Role of the DEAD-box RNA helicase DDX5 (p68) in cancer DNA repair, immune suppression, cancer metabolic control, virus infection promotion, and human microbiome (microbiota) negative influence

Abstract

There is increasing evidence indicating the significant role of DDX5 (also called p68), acting as a master regulator and a potential biomarker and target, in tumorigenesis, proliferation, metastasis and treatment resistance for cancer therapy. However, DDX5 has also been reported to act as an oncosuppressor. These seemingly contradictory observations can be reconciled by DDX5's role in DNA repair. This is because cancer cell apoptosis and malignant transformation can represent the two possible outcomes of a single process regulated by DDX5, reflecting different intensity of DNA damage. Thus, targeting DDX5 could potentially shift cancer cells from a growth-arrested state (necessary for DNA repair) to apoptosis and cell killing. In addition to the increasingly recognized role of DDX5 in global genome stability surveillance and DNA damage repair, DDX5 has been implicated in multiple oncogenic signaling pathways. DDX5 appears to utilize distinct signaling cascades via interactions with unique proteins in different types of tissues/cells to elicit opposing roles (e.g., smooth muscle cells versus cancer cells). Such unique features make DDX5 an intriguing therapeutic target for the treatment of human cancers, with limited low toxicity to normal tissues. In this review, we discuss the multifaceted functions of DDX5 in DNA repair in cancer, immune suppression, oncogenic metabolic rewiring, virus infection promotion, and negative impact on the human microbiome (microbiota). We also provide new data showing that FL118, a molecular glue DDX5 degrader, selectively works against current treatment-resistant prostate cancer organoids/cells. Altogether, current studies demonstrate that DDX5 may represent a unique oncotarget for effectively conquering cancer with minimal toxicity to normal tissues.

Keywords: Cancer metabolic control; Cancer therapeutics; DDX5; DDX5 degrader FL118; DNA repair; Immune suppression; Microbiota negative influence; Oncotarget; Virus infection promotion; p68.

Fig. 1

As a TF co-activator and an RNA helicase, DDX5 (p68) has diverse functions to act as an oncogenic master regulator: A DDX5-mediated transcriptional activation of many oncogenic genes by interacting with different TFs (e.g., c-Myc) together with CDK9/cyclin T1 complex, which bridges basic transcription machinery on the oncogene promoters; B DDX5 regulation of mRNA stability by interacting with RNA stability modulators (e.g., interacting with the IGF2BPs complex); C DDX5-regulated DNA conformational changes. This is achieved (i) through interaction with various topoisomerases (TOPs) to alter DNA topological structure, and (ii) through interaction with toposome (e.g., interacting with topoisomerase 2α-containing complex) for chromosome segregation; D DDX5 interaction with DNA repair regulators (Table 1) and involvement in DNA repair and genome surveillance. For example, DDX5 participates in NHEJ DNA repair by interacting with NF45/NF90 and Ku70/Ku80 complexes, and DDX5 participates in NER and facilitates DNA repair by interacting with replication factor C (RFC) proteins; E DDX5 regulation of the splicing of pre-RNA, microRNA (miRNA) and circular RNA (circRNA; e.g., Wang et al., Aging US 2023, 15:2525–40); and F DDX5 regulation of ribosome biogenesis through interacting with other regulators such as nucleophosmin 1 (NPM1) on the ribosomal DNA promoter. Note: The diverse functions of DDX5 presented herein were based on many literature reports but primarily based on the recent publication from Le et al., Mol Ther 2023 Feb 1; 31(2):471–486

Fig. 2

DDX5 plays a central role in resolving DNA replication / transcription-coupled DNA/RNA R-loop formation: A1 PRMT5-mediated methylation of DDX5 on the arginine residue in RGG motif is required for DDX5 to recruit XRN2 to become a complex to resolve DNA/RNA R-loop formation. A2 Thrap3 interaction with the RGG motif-methylated DDX5 to recruit XRN2 to resolve R-loop formation. B UAF1 is a mediator to form a ATAD5-UAF1-DDX5 complex to resolve R-loop formation. C Sox2 interacted with and inhibited DDX5 and thus stabilized R-loop formulation for somatic cell reprogramming into iPSCs. D BTAC2 interacts with and helps retain DDX5 to help DDX5 on DNA damage sites to resolve R-loop formulation. E1 TOP3B interacts with DDX5 to resolve R-loop formation. E2 TCOF1 interacts with DDX5 to resolve R-loop formation. F THOC5 recruits both DDX5 and DDX5 paralog DDX17 to resolve R-loop formation. G LncRNA Lnc530 recruits DDX5 and TDPJ-43 to prevent R-loop formation. H Hypoxia decreases DDX5 expression and thus, blocks DDX5 from accessing various forms of DNA

Fig. 3

DDX5 functionally suppresses immune system: A DDX5 is involved in transcriptional expression of IL-1beta to recruit neutrophils to glioma. B DDX5 stabilizes STING protein and blocks IFN-beta production to inhibit innate immune responses (IIRs). C PARP1 binds to and ADP-ribosylates DDX5 to inhibit DDX5 and promote CD24 expression for immune suppression. D1 DDX5 inhibits HIF1α-mediated IL-10 expression and contact-dependent suppressor function in RORγt⁺ T_reg and promotes T cell–mediated inflammation in the intestine. D2 DDX5 resolves the R-loop to block RNA Pol II loading and inhibit Hif1α transcription

Fig. 4

DDX5 plays a role in cancer metabolic control: A DBP2, a DDX5 yeast ortholog, promotes glucose-dependent gene expression and upregulates the expression levels of HXTs. B1 DDX5 or DBP2 is required for efficient glucose import. B2 DDX5 promotes glycolysis. C DDX5 recruiting with UCP2 at least partially contributes to the metabolic plasticity of NSCLCs via the AKT/mTOR pathway. D CSN6-mediated induction of PHGDH and metabolic reprogramming relies on DDX5, specifically, the E3 ligase β-Trcp interacts with, ubiquitinates and degrades DDX5, which can be blocked by CSN6 to stabilize DDX5 protein and in turn promote DDX5-mediated PHGDH mRNA stabilization, leading to metabolic reprogramming in CRC cells thus, promoting tumorigenesis reflected by poor CRC patient prognosis

Fig. 5

DDX5 modulates virus infection and replication: A DDX5 suppresses IFN-I antiviral IIRs by interacting with PP2A-Cβ to deactivate IRF3 to inhibit IFN-I production. B DDX5 inhibits antiviral innate immunity by promoting m6A-methylated antiviral transcripts. (i) DDX5 interacts with METTL3 to regulate methylation of mRNA through affecting the METTL3-METTL14 heterodimer complex; (ii) DDX5 promotes m6A modification and nuclear export of DHX58, p65, and IKKγ transcripts by binding the conserved UGCUGCAG element; (iii) stable IKKγ and p65 transcripts underwent YTHDF2-dependent mRNA decay, whereas DHX58 translation was promoted, resulting in the inhibited antiviral IIRs by DDX5 blocking the p65 pathway and activating the DHX58-TBK1 pathway. As a result, DDX5 suppresses antiviral innate immunity. C DDX5 suppressed IFN-β production and inhibited the expression of IRF1 and thus, promoted MDV replication. D DDX1, DDX5 and DDX6 promoted SARS-CoV-2 infection and replication by suppressing host IIRs, while DDX21 and MOV10 suppressed SARS-CoV-2 infection and replication. E DDX5 suppresses antiviral innate immunity and promotes replication of IAV

Fig. 6

A DDX5 is involved in intestinal inflammation to course colitis and tumorigenesis: B1 DDX5 inhibits lipid and protein metabolism in intestine tuft cells by blocking the expression of genes involved in transmembrane transport and lipid metabolism. B2 High succinate results in tuft cell hyperplasia, which leads to ileitis as well as tumorigenesis. B3 DDX5 promotes Wnt signaling, angiogenesis and integrin signaling, while suppressing transmembrane transport, cytokine signaling and metabolism in cancer cells and high succinate could mimic DDX5’s such effects and roles, which will need further investigation

Fig. 7

DDX5 promotes inflammation and tumorigenesis in colon and small intestine: A DDX5 binds to and promotes C3 mRNA expression, and in turn induces inflammation and tumorigenesis in colon. B DDX5 binds to and promotes FABP1 mRNA expression, and in turn (1) induces inflammation and tumorigenesis and (2) plays a role in lipid homeostasis in small intestine

Fig. 8

FL118 is potentially a superior anticancer drug against soft tissue sarcoma (STS): A, B FL118 (but not DOX) exhibited excellent efficacy against STS (A) with acceptable toxicity (B). The HT1080 STS cells (2 × 10⁶ per tumor site) mixed with 50% Matrigel were subcutaneously injected into 2–3 severe combined immunodeficiency (SCID) mice in the flank area to establish xenograft tumors. The established STS tumors were maintained on SCID mice. STS tumor SCID mice for the planned experimental studies were set up from the STS tumor‐maintained mice. Treatment with vehicle, DOX or FL118 at dose level of 5 mg/kg (DOX’s MTD – maximum tolerated dose) were started when tumors were grown into the size of 150—200 mm³. The schedule and route were weekly × 3 via intravenous (i.v.) administration (arrowed). A STS tumor change curves after vehicle, DOX and FL118 treatment. Each tumor curve is the mean tumor size + SD from 3 SCID mice. B Mouse body weight change curves after vehicle, DOX and FL118 treatment. Each body weight change curve is the mean body weight change + SD from 3 SCID mice. C, D FL118 exhibited high efficacy to regress STS tumors at FL118’s sub-MTD (C) with acceptable toxicity (D). HT1080 STS tumor establishment, experimental tumor mouse set up and treatment are the same as in A and B. The schedule and route were weekly × 4 via oral administration (arrowed). C STS tumor change curves after treatment with vehicle or FL118 at different dose levels as shown. Each tumor curve is the mean tumor size + SD from 3 SCID mice. D Mouse body weight change curves after treatment with vehicle or FL118 at different dose levels as shown. Each body weight change curve is the mean body weight change + SD from 3 SCID mice

Fig. 9

FL118 preferentially inhibits AR^−/lo LAPC9-AI cells in organoid screening assays: A Experimental schema. LAPC9-AD/AI cells were purified out from the maintenance tumors. Cytotoxic/cytostatic effects of drug (FL118) treatment were measured using resazurin assays in triplicate culture (5,000 cells/well in 96-well plate). B Representative LAPC9-AD and LAPC9-AI organoid images 6 days after FL118 treatment. C, D FL118 exhibited a more prominent inhibitory effect on AR^−/lo LAPC9 than AR.⁺ LAPC9-AD organoids. Shown in C are bar graphs of LAPC9-AD (top) and LAPC9-AI (bottom) organoids in the presence of vehicle control (DMSO) or increasing concentrations of FL118. Data are normalized to DMSO-treated control samples and presented as mean ± SD. Shown in D are the dose–response curves generated using relative viable cell numbers after treatment with increasing concentrations of FL118 for 4 days. The inhibitory effects (IC₅₀) of FL118 on LAPC9-AI compared to LAPC9-AD are statistically highly significant (P < 0.0001; two-way ANOVA)

Fig. 10

DDX5 mRNA levels are upregulated in PCa and correlate with tumor grade: A, B Increased DDX5 mRNA levels in human PCa. Shown are DDX5 mRNA levels in two different tumor-normal (N) comparisons from the TCGA-PRAD dataset, i.e., 422 treatment-naive PCa (Pri-PCa; A) and 495 total PCa patients in PRAD which includes the 422 Pri-PCa as well as PCa treated with hormone therapy [71] and chemotherapy [2]. **P < 0.01 (Student’s t-test). C Increasing DDX5 mRNA levels in human PCa correlate with tumor grade as indicated by increasing Gleason Scores (GS). DDX5 mRNA levels and corresponding PCa patients’ tumor grade were extracted from UCSC Xena database (http://xena.ucsc.edu/). Data is presented as mean ± SD. Statistical analyses were performed using GraphPad Prism software or using R. Student’s t-test was used to compare PCa of various grades to N while Jonckheere-Terpstra’s trend (J-T) test was used to calculate the statistical significance of the trend across all different groups. *P < 0.05; **P < 0.01; **** P < 0.0001

Fig. 11

DDX5 (also called p68 in early studies) plays important roles in multiple treatment resistant mechanisms: This includes, but may not be limited to, promoting DNA repair and R-loop resolution during DNA replication and gene transcription (this review); promoting immune suppression (this review); controlling cancer metabolism (this review); promoting oncogene expression (e.g., survivin, Mcl1, XIAP, cIAP2, MdmX, ERCC6, c-Myc, mutant Kras (mKras), etc., which can be indirectly inhibited by FL118 [46, 133, 139, 140]); controlling various types of RNA metabolism (e.g., pre-RNA, long and short/small non-coding RNA) and ribosome biogenesis [141]; promoting virus infection and replication (this review); and negatively influences microbiota (this review). On the other hand, while FL118 targets DDX5, FL118 could also bypass many other treatment resistant mechanisms (e.g., overexpression (OE) of ABCG2/BCRP [142, 143] and/or OE of P-gp/MDR-1 [142]; null/mutated p53 (n/mut p53) [140]; cancer stem cell (CSC)-induced treatment resistance [144], etc.)