Rational Design and Tuning of Functional RNA Switch to Control an

Jun 30, 2015 - Frontier Institute for Biomolecular Engineering Research (FIBER), Konan ... Hyun Gyu Lim , Sungho Jang , Sungyeon Jang , Sang Woo Seo ,...
0 downloads 0 Views 2MB Size
Download PDF
Article pubs.acs.org/ac

Rational Design and Tuning of Functional RNA Switch to Control an Allosteric Intermolecular Interaction Tamaki Endoh† and Naoki Sugimoto*,†,‡ †

Frontier Institute for Biomolecular Engineering Research (FIBER), Konan University, 7-1-20 Minatojimaminamimachi, Kobe, 650-0047, Japan ‡ Graduate School of Frontiers of Innovative Research in Science and Technology (FIRST), Konan University, 7-1-20 Minatojimaminamimachi, Kobe, 650-0047, Japan S Supporting Information *

ABSTRACT: Conformational transitions of biomolecules in response to specific stimuli control many biological processes. In natural functional RNA switches, often called riboswitches, a particular RNA structure that has a suppressive or facilitative effect on gene expression transitions to an alternative structure with the opposite effect upon binding of a specific metabolite to the aptamer region. Stability of RNA secondary structure (−ΔG°) can be predicted based on thermodynamic parameters and is easily tuned by changes in nucleobases. We envisioned that tuning of a functional RNA switch that causes an allosteric interaction between an RNA and a peptide would be possible based on a predicted switching energy (ΔΔG°) that corresponds to the energy difference between the RNA secondary structure before (−ΔG°before) and after (−ΔG°after) the RNA conformational transition. We first selected functional RNA switches responsive to neomycin with predicted ΔΔG° values ranging from 5.6 to 12.2 kcal mol−1. We then demonstrated a simple strategy to rationally convert the functional RNA switch to switches responsive to natural metabolites thiamine pyrophosphate, Sadenosyl methionine, and adenine based on the predicted ΔΔG° values. The ΔΔG° values of the designed RNA switches proportionally correlated with interaction energy (ΔG°interaction) between the RNA and peptide, and we were able to tune the sensitivity of the RNA switches for the trigger molecule. The strategy demonstrated here will be generally applicable for construction of functional RNA switches and biosensors in which mechanisms are based on conformational transition of nucleic acids.

R

systems.7 A useful characteristic of these systems is that the intrinsic aptamer domain, which is a recognition unit for the target molecule, can be converted to another aptamer domain to alter target molecule specificity.8 Artificial RNA functional switches that respond to various biologically relevant molecules have been developed by using aptamer domains obtained by a synthetic evolution of ligands by exponential enrichment (SELEX) technology.9 We recently demonstrated that not only RNA but also protein functions can be regulated by designing RNA conformational switches that induce or suppress allosteric RNA−peptide interactions.10−12 Proteins, which output fluorescence, luminescence, and activation of transcription, have been engineered to be responsive to a specific ligand or oligonucleotide by connection to an RNA binding peptide. To expand the applicability of these RNAbased functional systems, a rational and general strategy to design the RNA conformational switches is required.

ecent studies have revealed that conformational changes in the three-dimensional structures of nucleic acids are able to control gene expression in response to specific stimuli.1,2 For instance, riboswitches are RNA elements generally located in untranslated regions of mRNAs that regulate gene expressions in response to intracellular metabolites.3,4 Riboswitches consist of three contiguous RNA elements: an aptamer domain, a switching sequence, and an expression platform. The switching sequence overlaps with both the aptamer domain and the expression platform. Riboswitches that control transcription, translation, or post-transcriptional mRNA processing rely on binding of a specific metabolite to the aptamer domain and subsequent conformational transition that affects the expression platform.3 During the conformational transition, the switching sequence transduces the binding event at the aptamer domain to the expression platform by exchange of base pairing partners. Conformational transitions of RNAs in response to specific molecules have been applied in artificial systems that enable allosteric regulation of RNA functions.5,6 Examples of such artificial systems are allosteric ribozymes and synthetic riboswitches that are used for biosensors and gene regulation © XXXX American Chemical Society

Received: February 25, 2015 Accepted: June 30, 2015

A

DOI: 10.1021/acs.analchem.5b00765 Anal. Chem. XXXX, XXX, XXX−XXX

Article

Analytical Chemistry

Scheme 1. Experimental Scheme for Rational Design of RNA Conformational Switches Applicable for Replacing the Aptamer Domains

Supporting Information. The template containing the random sequence region used for the PCR was present at 2.5 pmol (1.51 × 1012 copies). DNA templates for aptamer-TARs were prepared by primer extension using primes listed in Table S2, Supporting Information. RNAs were prepared by in vitro transcription using ScriptMAX Thermo T7 Transcription Kit (Toyobo) and purified using a denaturing polyacrylamide gel. Selection of RNA. Selection of RNAs that allosterically interact with Tat peptide in response to neomycin was performed by repeating negative and positive selections as described in a previous report.18 For negative selection, the RNA library (20 nM) was mixed with 500 nM biotin-labeled Tat peptide in a selection buffer (20 mM phosphate, pH 7.4, 100 mM NaCl, 1 mM MgCl2, 100 ng/μL tRNA, 50 ng/μL glycogen, and 0.001% Tween 20) in a final volume of 100 μL and incubated at 37 °C for 30 min. An aliquot (62.5 μL) of streptavidin-coated magnetic beads (Dynabeads M-280 Streptavidin: Invitrogen) was added to the reaction mixture. After incubation at 37 °C for 30 min, RNAs that did not bind to Tat peptide were collected from the supernatant and purified by ethanol precipitation. In the positive selection step, the RNA library was incubated in selection buffer (final volume of 100 μL) containing 20 μM neomycin and 10 nM biotin-labeled Tat peptide at 37 °C for 30 min. An aliquot (1.25 μL) of the streptavidin-coated magnetic beads was added to the reaction mixture. After incubation at 37 °C for 30 min, the supernatant was removed and the magnetic beads were washed three times in selection buffer containing 20 μM neomycin. RNAs immobilized on the beads were eluted with 1 M NaCl and purified by ethanol precipitation. DNA complementary to the selected RNAs was synthesized by reverse transcriptase (Rever Tra Ace: Toyobo) using RT primer (Table S1, Supporting Information). The RNA library for the next selection was transcribed from double-stranded DNA amplified from the complementary DNA by PCR using PCR-sense and RT primers (Table S1, Supporting Informa-

Since basic physical research has led to a quantitatively understanding of the thermodynamic parameters of RNA base pairing interactions,13 it is now possible to predict secondary structures of short RNA molecules with accuracy using algorithm such as Mfold.14 This approach enables prediction of RNA secondary structures of arbitrarily selected sequences with the lowest or near lowest free energies. Although the stability of nucleic acids (−ΔG°) depends on various factors in solution such as temperature, ionic strength, and the presence of cosolutes,15 the differences in predicted stabilities between alternative secondary structures before and after the conformational transitions, or the switching energy (ΔΔG°), can guide rational design of functional RNA switches. In addition, tuning of the ΔΔG° values should enable functional modulation of the RNA switches.16,17 In this study, we performed rational design and tuning of functional RNA switches that induce an allosteric RNA− peptide interaction between the trans-activation responsive element (TAR) RNA and trans activator of transcription (Tat) peptide derived from bovine immunodeficiency virus (BIV) in response to a specific trigger molecule. Switch design was based on the ΔΔG° between RNA structures before and after the RNA conformational transition. RNA sequences that bind Tat peptide in response to neomycin were first selected from an RNA library consisting of a neomycin-specific aptamer linked to TAR RNA by a randomized switching region. Then, based on the ΔΔG° value of the selected RNAs, the functional RNA switch responsive to neomycin was converted to switches responsive to other trigger molecules using aptamer domains specific for natural metabolites thiamine pyrophosphate (TPP), S-adenocyl methionine (SAM), and adenine (Scheme 1). The strategy demonstrated here will enable design of functional RNA switches responsive to trigger molecules of interest.

■

EXPERIMENTAL SECTION Preparation of RNA. The DNA template for the RNA library was prepared by PCR using primers listed in Table S1, B

DOI: 10.1021/acs.analchem.5b00765 Anal. Chem. XXXX, XXX, XXX−XXX

Article

Analytical Chemistry tion) using ScriptMAX Thermo T7 Transcription Kit and was purified using the RNeasy Mini Kit (Qiagen). Assay for Interaction between Tat Peptide and Aptamer-TARs. Carboxytetramethylrhodamine (TAMRA)labeled Tat peptide (TMR-Tat: TAMRAGSGRPRGTRGKGRRIRRR), purchased from Scrum Inc., was mixed with RNAs in a buffer containing 20 mM phosphate, pH 7.4, 100 mM NaCl, 0.005% Tween 20, and indicated concentrations of MgCl2, tRNA, and trigger molecules. Fluorescence intensities from TMR-Tat were measured at 25 °C using a microwell plate reader with 555 nm excitation and 600 nm emission. The observed association constant at 25 °C (Kobs25) between TMR-Tat and RNAs was calculated as described previously.11,18 The fluorescence increase in TMR-Tat signal as a function of the concentration of RNAs was fitted assuming an equilibrium of a simple 1:1 binding reaction as follows:

Figure 1. (a) Sequence and secondary structure of neomycin-specific aptamer. (b) Sequence and secondary structure of TAR RNA derived from BIV. (c) Sequence of RNA library with neomycin-specific aptamer in green, randomized switching sequence in red, and TAR RNA in blue. Recognition sites of Af l II and BamH I used in cloning after selection are underlined.

⎧ ⎛ F − Finitial ⎞ ⎪⎛ F = Finitial + ⎜ final ⎟ × ⎨⎜[Tat] + [RNA] ⎝ 2[Tat] ⎠ ⎪⎝ ⎩ +

−

⎞ ⎟ Kobs25 ⎠ 1

Table 1. Kobs25 Values of Wild-Type TAR RNA and RNA Libraries in the Absence and Presence of Neomycin

⎫ 2 ⎛ ⎪ 1 ⎞ ⎜[Tat] + [RNA] + ⎟ − 4[Tat][RNA] ⎬ Kobs25 ⎠ ⎪ ⎝ ⎭

Kobs25 (× 108 M−1) RNA

where F is the fluorescence signal of TMR-Tat at each concentration of RNA, Finitial is the initial fluorescence signal of TMR-Tat without RNA, Ffinal is the fluorescence signal after fluorescence increase is saturated, [Tat] is the total TMR-Tat concentration, and [RNA] is the concentration of RNA. Limits of detection (LODs) of the trigger molecules were calculated using the following equation:

wild-type TAR RNA RNA library G0 G1 G2 G3 G4 G5

LOD = 3.3σ /S

where σ is standard deviation of fluorescence intensity of TMRTat in the absence of trigger molecule, and S is initial slope of fluorescence increase that depends on the concentration of the trigger molecule.

without neomycin

with 20 μM neomycin

relative Kobs25

2.57 ± 0.33

2.48 ± 0.65

0.97

± ± ± ± ± ±

0.83 0.82 2.51 14.4 15.2 15.6

0.19 0.23 0.55 0.18 0.17 0.15

± ± ± ± ± ±

0.03 0.01 0.07 0.01 0.01 0.01

0.16 0.19 1.37 2.63 2.66 2.27

0.01 0.02 0.22 0.28 0.42 0.32

indicates the library after five rounds of selection). The Kobs25 of the G0 library in the absence of the neomycin was 0.19 × 108 M−1. The value was substantially lower than that of wild-type TAR RNA, which was 2.57 × 108 M−1. The results suggest that the 15-nucleotides random sequence was sufficient to disturb formation of the TAR RNA structure in the absence of the neomycin in most of the population. Kobs25 of the G0 library in the presence of 20 μM neomycin (0.16 × 108 M−1) was a slightly lower than that in the absence of neomycin. The ratio of Kobs25 in the presence and absence of the trigger molecule (relative Kobs25) was 0.83 (Table 1). The Kobs25 of wild-type TAR RNA was almost the same irrespective of the presence of neomycin (Figure S2, Supporting Information). These results suggest that the interaction of the G0 library and TMR-Tat was slightly allosterically suppressed, although this was not significant within error. The Kobs25 values in the absence of neomycin increased after the first two rounds of selection but subsequently decreased. The Kobs25 values in the presence of neomycin increased in each of the first three rounds of selection, and Kobs25 values of the G3 to G5 libraries were almost the same as that of the wild-type TAR RNA (2.48 × 108 M−1) (Table 1). The relative Kobs25 value increased by 15.6-fold from G0 to G5 and indicates successful enrichment of RNAs that allosterically interact with Tat peptide in response to neomycin. The functional RNA

■

RESULTS AND DISCUSSION Selection of RNAs That Allosterically Interact with Tat Peptide in Response to Neomycin. Natural functional riboswitches normally consist of a 5′ aptamer domain and a 3′ expression platform. The switching sequence overlaps both the aptamer domain and the 5′ region of the expression platform. In this study, we designed an RNA library consisting of a 5′ aptamer domain specific for neomycin and a 3′ TAR RNA domain connected by a switching sequence of 15-nucleotides that was randomized by mimicking natural riboswitches (Figure 1). Nucleotides at the 3′ region of the aptamer domain and 5′ region of the TAR RNA domain were randomized to provide diversity in the initial RNA secondary structures. Before and after selection, the interaction of the RNA libraries and Tat peptide was evaluated using the fluorescently labeled TMR-Tat, which increases in fluorescence upon binding to wild-type TAR RNA (Figures S1 and S2, Supporting Information). Kobs25 in the absence and presence of neomycin were calculated from the fluorescence signals as a function of concentration (Table 1, G0 indicates the initial library and G5 C

DOI: 10.1021/acs.analchem.5b00765 Anal. Chem. XXXX, XXX, XXX−XXX

Article

Analytical Chemistry

Figure 2. (a) Switching sequences selected from the G5 RNA library. Sequences for neomycin-specific aptamer and TAR RNA are indicated in green and blue, respectively. The region indicated by the arrow under the sequences shows the region conserved in the selected switching sequences. Black lines above the sequences indicate potential base pairs between the conserved sequence and 5′ region of TAR RNA. (b) The predicted most stable secondary structure of G5-a. (c) The predicted secondary structure when formation of the neomycin-specific aptamer and of the TAR RNA conformations is enforced. Gray lines in part b indicate base pairs formed in the conformation shown in part c.

switch responsive to neomycin was drastically enriched during the third selection round as Kobs25 values in the absence and presence of neomycin decreased and increased, respectively, in that round of selection. Since the increase in relative Kobs25 values were almost saturated after the third selection round, the five selection cycles sufficiently enriched the library in RNA switches responsive to neomycin (Figure S1, Supporting Information). Switching Sequences Selected from RNA Libraries. The G5 library was cloned into a plasmid vector and sequenced (Figure 2a). The sequence 5′-GAGCURY-3′ (Figure 2a, indicated by an arrow), where R indicates guanine or adenine and Y indicates cytosine or uracil, was well conserved in the middle of the randomized region. All of the sequences tested individually allosterically interacted with Tat peptide in response to neomycin (data not shown). The DNA template for the G0 library was prepared from 1.51 × 1012 copies of template oligonucleotide, and the first selection was started from 1.20 × 1012 copies of G0 library; these numbers are more than 1000 times of the diversity of 15-nucleotides random sequence (1.07 × 109). Thus, the conserved sequence in the G5 library was significantly enriched from all potential sequences. The most stable secondary structures of the selected RNAs were predicted using Mfold (Figure 2b and Figure S3, Supporting Information). In all of the secondary structures, the conserved regions were predicted to form base pairs with regions of either the neomycin-specific aptamer or the TAR RNA. These secondary structures suggest that in the absence of neomycin the selected switching region base pairs with neighboring aptamer or TAR RNA domain to inhibit the interaction with Tat peptide. In RNAs in which the interaction between the TAR RNA region and Tat peptide can be allosterically induced by neomycin, both the aptamer and the TAR RNA domains must form their natural secondary structures (Figure 1a,b). The

secondary structures of the selected RNAs adopted after binding to neomycin were also predicted using Mfold (Figure 2c and Figure S4, Supporting Information). The structures were predicted by constraining the closing stem region of the neomycin aptamer to form base pairs. The predicted structures of G5-a before and after neomycin binding showed that the conserved sequence, which was forming base pairs with the 5′ region of TAR RNA, exchanged base pairing partners with the region at 5′ of the neomycin aptamer, and the sequence at the 3′ end of the switching region formed base pairs with 3′ end of the TAR RNA (Figure 2b,c). Similar exchanges in base paring partners were also observed in other selected RNAs (Figure S4, Supporting Information); G5-f included one A-C mismatch in the closing stem region of the aptamer. These results indicate that the conserved switching sequence is selected not only due to its ability to disrupt the natural aptamer and TAR RNA structures but also for its ability to stabilize these structures in the presence of the trigger molecule. In the predicted structures after the conformational transition, each of the selected RNAs included one or two more nucleotides in the bulged region of TAR RNA domain than found in the wild-type TAR RNA structure (Figures 1b and 2c and Figure S4, Supporting Information). A modified TAR RNA consisting of the UAGU bulge had greater affinity for Tat peptide than the wild-type TAR RNA (Figure S5, Supporting Information). Therefore, the selected switching sequences also enhanced the interaction between RNA and Tat peptide after the conformational transition induced by neomycin binding. Design of Functional RNA Switches Responsive to Natural Metabolites. Table 2 shows predicted stabilities before (−ΔG°before) and after (−ΔG°after) the RNA conformational transition of G5-a to G5-f and ΔΔG° values for the conformational transitions, which we refer to as the switching energy. ΔΔG° values suggested that the RNA structures were D

DOI: 10.1021/acs.analchem.5b00765 Anal. Chem. XXXX, XXX, XXX−XXX

Article

Analytical Chemistry

UGAGCUACUCGUA-3′, conserved in RNAs selected to respond to neomycin, was used as a switching sequence. The sequences differed in the ligand is length of base pairs formed in the closing stem region of the aptamers after the conformational transition (Figure 3, right-most structures in each panel). None of the RNAs were predicted to form the TAR RNA structure prior to ligand binding. Instead, a structure in which the switching sequence and the 5′ region of TAR RNA were predicted to form base pairs (Figure 3, left-most structures in each panel) as predicted for G5-a (Figure 2b). −ΔG°before and −ΔG°after values of the aptamer-TARs are shown in Table 3. All of these aptamer-TARs were designed to

Table 2. Predicted Stabilities of Selected RNAs before and after RNA Conformational Transitiona RNA

−ΔG°before (kcal mol−1)

−ΔG°after (kcal mol−1)

ΔΔG° (kcal mol−1)

G5-a G5-b G5-c G5-d G5-e G5-f G5-g

23.5 23.2 21.4 23.9 22.1 19.9 18.0

15.6 17.6 14.3 16.6 14.3 7.7 10.0

7.9 5.6 7.1 7.3 7.8 12.2 8.0

a

Stabilities of secondary structures in Figure 2b,c and Figures S3 and S4 in the Supporting Information were predicted using Mfold.14

Table 3. Observed Association Constants between AptamerTARs and TMR-Tat at 25 °C

5.6 to 12.2 kcal mol−1 less stable after the conformational transition than before. Previous studies have demonstrated that binding of neomycin stabilizes the neomycin specific aptamer.19 It is considered that the neomycin binding provided supportive energy that caused the RNA conformational transition and induced the allosteric interaction of the TAR RNA region with the Tat peptide. We hypothesized that functional RNA switches that respond to other trigger molecules can be designed based on the ΔΔG° value. We designed nine aptamer-TARs (TPP-TAR-4bp, TPPTAR-5bp, TPP-TAR-6bp, ade-TAR-4bp, ade-TAR-5bp, and ade-TAR-6bp, SAM-TAR-3bp, SAM-TAR-4bp, SAM-TAR5bp) by replacing the neomycin-specific aptamer domain with the natural aptamer domain of riboswitches specific for thiamine pyrophosphate (TPP), S-adenosyl methionine (SAM), and adenine (ade).20 Figure 3 shows predicted secondary structures of the designed RNAs before and after the RNA conformational transition. The sequence 5′-

RNA TPP-TAR4bp TPP-TAR5bp TPP-TAR6bp ade-TAR4bp ade-TAR5bp ade-TAR6bp SAM-TAR3bp SAM-TAR4bp SAM-TAR5bp

−ΔG°before (kcal mol−1)

−ΔG°after (kcal mol−1)

ΔΔG° (kcal mol−1)

Kobs25 (× 107 M−1)

38.72

24.44

14.28

1.11 ± 0.32

36.03

25.34

10.69

1.57 ± 0.49

35.78

28.54

7.24

4.72 ± 0.57

24.44

14.85

9.59

1.79 ± 0.11

24.83

15.75

9.08

2.65 ± 0.47

24.24

17.25

6.99

6.58 ± 1.79

31.90

23.70

8.20

2.96 ± 0.77

32.20

25.30

6.90

2.46 ± 0.28

31.00

26.20

4.80

5.59 ± 0.82

Figure 3. Predicted secondary structures of RNA conformational switches before (left) and after binding to TPP, SAM, and adenine (right). Aptamer domains specific for TPP, SAM, and adenine are shown in purple, orange, and light blue, respectively. The switching sequence is shown in red and the TAR RNA domain in dark blue. E

DOI: 10.1021/acs.analchem.5b00765 Anal. Chem. XXXX, XXX, XXX−XXX

Article

Analytical Chemistry

In addition, we expected that modulation of ΔΔG° values would tune the sensitivities of the RNA conformational switches because ΔΔG° should reflect the energy barrier for the RNA conformational transition (Figure 4b). Aptamer-TARs with lower ΔΔG° values will transition and cause the allosteric interaction with the Tat-peptide at lower concentrations of target molecules. Detection of a Specific Trigger Molecule by Fluorescence Signal. The aptamer-TARs were mixed with TMRTat in the presence of varying concentrations of cognate and noncognate trigger molecules. TMR-Tat and RNA concentrations were 2.5 and 5 nM, respectively. At these concentrations less than 25% of TMR-Tat forms the RNA− peptide complex in the absence of the target molecule. Thus, comparatively large fluorescence changes can be expected after the aptamer-TARs change conformation in response to the target molecule. The fluorescence signals of TMR-Tat increased in response to the cognate trigger molecules in a concentration-dependent manner for all aptamer-TARs (Figure 5). In the case of aptamer-TARs specific for adenine (ade-TAR4bp, -5bp, and -6bp), fluorescence increases were observed in the presence of SAM and to a lesser extent with increasing concentration of adenine. SAM was likely hydrolyzed to adenine and S-ribosylmethionine during incubation.22 ΔΔG° value of the aptamer-TARs reflected the sensitivity of the aptamer-TARs to the trigger molecules. Aptamer-TARs having lower ΔΔG° values were more sensitive to the concentration of the trigger molecule as expected. Energy input required to cause the RNA conformational transition depends on the ΔΔG° value (Figure 4b); therefore, in the case of the aptamerTARs with high ΔΔG° values, a higher concentration of target molecule was required to induce the allosteric interaction with Tat-peptide. Detection limits of the target molecules were evaluated from the initial slopes of analyses over a range of target concentrations. The detection limits for TPP, adenine, and SAM of TPP-TAR-6bp, ade-TAR-6bp, and SAM-TAR-5bp, respectively, were 4.84, 169, and 140 nM. These values are comparable to or lower than those of previously reported fluorescence sensors based on conformational changes or hybridization switching of nucleic acids.23,24 To fully utilize the intense affinity of the aptamer for sensing, design of aptamerTARs with ΔΔG° value as small as possible will be required. However, RNA conformational switches with small ΔΔG° values increase the Kobs value and basal fluorescence signal in the absence of the target molecule (Figure 5) resulting in an undesirable small amplitude in the fluorescence signal change. To design the RNA conformational switches appropriate for sensing the target molecules, tuning the sensitivity in response to the target concentration range is important. As shown here, ΔΔG° values will provide a useful guide to tune the sensitivity.

have ΔΔG° values close to those of the selected RNAs G5-a to G5-f. The −ΔG°after and ΔΔG° values of the aptamer-TARs were increased and decreased, respectively, with increasing the length of the base pairs in the closing stem region of aptamer domains. Correlation of ΔΔG° and Binding Affinity of RNA− Peptide Interaction. Fluorescence signals of TMR-Tat mixed with the aptamer-TARs were measured at 25 °C (Figure S6, Supporting Information) and Kobs25 values were calculated from the signals observed over a range of RNA concentrations. In general, the Kobs25 values increased with decreasing ΔΔG° value of the aptamer-TARs (Table 3). Interaction energy (ΔG°interaction) of the aptamer-TARs and the modified TAR RNA with a UAGU bulge for the Tat peptide in the absence of the trigger molecule was calculated using the equation ΔG°interaction = −RT ln Kobs25, where R is the gas constant and T is temperature. These values were plotted versus the ΔΔG° values of the aptamer-TARs (Figure 4a). The ΔΔG°

Figure 4. Relationship between ΔΔG° and ΔG°interaction. (a) ΔG°interaction values of TPP-TAR-4bp (purple diamond), TPP-TAR5bp (purple triangle), TPP-TAR-6bp (purple circle), ade-TAR-4bp (blue diamond), ade-TAR-5bp (blue triangle), ade-TAR-6bp (blue circle), SAM-TAR-3bp (orange diamond), SAM-TAR-4bp (orange triangle), SAM-TAR-5bp (orange circle), and modified TAR RNA with a UAGU bulge (pink square) calculated from the dependence of fluorescence increase on RNA concentration in the absence of the trigger molecules (Figures S5 and S6 in the Supporting Information) were plotted vs ΔΔG° values predicted by Mfold. (b) Schematic of reaction studied.

value of the modified TAR RNA was assumed to be zero because it has no switching energy. Figure 4a shows that correlation between ΔG°interaction and ΔΔG° is linear. This correlation suggests that the switching energy determines the basal binding affinity between the RNA and Tat peptide (Figure 4b). We previously reported that the correlation between the predicted and actual stabilities of RNA structures is linear in different ionic strengths.21 Thus, although the slope of the plots in Figure 5a is not 1, because the ΔΔG° values were predicted using thermodynamic parameters based on standard conditions of 1 M NaCl at 37 °C, different from the experimental buffer solution, the ΔΔG° can be used as a reference energy for controlling the basal RNA−peptide interaction in various solution conditions.

■

CONCLUSION Allosteric interactions control many biomolecular functions.25 This type of interaction enables an input molecular interaction to transduce a signal output mediated by other interactions.16 Here, we designed functional RNA switches that allosterically induce an interaction of TAR RNA and Tat peptide in response to various trigger molecules. In these RNAs, the mechanism of signal transduction relies on a simple RNA conformational transition. The key factor for design of the functional RNA switches was the switching energy (ΔΔG°). We demonstrated that rational tuning of an allosteric interaction in response to F

DOI: 10.1021/acs.analchem.5b00765 Anal. Chem. XXXX, XXX, XXX−XXX

Article

Analytical Chemistry

Figure 5. Relative fluorescence intensities of TMR-Tat mixed with aptamer-TARs and varying concentrations of TPP (purple), SAM (orange), and adenine (blue). TMR-Tat (2.5 nM) was mixed with 5 nM of (a) TPP-TAR-4bp, (b) TPP-TAR-5bp, (c) TPP-TAR-6bp, (d) ade-TAR-4bp, (e) adeTAR-5bp, (f) ade-TAR-6bp, (g) SAM-TAR-3bp, (h) SAM-TAR-4bp, and (i) SAM-TAR-5bp in a buffer containing 20 mM phosphate, pH 7.4, 100 mM NaCl, 5 mM MgCl2, 100 ng/μL tRNA, and 0.005% Tween 20 at 25 °C. Values are mean ± S. D. of triplicate experiments.

■

the trigger molecules was possible. Simple addition of base pairs in the closing stem of the aptamer domain, which decreased the ΔΔG° value, controlled the sensitivity of the fluorescence detection of the target molecule. That sensitivities depend on the ΔΔG° values suggest that controlling and expanding the dynamic range of molecular detection will be possible by designing RNA switches with different ΔΔG° values.26 As aptamers are known that bind to a variety of natural and artificial molecules,27 the simple strategy demonstrated here can be applied for construction of biosensors and gene regulation systems mediated by interaction of TAR RNA and Tat peptide10,11 and also by other intermolecular interactions, such as interactions of RNAs with fluorescent probes, Dicer, the RNA-induced silencing complex (RISC), RNases, and other proteins known to affect gene expression.1,6,24,28

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Fax: (+) 81-78-303-1495. Author Contributions

The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This work was supported in part by Grants-in-Aid for Scientific Research and MEXT (Japan)-Supported Program for the Strategic Research Foundation at Private Universities, The Hirao Taro Foundation of KONAN GAKUEN for Academic Research. We thank Nobuaki Hattori, Misa Kinoshita, and Yuki Miyashige for their help with experiments.

ASSOCIATED CONTENT

S Supporting Information *

■

DNA oligonucleotides for construction of RNA library, fluorescence intensities of TMR-Tat mixed with the G0 library to the G5 library, wild-type TAR RNA, modified TAR RNA with UAGU bulge and aptamer-TARs, and secondary structures of G5-b to G5-g before and after RNA conformational transition. The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/ acs.analchem.5b00765.

REFERENCES

(1) Kedde, M.; van Kouwenhove, M.; Zwart, W.; Oude Vrielink, J. A.; Elkon, R.; Agami, R. Nat. Cell Biol. 2010, 12, 1014−20. (2) Ray, P. S.; Jia, J.; Yao, P.; Majumder, M.; Hatzoglou, M.; Fox, P. L. Nature 2009, 457, 915−9. Miyazaki, Y.; Garcia, E. L.; King, S. R.; Iyalla, K.; Loeliger, K.; Starck, P.; Syed, S.; Telesnitsky, A.; Summers, M. F. J. Mol. Biol. 2010, 396, 141−52. Grundy, F. J.; Henkin, T. M.

G

DOI: 10.1021/acs.analchem.5b00765 Anal. Chem. XXXX, XXX, XXX−XXX

Article

Analytical Chemistry Crit. Rev. Biochem. Mol. Biol. 2006, 41, 329−38. Dethoff, E. A.; Chugh, J.; Mustoe, A. M.; Al-Hashimi, H. M. Nature 2012, 482, 322−30. (3) Barrick, J. E.; Breaker, R. R. Genome Biol. 2007, 8, R239. Serganov, A.; Nudler, E. Cell 2013, 152, 17−24. (4) Montange, R. K.; Batey, R. T. Annu. Rev. Biophys. 2008, 37, 117− 33. Serganov, A. Curr. Opin. Struct. Biol. 2009, 19, 251−9. (5) Vinkenborg, J. L.; Karnowski, N.; Famulok, M. Nat. Chem. Biol. 2011, 7, 519−27. Wieland, M.; Fussenegger, M. Curr. Opin. Biotechnol. 2010, 21, 760−5. Kang, H. J.; Kendrick, S.; Hecht, S. M.; Hurley, L. H. J. Am. Chem. Soc. 2014, 136, 4172−4185. Kendrick, S.; Kang, H. J.; Alam, M. P.; Madathil, M. M.; Agrawal, P.; Gokhale, V.; Yang, D.; Hecht, S. M.; Hurley, L. H. J. Am. Chem. Soc. 2014, 136, 4161−71. (6) An, C. I.; Trinh, V. B.; Yokobayashi, Y. RNA 2006, 12, 710−716. Babiskin, A. H.; Smolke, C. D. Nucleic Acids Res. 2011, 39, 5299−311. (7) Berens, C.; Groher, F.; Suess, B. Biotechnol. J. 2015, 10, 246−57. Link, K. H.; Breaker, R. R. Gene Ther. 2009, 16, 1189−201. (8) Seetharaman, S.; Zivarts, M.; Sudarsan, N.; Breaker, R. R. Nat. Biotechnol. 2001, 19, 336−41. (9) Zhou, W.; Huang, P. J.; Ding, J.; Liu, J. Analyst 2014, 139, 2627− 2640. Ellington, A. D.; Szostak, J. W. Nature 1990, 346, 818−22. Tuerk, C.; Gold, L. Science 1990, 249, 505−10. (10) Endoh, T.; Funabashi, H.; Mie, M.; Kobatake, E. Anal. Chem. 2005, 77, 4308−14. Endoh, T.; Mie, M.; Funabashi, H.; Sawasaki, T.; Endo, Y.; Kobatake, E. Bioconjugate Chem. 2007, 18, 956−62. Endoh, T.; Shintani, R.; Mie, M.; Kobatake, E.; Ohtsuki, T.; Sisido, M. Bioconjugate Chem. 2009, 20, 2242−6. (11) Endoh, T.; Sugimoto, N. ChemBioChem 2011, 12, 1174−8. (12) Endoh, T.; Sugimoto, N. ChemMedChem 2014, 9, 2045−8. (13) Laing, C.; Schlick, T. J. Phys.: Condens. Matter 2010, 22, 283101. Mathews, D. H.; Sabina, J.; Zuker, M.; Turner, D. H. J. Mol. Biol. 1999, 288, 911−40. (14) Zuker, M. Nucleic Acids Res. 2003, 31, 3406−15. (15) Nakano, S.; Sugimoto, N. In Bottom-Up Nanofabrication: Supramolecules, Self-Assemblies and Organized Films; Ariga, K., Nalwa, H. S., Eds.; American Scientific Publishers: Stevenson Ranch, CA, 2009; pp 191−215. Tateishi-Karimata, H.; Nakano, S.; Sugimoto, N. Quantitative Analyses of Nucleic Acid Stability Under the Molecular Crowding Condition Induced by Cosolutes. In Current Protocols in Nucleic Acid Chemistry; John Wiley & Sons, Inc.: New York, 2013; Supplement 53, Chapter 7, Unit 7.19, 7.19.1−7.19.17. (16) Vallee-Belisle, A.; Ricci, F.; Plaxco, K. W. Proc. Natl. Acad. Sci. U. S. A. 2009, 106, 13802−7. (17) Porchetta, A.; Vallee-Belisle, A.; Plaxco, K. W.; Ricci, F. J. Am. Chem. Soc. 2012, 134, 20601−4. (18) Endoh, T.; Sugimoto, N. PLoS One 2013, 8, e60222. (19) Weigand, J. E.; Schmidtke, S. R.; Will, T. J.; Duchardt-Ferner, E.; Hammann, C.; Wohnert, J.; Suess, B. Nucleic Acids Res. 2011, 39, 3363−72. (20) Edwards, T. E.; Ferre-D’Amare, A. R. Structure 2006, 14, 1459− 1468. Lu, C.; Smith, A. M.; Fuchs, R. T.; Ding, F.; Rajashankar, K.; Henkin, T. M.; Ke, A. Nat. Struct. Mol. Biol. 2008, 15, 1076−83. Serganov, A.; Yuan, Y. R.; Pikovskaya, O.; Polonskaia, A.; Malinina, L.; Phan, A. T.; Hobartner, C.; Micura, R.; Breaker, R. R.; Patel, D. J. Chem. Biol. 2004, 11, 1729−41. (21) Nakano, S.; Fujimoto, M.; Hara, H.; Sugimoto, N. Nucleic Acids Res. 1999, 27, 2957−2965. (22) Parks, L. W.; Schlenk, F. J. Biol. Chem. 1958, 230, 295−305. (23) Lau, P. S.; Coombes, B. K.; Li, Y. Angew. Chem., Int. Ed. 2010, 49, 7938−42. (24) Paige, J. S.; Nguyen-Duc, T.; Song, W.; Jaffrey, S. R. Science 2012, 335, 1194. (25) Changeux, J. P. Protein Sci. 2011, 20, 1119−24. Lindsley, J. E.; Rutter, J. Proc. Natl. Acad. Sci. U. S. A. 2006, 103, 10533−5. (26) Idili, A.; Vallee-Belisle, A.; Ricci, F. J. Am. Chem. Soc. 2014, 136, 5836−9. (27) Thompson, A. D.; Dugan, A.; Gestwicki, J. E.; Mapp, A. K. ACS Chem. Biol. 2012, 7, 1311−20. Nussinov, R.; Tsai, C. J.; Ma, B. Annu. Rev. Biophys. 2013, 42, 169−89.

(28) Saito, H.; Fujita, Y.; Kashida, S.; Hayashi, K.; Inoue, T. Nat. Commun. 2011, 2, 160.

H

DOI: 10.1021/acs.analchem.5b00765 Anal. Chem. XXXX, XXX, XXX−XXX