A novel target for cancer therapy on the RNA stimulation by FUBP1- interacting repressor (FIR) via TFIIH/P62

doi:10.21203/rs.3.rs-1722555/v1

Download PDF

Article

A novel target for cancer therapy on the RNA stimulation by FUBP1- interacting repressor (FIR) via TFIIH/P62

https://doi.org/10.21203/rs.3.rs-1722555/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this older preprint version

Read the latest preprint version →

This study indicated the novel impaired mechanism of ribosomal RNA (rRNA) transcription that causes cancer development. Human transcription factor IIH (TFIIH) is an essential factor with a seven-subunit core, including the P62 subunit, and is pivotal for DNA repair and transcription. Three RNA polymerases (RNAPs), RNAPI, II and III, possess a common RPB6 subunit that interacts with P62 of TFIIH. The FUBP1-interacting repressor (FIR) is a transcriptional suppressor of c-myc via suppressing DNA helicase activity of TFIIH. This study showed that FIR contains a highly conserved acidic string region, like the region of the RPB6 subunit of RNAPI. Knockdown of FIR by siRNA suppressed P62 expression in vivo in cancer cells. The isothermal titration calorimetry (ITC) measurements revealed the molecular interaction between the authentic FIR rather than FIRΔexon2, a dominant-negative splicing form of FIR that lacks the transcriptional repression domain exon2, and P62. Furthermore, knockdown of FIR and FIRΔexon2 by siRNA led to dynamic changes in ribosomal RNA (rRNA) expression assessed by RNA-seq analysis. A qRT-PCR analysis confirmed that FIR and FIRΔexon2 significantly changed the expression of numerous rRNAs on specific chromosomes. Moreover, small molecular chemical compounds against FIR and FIRΔexon2 were screened among 23,275 chemicals by Natural Product Depository (NPDepo) Array at RIKEN (Japan). Among them, BK697 that potentially binds FIR and FIRΔexon2 significantly inhibited tumor cell growth with the suppression of ribosomal proteins, FIR, and FIRΔexon2 expression. In summary, FIR and FIRΔexon2 provide a novel insight into cell proliferation by stimulating dynamic ribosome biosynthesis and a hint of therapeutic strategy.

Carcinogenesis

FUBP1-interacting repressor (FIR)

Ribosome biosynthesis (RiBi)

Ribosomal RNA (rRNA)

ribosomal proteins (RPs)

differentially expressed gene (DEG)

transcription factor IIH (TFIIH)

P62 PH （pleckstrin homology）domain

RNA polymerase (RNAP)

Ribosomal RNA (rRNA) transcription and ribosomal proteins (RPs) synthesis, and ribosome biosynthesis (RiBi) must be tightly regulated and controlled via a concerted feedback mechanism for the cell survival¹. Insufficient rRNA synthesis results in impaired RiBi that will arise “ribosomopathies” ². In humans, three RNA polymerases (RNAPs) are critical for synthesizing RNAs, such as RNAPI for rRNA, RNAPII for mRNA, and RNAPIII for tRNA. The FUBP1-interacting repressor (FIR) is an exon5-lacking splicing variant of poly (U)-binding splicing factor 60 (PUF60)³. FIR suppressed the c-myc transcription by repressing DNA helicase activity of P89/TFIIH⁴. FIRΔexon2, a dominant-negative splicing form of FIR that lacks transcriptional repression domain exon2, is expressed in human cancers⁵. FIR and PUF60 are single-strand (ss) nucleic acid (DNA/RNA)-binding proteins; P62 and P89, the subunits of the transcription factor of IIH (TFIIH) complex, were found to be consistently coimmunoprecipitated with FIR³. FIR total (homo) knockout, FIR^−/−, mouse is embryonic lethal, suggesting that FIR is essential for embryogenesis and compound with TP53 homo knockout (FIR^+/−TP53^−/−) mice promote the organ invasion of T cell acute lymphoblastic leukemia⁶. TFIIH is an elongation factor of RNAPI and the P62 of TFIIH interacts with RPB6, a common amino-terminal tail of the three RNAPI, II, and III^7,8. Ribosomes comprise large and small subunits (60S and 40S) that form a complex of proteins and RNAs ⁹. In human cells, the small 40S ribosomal subunit contains one rRNA (18S) and 33 RPs, whereas the large 60S subunit is built of three rRNAs (28S, 5.8S, and 5S) and 47 RPs¹⁰. In cancer cells, rRNA expression is upregulated and the production of RPs is related to apoptosis inhibition and cell cycle arrest induction^11,12. rRNA is transcribed from ribosomal DNA (rDNA) or genes for rRNA by RNAPI in the nucleolus, and the RPs that constitute the ribosome are generated. However, the impaired mechanism underlying the dynamic regulation of RPs in development, carcinogenesis or ribosomopathy remains largely unexplored.

To develop novel cancer therapy targeting ribosome synthesis, small molecular chemical compounds against FIR and FIRΔexon2 were screened among 23,275 natural chemicals of NPDepo (RIKEN, Wako, Saitama, Japan)^13–15. After identifying the candidates of natural chemicals that potentially targets FIR and FIRΔexon2, the low molecular weight artificial chemical, called BK697, was synthesized by in silico screening and examined the effects on tumor cell growth in vitro.

This study was conducted to investigate how the FIR and FIRΔexon2 are involved in dynamic RiBi though the interaction with the P62/PH (pleckstrin homology)domain of TFIIH. A novel role of FIR or FIRΔexon2 in rRNA transcription and RiBi was proposed by the crystal structure of FIR and comprehensive RNA-sequencing of affected genes. We discuss the dynamics of genes for rRNA expression by FIR stimulation in cancers, neural development or ribosomopathy.

FIR contains the highly conserved acidic string region that potentially interacts with P62 of TFIIH.

The N-terminal tail (NTT) of RPB6 interacts with the PH （pleckstrin homology）domain of the P62 subunit of TFIIH⁸. These FIR family members (FIR, PUF60, and FIRΔexon2) contain three RRMs (RNA recognition motifs; RRM1, RRM2, and RRM3) (Figure 1A,1B). RRM1 binds to the c-myc promotor and RRM2 binds to a transcriptional activator^4,16,17. FIR contains the highly conserved acidic string region (376-KKEKEEEELFPESERPEM-394) between RRM2 and RRM3/UHM (Figure 1A). As the crystal structure is unavailable for the acidic string region, this region is flexible and easily changes its conformation (Figure 1B). Partial domain structures of FIR, RRM1 (PDB#:2QFJ)¹⁸, RRM2 (PDB#:2KXH)¹⁶, and UHM (PDB#:3DXB)¹⁹, were previously elucidated by X-ray crystallography. The FIR protein binds ssDNA of FUSE as a dimer in which only the RRM1 and RRM2 domain of each subunit interacts with ssDNA (Figure 1C)¹⁸. RNAPI, II, and III contain the common NTT in RPB6 (Figure 1D). Interestingly, these acidic strings are highly conserved among FIR, RPB6, UVSSA, XPC, TFIIEα, TP53, and DP1 (Figure 1D). FIRΔexon2, a splicing variant form of FIR that lacks exon2, is overexpressed in cancer cells as a dominant-negative form of FIR^5,6,20,²¹. Given, FIR is crucial for RNAPI expression in cellular development, FIR affects RNAPI expression in a variety of differentiated cells. For this purpose, HepG2 cells (human hepatoblastoma cells), 98G cells (human glioblastoma cells), HCT116 cells (human colorectal cancer cells), and Hela cells (human cervical squamous carcinoma cells) were examined in this study. As anticipated, the knockdown of FIR expression by siRNA led to significant suppression of P62 in HepG2 cells (Figure 2A) and in T98G cells (Figure 2B) as well as in HCT116 cells (Figure 2C). The significant suppression of P62 was determined by densitometry analysis, with β-actin used as an internal control (Figure 2D). These results indicated that FIR is required as a scaffold protein for the stable expression of P62 and the E (glutamic acid)-rich sequence (Figure 1D). Considering that the E-rich acidic amino acid sequence along with the aromatic (F) and hydrophobic (L) amino acids of FIR interacts with P62 of TFIIH (Figure 1D), the expression of RNAPI and rRNA should be affected by FIR and FIRΔexon2. In the case of FIR, 31 (Gr 1) + 16 (Gr 2) = 47 genes were affected by the overexpression of FIR-FLAG, whereas 16 (Gr 2) + 3 (Gr 3) = 19 genes were affected by the knockdown of FIR by siRNA in HeLa cells (Table 1). Similarly, in the case of FIRΔexon2, 22 (Gr 4) + 35 (Gr 5) = 57 genes were affected by the overexpression of FIRΔexon2-FLAG, whereas35 (Gr 5) + 8 (Gr 6) = 43 genes were affected by the knockdown of FIRΔexon2 by siRNA (Table1).

Evaluation of the interaction between FIR and P62 of TFIIH by the isothermal titration calorimetry thermogram

The isothermal titration calorimetry (ITC) thermogram, showing the titration curve of P62 of TFIIH with FIR, suggested a molecular interaction between the two proteins. The exothermic peaks appeared with the earlier injections, and the exothermicity of the injection decreased with subsequent injections. The dissociation constant was calculated as Ka = 2.34 × 10⁷ M⁻¹ from the ITC thermogram (Figure 3A). The binding affinity energy between P62 of TFIIH and FIR was enthalpically favorable as the ITC thermogram revealed that the interaction was exothermic. However, the binding of P62 of TFIIH and FIRΔexon2 was not significant (Figure 3B). These results indicated that FIR interacts with P62 of TFIIH, and the interaction capability with P62 is potentially different between FIR and FIRΔexon2. Why do both FIR and FIRΔexon2 display significant effects on rRNA expression, although only FIR interacts with P62 expression? On possibility is that FIR forms heterodimer with FIRΔexon2 upon RRM1 and RRM2²²; therefore, FIRΔexon2 potentially affects P62 expression through forming heterodimer with FIR on RRM1 and RRM2.

FIR and FIRΔexon2 dynamically stimulated rRNA expression

The expressions of RPL30, RPL37A, RPL38, RPS14, and RPS29 mRNAs were decreased significantly by the overexpression of both FIR-FLAG and FIRΔexon2-FLAG. Similarly, the rRNA expressions of RPL19, RPL37, RPS6, RPS10, RPS15A, and RPS21 were decreased significantly by the overexpression of FIRΔexon2-FLAG in HCT116 cells (Figure 4A). The knockdown or overexpression of FIRΔexon2 significantly decreased the rRNA expressions of RPL19, RPS6, and RPS10 by siRNA. (Figure 4B), indicating FIRΔexon2 is required for sustained rRNA expressions of RPL19, RPS6, and RPS10. The rRNA expressions of RPL37, RPS10, RPS15A, RPL6, RPL29, and RPL23A were increased by the knockdown of both of FIR and FIRΔexon2. Furthermore, we performed RT-PCR to analyze the expression of RPS15A, RPS21, and RPS29 in HeLa cells (Figure S1C, D, and E). The relative RPS21 and RPS29 mRNA expressions were decreased significantly with the overexpression of both FIR and FIRΔexon2-FLAG (top, Figure S1C, D, and E). In contrast, the relative RPS15A mRNA expression was decreased significantly by siFIRΔexon2 (bottom, Figure S1C, D, and E). Knockdown of FIR or FIRΔexon2 by siRNA affected the mRNA expressions of RPS15A, RPS21, and RPS29 in HepG2 cells (Figure S1F). Knockdown of FIR by siRNA affected the mRNA expressions of RPS15A, RPS21, and RPS29 in T98G cells (human neuroblastoma cells) (Figure S1G). The results of comprehensive RNA-seq analysis demonstrated that numerous rRNA-processing proteins were prominently affected by FIR or FIRΔexon2 (Table S1). Numerous RPs were coimmunoprecipitated with FIR and FIRΔexon2 (Table S2)^{22, 23}, suggesting that FIR and FIRΔexon2 coexist with RPs and participate in dynamic rRNA expression.

FIR or FIRΔexon2 affected the dynamic expression of various genes for rRNA synthesis.

The RPs that constitute the ribosome are generated from the rRNAs transcribed from rDNA by RNAPI in the nucleolus. Given that FIR participates in dynamic rRNA transcription, the accessibility of FIR to genes for rRNA synthesis on the specific chromosome would be related to RP synthesis. Overlapped genes affected by both overexpression or knockdown of FIR (Figure S2A) and FIRΔexon2 (Figure S2B) were located on chromosomes 17, 12, 19, 5, 2, 6, 9 and 1 (Figure S2). On the other hand, five human acrocentric chromosomes (13-15, 21 and 22) contain tandemly repeated clusters of genes for rRNA in the satellite region of their short arm²⁴. Robertsonian translocations, centric fusion of the two chromosomes among above five acrocentric chromosomes, indicated 20% lower number of active ribosomal genes than those in the controls²⁴. In such cases, insufficient rRNA synthesis resulting in impaired RiBi called “ribosomopathies”². These results strongly suggested that FIR and FIRΔexon2 targets genes for rRNA synthesis on specific chromosomes that causes “ribosomopathies”.

FIR or FIRΔexon2 in affects the genes responsible for rRNA expression at least partially through a c-Myc-independent mechanism.

c-Myc enhances RNAPI activation by binding the rDNA promoter²⁵. c-Myc regulates protein synthesis by stimulating RPs; RPS and RPL, small and large ribosomal subunits, respectively²⁶. As FIRΔexon2 suppressed c-myc gene transcription as a dominant-negative form of FIR⁵, c-Myc expression was upregulated by the knockdown of FIR by siRNA. The affected genes for rRNA due to both increased and decreased expression of FIR were RPL26 and RPS27, and RPL39. The affected gene transcription due to both increased and decreased expression of FIRΔexon2 were RPL11, RPL3, RPL41, RPL7, RPL7A, RPS25, and RPS3A (Figure 5A, 5B). However, the majority of RPS and PRL genes were independent of c-Myc expression (Figure 5C, D). Here, c-Myc expression was evaluated using the average of FPKM in RPL or RPS genes that were commonly detected in duplicated experiments by the knockdown of FIR or FIRΔexon2 siRNA in HeLa cells under a cut-off level FDR of <0.05 (Table S2). These results indicated that FIR or FIRΔexon2 engages in rRNA expression at least partially through a c-Myc-independent mechanism (Figure 5).

Cell growth inhibition by an in silico screened chemical BK697 against FIR and FIRΔexon2

Small molecular weight chemical compounds potentially bound to FIR and or FIRΔexon2 was previously identified from the Natural Product Depository (NPDepo) at RIKEN (Saitama, Japan), which was a collection of the isolates from natural products (Table S3)^13-15. A small molecular weight chemical that potentially interacting with FIR was identified by NPDepo screening (A01 of Table S3). From computer screening to search synthesized chemicals that mimicking the structure of the identified natural chemical compound A01 (1,4a-Dimethyl-2,3,4,4a,9,9a-hexahydro-1H-fluorene-1,9-dicarboxylic acid) (Table S3) by Namiki database (Namiki Shoji Co., Ltd., Tokyo, Japan) (Figure 6A). The commercially available chemicals were screened that composed of the structure of A01 (Figure 6B-(A)) and identified some chemicals (Figure 6B-(B), (B’)). Chemical skeleton of the two synthesized compounds (Figure 6B-(B), (B’)) were regarded as a Trp (W) and Asp (D) combination (WD motif) mimicking form (Figure 6B-(C))²⁷. Affiliated chemicals were screened and indicated that BK697 (Figure 6B-(D)) effectively suppressed HepG2 cell growth with IC₅₀ of 17.8 mM (Figure 6C). BK697 suppressed FIR and FIRΔexon2 expression in HepG2 cells of their protein (Figure 6D) and mRNA level (Figure 6E). c-Myc expression was decreased along with FIR and FIRΔexon2 suppression by BK697 in Hela cells (Figure 6F). RPS6, RPS10, RPS15A, RPL30, and RPL23A expression was reduced by BK697 with IC₅₀ of 17.8 mM in HepG2 cells (Figure 6G). Clinically, BK697 and its derivatives are potential candidate anticancer drugs for cancers targeting FIR and FIRΔexon2 suppression.

FIR and FIRΔexon2 potentially interact with RPs and monitor their expression

RPs have been identified as coimmunoprecipitated proteins with FIR and FIRΔexon2 (Table S1) ^20,22,23. This supports the notion that FIR and FIRΔexon2 stimulate rRNA expression through P62 binding site between RRM2 and RRM3/UHM (U2AF homology motif) (Figure S3A) and potentially interact with RPs (Tables S1 & S2). Remarkably, the pathogenic germline variants of FIR/PUF60 gene have been accumulated as a cause of Verheij syndrome or CHARGE syndrome, which are congenital diseases related to neural development (Figure S3B). Since ribosome synthesis is critical for neural development²⁸, FIR family potentially bridges neural disease and cancer development through dynamic rRNA stimulation (Figure 7).

FIR and FIRΔexon2 dynamically stimulate RPs and cellular proliferative gene expression.

FIR/PUF60 and c-Myc are located at 8q24.3. FIR contains the highly conserved acidic amino acid string region, a similar sequence of the RPB6 subunit that interacts with P62 of TFIIH. FIRΔexon2, unlike FIR, interacts with P62 of TFIIH and stimulates RP small subunits (RPSs) and those of large subunits (RPLs) (Figure 7) in the nucleolus. Previous studies reports have supported that the FIR family potentially interacts with RPs^20,22,23. Therefore, the FIR family is a bona fide modulator of rRNA transcription of RP via P62/TFIIH. Consequently, FIR and FIRΔexon2 monitor the expression of RPs and participate in RiBi by regulating rRNA expression through the P62 PH domain of TFIIH. These results suggested that FIR and FIRΔexon2 involve in the rRNA expression and cellular proliferative gene expression.

The study has elucidated the novel mechanism underlying the role of FIR and FIRΔexon2 in RP expression. FIR and FIRΔexon2 participate in RP expression through the P62 subunit of TFIIH (Fig. 1). FIR contains the highly conserved E-rich acidic region (376-KKEKEEEELFPESERPEM-394) between RRM2 and UHM (Fig. 1A). The P62 subunit of TFIIH directly interacts with RPB6 of RNAPI⁸, and FIR contains the conserved acidic string region that potentially interacts with P62. The ITC thermogram indicated the direct interaction between FIR and P62 of TFIIH (Fig. 2A). In addition, siRNA of FIR significantly suppressed P62 expression in HepG2 cells (Fig. 1D), in T98G cells (Fig. 1E), and in HCT116 cells (Fig. 1F&G). The comprehensive RNA-seq analysis demonstrated that numerous rRNAs were prominently affected by FIR and FIRΔexon2 (Table 1), indicating that FIR and FIRΔexon2 stimulated dynamic ribosomal expression.

RT-PCR was performed to determine the expression of RPL19, RPL23A, RPL30, RPL36, RPL37, RPL37A, RPL38, RPS6, RPS10, RPS14, RPS15A, RPS16, RPS19, RPS21, and RPS29 mRNAs in HCT116 cells (Fig. 3A, B). The relative RPS21 and RPS29 mRNA expressions were significantly decreased by both FIR and FIRΔexon2-FLAG overexpression, whereas siFIRΔexon2 significantly decreased the relative RPS15A mRNA expression in HeLa cells (Figure S1C&D). These results indicated that FIR and FIRΔexon2 significantly stimulated the mRNA expression of various RPs. The comprehensive RNA-seq analysis demonstrated that the mRNA expression of RPs was prominently affected by FIR and FIRΔexon2 (Table S1). Interestingly, the commonly affected rDNA by FIR and FIRΔexon2 were observed on chromosomes 17, 12, 19, and 1, suggesting that FIR and FIRΔexon2 stimulate rDNA or genes for rRNA transcription on specific chromosomes through at least partially a MYC–independent mechanism (Fig. 5).

FIRΔexon2 participates in the multistep posttranscriptional regulation of BRG1, affecting EMT and promoting tumor proliferation and invasion of gastric cancer cells²⁰. Clinically, pathogenic variants in some of ribosome genes cause Diamond–Blackfan anemia. It is characterized by bone marrow failure and developmental anomalies with rare cancer predisposition syndromes²⁹. The downregulation of RiBi contributes to mouse forebrain development²⁸. Defects and mutations in PUF60, identical to FIR, have been detected in human congenital diseases such as Verheij and CHARGE syndrome^30,31. CHARGE syndrome, caused by pathogenic variants of chromodomain helicase DNA-binding protein 7 (CHD7) gene affects newborns manifesting congenital features such as coloboma of the eye, heart defects, atresia of the nasal choanae, and retardation of growth and/or development³². Moreover, recent genome-wide CRISPR screening identified Puf60, identical to FIR, as a novel stemness (self-renewal) gene^33,34. Moreover, an enrichment network analysis demonstrated that the DOWN gene group could be classified into three categories, viz., “cell cycle,” “metabolism,” and “DNA damage and repair.” The “Metabolism” group includes RPS and RPL biosynthesis genes, rRNA processing, and RNA splicing³⁴.

The complex of RNAP and RNAPI were included in the Cajal body (CB). The small CB-specific RNA (scaRNA) in CB was associated with the synthesis of snRNP in the nucleus and the biochemical modification of specific nucleotides of rRNA and snRNP is a splicing factor (Fig. 7). However, defects and pathogenic variants in PUF60 or chromodomain helicase DNA-binding protein 7 (CHD7) have been found in human congenital diseases, Verheij syndrome, and CHARGE syndrome^30,31. Verheij and CHARGE syndromes are detected in newborns manifesting congenital features such as coloboma of the eye, heart defects, atresia of the nasal choanae, retardation of growth and/or development, genital and/or urinary abnormalities, and ear abnormalities and deafness^32,35,36. The pathogenic variants of PUF60/FIR for Verheij syndrome and CHARGE syndrome generate truncated FIR protein lacking the P62-binding (Figure S3B). PUF60 has a spicing activity via interacting with U2AF65^19,37. Mutations in CHD7 observed in CHARGE syndrome also affect BRG1 expression; therefore, Verheij syndrome and CHARGE syndrome potentially have a common pathogenic mechanism³⁸. FIRΔexon2 is involved in Brahma-related gene 1 (BRG1) in epithelial-mesenchymal transition (EMT) along with CHD7²⁰. PUF60 plays a critical role in the mitochondrial homeostasis by regulating the mRNA stability of Drp1 and links circadian clock and cell metabolism³⁹. Although further studies are required, the dynamic stimulation of rRNA by FIR would provide a novel insight into neural development and carcinogenesis.

RPs (indicated in bold italic with asterisks* in Table S1) were coimmunoprecipitated with FIR or FIRΔexon2²². Pathogenic variants of rRNA genes lead to Diamond–Blackfan anemia coexisting with sarcoma, T-cell acute leukemia, and gastrointestinal cancer²⁹. This study elucidated the novel mechanism indicating that FIR and FIRΔexon2 are critical for human diseases by regulating dynamic rRNA expression.

A novel role of FIR or FIRΔexon2 in genes for rRNA transcription and RiBi through interacting with P62/PH (pleckstrin homology) domain was proposed by the crystal structure of FIR and comprehensive RNA-sequencing of affected genes. This study elucidated the novel mechanism indicating that FIR and FIRΔexon2 are critical for carcinogenesis by regulating dynamic rRNA expression. Small molecular chemical compounds against FIR and FIRΔexon2 were screened among 23,275 chemicals by Natural Product Depository (NPDepo) Array at RIKEN (Japan). Among them, BK697 that potentially binds FIR and FIRΔexon2 significantly inhibited tumor cell growth with the suppression of ribosomal proteins, FIR, and FIRΔexon2 expression. The dynamic stimulation of rRNA by FIR would provide a novel insight into novel therapeutic targets for cancer treatment.

ITC measurement

The binding affinity between FIR∆exon2 and FBW7 was measured using an ITC technique in the MicroCal VP-ITC system (Malvern Panalytical, UK). The sample cells were filled with 1400 μL of 50 mM phosphate buffer, pH 7.4, containing fifteen μM of the purified Skp1-FWB7 complex. The measurement of binding affinity was performed at 30°C. A solution of 50 mM phosphate buffer, pH 7.4, containing 300 μM FIR∆exon2 was injected into the sample cells using syringe for titration. The injection volumes were ten μL each, the injection time was 20 s and 150 s. The titration was repeated twenty-five times.

Binding affinity between FIR∆exon2 and TFIIH P62 PH domain

The titration curve of TFIIH P62 PH domain with FIR ∆exon2 suggested the presence of a molecular interaction between the two proteins. The exothermic peaks were observed with earlier injections. The exothermicity of the injection was decreased with subsequent injections. The dissociation constant was calculated as 1.3 × 10⁶ M⁻¹ from the ITC thermogram. The binding of TFIIH P62 PH domain and FIR∆exon2 was enthalpically favorable because the ITC measurement showed that the interaction was exothermic.

Cell lines, plasmids, and siRNA transfection

HeLa, HepG2, HCT116, and T98G cells were cultured in Iscove's modified Dulbecco's medium supplemented with 10% FBS and 1% penicillin-streptomycin (ThermoFisher Scientific Diagnostics). Cells were cultured at 37°C in 5% CO₂. Plasmids have been described in a previous report.²¹ Briefly, FIR cDNA was inserted into a p3xFLAG-CMV-14 vector (Sigma), and FIRΔexon2 cDNA was inserted into a pcDNA 3.1 plasmid (Invitrogen).

Western blotting

Cells were dissolved with 1:20 β-mercaptoethanol in a 2× sample buffer and incubated at 100°C for 5 min. The cell lysates were measured for protein concentration using the Bradford protein assay (Bio-Rad), and ten μg of cell lysates was separated using SDS-PAGE on 7.5% or 10%–20% XV PANTERA gels. These gels were transferred onto polyvinylidene fluoride membranes using a tank transfer apparatus. The membranes were blocked with 0.5% skimmed milk in phosphate-buffered saline (PBS) for 1 h at room temperature before overnight incubation with primary antibodies at 4°C. This was followed by three 10-min washes with 0.1% Tween 20 in PBS. The membranes were incubated with commercial secondary antibodies for 1 h at room temperature, followed by three 15-min washes with 0.1% Tween 20 in PBS. The primary mouse monoclonal antibody against the FIR C-terminus (total FIRs 6B4) was prepared by Dr. Nozaki. The other primary and secondary antibodies used in this study are listed in Table S1. Antigens were detected using Amersham ECL western blotting detection reagents (GE Healthcare).

RNA extraction and quantitative PCR

Total RNA was extracted from HeLa cells using the MagNA pure compact RNA isolation kit (Roche Diagnostics), and cDNA was synthesized using the first strand cDNA synthesis kit for reverse transcription (Roche Diagnostics). The primer sets for the quantitative PCR (qPCR) of FIR, FIRΔexon2, RPS21, RPS29, RPS15A, FTH1, MT1E, SLC3A2, TUBA1C, TUBB, FSTL1, INSIG1, and SNHG1 are described in Table S6, and the internal control gene was HPRT.

RNA-seq and data analysis

RNA-seq analysis was performed on HeLa cells transfected with FIR-FLAG, FIRΔexon2-FLAG, FLAG vector, or untreated samples, and with siFIR, siFIRΔexon2, or siGL2. RNA-seq libraries were constructed from the total RNA samples using the TruSeq Stranded mRNA Sample Prep Kit (Illumina), according to the manufacturer’s protocol. Deep sequencing was performed on the Illumina HiSeq1500 or NextSeq500 platforms. Sequenced reads from RNA-seq data were aligned by using Hisat2, and Cufflinks was used for transcript assembly. Gene expression levels were expressed as fragments per kilobase of exon per million mapped sequence reads (FPKM). Differential expression analysis was conducted with the TCC software package in R 3.4, and the upregulated or downregulated genes were analyzed using DAVID Bioinformatics Resources (http://david.abcc.ncifcrf.gov/).

Procedure of in silico screening

Small molecular chemical compounds against His-tagged FIR (His-FIR, 645 mg/mL) and -FIRΔexon2 (His-FIRΔexon2, 652 mg/mL) were screened among 23,275 chemicals of Natural Product Depository (NPDepo) at RIKEN (Japan) as described previously^13-15. In the process for searching potent compounds, in silico screening was performed from the commercial chemical database. First, 1,000 compounds were selected from the Namiki database that contains 5 million chemical entries, from the viewpoint of structural similarity to natural product that was identified to be bound to FIR in our previous work. Second, 125 compounds were extracted from the selected 1,000 chemicals in terms of the electrostatic potential caused by the distribution of positive and negative charges. Finally, 5 compounds were purchased from a supplier for experimental assay. Namiki database (Namiki Shoji Co., Ltd., Tokyo, Japan, https://www.namiki-s.co.jp/english/) was used for screening candidate chemicals that was composed of commercially available chemicals.

MTS assay (Cell proliferation assay)

One day before the chemical treatment, cells were cultured in 100 mL medium in flat-bottomed 96-well plates so that the cells will reach 40%-80% confluent at the time of chemical treatment. After 24 h incubation at 37 °C/5% CO₂, cells were treated with chemicals including BK697. After 24 h incubation at 37 °C, CellTiter 96® AQueous One Solution Reagent (Promega, Madison, WI, USA) was added to each well according to the manufacturer’s instructions. Briefly, CellTiter 96® AQueous One Solution Reagent was warmed up and added to each well (20 mL/well), incubated for 1 h at 37 °C. Then, 10% SDS solution was added to each well (25 mL/well). Cell viability was determined by measuring the absorbance at 490 nm using a 550 Bio-Rad plate reader. All samples are tested in duplicate, absorbencies were tested 3 times. Same volume of DMSO was used as negative control. Same volume of 3% H₂O₂ was used as positive control.

Statistical analyses

The expression of FIR and FIRΔexon2 mRNA was compared with that of FIR-FLAG, FIRΔexon2-FLAG, FLAG vector, and the untreated sample, or that of siGL2, siFIR, and siFIRΔexon2 using Student’s t-tests. Statistical analyses were conducted using GraphPad Prism version 6.0 for Windows (GraphPad Software).

Author Contributions: K.M. designed the entire study. K.K. principally performed the experiment, A.O., M.F. B.R., and A.K. supported RNA-seq, and data analysis, N.T. and S.K accomplished western blot and immunohistochemistry, T.M and T.T. helped scientific discussion. T.S and M.U supported additional experiments to prepare the manuscript.

Acknowledgments: The authors appreciate Dr Hiroyuki Osada and his coworkers (RIKEN, Wako, Saitama, Japan) for screening compounds bound to FIR/FIRΔexon2 from a chemical library which was a collection of the isolates from natural products. This study was supported in part by Grant-in-Aid 26460667 for priority areas in cancer research from “the Ministry of Education, Science, Sports and Culture of Japan” (KAKENHI), AMED (Japan Agency for Medical Research and Development), “Chiba Foundation for Health Promotion & Disease Prevention,” and was partly supported by Extramural Collaborative Research Grant of Cancer Research Institute, Kanazawa University to K.M.

Conflicts of Interest: The authors have no potential conflicts of interest related to this study to disclose.

Piazzi, M., Bavelloni, A., Gallo, A., Faenza, I. & Blalock, W.L. Signal Transduction in Ribosome Biogenesis: A Recipe to Avoid Disaster. Int J Mol Sci. 20, 2718; https://doi.org/10.3390/ijms20112718. (2019).
Poot, M. & Hochstenbach, R. Prevalence and Phenotypic Impact of Robertsonian Translocations. Mol Syndromol. 12, 1–11 (2021).
Liu, J. et al. The FBP interacting repressor targets TFIIH to inhibit activated transcription. Mol Cell. 5, 331–334 (2000).
Liu, J. et al. Defective interplay of activators and repressors with TFIH in xeroderma pigmentosum. Cell. 104, 353–363 (2001).
Matsushita, K. et al. An essential role of alternative splicing of c-myc suppressor FUSE-binding protein-interacting repressor in carcinogenesis. Cancer Res. 66, 1409–1417 (2006).
Matsushita, K. et al. SAP155-mediated splicing of FUSE-binding protein-interacting repressor (FIR) serves as a molecular switch for c-myc gene expression. Mol Cancer Res. 10, 787–799 (2012).
Assfalg, R. et al. TFIIH is an elongation factor of RNA polymerase I. Nucleic Acids Res. 40, 650–659 (2012).
Okuda, M., Suwa, T., Suzuki, H., Yamaguchi, Y. & Nishimura, Y. Three human RNA polymerases interact with TFIIH via a common RPB6 subunit. Nucleic Acids Res. 50, 1–16 (2022).
Wilson, D.N. & Doudna Cate, J.H. The structure and function of the eukaryotic ribosome. Cold Spring Harb Perspect Biol. 4, a011536; https://doi.org/10.1101/cshperspect.a011536. (2012).
Wandrey, F. et al. The NF45/NF90 heterodimer contributes to the biogenesis of 60S ribosomal subunits and influences nucleolar morphology. Mol Cell Biol. 35, 3491–3503 (2015).
Wang, M., Parshin, A.V., Shcherbik, N. & Pestov, D.G. Reduced expression of the mouse ribosomal protein Rpl17 alters the diversity of mature ribosomes by enhancing production of shortened 5.8S rRNA. RNA. 7, 1240–1248 (2015).
Yang, J., Chen, Z., Liu, N. & Chen, Y. Ribosomal protein L10 in mitochondria serves as a regulator for ROS level in pancreatic cancer cells. Redox Biol. 19, 158–165 (2018).
Kanoh N. et al. Immobilization of natural products on glass slides by using a photoaffinity reaction and the detection of protein-small-molecule interactions.Angew Chem Int Ed Engl. 42(45), 5584–5587 (2003).
Osada H. Introduction of new tools for chemical biology research on microbial metabolites. Biosci Biotechnol Biochem. 74(6), 1135–1140 (2010).
Kawatani M. et al. Identification of matrix metalloproteinase inhibitors by chemical arrays. Biosci Biotechnol Biochem. 79(10), 1597–1602 (2015).
Cukier, C.D. et al. Molecular basis of FIR-mediated c-myc transcriptional control. Nat Struct Mol Biol. 17, 58–64 (2010).
Levens, D., Baranello, L. & Kouzine, F. Controlling gene expression by DNA mechanics: emerging insights and challenges. Biophys Rev. 8, 259–268 (2016).
Crichlow, G.V. et al. Dimerization of FIR upon FUSE DNA binding suggests a mechanism of c-myc inhibition. EMBO J. 27, 277–289 (2008).
Corsini, L. et al. Dimerization and Protein Binding Specificity of the U2AF Homology Motif of the Splicing Factor Puf60. J Biol Chem. 284, 630–639 (2009).
Ailiken, G. et al. Posttranscriptional regulation of BRG1 by FIR∆exon2 in gastric cancer. Oncogenesis. 9, 26; https://doi.org/10.1038/s41389-020-0205-4. (2020).
Kajiwara, T. et al. SAP155-mediated c-myc suppressor far-upstream element-binding protein-interacting repressor splicing variants are activated in colon cancer tissues. Cancer Sci. 104, 149–156 (2013).
Matsushita, K. et al. SAP155-mediated splicing of FUSE-binding protein-interacting repressor (FIR) serves as a molecular switch for c-myc gene expression. Mol Cancer Res. 10, 787–799 (2012).
Kimura, A. et al. FIR haplodeficiency promotes splicing to pyruvate kinase M2 in mice thymic lymphoma tissues revealed by six-plex tandem mass tag quantitative proteomic analysis. Oncotarget. 8, 67955–67965 (2017).
Lyapunova, N.A., Porokhovnik, L.N., Kosyakova, N.V., Mandron, I.A. & Tsvetkova, T.G. Effects of the copy number of ribosomal genes (genes for rRNA) on viability of subjects with chromosomal abnormalities. Gene. 611, 47–53 (2017).
Grandori, C. et al. c-Myc binds to human ribosomal DNA and stimulates transcription of rRNA genes by RNA polymerase I. Nat Cell Biol. 7, 311–318 (2005).
van Riggelen, J., Yetil, A. & Felsher, D.W. MYC as a regulator of ribosome biogenesis and protein synthesis. Nat Rev Cancer. 10, 301–309 (2010).
Ogura Y. et al. Disturbed alternative splicing of FIR (PUF60) directed cyclin E overexpression in esophageal cancers. Oncotarget. 9(33), 22929–22944 (2018).
Chau, K.F. et al. Downregulation of ribosome biogenesis during early forebrain development. Elife. 7, e36998; https://doi.org/10.7554/eLife.36998. (2018).
Vlachos, A. et al. Increased risk of colon cancer and osteogenic sarcoma in Diamond-Blackfan anemia. Blood. 132, 2205–2208 (2018).
Verheij, J.B. et al. An 8.35 Mb overlapping interstitial deletion of 8q24 in two patients with coloboma, congenital heart defect, limb abnormalities, psychomotor retardation, and convulsions. Eur J Med Genet. 52, 353–357 (2009).
Dauber, A. et al. SCRIB and PUF60 are primary drivers of the multisystemic phenotypes of the 8q24.3 copy-number variant. Am J Hum Genet. 93, 798–811 (2018).
Pagon, R.A.; Graham, J.M. Jr.; Zonana, J. & Yong, S.L. Coloboma, congenital heart disease, and choanal atresia with multiple anomalies: CHARGE association. J Pediatr. 99(2), 223–227 (1981).
Zentner, G.E. et al. CHD7 functions in the nucleolus as a positive regulator of ribosomal RNA biogenesis. Hum Mol Genet. 19, 3491–3501 (2010).
Zhang, Y. et al. Genome-Wide CRISPR Screen Identifies Puf60 as a Novel Stemness Gene of Mouse Embryonic Stem Cells. Stem Cells Dev. 31(5–6), 132–142 (2022).
El Chehadeh, S. et al. Dominant variants in the splicing factor PUF60 cause a recognizable syndrome with intellectual disability, heart defects and short stature. Eur J Hum Genet. 25, 43–51 (2016).
Kampen, K.R., Sulima, S.O., Vereecke, S. & De Keersmaecker, K. Hallmarks of ribosomopathies. Nucleic Acids Res. 48,1013–1028 (2020).
Page-McCaw, P.S., Amonlirdviman, K. & Sharp, P.A. PUF60: a novel U2AF65-related splicing activity. RNA. 5, 1548–1560 (1999).
Yang, P. et al. Integrative analysis identifies co-dependent gene expression regulation of BRG1 and CHD7 at distal regulatory sites in embryonic stem cells. Bioinformatics. 33, 1916–1920 (2017).
Xu, L. et al. CLOCK regulates Drp1 mRNA stability and mitochondrial homeostasis by interacting with PUF60. Cell Rep. 39, 110635; https://doi.org/10.1016/j.celrep.2022.110635. (2022).

Table 1 is in the supplementary files section.

No competing interests reported.

KitamuraKetalTable1.xlsx
Table 1. RPs genes affected by overexpression or knockdown of FIR or FIRΔexon2 (HeLa cells).
FigureS1.jpg
Figure S1. rRNA expression affected by the knockdown of FIR or FIRΔexon2 in human cell lines. (A) Overexpression or knockdown by siRNA of FIR or FIRΔexon2 affected the expression of various rRNAs in HeLa cells as listed in Table 1. Gr1 (31 genes); RPL12, RPL13, RPL13A, RPL14, RPL17, RPL18, RPL23, RPL24, RPL27, RPL28, RPL29, RPL31, RPL35, RPL35A, RPL6, RPL8, RPLP0, RPLP1, RPS11, RPS12, RPS15, RPS17, RPS17L, RPS2, RPS28, RPS3, RPS5, RPS7, RPS8, RPS9, RPSA. Gr2 (16 genes); RPL18A, RPL19, RPL23A, RPL30, RPL36, RPL37, RPL37A, RPL38, RPS10, RPS14, RPS15A, RPS16, RPS19, RPS21, RPS29, RPS6. Gr3 (3 genes); RPL26, RPL39, RPS27. (B) Gr4 (22 genes); RPL10A, RPL15, RPL17, RPL18, RPL18A, RPL24, RPL29, RPL34, RPL36, RPL6, RPLP1, RPS15, RPS16, RPS19, RPS2, RPS20, RPS27A, RPS4X, RPS5, RPS7, RPS9, RPSA. Gr5 (35 genes); RPL12, RPL13A, RPL14, RPL19, RPL21, RPL23, RPL23A, RPL26, RPL27, RPL30, RPL31, RPL35, RPL35A, RPL37, RPL37A, RPL38, RPL4, RPL5, RPL9, RPLP0, RPS10, RPS11, RPS12, RPS13, RPS14, RPS15A, RPS17, RPS17L, RPS18, RPS21, RPS24, RPS27, RPS29, RPS6, RPS8. Gr6 (8 genes); RPL11, RPL3, RPL39, RPL41, RPL7, RPL7A, RPS25, RPS3A. (C)(D)(E) The FIR family affected ribosomal RNA expression. FIR or FIRΔexon2 affected the expression of endogenous ribosomal RNA. The relative RPS21 (D) and RPS29 (E) mRNA expressions were significantly decreased by either FIR or FIRΔexon2-FLAG overexpression, whereas the relative RPS15A mRNA expression was significantly decreased by siFIRΔexon2 in HeLa cells. (F) The effect of RPS21, RPS29 and RPS15A mRNA by siRNA of FIR or FIRΔexon2 in HepG2 cells. (G) The effect of RPS21, RPS29 and RPS15A mRNA by siRNA of FIR in T98G cells.
FigureS2.jpg
Figure S2. rDNA genes are located on specific chromosomes affected by FIR and FIRΔexon2. (A) Bar graph showing the number of upregulated differentially expressed genes (DEGs) on the chromosome in the case of FIR. Among 66 affected genes for rRNA, 48 genes were on chromosomes 17, 19, 1, 12, 5, 16, 2, 6, 9, and 11, that is 72.7% (48/66) of top 10 chromosomes. (B) Bar graph showing the number of upregulated DEGs on the chromosome in the case of FIRΔexon2. Among 100 affected rRNA genes, 70 genes were on chromosomes 17, 1, 12, 6, 19, 5,2, 9, 11, and 3, that is 70.0% (70/100) of top 10 chromosomes.
FigureS3.jpg
Figure S3. Structure of PUF60/FIR gene and reported pathogenic variants in congenital diseases. Structure of PUF60/FIR. PUF60, FIR, and FIR∆exon2 are splicing variants. (B) pathogenic variants of PUF60/FIR were found in congenital diseases, Verheij syndrome, and CHARGE syndrome. (https://www.ncbi.nlm.nih.gov/clinvar/?term=PUF60%5Bgene%5D&redir=gene). These pathogenic variants generated truncated FIR protein lacking the P62-binding site and RRM3/UHM (U2AF homology motif).
KitamuraKetalTableS1.xlsx
Table S1. Differentially expressed gene (DEG) analysis of ribosomal protein (RP) gene list.
KitamuraKetalTableS2.xls
Table S2. Previously reported immunoprecipitated ribosomal proteins with FIR or FIRΔexon2.
KitamuraKetalTableS3.xls
Table S3. Structures of small molecular weight chemicals that were interacted with His-FIR or His-FIRDexon2 screened by NPDepo.
KitamuraKetalTableS4.xlsx
Table S4. List of antibodies, siRNA and PCR primers used in this study.

Download PDF

Version 1

posted

You are reading this older preprint version

Read the latest preprint version →

A novel target for cancer therapy on the RNA stimulation by FUBP1- interacting repressor (FIR) via TFIIH/P62

Status:

Version 1

Abstract

Figures

Introduction

Results

Discussion

Conclusions

Materials And Methods

Declarations

References

Tables

Additional Declarations

Supplementary Files

Status:

Version 1