Portal de Periódicos da CAPES

Enhanced immunoprecipitation techniques for the identification of RNA-binding protein partners: IGF2BP1 interactions in mammary epithelial cells

2022; Elsevier BV; Volume: 298; Issue: 3 Linguagem: Inglês

10.1016/j.jbc.2022.101649

1083-351X

Saja A. Fakhraldeen, Scott M. Berry, David J. Beebe, Avtar Roopra, Celia M. Bisbach, Vladimir S. Spiegelman, Natalie M. Niemi, Caroline M. Alexander,

Cancer-related molecular mechanisms research

RNA-binding proteins (RBPs) regulate the expression of large cohorts of RNA species to produce programmatic changes in cellular phenotypes. To describe the function of RBPs within a cell, it is key to identify their mRNA-binding partners. This is often done by crosslinking nucleic acids to RBPs, followed by chemical release of the nucleic acid fragments for analysis. However, this methodology is lengthy, which involves complex processing with attendant sample losses, thus large amounts of starting materials and prone to artifacts. To evaluate potential alternative technologies, we tested “exclusion-based” purification of immunoprecipitates (IFAST or SLIDE) and report here that these methods can efficiently, rapidly, and specifically isolate RBP–RNA complexes. The analysis requires less than 1% of the starting material required for techniques that include crosslinking. Depending on the antibody used, 50% to 100% starting protein can be retrieved, facilitating the assay of endogenous levels of RBPs; the isolated ribonucleoproteins are subsequently analyzed using standard techniques, to provide a comprehensive portrait of RBP complexes. Using exclusion-based techniques, we show that the mRNA-binding partners for RBP IGF2BP1 in cultured mammary epithelial cells are enriched in mRNAs important for detoxifying superoxides (specifically glutathione peroxidase [GPX]-1 and GPX-2) and mRNAs encoding mitochondrial proteins. We show that these interactions are functionally significant, as loss of function of IGF2BP1 leads to destabilization of GPX mRNAs and reduces mitochondrial membrane potential and oxygen consumption. We speculate that this underlies a consistent requirement for IGF2BP1 for the expression of clonogenic activity in vitro. RNA-binding proteins (RBPs) regulate the expression of large cohorts of RNA species to produce programmatic changes in cellular phenotypes. To describe the function of RBPs within a cell, it is key to identify their mRNA-binding partners. This is often done by crosslinking nucleic acids to RBPs, followed by chemical release of the nucleic acid fragments for analysis. However, this methodology is lengthy, which involves complex processing with attendant sample losses, thus large amounts of starting materials and prone to artifacts. To evaluate potential alternative technologies, we tested “exclusion-based” purification of immunoprecipitates (IFAST or SLIDE) and report here that these methods can efficiently, rapidly, and specifically isolate RBP–RNA complexes. The analysis requires less than 1% of the starting material required for techniques that include crosslinking. Depending on the antibody used, 50% to 100% starting protein can be retrieved, facilitating the assay of endogenous levels of RBPs; the isolated ribonucleoproteins are subsequently analyzed using standard techniques, to provide a comprehensive portrait of RBP complexes. Using exclusion-based techniques, we show that the mRNA-binding partners for RBP IGF2BP1 in cultured mammary epithelial cells are enriched in mRNAs important for detoxifying superoxides (specifically glutathione peroxidase [GPX]-1 and GPX-2) and mRNAs encoding mitochondrial proteins. We show that these interactions are functionally significant, as loss of function of IGF2BP1 leads to destabilization of GPX mRNAs and reduces mitochondrial membrane potential and oxygen consumption. We speculate that this underlies a consistent requirement for IGF2BP1 for the expression of clonogenic activity in vitro. RNA-binding proteins (RBPs) are critical post-transcriptional regulators of gene expression in normal and pathological cellular contexts (1Gerstberger S. Hafner M. Ascano M. Tuschl T. Evolutionary conservation and expression of human RNA-binding proteins and their role in human genetic disease.Adv. Exp. Med. Biol. 2014; 825: 1-55Google Scholar). At least 1542 RBPs govern RNA metabolism at myriad stages of splicing, export, storage, transport, and translation (2Gerstberger S. Hafner M. Tuschl T. A census of human RNA-binding proteins.Nat. Rev. Genet. 2014; 15: 829-845Google Scholar). Often, RBPs bind select RNA species to modulate their expression, localization, and/or stability, occasionally via highly specific and conserved sequence motifs. However, more typically, RBPs bind RNA species via short and degenerate sequences that are not easy to recognize prospectively (1Gerstberger S. Hafner M. Ascano M. Tuschl T. Evolutionary conservation and expression of human RNA-binding proteins and their role in human genetic disease.Adv. Exp. Med. Biol. 2014; 825: 1-55Google Scholar). Aberrant RBP activity is responsible for such important phenotypes as fragile X syndrome (via the RBP fragile X mental retardation protein) (3Darnell J.C. Klann E. The translation of translational control by FMRP: Therapeutic targets for FXS.Nat. Neurosci. 2013; 16: 1530-1536Google Scholar), and splicing reactions of cancer-associated tumor drivers, such as androgen receptor (via the RBP DEAD-box helicase 3 X-linked) (4Vellky J.E. McSweeney S.T. Ricke E.A. Ricke W.A. RNA-binding protein DDX3 mediates posttranscriptional regulation of androgen receptor: A mechanism of castration resistance.Proc. Natl. Acad. Sci. U. S. A. 2020; 117: 28092-28101Google Scholar). It is therefore important to define the cohort of mRNA-binding partners that are bound by each specific RBP, since these partners are likely to be affected by altered RBP expression or activity. The cohort of mRNA species bound by a given RBP can be highly cell type specific for reasons that are not yet understood. RBPs sometimes stabilize mRNA species; this is deduced from the demonstration of a direct binding interaction, together with decreased abundance upon RBP knockdown/knockout (1Gerstberger S. Hafner M. Ascano M. Tuschl T. Evolutionary conservation and expression of human RNA-binding proteins and their role in human genetic disease.Adv. Exp. Med. Biol. 2014; 825: 1-55Google Scholar, 5Hafner M. Landthaler M. Burger L. Khorshid M. Hausser J. Berninger P. Rothballer A. Ascano M. Jungkamp A.C. Munschauer M. Ulrich A. Wardle G.S. Dewell S. Zavolan M. Tuschl T. PAR-CliP--a method to identify transcriptome-wide the binding sites of RNA binding proteins.J. Vis. Exp. 2010; 2: 2034Google Scholar). However, other regulatory activities that do not result in altered RNA abundance are much more difficult to identify, for example, regulation of RNA localization and delivery of target proteins to specific subcellular structures (6Eom T. Antar L.N. Singer R.H. Bassell G.J. Localization of a beta-actin messenger ribonucleoprotein complex with zipcode-binding protein modulates the density of dendritic filopodia and filopodial synapses.J. Neurosci. 2003; 23: 10433-10444Google Scholar). Some studies have selected specific mRNAs with obvious roles to show that RBP–mRNA interactions regulate function; for example, Conway et al. (7Conway A.E. Van Nostrand E.L. Pratt G.A. Aigner S. Wilbert M.L. Sundararaman B. Freese P. Lambert N.J. Sathe S. Liang T.Y. Essex A. Landais S. Burge C.B. Jones D.L. Yeo G.W. Enhanced CLIP uncovers IMP protein-RNA targets in human pluripotent stem cells important for cell adhesion and survival.Cell Rep. 2016; 15: 666-679Google Scholar) demonstrated adhesion defects after disruption of an insulin-like growth factor 2 mRNA-binding protein 1 (IGF2BP1)–integrin subunit beta 5 mRNA in embryonic stem cells. Most studies rely upon UV-induced crosslinking coupled with immunoprecipitation techniques to define mRNA-binding partners for RBPs. By exploiting the unique chemical reactivity of RNA for protein, irreversible crosslinks can be formed between RNA and protein moieties that lie in close proximity (1Gerstberger S. Hafner M. Ascano M. Tuschl T. Evolutionary conservation and expression of human RNA-binding proteins and their role in human genetic disease.Adv. Exp. Med. Biol. 2014; 825: 1-55Google Scholar). This technique was widely adopted after concerns were raised about the potential for switching of RBP-binding partners during incubations (8Mili S. Steitz J.A. Evidence for reassociation of RNA-binding proteins after cell lysis: Implications for the interpretation of immunoprecipitation analyses.RNA. 2004; 10: 1692-1694Google Scholar). However, it is not trivial to reverse these crosslinks sufficiently to release and identify the bound mRNA species, and the yield of input RBP that emerges after the extensive processing reactions is substantially less than 1% of total (9Zeliadt N. Capturing complexes: Techniques for analyzing RNA-protein interactions.The Scientist. October 1, 2014; Google Scholar). Various versions of photoactivatable ribonucleoside–enhanced crosslinking and immunoprecipitation (PAR–CLIP) protocols have been described and applied to the analysis of RBPs; the strengths and weaknesses of each have been reviewed (5Hafner M. Landthaler M. Burger L. Khorshid M. Hausser J. Berninger P. Rothballer A. Ascano M. Jungkamp A.C. Munschauer M. Ulrich A. Wardle G.S. Dewell S. Zavolan M. Tuschl T. PAR-CliP--a method to identify transcriptome-wide the binding sites of RNA binding proteins.J. Vis. Exp. 2010; 2: 2034Google Scholar, 9Zeliadt N. Capturing complexes: Techniques for analyzing RNA-protein interactions.The Scientist. October 1, 2014; Google Scholar, 10Van Nostrand E.L. Huelga S.C. Yeo G.W. Experimental and computational considerations in the study of RNA-binding protein-RNA interactions.Adv. Exp. Med. Biol. 2016; 907: 1-28Google Scholar, 11Ascano M. Gerstberger S. Tuschl T. Multi-disciplinary methods to define RNA-protein interactions and regulatory networks.Curr. Opin. Genet. Dev. 2013; 23: 20-28Google Scholar, 12Konig J. Zarnack K. Luscombe N.M. Ule J. Protein-RNA interactions: New genomic technologies and perspectives.Nat. Rev. Genet. 2011; 13: 77-83Google Scholar). In general, the limitations of crosslinking protocols fall into various classes: the loss of unstable RNA species during long processing procedures, loss of mRNAs with indirect or low-affinity interactions during washing of immunoprecipitates, artifacts created by crosslinking and extensive derivatization processes, and the requirement for an impractically high starting numbers of labeled cultured cells overexpressing the RBP of interest. Indeed, in many key cell types, RBP interactions cannot be studied because of the requirement for up to 1 g of starting protein lysate. Here, we have applied two Exclusion-Based Sample Preparation (ESP) technologies to identify mRNA-binding partners for an exemplar RBP; the simple and expedited processing that is required for the techniques used in this article versus those required for PAR–CLIP is summarized in Figure 1. The focus of this study is the RBP IGF2BP1 (IGF2 mRNA-binding protein-1 [IMP-1]), which is known by several other names, depending on the activity ascribed by several independent investigators (13Hansen T.V. Hammer N.A. Nielsen J. Madsen M. Dalbaeck C. Wewer U.M. Christiansen J. Nielsen F.C. Dwarfism and impaired gut development in insulin-like growth factor II mRNA-binding protein 1-deficient mice.Mol. Cell. Biol. 2004; 24: 4448-4464Google Scholar, 14Nielsen J. Christiansen J. Lykke-Andersen J. Johnsen A.H. Wewer U.M. Nielsen F.C. A family of insulin-like growth factor II mRNA-binding proteins represses translation in late development.Mol. Cell. Biol. 1999; 19: 1262-1270Google Scholar, 15Ross A.F. Oleynikov Y. Kislauskis E.H. Taneja K.L. Singer R.H. Characterization of a beta-actin mRNA zipcode-binding protein.Mol. Cell. Biol. 1997; 17: 2158-2165Google Scholar, 16Bernstein P.L. Herrick D.J. Prokipcak R.D. Ross J. Control of c-myc mRNA half-life in vitro by a protein capable of binding to a coding region stability determinant.Genes Dev. 1992; 6: 642-654Google Scholar). Thus, coding region determinant-binding protein was originally defined as a regulator of stability of MYC mRNA and a modulator of βTrCP, a ubiquitin ligase regulating Wnt signaling (17Elcheva I. Goswami S. Noubissi F.K. Spiegelman V.S. CRD-BP protects the coding region of betaTrCP1 mRNA from miR-183-mediated degradation.Mol. Cell. 2009; 35: 240-246Google Scholar, 18Leeds P. Kren B.T. Boylan J.M. Betz N.A. Steer C.J. Gruppuso P.A. Ross J. Developmental regulation of CRD-BP, an RNA-binding protein that stabilizes c-myc mRNA in vitro.Oncogene. 1997; 14: 1279-1286Google Scholar); other names reflect other functions, including the regulated delivery of β-actin to cell lamellae (zipcode-binding protein-1, ZBP1; the complete list of alternate names [IGF2BP1 or IMP-1; CRD-BP; and VICKZ1] is provided in the Abbreviations section). Although often described as showing an oncofetal expression pattern (expressed in embryo, not adult tissues, and re-expressed by tumors), we demonstrated significant and approximately similar expression in breast tumor and nontumor cells. The predominant isomer expressed in both conditions was an N-truncated variant initiating from an internal promoter, containing all the mRNA-binding KH domains (19Fakhraldeen S.A. Clark R.J. Roopra A. Chin E.N. Huang W. Castorino J. Wisinski K.B. Kim T. Spiegelman V.S. Alexander C.M. Two isoforms of the RNA binding protein, coding region determinant-binding protein (CRD-BP/IGF2BP1), are expressed in breast epithelium and support clonogenic growth of breast tumor cells.J. Biol. Chem. 2015; 290: 13386-13400Google Scholar). Expression levels were approximately 100× lower than for embryonic cells (either mouse embryonic fibroblasts [MEFs] or 293T cells). Nonetheless, this protein regulated clonogenic growth in vitro (19Fakhraldeen S.A. Clark R.J. Roopra A. Chin E.N. Huang W. Castorino J. Wisinski K.B. Kim T. Spiegelman V.S. Alexander C.M. Two isoforms of the RNA binding protein, coding region determinant-binding protein (CRD-BP/IGF2BP1), are expressed in breast epithelium and support clonogenic growth of breast tumor cells.J. Biol. Chem. 2015; 290: 13386-13400Google Scholar). Indeed, IMP-1 has been shown to be required for clonogenic activity in many tumor cell types, suggesting that it enables some fundamental property required for clonogenic growth (20Buszczak M. Signer R.A. Morrison S.J. Cellular differences in protein synthesis regulate tissue homeostasis.Cell. 2014; 159: 242-251Google Scholar). Other tumor-associated activities have also been linked to IMP-1 expression (21Elcheva I. Tarapore R.S. Bhatia N. Spiegelman V.S. Overexpression of mRNA-binding protein CRD-BP in malignant melanomas.Oncogene. 2008; 27: 5069-5074Google Scholar, 22Ross J. Lemm I. Berberet B. Overexpression of an mRNA-binding protein in human colorectal cancer.Oncogene. 2001; 20: 6544-6550Google Scholar, 23Kobel M. Weidensdorfer D. Reinke C. Lederer M. Schmitt W.D. Zeng K. Thomssen C. Hauptmann S. Huttelmaier S. Expression of the RNA-binding protein IMP1 correlates with poor prognosis in ovarian carcinoma.Oncogene. 2007; 26: 7584-7589Google Scholar, 24Doyle G.A. Bourdeau-Heller J.M. Coulthard S. Meisner L.F. Ross J. Amplification in human breast cancer of a gene encoding a c-myc mRNA-binding protein.Cancer Res. 2000; 60: 2756-2759Google Scholar, 25Ioannidis P. Mahaira L.G. Papadopoulou A. Teixeira M.R. Heim S. Andersen J.A. Evangelou E. Dafni U. Pandis N. Trangas T. 8q24 Copy number gains and expression of the c-myc mRNA stabilizing protein CRD-BP in primary breast carcinomas.Int. J. Cancer. 2003; 104: 54-59Google Scholar). Target mRNAs bind the KH repeat domains of IMP-1 via combinatorial interactions through a looped tertiary structure with short consensus sequences. This complex interaction makes IMP-1 mRNA-binding partners difficult to predict a priori (26Chao J.A. Patskovsky Y. Patel V. Levy M. Almo S.C. Singer R.H. ZBP1 recognition of beta-actin zipcode induces RNA looping.Genes Dev. 2010; 24: 148-158Google Scholar, 27Farina K.L. Huttelmaier S. Musunuru K. Darnell R. Singer R.H. Two ZBP1 KH domains facilitate beta-actin mRNA localization, granule formation, and cytoskeletal attachment.J. Cell Biol. 2003; 160: 77-87Google Scholar). Since endogenous IMP-1 is expressed at such low levels in breast epithelial (EP) cells, the isolation of binding partners has been a challenge. Therefore, we turned to ESP preparation technologies because of their inherent sensitivity, speed, parallel processing capacity, and potential for multiple endpoint assays (28Sperger J.M. Strotman L.N. Welsh A. Casavant B.P. Chalmers Z. Horn S. Heninger E. Thiede S.M. Tokar J. Gibbs B.K. Guckenberger D.J. Carmichael L. Dehm S.M. Stephens P.J. Beebe D.J. et al.Integrated analysis of multiple biomarkers from circulating tumor cells enabled by exclusion-based analyte isolation.Clin. Cancer Res. 2017; 23: 746-756Google Scholar). Briefly, these techniques allow for the gentle and reliable extraction of analyte-bound paramagnetic particles (PMPs) by magnetically immobilizing and removing PMPs from incubation and wash buffers, thereby minimizing the time spent in large wash buffer volumes. By leveraging the surface tension of fluids, samples can be purified within seconds, prohibiting dissociation (and reassociation with noncognate targets), which typically occurs during typical RNA immunoprecipitation (RIP) protocols. We previously showed that low-affinity interactions dissociated after 10 min of wash incubation (29Berry S.M. Chin E.N. Jackson S.S. Strotman L.N. Goel M. Thompson N.E. Alexander C.M. Miyamoto S. Burgess R.R. Beebe D.J. Weak protein-protein interactions revealed by immiscible filtration assisted by surface tension.Anal. Biochem. 2014; 447: 133-140Google Scholar). The two techniques tested here evaluate two versions of this hydrophilic–hydrophobic ESP patterning, one oil based (Immiscible Filtration Assisted by Surface Tension [IFAST]) and the other air based (Sliding Lid for Immobilized Droplet Extractions [SLIDE]). The mRNA-binding partners identified by this analysis include a group of mRNAs encoding proteins destined for mitochondria and several mRNAs encoding proteins involved in glutathione metabolism such as the selenoprotein glutathione peroxidases GPX-1 and GPX-2 (important for the detoxification of superoxides (30Kipp A.P. Selenium-dependent glutathione peroxidases during tumor development.Adv. Cancer Res. 2017; 136: 109-138Google Scholar)). We found that mitochondrial function and GPX-2 mRNA stability required IMP-1 in breast EP cells, confirming previously described regulatory binding reactions of mRNAs with related functions in other cell types (31Janiszewska M. Suva M.L. Riggi N. Houtkooper R.H. Auwerx J. Clement-Schatlo V. Radovanovic I. Rheinbay E. Provero P. Stamenkovic I. Imp2 controls oxidative phosphorylation and is crucial for preserving glioblastoma cancer stem cells.Genes Dev. 2012; 26: 1926-1944Google Scholar, 32Cruz-Zaragoza L.D. Dennerlein S. Linden A. Yousefi R. Lavdovskaia E. Aich A. Falk R.R. Gomkale R. Schondorf T. Bohnsack M.T. Richter-Dennerlein R. Urlaub H. Rehling P. An in vitro system to silence mitochondrial gene expression.Cell. 2021; 184: 5824-5837.e15Google Scholar). Given the high sensitivity and accuracy of these ESP techniques, we propose that this technology will be a useful approach to dissecting RBP function in general, either alone or as a complement to techniques that provide accurate binding site predictions derived from crosslinking (CLIP) studies. IFAST devices are fabricated from polypropylene via injection compression molding (DTE Research and Design, LLC) and consist of linearly aligned wells (volume of 5–15 μl) connected by trapezoidal microfluidic conduits (Fig. 2). These wells are flanked by a larger input well (up to 200 μl volume) on one end and an output well of designer-specified volume (5–10 μl) on the other. The preincubated cell lysate–antibody–PMP (prepared as described in the Experimental procedures section) mixture is transferred to the input well, and PMP-bound biomolecular complexes are purified by a magnet-based pull through the intermediate wells, which consist of alternating solutions of oil (Fluorinert FC-40 oil; Sigma–Aldrich), and aqueous wash phases, to the output well. Note that this requires no pipetting or additional handling beyond the initial loading of the device and takes an average of 20 to 30 s. The utility of this device for identifying valid biological interactions (including weak interactions), for streamlining multiplexed assays of analytes, and for the detection of viral RNAs for clinical diagnostics has been previously demonstrated (29Berry S.M. Chin E.N. Jackson S.S. Strotman L.N. Goel M. Thompson N.E. Alexander C.M. Miyamoto S. Burgess R.R. Beebe D.J. Weak protein-protein interactions revealed by immiscible filtration assisted by surface tension.Anal. Biochem. 2014; 447: 133-140Google Scholar, 33Berry S.M. Regehr K.J. Casavant B.P. Beebe D.J. Automated operation of immiscible filtration assisted by surface tension (IFAST) arrays for streamlined analyte isolation.J. Lab. Autom. 2013; 18: 206-211Google Scholar, 34Berry S.M. Singh C. Lang J.D. Strotman L.N. Alarid E.T. Beebe D.J. Streamlining gene expression analysis: Integration of co-culture and mRNA purification.Integr. Biol. (Camb.). 2014; 6: 224-231Google Scholar, 35Berry S.M. LaVanway A.J. Pezzi H.M. Guckenberger D.J. Anderson M.A. Loeb J.M. Beebe D.J. HIV viral RNA extraction in wax immiscible filtration assisted by surface tension (IFAST) devices.J. Mol. Diagn. 2014; 16: 297-304Google Scholar, 36Moussavi-Harami S.F. Annis D.S. Ma W. Berry S.M. Coughlin E.E. Strotman L.N. Maurer L.M. Westphall M.S. Coon J.J. Mosher D.F. Beebe D.J. Characterization of molecules binding to the 70K N-terminal region of fibronectin by IFAST purification coupled with mass spectrometry.J. Proteome Res. 2013; 12: 3393-3404Google Scholar). To avoid the use of oil-based exclusion, we used a SLIDE device, which depends instead on air-based exclusion (37Casavant B.P. Guckenberger D.J. Beebe D.J. Berry S.M. Efficient sample preparation from complex biological samples using a sliding lid for immobilized droplet extractions.Anal. Chem. 2014; 86: 6355-6362Google Scholar). This has the advantage of eliminating oil from the purification process and the pull-through lysate. The SLIDE device consists of a handle and a base, each with movable magnets within them (37Casavant B.P. Guckenberger D.J. Beebe D.J. Berry S.M. Efficient sample preparation from complex biological samples using a sliding lid for immobilized droplet extractions.Anal. Chem. 2014; 86: 6355-6362Google Scholar) (commercial name is EXTRACTMAN from Gilson). A polypropylene well plate (provided by Gilson, Inc) is loaded with samples containing PMPs, wash buffer, and elution buffer. By sliding the SLIDE handle across the base, PMPs are rapidly and efficiently transferred between reagents in series. Importantly, PMPs are collected on a disposable PMP collection strip, which comprises highly polished uncharged polypropylene. Thus, this hydrophobic PMP collection minimizes carryover of aqueous material as the SLIDE handle moves between reagents. In RIP experiments, the input wells of this device are loaded with cell lysate, and the RBP complex–bound PMP beads are moved through adjacent wells containing wash buffer as described previously (Fig. 3); total time for exclusion purification is approximately 20 s. By passing the PMPs carrying immunoprecipitation complexes through oil (IFAST) or air (SLIDE) by attraction to a magnet, the aqueous dead volumes are minimized, reducing the time and handling required to dilute out associated fluids (i.e., to wash immunoprecipitates). The internal aqueous volume of PMPs is approximately 115 nl for each 5 μl volume of beads. We optimized the protocol for this specific buffer composition, given that the residual surface volume determines surface tension (increased by higher salt and decreased by detergent). Samples were processed simultaneously for up to four immuno-PMP lysates (directly in parallel), whereas samples were processed individually in the IFAST devices. We tested the efficiency of the recovery of endogenous IMP-1 protein by immunoprecipitation using IFAST, first for two cell types, 293T human embryonic kidney cells and MEFs, and then for a tagged IMP-1 protein (applying a different antibody, anti-FLAG) expressed in cultured mouse mammary EP cells (Fig. 4, A and B) (38Jonson L. Vikesaa J. Krogh A. Nielsen L.K. Hansen T. Borup R. Johnsen A.H. Christiansen J. Nielsen F.C. Molecular composition of IMP1 ribonucleoprotein granules.Mol. Cell. Proteomics. 2007; 6: 798-811Google Scholar, 39Hafner M. Landthaler M. Burger L. Khorshid M. Hausser J. Berninger P. Rothballer A. Ascano Jr., M. Jungkamp A.C. Munschauer M. Ulrich A. Wardle G.S. Dewell S. Zavolan M. Tuschl T. Transcriptome-wide identification of RNA-binding protein and microRNA target sites by PAR-CLIP.Cell. 2010; 141: 129-141Google Scholar). Using the anti-IMP-1 antibody to purify endogenous IMP-1 from 293T cells and MEFs, the yield of purified protein was 50% to 60% of total input; for the high-affinity anti-FLAG antibody, yield was even higher and losses were insignificant. Vinculin was used as an indicator of nonspecific protein adsorption and was not detectable. We also tested the efficiency of the affinity purification by assaying residual antibody in the unbound fraction and found almost no losses for the immunocomplexes during extraction from the cell lysates (Fig. 4C). We next evaluated the efficiency of recovery when shorter times were allowed for immune complexation. Maximal recovery was found for overnight incubation, but significant recovery was obtained using only 30 min of binding (24% for 30 min compared with 60% recovery for overnight complexation) (Fig. 4D). For unstable RNAs or rapidly reassociating species, these short preincubation times could be particularly important. To evaluate the efficiency of recovery for cells with low endogenous levels of IMP-1, we tested a nontumorigenic mouse mammary EP cell strain, EP cells (40Chin E.N. Martin J.A. Kim S. Fakhraldeen S.A. Alexander C.M. Lrp5 has a wnt-independent role in glucose uptake and growth for mammary epithelial cells.Mol. Cell. Biol. 2015; 36: 871-885Google Scholar). Although IMP-1 is typically 100× less expressed in cells derived from adults compared with fetal cells, IMP-1 is still functionally important, at least for the expression of clonogenicity in vitro (19Fakhraldeen S.A. Clark R.J. Roopra A. Chin E.N. Huang W. Castorino J. Wisinski K.B. Kim T. Spiegelman V.S. Alexander C.M. Two isoforms of the RNA binding protein, coding region determinant-binding protein (CRD-BP/IGF2BP1), are expressed in breast epithelium and support clonogenic growth of breast tumor cells.J. Biol. Chem. 2015; 290: 13386-13400Google Scholar). We showed that the efficiency of pull through of IMP-1 by IFAST from mouse mammary EP cells was approximately the same as for cell lines with higher endogenous levels of IMP-1 (shown in Fig. 4); here measured at 62% by Western blotting (Fig. 5A). Using the IFAST protocol, fivefold more RNA was pulled through with the IMP-1 immunocomplexes than with the control (immunoglobulin G [IgG])-bound PMPs (1.1% compared with 0.2% for anti-IgG) (Fig. 5A). A “standard” immunoprecipitation protocol without crosslinking was compared with IFAST-purified RIP complexes; in other words, we used typical serial pipetting operations to conduct sequential and manual washes of each immunoprecipitate-bound PMP sample in individual Eppendorf tubes. We found broadly similar efficiency for recovery of both RBP and the total associated RNA (Fig. 5B). We also tested whether the RNAs pulled through by this enhanced immunoprecipitation protocol included mRNA-binding partners previously characterized as IMP-1-binding partners in 293T cells (38Jonson L. Vikesaa J. Krogh A. Nielsen L.K. Hansen T. Borup R. Johnsen A.H. Christiansen J. Nielsen F.C. Molecular composition of IMP1 ribonucleoprotein granules.Mol. Cell. Proteomics. 2007; 6: 798-811Google Scholar). All 10 mRNA species surveyed were significantly pulled through by SLIDE–RIP (Fig. 5C). Ranked gene lists of bound and unbound mRNA fractions were compared for IMP-1-associated or IgG-associated IFAST-purified RIP fractions to identify species that showed a significant change in rank listing (p < 0.01). The IMP-1 gene list included 1343 genes, of which 443 (approximately 35%) overlapped with the gene list from IgG control fractions (1170 genes). These “sticky mRNAs” were subtracted from the total to generate a list of 900 potentially specific mRNAs in IMP-1-associated complexes. The fold enrichment of these 900 mRNA species (all greater than twofold) is illustrated in Figure 6A, and the mRNAs most highly enriched are shown in Figure 6B (greater than fourfold). To verify the array analysis of RIP fractions, we selected >30 mRNA species for evaluation by quantitative RT–PCR (qRT–PCR), including enriched and excluded mRNAs (Fig. 6C). For the purpose of illustration, we set a threshold on this confirmation assay; this threshold excludes 93% of mRNAs not enriched by array analysis and also increases the stringency of inclusion in the specifically enriched fraction. To test whether this group of genes includes mRNAs significantly associated with specific cellular processes, we analyzed the group of 900 genes by Search Tool for the Retrieval of Interacting Genes/Proteins analysis (Fig. 7). We found significant enrichment of genes involved in glutathione metabolism, including GPX-1 and GPX-2, which together catalyze the reduction of organic hydroperoxides and H2O2 by glutathione, protecting cells against oxidative damage (labeled “glutathione metabolism”). Also enriched were glutathione-S-transferases (GSTs) of the mu, theta, and omega classes (GST Mu-1, GST Mu-5, GST theta-1, and GST omega-2), involved in detoxification of electrophilic compounds (including carcinogens, therapeutic drugs, environmental toxins, and products of oxidative stress by conjugation with microsomal GST), an