PMCCPMCCPMCC

Search tips
Search criteria 

Advanced

 
Logo of jbcThe Journal of Biological Chemistry
 
J Biol Chem. 2016 March 4; 291(10): 5355–5372.
Published online 2015 January 5. doi:  10.1074/jbc.M115.711499
PMCID: PMC4777866

RNA Helicase Associated with AU-rich Element (RHAU/DHX36) Interacts with the 3′-Tail of the Long Non-coding RNA BC200 (BCYRN1)*

Abstract

RNA helicase associated with AU-rich element (RHAU) is an ATP-dependent RNA helicase that demonstrates high affinity for quadruplex structures in DNA and RNA. To elucidate the significance of these quadruplex-RHAU interactions, we have performed RNA co-immunoprecipitation screens to identify novel RNAs bound to RHAU and characterize their function. In the course of this study, we have identified the non-coding RNA BC200 (BCYRN1) as specifically enriched upon RHAU immunoprecipitation. Although BC200 does not adopt a quadruplex structure and does not bind the quadruplex-interacting motif of RHAU, it has direct affinity for RHAU in vitro. Specifically designed BC200 truncations and RNase footprinting assays demonstrate that RHAU binds to an adenosine-rich region near the 3′-end of the RNA. RHAU truncations support binding that is dependent upon a region within the C terminus and is specific to RHAU isoform 1. Tests performed to assess whether BC200 interferes with RHAU helicase activity have demonstrated the ability of BC200 to act as an acceptor of unwound quadruplexes via a cytosine-rich region near the 3′-end of the RNA. Furthermore, an interaction between BC200 and the quadruplex-containing telomerase RNA was confirmed by pull-down assays of the endogenous RNAs. This leads to the possibility that RHAU may direct BC200 to bind and exert regulatory functions at quadruplex-containing RNA or DNA sequences.

Keywords: G-quadruplex, long noncoding RNA (long ncRNA, lncRNA), RNA helicase, RNA structure, RNA-protein interaction, BC200, BCYRN1, DHX36, RHAU, telomerase RNA

Introduction

RNA helicase associated with AU-rich element (RHAU),8 also known as DHX36 and G4R1, is an ATP-dependent DEAH-box RNA helicase initially characterized as a regulator of mRNA stability via binding to the AU-rich element of urokinase plasminogen activator mRNA (1). Subsequent to this, the discovery that RHAU possesses the ability to unwind both DNA and RNA quadruplexes shifted the focus of RHAU to a role in quadruplex biology (2,4).

Quadruplexes are stable secondary structures that occur in guanine-rich nucleic acids through the alignment and hydrogen bonding of guanines in multiple tetrad planes (5). A typical requirement for quadruplex formation is the presence of four tracts of three or more guanines that align in parallel or anti-parallel strands with short (<7-nt) interspersed loops (6); however, quadruplexes with longer loops have also been reported (7, 8). Quadruplexes have been shown to possess myriad roles in gene transcription and post-transcriptional regulation of both coding and non-coding RNAs (9,16).

The quadruplex binding specificity of RHAU is mediated by a short N-terminal motif, referred to as the RHAU-specific motif (RSM). In addition to the RSM, RHAU contains a catalytic helicase core (DEXDc and HELICc domains) as well as an HA2 domain and oligosaccharide binding (OB) fold. Whereas the RSM confers quadruplex interacting specificity, maximal quadruplex binding affinity requires the presence of the helicase core (17, 18). The DHX36 gene generates two known splice variants of RHAU. Whereas isoform 1 localizes to both nucleus and cytoplasm, isoform 2 lacks 14 amino acids within the helicase domain and is predominantly expressed in the cytoplasm (1, 19).

Following its initial characterization as a quadruplex-binding helicase, RHAU has been implicated in quadruplex-mediated transcriptional and post-transcriptional gene regulation (19,21). RHAU has also garnered attention for complicity with several non-coding RNAs, both in the regulation of subcellular localization of specific microRNAs and in the remodeling of quadruplexes within the 5′-region of the telomerase RNA (22,25). Furthermore, RHAU has been related to critical roles in development, differentiation, and immunity (26,29). As such, RHAU appears to perform multiple functions in a number of cellular contexts. Because RHAU is a key helicase that possesses the ability to unwind RNA quadruplexes, it is a valuable tool to study the function of quadruplexes in RNA biology. To further expand upon the role of RHAU in this context, we have performed an RNA co-immunoprecipitation screen to identify novel RHAU-interacting RNAs (21). This screen identified brain cytoplasmic RNA 1 (BCYRN1, BC200) as an RNA specifically enriched upon RHAU immunoprecipitation.

The BC200 gene produces a 200-nucleotide primate-specific transcript expressed predominantly in neural tissue as well as germ cells (30,32). The BC200 RNA possesses three distinct structural regions. The first 120 nucleotides of BC200 are homologous to the left monomer of Alu-J repetitive elements and are predicted to fold into a distinct and conserved structure (Alu domain) similar to that of the 7SL RNA of the signal recognition particle (31, 33). Immediately following the Alu domain resides a 40-nt adenosine-rich stretch followed by 40 additional nucleotides of unique sequence that contain a run of 12 consecutive cytosines (31, 32). Although its function has yet to be clearly defined, the BC200 RNA is postulated to play a role in transport of mRNAs as well as subcellular regulation of translation (32, 34).

BC200 is aberrantly expressed in a wide variety of carcinomas (35,37). In addition to its disregulation in cancer, BC200 expression is altered in neurodegenerative disease (38,40). Increased expression of BC200 in tumor cells and reduced expression in the context of neurodegenerative disease is suggestive of a role in the regulation of a subset of genes governing cell survival and proliferation. With altered expression in several disease states and little known about the function of BC200, we set out to evaluate the relationship between the RNA helicase RHAU and this non-coding RNA and assess the implications of the relationship for their respective functions.

In this report, we validate the interaction between BC200 and RHAU in four immortalized cell lines (HEK293T, MCF-7, HeLa, and SK-BR-3). We present evidence that BC200 does not contain a guanine quadruplex and furthermore interacts with RHAU independent of the quadruplex interaction motif (RSM). Digestion of the BC200 RNA with single-strand-specific RNases confirms the basic structural features predicted by homology modeling. Segmentation of the RNA into discrete domains combined with RNase footprinting assays support an interaction dependent upon the adenosine-rich region of BC200. Modifications of the RHAU protein indicate that the interaction is specific to RHAU isoform 1 and requires the presence of the HA2 domain and OB fold within the C terminus of the protein. We conclude that RHAU expression/binding does not regulate BC200 RNA stability and that BC200 does not impair RHAU helicase activity. Finally, we demonstrate that the cytosine-rich tract near the 3′-end of the BC200 RNA can act as a binding partner to unwound quadruplex RNA. These data raise the hypothesis that BC200 may interact with G-rich sequences and its localization at these sites may be facilitated by the quadruplex helicase RHAU.

Experimental Procedures

Cell Culture and Reagents

The HEK293T cell line was a gift from Dr. Thomas Klonisch; the HeLa, MCF-7, T47D, MDA-MB-231, and SK-BR-3 cell lines were a gift from Dr. Spencer Gibson; and the A549 cell line was a gift from Dr. Peter Pelka. The monoclonal murine anti-RHAU hybridoma was a kind gift from Dr. Yoshikuni Nagamine. Cell culture and monoclonal antibody purification were performed as described previously (23). The following additional antibodies were used: mouse anti-synthetic hapten (isotype control; Abcam, Toronto, Canada), mouse anti-GAPDH (PIMA515738, Thermo Fisher Scientific, Ottawa, Canada), mouse anti-His6 tag antibody (ab18184, Abcam), mouse anti-α-tubulin (T6074, Sigma-Aldrich, Oakville, Canada), and mouse anti-FLAG® M2 antibody (F3165, Sigma-Aldrich). Synthetic RNAs and DNA primers were purchased from Integrated DNA Technologies (Coralville, IA). Plasmids containing the BC200 sequence fused to a T7 promoter were synthesized by Genscript Inc. (Piscataway, NJ). The quadruplex stain n-methyl-mesoporphyrin IX was purchased from Frontier Scientific (Logan, UT). SYBR Gold and Lipofectamine RNAiMax were purchased from Life Technologies (Burlington, Canada). pCp-Biotin and pCp-Cy5 were purchased from Jena Bioscience (Jena, Germany). All standard laboratory chemicals and reagents were purchased from Thermo Fisher Scientific. The recombinant full-length RHAU protein, RHAUE335A mutant, RHAUiso.2, RHAUΔRSM, RHAU(1–614), RHAU(1–760), and RHAU(53–105) truncations were cloned using standard molecular biology techniques and expressed and purified as described previously (23, 41).

RNA Immunoprecipitation for RT-PCR Analysis and Western Blotting

RNA immunoprecipitation experiments, RT-PCR analysis, and Western blotting were performed as described previously (21, 23). For RHAU(53–105) competition assays, recombinant RHAU(53–105) was added to cell lysates at the indicated concentrations. For His6-RHAU(53–105) immunoprecipitations, His6-RHAU(53–105) was supplemented to the cell lysates at a concentration of 250 nm. Recombinant RHAU proteins (RHAU truncations, isoform 1, isoform 2, and deletion mutants) were immunoprecipitated with anti-FLAG M2 antibody. Cycling conditions were as follows: 95 °C for 5 min followed by 40 cycles of 95 °C for 10 s and 58 °C for 30 s. PCR products were analyzed by agarose gel electrophoresis following amplification to ensure reaction specificity. The primer sequences used are as follows: telomerase RNA (hTR)-forward, TCTAACCCTAACTGAGAAGGGCGT; hTR-reverse, TGCTCTAGAATGAACGGTGGAAGG; BCYRN1-forward, TAATCCCAGCTCTCAGGGAGGCTAA; BCYRN1-reverse, GGTTGTTGCTTTGAGGGAAGTTACGC; GAPDH-forward, ACCCACTCCTCCACCTTTG; GAPDH-reverse, CTCTTGTGCTCTTGCTGGG.

In Vitro Transcription of BC200 RNAs

The pUC57 plasmid containing the BC200 sequence and various truncations were synthesized by Genscript Inc. (Piscataway, NJ). Constructs contained a T7 promoter and DraI (BC200 and BC(45–200)), BsaI (BC(122–200) and BC50), or BbsI (BC119, BC(51–119), and BC185) linearization restriction endonuclease cutting sites. The 3′-end of the T7 promoter was modified to accommodate in vitro transcription without the addition of aberrant 5′-nucleotides (TATAGG for BC200, BC50, BC119, BC185, and BC(45–200); TATAGA for BC(51–119) and BC(122–200)). Run-off transcription reactions and native RNA purification by size exclusion FPLC were performed as described previously (42). BC200 RNAs were stored in phosphate-buffered saline (137 mm NaCl, 2.7 mm KCl, 10 mm Na2HPO4, 1.8 mm KH2PO4, pH 7.4). The PITX1 Q1, PITX1 Q1Mut, PITX1 Q2, PITX1 Q3, hTR(10–43), hTR(1–20), and 25P1 RNAs were purchased from Integrated DNA Technologies and were described previously (21, 23).

In-gel Quadruplex Staining and Electrophoretic Mobility Shift Assays (EMSAs)

In-gel quadruplex staining with the fluorescent dye n-methyl mesoporphyrin IX was performed as described previously (23). To assess direct protein-RNA interactions, EMSAs were performed with constant RNA concentrations of either 150 nm, in the case of quadruplex forming RNAs, or 50 nm in the case of the BC200 RNAs. Quadruplex-forming RNAs required higher concentration due to poor staining efficiency with the fluorescent dye SYBR Gold. RHAU, RHAU(53–105), or an equal volume of protein storage buffer was added to the binding reactions at the indicated concentrations. Binding reactions were performed in phosphate-buffered saline (PBS) for 10 min at room temperature, after which the products were resolved by native Tris-borate EDTA (TBE) polyacrylamide gel electrophoresis (TBE-PAGE), stained with the fluorescent nucleic acid dye SYBR Gold, and imaged on a Fluorchem Q imaging system using the Cy2 excitation LEDs and emission filters (Protein Simple, San Jose CA). Quantification was performed by densitometry analysis of the free RNA on three independent replicates per condition.

Circular Dichroism Spectroscopy

All CD spectra were obtained using a J-810 spectropolarimeter (Jasco Inc.) and 0.1-cm quartz cell with a 200-μl volume (Hellma). Collection parameters were constant for all samples and include a data pitch of 0.5 nm (spectral band width of 1 nm) using continuous mode with a scan speed of 50 nm/min, response time of 1 s, and accumulation of nine scans. BC200 RNA and truncations were in PBS buffered to pH 7.4 with sample concentrations in the 5–15 μm range. Postcollection, data were converted to m−1 cm−1 absorbance units.

RNA Labeling and Streptavidin Pull-down Assays

To perform streptavidin pull-down assays, RNAs were biotinylated at the 3′-end with T4 RNA ligase (Thermo Fisher Scientific). RNAs were added to a 50-μl reaction containing 50 mm Tris/Tris-HCl (pH 7.5), 10 mm MgCl2, 10 mm DTT, 1 mm ATP, 10% PEG, 100 μg/ml BSA, 50 units of T4 RNA ligase, 3 μm RNA, and 30 μm pCp-Biotin. Reactions were incubated at 12 °C for 3 h, after which the biotinylated RNAs were purified using the GeneJET RNA Cleanup and Concentration microkit (Thermo Fisher Scientific). Biotinylation efficiency was assessed by performing EMSAs with the purified biotinylated RNA and a 2-fold excess of recombinant Streptavidin protein. Streptavidin pull-down assays were performed as described previously (21). 3′-Fluorescent labeling of RNAs was carried out in a similar manner as biotinylation with pCp-Cy5 used in place of pCp-Biotin. 5′-Fluorescent labeling of BC200 was performed by ligation of a synthetic 5′-Cy5-labeled RNA comprising nucleotides 1–44 of BC200 to the in vitro transcribed BC(45–200) as described above. 5′-Labeled BC200 was purified by size exclusion FPLC. Due to low labeling efficiency and incomplete separation, a significant fraction of unlabeled BC(45–200) was present in the final product.

siRNA, Plasmid Transfections, and Western Blotting

Transfection of siRNAs and plasmids and co-transfection experiments were performed in the same manner as described previously (21). Preparation of cell lysates and Western blotting conditions were also described previously (23).

RNase Digestion of the BC200 RNA

To perform RNase H-mediated digestion, DNA oligonucleotides corresponding to the reverse complement of the BC200 RNA spanning the entire RNA length in 20-nucleotide segments were purchased from Integrated DNA Technologies. The BC200 RNA (200 nm) was combined with each DNA oligonucleotide (250 nm) in separate digestion reactions. 30-μl reactions were carried out in PBS containing the RNA, oligonucleotides, 10 mm DTT, and 5 mm MgCl2. Reactions were incubated for 30 min at 37 °C, following which reaction products were separated by native TBE-PAGE and stained with SYBR Gold.

Limited digestion of the BC200 RNA with RNase T1 (Thermo Fisher Scientific) was performed by combining 0.5 units of RNase T1 with the 5′- and 3′-labeled BC200 RNA (100 nm) in PBS (30-μl reaction volume) for 25 min at room temperature. RNase A digestion was performed by combining 25 pg of RNase A (Thermo Fisher Scientific) with end-labeled BC200 (100 nm) in PBS (30-μl reaction volume) for 15 min at room temperature. After incubation, an equal volume of denaturing RNA Load Dye (95% formamide, 0.01% SDS, 0.5 mm EDTA, 0.025% (w/v) Orange G) was added, and samples were heated to 95 °C for 5 min and separated by denaturing TBE-urea PAGE on 10% acrylamide gels. For RNA footprinting assays, the BC200 RNA was preincubated with 300 nm RHAU for 5 min prior to the addition of RNases.

Helicase Assays

RNA helicase assays were performed by combining a quadruplex-forming RNA derived from the 5′-end of the telomerase RNA (hTR(10–43)) with a complementary sequence to which it hybridizes in the full-length telomerase RNA (25P1) (23). hTR(10–43) was synthesized with a 3′ Cy3 fluorescent tag, and 25P1 was synthesized with a 5′ Cy5 fluorescent tag. RNA sequences are shown in Fig. 10B. To unwind the hTR quadruplex and promote duplex formation with 25P1, 16 nm RHAU and 1 mm ATP were added to a reaction containing hTR(10–43) (200 nm) and 25P1 (250 nm) in 50 mm Tris acetate, pH 7.8, 100 mm KCl, 10 mm NaCl, 3 mm MgCl2, 70 mm glycine, and 10% glycerol. Reactions were incubated for 30 min at 30 °C. To assess the impact of BC200 on helicase activity, BC200 was supplemented to reactions at 250 nm, 500 nm, and 1 μm. Control reactions with the non-hydrolyzable ATP analog AMP-PNP were also performed. Time course analyses of helicase activity were performed as above with delayed addition of RHAU and ATP to the reaction at the corresponding time intervals. Reaction products were separated by native TBE-PAGE and visualized by fluorescent excitation with Cy3 and Cy5 LEDs and emission filters and/or staining with SYBR Gold.

FIGURE 10.
BC200 does not impact RHAU helicase activity but rather acts to bind unfolded quadruplex RNA. A, RNA helicase assay demonstrating the ATP-dependent ability of RHAU to unwind the hTR(10–43) quadruplex. hTR(10–43)-Cy3 (200 nm) and the complementary ...

Pull-down of Endogenous BC200 and hTR RNAs with Biotinylated Locked Nucleic Acid (LNA)

3′-Triethylene glycol-biotinylated LNA probes (Exiqon) specific for BC200 and hTR were incubated with 5 mg of HEK293T cell lysate at a concentration of 300 nm in a 0.5-ml volume for 30 min at room temperature. RNA complexes were captured with streptavidin magnetic beads (Thermo Fisher Scientific) and purified as described previously (21). Bound proteins were identified by SDS-PAGE/Western blotting, and bound RNAs were identified by qPCR as described above. LNA probe sequences are as follows: hTR, TGCTCTAGAATGAACGGTGGAA-Biotin; BC200,TTGAGGGAAGTTACGCTTATT-Biotin; Scramble, GTGTAACACGTCTATACGCCCA-Biotin. Alternative primer sets were used for qPCR to avoid the LNA targeting site: BCYRN1-forward, GCCTGTAATCCCAGCTCTCA; BCYRN1-reverse, CCAGGCAGGTCTCGAACTC; hTR-forward, GCGAAGAGTTGGGCTCTG; hTR-reverse, ACTCGCTCCGTTCCTCTT.

Results

RHAU Interacts with the BC200 RNA

We previously reported an interaction between RHAU and the BC200 RNA as part of an RNA immunoprecipitation screen for novel RHAU-interacting RNAs (21). To validate this interaction, we performed RHAU immunoprecipitations with HEK293T cells and assessed BC200 binding to RHAU by quantitative PCR analysis of the co-precipitated RNA. BC200 demonstrated a ~40-fold average enrichment in RHAU-precipitated RNA relative to an approximately equal quantity of total RNA (25 ng) (Fig. 1A). This is in contrast to a negative control, GAPDH mRNA, which demonstrated only marginal enrichment relative to input. Furthermore, immunoprecipitation with an isotype control antibody failed to enrich either of these RNAs. The enrichment of BC200 by RHAU immunoprecipitation (IP) is in a similar range as previously reported RHAU-interacting RNAs (21, 23). Specificity of the PCR was confirmed by agarose gel electrophoresis of the PCR products (Fig. 1B). Efficiency and specificity of the RHAU and isotype control immunoprecipitations were confirmed by Western blotting of the pre-IP, IP, and post-IP samples (Fig. 1C).

FIGURE 1.
RHAU interacts with the BC200 RNA. A, RHAU was immunoprecipitated from HEK293T cell lysates, and co-precipitated RNA was used as a template for reverse transcription/real-time PCR with primers specific for BC200 and GAPDH. Data represent the mean -fold ...

To confirm that the interaction was not unique to the HEK293T cell line, we assessed BC200 expression in a panel of cell lines including HeLa, A549, MCF-7, SK-BR-3, MDA-MB-231, and T47D cells (Fig. 1D). Cell lines expressing moderate (HeLa), high (MCF-7), and low (SK-BR-3) levels of BC200 were employed for further RHAU immunoprecipitations. In all of these cases, RHAU immunoprecipitation demonstrated a significant enrichment of the BC200 RNA with data normalized to GAPDH mRNA for each cell line (Fig. 1E). These results confirm a specific interaction between the endogenously expressed RHAU protein and BC200 RNA.

Interaction of RHAU with the BC200 RNA Is Independent of the RSM

RHAU has been shown to bind quadruplex RNA via a short N-terminal motif referred to as the RSM (17). An N-terminal fragment of RHAU containing this motif (RHAU(53–105)) is sufficient to bind quadruplexes, albeit with lower affinity than the full-length protein (17, 18). We have previously demonstrated the ability of this truncation to disrupt quadruplex-mediated interactions with the full-length protein (21). To assess whether the interaction between RHAU and the BC200 RNA is mediated via the RSM, we performed a competition assay with excess RHAU(53–105) and assessed RHAU binding to a known quadruplex-containing RNA, hTR, alongside BC200. RNA immunoprecipitations were performed in the presence of 0, 5, and 10 μm RHAU(53–105) supplemented to the cell lysates prior to immunoprecipitation. Whereas RHAU(53–105) drastically reduced enrichment of hTR by RHAU-IP, BC200 enrichment remained within error of the untreated samples (Fig. 2A). This finding suggests that the interaction with BC200 is not quadruplex-mediated and does not require the RSM. The efficiency of RHAU immunoprecipitation was not impacted by the presence of RHAU(53–105) as is demonstrated by Western blotting of the pre-IP, IP, and post-IP samples (Fig. 2B).

FIGURE 2.
RHAU binds BC200 in a manner independent of the RHAU-specific motif. A, RHAU was immunoprecipitated from HEK293T cell lysates supplemented with 0, 5, and 10 μm recombinant RHAU(53–105). Co-precipitated RNA was quantified by quantitative ...

To further rule out the possibility that the RSM participates in the interaction, we performed a complementary experiment whereby RNA co-immunoprecipitations were performed with wild-type RHAU and a RHAU mutant lacking an intact RSM (amino acids 54–63 deleted, RHAUΔRSM). RHAU and RHAUΔRSM cDNAs containing silent mutations in an siRNA targeting region were transfected into HEK293T cells. Cells were co-transfected with RHAU siRNA to knock down the endogenous protein. RHAU siRNA alone strongly reduced enrichment of both the telomerase RNA and BC200 (Fig. 2C), and enrichment was rescued in both cases by co-transfection of the RHAU cDNA (Fig. 2C). Upon transfection of the RHAUΔRSM cDNA, no recovery was seen for hTR; however, enrichment of BC200 was enhanced in an identical manner as the wild-type cDNA. RHAU knockdown and the expression and immunoprecipitation of endogenous RHAU as well as exogenous wild-type RHAU and RHAUΔRSM were assessed by Western blotting (Fig. 2D).

The BC200 RNA Does Not Adopt a Quadruplex-containing Secondary Structure

Although the above data demonstrate that the interaction of BC200 with RHAU is independent of the RSM, this does not necessarily rule out the presence of quadruplexes within the BC200 RNA. To test whether BC200 contained any quadruplex-forming regions, we separated the BC200 RNA and several quadruplex-containing RNAs (derived from the PITX1 mRNA) by native TBE-PAGE and stained the RNAs with a quadruplex-specific fluorescent dye n-methyl mesoporphyrin IX (43). Whereas the positive control RNAs (PITX1 Q1–Q3) exhibited intense staining with the dye, the BC200 RNA and a negative control RNA (PITX1 Q1Mut) did not (Fig. 3A). Following fluorescent visualization, the gel was stained for total RNA with toluidine blue (Fig. 3A, gel 2).

FIGURE 3.
The BC200 RNA does not adopt a quadruplex structure. A, native TBE polyacrylamide gel electrophoresis of BC200 alongside quadruplexes derived from the PITX1 mRNA 3′-UTR (PITX1 Q1–Q3) and a quadruplex-deficient mutant (PIX1 Q1Mut) were ...

Recombinant RHAU(53–105) exhibits in vitro affinity for a wide variety of DNA and RNA inter- and intramolecular quadruplexes (1718, 21, 23). To validate the above result, we performed an EMSA with a known quadruplex alongside BC200 under increasing concentrations of RHAU(53–105). Whereas PITX1 Q2 demonstrated a complete shift at 450 nm RHAU(53–105), the BC200 RNA did not exhibit any detectable interaction with protein concentrations as high as 750 nm (Fig. 3B).

To further test whether the endogenous BC200 RNA contains secondary structure elements with RSM affinity not present in the in vitro transcribed BC200, we performed an RNA immunoprecipitation of endogenous BC200 with His6-RHAU(53–105) (Fig. 3C). Immunoprecipitations with His6-RHAU(53–105) were carried out under control and RHAU-knockdown conditions. Whereas His6-RHAU(53–105) demonstrated efficient binding to an RNA containing a confirmed RNA quadruplex (hTR), the RSM was unable to appreciably enrich either BC200 or GAPDH (Fig. 3, C and D). These data solidify the notion that the BC200 RNA does not contain a quadruplex, and the interaction with RHAU is independent of the RSM.

RHAU Binds BC200 and Quadruplexes Directly in Vitro and Concomitant Binding of Both BC200 and Quadruplex RNA Is Not Observed

Because the data indicate that BC200 does not contain a quadruplex and also does not bind RHAU via the RSM, we next assessed whether BC200 demonstrated affinity for RHAU in a direct in vitro binding assay. EMSAs were performed with RHAU and BC200 alongside a set of previously reported quadruplex-forming RNAs. RNAs were incubated with increasing concentrations of full-length recombinant RHAU, and binding reactions were separated by native TBE-PAGE. Binding reactions were repeated in triplicate, and the fraction of free RNA was quantified at each concentration of RHAU (Fig. 4A). Representative gels from each condition are shown in Fig. 4B. The EMSA data demonstrate that RHAU is capable of binding BC200 directly in vitro, with a Kd in the nanomolar range.

FIGURE 4.
The BC200 RNA binds RHAU directly in vitro. A, densitometry analysis of RNA fraction bound in EMSAs of a panel of quadruplex-forming RNAs and BC200 with the full-length recombinant RHAU protein. The RNA fraction bound was calculated by performing densitometry ...

The results of Figs. 2 and and33 clearly demonstrate that the interaction between RHAU and BC200 is independent of the RSM, a region of the protein critical for quadruplex binding. We therefore wanted to test whether RHAU was able to bind both quadruplex RNA and BC200 simultaneously. We performed an EMSA experiment in which Cy5-labeled BC200 and Cy3-labeled hTR(10–43) were combined in a binding reaction with limiting RHAU. In Fig. 4C, BC200 and hTR(10–43) (150 nm) were run separately and together in the presence and absence of 200 nm RHAU. Whereas nearly all of the free BC200 and hTR(10–43) form a complex in the presence of 200 nm RHAU, the combined binding reaction resulted in displacement of a fraction of BC200. Extended separation of the protein-RNA complexes (6% native TBE) shows that hTR(10–43) and BC200 form distinct complexes that migrate with no overlap of the fluorescent signal. Because the presence of quadruplex RNA reduces the binding of BC200 and because a protein-shifted band with both Cy3 and Cy5 fluorescence is not observed, RHAU does not appear to readily bind both RNAs simultaneously under the conditions tested.

The Interaction between RHAU and BC200 Is Dependent upon a Region in the C Terminus of RHAU and Exhibits Isoform-specific Differences in Binding

We next generated a series of RHAU deletion mutants to narrow down the region of binding on the protein. Truncations were made based upon domain boundaries identified by the NCBI Conserved Domain Database (Fig. 5A). Plasmid constructs were produced to express several C-terminally FLAG-tagged RHAU variants, with two truncations (RHAU(1–614) and RHAU(1–760)) demonstrating stable and soluble protein expression. Two additional constructs (RHAU917 and RHAU(210–917)) failed to demonstrate appreciable expression (data not shown). We also tested a panel of additional RHAU variants (RHAU isoform 1, RHAU isoform 2, RHAUΔRSM, and RHAUE335A (helicase-dead)) alongside these truncations.

FIGURE 5.
The interaction between RHAU and BC200 relies on a region within the C terminus of RHAU, and RHAU isoform 2 is unable to bind BC200. A, schematic of the RHAU domain structure as predicted by the NCBI conserved domain database. B, RT-PCR analysis of BC200 ...

To assess the interaction between RHAU and BC200, an RNA co-immunoprecipitation assay was performed with HEK293T cells transfected with plasmids to express the above-mentioned RHAU variants. 24 h post-transfection, RHAU and the variants were immunoprecipitated from cell lysates with anti-FLAG tag antibodies, and the co-precipitating RNA was analyzed by quantitative real-time PCR with primers specific to BC200 and hTR (Fig. 5, B and C). As expected, RHAUΔRSM demonstrated similar binding to BC200 as RHAU isoform 1; however, the interaction with BC200 was completely lost in the case of RHAU isoform 2, RHAU(1–614), and RHAU(1–760). In contrast, the helicase-dead mutant (RHAUE335A) demonstrated a ~6-fold increase in BC200 enrichment (Fig. 5B). In the case of hTR, the results were generally as expected, with all RHAU constructs except RHAUΔRSM showing an interaction (Fig. 5C). Expression and immunoprecipitation efficiency of all RHAU variants were assessed by Western blotting (Fig. 5D).

Because the data suggest that RHAU(1–614) lacks the region necessary to interact with BC200, and RHAU isoform 2 (Δ517–530) also does not interact, we wished to determine whether the loss of interaction with isoform 2 was due to amino acids 517–530 being critical for binding BC200 or whether an altered cellular distribution of isoform 2 and/or a change in interaction partners was the cause for failed enrichment. To further assess the interaction between RHAU isoform 2 and BC200, RHAU isoform 2 was purified from HEK293T cells and used for in vitro EMSAs. EMSA results demonstrate that, whereas the in-cell interaction between isoform 2 and BC200 is abolished, the in vitro affinity for BC200 (and the hTR(10–43) quadruplex) is unchanged (data not shown). In light of this, the localization and interacting partners of RHAU isoform 2 are quite likely very distinct from that of RHAU isoform 1 despite only a small 14-amino acid deletion.

Analysis of the in Vitro Transcribed BC200 RNA by Limited RNase Digestion Supports the Predicted RNA Secondary Structure

The predicted secondary structure of BC200 consists of a 5′-structured domain (Alu domain, nucleotides 1–119) homologous to the 7SL RNA of the signal recognition particle and an unstructured 3′-tail (nucleotides 120–200). This 5′-region is predicted to contain two short stem-loops that form a pseudoknot as well as a third extended stem loop containing two bulges (Fig. 6A) (31).

FIGURE 6.
Analysis of the in vitro transcribed BC200 RNA. A, schematic representing the predicted secondary structure of the BC200 RNA. Nucleotide numbers are presented in red type. Nucleotides participating in predicted pseudoknot formation are highlighted in ...

We probed the secondary structure of the in vitro transcribed RNA by RNase H-directed digestion with 20-mer antisense DNA oligonucleotides covering the length of the RNA. Consistent with the predicted structure, the first 120 nucleotides of BC200 resisted RNase H digestion, presumably due to inaccessibility of the duplexed RNA to the DNA oligonucleotides (Fig. 6B). The 3′-end of the RNA comprising nucleotides 121–200 appears to be primarily in an accessible conformation, because digestion of the RNA with RNase H and DNA oligonucleotides complementary to this region was nearly complete (Fig. 6B).

To further probe the BC200 secondary structure, limited digestion was performed with RNase T1 (specific for single-stranded guanosine residues) and RNase A (preferential for single-stranded cytosine and uridine residues). Digestion was performed on 5′ and 3′ Cy5-labeled BC200 to eliminate any bias for digestion sites near the label (Fig. 6C, left and middle panels). The isolated structured region (BC119) was also 3′-Cy5-labeled and digested with 20-fold higher concentrations of RNase (Fig. 6C, right panel). The RNA ladder labels are inverted in the case of 3′-Cy5 labels to facilitate band identification.

In the context of the full-length RNA (Fig. 6C, left and middle panels), RNase digestion is almost exclusively restricted to the 3′-tail with the exception of high accessibility between nucleotides 80 and 90, corresponding to the predicted loop of the largest stem-loop structure. Digestion of the structured region alone permitted the use of higher RNase concentrations and revealed additional accessible nucleotides (right panel). These data reveal some discrepancies in that the predicted bulges of the longest stem-loop are not accessible to nuclease digestion, whereas several nucleotides expected to be base-pairing were. All nucleotides accessible to RNase digestion are marked with a red arrow in the predicted structure of BC200 (Fig. 6A).

Despite some incongruity, these data largely support the predicted secondary structure. The formation of a pseudoknot near the 5′-end of the RNA is supported by very little accessibility, even with elevated RNase concentration. Digestions are in agreement with predominant base-pairing within the first 120 nucleotides, an accessible loop between nucleotides 80 and 90, and a 3′-tail that is in a more open conformation.

RNA truncations based on the above-mentioned predicted structure were in vitro transcribed and purified by size exclusion chromatography. Truncations made consist of BC50, BC(45–200), BC119, BC(51–119), BC(122–200), and BC185. Analysis of the RNAs by native and denaturing TBE-PAGE demonstrates a pattern in keeping with both the expected lengths and secondary structures (Fig. 7A). This is exemplified in the BC(122–200) RNA, a 79-nucleotide fragment without extensive predicted secondary structure that migrates slower on a native gel than BC119, a 119-nt RNA that is predicted to be predominantly double-stranded. Under denaturing conditions, the RNAs migrate appropriately according to sequence length.

FIGURE 7.
CD spectra of purified BC200 RNA truncations in PBS, pH 7.4, at room temperature. A, BC200 and various RNA truncations were in vitro transcribed and analyzed by native and denaturing TBE polyacrylamide gel electrophoresis. 2 μg of each RNA was ...

The BC200 RNA and all truncations were analyzed by CD spectroscopy to ensure the validity of the truncations used (Fig. 7, B–D). The BC200 spectrum demonstrates features consistent with A-form RNA with a dominant positive peak at 266 nm and negative peak at 210 nm, whereas the analysis of the 3′-tail truncation yielded spectra consistent with previously calculated and experimental spectra for polyadenosine chains (44). Finally, the sum of the individual spectra of BC119 and BC(122–200) as well as BC50 and BC(45–200) yield a spectrum similar to full-length BC200, suggestive that the truncation of the RNA has not grossly impacted the secondary structure of the intact BC200 (Fig. 7, C and D).

RHAU Binds to the Adenosine-rich Region of BC200 Spanning Nucleotides 120–184

The above-described RNAs were 3′-biotinylated with T4 RNA ligase. Biotinylated RNAs were added to HEK293T cell lysate at a concentration of 250 nm. Streptavidin pull-down assays demonstrate that BC200, BC(45–200), BC(122–200), and BC185 all bind endogenous RHAU but not GAPDH (Fig. 8A). The interaction with RHAU was greatly reduced in the case of BC50, BC119, and BC(51–119). To control for specificity of the interaction, Western blots were reprobed with an antibody to GAPDH, and a pull-down was performed in parallel with streptavidin-conjugated beads alone (Fig. 8A). These data suggest that RHAU interacts with the BC200 RNA between nucleotides 120 and 184. Biotinylation efficiency was confirmed by EMSAs with recombinant streptavidin protein (data not shown).

FIGURE 8.
The BC200 RNA and truncations containing the adenosine-rich region bind RHAU. A, streptavidin pull-down assays were performed with the 3′-biotinylated BC200 RNA and the indicated truncations in HEK293T cell lysates. 250 nm RNA was added to 500 ...

To confirm the consistency of the above data in direct interactions, EMSAs were performed with purified full-length RHAU (Fig. 8, B–D). A nearly complete band shift was observed in a binding reaction containing 50 nm BC200 and 100 nm RHAU (Fig. 8B, lane 11). This condition was tested with each of the BC200 truncations (Fig. 8C), and the data were in full agreement with the above-described streptavidin pull-down assay. Each RNA was further tested by performing EMSAs over a full range of RHAU concentrations. Four sets of binding reactions were performed for each condition, and the average RNA fraction bound was determined by densitometry (Fig. 8D).

To confirm that the interaction is facilitated through the 3′-tail in the context of the full-length RNA, footprinting experiments were performed to identify residues protected from RNase digestion by RHAU binding. 5′-Cy5- and 3′-Cy5-labeled BC200 RNA (100 nm) were preincubated with RHAU (300 nm) prior to RNase digestion. Following preincubation, RNase T1 or RNase A was added, and digestion products were analyzed by denaturing TBE-PAGE (Fig. 9A). Protection was most evident at nucleotide 129, with protection also observed at 114, 120, 159, and 165, as indicated by red arrows (Fig. 9B), in agreement with the experiments performed with BC200 truncations.

FIGURE 9.
RNA footprinting analysis of the RHAU-BC200 complex. A, 5′-Cy5 BC200 and 3′-Cy5 BC200 (100 nm) were untreated or preincubated with 300 nm RHAU prior to digestion with the indicated RNases. Digestion products were separated by denaturing ...

BC200 Does Not Inhibit RHAU Helicase Activity but Acts as a Binding Partner of Unwound Quadruplex RNA in Vitro

RHAU has previously been reported to promote RNA decay via binding to an AU-rich element in the urokinase plasminogen activator mRNA (1). To determine whether RHAU impacts steady state levels of the BC200 RNA, we performed RHAU knockdown and overexpression and evaluated BC200 RNA levels by qPCR. Because RHAU did not appear to impact the steady state levels of BC200 over the course of 72 h (data not shown), we next assessed whether BC200 was capable of inhibiting the quadruplex helicase activity of RHAU. We have employed an assay in which a Cy3-labeled quadruplex formed by nucleotides 10–43 of the telomerase RNA (hTR(10–43)) is unwound by RHAU to promote duplex formation with a complementary cytosine-rich Cy5-labeled oligonucleotide (25P1) (Fig. 10A). The 25P1 RNA is derived from a downstream sequence within the telomerase RNA that participates in a long range base-pairing helix with the 5′-end of the RNA. Thus, it is proposed to be the natural base-pairing partner to the quadruplex-prone region of hTR (45,48). We (23) and others (47, 49) have shown that quadruplexes within hTR disrupt the formation of this P1 helix. Upon introduction of RHAU and ATP, >90% of the hTR(10–43) forms a double-stranded RNA with 25P1 (Fig. 10, A and B). To determine whether BC200 inhibits RHAU helicase activity, BC200 was introduced into this helicase assay at equimolar concentrations to hTR(10–43) as well as at a 2- and 4-fold excess (Fig. 10C). In this context, BC200 did not inhibit helicase activity but rather became the preferred binding partner of hTR(10–43), completely displacing the 25P1 RNA. This assay was repeated in the absence of 25P1, and the results indicate that BC200 acts as an efficient binding partner to unwound quadruplex RNA (Fig. 10D).

The Interaction between BC200 and hTR(10–43) Is Facilitated by Quadruplex Unwinding and Relies upon a Cytosine-rich Region at the 3′-End of the BC200 RNA

Because we have demonstrated that BC200 binding to hTR(10–43) is enhanced by RHAU helicase activity, we wished to assess whether this was indeed due to unwinding of the hTR(10–43) quadruplex. To test this, we heated hTR(10–43) to 95 °C in the presence of either 100 mm KCl or 100 mm LiCl. Whereas LiCl has been shown to be unfavorable for quadruplex formation, quadruplex formation is promoted by the presence of KCl (23, 49). Following heating in the presence of LiCl, hTR(10–43) demonstrated an enhanced ability to bind BC200 in the absence of RHAU and ATP, consistent with the hypothesis that the interaction is promoted by quadruplex disruption (Fig. 11A).

FIGURE 11.
The 3′-cytosine-rich stretch of BC200 is essential for binding/stabilization of unwound hTR(10–43) quadruplex. A, hTR(10–43) was heated to 95 °C in the presence of 100 mm KCl or 100 mm LiCl. 200 nm BC200 was combined with ...

30 nucleotides at the 3′-end of BC200 contain 18 cytosines, 12 of which form a continuous run from nucleotides 185–197. Because unwinding of quadruplex results in the presentation of guanosine-rich single-stranded RNA, it was expected that the interaction with BC200 would be mediated by this cytosine-rich stretch of the RNA. To test this hypothesis, RHAU helicase assays were performed with hTR(10–43) along with BC200, BC(45–200), BC(122–200), and BC185 as the available binding partner. Whereas BC200, BC(45–200), and BC(122–200) all demonstrated a RHAU/ATP-dependent interaction with hTR(10–43), BC185 was not able to detectably bind hTR(10–43) in either case (Fig. 11B). This supports the hypothesis that BC200 is stabilizing the unwound hTR quadruplex through interactions with cytosine-rich nucleotides at the 3′-end of the RNA.

Because BC200 was able to out-compete 25P1 for binding to hTR(10–43), we wished to assess the relative efficiency of both RNAs to act as unwound quadruplex acceptors in an RHAU helicase assay. Alongside these RNAs, we also tested the isolated cytosine-rich fragment of BC200 (BC(185–200)). Helicase assays were performed with 200 nm hTR(10–43) combined with 250 nm 25P1, BC200, or BC(185–200) RNA at two concentrations of RHAU (8 and 16 nm). A time course of quadruplex unwinding was performed and evaluated by native TBE-PAGE of reaction products. Quantification of the free hTR(10–43) by densitometry revealed that inclusion of either BC200 or BC(185–200) resulted in enhanced quadruplex unwinding as compared with 25P1 (Fig. 11, C and D). This was particularly apparent at limiting concentrations of RHAU (8 nm), whereupon, following a 60-min reaction time, ~60% of the quadruplex was converted to duplex in the case of 25P1, whereas greater than 80% conversion took place with BC200 and BC(185–200) (Fig. 11C). These results indicate that BC200 acts as an efficient binding partner and stabilizer of unwound quadruplex RNA in vitro.

An Interaction between BC200 and the Telomerase RNA Is Supported by Endogenous RNA Pull-down Assays

Because the in vitro data demonstrated an interaction between BC200 and hTR, we wished to test the biological relevance of this by assessing whether this interaction is observed with the endogenous RNAs. To perform endogenous RNA pull-down assays, biotinylated LNA oligonucleotides complementary to the respective RNAs were employed in a streptavidin pull-down assay. Evaluation of RNA quantity in the pre- and post-pull-down lysates by qPCR confirmed that ~50% of the endogenous RNAs were captured on the beads (Fig. 12A). To control for nonspecific interactions, pull-downs were performed with the magnetic beads alone and a scrambled LNA-biotin probe.

FIGURE 12.
Endogenous BC200 and hTR are co-enriched. A, HEK293T cell lysates were supplemented with no LNA (Beads), a scrambled LNA-biotin, or biotinylated LNAs targeting BC200 and hTR. RNA was quantified by real-time PCR prior to and following pull-down of endogenous ...

Quantification of the bead-bound RNA was performed by qPCR with gene-specific primers relative to 25 ng of input RNA. All samples were eluted in equal volumes, and 5% of pull-down was used as a template. As expected, the BC200 and hTR LNA probes enriched their direct target RNAs strongly (>1000-fold; data not shown). Pull-down of the BC200 RNA demonstrated ~80-fold enrichment of the hTR RNA (Fig. 12B), whereas pull-down of the hTR RNA demonstrated ~15-fold enrichment of BC200 (Fig. 12C). Both RNAs demonstrated co-enrichment that was substantially higher than that of the scrambled LNA probe, supportive of the possibility that an interaction between the RNAs exists in a cellular environment.

Analysis of the proteins bound to the pulled-down endogenous RNAs by Western blotting demonstrates specific binding of RHAU to both RNAs, providing a further confirmation of the RHAU-BC200 interaction (Fig. 12D).

Discussion

In recent years, G4-quadruplexes have emerged from being an obscure structure formed by guanine-rich oligonucleotides in vitro to represent prevalent motifs found in regulatory regions throughout both the human genome and transcriptome (9, 11, 16, 50, 51). Although several helicases have been identified that unwind DNA quadruplexes, RHAU remains the primary helicase studied to act upon quadruplex RNA (52,54). In light of this, we have focused our interest on studying RHAU-binding RNAs to gain insight into the role of quadruplexes in RNA biology. We have previously reported the identification of several novel RHAU-binding RNAs through an RNA-RHAU co-immunoprecipitation screen (21). The BC200 RNA was identified in this screen to demonstrate an affinity for RHAU and thus became a focus of our interest. In this study, we have validated and examined the interaction between BC200 and RHAU and have begun to explore the biological implications of this binding event.

The BC200 RNA is a long non-coding RNA expressed in neuronal tissue and germ cells that is aberrantly expressed in a wide range of immortalized/tumor cell lines. The potential involvement of BC200 in several disease states as well as a lack of complete understanding of its function made it an attractive RNA to study. Upon confirming and validating an interaction between BC200 and RHAU, we set out to determine whether the BC200 RNA secondary structure contains any quadruplex elements. Analysis of the BC200 RNA sequence with the quadruplex-predicting QGRS Mapper software did not identify any region of the RNA prone to fold in this manner (55). This in silico prediction was confirmed by a number of studies on both the in vitro transcribed RNA as well as the endogenously expressed RNA. The in vitro transcribed RNA failed to stain with an established quadruplex-specific fluorescent dye. BC200 also demonstrated little to no affinity for the quadruplex interaction motif of RHAU (RHAU(53–105)) and deletion of the RSM did not impact binding in cell lysate. Furthermore, CD spectra of BC200 and six truncations of the RNA were not indicative of quadruplex formation. The BC200 spectrum demonstrates features consistent with A-form RNA with a dominant positive peak at 266 nm and negative peak at 210 nm. This is unlike the spectra typically observed for quadruplex RNA, which demonstrates a maximum and minimum at 264 and 239.5 nm, respectively, as well as hallmark positive ellipticity at 210 nm (56, 57). RHAU(53–105) binds a wide range of inter- and intramolecular DNA and RNA quadruplexes (17, 18, 21, 23, 58). In the case in which the in vitro transcribed RNA failed to fold in a similar manner as the endogenous RNA, we also employed RHAU(53–105) to determine whether the endogenously expressed BC200 contained any quadruplexes. In this test as well, BC200 failed to display evidence of folding into an RNA quadruplex. Combined with the in silico predictions, our data are strongly in support of the notion that the BC200 RNA does not contain a quadruplex structure.

Despite the lack of a quadruplex-forming region, BC200 demonstrates a direct in vitro affinity toward RHAU, albeit with relatively lower affinity than tested quadruplex RNAs. Although quadruplexes and BC200 require unique regions of RHAU to bind with full affinity, RHAU is unable to bind both quadruplex RNA and BC200 RNA simultaneously. This may be due to a shared secondary interaction site, possibly the helicase domains (DEXDc and HELICc).

After ruling out involvement of the RSM, further truncations of RHAU were made that support an interaction that is dependent upon the C terminus of the protein; however, the structural features that allow for BC200 interaction with RHAU remain to be determined and will be the subject of future study. Because loss of both the OB fold (RHAU(1–760)) and the HA domain (RHAU(1–614)) disrupts BC200 binding, the interaction may be dependent upon the presence of both domains, or proper folding in this region requires an intact C terminus. Interestingly, RHAU isoform 2, which lacks amino acids 517–530, did not bind BC200 in vivo but retained a similar affinity in vitro. This result would suggest that the interaction with BC200 in the cell is also governed by the subcellular localization of RHAU and/or the presence of additional protein and/or RNA binding partners. Regardless, a simple scenario whereby amino acids 517–530 are key to the binding event is unlikely in that RHAU(1–614) and RHAU(1–760) retain these amino acids but cannot bind to BC200. Whether this interaction is mediated through the HA2 domain and/or OB fold or an additional binding/specificity motif remains to be assessed. A further observation from these experiments was a significant increase in BC200 enrichment by the helicase-dead mutant of RHAU (RHAUE335A). This is suggestive that the helicase activity of RHAU results in displacement of bound BC200 and warrants further investigation.

We have narrowed the region of BC200 that is necessary and sufficient for RHAU interaction down to an adenosine-rich stretch of nucleotides within the last 80 nucleotides of the RNA. The BC200 RNA is predicted by sequence conservation among primates and sequence homology to the 7SL RNA of the signal recognition particle to adopt a secondary structure in which the 5′-Alu domain contains three stem-loops with the first two stem-loops forming a pseudoknot structure (31, 33). Whereas the first 120 nucleotides of BC200 are predicted to be well structured, the remaining 80 nucleotides containing the adenosine-rich and 3′-cytosine-rich regions are expected to remain predominantly single-stranded (31). Probing of the in vitro transcribed BC200 RNA with hybridizing DNA oligonucleotides and RNase H as well as single-strand-specific RNases generated results consistent with this prediction. Incomplete digestion of nucleotides 181–200 by RNase H may be explained by the propensity of the G-rich antisense oligonucleotide to form higher order structures, presumably intermolecular G-quadruplexes; however, a lack of digestion of residues in this region by RNase A supports the possibility that an RNase-resistant structure is formed near the 3′-end of the RNA. Although the RNase digestion data were in general agreement with the predicted RNA structure, the digestion pattern did not agree with the expected bulge pattern within the stem-loop comprising nucleotides 51–119. Higher resolution methods will be required to further clarify the structural details.

Because our in vitro transcribed RNA was generally consistent with the predicted secondary structure, we used the predicted structure to design a series of six truncations of the BC200 RNA. Circular dichroism analysis of the 5′-region of the RNA yielded spectra consistent with A-form RNA, whereas CD spectroscopy of the tail truncation yielded spectra consistent with previously calculated and experimental spectra for polyadenosine chains, which is not unexpected, considering that a significant portion (~50%) is composed of adenosine, with multiple runs of consecutive adenosines. This, as well as the red shifted local minimum (247 nm), indicate that the tail region is predominantly single-stranded and that the dichroic band observed in this region is due to “vertical” base-stacking interactions and is not consistent with duplex formation. Furthermore, being able to recapitulate the BC200 spectrum through the addition of the normalized BC119 and 122–200 as well as BC50 and BC(45–200) spectra indicates that our truncations have not significantly altered the structure of the RNA (44).

The various truncated BC200 RNAs containing the predicted secondary structure elements between their respective transition points allowed us to narrow the site of interaction down to the likely single-stranded adenosine-rich stretch (nucleotides 122–200). This was supported by streptavidin pull-down assays with the endogenously expressed protein as well as EMSAs of the truncated RNAs with recombinant RHAU and RNA footprinting analysis.

RHAU has been shown to specifically bind several non-quadruplex RNAs (1, 25); however, this is the first time it has been shown to have affinity for adenosine-rich single-stranded RNA. Whether this affinity is specific to the adenosine-rich stretch of BC200 or RHAU binds a broad spectrum of RNAs containing similar repetitive elements remains to be seen. Interestingly, although BC(122–200) demonstrated full affinity for RHAU in EMSAs, the streptavidin pull-down assays revealed reduced RHAU binding compared with BC200 and BC185. Therefore, it is possible that the interaction with BC200 is enhanced in the context of the full-length RNA through contacts with other regions of the RNA or perhaps additional BC200 and/or RHAU protein binding partners.

Having validated the interaction between BC200 and RHAU, we sought to determine the functional consequences of this binding event. Because RHAU has previously been shown to destabilize RNAs through exosome recruitment (1), we tested whether RHAU expression had any implications for the steady state levels of BC200. Finding that RHAU did not impact BC200 expression to a measurable degree, we next wished to determine whether BC200 impeded RHAU quadruplex helicase activity. Testing this hypothesis in an RHAU-dependent quadruplex unwinding assay demonstrated that the cytosine-rich 3′-end of BC200 is capable of binding unfolded quadruplex. Not only did BC200 bind the unwound hTR quadruplex; it did so in a more efficient manner than the expected natural base-pairing region of hTR (25P1). This is despite the fact that 25P1 is predicted to have 19 base pairs with hTR(10–43) (16 Watson-Crick, 14 of which are G-C and three of which are G-U wobble base pairs), whereas the 12 consecutive cytosines of BC200 can expect to have a maximum of 12 G-C base-pairs and would require bulging of the interspersed non-guanine nucleotides. This may be due to the fact that RHAU has affinity for both hTR and BC200 but does not appreciably interact with the 25P1 RNA. To test this, we employed BC(185–200), which exhibits substantially decreased affinity for RHAU as compared with BC200. In this test, BC(185–200) was as effective in the helicase assays as BC200; therefore, a simple explanation based on the affinity of RHAU for both partners may not be accurate. It should be noted, however, that BC(185–200) retains some affinity for RHAU (unlike 25P1), and the CD spectra indicate it to be structurally distinct in isolation as compared with in the context of the larger RNA (data not shown).

The inability of BC185 and the ability of BC(185–200) to act as a binding partner to unfolded hTR quadruplex confirm the notion that the cytosine-rich stretch near the 3′-end of the RNA is critical for unwound quadruplex stabilization. Furthermore, the ability of BC(45–200) to interact with similar efficiency as BC200 rules out involvement of the 5′-end of the RNA. Interestingly, although BC(122–200) was able to bind the unwound hTR, it did so to a lesser extent than BC200 or BC(45–200), despite containing the intact cytosine-rich region and maintaining full affinity for RHAU. This raises a possibility that the folding of BC(122–200) is not entirely representative of the arrangement of this region within the context of the full-length RNA.

The RHAU and ATP-dependent interaction between BC200 and hTR(10–43) raises the possibility that BC200 stabilizes unwound quadruplex RNA during RNA remodeling events and/or that BC200 may exert translational regulatory properties at mRNAs bearing guanine-rich stretches of nucleotides. In the former case, BC200 may be acting as an RNA co-factor to RHAU, because we have demonstrated that, in the absence of a complementary oligonucleotide, intramolecular quadruplexes rapidly refold despite the presence of RHAU and ATP (21, 23). The latter case is intriguing because a previous study has postulated that BC200 and its rodent analog BC1 act to specify mRNA targeting of RNA-binding proteins. This study was performed with the Fragile X mental retardation protein, a published BC200-interacting protein that, like RHAU, has been shown to bind quadruplex RNA (59,63). If this is the case, it can be predicted that BC200 exerts regulatory functions at both U-rich and G-rich stretches of RNA. In the case of G-rich binding, the presence of a quadruplex helicase would be an expected requirement for BC200 to efficiently base-pair at such sites. Unfortunately, at this time, not a single known mRNA target of BC200 has been validated. We have, however, demonstrated data in vitro and with endogenous RNAs supporting a putative interaction between BC200 and hTR, a result warranting further investigation. These results provide evidence to support the hypothesis that BC200 can act as a binding partner to unwound quadruplex RNA. As such, we are currently investigating the in-cell RNA/DNA targets of BC200 to test this hypothesis on other non-coding RNA/mRNA candidates. Further study in this area will expand both our knowledge of the role of BC200 in regulation of gene expression and our understanding of a role for BC200 in relation to quadruplex RNA, the quadruplex helicase RHAU, and the telomerase RNA.

Author Contributions

E. P. B. performed the majority of experimental work as well as data analysis and authored the manuscript. E. K. S. M. performed circular dichroism experimentation and data analysis. R. H. assisted in in vitro RNA transcription and preliminary interaction studies. S. R. D., E. O. A., and E. D. assisted in RNA transcription and purification. M. M. assisted with recombinant protein expression and purification. J. S. and S. A. M. provided funding, coordinated the project, and participated in authoring of the manuscript. All authors reviewed the results and approved the final version of the manuscript.

Acknowledgments

We thank Dr. Yoshikuni Nagamine for the monoclonal anti-RHAU 12F33 hybridoma. We also thank Drs. Thomas Klonisch, Spencer Gibson, and Peter Pelka for the provided cell lines.

*This work was supported by a grant from the Cancer Research Society. The authors declare that they have no conflicts of interest with the contents of this article.

8The abbreviations used are:

RHAU
RNA helicase associated with AU-rich element
RSM
RHAU-specific motif
OB
oligosaccharide binding
BCYRN1
brain cytoplasmic RNA 1
TBE
Tris-borate EDTA
hTR
human telomerase RNA
LNA
locked nucleic acid
IP
immunoprecipitation
qPCR
quantitative PCR
AMP-PNP
5′-adenylyl-β,γ-imidodiphosphate.

References

1. Tran H., Schilling M., Wirbelauer C., Hess D., and Nagamine Y. (2004) Facilitation of mRNA deadenylation and decay by the exosome-bound, DExH protein RHAU. Mol. Cell 13, 101–111 [PubMed]
2. Vaughn J. P., Creacy S. D., Routh E. D., Joyner-Butt C., Jenkins G. S., Pauli S., Nagamine Y., and Akman S. A. (2005) The DEXH protein product of the DHX36 gene is the major source of tetramolecular quadruplex G4-DNA resolving activity in HeLa cell lysates. J. Biol. Chem. 280, 38117–38120 [PubMed]
3. Creacy S. D., Routh E. D., Iwamoto F., Nagamine Y., Akman S. A., and Vaughn J. P. (2008) G4 resolvase 1 binds both DNA and RNA tetramolecular quadruplex with high affinity and is the major source of tetramolecular quadruplex G4-DNA and G4-RNA resolving activity in HeLa cell lysates. J. Biol. Chem. 283, 34626–34634 [PMC free article] [PubMed]
4. Chen M. C., Murat P., Abecassis K., Ferré-D'Amaré A. R., and Balasubramanian S. (2015) Insights into the mechanism of a G-quadruplex-unwinding DEAH-box helicase. Nucleic Acids Res. 43, 2223–2231 [PMC free article] [PubMed]
5. Burge S., Parkinson G. N., Hazel P., Todd A. K., and Neidle S. (2006) Quadruplex DNA: sequence, topology and structure. Nucleic Acids Res. 34, 5402–5415 [PMC free article] [PubMed]
6. Simonsson T. (2001) G-quadruplex DNA structures: variations on a theme. Biol. Chem. 382, 621–628 [PubMed]
7. Beaudoin J. D., Jodoin R., and Perreault J. P. (2014) New scoring system to identify RNA G-quadruplex folding. Nucleic Acids Res. 42, 1209–1223 [PMC free article] [PubMed]
8. Jodoin R., Bauer L., Garant J. M., Mahdi Laaref A., Phaneuf F., and Perreault J. P. (2014) The folding of 5′-UTR human G-quadruplexes possessing a long central loop. RNA 20, 1129–1141 [PubMed]
9. Millevoi S., Moine H., and Vagner S. (2012) G-quadruplexes in RNA biology. Wiley Interdiscip. Rev. RNA 3, 495–507 [PubMed]
10. Beaudoin J. D., and Perreault J. P. (2010) 5′-UTR G-quadruplex structures acting as translational repressors. Nucleic Acids Res. 38, 7022–7036 [PMC free article] [PubMed]
11. Lam E. Y., Beraldi D., Tannahill D., and Balasubramanian S. (2013) G-quadruplex structures are stable and detectable in human genomic DNA. Nat. Commun. 4, 1796. [PMC free article] [PubMed]
12. Bugaut A., and Balasubramanian S. (2012) 5′-UTR RNA G-quadruplexes: translation regulation and targeting. Nucleic Acids Res. 40, 4727–4741 [PMC free article] [PubMed]
13. Raiber E. A., Kranaster R., Lam E., Nikan M., and Balasubramanian S. (2012) A non-canonical DNA structure is a binding motif for the transcription factor SP1 in vitro. Nucleic Acids Res. 40, 1499–1508 [PMC free article] [PubMed]
14. Shahid R., Bugaut A., and Balasubramanian S. (2010) The BCL-2 5′ untranslated region contains an RNA G-quadruplex-forming motif that modulates protein expression. Biochemistry 49, 8300–8306 [PMC free article] [PubMed]
15. Bugaut A., Rodriguez R., Kumari S., Hsu S. T., and Balasubramanian S. (2010) Small molecule-mediated inhibition of translation by targeting a native RNA G-quadruplex. Org. Biomol. Chem. 8, 2771–2776 [PMC free article] [PubMed]
16. Huppert J. L., and Balasubramanian S. (2005) Prevalence of quadruplexes in the human genome. Nucleic Acids Res. 33, 2908–2916 [PMC free article] [PubMed]
17. Lattmann S., Giri B., Vaughn J. P., Akman S. A., and Nagamine Y. (2010) Role of the amino terminal RHAU-specific motif in the recognition and resolution of guanine quadruplex-RNA by the DEAH-box RNA helicase RHAU. Nucleic Acids Res. 38, 6219–6233 [PMC free article] [PubMed]
18. Meier M., Patel T. R., Booy E. P., Marushchak O., Okun N., Deo S., Howard R., McEleney K., Harding S. E., Stetefeld J., and McKenna S. A. (2013) Binding of G-quadruplexes to the N-terminal recognition domain of the RNA helicase associated with AU-rich element (RHAU). J. Biol. Chem. 288, 35014–35027 [PMC free article] [PubMed]
19. Iwamoto F., Stadler M., Chalupníková K., Oakeley E., and Nagamine Y. (2008) Transcription-dependent nucleolar cap localization and possible nuclear function of DExH RNA helicase RHAU. Exp. Cell Res. 314, 1378–1391 [PubMed]
20. Huang W., Smaldino P. J., Zhang Q., Miller L. D., Cao P., Stadelman K., Wan M., Giri B., Lei M., Nagamine Y., Vaughn J. P., Akman S. A., and Sui G. (2012) Yin Yang 1 contains G-quadruplex structures in its promoter and 5′-UTR and its expression is modulated by G4 resolvase 1. Nucleic Acids Res. 40, 1033–1049 [PMC free article] [PubMed]
21. Booy E. P., Howard R., Marushchak O., Ariyo E. O., Meier M., Novakowski S. K., Deo S. R., Dzananovic E., Stetefeld J., and McKenna S. A. (2014) The RNA helicase RHAU (DHX36) suppresses expression of the transcription factor PITX1. Nucleic Acids Res. 42, 3346–3361 [PMC free article] [PubMed]
22. Lattmann S., Stadler M. B., Vaughn J. P., Akman S. A., and Nagamine Y. (2011) The DEAH-box RNA helicase RHAU binds an intramolecular RNA G-quadruplex in TERC and associates with telomerase holoenzyme. Nucleic Acids Res. 39, 9390–9404 [PMC free article] [PubMed]
23. Booy E. P., Meier M., Okun N., Novakowski S. K., Xiong S., Stetefeld J., and McKenna S. A. (2012) The RNA helicase RHAU (DHX36) unwinds a G4-quadruplex in human telomerase RNA and promotes the formation of the P1 helix template boundary. Nucleic Acids Res. 40, 4110–4124 [PMC free article] [PubMed]
24. Sexton A. N., and Collins K. (2011) The 5′ guanosine tracts of human telomerase RNA are recognized by the G-quadruplex binding domain of the RNA helicase DHX36 and function to increase RNA accumulation. Mol. Cell. Biol. 31, 736–743 [PMC free article] [PubMed]
25. Bicker S., Khudayberdiev S., Weiss K., Zocher K., Baumeister S., and Schratt G. (2013) The DEAH-box helicase DHX36 mediates dendritic localization of the neuronal precursor-microRNA-134. Genes Dev. 27, 991–996 [PubMed]
26. Yoo J. S., Takahasi K., Ng C. S., Ouda R., Onomoto K., Yoneyama M., Lai J. C., Lattmann S., Nagamine Y., Matsui T., Iwabuchi K., Kato H., and Fujita T. (2014) DHX36 enhances RIG-I signaling by facilitating PKR-mediated antiviral stress granule formation. PLoS Pathog. 10, e1004012. [PMC free article] [PubMed]
27. Lai J. C., Ponti S., Pan D., Kohler H., Skoda R. C., Matthias P., and Nagamine Y. (2012) The DEAH-box helicase RHAU is an essential gene and critical for mouse hematopoiesis. Blood 119, 4291–4300 [PubMed]
28. Gao X., Ma W., Nie J., Zhang C., Zhang J., Yao G., Han J., Xu J., Hu B., Du Y., Shi Q., Yang Z., Huang X., and Zhang Y. (2015) A G-quadruplex DNA structure resolvase, RHAU, is essential for spermatogonia differentiation. Cell Death Dis. 6, e1610. [PMC free article] [PubMed]
29. Kim T., Pazhoor S., Bao M., Zhang Z., Hanabuchi S., Facchinetti V., Bover L., Plumas J., Chaperot L., Qin J., and Liu Y. J. (2010) Aspartate-glutamate-alanine-histidine box motif (DEAH)/RNA helicase A helicases sense microbial DNA in human plasmacytoid dendritic cells. Proc. Natl. Acad. Sci. U.S.A. 107, 15181–15186 [PubMed]
30. Watson J. B., and Sutcliffe J. G. (1987) Primate brain-specific cytoplasmic transcript of the Alu repeat family. Mol. Cell. Biol. 7, 3324–3327 [PMC free article] [PubMed]
31. Skryabin B. V., Kremerskothen J., Vassilacopoulou D., Disotell T. R., Kapitonov V. V., Jurka J., and Brosius J. (1998) The BC200 RNA gene and its neural expression are conserved in Anthropoidea (primates). J. Mol. Evol. 47, 677–685 [PubMed]
32. Tiedge H., Chen W., and Brosius J. (1993) Primary structure, neural-specific expression, and dendritic location of human BC200 RNA. J. Neurosci. 13, 2382–2390 [PubMed]
33. Labuda D., and Zietkiewicz E. (1994) Evolution of secondary structure in the family of 7SL-like RNAs. J. Mol. Evol. 39, 506–518 [PubMed]
34. Kondrashov A. V., Kiefmann M., Ebnet K., Khanam T., Muddashetty R. S., and Brosius J. (2005) Inhibitory effect of naked neural BC1 RNA or BC200 RNA on eukaryotic in vitro translation systems is reversed by poly(A)-binding protein (PABP). J. Mol. Biol. 353, 88–103 [PubMed]
35. Chen W., Böcker W., Brosius J., and Tiedge H. (1997) Expression of neural BC200 RNA in human tumours. J. Pathol. 183, 345–351 [PubMed]
36. Iacoangeli A., Lin Y., Morley E. J., Muslimov I. A., Bianchi R., Reilly J., Weedon J., Diallo R., Böcker W., and Tiedge H. (2004) BC200 RNA in invasive and preinvasive breast cancer. Carcinogenesis 25, 2125–2133 [PubMed]
37. Perez D. S., Hoage T. R., Pritchett J. R., Ducharme-Smith A. L., Halling M. L., Ganapathiraju S. C., Streng P. S., and Smith D. I. (2008) Long, abundantly expressed non-coding transcripts are altered in cancer. Hum. Mol. Genet. 17, 642–655 [PubMed]
38. Mus E., Hof P. R., and Tiedge H. (2007) Dendritic BC200 RNA in aging and in Alzheimer's disease. Proc. Natl. Acad. Sci. U.S.A. 104, 10679–10684 [PubMed]
39. Lukiw W. J., Handley P., Wong L., and Crapper McLachlan D. R. (1992) BC200 RNA in normal human neocortex, non-Alzheimer dementia (NAD), and senile dementia of the Alzheimer type (AD). Neurochem. Res. 17, 591–597 [PubMed]
40. Wu P., Zuo X., Deng H., Liu X., Liu L., and Ji A. (2013) Roles of long noncoding RNAs in brain development, functional diversification and neurodegenerative diseases. Brain Res. Bull. 97, 69–80 [PubMed]
41. Booy E. P., McRae E. K., and McKenna S. A. (2015) Biochemical characterization of G4 quadruplex telomerase RNA unwinding by the RNA helicase RHAU. Methods Mol. Biol. 1259, 125–135 [PubMed]
42. Booy E. P., Meng H., and McKenna S. A. (2012) Native RNA purification by gel filtration chromatography. Methods Mol. Biol. 941, 69–81 [PubMed]
43. Arthanari H., Basu S., Kawano T. L., and Bolton P. H. (1998) Fluorescent dyes specific for quadruplex DNA. Nucleic Acids Res. 26, 3724–3728 [PMC free article] [PubMed]
44. Olsthoorn C. S., Bostelaar L. J., De Rooij J. F., Van Boom J. H., and Altona C. (1981) Circular dichroism study of stacking properties of oligodeoxyadenylates and polydeoxyadenylate: a three-state conformational model. Eur. J. Biochem. 115, 309–321 [PubMed]
45. Theimer C. A., and Feigon J. (2006) Structure and function of telomerase RNA. Curr. Opin. Struct. Biol. 16, 307–318 [PubMed]
46. Li X., Nishizuka H., Tsutsumi K., Imai Y., Kurihara Y., and Uesugi S. (2005) Effects on telomerase activity of the 5′-terminal region of human telomerase RNA. Nucleic Acids Symp. Ser. 49, 313–314 [PubMed]
47. Chen J. L., and Greider C. W. (2003) Template boundary definition in mammalian telomerase. Genes Dev. 17, 2747–2752 [PubMed]
48. Chen J. L., Blasco M. A., and Greider C. W. (2000) Secondary structure of vertebrate telomerase RNA. Cell 100, 503–514 [PubMed]
49. Gros J., Guédin A., Mergny J. L., and Lacroix L. (2008) G-quadruplex formation interferes with P1 helix formation in the RNA component of telomerase hTERC. Chembiochem 9, 2075–2079 [PubMed]
50. Biffi G., Tannahill D., McCafferty J., and Balasubramanian S. (2013) Quantitative visualization of DNA G-quadruplex structures in human cells. Nat. Chem. 5, 182–186 [PMC free article] [PubMed]
51. Bochman M. L., Paeschke K., and Zakian V. A. (2012) DNA secondary structures: stability and function of G-quadruplex structures. Nat. Rev. Genet. 13, 770–780 [PMC free article] [PubMed]
52. Castillo Bosch P., Segura-Bayona S., Koole W., van Heteren J. T., Dewar J. M., Tijsterman M., and Knipscheer P. (2014) FANCJ promotes DNA synthesis through G-quadruplex structures. EMBO J. 33, 2521–2533 [PMC free article] [PubMed]
53. Gray L. T., Vallur A. C., Eddy J., and Maizels N. (2014) G quadruplexes are genomewide targets of transcriptional helicases XPB and XPD. Nat. Chem. Biol. 10, 313–318 [PMC free article] [PubMed]
54. Paeschke K., Bochman M. L., Garcia P. D., Cejka P., Friedman K. L., Kowalczykowski S. C., and Zakian V. A. (2013) Pif1 family helicases suppress genome instability at G-quadruplex motifs. Nature 497, 458–462 [PMC free article] [PubMed]
55. Kikin O., D'Antonio L., and Bagga P. S. (2006) QGRS Mapper: a web-based server for predicting G-quadruplexes in nucleotide sequences. Nucleic Acids Res. 34, W676–W682 [PMC free article] [PubMed]
56. Kypr J., Kejnovská I., Renciuk D., and Vorlícková M. (2009) Circular dichroism and conformational polymorphism of DNA. Nucleic Acids Res. 37, 1713–1725 [PMC free article] [PubMed]
57. Vorlíčková M., Kejnovská I., Sagi J., Renčiuk D., Bednářová K., Motlová J., and Kypr J. (2012) Circular dichroism and guanine quadruplexes. Methods 57, 64–75 [PubMed]
58. Ariyo E. O., Booy E. P., Patel T. R., Dzananovic E., McRae E. K., Meier M., McEleney K., Stetefeld J., and McKenna S. A. (2015) Biophysical characterization of G-quadruplex recognition in the PITX1 mRNA by the specificity domain of the helicase RHAU. PLoS One 10, e0144510. [PMC free article] [PubMed]
59. Bole M., Menon L., and Mihailescu M. R. (2008) Fragile X mental retardation protein recognition of G quadruplex structure per se is sufficient for high affinity binding to RNA. Mol. Biosyst. 4, 1212–1219 [PubMed]
60. Blice-Baum A. C., and Mihailescu M. R. (2014) Biophysical characterization of G-quadruplex forming FMR1 mRNA and of its interactions with different fragile X mental retardation protein isoforms. RNA 20, 103–114 [PubMed]
61. Suhl J. A., Chopra P., Anderson B. R., Bassell G. J., and Warren S. T. (2014) Analysis of FMRP mRNA target datasets reveals highly associated mRNAs mediated by G-quadruplex structures formed via clustered WGGA sequences. Hum. Mol. Genet. 23, 5479–5491 [PMC free article] [PubMed]
62. Zhang Y., Gaetano C. M., Williams K. R., Bassell G. J., and Mihailescu M. R. (2014) FMRP interacts with G-quadruplex structures in the 3′-UTR of its dendritic target Shank1 mRNA. RNA. Biol. 11, 1364–1374 [PMC free article] [PubMed]
63. Zalfa F., Adinolfi S., Napoli I., Kühn-Hölsken E., Urlaub H., Achsel T., Pastore A., and Bagni C. (2005) Fragile X mental retardation protein (FMRP) binds specifically to the brain cytoplasmic RNAs BC1/BC200 via a novel RNA-binding motif. J. Biol. Chem. 280, 33403–33410 [PubMed]

Articles from The Journal of Biological Chemistry are provided here courtesy of American Society for Biochemistry and Molecular Biology