Published on 03 September 2014. Downloaded by The University of British Columbia Library on 11/10/2014 01:44:52.

Organic & Biomolecular Chemistry View Article Online

PAPER

Cite this: Org. Biomol. Chem., 2014, 12, 8422

View Journal | View Issue

Multi-channel colorimetric and fluorescent probes for differentiating between cysteine and glutathione/homocysteine† Lun Song,a Ti Jia,b Wenjia Lu,a Nengqin Jia,b Weibing Zhang*a,c and Junhong Qian*a,c Three fluorescent probes TP1–3 for thiols were rationally designed and synthesized to distinguish cysteine (Cys) from glutathione (GSH)/homocysteine (Hcy). TP1–3 are almost non-fluorescent and colorless 4-nitro-1,8-naphthalimide derivatives. Upon the substitution of nitro by Cys, TP1–3 were transformed into weakly fluorescent green-emitting 4-amino analogs via highly fluorescent blue-emitting thioether intermediates. The three-channel signaling capability allows discrimination between Cys and GSH/Hcy. The fluorescence intensity at 498 nm was linearly proportional to GSH concentration in

Received 13th June 2014, Accepted 2nd September 2014

the range of 0–20 μM, and the detection limit was 5 × 10−8 mol L−1. A good linear relationship between

DOI: 10.1039/c4ob01219d

A446/A350 and Cys concentration was found in the range of 0–70 μM, and the detection limit was 2 × 10−7 mol L−1. Moreover, TP3 was used for living cell imaging as well as for detecting mercapto-containing

www.rsc.org/obc

proteins.

Introduction Low-molecular-weight biothiols cysteine (Cys), homocysteine (Hcy) and glutathione (GSH) play vital but distinct roles in biological systems.1 The abnormal levels of these thiols in plasma are related to different diseases.2 For example, the deficiency of Cys may lead to liver damage, slowed growth, hematopoiesis decrease and leucocyte loss;3 the elevated levels of Hcy have been a risk factor for cardiovascular and Alzheimer’s diseases,4 while abnormal levels of GSH are related to cancer, aging and other sicknesses.5 Therefore, selective detection and quantification of biological thiols are of considerable importance and have attracted much attention. Numerous methods have been developed for biothiol detection.6 Among those, the fluorescence technique shows some advantages over other methods, such as sensitivity, selectivity,

a Shanghai Key Laboratory of Functional Materials Chemistry, School of Chemistry and Molecular Engineering, East China University of Science and Technology, Shanghai, 200237, China. E-mail: [email protected], [email protected] b The Education Ministry Key Laboratory of Resource Chemistry and Shanghai Key Laboratory of Rare Earth Functional Materials, Department of Chemistry, College of Life and Environmental Sciences, Shanghai 200234, China c State Key Laboratory of Food Science and Technology, Nanchang University, Nanchang 330047, China † Electronic supplementary information (ESI) available: Experimental details of synthetic procedures, characterization and cell cultures, and fluorescence titrations. See DOI: 10.1039/c4ob01219d

8422 | Org. Biomol. Chem., 2014, 12, 8422–8427

real time detection capability, in vivo imaging, and so on.7 A wide variety of colorimetric and fluorescent probes for thiols have been reported.8 Most of these probes responded to all the thiol-containing amino acids; only a few of them could distinguish Cys/Hcy/GSH from each other.9 It is of great interest to discriminate one from other thiols because of their different physiological and pathological roles. In our previous work, 4-nitro-1,8-naphthalic anhydride (NNA) was employed as a fluorescent probe to detect Cys over Hcy/GSH, and the sensing mechanism involves an intramolecular aromatic nucleophilic substitution (SNAr).10 However, this determination of Cys was conducted at 50 °C and in DMF, which was not good enough for application in biosamples. Shortly afterwards, several fluorescent probes were developed for discriminating Cys/Hcy/GSH from each other in aqueous solution based on this mechanism.9c–f In the present work, we modified NNA with different substitutes in view of the following considerations: (1) the substitution of nitro by thiol is an aromatic nucleophilic substitution, which is facilitated by an electron-withdrawing group; (2) the introduction of a hydrophilic group enhances the solubility of the probe and promotes the potential for applications in cell-imaging and biosample detection. Herein, 2-amino-pyrimidine, 2-aminothiazole and 2-amino-pyridine were selected to modify 4-nitro1,8-naphthalic anhydride because of their electron-withdrawing nature and good H-bond acceptors. 2-Amino-pyridine was further methylated to its corresponding ammonium salt. We envisaged that the reactions between thiols and the probes

This journal is © The Royal Society of Chemistry 2014

View Article Online

Published on 03 September 2014. Downloaded by The University of British Columbia Library on 11/10/2014 01:44:52.

Organic & Biomolecular Chemistry

Scheme 1

The chemical structures of the three dyes studied.

TP1–3 (Scheme 1) could be conducted under milder conditions with improved detection kinetics and improved selectivity between Cys and GSH/Hcy.

Results and discussion Colorimetric and fluorescent dual-responses of TP1–3 toward Cys Initially, the spectral responses of TP1–3 towards Cys were measured in DMF at 25 °C to study the substitute effect on the reaction rate. The spectral behaviors of all three compounds were similar to that of NNA.10 Upon the addition of Cys, the absorption peak of TP1 at 350 nm decreased steadily with a new peak rising at 395 nm, which is ascribed to the thioether intermediate (Fig. 1a). After 3 min, the absorption at 395 nm decreased and a second new peak at 446 nm increased, which is from the 4-amino analog. The formation of 4-amino analog is accompanied by a yellow color, which is easily noticed by the naked eye. TP1 is non-fluorescent and therefore its disappearance cannot be recorded by fluorescence spectroscopy. The compounds thioether intermediate and 4-amino analog are fluorescent with corresponding emission maxima at 480 and 521 nm when excited at 380 and 420 nm respectively. Their formation or consumption could be conveniently and

Paper

Fig. 2 Time-dependent absorbance (a, λab = 446 nm) and normalized emission intensity (b, λem = 535 nm) of different probes in the presence of 400 μM Cys in DMF. [ probe] = 20 μM, λex = 420 nm, 25 °C.

sensitively monitored via fluorescence spectroscopy. When TP2 or TP3 were used instead of TP1, similar results were obtained but with shorter reaction times (Fig. S1–2†). As expected, the reaction rates of the three probes were much faster than that of NNA. It only took 5, 3 or 1 min respectively for the thioether intermediates of TP1, TP2 and TP3 to reach their maxima. Under the same conditions, 15 min was needed for NNA. 30, 15 or 6 min were needed respectively for the signals of the ultimate 4-amino analogs of TP1, TP2 and TP3 to level off compared to ca. 100 min for NNA.10 Fig. 2 demonstrated that TP3 was the most active; consequently, it was used to detect thiols in the following experiments. Effect of water content on the reaction between TP3 and thiols The effect of water content in the solvent on the reaction between TP3 and Cys is shown in Fig. S3.† It is clear that the reaction slightly slows down with the addition of water: it took 30 min or 60 min for the signal of 4-amino analog to reach the plateau in 20 : 80 PBS–DMF (v/v) or 30 : 70 PBS–DMF (6 min in DMF), respectively. However, the absorbance at 446 nm decreases dramatically when the water content is elevated to 30% (Fig. S3a†). Time-dependent UV-vis spectra of TP3 were measured to study its stability in different media. The results showed that about 5% of TP3 was hydrolyzed in 8 : 2 DMF–PBS (v/v) after 1 hour, while about 30% of TP3 was hydrolyzed to produce by-product in 7 : 3 DMF–PBS within 30 min (Fig. S4†). Therefore, 8 : 2 DMF–PBS was used as the reaction medium. The hydrolyzed product is non-fluorescent and has no absorbance above 400 nm (Fig. S4†) and the interference by the hydrolysis of TP3 in the detection of thiols is negligible. Sensing mechanism of TP1–3 toward thiols

Fig. 1 The time-dependent UV-vis (a, b) and fluorescence (c, λex = 380 nm; d, λex = 420 nm) spectra of TP1 (20 μM) in the presence of 400 μM Cys in DMF at 25 °C.

This journal is © The Royal Society of Chemistry 2014

The effects of GSH and Hcy on the spectral properties of TP3 were also studied in 4 : 1 DMF–PBS. With 20 equiv. of GSH, the original absorption peak of TP3 (at 350 nm) decreased rapidly, and a new band at ca. 402 nm appeared and its signal reached a maximum very quickly (less than 5 min, Fig. S5a†). Meanwhile, more than 90-fold enhancement in fluorescence was observed with the emission maximum shifted from 450 to 498 nm (Fig. S5b†). A clear color change from dark to blue was observed under illumination with a UV lamp. Hcy induced similar spectral changes of TP3, but the absorbance at 402 nm

Org. Biomol. Chem., 2014, 12, 8422–8427 | 8423

View Article Online

Published on 03 September 2014. Downloaded by The University of British Columbia Library on 11/10/2014 01:44:52.

Paper

Scheme 2

Organic & Biomolecular Chemistry

The proposed reaction mechanism of TP3 with GSH/Cys.

and the fluorescence intensity at 498 nm decreased slowly. At the same time, a shoulder peak in the absorption spectrum at ∼460 nm emerged at the same time (Fig. S6b†), which could be ascribed to the 4-amino analog (compound 2 in Scheme 2). The reaction mechanism of TP3 with GSH and Cys is proposed in Scheme 2. The nitro group in TP3 was initially replaced by the thiol in Cys to produce the 4-sulfhydryl substitute (1, λex = 480 nm, λab = 395 nm, colorless), and the subsequent intramolecular SNAr substitution of sulfhydryl by the amino group through an intramolecular nucleophilic aromatic substitution via a 5-membered ring transition state led to 4-amino product (2, λem = 521 nm, λab = 446 nm, yellow). In the case of Hcy, the intramolecular substitution reaction occurred at a slower rate, because the 6-membered ring was kinetically less favorable. However, due to the unstable 10-membered ring in GSH, it is unlikely for the 4-sulfhydryl product (3) to give 4-amino derivative (2) analogously. To gain more insight into the sensing mechanism, HPLC traces of the reaction processes were performed (Fig. S7†). The peak of TP3 was at about 3.43 min; after reacting with GSH, the original peak decayed and a new peak at about 2.99 min emerged and developed. No other new peak formed during the experimental period (Fig. S7c†). In the case of Cys, the new peak at 2.81 min reached its maximum within 1 min and then started to decrease gradually in intensity. Meanwhile, a second new peak at about 3.48 min appeared and increased its intensity progressively (Fig. S7a–b†). The peaks at about 2.81 min and 3.48 min were believed to be 4-mercapto and 4-amino substitutes, respectively. NMR titration experiments were carried out in 4 : 1 DMFd6–D2O to provide some information of the reaction products of TP3 with GSH/Cys (Fig. S8†). From Fig. S8† it can be seen that the substitution of the electron-withdrawing group 4-nitro with GSH/Cys made all the MNR signals shift to the upfield due to the electron-donor property of either the amino or mercapto group. The peaks at 9.56 ppm (1H, Hj), 9.10 ppm (1H, Hg), 8.74 ppm (2H, Hf and Hh) and 8.54 ppm (1H, He) shifted slightly to the upfield, while the peak at 8.35 ppm (1H, Hd) shifted markedly to the upfield (8.13 ppm). At the same time, the signal of the methyl (4.61 ppm) in TP3 shifted to 4.57 ppm. No other new signal at