Desthiobiotin labeling of a synthetic 25-nucleotide RNA oligo, which contains an adenine-rich element (ARE) motif, allows specific binding of cytosolic ARE-binding protein.
The in vitro RNA-pulldown is still largely used in the first steps of protocols aimed at identifying RNA-binding proteins that recognize specific RNA structures and motifs. In this RNA-pulldown protocol, commercially synthesized RNA probes are labeled with a modified form of biotin, desthiobiotin, at the 3′ terminus of the RNA strand, which reversibly binds to streptavidin and thus allows elution of proteins under more physiological conditions. The RNA-desthiobiotin is immobilized through interaction with streptavidin on magnetic beads, which are used to pull down proteins that specifically interact with the RNA of interest. Non-denatured and active proteins from the cytosolic fraction of mesothelioma cells are used as the source of proteins. The method described here can be applied to detect the interaction between known RNA binding proteins and a 25-nucleotide (nt) long RNA probe containing a sequence of interest. This is useful to complete the functional characterization of stabilizing or destabilizing elements present in RNA molecules achieved using a reporter vector assay.
Gene expression and the final level of the gene product can be tightly regulated by affecting the mRNA stability and mRNA translation rate1. These post-transcriptional regulatory mechanisms are exerted through the interactions of non-coding RNA and/or RNA-binding proteins (RBPs) with targeted mRNA. It is usually the 3' untranslated region of mRNA (3' UTR – belonging to the non-coding portion of the genome2) that contains specific cis-regulatory elements (CRE), which are recognized by trans-acting factors such as miRNA or RBPs3. The best-studied cis-element within the 3' UTR, is the adenine-rich element (ARE) motif, which is recognized by specific AU-binding proteins (AUBP), and, in turn, induces either mRNA degradation/deadenylation (ARE-mediated decay) or mRNA stabilization4.
The size of the 3' UTR of calretinin mRNA (CALB2) is 573 bp long and contains a putative AUUUA pentamer, as predicted by AREsite2, a bioinformatic tool5. Consistent with the presence of a putative ARE motif, the pmirGLO vector-reporter assay demonstrated a stabilization role of this element within CALB2 mRNA6. Finally, the in vitro RNA-pulldown was used to identify the AUBP that stabilizes calretinin mRNA through the ARE motif.
Since all non-coding RNAs interact with proteins7, the in vitro RNA-pulldown is a good way and first-of-choice assay for identifying RNA-interactors to aid in deciphering molecular mechanisms. In this RNA-pulldown method, commercially synthesized RNA probes, which were labeled with a modified form of biotin (desthiobiotin) at the 3' terminus of the RNA strand, were used. The RNA-desthiobiotin is immobilized through interaction with streptavidin on magnetic beads, which are used to pull down proteins that specifically interact with the bound-RNA of interest. Non-denatured and active proteins from the cytosolic fraction of mesothelioma cells are used as the source of proteins. Such RNA-bound proteins are eluted from the magnetic beads, run through a 12% SDS-PAGE gel, transferred to a membrane, and probed with different antibodies.
In the standard streptavidin-biotin affinity purification procedure, harsh denaturation conditions are required to disrupt the strong irreversible biotin-streptavidin bond to elute the bound proteins8, which could lead to the dissociation of protein complexes. Unlike biotin, desthiobiotin reversibly binds to streptavidin and is competitively displaced with a buffered solution of biotin, allowing for the gentle elution of proteins, and avoiding the isolation of naturally biotinylated molecules9, suggesting that the technique is ideal for isolating native protein complexes under native conditions.
In vitro binding conditions and stringency, which are determined by salt concentration, reducing agents and detergent percentage, should be close to those present in the cellular context in order to identify true in vivo interactions. The binding conditions implemented herein have been previously demonstrated as appropriate for the identification of the HuR as an AU-binding protein10. This approach could save time, since optimization of proper binding conditions can be time-consuming and challenging. In addition, this method could be used as a starting protocol for any RNA-pulldown experiment and can be gradually optimized by changing the concentration of salts and detergents, changing the glycerol percentage, and adding other salts. Moreover, we demonstrated that even a short 25-nucleotide RNA-probe harboring a pentamer ARE-motif could be used to demonstrate interaction with a specific AUBP.
1. Preparation of Cytosolic and Nuclear Protein Fraction
2. Labeling of RNA with Desthiobiotin
3. RNA-Protein Pulldown
4. Polyacrylamide Electrophoresis and Western Blot Analysis
NOTE: See Table 2 for recipes for the buffers used in this section. Perform a Western blot according to the Laemmli method12.
In this experiment, a 25-nt long fragment of calretinin 3' UTR harboring ARE motif (CALB2 3' UTR (ARE) 25-nt) was used to test whether it binds specifically to the Human-antigen R (HuR) protein, a known mRNA stabilizer. To test the specificity of the ARE element, a 25-nt RNA probe CALB2 3' UTR (mtARE) containing an ARE-motif mutation, which was previously shown to abolish the stabilization effect of the ARE motif, was used6. The third RNA probe represents the negative control, which is a 28-nt unrelated RNA that contains the well-defined iron-responsive element (Unrelated RNA (IRE)), known to bind cytosolic iron-responsive protein13,14. Since HuR is predominately localized within the nucleus but functions as a mRNA-stabilizer in the cytosol15, nuclear/cytosolic extraction was performed to obtain active proteins from the cytosol.
To demonstrate the purity of the nuclear/cytosolic fractions, proteins were analyzed by Western blot, which showed that α-tubulin was only detected in the cytosolic fraction, whereas PARP protein was detected only in the nuclear fraction, as anticipated (Figure 1). The eluate from the CALB2 3' UTR ARE-probe demonstrated binding of HuR whereas HuR was absent in the eluate from the mutant probe CALB 3' UTR (mtARE). HuR was also absent in the eluate from the unrelated RNA probe which binds the iron-responsive protein (Figure 2A). To further demonstrate the specificity between the CALB 3' UTR (ARE) and HuR, the membrane was additionally probed with anti-mesothelin (MSLN) antibody, as this protein does not interact with RNA. Eluates from all three samples showed no presence of mesothelin. Taken together, this indicates that the stabilizing ARE motif within CALB2 3'UTR can specifically bind HuR protein.
Coomassie staining of the gel (Figure 2B) shows that equal amounts of proteins were used to incubate with the three different RNA probes. Because of the low amount of cytosolic extract used and transfer to the membrane, this staining did not allow for detection of proteins in the eluate (E) lanes, which were otherwise detected by Western blot analysis.
Figure 1: Western blot analysis of 5 µg of cytosolic and nuclear protein fraction. PARP protein is present in the nuclear fraction (N) but absent from the cytosolic (C) fraction. α-tubulin is present in the cytosolic fraction (C) but absent in the nuclear fraction (N). Please click here to view a larger version of this figure.
Figure 2: Western blot shows that HuR is captured by 25-nt CALB2 3' UTR (ARE). A) FT: flow through after incubation with RNA probe; E: proteins eluted upon incubation and binding to RNA probes. Probes: CALB2 3' UTR (mtARE) – 25-nt fragment of calretinin 3' UTR that contains 3 mutated nucleotides within the ARE motif; UR RNA (IRE) – 28-nt RNA probe with a well-defined iron-responsive element that binds the iron-responsive proteins. The membrane was further probed for mesothelin (MSLN), a protein that does not interact with RNA. B) Coomassie staining of the gel after protein transfer, demonstrating that equal amounts of proteins were incubated with RNA probes. Please click here to view a larger version of this figure.
RNA probe | Sequence | Concentration |
CALB2 3’ UTR (ARE) 25nt | UCGCUGUAUGAUUUAGGCUUCUAUG | 10 µM |
CALB2 3’ UTR (mtARE) 25nt | UCGCUGUAUGGUCUGGGCUUCUAUG | 10 µM |
Unrelated RNA – IRE 28nt | UCCUGCUUCAACAGUGCUUGGACGGAAC | 10 µM |
Table 1: Sequences of RNA probes
Lower Tris buffer for running gel | |
1.5 M | Tris Base |
0.40% | SDS solution |
addjust pH to 8.8 using HCL (6 mol/L) | |
fill up with | dH2O |
Upper Tris for stacking gel SDS-PAGE | |
500 mM | Tris base |
0.4% | SDS solution |
Adjust pH to 6.8 using HCL (6 mol/L) | |
fill up with | dH2O |
10X Running buffer | |
250 mM | Tris base |
1.92 M | Glycine |
Adjust pH to 8.3 using HCL (6 mol/L) | |
fill up with | dH2O |
Filter or autoclave and store at 4°C | |
1X Running buffer | |
100 mL | 10X Running buffer |
0.05% | SDS solution |
Fill up to 1 L with | dH2O |
10X Transfer Buffer | |
480 mM | Tris base |
390 mM | Glycine |
0.38% | SDS solution |
fill up with | dH2O |
Note: Filter 0.22 um and Store at 4 °C | |
1X Transfer buffer (for semi-dry system) | |
20 mL | 10X Transfer Buffer |
40 mL | methanol |
fill up to 200 mL | dH20 |
10X TBS (Tris-buffered saline) | |
250 mM | Tris base |
1.36 M | NaCl |
27 mM | KCl |
Adjust pH to 7.4 | |
fill up with | dH2O |
Filter | |
1X TTBS (Tris-buffer saline with 0.1 Tween-20) | |
1X | 10X TBS |
0.10% | Tween-20 |
fill up with | dH2O |
RNA Capture Buffer (1X) | |
20 mM | Tris (pH 7.5) |
1 M | NaCl |
1 mM | EDTA |
Protein-RNA Binding Buffer (10x) | |
200 mM | Tris (pH 7.5) |
500 mM | NaCl |
20 mM | MgCl2 |
1% | Tween-20 |
Wash Buffer (1x) | |
20 mM | Tris (pH 7.5) |
10 mM | NaCl |
0.1% | Tween-20 |
Elution Buffer | |
4 mM | biotin |
20 mM | Tris (pH 7.5) |
50 mM | NaCl |
Coomassie solution | |
0.02% | Coomassie G-250 |
5% | Aluminiumsulfate-(14-18)-hydrate |
10% | EtOH |
2% | ortho-phosphoric acid |
Table 2: Recipes
3' UTRs belong to the non-coding genome3, and all non-coding RNAs can interact with proteins in order to exert their function7. When the mammalian genome was found to be pervasively transcribed and produced a significant portion of long noncoding RNAs16, emerging evidences demonstrated that these long-noncoding RNAs function in regulating gene expression as they interact with chromatin-remodeling complexes17. This knowledge was gained by utilizing the RNA-pulldown assay. Thus, to start deciphering interactors of a specific RNA, the first step could be in vitro biochemical assays such as RNA-pulldown, as it is simple to perform.
Of note, not only sequence motifs but also the secondary structure of RNA plays a critical role in RNA functionality as they form domains essential for interacting with proteins. The secondary structure can vary in physiological and in vitro conditions, and RNA-pulldown may lead to the identification of falsepositives. For example, RNA-pulldown identified EZH2, a component of the polycomb repressive complex 2 (PRC2), as an interactor of X inactive specific transcript (Xist)18, whereas a recent study that used RNA antisense purification coupled with mass spectrometry (RAP-MS) identified other interactors such as SHARP (SMRT and HDAC associated repressor protein), SAF-1 (scaffold attachment factor A) and LBR (lamin B receptor)19. Therefore, if no other functional tests support the interaction between an RNA and a given protein, findings should be further supplemented with a complementary protein-centric approach, such as immunoprecipitating the given protein, followed by extraction and characterization of the bound RNA.
The herein protocol presented has been used to detect HuR binding to the ARE motif in CALB2 3' UTR, which had previously been functionally characterized by pmiRGLO-luciferase assay6. Due to the small amount of proteins, this protocol is not suitable for downstream analysis such as mass-spectrometry where larger protein amounts are necessary. Instead, the protocol can be used to revalidate the RNA-interactors identified by mass spectrometry, but using a significantly lower amount of proteins.
Since the method includes working with RNA, which is susceptible to degradation, we recommend cleaning the working surfaces with a RNase-decontaminating solution in addition to standard laboratory good practice. If possible, perform preparation of RNA labeling and purification under laminar flow. The purity of the RNA oligo may also affect the binding conditions; thus, the commercially synthesized probes were HPLC purified. If modified, longer and extremely pure RNA oligos are required, use PAGE purification method. Prepare aliquots of native proteins by snap freezing, therefore avoiding damaging of proteins due to the slow freezing approach and repeated freezing and thawing.
The authors have nothing to disclose.
This work was supported by the Swiss National Science Foundation Sinergia grant CRSII3 147697, the Stiftung für Angewandte Krebsforschung and Zürich Krebsliga.
NE-PER Nuclear and cytoplasmic extraction kit | Thermo Fisher Scientific | 78833 | |
Pierce Protease Inhibitor Tablets, EDTA-free | Thermo Fisher Scientific | A32965 | |
Pierce BCA Protein Assay Kit | Thermo Fisher Scientific | 23225 | |
Pierce Magnetic RNA-protein Pull-Down Kit | Thermo Fisher Scientific | 20164 | Includes magnetic beads and therefore the kit need to be stored at 4 °C |
Pierce RNA 3’ End Desthiobiotinylation Kit | Thermo Fisher Scientific | 20163 | The kit is a part of the "Pierce Magnetic RNA-protein Pull-Down Kit", and needs to be stored at -20 °C unlike the pull-down kit (4 °C). |
DynaMag-2, magnetic stand | Thermo Fisher Scientific | 12321D | |
Anti – HuR (MOUSE) monoclonal antibody | Thermo Fisher Scientific | – | The antibody is included in the Pierce Magnetic RNA-protein Pull-Down Kit – 1:1000 in 5% BSA 1x TTBS – rabbit anti-mouse secondary antibody 1:10,000 in 5% BSA 1x TTBS |
Anti – α – tubulin (MOUSE) monoclonal antibody | Santa Cruz | 8035 | 1:1000 in 5% BSA 1x TTBS – rabbit anti-mouse secondary antibody 1:10,000 in 5% BSA 1x TTBS |
Anti – PARP (RABBIT) Polyclonal antibody | Cell Signaling | 9542 | 1:1000 in 5% BSA 1x TTBS – goat anti-rabbit secondary antibody 1:10,000 in 5% BSA 1x TTBS |
Anti – Mesothelin (MOUSE) Monoclonal Antibody | Rockland | 200-301-A87 | |
Secondary goat anti-rabbit antibody | Cell Signaling | 7074 | |
Secondary rabbit anti-mouse antibody | Sigma-Aldrich | A-9044 | |
RPMI – 1640 medium | Sigma-Aldrich | R8758 | |
FBS – Filtrated Bovine Serum | Pan Biotech | P40-37500 | |
Penicillin – streptomycin (100x) | Sigma-Aldrich | P4333 | |
L-Glutamine solution (200 mM) | Sigma-Aldrich | G7513 | |
0.25% Trypsin-EDTA (1x) | Gibco by Life technologies | 25200-056 | |
Mini-PROTEAN 3 Cell | Bio-Rad | 1653301 | |
Mini Protean System Glass Plates | Bio-Rad | 1653308 | |
Mini-PROTEAN Spacer Plates with 1.5 mm integrated spacer | Bio-Rad | 1653312 | |
Mini-PROTEAN Comb, 10-well, 1.5 mm, 66 µL | Bio-Rad | 1653365 | |
Mini-PROTEAN Tetra Cell Casting Modules | Bio-Rad | 1658050 | |
RNaseZap RNase Decontamination Solution | Thermo Fisher Scientific | AM9780 | |
Rotiphorese gel 30 – aqueous 30% acrylamide and bisacrylamide stock solution at a ration of 37.5:1 | Roth | 3029 | |
20% SDS | PanReac AppliChem | A3942 | |
PVDF transfer membrane | Perkin Elmer | NEF1002 | Need to be activated by incubating in pure methanol for 1 min followed by washing in water for 2 min |
Trans-Blot SD semi-dry transfer cell | Bio-Rad | 1703940 | |
Clarity Western ECL Substrate | Bio-Rad | 1705061 | |
FusionFX Digital Imager | Vilber | – | |
RNA oligo synthesis | Mycrosynth AG, Switzerland | – | the synthesis scale – 0.04 µmol – HPLC purified |
Bovine serum albumin | Sigma-Aldrich | A7030 | |
Hydrochloric Acid (HCl) 6 M | PanReac AppliChem | 182883.1211 | |
Ultra Pure 1 M Tris, pH 7.5 | Thermo Fisher Scientific | 15567027 | |
Glycerol | Thermo Fisher Scientific | 17904 | 50% sterile aliquots to be stored at -20°C |
Pierce Streptavidin magnetic beads | Thermo Fisher Scientific | 88817 | Please note that these magnetic beads have an average diameter of 1 µm (range from 0.5 – 1.5 µm). Furthermore, Dynabeads-M280 Streptavidin, which are beads of homogenouse size 2.8 µm, are also used in RNA-pulldown but be aware that this may affect purification process. |
TEMED | Sigma-Aldrich | T9281 | |
Ammonium persulfate | Sigma-Aldrich | A3678 | |
Coomassie G-250 | Applichem | A3480 | |
Aluminiumsulfate-(14-18)-hydrate | Sigma-Aldrich | 368458 | |
ortho-phosphoric acid | Applichem | A0637 |