. 4D), suggesting that acetylated KHSRP is related to the malignancy of PCa. To further exclude the variation of KHSRP acetylation level caused by the overall KHSRP expression, we screened the samples with a similar term of KHSRP (The difference of KHSRP intensity score between cancer and non-cancer samples ≤ 1/3). The intensity score of KHSRP acetylation in these samples showed that KHSRP acetylation in cancer tissues is higher than that in non-cancer tissues, indicating that KHSRP acetylation may promote the tumorigenesis of PCa (Fig. 4E,F). These results suggested that KHSRP acetylation might be critical in driving PCa growth and malignancy.

[IMAGE OMITTED. SEE PDF]

KHSRP acetylation tightly responds to androgen stimuli or AR activity in PCa

Considering that AR plays a crucial role in PCa development and progression, we intended to identify whether the pro-oncogenic effect of KHSRP acetylation is related to the AR signaling pathway. By treating androgen-dependent LNCaP cells with different concentrations of dihydrotestosterone (DHT), an active form of androgen, we found that KHSRP acetylation could be significantly induced in a dose-dependent manner (Fig. 5A). Importantly, 1 nm DHT was the median DHT level in the blood of PCa patients, and a high DHT concentration is related to worse PCa malignancy [25]. KHSRP acetylation could be induced upon androgen stimuli in the androgen-dependent or independent cell lines LNCaP or DU145 but not 22RV1 cells (Fig. 5B), suggesting that the regulatory function of KHSRP acetylation in response to androgen stimuli is complex in different types of PCa cells. One possible reason is the sensitivity difference among variant or non-variant AR upon androgen stimuli in those diverse cell lines [26]. Intriguingly, KHSRP acetylation was significantly suppressed in LNCaP cells treated with ARN-509, the androgen receptor inhibitor (Fig. 5C). Moreover, the KHSRP acetylation induced by DHT can be reduced again under the additional ARN-509 treatment in both LNCaP and DU145 cells (Fig. 5D,E), confirming that DHT-induced alteration of KHSRP acetylation is tightly associated with AR regulation. It was worth noting that the higher KHSRP acetylation level was accompanied by the lower SIRT7 expression (Fig. 5B,F). To address how DHT affects KHSRP acetylation, we verified that neither the enzymatic activity nor the protein level of SIRT7 in LNCaP cells was impaired after DHT treatment on account of the unchanged acetylation level of histone protein H3K18, a specific substrate of SIRT7 [27] (Fig. 5F,G). However, the interaction between KHSRP and SIRT7 was attenuated upon the DHT stimulation (Fig. 5H and Fig. S3B) and recovered by additional ARN-509 treatment (Fig. 5I). These results revealed that KHSRP acetylation is closely related to androgen stimulation and AR regulation by impairing the binding ability of SIRT7 to KHSRP.

[IMAGE OMITTED. SEE PDF]

KHSRP acetylation is highly associated with DNA damage response

RNA-seq was conducted in the stable LNCaP cell lines to determine the underlying mechanism of KHSRP acetylation in PCa tumorigenesis. Data analysis showed no valuable difference in miRNA biogenesis in cell lines between KHSRP-WT and KHSRP-K205R (Fig. S4). However, KEGG enrichment analysis of differentially expressed mRNA transcripts suggested that KHSRP acetylation-associated genes were significantly enriched in clusters of Homologous recombination, Mismatch repair, Base excision repair, or Fanconi anemia (Fig. 6A). Intriguingly, all those clusters are the crucial components of the DNA damage response (DDR) pathway. Furtherly, we analyzed the differential gene datasets by GSEA and found that these genes were significantly enriched in the DNA repair and damage response-related pathways (Fig. 6B). Since AR has been reported to induce expression of DNA repair genes [28], and KHSRP acetylation is coincidentally related to AR response, we naturally hypothesize that AR-regulated DNA damage response may account for KHSRP acetylation. By aligning differentially expressed genes regulated by KHSRP acetylation with androgen-induced DDR genes from the previous report [28], we surprisingly found a common list of 38 genes (Table S2).

[IMAGE OMITTED. SEE PDF]

Moreover, a total of 25 downregulated genes due to the deacetylation in cells were further screened and validated by qRT-PCR (Fig. 6C–E). Intriguingly, 78% of these 25 genes contain ARE elements in the 3′-UTR of their mRNAs (Fig. 6C and Table 1). To verify further, we found that the LNCaP cells with KHSRP-K205R presented a higher expression of the specific DNA damage marker γ-H2AX than those with KHSRP-WT upon etoposide treatment (Fig. 6F). The qRT-PCR analysis proved the most mRNA of DDR genes degraded faster in LNCaP-K205R cells than that in LNCaP-WT cells (Fig. 6G–J and Fig. S5A–L). To trace the binding ability of KHSRP to mRNAs, the RIP assay revealed that KHSRP significantly binds to the target mRNAs by qRT-PCR detection after immunoprecipitation, particularly the KHSRP-K205R even which promotes the mRNA decay in PCa cells (Fig. 6K). These demonstrated that KHSRP acetylation functions as a downstream regulator of AR and is highly associated with DNA damage response through impeding KHSRP-regulated mRNA decay.

Table 1 The mRNA level of DDR genes and ARE consensus analysis.

Gene symbol	log2Fold change	P-value	ARE
Num.	Consensus (red label)
ALKBH2	−1.034692048	0.0001392860	Null	/
APTX	−0.784623032	0.0344756510	2	CACAAATTTATTCTA
CHEK1	−0.687565841	0.0057937680	1	TTCCAATTTATTTTG
FANCI	−1.561488453	0.0002959490	2	TTCTGATTTACTTGT
MAD2L1	−0.381103256	0.0349528570	11	CATGAATTTATTGCA
MBD4	−2.105604939	0.0000000003	1	TATATATTTAAAAAA
MCM7	−2.771812223	0.0200371610	Null	/
MSH5	−2.113645974	0.0000001980	Null	/
PARP1	−0.25392718	0.0248666080	1	ATACTATTTAGATTT
POLD2	−0.825922184	0.0420094630	Null	/
POLR2I	−0.310788894	0.0488854300	Null	/
RAD18	−2.433783266	0.0144478800	8	AAGAAATTTATGATT
RAD21	−0.20923922	0.0000111000	4	GTTTAATTTAAAACT
RAD51AP1	−3.245648393	0.0467419350	4	TTTATATTTACATTG
RAD54B	−1.693269106	0.0004121380	5	AAGCTATTTATGGCA
RECQL	−0.242384411	0.0198644550	2	AACATATTTATGTTT
RFC1	−1.08034092	0.0463407140	5	CTGGAATTTAGATGT
RFC3	−0.352778699	0.0000539000	7	AGGAGATTTACACAT
RFC5	−0.888966593	0.0069535500	7	CAGGCATTTAAAAAG
RIF1	−10.10402023	0.0024179140	23	GCAGAATTTACTAAG
SSBP1	−1.278927643	0.0354741280	Null	/
TDG	−1.93971224	0.0465848560	Null	/
TDP1	−0.378481333	0.0335728650	3	GTGAAATTTAAGTGT
UPF1	−0.239964813	0.0470835640	3	CCTTCATTTAAAGAA
XRCC5	−0.458643659	0.0003235120	2	GAGAAATTTACTACA

Discussion

Cancer cells have the characteristic of genomic instability and are usually accompanied by defects in the DDR pathway, making it easier to accumulate more damaged DNA than normal cells [29,30]. Defects in the DDR pathway have been a prominent feature, with about 10% of primary tumors and 25% of distant metastases in PCa [31,32]. Clinical practice shows that the development and progression of PCa can be driven by AR activity at all stages of PCa. The most practical treatment option for locally advanced and metastatic PCa is ADT or direct AR antagonism [33,34]. Although these treatments are initially effective, resistant tumors eventually emerge through multiple mechanisms with no durable therapeutic options [33]. Increasing evidence has revealed that PCa events are closely related to AR-regulated DDR, and targeting DDR disorder shows an attractive prospect for clinical therapy [30,33–36]. However, the biological relevance or precise mechanism between DDR and AR activity is still poorly understood.

This work reports that KHSRP, a multifunctional single-strand nucleic acid-binding protein, can be significantly acetylated in vitro and in vivo (Fig. 1), and SIRT7 is a direct deacetylase of KHSRP (Fig. 2). As reported literature, KHSRP is required for miRNA biogenesis and mRNA decay in cells [15–17]. KHSRP deficiency can cause multiple physiological abnormalities and even cancer disease [17]. However, we find that the aberrant acetylation of KHSRP is the authentic and intrinsic driver for tumor growth in PCa (Fig. 3A–C), which is also confirmed in PCa patient samples according to the Gleason Score (Fig. 4D,E). Those data reveal that KHSRP acetylation may play a critical role in the development and progression of PCa. More importantly, KHSRP acetylation can either be induced upon the androgen stimuli or depressed by the AR inhibitor (Fig. 5). A possible reason is the alternated binding ability to KHSRP of the deacetylase SIRT7 alone with the AR activity change (Fig. 5H,I). To some extent, these shreds of evidence explain the scientific nature and effectiveness of the androgen deprivation therapy for PCa in clinical. Moreover, the acetylation level of KHSRP in androgen-independent DU145 cells was significantly higher than that in androgen-dependent LNCaP cells (Fig. 5D), which seems to indicate that KHSRP acetylation is a necessary factor in PCa occurrence and development and AR activity may function as a regulating switch in this process.

More intriguingly, androgen can stimulate KHSRP acetylation in androgen-dependent LNCaP cells and even more in androgen-independent DU145 cells (Fig. 5A–E). It is traditionally considered that DU145 cells do not contain the expression of AR; however, Alimirah et al. [26] reported that DU145 and PC3 cells do not express full-length AR but express the truncated AR, which could not be recognized by the conventional antibodies. According to our present data, therefore, the truncated AR in DU145 cells may also respond to the androgen stimuli, but the response pattern may differ from that of the full-length AR in LNCaP cells. Oddly, KHSRP acetylation in another androgen-dependent 22RV1 cell with full-length AR still cannot be induced by androgen stimulation (Fig. 5B). The great possibility is that SIRT7 expression is much higher in 22RV1 than in LNCaP or DU145 cells (Fig. 5F), resulting to androgen cannot efficiently induce the KHSRP acetylation. All these hints that SIRT7 may have a tremendous biological link within both androgen-dependent and -independent PCa cells, at least, the SIRT7-KHSRP-DDR signaling.

AR signaling in PCa cells has been connected with numerous aspects of DDR pathways, including ATM-Chk2 regulated signaling for DDR initiation [37], poly(ADP-ribose) polymerase function [38], and non-homologous end-joining recombination [28]. Alterations in DDR pathways are thought to be associated with the risk of PCa development, progression, and aggressiveness [30]. DDR defects have been applied for the common treatment in patients with advanced ovarian cancer, like Platinum-based therapies combined with Paclitaxel, which is aimed to cause DNA inter- and intra-strand crosslinks, and the defective DDR system cannot repair such damage [39]. The inhibitor of PARP, an essential gene in the DDR pathway, has also been approved to treat BRCA1/2 deficient PCa [12]. In this present study, we revealed that KHSRP acetylation involves a significant DDR response and could specifically down-regulate the expression of DDR-relative genes (e.g., PARP1, CHEK1, FANCI, RAD21, and RAD54B) (Fig. 6D). Since KHSRP acetylation can respond to androgen stimuli or AR activity (Fig. 5), we believe that KHSRP acetylation may play a critical regulatory role in the AR-mediated DDR and is required for tumor growth in PCa.

KHSRP is an essential regulator of precursor miRNA processing, but we did not observe the miRNA alteration resulting from KHSRP acetylation (Fig. S4). Alternatively, KHSRP acetylation mainly impairs its mRNA decay function and tends to regulate the expression of specific genes by impeding the bind with the AREs in the 3′-UTR regions of the DDR-related genes (Fig. 6C and Table 1), which is because the deacetylation on KHSRP can enhance its affinity to the specific mRNAs or the ability to recognize and decay mRNA substrates (Fig. 6K).

Conclusions

Our study found that KHSRP acetylation as a downstream regulator of AR can widely promote the expression of DDR-related genes, especially the AREs-containing genes, to drive tumor growth in PCa (Fig. 7). This study sheds light on a novel insight into KHSRP biology to deeply comprehend the mechanism of PCa development, and it provides an attractive strategy for PCa treatment, particularly castration-resistant PCa, targeting KHSRP acetylation in the AR-mediated DDR pathway.

[IMAGE OMITTED. SEE PDF]

Acknowledgements

This work was supported by grants from the National Natural Science Foundation of China (81802563 to HY, 81672708 to MX) and Seed Founding of Shanghai Ninth People's Hospital, Shanghai Jiao Tong University School of Medicine (JYZZ008G to HY).

Conflict of interest

The authors declare no conflict of interest.

Author contributions

HY and RC performed most of the experiments and wrote the manuscript; BC and QW helped with all experiments and contributed to the implementation and interpretation of the results; MW, JA, and WA analyzed the RNA-seq data; YT helped with the Animal Xenograft experiment; JY gave the academic suggestion and technical guidance. MX and HY designed and guided the whole study; MX, BJ, YZ, and HY contributed to the project implementation, discussion of the results, providing the ethics committee approval, and responding the peer-review comment; MX is responsible for manuscript revision, editing, and final draft; All authors read and approved the final manuscript.

Peer review

The peer review history for this article is available at .

Data accessibility

All data or datasets used and analyzed in this study are available on reasonable requirements from the corresponding authors.

References

Zhu Y, Mo M, Wei Y, Wu J, Pan J, Freedland SJ, et al. Epidemiology and genomics of prostate cancer in Asian men. Nat Rev Urol. 2021;18:282–301.

Zhao SG, Chen WS, Li H, Foye A, Zhang M, Sjöström M, et al. The DNA methylation landscape of advanced prostate cancer. Nat Genet. 2020;52:778–789.

Knudsen KE, Scher HI. Starving the addiction: new opportunities for durable suppression of AR signaling in prostate cancer. Clin Cancer Res. 2009;15:4792–4798.

Denis LJ, Griffiths K. Endocrine treatment in prostate cancer. Semin Surg Oncol. 2000;18:52–74.

Komura K, Sweeney CJ, Inamoto T, Ibuki N, Azuma H, Kantoff PW. Current treatment strategies for advanced prostate cancer. Int J Urol. 2018;25:220–231.

Knudsen KE, Penning TM. Partners in crime: deregulation of AR activity and androgen synthesis in prostate cancer. Trends Endocrinol Metab. 2010;21:315–324.

Yuan X, Balk SP. Mechanisms mediating androgen receptor reactivation after castration. Urol Oncol. 2009;27:36–41.

Yuan X, Cai C, Chen S, Chen S, Yu Z, Balk SP. Androgen receptor functions in castration‐resistant prostate cancer and mechanisms of resistance to new agents targeting the androgen axis. Oncogene. 2014;33:2815–2825.

Sung H, Ferlay J, Siegel RL, Laversanne M, Soerjomataram I, Jemal A, et al. Global cancer statistics 2020: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin. 2021;71:209–249.

Bartek J, Mistrik M, Bartkova J. Androgen receptor signaling fuels DNA repair and Radioresistance in prostate cancer. Cancer Discov. 2013;3:1222–1224.

Goodwin JF, Schiewer MJ, Dean JL, Schrecengost RS, de Leeuw R, Han SM, et al. A hormone‐DNA repair circuit governs the response to genotoxic insult. Cancer Discov. 2013;3:1254–1271.

Fenton SE, Chalmers ZR, Hussain M. PARP inhibition in advanced prostate cancer. Cancer J. 2021;27:457–464.

Zhang X, Wan G, Berger FG, He X, Lu X. The ATM kinase induces microRNA biogenesis in the DNA damage response. Mol Cell. 2011;41:371–383.

Sidali A, Teotia V, Solaiman NS, Bashir N, Kanagaraj R, Murphy JJ, et al. AU‐rich element RNA binding proteins: at the crossroads of post‐transcriptional regulation and genome integrity. Int J Mol Sci. 2021;23: [eLocator: 96].

Briata P, Chen CY, Ramos A, Gherzi R. Functional and molecular insights into KSRP function in mRNA decay. Biochim Biophys Acta. 2013;1829:689–694.

Trabucchi M, Briata P, Garcia‐Mayoral M, Haase AD, Filipowicz W, Ramos A, et al. The RNA‐binding protein KSRP promotes the biogenesis of a subset of microRNAs. Nature. 2009;459:1010–1014.

Briata P, Bordo D, Puppo M, Gorlero F, Rossi M, Perrone‐Bizzozero N, et al. Diverse roles of the nucleic acid‐binding protein KHSRP in cell differentiation and disease. Wiley Interdiscip Rev RNA. 2016;7:227–240.

Gherzi R, Lee KY, Briata P, Wegmüller D, Moroni C, Karin M, et al. A KH domain RNA binding protein, KSRP, promotes ARE‐directed mRNA turnover by recruiting the degradation machinery. Mol Cell. 2004;14:571–583.

Chou CF, Mulky A, Maitra S, Lin WJ, Gherzi R, Kappes J, et al. Tethering KSRP, a decay‐promoting AU‐rich element‐binding protein, to mRNAs elicits mRNA decay. Mol Cell Biol. 2006;26:3695–3706.

Chen CY, Gherzi R, Ong SE, Chan EL, Raijmakers R, Pruijn GJ, et al. AU binding proteins recruit the exosome to degrade ARE‐containing mRNAs. Cell. 2001;107:451–464.

Kung YA, Hung CT, Chien KY, Shih SR. Control of the negative IRES trans‐acting factor KHSRP by ubiquitination. Nucleic Acids Res. 2017;45:271–287.

Yuan H, Deng R, Zhao X, Chen R, Hou G, Zhang H, et al. SUMO1 modification of KHSRP regulates tumorigenesis by preventing the TL‐G‐rich miRNA biogenesis. Mol Cancer. 2017;16:157.

Briata P, Forcales SV, Ponassi M, Corte G, Chen CY, Karin M, et al. p38‐dependent phosphorylation of the mRNA decay‐promoting factor KSRP controls the stability of select myogenic transcripts. Mol Cell. 2005;20:891–903.

Lu W, Wang Q, Xu C, Yuan H, Fan Q, Chen B, et al. SUMOylation is essential for Sirt2 tumor‐suppressor function in neuroblastoma. Neoplasia. 2021;23:129–139.

Miyoshi Y, Umemoto S, Hiroji U, Honma S, Shibata Y, Kubota Y. High TESTOSTERONE levels in prostate tissue obtained by needle biopsy correlate with poor prognosis factors in prostate cancer patients. J Urol. 2013;189: [eLocator: E872].

Alimirah F, Chen J, Basrawala Z, Xin H, Choubey D. DU‐145 and PC‐3 human prostate cancer cell lines express androgen receptor: implications for the androgen receptor functions and regulation. FEBS Lett. 2006;580:2294–2300.

Barber MF, Michishita‐Kioi E, Xi Y, Tasselli L, Kioi M, Moqtaderi Z, et al. SIRT7 links H3K18 deacetylation to maintenance of oncogenic transformation. Nature. 2012;487:114–118.

Polkinghorn WR, Parker JS, Lee MX, Kass EM, Spratt DE, Iaquinta PJ, et al. Androgen receptor signaling regulates DNA repair in prostate cancers. Cancer Discov. 2013;3:1245–1253.

Prince EW, Balakrishnan I, Shah M, Mulcahy Levy JM, Griesinger AM, Alimova I, et al. Checkpoint kinase 1 expression is an adverse prognostic marker and therapeutic target in MYC‐driven medulloblastoma. Oncotarget. 2016;7:53881–53894.

Schiewer MJ, Knudsen KE. DNA damage response in prostate cancer. Cold Spring Harb Perspect Med. 2019;9: [eLocator: a030486].

Robinson D, Van Allen EM, Wu YM, Schultz N, Lonigro RJ, Mosquera JM, et al. Integrative clinical genomics of advanced prostate cancer. Cell. 2015;161:1215–1228.

Armenia J, Wankowicz SAM, Liu D, Gao J, Kundra R, Reznik E, et al. The long tail of oncogenic drivers in prostate cancer. Nat Genet. 2018;50:645–651.

Einstein DJ, Arai S, Balk SP. Targeting the androgen receptor and overcoming resistance in prostate cancer. Curr Opin Oncol. 2019;31:175–182.

Reddy V, Iskander A, Hwang C, Divine G, Menon M, Barrack ER, et al. Castration‐resistant prostate cancer: androgen receptor inactivation induces telomere DNA damage, and damage response inhibition leads to cell death. PLoS ONE. 2019;14: [eLocator: e0211090].

Zhang W, van Gent DC, Incrocci L, van Weerden WM, Nonnekens J. Role of the DNA damage response in prostate cancer formation, progression and treatment. Prostate Cancer Prostatic Dis. 2020;23:24–37.

Lam HM, Nguyen HM, Labrecque MP, Brown LG, Coleman IM, Gulati R, et al. Durable response of enzalutamide‐resistant prostate cancer to Supraphysiological Testosterone is associated with a multifaceted growth suppression and impaired DNA damage response transcriptomic program in patient‐derived xenografts. Eur Urol. 2020;77:144–155.

Thomas LN, Morehouse TJ, Too CK. Testosterone and prolactin increase carboxypeptidase‐D and nitric oxide levels to promote survival of prostate cancer cells. Prostate. 2012;72:450–460.

Schiewer MJ, Goodwin JF, Han S, Brenner JC, Augello MA, Dean JL, et al. Dual roles of PARP‐1 promote cancer growth and progression. Cancer Discov. 2012;2:1134–1149.

Lord CJ, Ashworth A. The DNA damage response and cancer therapy. Nature. 2012;481:287–294.