Click Chemistry-Mediated Complementation Assay for RNA–Protein

Letter Cite This: ACS Comb. Sci. 2019, 21, 522−527

pubs.acs.org/acscombsci

Click Chemistry-Mediated Complementation Assay for RNA−Protein Interactions Emily J. Sherman,†,⊥ Daniel A. Lorenz,†,⊥ and Amanda L. Garner*,†,‡ †

Program in Chemical Biology and ‡Department of Medicinal Chemistry, College of Pharmacy, University of Michigan, Ann Arbor, Michigan 48109, United States

Downloaded via NOTTINGHAM TRENT UNIV on August 13, 2019 at 13:03:32 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.

S Supporting Information *

ABSTRACT: Click chemistry-based assays are a growing class of biochemical assay for facilitating the discovery of modulators of important biological processes. To date, most have relied on the use of immobilized biomolecules, which increases the cost of the assay and decreases throughput because of the necessary washing steps. To overcome these challenges, we have developed a click chemistry-mediated complementation assay that retains many of the advantages of the previous technology, including catalytic signal amplification for assay robustness and applicability to full-length biomolecules, but that can be performed in a homogeneous format. As demonstration of this methodology, we have developed a new high-throughput screening method for RNA−protein interactions using the interaction of Lin28 with the premicroRNA, prelet-7, as a model. KEYWORDS: RNA−protein interactions, RNA−binding proteins, click chemistry, high-throughput screening

C

decreases its potential for application to ultra-HTS formats and screening very large compound libraries (>100 000). These disadvantages inspired us to design a new click chemistry assay that retains the high sensitivity of catELCCA but can be performed in a homogeneous format. Using an RNA−protein interaction as a model, herein, we report a click chemistry-based split enzyme assay with applicability to HTS and inhibitor discovery. A commonly used technique for studying biomolecular interactions is the protein complementation assay.10 Recently, a new complementation assay based on split nanoluciferase (NanoLuc) was reported, where the enzyme was divided into two subunits, a 1.3 kDa SmBiT and 18 kDa LgBiT.11 The NanoBiTs have weak intrinsic affinity for one another (Kd = 190 μM), which is enhanced upon fusion of each fragment to protein-interacting partners to generate catalytic signal amplification from the reassembled enzyme.11 Because biochemical complementation assays have not yet been reported for RNA−protein interactions (RPIs),12−14 we sought to utilize the NanoBiT system to design a new HTS assay for RPIs. Unlike protein labeling, however, modification of small RNAs ( 0.5,8 and negligible compound interference in comparison to traditional fluorescence-based assays, such as fluorescence polarization and fluorescence resonance energy transfer.1 Although cat-ELCCA has been successful in identifying and characterizing small molecule and peptide inhibitors,4,7,9 a limiting factor in its use for HTS is its reliance on streptavidincoated well plates, which are both costly and limited to 96- or 384-well format from commercial vendors. Additionally, similar to enzyme-linked immunosorbent assays (ELISA), cat-ELCCA requires several washing steps that further © 2019 American Chemical Society

Received: April 12, 2019 Revised: May 28, 2019 Published: June 3, 2019 522

DOI: 10.1021/acscombsci.9b00071 ACS Comb. Sci. 2019, 21, 522−527

Letter

ACS Combinatorial Science

Figure 1. Comparison of click chemistry-based biochemical assays. X and Y represent click chemistry handles. Z represents a click chemistry product. (a) Generic scheme for cat-ELCCA and (b) click chemistry-mediated NanoLuc complementation assay for RPIs.

chemistry handle, as in cat-ELCCA,5,6 and the RBP generated as a LgBiT fusion protein. The RPI could then be detected following assembly of NanoLuc and a click chemistry reaction of the RNA with a labeled SmBiT (Figure 1b). This click chemistry step would enable stability of the complex to support robust signal generation, even with weak RPIs. To demonstrate proof-of-concept for this approach, we first generated HaloTag fusion proteins of LgBiT and SmBiT to examine the efficiency of different click chemistry reactions for stabilizing the split NanoLuc enzyme. HaloTag was chosen due to its ability to be readily labeled with chloroalkane-tagged ligands.15 As bioorthogonal chemistries, we examined the copper(I)-catalyzed alkyne−azide cycloaddition (CuAAC) reaction,16,17 strain-promoted AAC (SPAAC) reaction,18 and inverse-electron demand Diels−Alder (IEDDA) reaction of a trans-cyclooctene (TCO) and methyltetrazine (mTet) (Figure 2a; structures of HT click chemistry ligands can be found in Figure S3).19,20 Upon incubation of a 1:1 mixture of each labeled BiT with the appropriate click chemistry reagents and treatment with Nano-Glo luciferase reagent, chemiluminescence signal was observed. Similar to our work with catELCCA,6 we found that the IEDDA reaction yielded a much greater signal-to-background ratio (S/B) than the CuAAC and SPAAC reactions with S/B values of 3500, 1500, and 1700, respectively, in comparison to a reagent only control (Figure 2b). Importantly, using the IEDDA assay as a model, further characterization revealed that signal generation was dependent upon LgBiT-SmBiT assembly, as very little signal was observed in the presence of only one of the BiTs (Figure 2c). Moreover, these studies also demonstrated that the click chemistry reaction was necessary for robust signal production, and a 7.3fold enhancement in chemiluminescence was observed when both BiTs were labeled with the appropriate IEDDA handle (Figure 2c). Separately, we also determined that the reaction

proceed quickly with maximum signal generated within 10 min that was stable for up to 30 min (Figure S4). Encouraged by these promising proof-of-concept data, we were excited to apply this IEDDA-based complementation strategy for RPI assay development. As a model, we selected the interaction of the RBP, Lin28, with pre-let-7d, a precursor microRNA (miRNA or miR), as we previously developed a cat-ELCCA for this RPI and used it in a HTS campaign.7 Lin28 functions as an inhibitor of the processing of select let-7 family members by binding to the terminal loop of the primary and precursor intermediates and inducing their degradation.21−25 As let-7s serve as tumor suppressor miRNAs, aberrant Lin28 activity has been linked to a number of human cancers, making Lin28 a promising therapeutic target.26 In addition to Lin28, many other RBPs have been identified that regulate the biogenesis of miRNAs;27,28 thus, we envisioned that development of a new screening technology for the discovery of small molecule modulators of these interactions would be useful as the field advances. To begin our assay development, we expressed and purified Lin28 as a N-terminal fusion protein with LgBiT (LgBiTLin28) and SmBiT as a C-terminal fusion with HaloTag (SmBiT-HT).15 SmBiT-HT was subsequently labeled with a mTet HT ligand (SmBiT-HT-mTet).4 Pre-let-7d containing an aminoallyl uridine modification in the terminal loop was then conjugated to TCO similar to our work with catELCCA.6,9 With each of the components in hand, we then tested our assay design, which is shown in Figure 3a. To our delight, upon incubation of LgBiT-Lin28, SmBiT-HT-mTet, and pre-let-7d-TCO and treatment with the Nano-Glo reagent, 10−11-fold enhancements in chemiluminescence signal were observed over controls lacking RNA or SmBiT (Figure 3b). Importantly, we also observed selectivity with respect to pre523

DOI: 10.1021/acscombsci.9b00071 ACS Comb. Sci. 2019, 21, 522−527

Letter

ACS Combinatorial Science

Figure 2. Exploration of click chemistry reactions for mediating robust reassembly of the split NanoLuc. (a) Model reactions using HaloTag (HT)tagged LgBiT and SmBiT. (b) Chemiluminescence signal generated upon click chemistry reaction of a 1:1 mixture of each of the BiTs (500 nM) after incubation for 30 min at room temperature and subsequent treatment with Nano-Glo luciferase reagent. ****: p < 0.0001. Control refers to Nano-Glo luciferase reagent in buffer. (c) Characterization of the IEDDA assay to demonstrate that signal generation is dependent upon NanoBiT complementation and click chemistry-mediated stabilization of the reassembled NanoLuc. The assay was performed as described in (b). ****: p < 0.0001.

signal generation was due to the formation of a 1:1 complex between pre-let-7d and Lin28, we performed the assay with varying stoichiometry of LgBiT-Lin28 and pre-let-7d-TCO but with constant SmBiT-HT-mTet. As shown in Figure 4a, no difference in chemiluminescence was observed with increased

let-7d, and a 4−5-fold increase in signal was observed over control experiments performed using pre-miR-21 and -10b that show minimal binding to Lin28 (Figure 3b). Encouraged by these preliminary results, we then went on to further characterize our new RPI assay system. To determine if 524

DOI: 10.1021/acscombsci.9b00071 ACS Comb. Sci. 2019, 21, 522−527

Letter

ACS Combinatorial Science

Figure 3. Click chemistry-mediated complementation assay for the pre-let-7d-Lin28 RPI. (a) Assay scheme. (b) Assay characterization to determine the significance of the IEDDA reaction and RPI in signal generation. Equimolar mixtures of each protein or RNA (500 nM) as described in the x-axis were incubated for 30 min at room temperature prior to addition of Nano-Glo luciferase reagent and chemiluminescence detection. ****: p < 0.0001.

Because our goal is to develop an assay that can be used for inhibitor discovery, we next examined antagonists of the prelet-7d-Lin28 interaction. We first tested the ability of HTLin287 to compete with LgBiT-Lin28. As can be seen in Figure 5a, a dose-dependent decrease in chemiluminescence signal was observed with increasing concentration of HT-Lin28. Of significance, a similar effect was found with a small molecule Lin28 inhibitor previously reported by our group, CCG233094 (Figure 5b).7 Finally, to determine the HTS suitability of the assay, we miniaturized it for 384-well plate format and measured the Z′ factor using automated liquid handling.8 From this analysis, we found that the assay performed excellently with a Z′ factor of 0.86 and S/B of 7.7. Importantly, this S/B was similar to that measured using the 96-well plate format (S/ B of 10; Figure 3b), demonstrating maintained sensitivity upon miniaturization. In comparison to cat-ELCCA, this assay has an improved Z′ factor, but decreased S/B (0.5 and 76 for catELCCA, respectively),7 potentially because of its dependence upon ternary complex formation rather than a direct click reaction with the detection enzyme. Future efforts will explore the dependence on SmBiT-HT-mTet concentration in this miniaturized assay system. Despite this, these represent

RNA or protein concentration, indicating that the assay is driven through formation of a 1:1 pre-let-7d-Lin28 RPI. We also tested the impact of increased SmBiT-HT-mTet concentration, which resulted in a 1.5-fold enhancement in chemiluminescence (Figure 4a), highlighting the potential rate-determining nature of ternary complex formation. We next confirmed that signal generation was dependent upon the presence of RNA. Importantly, upon titration of prelet-7d-TCO, a dose-responsive increase in chemiluminescence signal was observed (Figure 4b). From this data, we measured an apparent Kd (Kd,app) value of 198 nM, which is in agreement with our previous measurement as determined via cat-ELCCA (Kd,app = 106 nM).7 To further characterize the RNAdependence of the assay and its ability to monitor real-time changes in the RPI, we treated with RNase A. As shown in Figure 4c, chemiluminescence signal was decreased following RNase A-catalyzed pre-let-7d degradation in a time-dependent manner. Together with the results shown in Figure 3b, these experiments provide key data demonstrating that the mechanism of RPI- and click chemistry-driven catalytic signal amplification is as designed. 525

DOI: 10.1021/acscombsci.9b00071 ACS Comb. Sci. 2019, 21, 522−527

Letter

ACS Combinatorial Science

Figure 5. Inhibitor characterization and HTS potential. (a) Competition with HT-Lin28. (b) Competition with small molecule Lin28 inhibitor CCG-233094. In panels a and b, inhibitor was added following addition of LgBiT-Lin28, SmBiT-HT-mTet, and pre-let-7dTCO (500 nM), and the assay was performed as described. (c) Z′ factor using automated liquid handling in 384-well plate format (data from a quarter plate is shown). Figure 4. Assay characterization. (a) Dependence on stoichiometric pre-let-7d-Lin28 binding. Varying mixtures of RNA and protein, as described in the x-axis (500 or 1000 nM), were incubated for 30 min at room temperature prior to addition of Nano-Glo luciferase reagent and chemiluminescence detection. Chemiluminescence values are normalized to the results obtained using a 1:1:1 mixture of and prelet-7d-TCO, LgBiT-Lin28, and SmBiT-HT-mTet (500 nM). (b) Dependence on the concentration of pre-let-7d-TCO in the presence of 500 nM LgBiT-Lin28 and SmBiT-HT-mTet. 95% confidence interval for Kd,app is 171−234 nM with a Hill slope of 1.5. (c) Signal depletion following treatment with RNase A. T0 = signal after addition of RNase A. T5 = signal at 5 min post addition of RNase A. Control refers to the assay performed using an equimolar mixture of LgBiT-Lin28, SmBiT-HT-mTet, and pre-let-7d-TCO (500 nM) as described above, which also served as a normalization control.

fluorescence-based methods, as no washing steps are included to remove interferents. On the basis of its excellent assay statistics and adaptability to automated liquid handling, we envision ready application to ultra-HTS formats for RPI inhibitor discovery. Moreover, as chemical RNA synthesis can facilitate the generation of various labeled RNAs and both Nand C-terminal HT fusions are available, this approach should be adaptable to other RPI systems for HTS design and development. As recent studies have shown that RNAs are invariably bound to and often modified by RBPs, RPIs play key roles in regulating many aspects of coding and noncoding RNA biology and are likely to be exciting drug targets in the future.29 It is our hope that technologies like that described will enable such efforts.

■

promising assay statistics, indicating the potential of this new click chemistry-based biochemical assay for RPI HTS.1,7 In conclusion, we have designed a click chemistry-mediated complementation assay for RPIs that can be performed in a homogeneous format. This new assay has many of the same advantages as cat-ELCCA, such as catalytic signal amplification to obtain a robust and highly sensitive assay system and capability to use full-length biomolecules. Unlike cat-ELCCA, however, it will be subject to compound interference similar to

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acscombsci.9b00071. General materials and methods, protein expression and purification, bioconjugation methods, assay protocols, and supplemental figures (PDF) 526

DOI: 10.1021/acscombsci.9b00071 ACS Comb. Sci. 2019, 21, 522−527

Letter

ACS Combinatorial Science

■

measurement of protein interactions in cells. ACS Chem. Biol. 2016, 11, 400−408. (12) Rackham, O.; Brown, C. M. Visualization of RNA-protein interactions in living cells: FMRP and IMP1 interact on mRNAs. EMBO J. 2004, 23, 3346−3355. (13) Han, Y.; Wang, S.; Zhang, Z.; Ma, X.; Li, W.; Zhang, X.; Deng, J.; Wei, H.; Li, Z.; Zhang, X.-E.; Cui, Z. In vivo imaging of proteinprotein and RNA-protein interactions using novel far-red fluorescence complementation systems. Nucleic Acids Res. 2014, 42, No. e103. (14) Porter, J. R.; Stains, C. I.; Jester, B. W.; Ghosh, I. A general and rapid cell-free approach for the interogation of protein-protein, protein-DNA, and protein-RNA interactions and their antagonists utilizing split-protein reporters. J. Am. Chem. Soc. 2008, 130, 6488− 6497. (15) Los, G. V.; Encell, L. P.; McDougall, M. G.; Hartzell, D. D.; Karassina, N.; Zimprich, C.; Wood, M. G.; Learish, R.; Ohana, R. F.; Urh, M.; Simpson, D.; Mendez, J.; Zimmerman, K.; Otto, P.; Vidugiris, G.; Zhu, J.; Darzins, A.; Klaubert, D. H.; Bulleit, R. F.; Wood, K. V. HaloTag: a novel protein labeling technology for cell imaging and protein analysis. ACS Chem. Biol. 2008, 3, 373−382. (16) Rostovtsev, V. V.; Green, L. G.; Fokin, V. V.; Sharpless, K. B. A stepwise Huisgen cycloaddition process: copper(I)-catalyzed regioselective ‘ligation’ of azides and terminal alkynes. Angew. Chem., Int. Ed. 2002, 41, 2596−2599. (17) Tornøe, C. W.; Christensen, C.; Meldal, M. Peptidotriazoles on solid phase: [1,2,3]-triazoles by regiospecific copper(I)-catalyzed 1,3dipolar cycloadditions of terminal alkynes to azides. J. Org. Chem. 2002, 67, 3057−3064. (18) Ning, X.; Guo, J.; Wolfert, M. A.; Boons, G.-J. Visualizing metabolically labeled glycoconjugates of living cells by copper-free and fast Huisgen cycloadditions. Angew. Chem., Int. Ed. 2008, 47, 2253−2255. (19) Blackman, M. L.; Royzen, M.; Fox, J. M. Tetrazine ligation: fast bioconjugation based on inverse-electron-demand Diels-Alder reactivity. J. Am. Chem. Soc. 2008, 130, 13518−13519. (20) Devaraj, N. K.; Upadhyay, R.; Haun, J. B.; Hilderbrand, S. A.; Weissleder, R. Fast and sensitive pretargeted labeling of cancer cells through a tetrazine/trans-cyclooctene cycloaddition. Angew. Chem., Int. Ed. 2009, 48, 7013−7016. (21) Viswanathan, S. R.; Daley, G. Q.; Gregory, R. I. Selective blockade of microRNA processing by Lin28. Science 2008, 320, 97− 100. (22) Newman, M. A.; Thomson, J. M.; Hammond, S. M. Lin-28 interaction with the let-7 precursor loop mediates regulated microRNA processing. RNA 2008, 14, 1539−1549. (23) Heo, I.; Joo, C.; Cho, J.; Ha, M.; Han, J.; Kim, V. N. Lin28 mediates the terminal uridylation of let-7 precursor microRNA. Mol. Cell 2008, 32, 276−284. (24) Heo, I.; Joo, C.; Kim, Y.-K.; Ha, M.; Yoon, M.-J.; Cho, J.; Yeom, K.-H.; Han, J.; Kim, V. N. TUT4 in concert with Lin28 suppressess microRNA biogenesis through pre-microRNA uridylation. Cell 2009, 138, 696−708. (25) Ustianenko, D.; Chiu, H.-S.; Treiber, T.; Weyn-Vanhentenryck, S. M.; Treiber, N.; Meister, G.; Sumazin, P.; Zhang, C. LIN28 selectively modulates a subclass of let-7 microRNAs. Mol. Cell 2018, 71, 271−283. (26) Balzeau, J.; Menezes, M. R.; Cao, S.; Hagan, J. P. The LIN28/ let-7 pathway in cancer. Front. Genet. 2017, 3, 31. (27) Treiber, T.; Treiber, N.; Plessmann, U.; Harlander, S.; Daiß, J.L.; Eichner, N.; Lehmann, G.; Schall, K.; Urlaub, H.; Meister, G. A compendium of RNA-binding proteins that regulate microRNA biogenesis. Mol. Cell 2017, 66, 270−284. (28) Treiber, T.; Treiber, N.; Meister, G. Regulation of microRNA biogenesis and its crosstalk with other cellular pathways. Nat. Rev. Mol. Cell Biol. 2019, 20, 5−20. (29) Gerstberger, S.; Hafner, M.; Tuschl, T. A census of human RNA-binding proteins. Nat. Rev. Genet. 2014, 15, 829−845.

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. ORCID

Amanda L. Garner: 0000-0002-0870-3347 Author Contributions ⊥

E.J.S. and D.A.L. contributed equally.

Funding

This work was supported through a Catalyst Award from the Dr. Ralph and Marian Falk Medical Research Trust (A.L.G.) and the NSF (E.J.S.). Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS We thank James Song for providing HT click chemistry ligands, Evan Barnes for preliminary experiments, and Steve Vander Roest for assistance with HTS equipment setup.

■

ABBREVIATIONS cat-ELCCA, catalytic enzyme-linked click chemistry assay; HT, HaloTag; IEDDA, inverse-electron demand Diels−Alder; HTS, high-throughput screening; NanoLuc, nanoluciferase; RBP, RNA-binding protein; RPI, RNA−protein interaction; miRNA, microRNA

■

REFERENCES

(1) Garner, A. L. cat-ELCCA: catalyzing drug discovery through click chemistry. Chem. Commun. 2018, 54, 6531−6539. (2) Garner, A. L.; Janda, K. D. cat-ELCCA: a robust method to monitor the fatty acid acyltransferase activity of ghrelin Oacyltransferase (GOAT). Angew. Chem., Int. Ed. 2010, 49, 9630− 9634. (3) Garner, A. L.; Janda, K. D. A small molecule antagonist of ghrelin O-acyltransferase (GOAT). Chem. Commun. 2011, 47, 7512− 7514. (4) Song, J. M.; Menon, A.; Mitchell, D. C.; Johnson, O. T.; Garner, A. L. High-throughput chemical probing of full-length protein-protein interactions. ACS Comb. Sci. 2017, 19, 763−769. (5) Lorenz, D. A.; Song, J. M.; Garner, A. L. High-throughput platform assay technology for the discovery of pre-microRNAselective small molecule probes. Bioconjugate Chem. 2015, 26, 19−23. (6) Lorenz, D. A.; Garner, A. L. A click chemistry-based microRNA maturation assay optimized for high-throughput screening. Chem. Commun. 2016, 52, 8267−8270. (7) Lorenz, D. A.; Kaur, T.; Kerk, S. A.; Gallagher, E. E.; Sandoval, J.; Garner, A. L. Expansion of cat-ELCCA for the discovery of small molecule inhibitors of the pre-let-7-Lin28 RNA-protein interaction. ACS Med. Chem. Lett. 2018, 9, 517−521. (8) Zhang, J.-H.; Chung, T. D. Y.; Oldenburg, K. R. A simple statistical parameter for use in evaluation and validation of high throughput screening assays. J. Biomol. Screening 1999, 4, 67−73. (9) Lorenz, D. A.; Vander Roest, S.; Larsen, M. J.; Garner, A. L. Development and implementation of an HTS-compatible assay for the discovery of selective small-molecule ligands for pre-microRNAs. SLAS Disc. 2018, 23, 47−54. (10) Michnick, S. W.; Ear, P. H.; Manderson, E. N.; Remy, I.; Stefan, E. Universal strategies in research and drug discovery based on protein-fragment complementation assays. Nat. Rev. Drug Discovery 2007, 6, 569−582. (11) Dixon, A. S.; Schwinn, M. K.; Hall, M. P.; Zimmerman, K.; Otto, P.; Lubben, T. H.; Butler, B. L.; Binkowski, B. F.; Machleidt, T.; Kirkland, T. A.; Wood, M. G.; Eggers, C. T.; Encell, L. P.; Wood, K. V. NanoLuc complementation reporter optimized for accurate 527

DOI: 10.1021/acscombsci.9b00071 ACS Comb. Sci. 2019, 21, 522−527