Coupling the folding of a β-hairpin with chelation-enhanced luminescence of Tb(iii) and Eu(iii) ions for specific sensing of a viral RNA

Rational modification of a natural RNA-binding peptide with a lanthanide EDTA chelator, and a phenanthroline ligand yields a highly selective luminescent sensor.


General
All reagents were acquired from commercial sources: DMF and TFA were purchased from Scharlau, CH 2 Cl 2 from Panreac and CH 3 CN from Merck. The rest of the reagents were acquired from Sigma-Aldrich. Reactions were monitored by analytical RP-HPLC with an Agilent 1100 series LC/MS using an Eclipse XDB-C 18 (4.6 × 150 mm, 5 μm) analytical column. Compounds were detected by UV absorption (220, 270, 380 and 495 nm); the standard conditions for analytical RP-HPLC consisted on an isocratic regime during the first 5 min, followed by a linear gradient from 5 to 95% of solvent B for 30 min at a flow rate of 1 mL/min (A: water with 0.1% TFA, B: acetonitrile with 0.1% TFA).
Electrospray Ionization Mass Spectrometry (ESI/MS) of compounds EDTA-Tat-Lys(ϕ)/Eu(III) and EDTA-Tat-Lys(ϕ)/Tb(III) was performed with a Bruker Amazon IT/MS in positive scan mode using direct injection of the peptide-lanthanide solution into the MS.
Purification of compounds was carried out by column chromatography using as stationary phase silica gel 60 SDS type, 230-400 mesh (Merck) and using mixtures of MeOH/CHCl 3 as eluents; or by preparative purification system Büchi Sepacore, consisting of a pump manager C-615 with two pump modules C-605 for binary solvent gradients, a C-660 fraction collector, and C-635 UV detector. Purification was carried out using an isocratic regime during the first 5 min at 5% of solvent B, and then linear gradient from 5% to 75% of solvent B for 30 min at a flow rate of 30 mL/min (A: water with 0.1% TFA, B: methanol with 0.1% TFA). Separations were made on prepacked preparative cartridges (150 × 40 mm) with reverse phase RP-18 silica gel (Büchi #54863).
Reverse phase HPLC purification was performed on an Agilent 1100 series equipped with a binary pump system and a UV-visible detector using a Phenomenex Luna C18 100A (250 × 21.20 mm, 10 μm) preparative column. Purification was carried out using an isocratic regime during the first 5 min at 5% of solvent B and then linear gradients from 5% to 75% of solvent B for 30 min at a flow rate of 3 mL/min (A: water with 0.1% TFA, B: acetonitrile with 0.1% TFA). The fractions containing the products were freeze-dried, and their identity confirmed by ESI-MS and NMR.
NMR spectra were recorded on a Varian Mercury 300. Chemical shifts are given in ppm relative to TMS signal and coupling constants in Hz.
Time-gated emission measurements were made with a Varian Cary Eclipse Fluorescence Spectrophotometer. All measurements were made with a Hellma macro cuvette (111-QS) at 20 ºC, using the following settings: excitation wavelength 300 nm; excitation slit width 20.0 nm, emission slit width 10.0 nm; increment 1.0 nm; average time 0.2 s; total decay time 0.02 s; delay time 0.2 ms; PMT detector voltage 1000 V.
The steady state emission luminescence measurements were made in a Jobin-Yvon Fluoromax-3 (DataMax 2.20), coupled to a temperature controller Wavelength Electronics LFI-3751. All measurements were performed at 20 ° C in a Hellma macro cuvette (111-QS), using the following parameters: excitation wavelength of 280 nm, bandwidth in the excitation slit 3.0 nm, bandwidth in the emission slit 6.0 nm, increase 1.0 nm, 0.2 s integration time. The emission spectrum was recorded between 295 and 500 nm.
EMSA was performed with a BIO-RAS Mini Protean gel system, powered by an electrophoresis power supplies PowerPac Basic model, maximum power 150 V, frequency 50.60 Hz at 140 V (constant V). Binding reactions were performed over 20 min at 8ºC in 20 mM Tris⋅HCl (pH 7.5), 90 mM KCl, 1.8 mM MgCl 2 , 1.8 mM EDTA, 9% glycerol, 0.11 mg/mL BSA and 2.2 % NP-40. In the experiments we used 150 nM of the unlabeled TAT RNA. After incubation for 20 min products were resolved by PAGE using a 10% non-denaturing poliacrylamide gel and 0.5X TBE buffer for 45 min at 8 ºC, analyzed by staining with SyBrGold (Molecular Probes: 5 μL in 50 mL of 1X TBE) for 10 min, and visualized by fluorescence 5X TBE buffer (0.445 M Tris⋅HCl, 0.445 M Boric acid, 10 mM ETDA pH 8.0). UV measurements were made in a SmartSpec Plus spectrophotometer using a disposable cuvette trUView from BioRad. Concentrations were measured using the listed extinction coefficients: 287,100 M -1 cm -1 at 260 nm for TAR RNA 1 and 5,579 M -1 cm -1 at 278 nm for Trp. 2 Circular dichroism measurements were made with a Jasco J-715 coupled to a Neslab RTE-111 termostated water bath, using a Hellma macro cuvette (100-QS, 2 mm light pass). Measurements were made at 20 ºC. The spectra are the average of 4 scans.
General peptide synthesis procedures Peptides were synthesized as C-terminal amides (usually in a 0.05 mmol scale using the resin Fmoc-PAL-PEG from Life Technologies: 0.25 mmol/g) on a PS3 Peptide Synthesizer (Protein Technologies) following standard Fmoc solid phase synthesis protocols. All peptide synthesis reagents and amino acid derivatives were purchased from GL Biochem (Shanghai) and Novabiochem; amino acids were purchased as protected Fmoc amino acids with the standard side chain protecting scheme: Fmoc-Pro-OH, Fmoc-Gly-OH, Fmoc-Thr(tBu)-OH, Fmoc-Lys(Boc)-OH, Fmoc-Ile-OH and Fmoc-Arg(Pbf)-OH. All other chemicals were purchased from Aldrich or Fluka. All solvents were dry and synthesis grade.
Peptide bond-forming couplings were conducted for 40 min by using HBTU (10 equiv), DIEA (11 equiv) in DMF (2.5 mL) and 10 equivalents of the amino acids. Each amino acid was incubated for 2 min in the coupling mixture before being added onto the resin. After washing with DMF, the deprotection of the temporal Fmoc protecting group was performed by treating the resin with 20% piperidine in DMF solution for 15 min.
The final peptides were cleaved from the resin, and side-chain protecting groups were simultaneously removed using a standard TFA cleavage cocktail as outlined below. The High-Performance Liquid Chromatography (HPLC) was performed using an Agilent 1100 series Liquid Chromatograph Mass Spectrometer system. Analytical HPLC was run using a Zorbax Eclipse XDB-C 18 (5 μm) 4.6 × 150 mm analytical column from Agilent. The purification of the peptides was performed on a semipreparative Jupiter Proteo 90A (4 μm), 10 × 250 mm reverse-phase column from Phenomenex. The standard gradients used for analytical and preparative HPLC consisted on a linear gradient 5 → 95% CH 3 CN, 0.1% TFA /H 2 O, 0.1% TFA over 30 min, or on an isocratic regime during the first 5 min, followed by a linear gradient from 5% to 75% of solvent B for 30 min. Electrospray Ionization Mass Spectrometry (ESI/MS) was performed with an Agilent 1100 Series LC/MSD VL G1956A model in positive scan mode using direct injection of the purified peptide solution into the MS.
Cleavage and deprotection of the resin bound peptides: The resin-bound peptide dried under argon (≈ 0.025 mmol) was placed in a 50 mL falcon tube and suspended in 2 mL of the cleavage cocktail (100 μL of CH 2 Cl 2 , 50 μL of water, 50 μL of triisopropylsilane (TIS) and TFA to 2 mL), and the resulting mixture was shaken for 3 h. The resin was filtered, and the TFA filtrate was concentrated under argon current to a volume of approximately 1 mL, which was then added to ice-cold diethyl ether (20 mL). After 10 min, the precipitate was centrifuged and washed again with 10 mL of ice-cold ether. The solid residue was dried under argon, redissolved in CH 3 CN/H 2 O 1:1 (1 mL) and purified by semipreparative reverse-phase HPLC. The collected fractions were lyophilized and stored at −20 °C.

Synthesis
The target peptide W-Tat-K(DOTA[Tb]) was assembled following standard Fmoc/tBu solid-phase peptide synthesis procedures; the DOTA chelating unit was introduced into the peptide scaffold as a tri-tBuprotected acid into the side chain of an orthogonally-deprotected Lys residue, while the peptide was still attached to the solid support. After purification the peptide was complexed with TbCl 3 in HEPES buffer to give the desired chelate, which was then repurified by reverse-phase HPLC (see below).
Scheme S2. Synthesis of the peptide with the DOTA unit orthogonally attached to the Lys side chain.
Cleavage and deprotection of semipermanent protecting groups: The cleavage and deprotection of the resin-bound peptide was done following the experimental procedure already described. After reversephase HPLC purification the collected fractions were freeze-dried, and the desired peptide W-Tat-K(DOTA), was isolated as white solid in a 16% yield. Tat  Terbium chelation: The lyophilized peptide was dissolved in 200 μL HEPES buffer 50 mM, pH 7.6, NaCl 500 mM; and 300 μL of TbCl3 50 mM solution in HCl 1 mM was added. Then, the solution was brought of 1 mL in water. The mixture was shaken for 6 h and then HPLC-purified. After lyophilization, a white solid was obtained and identified as the desired terbium-peptide complex. Tat

Luminescence measurements and EMSA experiments
Having at hand the desired metallopeptide probe, we studied the effect of TAR RNA in its emission. As expected, the peptide W-Tat-K(DOTA[Tb]) was weakly emissive by itself. Unfortunately, addition of the target TAR BIV RNA did not promote any increase in its luminescence ( Figure S5a). Control electrophoretic mobility assays (EMSA) in polyacrylamide gel under non-denaturing conditions showed that the peptide was binding to the RNA. Thus, incubation of the BIV TAR hairpin oligonucleotide with increasing concentrations of the peptide W-Tat-K(DOTA[Tb]) gave rise to new slow-migrating bands, consistent with the formation of the expected peptide/RNA complex ( Figure S5b, band 2). Given that the probe appears to maintain the RNA binding capabilities of the parent Tat peptide ( Figure S5b, band 1), the lack of fluorescent emission suggests that the antenna and the Tb complex are too distant for an efficient transfer. Alternatively, the failing might arise from the rapid quenching of the excited state of the Trp by the RNA bases before the energy is transferred to the Tb(III) complex. This is consistent with reports of Trp quenching due to oligonucletoide binding, as well as on the quenching observed in the fluorescence emission of the Trp residue in our own system upon excitation at 300 nm ( Figure S5c). Figure S5a). To 3 mL of a 100 nM solution of the peptide W-Tat-K(DOTA[Tb]) in HEPES buffer (10 mM HEPES pH 7.6, 100 mM NaCl), 1 equivalent of TAR RNA was added. The emission spectra were recorded before and after the addition.

Design and structures
Despite the failing of the probe, the good RNA binding affinity of this first design encouraged us to explore other alternatives. Therefore, we tried adding spacers, such as 5-amino-3-oxapentanoic acid (O1Pen), or polyprolyne helices (P 4 ) between the Trp antenna and the peptide binding module, as well as switching the position of the antenna and the DOTA[Tb] unit. We also synthesized probes containing other antennas than Trp, such as 7-methoxycoumarin (Cou), thiazole orange (TO), or acridine (ACR), which are known to increase their emission upon binding to the RNA (see Table 1, below). Unfortunately, none of those new probes lighted up in the presence of the target TAR RNA hairpin. Table 1. Peptide probes used in this study. Cou = 7-methoxycoumarin, O1Pen = 5-amino-3-oxapentanoic acid, ACR = acridine, and TO = thiazole orange; ϕ = 1,10-phenanthroline-5-carbonyl group.

Synthesis of small molecular fragments and assembly of peptides
The molar extinction coefficient of compound 1 was obtained by linear regression analysis of the UV absorption values of samples of known concentrations by weight.  434nm = 7624 M -1 cm -1 was found for the acridine derivative 1.

Synthesis of 1-(carboxymethyl)-4-methyl-chinolinium bromide (2)
Compound 2 was synthesized following previously described procedures. 3 4-methylquinoline was added (6.2 g, 43 mmol) to a bromoacetic acid solution (6.1 g, 44 mmol) in ethyl acetate (~3 mL). The reaction mixture was stirred at rt overnight. The product appeared as a pale yellow precipitate, which was filtered and washed with EtOAc. The residue was adsorbed in silica and purified by flash column chromatography (20% MeOH/CHCl 3 ) to give a solid identified as the desired product (3.4 g, 35%).

Synthesis of thiobenzoxazole (3)
The benzothiazole derivative was synthesized following reported procedures. 3 Iodomethane (4.1 mL, 67 mmol) was added to a solution of 2-(methylthio)benzothiazole (6.0 g, 33 mmol) in EtOH (15 mL). The mixture was refluxed overnight affording a yellow precipitate, which was filtered and washed with cold EtOH. The resulting light-yellow solid was identified as the desired product (3, 2.8 g, 16% yield).

Synthesis of thiazole orange derivative 4
The synthesis of the thiazole orange derivative was performed as previously described. 4 To a solution of benzothiazole derivative 3 (200 mg, 0.9 mmol) and the quinoline 2 (140 mg, 0.7 mmol) in 14 mL of CH 2 Cl 2 , triethylamine (251 µL, 1.8 mmol) was added. The mixture was stirred in the dark at rt for 17 h. for 4 days to afford a precipitate, which was collected by filtration, washed with hexane and cold water, and freeze-dried to yield a red solid that was identified as the desired product (4, 73 mg, 30% yield).

Synthesis of peptide (DOTA [Tb])-Tat-W
The strategy for the synthesis of peptide (DOTA [Tb])-Tat-W consisted on the coupling of the DOTA unit (following the already described procedure) to the N-terminal end of the peptide after assembling the peptide chain by SPPS. After RP-HPLC purification, the terbium coordination proceeded as previously described.

Peptide chain elongation and Fmoc deprotection:
The peptide chains of each of the four peptides were synthesized automatically in solid phase using the resin PAL-PEG-PS as solid support and in a 0.05 mmol scale, following the general protocol for solid phase synthesis already described. Once the peptides were fully assembled in the solid phase, the N-terminal temporary protecting group was deprotected using the standard conditions (20% piperidine/ DMF) for the subsequent coupling of coumarine, thiazole orange or acridine derivatives over the liberated amines.
Scheme S5. SPPS and Fmoc deprotection for the subsequent coupling of the corresponding dyes.

Coupling of 7-methoxycoumarin-3-carboxilic acid:
The commercially available 7-methoxycoumarin-3carboxilic acid (11 mg, 0.05 mmol) and HATU (19 mg, 0.05 mmol) were dissolved in 250 µL of DMF. Then, 9.4 µL of DIEA (0.55 mmol) were added to the solution and after activating the mixture for 2 minutes, the solution was added over the peptide resins (≈ 0.025 mmol). The mixtures were shaken for two hours and then the resins were filtered and washed with DMF (3 × 3 mL × 3 min).
Scheme S6. Attachement of the coumarin unit to the N-terminal end of the peptide chains.
Scheme S7. Coupling of the acridine derivative 1 to the N-terminal end of the peptide.

Scheme S8.
Coupling of 4 to the N-terminal amine of the resin-bound peptide.

Alloc deprotection and DOTA tris (t-Bu) ester coupling:
Once that each of the dyes were coupled to the N-terminal end of the respective peptides, the side chain of the Lys(Alloc) residue is selectively deprotected for specific attachment of the DOTA unit. The experimental procedures for the deprotection of the orthogonally protected Lys(Alloc) side chain and DOTA coupling to Lys side chain, were the same than the previously described.
Scheme S9. Alloc deprotection and DOTA coupling following the already described procedures.

Eu 3+ coordination:
The lyophilized peptides were dissolved in 200 μL HEPES buffer 50 mM, pH 7.6, NaCl 500 mM; and 300 μL of EuCl 3 50 mM solution in HCl 1 mM were added. Then, the solutions were brought to 1 mL in water. The mixtures were shaken for 6 h and after checking by HPLC-MS that the coordination had occurred they were HPLC-purified. After lyophilization, a white solids were obtained and identified as the desired europium-peptide complexes,

Biophysical assays
Luminescence measurements

Time-gated and steady state experiments with peptide W-O1Pen-Tat-K(DOTA [Tb]): To 3 mL of a 100 nM solution of the peptide W-O1Pen-Tat-K(DOTA [Tb])
in HEPES buffer (10 mM HEPES pH 7.6, 100 mM NaCl), 1 equivalent of TAR RNA was added. The emission spectra were recorded before and after each addition.

Time-gated and steady state experiments with peptide (DOTA [Tb])-Tat-W:
To 3 mL of a 100 nM solution of the peptide (DOTA [Tb])-Tat-W in HEPES buffer (10 mM HEPES pH 7.6, 100 mM NaCl), 1 equivalent of TAR RNA was added. The emission spectra were recorded before and after each addition.

Time-gated and steady state experiments with peptide Cou-P 4 -Tat-P 4 -K(DOTA [Eu]):
To 3 mL of a 100 nM solution of the peptide Cou-P 4 -Tat-P 4 -K(DOTA [Eu]) in HEPES buffer (10 mM HEPES pH 7.6, 100 mM NaCl), 1 equivalent of TAR RNA was added. The emission spectra were recorded before and after each addition.

Time-gated and steady state experiments with peptide TO-Tat-K(DOTA [Eu]): To 3 mL of a 100 nM solution of the peptide TO-Tat-K(DOTA [Eu])
in HEPES buffer (10 mM HEPES pH 7.6, 100 mM NaCl), 1 equivalent of TAR RNA was added. The emission spectra were recorded before and after each addition.

Peptides with phenantroline antennas
4.1. Synthesis of the phenantroline derivative 8 Scheme S10. Synthetic route for the preparation of the phenantroline derivative 8.
The intermediate phenantroline 7 was synthesized as previously described. 5 Reaction of 5 (600 mg, 3.06 mmol) with potassium cyanide (1.2 g, 18 mmol) in water (aprox. 60 mL) at room temperature for 6 h affords a white precipitate that after being filtered and washed was lyophilized. In this way, 5-cyano-1,10phenantroline (6) was obtained directly in an overall yield of 80%, without isolation of the intermediates.  The molar extinction coefficient of compound 7 was obtained by linear regression analysis of the UV absorption values of samples of known concentrations by weight. ε 278nm = 33,847 M -1 cm -1 was found for the phenanthroline derivative 7. 7 (125 mg, 0.56 mmol) was refluxed in SOCl 2 (700 µL) for 5 h. Then, solvents were removed under reduced pressure. The product obtained (8) was used without further purification.

Synthesis of peptide EDTA[Tb]-Tat-Lys(ϕ)
The synthesis of EDTA-Tat-Lys(ϕ) peptide was carried out following the tactic described below, in which the phenanthroline unit was inserted in the side chain of a C-terminal Lys residue, which is introduced with its side chain protected with an Mtt group. Orthogonal deprotection of this group, while the residue is still attached to the solid support, allowed the introduction of the phenantroline unit.
Following the derivatization with the phenanthroline antenna, the rest of the peptide was assembled in solid phase and finally, the chelating ethylenediaminetetraacetic acid (EDTA) unit in the form of a dianhydride was attached to the N-terminus of the peptide to give the desired peptide EDTA-Tat-Lys(ϕ). Scheme S11. Synthethic route for the synthesis of peptide EDTA-Tat-Lys(ϕ).

Fmoc-Lys(Mtt)-OH coupling and orthogonal Mtt deprotection:
Fmoc-Lys(Mtt)-OH was manually coupled to the resin (0.05 mmol scale using as solid support the Fmoc-PAL-PEG-PS resin). The coupling was checked by a TNBS test. 6 Then, side chain of the Lys(Mtt) residue was selectively deprotected for specific attachment of the phenanthroline unit: the resin-bound peptide (≈ 0.05 mmol) was treated three times with a mixture of TFA (250 µL), triisopropylsilane (50 µL) and CH 2 Cl 2 (4.7 mL) for 2 min. The resin was then filtered and washed with CH 2 Cl 2 (2 × 5 mL × 3 min).
Phenantroline coupling to the Lys side chain: 8 (62.4 mg, 0.26 mmol) was dissolved in dry DMF (470 μL). N,N-diisopropylethylamine (DIEA) (146 μL , 0.85 mmol) was added to the solution and resulting mixture was added over the Mtt-deprotected peptide attached to the resin (≈ 0.05 mmol). N 2 was passed through the resin suspension for 1 h and was then filtered and washed with DMF (2 × 5 mL × 3 min).

Cleavage and deprotection of semipermanent protecting groups:
The resin-bound peptide was subjected to cleavage and deprotection by treatment with the cleavage cocktail under the conditions already described. After reverse-phase HPLC purification and lyophilization of the collected fractions, the desired peptide EDTA-Tat-Lys(ϕ) was isolated as white solid in a 10% yield. Tat

Data fitting:
The experimental data of the titration of complexes EDTA-Tat-Lys(ϕ)/Tb(III) and EDTA-Tat-Lys(ϕ)/Eu(III) with TAR RNA were fit to a 1:1 model for a luminescent ligand binding (EDTA-Tat-Lys(ϕ)/Tb(III) or EDTA-Tat-Lys(ϕ)/Eu(III)) to a unlabeled receptor (TAR RNA). If non-specific binding is ignored, then this interaction is described by the following equation, which was used to fit the experimental data using non-linear regression analysis: 7 Where R, concentration of the free receptor in the equilibrium (TAR RNA); R T , total receptor concentration; L, concentration of the free labeled ligand in the equilibrium (EDTA-Tat-Lys(ϕ)/Tb(III) or EDTA-Tat-Lys(ϕ)/Eu(III)); L T , total concentration of the labeled ligand; K D , dissociation constant of the interaction between the receptor and the ligand; F T , total observed emission, F 0 , adjustable parameter accounting for the background emission; F RL adjustable parameter for the labeled ligand-receptor complex molar emission.
Time-gated emission of the peptide complex EDTA-Tat-Lys(ϕ)/Eu with ssDNA, dsDNA and with a non specific RNA hairpin. To 3 mL of a 50 nM solutions of peptide EDTA-Tat-Lys(ϕ) and EuCl 3 in HEPES buffer (10 mM HEPES pH 7.6, 100 mM NaCl), aliquots of a ssDNA (TGG AGA TGA CTC ATC TCG TT) or dsDNA (ds(TGG AGA TGA CTC ATC TCG TT)) were added up to 0.6 and 0.7 μM respectively and the emission spectra were recorded before and after each addition. Phage P22 boxB nonspecific RNA hairpin (GCG CUG ACA AAG CGC) was also added over a 50 nM solution of peptide EDTA-Tat-Lys(ϕ) and EuCl 3 in HEPES buffer (10 mM HEPES pH 7.6, 100 mM NaCl) up to 0.7 μM, and the emission spectra were recorded before and after the addition.   Time-gated emission of EDTA-Tat-Lys(ϕ) in presence of cell lysates. The luminescence spectra of a 50 nM solutions of peptide EDTA-Tat-Lys(ϕ) and EuCl 3 , and peptide EDTA-Tat-Lys(ϕ) and TbCl 3 in HEPES buffer (10 mM HEPES pH 7.6, 100 mM NaCl) were measured before, and after the addition of 3.4 μL of a 147 ng/μL RNA stock solution of cell lysate (total RNA concentration approximately 170 ng/mL). The luminescence spectrum were recorded again after the addition 20 equivalents of TAR RNA to the EDTA-Tat-Lys(ϕ)/Eu and EDTA-Tat-Lys(ϕ)/Tb solutions in the presence of the lysate.
Preparation of cell lysates: Vero cells were grown in monolayers and lysed in buffer A (10 mM HEPES pH 7.2, 10 mM KCl, 100 mM NaCl, 0.5% Triton X100, and proteinase inhibitor cocktail by Sigma). Lysates were cleared by centrifugation, and their concentration was adjusted to contain approximately 147 nanograms per microliter of cellular RNA. Cellular extracts were kept on ice and immediately used for the experiments.

Time-gated emission of EDTA[TB]-Tat-Lys(ϕ) in complete cell lysates:
The peptide sensor EDTA-Tat-Lys(ϕ) and TbCl 3 were added over 1mL of three different cell lysates to a final concentration of 150 nM. The lysates were prepared from HeLa cells that had previously been transfected with TAR RNA, boxB RNA, or a blank buffer solution (no exogenous RNA).
Protocol for RNA transfection and preparation of the cell lysates: Monolayers of HeLa cells were transfected at 70-90% confluency following the manufacturer's instructions and using lipofectamine® 2000 (Thermo Fisher Scientific) and 2µg of the corresponding RNA per well were used. Cells were incubated at 37ºC for 4 hours and then the solution was removed. After washing each well twice with 10 mM HEPES pH 7.4, 10 mM KCl, 100 mM NaCl buffer, cells were lysed with 1 mL of 10 mM HEPES pH 7.4, 10 mM KCl, 100 mM NaCl, 0.5% Triton X100 buffer. Lysates were cleared by centrifugation, and kept on ice and immediately used for the experiments.
Calculation of the detection limit. 20 replicates of a blank sample (50 nM solution of EDTA-Tat-Lys(ϕ) and EuCl 3 ) were measured. The limit of detection was calculated from the value of the mean emission intensity at 616 nm plus five times the standard deviation.

Multicolor detection experiments
All the luminescence measurements next described were time-resolved experiments, using the settings previously described for time-gated emission measurements.
The emission is normalized to those of the peptides mixture in absence of RNA. As the 592 and 614 nm Eu emissions overlap with the Tb 5 D 4 ⟶ 7 F J bands (J = 4, 586 nm; J = 3, 621 nm), we monitored the 5 D 0 ⟶ 7 F 3 (655nm) Eu band, to which the Tb contribution is negligible. 9 Detection of boxB RNA hairpin: To a mixture of the peptide probes P22-N W [Tb] and EDTA-Tat-Lys(ϕ) -both 50 nM-in presence of EuCl 3 50 nM in HEPES buffer (HEPES 10 mM, NaCl 100 mM, pH 7.6), 1 equivalent of boxB RNA was added. The emission spectra were recorded before and after the RNA addition. Error bars show standard error based on three independent experiments.
Detection of TAR RNA hairpin: To a mixture of the peptide probes P22-N W [Tb] and EDTA-Tat-Lys(ϕ) -both 50 nM-in presence of EuCl 3 50 nM in HEPES buffer (HEPES 10 mM, NaCl 100 mM, pH 7.6), 1 equivalent of TAR RNA was added. The emission spectra were recorded before and after the RNA addition. Error bars show standard error based on three independent experiments.
Detection of both boxB and TAR RNA hairpins: To a mixture of the peptide probes P22-N W [Tb] and EDTA-Tat-Lys(ϕ) -both 50 nM-in presence of EuCl 3 50 nM in HEPES buffer (HEPES 10 mM, NaCl 100 mM, pH 7.6), 1 equivalent of boxB RNA and 1 equivalent of TAR RNA were successively added. 10 The emission spectra were recorded before and after each addition. Error bars show standard error based on six independent experiments.

Determination of luminescence quantum yields
The luminescence quantum yields were calculated following already described procedures 11 , using equation (1), were is the integral photon flux, is the absorption factor, is the refractive index of the solvent and is the quantum yield. S denotes de sample and R denotes the standard. An aqueous air-