This document is downloaded from DR‑NTU (https://dr.ntu.edu.sg)Nanyang Technological University, Singapore.

Organic additives stabilize RNA aptamer bindingof malachite green

Zhou, Yubin; Chi, Hong; Wu, Yuanyuan; Marks, Robert S.; Steele, Terry W. J.

2016

Zhou, Y., Chi, H., Wu, Y., Marks, R. S., & Steele, T. W. J. (2016). Organic additives stabilizeRNA aptamer binding of malachite green. Talanta, 160, 172‑182.

https://hdl.handle.net/10356/80594

https://doi.org/10.1016/j.talanta.2016.06.067

© 2016 Elsevier. This is the author created version of a work that has been peer reviewedand accepted for publication by Talanta, Elsevier. It incorporates referee’s comments butchanges resulting from the publishing process, such as copyediting, structural formatting,may not be reflected in this document. The published version is available at:[http://dx.doi.org/10.1016/j.talanta.2016.06.067].

Downloaded on 14 Dec 2021 13:05:34 SGT

* Abbreviations

MG, malachite green; MGOH, malachite green carbinol base; MGA, malachite green

aptamer; ACN, acetonitrile; DMSO, Dimethyl sulfoxide; EtOH, ethanol.

1

Organic additives stabilize RNA aptamer binding of 1

malachite green 2

Yubin Zhoua, Hong Chia, Yuanyuan Wua, Robert S. Marksa,b, Terry W.J. Steelea,* 3

a School of Materials Science & Engineering, College of Engineering, Nanyang 4

Technological University, 50 Nanyang Avenue, Singapore 639798, Singapore 5

b Department of Biotechnology Engineering, Faculty of Engineering Sciences, Ben 6

Gurion University of the Negev, P.O. Box 653, Beer Sheva 84105, Israel 7

* Corresponding Author: 8

Terry W.J. Steele: wjsteele@ntu.edu.sg, 9

Tel.:65-65927594, fax: 65-67909081. 10

Abstract 11

Aptamer-ligand binding has been utilized for biological applications due to its specific 12

binding and synthetic nature. However, the applications will be limited if the binding 13

or the ligand is unstable. Malachite green aptamer (MGA) and its labile ligand 14

malachite green (MG) were found to have increasing apparent dissociation constants 15

(Kd) as determined through the first order rate loss of emission intensity of the MGA-16

MG fluorescent complex. The fluorescent intensity loss was hypothesized to be from 17

the hydrolysis of MG into malachite green carbinol base (MGOH). Random screening 18

organic additives were found to reduce or retain the fluorescence emission and the 19

calculated apparent Kd of MGA-MG binding. The protective effect became more 20

apparent as the percentage of organic additives increased up to 10% v/v. The 21

mechanism behind the organic additive protective effects was primarily from a ~5X 22

increase in first order rate kinetics of MGOHMG (kMGOHMG), which significantly 23

changed the equilibrium constant (Keq), favoring the generation of MG, versus 24

MGOH without organic additives. A simple way has been developed to stabilize the 25

apparent Kd of MGA-MG binding over 24 hours, which may be beneficial in 26

stabilizing other triphenylmethane or carbocation ligand-aptamer interactions that are 27

susceptible to SN1 hydrolysis. 28

Keywords 29

2

Malachite green aptamer, malachite green, ligand hydrolysis, adaptive binding, 30

equilibrium 31

1. Introduction 32

Oligonucleotide aptamers have been widely studied in biological applications such 33

as biosensing [1-4], cell imaging [5, 6], drug/siRNA delivery [7, 8] and in vivo 34

therapeutics [9], due to their specific recognition, adaptive binding and synthetic 35

nature [10, 11]. Compared to antibodies, aptamers possess advantages such as low 36

molecular weights, facile modification and low immunogenicity [12, 13]. However, 37

lability of ligands affects the binding stability with aptamers, limiting their 38

applications [14]. In this study, malachite green aptamer (MGA) with its labile ligand 39

malachite green (MG) was chosen as a model system to study the aptamer-ligand 40

binding stability over time based on MG’s enhancement of fluorescence upon aptamer 41

adaptive binding. The MGA-MG complex is fluorescent due to an extended π-system 42

of the MG that is forced to become planar by the dynamic binding of the MGA [15, 43

16]. 44

MGA has been utilized for sensing the banned fungicide MG [17] with a limit of 45

detection down to 5 nM [15]. Detection of such low concentrations is due to the 46

unique enhanced quantum yield (2000X) upon MGA adaptive folding around MG as 47

seen in Fig. 1B [16]. MG is currently employed in as a dye in paper, leather, inks, 48

waxes, and textiles and is known for its use in algal inhibition in fish farming, 49

although this practice has been banned in most developing countries due to acute 50

toxicities in fish and humans [18]. MG belongs to a large family of triphenylmethane 51

compounds that have considerable economic importance, as they are used in the 52

printing industry (to brighten carbon black), sensor applications and as chrome dyes 53

for wool and color formers in self-replicating/thermoresponsive papers. The latest 54

known commercial statistics shows over 25 kilotons of product per year of various 55

triphenylmethanes with unknown amounts ending in industrial waste waters and 56

treatment plants [19]. Thus there exists a need for development of stable, fast, and 57

accurate sensor development towards triphenylmethane derivatives, all of which are 58

more or less susceptible to hydrolysis attack. Detection assays must then incorporate 59

aspects of portability, specificity, and sensitive quantitation of MG in a range of 60

aqueous environments. Additionally, it has been reported that MG can convert to its 61

derivatives such as malachite green carbinol base (MGOH) [20, 21] and 62

3

leucomalachite green [22], which affects the sensitivity and specificity of the MGA-63

MG binding complex (Fig. 1A). 64

Fig. 1

65

However, assessments of temporal Kd of MG to MGA and prevention of MG 66

hydrolysis (and related triphenylmethane derivatives) have not been investigated to 67

our knowledge, despite the economic importance of detecting triphenylmethane 68

derivatives in dye industrial wastewaters and within the food supply. 69

In this work, the adaptive binding behavior of MG to its RNA aptamer was 70

assessed by for temporal fluorescence with various organic additives added to prevent 71

the decreasing fluorescence intensities. The apparent dissociation constant, Kd was 72

determined to be comparable to the reported value of 200 nM [23] at time 0. 73

However, the fluorescence intensity and the apparent Kd decreased as a function of 74

time, but stabilized after 5 h, which suggested a change in equilibrium in the RNA 75

aptamer or MG. We noted that when organic additives were initially employed to 76

dissolve various triphenylmethane dyes, the Kd was less sensitive to time. 77

Investigation of the organic additives allowed more sensitive and reproducible 78

determination of MG, which is a simple strategy with little to no impact on Kd. The 79

increase in stability minimizes deviations in triphenylmethane detection and facilitates 80

the further development of RNA aptamer-based applications. More importantly, the 81

discovery of kinetic stabilization of the MGA-MG model may provide a simple 82

method of stabilization for triphenylmethane, triarylmethane, or other carbocation-83

susceptible ligands that are known to undergo hydrolysis. 84

2. Materials and methods 85

2.1. Materials 86

Desalted RNA aptamer (MGA) (5’-87

GGAUCCCGACUGGCGAGAGCCAGGUAACGAAUGGAUCC-3’) was purchased 88

from AITbiotech Pte Ltd Singapore. Malachite green oxalate (MG) and malachite 89

green carbinol base (MGOH) were purchased from Sigma Aldrich, Singapore. 90

Dimethyl sulfoxide (DMSO) was purchased from Alfa Aesar, USA. Ethanol (EtOH) 91

was purchased from Fisher Scientific, Singapore. HPLC grade acetonitrile (ACN) was 92

purchased from Tedia, USA. Polystyrene microplates (96-well) were purchased from 93

Corning, USA. All chemicals and materials were used as received. 94

4

2.2. Kd determination of MG to MGA 95

Binding buffer (pH = 6.7 unless otherwise specified) which consisted of 5.65 mM 96

NaH2PO4, 4.35 mM Na2HPO4, 10 mM MgCl2 and 1 mM NaCl was prepared 97

according to the literature [23, 24]. The effect of pH on binding was studied and the 98

results (Fig. S1) indicated that the suitable range for binding was pH 6-8.5. Thus, pH 99

at 6.7 was chosen for comparison to literature [23]. Malachite Green Aptamer (MGA) 100

was denatured at 95 ℃ for 1 min prior to the binding experiment. The binding of MG 101

to MGA was determined by fluorescence enhancement [16]. Kd was determined 102

varying concentrations of MGA with MG concentration (150 nM) held constant in 103

binding buffer with or without organic additive present at 25 ℃. MG fluorescence was 104

immediately assayed or time delayed from 0.5-24 h in 96-well polystyrene plates with 105

a Tecan Microplate spectrofluorometer (Infinite M200, Tecan Asia PTE LTD, 106

Singapore) at ex/em setting of 620 nm/656 nm. Based on the MG fluorescence 107

enhancement upon aptamer binding described above [16], fluorescence intensity of 108

MG (in fixed concentration) rises with increasing MGA concentration, and reaches a 109

plateau when 100% MG are bound. Thus, the fluorescence signal of MG against 110

added MGA concentration can be utilized for monitoring the binding. By combining 111

the change in fluorescence intensity with the known concentrations of MGA and MG 112

in the binding system, dissociation constant Kd value can be determined by the non-113

linear fitting in section 2.3 based on the literature by Wang et al [25, 26]. 114

Additionally, to test the recovery effect of organic additive on fluorescence (of 115

MGA-MG complex) after reduction, DMSO was added into the MGA-MG binding 116

complex 5 h after the binding was performed. 117

2.3. Fitting of dissociation constant Kd 118

The equation (Equation 1) used for fitting of Kd is displayed below, which is based 119

on previous investigations by Wang et al [25, 26].: 120

F = F0 + (Fmax − F0) ×([MG]0 + [MGA] + 𝐾d) − √([MG]0 + [MGA] + 𝐾d)2 − 4[MG]0 × [MGA]

2 × [MG]0 (1)

F and F0 are fluorescence units (FLU) in the presence and absence of aptamer, 121

respectively. Fmax is the FLU value upon 100% binding to aptamer. [MG]0 and 122

[MGA] are the initial concentrations within the binding system, and the Kd is the 123

dissociation constant. Non-linear fitting and statistical analysis were performed by 124

OriginPro 9.1. 125

2.4. HPLC-DAD study on stability of MG 126

5

An Agilent 1200 series high-performance liquid chromatography system with 127

diode array detector (DAD) was used to monitor the stability of MG. A UltisilTM C18 128

column (Welch Materials Inc.) of 5 µm and 4.6 × 250 mm was used for 129

chromatographic separation. The system was operated with a mobile phase flow rate 130

of 1.0 mL•min-1, and the diode array detector was set to continuously wavelength 131

scans of 200 to 800 nm with λmax of 265 and 620 nm for MGOH and MG, 132

respectively. Fresh MG (30 µM) in binding buffer was injected (20 µL) at 0 h and 24 133

h and MGOH (30 µM) as a standard in acetonitrile was injected (20 µL) at 0h. Elution 134

profile was 50 % mobile phase A (25 mM ammonium acetate, pH 6.8) / 50 % mobile 135

phase B (acetonitrile) for 5 min, a gradient to 15 % A / 85% B from 5 to 15 min, a 15 136

% A / 85% B from 15 to 18 min, followed by a ramp back to initial conditions from 137

18 to 21 min and held for 4 min thereafter. 138

2.5. Kinetic study on conversion between MG and MGOH 139

MG and MGOH were dissolved in pH 6.7 binding buffer with or without 10% 140

(V/V, the same to all description on percentage of organic additive in this article) 141

organic additive respectively. Then the sample was transferred to a quartz cuvette and 142

placed in a Varian Cary 50 UV-Vis spectrophotometer. The UV-Vis spectra were 143

recorded every 5 min for 12 hours. Absorbance maximum of MG at 618 nm was used 144

to calculate the change of MG. The decrease in MG at MGMGOH and the increase 145

in MG at MGOHMG were fitted by OriginPro 9.1 with first order equation. 146

2.7 Statistics 147

All statistics were performed with OriginPro 9.1 (OriginLab Corporation, 148

Northampton, MA 01060 USA). Linear regressions R2 were determined by ordinary 149

least squares. Statistical differences between two groups were performed by Student’s 150

t-test with significance level P < 0.05 (two-tailed test). 151

3. Results 152

3.1. Fluorescence intensity and apparent affinity of MGA-MG diminish over time 153

Based on the fluorescence enhancement of MG, the initial dissociation constant (Kd) 154

is determined to be 200 nM as seen in Fig. 2A and 2B, which compares well to the 155

analysis by Costa et al [23]. However, fluorescence intensity and apparent Kd of MG 156

decrease with time, but reach a plateau at 5 h as shown in Fig. 2A. At 24 h, 40.8% of 157

the fluorescence intensity at maximum [MGA]/ratio (6000 nM, MGA:MG = 40:1) 158

remains while the apparent Kd increases > 10000 nM. The fluorescence intensity at 159

maximum [MGA]/ratio undergoes a first order decay with k = 2.5×10-4 s-1 (Fig. 2C). 160

6

This decrease is attributed to the hydrolysis of MG in aqueous environment. Minute 161

changes in pH are also ruled out, as only minor shifts in Kd are seen across pH 6-8.5 162

(Fig. S1). Thus, the shifts in Kd are attributed to the flattened asymptote to linear-163

profiles after FLU decay. 164

Fig. 2

165

3.2. Organic additives prevent fluorescence decay while retaining Kd 166

Binding buffer with 3%-10% v/v acetonitrile (ACN), dimethyl sulfoxide (DMSO) 167

or ethanol (EtOH), additives typically employed to increase solubility of organic 168

ligands, is mixed into the MGA-MG binding experiments (Fig. 3) towards assessment 169

of temporal fluorescent intensity. The MGA-MG fluorescence displays observable 170

increases across the largest concentrations/ratios of MGA:MG. With organic 171

addtives, > 9000 FLU are observed at [MGA] above 3000 nM as seen in Fig. 3A-C, 172

about 40% higher than 5500 FLU at time 0 h in Fig. 2A. This suggests MGMGOH 173

conversion is present even before the time 0 h measurements in Fig. 2A were taken. 174

The addition of organic additives changes the equilibrium constant, decay kinetics, or 175

both. In other words, known stocks of [MG] can never truly be prepared in aqueous 176

solutions, because of the fast hydrolysis equilibrium. To further quantitate the effects 177

of organic additives, the apparent Kd values are calculated and presented in Fig. 3D. 178

The apparent Kd shows a strong correlation with percentage of ACN, but weak to 179

moderate correlation in either DMSO or EtOH. Thus, only ACN weakens the overall 180

Kd, while DMSO and EtOH show little to no changes in Kd. 181

Fig. 3

182

3.3. Organic additives retain over >80% of fluorescence intensity after 24 h 183

The stability of MGA-MG binding after 24 h in the organic additive-buffers is 184

investigated towards preventing fluorescence decay. Fig. 4A-C illustrates the MGA-185

MG Kd fittings at 0 h and 24 h in the presence of ACN from 3 - 10% v/v. Similar 186

trends are found for DMSO and EtOH (data not shown). Fig. 4B,C summarize the 187

apparent Kd and FLUs at the highest [MGA]/ratio at 0 h and 24 h vs. conc. of organic 188

additive. FLU and apparent Kd are less prone to decay or increase, respectively, as the 189

organic additive increases, suggesting a change in Keq, decay kinetics, or both. For 190

example, with no additives, the FLU at maximum [MGA]/ratio decreases to 40.8% of 191

7

the original value and the apparent Kd increases ~ 65-fold after 24 h, Fig. 2A. The 192

addition of the 8-10% v/v DMSO additive shows no statistical difference in FLU units 193

after 24 h (at the 40:1 ratio), but a slight shift in the overall binding profile (which 194

incorporates all FLU ratios), doubles the Kd. EtOH has the inverse effect. Overall, 195

organic additives cause considerable shifts on the FLU decay kinetics as shown in 196

Table 1. The first order decay kinetics, as measured by FLU, decreased as the v/v% 197

increased from 3 to 10%, which indicates organic additives shifted the Keq 198

(MGMGOH) towards MG, and increasing the overall conc. of the fluorescent 199

MGA-MG complex. 200

Fig. 4

201

Table 1

202

3.4. Loss of MG fluorescence is attributed to hydrolysis, formation of MGOH 203

HPLC in-line with a diode array absorbance detector allows the quantification and 204

identification of MG derivatives through their absorbance spectra. MG (λmax = 620 205

nm) and MGOH (λmax = 265 nm) absorbance spectra are displayed in the Fig. 5A,B 206

insets. HPLC chromatogram overlays of MG 0 h, MG 24 h, MGOH standard is shown 207

in Fig. 5A. Under the HPLC separation conditions (see Methods), MG 0 h displays a 208

major peak at 11.1 min with a minor observed peak at 18.2 min attributed to the 209

MGOH (malachite green carbinol base). 210

Fig. 5

211

The MG peak decreases 36.3 % after 24 h (Fig. 5A) while the MGOH peak 212

increases 38.6%, nearly the same amount as seen in Fig. 5B. The MGOH peak in the 213

MG 0h sample has 19% AUC compared to the 30 µM standard MGOH; in other 214

words 5.7 µM MGOH is detected in the (as prepared) 30 µM MG 0h sample. In the 215

MG 24h sample, MGOH conc. nearly triples to 17.1 µM, and there is no observed 216

formation of other minor peaks during 0-25 min elution. By plotting 100% MG at 217

10,000 FLU (Fig. 4C), 81% MG at 5,500 FLU (Fig. 2A and 5B), and 43% MG at 218

2,000 FLU (Fig. 2A and 5A,B), a reasonable linear fit of R2 = 0.86 is noted. 219

3.5. Organic additives shift the Keq towards MG formation 220

8

The kinetics of conversion (without aptamer) of MGMGOH and MGOHMG 221

with and without 10% v/v organic additive (ACN, DMSO, EtOH) is conducted by 222

following the lambda max of MG, as seen in Fig. 6. The recorded wavelength scans 223

from 400 nm to 800 nm are seen in Fig. S2, which show similar results as noted in the 224

HPLC characterization above, that MG cannot be prepared pure in aqueous solutions 225

due to a t1/2 of ~19 min (ln 2/6.2× 10-4 s-1), while MGOH can (t1/2 = 2.8 h). In buffer 226

without any organic additive, the conversion of MGMGOH occurs nine times faster 227

(Fig. 6A, kMGMGOH = 6.2× 10-4 s-1) than the conversion of MGOHMG (Fig. 6B, 228

kMGOHMG = 0.69× 10-4 s-1). Additionally, the conversion of MGMGOH reaches 229

equilibrium after 2 h, while the conversion of MGOH to MG does not reach 230

equilibrium even after 12 h, as seen in Fig. 6B. When MGMGOH in the presence 231

of 10% organic additives (Fig. 6A), the kMGMGOH is retarded by 23%, 9%, and 5% 232

for ACN, DMSO, and EtOH, respectively. From the FLU decay kinetics, the aptamer 233

retards the kMGMGOH by 60% with no additives present (Fig. 2C and 6A). 234

Surprisingly, the conversion of MGOHMG in the presence of 10% organic 235

additives (Fig. 6B) displays a comparably large shift in kMGOHMG. Organic additives 236

increased kMGOHMG over buffer by 360%, 450%, and 460% for ACN, DMSO, and 237

EtOH, respectively. This acceleration on MGOHMG conversion by organic 238

additives results in a large shift in Keq compared to the MG in buffer or buffer + 239

MGA. This shift in Keq is summarized in Table 2. To investigate any effects the 240

aptamer has on kMGOHMG, various ratios of MGA were mixed with freshly prepared 241

MGOH and the MGA-MG fluorescent complex was monitored over time under 242

various MGOH:MGA ratios. At high ratios of 20 and 40:1, there was no statistical 243

difference in the FLU rate of MGA-MG fluorescent complex formation versus the 244

ABS rate of MG formation, as seen in Fig. S3. Thus, high aptamer ratios have no 245

effect on kMGOHMG kinetics. 246

Table 2 summarizes kMGMGOH in different additives and in the presence and 247

absence of aptamer (MGA). The molecular properties of water and organic additives 248

are also listed for any evident correlations. Without aptamer, organic additives had 249

only a slight effect on MGMGOH conversion. However, in the presence of 250

aptamer, the conversion was 40% slower, indicating the protective environment the 251

aptamer provides the MG in terms of hydrolysis. More importantly, this protection 252

effect combined with the organic additives provides a synergistic outcome towards the 253

stable formation of the MGA-MG fluorescent complex. The adjusted kMGMGOH as 254

9

measured by FLU was 500% slower in 10% ACN and 2950% slower with 10% EtOH. 255

No significant decay in fluorescence was observed with the addition of 10% DMSO, 256

preventing any measurement of kMGMGOH under these conditions. Additionally, 257

Pearson’s R values indicate strong correlations between kinetic constant and intrinsic 258

properties of the organic additives. We speculate that this may allow prediction of 259

even more effective additives. Without aptamer, no distinct correlation between 260

kinetic constant and intrinsic properties is observed (0.5 > R > -0.5). However, in the 261

presence of aptamer, kinetic constant has a strong correlation with molecular polar 262

surface area (MPSA) (R > 0.99) and a strong inverse correlation to additive volume (R 263

< -0.99). The 10% DMSO is excluded in calculation due to its unavailable kinetic 264

constant. The lack of decay follows the strong correlation of the intrinsic molecular 265

properties of DMSO (small MPSA and relatively large volume). 266

Fig. 6

267

Table 2

268

3.6 Organic additives reverse decay of fluorescence through MGOH dehydration 269

The shift in Keq values as seen in Table 2 suggests that organic additives can 270

immediately change the reaction kinetics to favor MG as the major equilibrium 271

product. This hypothesis is challenged by the addition of 3-10% v/v DMSO to 272

MGA:MG aqueous buffer solutions that have reached a stable FLU after 5 h of 273

incubation. Fig. 7 displays FLU recovery of 3 and 5% DMSO after 1 and 12 h 274

incubation. The results support an immediate shift in Keq by organic additives, with a 275

minimum of 5% v/v DMSO to recover the 0 h fluorescence intensity after 12 h. 276

Higher conc. of DMSO yielded FLU values that approached 10,000 FLU units (at 277

MGA:MG ratios > 20:1), similar to the values seen in Fig. 4C (data not shown). 278

Fig. 7

279

4. Discussion 280

MG and its aptamer have been utilized for various applications such as cell 281

imaging and probing for RNA nanoparticles based on enhanced fluorescence of MG 282

after induced binding [14, 27, 28]. However, due to the known instability of MG in 283

10

aqueous buffers, MG and MGA-MG binding will decrease over time. However the 284

temporal Kd or fluorescent intensity of the MGA-MG complex is seldom reported. 285

Wang et al. has experimentally and theoretically argued that MGA prevents MG 286

hydrolysis by measuring absorbance maxima in bound vs. unbound MG—no data on 287

changes in fluorescence was described [29]. Decreased hydrolysis kinetics was also 288

observed by FLU measurements where MGA is in buffer and buffer + organic 289

additives, but our work differs in conclusions drawn. Additionally, we also monitored 290

the apparent Kd of MGA-MG over time, and found that fluorescence intensity 291

decreased while apparent Kd increased over time. Moreover, we report for the first 292

time how the presence of common organic additives prevented temporal shifts in 293

fluorescence intensity and Kd within a 24 h timeframe. This stabilization is beneficial 294

for MGA-based applications towards accurate and repeatable detection assays based 295

on MG’s fluorescence enhancement by MGA [30, 31]. Finally, a synergistic 296

stabilization mechanism is proposed based on both fluorescence and absorbance 297

kinetics combined with previous reports on MG kinetics in aqueous environments [32, 298

33]. 299

4.1. Hydrolysis of MG to MGOH prevents aptamer binding 300

The Kd was determined by MG’s unique fluorescence enhancement upon induced 301

aptamer binding, but when the FLU signal decays, the dissociation constant will be 302

directly affected. The fluorescent response of MG upon binding to MGA is illustrated 303

in Fig. 2. In Fig. 2A, the increase in apparent Kd and subsequent decrease in 304

fluorescence intensity through 24 h (at 40:1 ratio) indicate an unstable association 305

between MGA and MG. In two preliminary experiments, MG and MGA were 306

separately added into the MGA-MG complex respectively after 24h to isolate the 307

cause of fluorescence decay. The addition of fresh MG recovered the fluorescence 308

intensity instantly, but continued to decay. The addition of fresh MGA displayed no 309

fluorescence recovery (Fig. S4). Such behavior indicated that the decrease in FLU 310

was due to events centered on MG but not the aptamer, ruling out RNA misfolding or 311

degradation. To investigate integrity of MGA in binding buffer (containing 10 mM 312

Mg2+) over 48 h, gel electrophoresis on MGA over time was performed. As seen in 313

Fig. S5, no obvious change in the MGA band was observed from 0 h to 48 h. The 314

binding interactions of RNA aptamers to ligands are dynamic and do not follow a 315

typical ‘lock and key’ static enzyme mechanism, but follows an ‘induced fit’ 316

mechanism that more closely resembles RNA folding around the ligand, where both 317

11

the RNA and ligand have induced changes in structure [34]. Thus, a clear explanation 318

of how RNA folding changes after a ligand hydrolysis event hasn’t been explored, but 319

creates another level complexity best removed if possible, which is yet another reason 320

for the addition of the organic additives shown herein. HPLC investigations display 321

only MG and MGOH as the major reactants and products. The 1:1 conversion of MG 322

peak to MGOH peak combined with the UV profile spectra confirms the conversion 323

of MG to MGOH with no other peaks of interests detected [21, 35, 36]. No other MG 324

variants (i.e. leucomalachite green, dimers) were detected, nullifying any 325

hypothesized side reactions. Another hypothesis, that aptamer prevented 326

MGOHMG dehydration, was proven false, as pure MGOH was incubated with 327

varying ratios of aptamer, with increasing fluorescence noted as MG was formed at 328

the predicted kinetics. The observation that FLU decay is completely reversible by 329

organic additives (see Fig. 7) speculates that MG undergoes a time dependent change 330

in electronic resonance or charge redistribution within the aptamer binding pocket. 331

This would lower the quantum yield or shift its fluorescent planar ‘induced fit’ 332

configuration. As seen in Fig. 8, we speculate that the hydrolysis intermediate, MG-333

OH2, is responsible for the FLU decay through electronic and charge redistribution 334

within the aptamer binding pocket, as the C+-OH2 carbocation is further stabilized and 335

becomes the preferred resonance structure. Shifts in MG 13C NMR peaks have 336

confirmed this shift in charge distribution between unbound and bound MG [34]. 337

However, due to the long timescale requirement of 13C measurements, MG 338

intermediates are unlikely to be distinguished from MG. Alternatively, the RNA base 339

stacking forces may be weakened through the more hydrophilic C+-OH2 carbocation, 340

relaxing the induced fluorescent planar geometry [37]. The addition of organic 341

additives prevents or reverses it. 342

Fig. 8

343

The hydrolysis of MG resulted in a significant impact in the apparent Kd (Fig. 2A), 344

due to the less available MG, reduced affinity MG-OH2 transition state, or both. The 345

binding of one ligand to MGA will reduce the competitiveness, even with ligands of 346

higher affinity, due to the induced fit model [23]. Thus, competitive binding 347

experiment (between MG and MGOH to MGA) was conducted by simultaneously 348

introducing various ratios of MGOH:MG with a fixed amount of MGA. As shown in 349

12

Fig. S6, up to 7-fold (limited by aqueous solubility) molar excess of MGOH was 350

added into a fixed MGA:MG ratio, but no significant changes were observed in FLU 351

(no competitive binding) of the MGA-MG complex. Even when MGOH and MGA 352

were the initial conditions (Fig. S3), the formation of MG and the subsequent MGA-353

MG fluorescent complex still matched absorbance kinetics (buffer-only kMGOHMG), 354

suggesting there is no competitive or non-specific interactions from buffer containing 355

primarily MGOH. The failed affinity of MGOH to MGA is speculated to be from 356

MGOH’s neutrally charged structure, as long range electrostatic forces provided by 357

cationic amines need to be present [15, 24]. 358

4.2. Organic additives shift Keq with limited impact on Kd 359

Organic additives are known to increase the Kd of molecular recognition elements 360

[15]. Towards MGA-MG binding at time 0 h, the organic additives increased Kd in the 361

following additive order: ACN > EtOH > DMSO. The lowered affinity was 362

proportional to the additive concentration, but increased the fluorescence intensity of 363

MGA-MG complex (Fig. 3A-C, compared to Fig. 2A). This paradox can be explained 364

by the shift in Keq and by noting that freshly prepared MG solutions always contained 365

various concentrations of MG and MGOH, as seen in Fig. 5. Fig. 6 demonstrates how 366

organic additives slow MGMGOH while accelerating the MGOHMG 367

conversion. Hence, the presence of organic additives results in synergistic shift 368

towards MG, thus enhances the formation of MGA-MG based fluorescence. Although 369

the aptamer binding is normally performed in aqueous only environments, the 370

presence of organic additives simplifies the Keq towards MG and allows a much 371

longer window of measurement towards assay development and may extend the 372

application of aptamers and more exotic ligands, i.e. modifications of hydrophobic 373

prodrugs for delivery [38]. Indeed, 40% of ‘new chemical entities’ developed by the 374

pharmaceutical industry are practically insoluble and organic additives are often 375

added to mimic in vivo solubility parameters, widening the window of application and 376

selection of ligands with poor aqueous solubility [39-42]. Additional benefits include 377

the facilitated competitive binding of organic ligands and antibacterial properties. 378

The loss of fluorescence intensity and increase in Kd were prevented upon addition 379

of organic additives. This discovery provides a simple way to stabilize aptamer 380

binding, which is beneficial for applications that rely on MGA for accurate detection 381

of MG. This inhibition strategy may also be applied to other triphenylmethane 382

compounds that are susceptible to hydrolysis, for instance, to stabilize aptasensors of 383

13

crystal violet [43-46]. To date, more than 105 triphenylmethane compounds ((Ph)3-C) 384

and 9,000 substructures of MG have been reported to date (SciFinder). 385

Triphenylmethane compounds have been commercially and analytically used as 386

bacteriological stains, pH indicators, therapeutic agent and food dyes [45]. However, 387

the use of some triphenylmethane compounds has been banned for farming due to 388

their potential toxicity, thus the detection and monitoring of triphenylmethane 389

compounds is of consumer importance [17]. Additionally, it is well known that the 390

triphenylmethane and other triarylmethane structures can form carbocations, which is 391

susceptible to nucleophilic attack (i.e. hydrolysis). The discovery in this work paves a 392

way to stabilize aptamer binding to ligands with triphenylmethane or similar 393

carbocation structures. 394

Moreover, the decreasing trend of rate constants with increasing additive 395

concentration (Table 1) shows a strong correlation. In other words, the rate constant 396

may be tuned by the type of organic additive, which may benefit alternative designs 397

for in-line or flow aptasensors, i.e. sensing within specific time frames. 398

4.3. Synergistic stabilization of ligand by aptamer and organic additives 399

To further understand the MG hydrolysis and MGOH dehydration, kinetics in both 400

directions was evaluated independently without aptamer by absorbance (Fig. 6). The 401

results indicate a relatively fast hydrolysis reaction from MG to MGOH in pH 6.7 402

buffer (Fig. 6A-B). The larger rate constant (compared to the one in aptamer binding 403

in Fig. 2C) indicates the aptamer can interfere with MG hydrolysis, but not prevent it 404

completely. Wang et al. has shown that the MGA could prevent MG well from 405

hydrolysis over time due to the steric hindrance effect [29]. However, our results 406

showed the protective effect is weaker, which is likely due to the difference in 407

experimental method of kinetics and affinity evaluation. Wang et al. evaluated MG 408

hydrolysis through changes in MG’s absorbance maxima, while we based ours on 409

fluorescence. Absorbance changes may not be apparent between complexed MG and 410

the intermediate MG-OH2, thus accounting for their perceived observation of 411

complete quenching of hydrolysis. If the bound intermediate MG-OH2 is responsible 412

for the FLU decay, Wang’s results provide further evidence that MG-OH2 413

intermediate can still provide enough affinity interactions to keep the aptamer folded, 414

while MGOH cannot. Without aptamer, the organic additives had a slight inhibition 415

on kinetics of MG to MGOH (Fig. 6A), but catalyze the dehydration of MGOH to 416

MG (Fig. 6B). Addition of organic additive not only accelerated the conversion of 417

14

MGOH to MG, but the conversion appeared to shift the equilibrium in favor of MG 418

(Table 2). The impact on kinetics and equilibrium defines how the organic additives 419

stabilized the fluorescence intensity of the MGA-MG complex over 24 h. 420

Synergistic stabilization on MG was observed by the combination of rates changes 421

through the organic additives and aptamer (Table 2). Pearson’s R correlation shows a 422

strong correlation between rate constants and the molecular properties of the organic 423

additives in the presence of aptamer. Less polar surface area, larger solvent volume, 424

or a combination of both, correlates with decreasing rate constants of MG hydrolysis. 425

These results are supported by Wang et al, showing minimal space between the 426

MGA-MG interface and the prevention of hydroxyl ion attack, which is an unlikely 427

mechanism of hydrolysis in a pH 6.7 buffer [29, 32]. Nguyen et al. argued that MGA 428

can change the conformation and charge distribution of MG upon binding [34], which 429

may result in a shift of hydrolysis kinetics and thermodynamics. However, it has not 430

yet been reported what the impact of organic additives may have on the stability of 431

MG inside the aptamer binding pocket. In this study, the results show that organic 432

additives; 1) Shift equilibrium to MG in the presence of aptamer, 2) had minor effect 433

on MG hydrolysis without aptamer, 3) catalyze MGOH dehydration and 4) had only 434

minor shifts on affinity and Kd. Based on these primary 4 observations, a mechanism 435

is proposed in Fig. 8. At pH 6.7, MG undergoes slow hydrolysis to intermediate MG-436

OH2, which transforms to MGOH immediately. Without MGA, the rate constants 437

kMGMGOH (hydrolysis) and kMGOHMG (dehydration) favor MGOH as the major 438

equilibrium product (Fig. 8A). Since [OH-] equals to 5.0 × 10-8 M at this pH, thus 439

kOH*[OH-] << kH2O and results in kobs ≈ kMGMGOH [47] (Fig. 8B). As proposed in Fig. 440

8C, MGA retards MG hydrolysis (where aptamer is in excess of MG) by retarding the 441

conversion of MG-OH2 to MGOH, which results in a decreased kMGMGOH, as 442

summarized in Table 2. However, this inhibition effect by MGA is insufficient to 443

make kMGMGOH ≤ kMGOHMG, and thus the conversion still favors to MGOH 444

generation, suggested by fluorescence decrease of MGA-MG complex over time as 445

displayed in Fig. 2A. With the presence of organic additives, kMGOHMG increases by 446

few hundred percent (as shown in Fig. 6B), which results in kMGMGOH ≥ kMGOHMG 447

combining with the effect from MGA. The synergistic effect of aptamer and organic 448

additives (decrease of kMGMGOH and increase of kMGMGOH) results in a Keq favoring 449

prevention of MG hydrolysis (Fig. 4) and MGOH dehydration. MGOH dehydration 450

and observed Keq was challenged by repeating the FLU decayed experimental 451

15

conditions in Fig. 2A, with the addition of DMSO (3-10% v/v) post decay. The 452

results in Fig. 7 show that a minimum of 5% v/v DMSO was enough to recover the 0 453

h fluorescence after an incubation time of 12 h. Higher conc. of DMSO yielded FLU 454

values that approached 10,000 FLU units (data not shown). 455

5. Conclusions 456

Aptamers as new materials for pharmaceutics and diagnostics have unique 457

advantages as molecular recognition elements. However, the aptamer-ligand affinity 458

is seldom investigated long term. In this work, we investigated the fluorescence 459

intensity of aptamer-malachite green complexes over 24 h and quantified the kinetics 460

of malachite green hydrolysis under various environmental conditions. The loss of 461

fluorescence intensity and increase in apparent Kd was due to the hydrolysis of 462

malachite green into speculated intermediates and malachite green carbinol base, as 463

indicated by spectrofluorometry, HPLC-diode array detector, and absorbance spectra. 464

Organic additives prevented the decay of the fluorescence intensities and were found 465

to minimize shifts in apparent Kd. A mechanism was proposed and challenged that 466

suggests shifts in Keq prevent the formation of malachite green carbinol base, or if 467

formed, cause dehydration back into the malachite green. This work demonstrated a 468

synergistic effect of both aptamer and organic additive on the prevention of 469

carbocation hydrolysis, allowing larger time frames for aptamer biosensor assays. 470

Acknowledgments 471

This work was supported by the Ministry of Education Tier 1 Grant RG54/13: 472

“Photochrome Aptamer Switch Assay: A Universal Bioassay Device”; and the 473

Campus for Research Excellence and Technological Enterprise (CREATE) 474

programme (13-04-00364 А), which is supported by the National Research 475

Foundation, Prime Minister’s Office, Singapore. 476

Appendix A. Supplementary materials 477

Please refer to the individual file for supplementary data. 478

479

References 480

[1] Y. Mao, J. Liu, D. He, X. He, K. Wang, H. Shi, L. Wen, Aptamer/target binding-481

induced triple helix forming for signal-on electrochemical biosensing, Talanta 143 482

(2015) 381-387. 483

16

[2] E. Babu, P. Muthu Mareeswaran, V. Sathish, S. Singaravadivel, S. Rajagopal, 484

Sensing and inhibition of amyloid-β based on the simple luminescent aptamer–485

ruthenium complex system, Talanta 134 (2015) 348-353. 486

[3] S.Y. Toh, M. Citartan, S.C.B. Gopinath, T.-H. Tang, Aptamers as a replacement 487

for antibodies in enzyme-linked immunosorbent assay, Biosens Bioelectron 64 (2015) 488

392-403. 489

[4] Y. Wang, Z. Ye, J. Ping, S. Jing, Y. Ying, Development of an aptamer-based 490

impedimetric bioassay using microfluidic system and magnetic separation for protein 491

detection, Biosens Bioelectron 59 (2014) 106-111. 492

[5] T. Moore, H. Chen, R. Morrison, F. Wang, J.N. Anker, F. Alexis, 493

Nanotechnologies for Noninvasive Measurement of Drug Release, Mol Pharmaceutics 494

11(1) (2014) 24-39. 495

[6] Q. Wu, L. Wu, Y. Wang, Z. Zhu, Y. Song, Y. Tan, X.-F. Wang, J. Li, D. Kang, 496

C.J. Yang, Evolution of DNA aptamers for malignant brain tumor gliosarcoma cell 497

recognition and clinical tissue imaging, Biosens Bioelectron 80 (2016) 1-8. 498

[7] B. Sivakumar, R.G. Aswathy, Y. Nagaoka, S. Iwai, K. Venugopal, K. Kato, Y. 499

Yoshida, T. Maekawa, D.N. Sakthi Kumar, Aptamer conjugated theragnostic 500

multifunctional magnetic nanoparticles as a nanoplatform for pancreatic cancer 501

therapy, RSC Advances 3(43) (2013) 20579-20598. 502

[8] J. Wu, C. Song, C. Jiang, X. Shen, Q. Qiao, Y. Hu, Nucleolin Targeting AS1411 503

Modified Protein Nanoparticle for Antitumor Drugs Delivery, Mol Pharmaceutics 504

10(10) (2013) 3555-3563. 505

[9] V.K. Sharma, P. Rungta, A.K. Prasad, Nucleic acid therapeutics: basic concepts 506

and recent developments, RSC Advances 4(32) (2014) 16618-16631. 507

[10] J. Ping, Y. Zhou, Y. Wu, V. Papper, S. Boujday, R.S. Marks, T.W.J. Steele, 508

Recent advances in aptasensors based on graphene and graphene-like nanomaterials, 509

Biosens Bioelectron 64(0) (2015) 373-385. 510

[11] J.-J. Zhang, J.-T. Cao, G.-F. Shi, K.-J. Huang, Y.-M. Liu, S.-W. Ren, A luminol 511

electrochemiluminescence aptasensor based on glucose oxidase modified gold 512

nanoparticles for measurement of platelet-derived growth factor BB, Talanta 132 513

(2015) 65-71. 514

[12] A. Chen, S. Yang, Replacing antibodies with aptamers in lateral flow 515

immunoassay, Biosens Bioelectron 71 (2015) 230-242. 516

[13] S.L. Richards, A.T. Cawley, R. Cavicchioli, C.J. Suann, R. Pickford, M.J. 517

Raftery, Aptamer based peptide enrichment for quantitative analysis of gonadotropin-518

releasing hormone by LC–MS/MS, Talanta 150 (2016) 671-680. 519

[14] V. Papper, O. Pokholenko, Y. Wu, Y. Zhou, P. Jianfeng, T.J. Steele, R. Marks, 520

Novel Photochrome Aptamer Switch Assay (PHASA) for Adaptive Binding to 521

Aptamers, J Fluoresc 24(6) (2014) 1581-1591. 522

[15] S.L. Stead, H. Ashwin, B. Johnston, A. Dallas, S.A. Kazakov, J.A. Tarbin, M. 523

Sharman, J. Kay, B.J. Keely, An RNA-aptamer-based assay for the detection and 524

analysis of malachite green and leucomalachite green residues in fish tissue, Anal 525

Chem 82(7) (2010) 2652-60. 526

[16] J.R. Babendure, S.R. Adams, R.Y. Tsien, Aptamers Switch on Fluorescence of 527

Triphenylmethane Dyes, J Am Chem Soc 125(48) (2003) 14716-14717. 528

[17] W.C. Andersen, S.B. Turnipseed, C.M. Karbiwnyk, R.H. Lee, S.B. Clark, W.D. 529

Rowe, M.R. Madson, K.E. Miller, Multiresidue method for the triphenylmethane dyes 530

in fish: Malachite green, crystal (gentian) violet, and brilliant green, Anal Chim Acta 531

637(1–2) (2009) 279-289. 532

17

[18] H. Wan, S. Weng, L. Liang, Q. Lu, J. He, Evaluation of the developmental 533

toxicity of leucomalachite green administered orally to rats, Food Chem Toxicol 534

49(12) (2011) 3031-3037. 535

[19] T. Gessner, U. Mayer, Triarylmethane and Diarylmethane Dyes, Ullmann's 536

Encyclopedia of Industrial Chemistry, Wiley-VCH Verlag GmbH & Co. KGaA2000. 537

[20] B. Samiey, L. Motamedi, Kinetics of malachite green fading in water–ethanol–2-538

propanol ternary mixtures, Int J Chem Kinet 43(8) (2011) 441-453. 539

[21] D. He, X. He, K. Wang, J. Cao, Y. Zhao, A Photon-Fueled Gate-Like Delivery 540

System Using i-Motif DNA Functionalized Mesoporous Silica Nanoparticles, Adv 541

Funct Mater 22(22) (2012) 4704-4710. 542

[22] K. Mitrowska, A. Posyniak, J. Zmudzki, Determination of malachite green and 543

leucomalachite green in carp muscle by liquid chromatography with visible and 544

fluorescence detection, J Chromatogr A 1089(1–2) (2005) 187-192. 545

[23] J.B. Da Costa, A.I. Andreiev, T. Dieckmann, Thermodynamics and Kinetics of 546

Adaptive Binding in the Malachite Green RNA Aptamer, Biochemistry-us 52(38) 547

(2013) 6575-6583. 548

[24] J. Bernard Da Costa, T. Dieckmann, Entropy and Mg2+ control ligand affinity 549

and specificity in the malachite green binding RNA aptamer, Mol BioSyst 7(7) (2011) 550

2156-63. 551

[25] Y. Wang, J. Killian, K. Hamasaki, R.R. Rando, RNA Molecules That 552

Specifically and Stoichiometrically Bind Aminoglycoside Antibiotics with High 553

Affinities†, Biochemistry-us 35(38) (1996) 12338-12346. 554

[26] Y. Wang, R.R. Rando, Specific Binding of Aminoglycoside Antibiotics to Rna, 555

Chem Biol 2(5) (1995) 281-290. 556

[27] M. Strackharn, S.W. Stahl, E.M. Puchner, H.E. Gaub, Functional Assembly of 557

Aptamer Binding Sites by Single-Molecule Cut-and-Paste, Nano Lett 12(5) (2012) 558

2425-2428. 559

[28] K.A. Afonin, M. Viard, A.Y. Koyfman, A.N. Martins, W.K. Kasprzak, M. 560

Panigaj, R. Desai, A. Santhanam, W.W. Grabow, L. Jaeger, E. Heldman, J. Reiser, W. 561

Chiu, E.O. Freed, B.A. Shapiro, Multifunctional RNA Nanoparticles, Nano Lett 14(10) 562

(2014) 5662-5671. 563

[29] T. Wang, J.A. Hoy, M.H. Lamm, M. Nilsen-Hamilton, Computational and 564

Experimental Analyses Converge to Reveal a Coherent Yet Malleable Aptamer 565

Structure That Controls Chemical Reactivity, J Am Chem Soc 131(41) (2009) 14747-566

14755. 567

[30] D.M. Kolpashchikov, Binary malachite green aptamer for fluorescent detection 568

of nucleic acids, J Am Chem Soc 127(36) (2005) 12442-12443. 569

[31] D.M. Kolpashchikov, M.N. Stojanovic, Boolean Control of Aptamer Binding 570

States, J Am Chem Soc 127(32) (2005) 11348-11351. 571

[32] R. Cigen, The influence of temperature on the prtolytic equilibriums and reaction 572

rate constants of basic tryphenylmehtane dyes, Acta Chem Scand 14 (1960) 979-993. 573

[33] S. Dasmandal, H.K. Mandal, S. Rudra, A. Kundu, T. Majumdar, A. Mahapatra, 574

Kinetic exploration supplemented by spectroscopic and molecular docking analysis in 575

search of the optimal conditions for effective degradation of malachite green, Rsc 576

Advances 5(48) (2015) 38503-38512. 577

[34] D.H. Nguyen, S.C. DeFina, W.H. Fink, T. Dieckmann, Binding to an RNA 578

aptamer changes the charge distribution and conformation of malachite green, J Am 579

Chem Soc 124(50) (2002) 15081-4. 580

[35] B. Samiey, M. Zafari, Kinetics of Malachite Green Fading in Water-ethanol-1-581

propanol Ternary Mixtures, J Chin Chem Soc-taip 59(1) (2012) 32-39. 582

18

[36] S. Chen, Y. Jiang, Z. Wang, X. Zhang, L. Dai, M. Smet, Light-Controlled Single-583

Walled Carbon Nanotube Dispersions in Aqueous Solution, Langmuir 24(17) (2008) 584

9233-9236. 585

[37] J. Flinders, S.C. DeFina, D.M. Brackett, C. Baugh, C. Wilson, T. Dieckmann, 586

Recognition of planar and nonplanar ligands in the malachite green-RNA aptamer 587

complex, Chembiochem 5(1) (2004) 62-72. 588

[38] N. de-los-Santos-Álvarez, M.J. Lobo-Castañón, A.J. Miranda-Ordieres, P. 589

Tuñón-Blanco, Modified-RNA Aptamer-Based Sensor for Competitive Impedimetric 590

Assay of Neomycin B, J Am Chem Soc 129(13) (2007) 3808-3809. 591

[39] J. Lee, M. Jo, T.H. Kim, J.-Y. Ahn, D.-k. Lee, S. Kim, S. Hong, Aptamer 592

sandwich-based carbon nanotube sensors for single-carbon-atomic-resolution 593

detection ofnon-polar small molecular species, Lab on a Chip 11(1) (2011) 52-56. 594

[40] Y. Ito, S. Fujita, N. Kawazoe, Y. Imanishi, Competitive Binding Assay for 595

Thyroxine Using in Vitro Selected Oligonucleotides, Anal Chem 70(16) (1998) 3510-596

3512. 597

[41] K.T. Savjani, A.K. Gajjar, J.K. Savjani, Drug solubility: importance and 598

enhancement techniques, ISRN Pharm 2012 (2012) 195727. 599

[42] D. Sharma, M. Soni, S. Kumar, G. Gupta, Solubility enhancement—eminent role 600

in poorly soluble drugs, Research Journal of Pharmacy and Technology 2(2) (2009) 601

220-224. 602

[43] C. Baugh, D. Grate, C. Wilson, 2.8 A crystal structure of the malachite green 603

aptamer, J Mol Biol 301(1) (2000) 117-28. 604

[44] L. García-Río, J.R. Leis, J.C. Mejuto, A. Navarro-Vázquez, J. Pérez-Juste, P. 605

Rodriguez-Dafonte, Basic Hydrolysis of Crystal Violet in β-Cyclodextrin/Surfactant 606

Mixed Systems, Langmuir 20(3) (2004) 606-613. 607

[45] B. Samiey, Z. Dalvand, Kinetics of Fading of Some Triphenylmethane Dyes: 608

Effects of Electric Charge, Substituent, and Aqueous Binary Mixtures of Dimethyl 609

Sulfoxide and 2-Propanol, Int J Chem Kinet 46(1) (2014) 60-70. 610

[46] Y. Wu, S. Zhan, H. Xing, L. He, L. Xu, P. Zhou, Nanoparticles assembled by 611

aptamers and crystal violet for arsenic(iii) detection in aqueous solution based on a 612

resonance Rayleigh scattering spectral assay, Nanoscale 4(21) (2012) 6841-6849. 613

[47] B.M. Fox, G. Hallas, J.D. Hepworth, D. Mason, The hydrolysis of brilliant green 614

and some derivatives. I. A kinetic and equilibrium study of the hydrolysis of brilliant 615

green, J Chem Technol Biot 30(1) (1980) 317-323. 616

617

618

19

Tables 619

620

Table 1. Decay rate constant (kMGMGOH) of fluorescence intensity at maximum 621

[MGA]/ratio with various percentages of organic additives. 622

Media kMGMGOH (×10-4) (s-1)

3% Acetonitrile 3.1±0.4

5% Acetonitrile 2.6±0.1 8% Acetonitrile 1.3±0.2

10% Acetonitrile 1.0±0.1

3% Dimethyl sulfoxide 2.7±0.6 5% Dimethyl sulfoxide 0.2±0.3*

8% Dimethyl sulfoxide < 0.01, NSD

10% Dimethyl sulfoxide < 0.01, NSD 3% Ethanol 4.0±2.1

5% Ethanol 1.5±0.1

8% Ethanol 0.4±0.1

10% Ethanol 0.2±0.1 * Fitting coefficient of determination R2 < 0.6. NSD: No statistical difference was seen between 0 and 623

24 h FLU. 624

625

20

Table 2. Rate constant (kMGMGOH) of MG hydrolysis with 10% percentages of 626

organic additives in the presence and absence of aptamer (maximum [MGA]/ratio), 627

and the theoretical properties of water and additives. 628

Media With aptamer Without aptamer Solvent Molecular

properties#

kMGMGOH (×10-4)

(s-1) Keq

kMGMGOH (×10-4)

(s-1) Keq

Molecular Polar

Surface

Area (MPSA)

(Å2)

Volume

(Å3)

buffer 2.5±0.2 3.62 6.2±0.1 8.99 Water 29.3 19.3 10% Acetonitrile 1.0±0.1 0.40 4.8±0.1 1.92 Acetonitrile 23.8 46.1

10% Ethanol 0.2±0.1 0.06 5.9±0.1 1.84 Ethanol 20.2 54.0

10% Dimethyl sulfoxide

NSD* < 0.05* 5.7±0.0 1.84 Dimethyl sulfoxide

17.1 71.4

Pearson’s R

correlation, kMGMGOH vs.

MPSA

+0.999** +0.189

Pearson’s R correlation,

kMGMGOH vs.

Solvent Volume

-0.991** -0.293

* No statistical difference in 0 and 24 h FLU values under 10% DMSO additive. 629

**: 10% Dimethyl sulfoxide is excluded in the correlation due to its unavailable kinetic constant. 630

#: Calculated by Molinspiration Cheminformatics 2015: www.molinspiration.com. 631

632

21

Figures 633

634

635

Fig. 1. A: Structure of malachite green aptamer (MGA), malachite green (MG) and 636

malachite green carbinol base (MGOH). B: Binding of MG to MGA, with 637

fluorescence obtained. 638

639

22

640

Fig. 2. A: Kd determination of MGA to MG in aqueous buffer from 0 to 24 h. B: An 641

illustration of the fitting of Kd (R2> 0.99). F0 and Fmax: the fluorescence intensity at 642

zero and infinite aptamer [MGA]; [MG]0 and [MGA]0: initial concentrations of MG 643

and MGA, respectively; Kd: dissociation constant. D: Normalized fluorescence decay 644

at the highest MGA concentration/ratio for MG. The fluorescent decrease has a high 645

correlation to first order decay. 646

647

648

23

649

Fig. 3. Apparent Kd in organic additives at time 0 h. A: In 3% and 10% ACN. B: In 650

3% and 10% DMSO. C: In 3% and 10% EtOH. D: Comparison of the apparent Kd 651

values * = p < 0.05: significant difference vs. apparent Kd with 0% organic additive. 652

Generally, the apparent Kd increases with the percentage of organic additives, while 653

acetonitrile displayed the largest shifts. 654

655

24

656

Fig. 4. A: Kd determination of MG to MGA with different percentage of ACN at 0 h 657

and 24 h. B: Plots of calculated Kd values vs. %v/v additive at 0 h and 24 h. C: 658

Fluorescence intensity at maximum [MGA]/ratio at 0 h and 24 h with various 659

percentages of organic additives. The decrease of fluorescence intensity and increase 660

of apparent Kd after 24 h becomes smaller and smaller as the percentage of organic 661

additive increases from 3% to 10%. NSD = no statistical difference, p < 0.05. 662

25

663

Fig. 5. HPLC of MG, MGOH and the monitoring of the conversion of MG. A: 664

Detection wavelength = 620 nm. B: Detection wavelength = 265 nm. As time goes on, 665

the MG peak decreases while the MGOH peak increases, which indicates the 666

conversion of MG to MGOH. 667

668

26

669

Fig. 6. Kinetic evolution between MG and MGOH. A: MG converts to MGOH in 670

binding buffer and with 10% organic additives. B: MGOH converts to MG in binding 671

buffer and with 10% organic additives. The kinetics of MG to MGOH is much larger 672

than MGOH to MG in binding buffer, while organic additives induces a slightly 673

smaller kinetics of the conversion of MG to MGOH, but a much larger kinetics of the 674

conversion of MGOH to MG. 675

676

27

677

Fig. 7. Addition of DMSO into MGA-MG complex at 5h. Fluorescence of MGA-MG 678

complex decreases significant at 5h, while the addition of DMSO at 5h recovers the 679

fluorescence. Stronger effect of recovery occurs while higher percentage of DMSO is 680

added. 681

682

28

683

Fig. 8. A: Equilibrium and intermediates of MG, MG-OH2 and MGOH. B: Rate of 684

MG elimination. At pH 6.7, [OH-] is 5.0 × 10-8 M and thus, kOH*[OH-] << kH2O [47]. 685

C: Proposed mechanism of the conversions between MG+, MG-OH2+ and MGOH in 686

the presence of MGA. In the absence of MGA, the rate constants kMGMGOH 687

(hydrolysis) and kMGOHMG (dehydration) favor MGOH as the major equilibrium 688

product (A). While excess MGA is present, it retards MG hydrolysis, which results in 689

a decreased kMGMGOH. With the presence of organic additives, kMGOHMG increases 690

by few hundred percent (C), resulting in a “stabilizing” effect. 691

692

1

Electronic Supplementary Information (ESI)

for

Organic additives stabilize RNA aptamer binding of malachite

green

Yubin Zhoua, Hong Chia, Yuanyuan Wua, Robert S. Marksa,b, Terry W.J. Steelea,*

a School of Materials Science & Engineering, College of Engineering, Nanyang

Technological University, 50 Nanyang Avenue, Singapore 639798, Singapore

b Department of Biotechnology Engineering, Faculty of Engineering Sciences, Ben

Gurion University of the Negev, P.O. Box 653, Beer Sheva 84105, Israel

* Corresponding Author:

Terry W.J. Steele: wjsteele@ntu.edu.sg,

Tel.:65-65927594, fax: 65-67909081.

2

Method for electrophoresis of MGA

MGA in binding buffer (containing 10 mM Mg2+) over 48 hours were analyzed by

electrophoresis. Samples were stained with AMRESCO EZ-Vision® One (with amaranth

indicator) and loaded on 2% agarose gel, and then electrophoresed at 100 V for 40

minutes in pH = 8.0 1X tris-acetate-EDTA (TAE) buffer. BioLabs 100 bp DNA ladder

(100-1517 bp) was used as marker for reference. The gel was visualized under 365 nm

UV exposure.

3

Fig. S1. Binding curves (A) and apparent Kd values (B) in different pH at time 0. The

result indicates that the suitable pH for the binding experiment is 6-8.5.

4

Fig. S2. UV-vis spectra of conversion between MG and MGOH as a function of time. A:

MG converted to MGOH. B: MGOH converted to MG.

5

Fig. S3. Rate constant of MGOH to MG (kMGOHMG) in binding buffer in the presence of

MGA.

6

Fig. S4. Fluorescence intensity of MGA-MG complex (32:1) at 24 h binding and with the

subsequent addition of original amount of MG (blue square) and MGA (red circle).

7

Fig. S5. Agarose gel electrophoresis of MGA in different time points in binding buffer

(with Mg2+ presented).

8

Fig. S6. Competitive binding of MGOH with MG to MGA. No significant change in

fluorescence intensity of MGA-MG was observed in the presence of 7-fold excess of

MGOH (poor solubility of MGOH in water limits the maximum ratio to 8X in our

experimental setup).