A facile ratiometric fluorescent chemodosimeter for hydrazine based on Ing-Manske hydrazinolysis and its applications in living cells

(1)

A facile ratiometric

ﬂuorescent chemodosimeter for hydrazine based

on Ing

eManske hydrazinolysis and its applications in living cells

Mandapati V. Ramakrishnam Raju

a

, Epperla Chandra Prakash

b

, Huan-Cheng Chang

b

,

Hong-Cheu Lin

a,*

a_{Department of Materials Science and Engineering, National Chiao Tung University, Hsinchu 300, Taiwan} b_{Institute of Atomic and Molecular Sciences, Academia Sinica, Taipei 106, Taiwan}

a r t i c l e i n f o

Article history:

Received 17 September 2013 Received in revised form 11 November 2013 Accepted 12 November 2013 Available online 21 November 2013 Keywords: Charge transfer Fluorescent probe IngeManske hydrazinolysis pH effects Semi-empirical calculations

Time resolved photoluminescence spectra

a b s t r a c t

A facile and sensitiveﬂuorescent probe for hydrazine was successfully constructed, displaying excellent colorimetric and ratiometric responses towards hydrazine via IngeManske hydrazinolysis conditions in semi-aqueous buffer solution. Semi-empirical calculations as well as spectroscopic results revealed the signalling mechanism of the current probe under hydrazinolysis conditions, in which hydrazine exclu-sively deprotected the phthalimide group by an intermediate of phthalhydrazide. Extensive screening of pH effects on the probe with the aid of proton nuclear magnetic resonance and mass spectrometry supported the distinctive and diverse ratiometric responses under hydrazinolysis and basic hydrolysis conditions. Time resolved photoluminescence measurements of the probe further conﬁrmed its discernible ratiometric responses probed at respective wavelengths. A distinctive ratiometric response under basic hydrolysis conditions and a successful utilization of probe towards hydrazine detection in living cells are demonstrated.

1. Introduction

Developing efficient and reaction specific synthetic probes with better sensitivity for the detection of small molecule based ana-lytes is of pivotal research interest owing to the toxic effects of many small molecules to humans and the environment[1]. Hy-drazine is a strong reducing agent and highly reactive base[2]; moreover, its widespread usage is inevitable due to its vital roles in chemical, pharmaceutical, and agricultural industries involving catalysts, corrosion inhibitors, and pesticides[3]. Hydrazine is a well-known high-energy fuel in rocket propulsion and missile systems due to its improved detonable properties[4]. However, hydrazine is extremely toxic and easily absorbed by oral, dermal, and inhalation exposure routes. Previous studies on laboratory animals suggested that hydrazine is highly neurotoxic, mutagenic, and carcinogenic [5]. Thus, developing reliable and real-time fluorometric detection methods for the specific detection of hy-drazine is warranted.

Conventionally, hydrazine was analysed by electrochemistry

[6], chromatography-mass spectrometric [7], titrimetric [8] and

gas chromatography[9]methods. However, those methods were often suffered in detecting hydrazine with low sensitivities. Despite their ease in detections with a trace amount of analytes by fluorometric methods possessing high sensitivity and selectivity functions; only a limited number of fluorescent small molecule based probes for hydrazine have been reported. Swager et al. developed the first fluorescent conjugated polymer for turn-on detection of trace amounts of hydrazine. [10] Chang et al. re-ported a selective detection of hydrazine by deprotection of a levulinate group[11]. Recently, Sessler and co-workers reported a trifluoroacetyl acetonate naphthalimide derivative that was formed afive membered heterocyclic compound, giving rise to a fluorescent turn-on response exclusively in the presence of hy-drazine[12].

Developing ratiometric and reaction specific fluorescent che-modosimeters are often bene_{ficial due to their specificity and} built-in correction for quantitative measurement by the ratio of fluo-rescence intensities at two different wavelengths [13]. Chemo-dosimeters appended with specific protection groups for selective detections via target specific deprotection for various analytes have often been utilized effectively[14e16]. However, to date there are only two reported ratiometric probes based on hydrazine mediated ester deprotection[17]and hydrazone formation[18]. However, to the best of our knowledge a renowned NH2 functional group

* Corresponding author. Tel.: þ886 3 5712121x55305; fax: þ886 3 5724727. E-mail address:[email protected](H.-C. Lin).

Contents lists available atScienceDirect

Dyes and Pigments

j o u r n a l h o m e p a g e : w w w . e l s e v i e r . c o m / l o c a t e / d y e p i g

(2)

synthon phthalimide [19] has never been explored in the speciﬁc ratiometric detection of hydrazine. Excellent photophysical properties and outstanding intramolecular charge transfer (ICT) structures of hydrophilic 4-aminonaphthalimide make them expedient candidates in designing novel ﬂuorescent probes[20]. However, facile ratiometric probes for hydrazine with selective and discriminative functions from other amine sources having potent biocompatibility within the biological pH range are required.

Herein, we developed a novel phthalimide protected 4-aminonaphthalimide for the speciﬁc and sensitive ratiometric detection of hydrazine via the IngeManske hydrazinolysis method

[21], a key step in Gabriel amine synthesis[22]and thus, enabling ICT as well as living cell permeability. Probe HZ was synthesized by appending phthalimide group via CuI promoted aryl halide nucle-ophilic substitution of compound 2 with potassium phthalimide in high boiling dimethylacetamide (DMA) solvent in a moderate yield as depicted inSchemes 1and2.

2. Experimental

2.1. General characterization methods

NMR spectra were recorded on Bruker Avance DRX300 Series (1H: 300 MHz;13C: 75 MHz) at a constant temperature of 25C. Chemical shifts were reported in parts per million from low to high ﬁeld and referenced to residual solvent (CDCl3, d6-DMSO: 1H

d

¼ 7.26, 2.49 ppm and13_C

d

¼ 77.23, 39.52 ppm, respectively). Coupling constant (J) were reported in hertz (Hz). UVeVis spectra were recorded on the Jasco UV-600 spectrophotometer using 1 cm quartz cuvette. Fluorescence measurements were conducted with HITACHI 7000 Series Spectrophotometer. All emission and excita-tion spectra were corrected for the detector response and the lamp output. Melting points were determined using a Fargo MP-2D

apparatus and are uncorrected. Elemental analyses were conduct-ed on HERAEUS CHN-OS RAPID elemental analyser. Infrarconduct-ed spec-troscopy data were collected using Perkin Elmer IR spectrophotometer. Solid sample were analysed using KBr pellet method. Time resolved photoluminescence (TRPL) spectra were measured using a home built single photon counting system with excitation from a 400 nm diode laser (Picoquant PDL-200, 50 ps fwhm, 2 MHz). The signals collected at the excitonic emis-sions of all sample solutions were connected to a time-correlated single photon counting card (TCSPC, Picoquant Timeharp 200). The emission decay data were analysed for hydrazine and HZ-hydroxide complex with biexponential kinetics, from which two decay components were derived; the lifetime values of (

s

1,

s

2) and pre-exponential factors (A1, A2) were determined. Confocal imaging was carried out using Leica TCS SP8 confocalﬂuorescence micro-scope, confocal ﬂuorescence imaging with using 60 times oil objective. Semi-empirical PM3 calculations were calculated using Gaussian-09 suite[23].

2.2. Materials

All the reagents were purchased from commercial sources and used without further purification. All the solvents were HPLC grade; anhydrous solvents were obtained by passing through activated alumina column purification system, further dried by standard drying procedures. Solvents were degassed by freeze/ thaw/pump cycle technique prior to use. 6-bromo-2-(2-(2-hydroxyethoxy)ethyl)-1H-benzo[de]isoquinoline-1,3(2H)-dione was prepared with a slight modification of previous literature

[24].

2.3. Stock solutions

Standard solution of probe HZ (100

m

M) were prepared in (1:9, v/v) in a mixture of water and ethanol solution. Prior to analysis the stock solution was diluted and pH of the solution was adjusted to about 7.2 using phosphate buffer saline (PBS) solution to deliver the ﬁnal concentration of the probe (5

m

M, pH¼ 7.2) in PBS-EtOH (1:9, v/v) solution. Hydrazine, other primary amines, metal ions, and anion stock solutions with concentration of (10 mM) were pre-pared, respectively in water. Before the titrations analytes were diluted to their desired volumes.

2.4. Cell culture and imaging

The human cervical cancer cell line (HeLa cells) were seeded onto cover slips at a concentration of (2 105 _{cells/mL) and} cultured in Dulbecco’s Modiﬁed Eagle’s Medium (DMEM) and 10%

Scheme 1. Hydrazine mediated phthalimide deprotection of probe HZ to form HZA.

Scheme 2. Synthesis of Probe HZ. Reagents and Conditions: (a) 2-(2-aminoethoxy)ethanol, EtOH, reﬂux, 4 h, 88%; (b) potassium phthalimide, CuI, DMA, reﬂux, 1 day, 55%; (c) N2H4,

(3)

fetal bovine serum in an incubator (37C, 5% CO2and 25% O2). After 30 h, the cover slips were rinsed slightly 3 times with PBS to remove the media and then cultured in PBS for later use. In view of imaging procedure, initially cells were incubated with 10

m

M of probe HZ alone for 30 min at 37C and observed under microscope and then again the samples were treated with hydrazine (25

m

M) and then incubated for 30 min and moved to the confocal stage. All the samples were slightly rinsed for 3 times with PBS buffer before observing them under the microscope. All the cell images were obtained with Leica TCS SP8 confocalﬂuorescence microscope us-ing 60 times oil objective.

2.5. Synthesis of Probe HZ

2.5.1. Synthesis of 6-bromo-2-(2-(2-hydroxyethoxy)ethyl)-1H-benzo[de]isoquinoline-1,3(2H)-dione (Compound 2)

A mixture of 6-bromobenzo[de]isochromene-1,3-dione (1, 2.0 g, 7.22 mmol, 1.0 equiv) and 2-(2-aminoethoxy)ethanol (0.79 g, 7.58 mmol, 1.05 equiv) in ethanol (60 mL) was heated under reﬂux for 4 h, and slowly cooled down to room temperature. The solution further cooled overnight in a freezer and the precipitated com-pound wasﬁltered and dried for overnight to get the compound 2 (2.31 g, 6.34 mol, 88%). Chemical formula: C16H14BrNO4,Molecular weight: 364.19; m.p.140.9e142.4_C. IR (KBr, cm1): 3513, 3081, 2908, 2865, 1692,1588; 1H NMR (300 MHz, CDCl3, 25C):

d

(ppm)¼ 8.65 (d, Jd¼ 8.9 Hz, 1H), 8.55 (d, Jd¼ 8.9 Hz, 1H), 8.40 (d, Jd¼ 8.9 Hz, 1H), 8.03 (d, Jd¼ 8.9 Hz, 1H), 7.83 (t, Jt¼ 9.0 Hz, 1H), 4.43 (t, Jt¼ 6.1 Hz, 2H), 3.86 (t, Jt¼ 6.1 Hz, 2H), 3.68 (t, Jt¼ 6.0 Hz, 4H), 2.51 (br, 1H, OH);13C NMR (75 MHz, CDCl3, 298 K):

d

(ppm)¼ 163.7, 133.2, 132.1, 131.3, 131.0, 130.4, 128.7, 128.0, 122.7, 121.9, 72.4, 68.3, 61.8, 39.7; MS (þESI-MS): (m/z): Calcd for C16H14BrNO4; 364.19; found: 364.0 [M]þ, 366.0 [Mþ 2]þ, 386 [Mþ Na]þ_{, [M}_{þ Naþ2]}þ_{; Anal. Calcd. for C}

16H14BrNO4: C, 52.77; H, 3.87, N, 3.85 found; C, 52.68; H, 3.85, N, 3.86.

2.5.2. Synthesis of 6-(1,3-dioxoisoindolin-2-yl)-2-(2-(2-hydroxyethoxy)ethyl)-1H-benzo[de]isoquinoline-1,3(2H)-dione (Probe HZ)

A mixture of compound 2 (1.0 g, 2.74 mmol, 1.0 equiv), potas-sium phthalimide (0.53 g, 2.88 mmol, 1.05 equiv), and CuI (0.575 g, 3.00 mmol, 1.1 equiv) were taken in over dried 100 mL RBF, and the mixture was applied to 3 freezeethawepump cycles. A freshly degassed DMA (60 mL) was added to the compound mixture and further degassed under argon atmosphere for 5 min, then reflux for 1 day. The solution was slowly cooled down to room temperature and poured into a beaker contains 200 g of crushed ice and stirred for 30 min, the resulted yellowish orange precipitate wasfiltered. The crude cake was dissolved in DCM (400 mL) and washed with brine solution (2 80 mL). The resulting solution was dried over MgSO4 and evaporated under vacuum. The crude product was subjected toflash column chromatography (silica gel, Hexane/EA: 8/2 to 6/4) to yield a pure yellow colouredfinal probe HZ (0.65 g, 1.51 mmol, 55%). Chemical formula: C24H18N2O6,Molecular weight: 430.12; m.p. 256.8e258.2_C. IR (KBr, cm1): 3515, 3091, 2942, 2863, 1727, 1609, 1401, 1224, 1046;1H NMR (300 MHz, CDCl3, 25C):

d

(ppm)¼ 8.72 (d, Jd¼ 8.9 Hz, 1H), 8.66 (d, Jd¼ 7.3 Hz, 1H), 8.05e8.00 (m, 3H), 7.91e7.88 (m, 2H), 7.79e7.73 (m, 2H), 4,47(t, Jt¼ 5.9 Hz, 2H), 3.86 (t,, Jt¼ 5.8 Hz, 2H), 3.67 (t, Jt¼ 3.9 Hz, 4H), 2.47 (br, 1H, OH);13C NMR (75 MHz, CDCl3, 298 K):

d

(ppm)¼ 167.1, 164.2, 163.8, 135.1, 134.4, 132.1, 131.8, 131.2, 129.5, 129.3, 128.9, 127.9, 127.8, 124.4, 123.5, 123.2, 72.4, 68.4,

Fig. 1. (a) Fluorescence spectra of probe HZ in the presence of hydrazine and other representative primary amines. [HZ] ¼ (5 mM), [hydrazine], and [primary amines]¼ (25mM) in a mixture of PBS buffer solutions (pH 7.2, 10 mM) and EtOH (1:9, v/v); (b) Bars represent theﬂuorescence intensity ratio in the presence and absence of various amines. Black bar represent the addition of primary amines (25mM) to probe HZ (5mM). Red bars represent the subsequent addition of hydrazine (25mM) to the solution. 0 ¼ hydrazine, 1 ¼ hydroxyl amine, 2 ¼ urea, 3 ¼ thiourea, 4¼ monomethylamine, 5 ¼ ethylenediamine, 6 ¼ 1,4-diaminobutane, 7 ¼ trans-1,2-diaminocyclohexane, 8¼ aqueous ammonia, and 9 ¼ guanidine nitrate, respectively. (c) and (d) UVevis (naked eye) and ﬂuorescence colour changes under (UV lamp 365 nm) of probe HZ (5mM) with various amines (25mM) sequentially from left to right hydroxyl amine, urea, thiourea, monomethylamine, ethylenediamine, hydrazine,

1,4-diaminobutane, trans-1,2-diaminocyclohexane, aqueous ammonia, and guanidine nitrate.lex¼ 405 nm, Slits: 5 nm/5 nm. (For interpretation of the references to colour

(4)

61.9, 39.8; MS (þESI-Ms): (m/z): Calcd for C24H18N2O6; 430.12; found: 431.1 [M þ 1]þ_{, 453.1 [M} _{þ Na]}þ_{; Anal. Calcd. for} C24H18N2O6: C, 66.97; H, 4.22 N, 6.51, found; C, 66.90; H, 4.20, N, 6.53.

2.5.3. Synthesis of 6-amino-2-(2-(2-hydroxyethoxy)ethyl)-1H-benzo[de]isoquinoline-1,3(2H)-dione (Compound HZA)

In an oven dried 25 mL RBF, probe HZ (0.050 g, 0.11 mmol, 1.0 equiv) was taken and dissolved in 8 mL of H2O/EtOH (1/9, v/ v) solution. Hydrazine (0.008 g, 0.24 mmol, 2.1 equiv, 0.5 M) solution was added in portion, an immediate colour change from deep yellow to pale yellow was observed. The reaction was continued for 15 min, at which time TLC showed complete deprotection of phthalimide. The resulted deep _fluorescent so-lution was evaporated under vacuum, and then DCM (25 mL) was added to the crude product and stirred for 30 min, the resulted precipitatefiltered through fine micron filter. The crude product was dissolved in minimum amount (4 mL) of MeOH and pentane (100 mL) was added and left under stirring for over-night. The resultedfine orange precipitate was filtered through fine micron filter and dried for overnight under vacuum to yield the final HZA product (0.024 g, 0.079 mmol, 70%). Chemical formula: C16H16N2O4, Molecular weight: 300.31; m.p. 204.6e 206.0C. IR (KBr, cm1): 3425, 3351, 3201, 2962, 2879, 1665, 1564, 1119; 1_{H NMR (300 MHz, d} 6-DMSO, 25C):

d

(ppm)¼ 8.60 (d, Jd¼ 8.9 Hz, 1H), 8.42 (d, Jd¼ 8.9 Hz, 1H), 8.18 (d, Jd ¼ 8.9 Hz, 1H), 7.64 (t, Jt¼ 8.9 Hz, 1H), 7.45(s, 2H, NH2), 6.83 (d, Jd¼ 8.1 Hz, 1H), 4.56 (s, 1H, OH), 4.19 (t, Jt¼ 6.0 Hz, 2H), 3.60 (t, Jt¼ 6.0 Hz, 2H), 3.44 (s, 4H);13C NMR (75 MHz, d6-DMSO3, 298 K):

d

(ppm)¼ 164.3, 163.3, 153.2, 134.5, 131.5, 130.2, 129.8, 124.4, 122.1, 119.8, 108.6, 107.8, 72.5, 67.5, 60.6, 38.8; MS (þESI-MS): (m/z): Calcd for C16H16N2O4; 300.31; found: 301.3 [M þ 1]þ, 323.1 [M þ Na]þ; Anal. Calcd. for C16H16N2O4: C, 63.99; H, 5.37 N, 9.33, found; C, 63.89; H, 5.35, N, 9.31.

3. Results and discussion

3.1. UVeVis and ﬂuorescence measurements of probe HZ

We primarily assessed the spectroscopic properties of the probe HZ in a mixture of phosphate buffer saline (PBS, pH¼ 7.2, 10 mM) and EtOH (1:9, v/v). The probe HZ (5

m

M) without hy-drazine exhibited a moderate UVevis absorption band and a ﬂuorescence emission band at 344 and 467 nm, respectively, owing to the electron withdrawing phthalimide protection group. However, upon the addition of hydrazine (20

m

M, 4.0 equiv) we noticed an immediate colour change from colourless (344 nm) to yellow colour (439 nm) with a noticeable red-shift in the ab-sorption band (Fig. S1). Concomitantly, a selective red-shift from 467 nm (blue) to 528 nm (yellowish green) was evidenced in the ﬂuorescence emission spectra (seeFig. S2). The perceptible red-shifts in both UVevis and ﬂuorescence cases could be ascribed to the hydrazine promoted phthalimide deprotection with the release of electron donating amino grouped compound HZA. These observations indicated that the current probe HZ could be employed as a sensitive ratiometric sensor under physiological conditions.

We further examinedﬂuorescence responses of the probe HZ over the various primary amine sources, such as hydroxylamine, urea, thiourea, monomethylamine, ethylenediamine, 1,4-diaminobutane, trans-1,2-diaminocyclohexane, ammonia, guani-dine nitrate, and hydrazine to substantiate the selectivity of probe HZ (Fig. 1(bed)). Upon the addition of 5 equiv of hydrazine, probe HZ (5

m

M) illustrated a discernible ratiometric red-shift inFig. 1(a).

However, under similar conditions the other primary amine sources merely showed trivial responses in the emission behaviour.

To evaluate the quantitative analysis of probe HZ, we further measured the absorption and ﬂuorescence changes of probe HZ (5

m

M) by increasing the hydrazine concentrations from 0 to 20

m

M. As shown inFig. S3, upon the addition of hydrazine we noticed a gradual decline in the absorption band at 344 nm and a simulta-neous increase of newly red-shifted absorption band at 439 nm. Likewise, theﬂuorescence emission band at 467 nm was gradually decreased with a concomitant upturn of a new red-shifted emis-sion band at 528 nm (Fig. 2(a)), indicating a lucid colorimetric and ratiometricﬂuorescence response of probe HZ.

With the addition of hydrazine (20

m

M), the emission intensity ratio at the two characteristic wavelengths of 467 and 528 nm increased to 15 fold (from 0.24 to 3.65). Importantly, the ﬂuores-cence response of probe HZ towards hydrazine showed a clear linear relationship (Fig. 2(b)) within the range of 0e7.5

m

M, which allowed us to determine the detection limit of probe HZ for hy-drazine (Fig. S4). Thus, the estimated detection limit was 4.2 nM (3

s

/slope), which is comparable to those of previously reported ratiometricﬂuorescent sensors for hydrazine[18b,18c]. Moreover,

Fig. 2. (a) Fluorescence spectra of probe HZ (5mM) upon the titration of hydrazine (0e 20mM) in a mixture of PBS buffer solutions (pH 7.2, 10 mM) in EtOH (1:9, v/v). Exci-tationl¼ 405 nm, Slit: 5 nm/5 nm; (b) Ratiometric calibration curve I528 nm/I467 nm.

(5)

the probe showed no interferences of other competing analytes and satisﬁed the monitoring of threshold limit value (10 ppb) of hy-drazine according to the U. S. Environmental Protection Agency (EPA)[25].

3.2. Theoretical and spectroscopic studies

To realize the ratiometric signalling mechanism of probe HZ, further strides were then made, in which the semi-empirical

Fig. 3. (a) and (b) Semi-empirical PM3 optimized HOMO and LUMO frontier molecular orbital distributions of probe HZ, respectively; (c) energy-minimized geometry of probe HZ. Dihedral angle between phthalimide and naphthalimide plane was denoted in the picture, Mulliken charges of phthalimide and naphthalimide carbonyl carbons were depicted in picture (c) Colour coding of atoms blue¼ N, red ¼ O, grey ¼ C, and white ¼ H, respectively. (For interpretation of the references to colour in this ﬁgure legend, the reader is referred to the web version of this article).

(6)

Fig. 4. (a) and (b) Semi-empirical PM3 optimized HOMO and LUMO frontier molecular orbital distributions of compound HZA, respectively; (c) energy-minimized geometry of compound HZA. Colour coding of atoms blue¼ N, red ¼ O, grey ¼ C, and white ¼ H, respectively. (For interpretation of the references to colour in this ﬁgure legend, the reader is referred to the web version of this article).

(7)

theoretical calculations of probe HZ and compound HZA were studied. We observed that both HOMO and LUMO molecular orbital distributions in probe HZ were mainly resided on the naphthalimide moiety as shown in Fig. 3, which possessed a dihedral angle of

F

¼ 71.8_{with the distorted phthalimide unit. A} further Mulliken charge analysis showed that the phthalimide carbonyl unit had a more electropositive character for carbon

d

þ ¼ 0.33 e in contrast to 0.31 e in the naphthalimide carbonyl unit.

The effective molecular orbital distribution across naph-thalimide moiety indicated a weak ICT between phnaph-thalimide and naphthalimide. As we anticipated the hydrazinolysis of probe HZ changed the fate such that, with an enabling ICT from free donor amine group to naphthalimide moiety in the resulted compound

Fig. 5.1_{H NMR (d}

6-DMSO, 300 MHz, 25C) stock plot. (a) pure HZA (3 mM), (b) probe HZ (3 mM) with the addition of hydrazine (2.0 equiv), and (c) probe HZ (3 mM).

(8)

HZA as depicted in Fig. 4. Moreover,1H NMR spectra of the free probe HZ, probe HZ-hydrazine complex, and isolated HZA in d6 -DMSO were compared in which the proton signals (a and b) cor-responding to phthalhydrazide along with free amino grouped naphthalimide derivative appeared in the spectrum with the addition of hydrazine to probe (Fig. 5). Furthermore, ESI-MS anal-ysis veriﬁed the release of phthalhydrazide during the hydrazine mediated phthalimide deprotection.

Based on these experimental and theoretical observations we outlined the plausible signalling mechanism. Deprotection of the phthalimide group of probe HZ proceeds ﬁrst at the carbonyl position of phthalimide by the nucleophilic addition of hydrazine to leave the intermediate 2-(hydrazinecarbonyl)-N-naphthalimidobenzamide. [26] The subsequent nucleophile attacked on the carbonyl group to generate phthalhydrazide and 4-aminonaphthalimide, which possessed a unique colorimetric and ratiometric response (Scheme 3).

3.3. Time effect on probe HZ ratiometric response

Considering the real-time detection of the probe HZ (6

m

M) towards hydrazine, time-dependentﬂuoresce ratiometric changes of two characteristic wavelengths at 467 and 528 nm in the pres-ence of hydrazine (30

m

M) in a mixture buffer (PBS, pH 7.2, 10 mM) and EtOH (1:9, v/v) solutions (Fig. S5(a)) were veriﬁed. Delightfully, within 15 min the ﬂuorescence ratiometric intensity (I528 nm/ I467 nm) was increased to 5 fold with a perceivable dynamic nature (Fig. S5(b)). Obviously, the crossover point at 750 s for the two characteristic wavelengths 467 and 528 nm indicated the release of compound HZA, with enabling an effective ICT induced hydrazine selective ratiometric response of the probe HZ. Thus, the probe HZ could be useful for real-time detection of trace amounts of hydrazine.

3.4. Screening of probe selectivity over competing cations and anions

To fortify the selectivity of probe HZ to other common cations and anions (10 equiv), we investigated theﬂuorescence behaviour of probe HZ. However, the tested cations, such as Naþ, Agþ, Ca2þ, Zn2þ, Cu2þ, Ni2þ, Cd2þ, Hg2þ, Pb2þ, Ag2þ, Fe3þand Al3þ, on the probe HZ and probe HZ-hydrazine complex could not induce any noticeable changes as shown inFig. S6. Similarly, we screened the effect of different anions, such as F, Cl, Br, I, AcO, NO3, H2PO4, N3, HCO3, ClO4, SO42and S2O82, on probe HZ and probe HZ-hy-drazine complex. As anticipated, none of these above anions could present distinct responses on probe HZ as well as probe HZ-hy-drazine complex as depicted inFig. S7. Based on these results it was inferred that probe HZ could be selectively and sensitively detect the hydrazine even in the presence of other competing cations and anions.

3.5. pH Effect on probe HZ and HZ-hydrazine complex

Since phthalimide was prone to basic hydrolysis and to further appreciate the probe HZ towards biological applications, we investigated the pH effects onfluorescence capabilities of HZ and HZ-hydrazine complex. The probe HZ possessed a stable response over the pH range of 1.0e10.0. Moreover, the HZ-hydrazine com-plex showed a stable ratiometric response within the biological pH range of 5.0_{e9.0 including acidic media as shown in}Fig. 6andFig. 7. However, both the free probe HZ and HZ-hydrazine complex dis-played a distinct ratiometric response (red colour) with a newly instigatedfluorescence band at 596 nm in high basic pH range of 12e14 in contrast to the green fluorescence of probe-HZ-hydrazine within the pH range of 1e10. This photophysical study gave a clue that the ratiometric response of probe HZ under basic hydrolysis condition was quite different in comparison with probe HZ hydrazinolysis.

Further spectroscopic (1H NMR & ESI-MS) analyses were con-ducted to obtain further insight into the distinctive ratiometric response of probe HZ in high basic solution. We noticed a different chemical shift pattern for probe HZ in the presence of high basic solution in contrast to hydrazinolysis as shown inFig. 8. A readily observed colour change from colourless to red as well bright red fluorescence compared with initial blue fluorescence was observed immediately after hydroxide addition in the NMR tube. ESI-MS analysis of probe HZ in both positive and negative modes were verified to know the plausible reaction fragments in acidic, basic, and hydrazinolysis conditions, as shown inFigs. S8eS10. Regardless of phthalimide itself in acidic nature, the probe could not present any observable changes in acidic media as we noticed in photo-physical titrations. However, we observed distinct mass fragments

Fig. 6. Fluorescence intensity changes of probe HZ (6mM) as a function of pH in a mixture of PBS buffer (pH 7.2, 10 mM) and EtOH (1:9), (v/v);lex¼ 405 nm, Slits: 5 nm/

5 nm.

Fig. 7. Fluorescence intensity changes of probe HZ (6mM) in the presence of hydrazine (12mM) as a function of pH in a mixture of PBS buffer (pH 7.2, 10 mM) and EtOH (1:9), (v/v);lex¼ 405 nm, Slits: 5 nm/5 nm.

(9)

in basic hydrolysis which was consistent with above photophysical and spectroscopic studies.

Based on this evidence we draw out the plausible intermediates in acidic, hydrazinolysis and basic conditions for probe HZ as depicted in Scheme 4. These results clearly suggested that the current probe could be employed in living cells with better cell permeability without interference from the pH effects within the biological pH range and acidic media. Although the probe has a pure aqueous solubility, it gave long-time and trivial responses towards hydrazine and other primary amines owing to the reduced nucleophilicity of amines by the strong water-amine H-bonding in aqueous solutions in contrast to ethanol buffer solutions.

3.6. Time resolved photoluminescence measurements

To corroborate the above photophysical studies, time resolved photoluminescence measurements (excited at 405 nm) were con-ducted for the free probe HZ, HZ-hydrazine complex, and HZ-hy-droxide complex by probing at 467, 528, and 596 nm, respectively. Probe HZ showed a monoexponentialﬂuorescence decay with the lifetime of (

s

1) 4.31 ns, but we observed a biexponential ﬂuores-cence decay with the life time values of

s

1 ¼ 7.91 ns (87.4%) and

s

2¼ 1.86 ns (12.6%) for HZ-hydrazine complex as shown inFig. 9. Lifetime component

s

2with the shorter value could be ascribed to the hydrazine mediated phthalimide deprotection with the release of electron donating amino group in compound HZA, and the lifetime component

s

1represents intrinsic fluorescence of naph-thalimide fluorophore. Significantly, we noticed a distinctive

lifetime pattern for HZ-hydroxide complex as shown inFig. S11. Time resolved_{ﬂuorescence became biexponential decay with the} life time values of

s

1¼ 18.38 ns (2.15%) and

s

2¼ 2.63 ns (97.85%). The larger variation of the lifetime components indicated a unique ratiometric response under the basic hydrolysis of probe HZ. Moreover, these drastic lifetime changes in contrast to HZ-hydra-zine could be attributed to the newly instigated redﬂuorescence peak at 596 nm. Based on the lifetime measurements, we can infer that probe HZ showed diverse and characteristic ratiometric re-sponses under hydrazinolysis and basic hydrolysis depending on the nucleophilicity of analytes. We were able to show the facile and differentiable ratiometric chemodosimeteric approach for the detection of hydrazine even in the presence of competing basic media.

3.7. Confocal imaging

Encouraged by the foregoing performance of probe HZ, we next sought to apply probe HZ forﬂuorescence ratiometric imaging of hydrazine in living cells. Hydrazine could be detected in the human cervical cancer cell line (HeLa cells). The Cells incubated with probe HZ (10

m

M) alone for 30 min at 37C showed blueﬂuorescence (Fig. 10(a) and (b)). However, a perceptible greenﬂuorescence was monitored in the cells after treatment with hydrazine (25

m

M) see

Fig. 10(c) and (d). Apparent changes denoted that probe HZ was cell membrane permeable and capable of ratiometric imaging of hy-drazine in the living cells.

Fig. 8.1_{H NMR (d}

6-DMSO, 400 MHz, 25C) stock plot. (a) pure HZA (3 mM); (b) probe HZ (3 mM) with the addition of hydrazine (2.0 equiv); (c) probe HZ with the addition of

(10)

4. Conclusions

In summary, we have developed a facile and sensitive fluo-rescent probe for hydrazine based on IngeManske hydrazinolysis method under mild conditions. The probe HZ showed a selective colorimetric and fluorescent ratiometric response towards hy-drazine in the semi-aqueous buffer solution with a low detection limit. The unique ratiometric response under basic hydrolysis further differentiated from probe HZ hydrazinolysis. Current probe showed a stable ratiometric response including acidic and biological pH ranges. Theoretical and time resolved photo-luminescence measurements further confirmed the distinctive ratiometric modes of probe HZ under hydrazinolysis and basic hydrolysis conditions. Hence, the hydrazinolysis-based ratio-metricfluorescent probe was developed for the first time in this report. Pivotal confocal imaging of hydrazine in living cells also demonstrated that probe HZ could be favourable for biological

(11)

Acknowledgements

We thank the National Science Council of Taiwan (ROC) for ﬁnancial supports of this project through NSC 101-2113-M-009-013-MY2 and NSC 99-2221-E-009-008-MY2.

Appendix A. Supplementary data

Supplementary data related to this article can be found athttp:// dx.doi.org/10.1016/j.dyepig.2013.11.015

References

[1] (a) Rettig W, Strehmel B, Schrader S, Seifert H. Applied ﬂuorescence in chemistry, biology and medicine. New York: Springer; 1999;

(b) Bufﬂe J, Horvai G. In situ monitoring of aquatic systems; chemical analysis and speciation. New York: Wiley; 2001;

(c) Diamond D, Coyle S, Scarmagnani S, Hayes J. Wireless sensor networks and chemo-/biosensing. Chem Rev 2008;108:652e79;

(d) Shapira A, Livney YD, Broxterman HJ, Assaraf YG. Nanomedicine for tar-geted cancer therapy: towards the overcoming of drug resistance. Drug Resist Updat 2011;14:150e63.

[2] (a) Schiessl HW. Kirk-Othmer encyclopedia of chemical technology. John Wiley & Sons, Inc.; 2000. p. 562;

(b) Garrod S, Bollard ME, Nicholls AW, Connor SC, Connelly J, Nicholson JK, et al. Integrated metabonomic analysis of the multiorgan effects of hydrazine toxicity in the rat. Chem Res Toxicol 2005;18:115e22.

[3] (a) Narayanan SS, Scholz F. A comparative study of the electrocatalytic ac-tivities of some metal hexacyanoferrates for the oxidation of hydrazine.

Electroanalysis 1999;11:465e9;

(b) Ragnarsson U. Synthetic methodology for alkyl substituted hydrazines. Chem Soc Rev 2001;30:205e13.

[4] (a) Hannum JAE. Recent developments in the technology of propellant hy-drazines. Chemical Propulsion Information Agency; June 1982. CPTR 82e15; (b) Schmidt EW. Hydrazine and its derivatives: preparation, properties, ap-plications. New York: Wiley; 1984;

(c) Wang J. Hydrazine detection using a tyrosinase-based inhibition biosensor. Anal Chem 1995;67:3824e7;

(d) Zelnick SD, Mattie DR, Stepaniak PC. Occupational exposure to hydrazines: treatment of acute central nervous system toxicity. Aviat Space Environ Med 2003;74:1285e91;

(e) Serov A, Kwak C. Direct hydrazine fuel cells: a review. Appl Catal B 2010;98:1e9.

[5] (a) Vernot EH, MacEwen JD, Bruner RH, Haus CC, Kinkead ER. Long-term inhalation toxicity of hydrazine. Fundam Appl Toxicol 1985;5:1050e64; (b) Lyon. International agency for research on cancer: re-evaluation of some organic chemicals, hydrazine, and hydrogen peroxideIn IARC monographs on the evaluation of carcinogenic risk of chemicals to humans, vol. 71. IARC; 1999. pp. 991e1013;

(c) Golabi SM, Zare HR. Electrocatalytic oxidation of hydrazine at a chlorogenic acid (CGA) modiﬁed glassy carbon electrode. J Electroanal Chem 1999;465: 168e76.

[6] (a) Hou W, Wang E. Flow-injection amperometric detection of hydrazine by electrocatalytic oxidation at a prussian blue-ﬁlm modiﬁed electrode. Anal Chim Acta 1992;257:275e80;

(b) Xia H, Li HL. Electrooxidation of hydrazine catalysed by 4-hydroxy-2,2,6,6-tetramethyl-piperidininyloxy (TEMPOL). J Electroanal Chem 1997;430:183e7; (c) Casella IG, Guascito MR, Salvi AM, Desimoni E. Catalytic oxidation andflow detection of hydrazine compounds at a nifion/ruthenium(III) chemically modified electrode. Anal Chim Acta 1997;354:333e41;

(d) Guerra SV, Xavier CR, Nakagaki S, Kubota LT. Electrochemical behavior of copper porphyrin synthesized into zeolite cavity: a sensor for hydrazine. Fig. 10. Confocal microscopic images of HeLa cells incubated with 10mM probe HZ for 30 min (b) and then further incubated with 25mM hydrazine for 30 min (d); (a), (c) Bright-ﬁeld transmission image of HeLa cells in (b) and (d), respectively.lex¼ 405 nm.

(12)

Electroanlaysis 1998;10:462e6;

(e) Li X, Zhang S, Sun C. Fabrication of covalently attached multilayerﬁlm electrode containing cobalt phthalocyanine and its electrocalalytic oxidation of hydrazine. J Electroanal Chem 2003;553:139e45;

(f) Revenga-Parra M, Lorenzo E, Pariente F. Synthesis and electrocalalytic ac-tivity towards oxidation of hydrazine a new family of hydroquinone salophen derivaives: application to the construction of hydrazine sensors. Sens Actua-tors B Chem 2005;107:678e87;

(g) Batchelor-McAuley B, Banks CE, Simm AO, Jones TGJ, Compton RG. The electroanalytical detection of hydrazine: a comparision of the use of palla-dium nanoparticles supported on boron-doped diamond and pallapalla-dium plated BDD microdisc array. Analyst 2006;131:106e10.

[7] Oh JA, Park JH, Shin HS. Sensitive determination of hydrazine in water by gas chromatography-mass spectrometry after derivatization of with ortho-phthalaldehyde. Anal Chim Acta 2013;769:79e83.

[8] He ZK, Fuhrmann B, Spohn U. Coulometric microﬂow titrations with chemi-luminescent and amperometric equivalence point detection: bromimetric titration of low concentration of hydrazine and ammonium. Anal Chim Acta 2000;409:83e91.

[9] Liu YY, Schmeltz I, Hoffmann D. Chemical studies on tobacco smoke: quan-titative analysis of hydrazine in tobacco and cigarette smoke. Anal Chem 1974;46:885e9.

[10] Thomas SW, Swager TM. Trace hydrazine detection withﬂuorescent conju-gated polymers: a turn-on sensory mechanism. Adv Mater 2006;18:1047e50. [11] Choi MG, Hwang J, Moon JO, Sung J, Chang SK. Hydrazine-selective chromo-genic andﬂuorogenic probe based on levulinated coumarin. Org Lett 2011;13: 5260e3.

[12] Lee MH, Yoon B, Kim JS, Sessler JL. Naphthalimide triﬂuoroacetyl acetonate: a hydrazine- selective chemodosimeteric sensor. Chem Sci 2013;4:4121e6. http://dx.doi.org/10.1039/c3sc51813b.

[13] (a) Chen X, Tian X, Shin I, Yoon J. Fluorescent and luminescent probes for detection of reactive oxygen and nitrogen species. Chem Soc Rev 2011;40: 4783e804;

(b) Chang JF, Zhou Y, Yoon J, Kim JS. Recent progress inﬂuorescent and colorimetric for detection of precious metal ions (silver, gold and platinum ions). Chem Soc Rev 2011;40:3416e29;

(c) Park S, Kim W, Swamy KMK, Lee HY, Jung JY, Kim G, et al. Rhodamine hydrazone derivatives bearing thiophene group asﬂuorescent chemosensors for Hg2þ. Dyes Pigment 2013;99:323e8.

[14] (a) Jiang J, Liu W, Cheng J, Yang L, Jiang H, Bai D, et al. A sensitive colorimetric and ratiometricﬂuorescent probe for mercury species in aqueous solution and living cells. Chem Commun 2012;48:8371e3;

(b) Das P, Mandal AK, Chandar NB, Baidya M, Bhatt HB, Ganguly B, et al. New chemodosimeteric reagents as ratiometric probes for cysteine and homocys-teine and possible detection in living cells and in blood plasma. Chem Eur J 2012;18:15382e93;

(c) Wang L, Zhou Q, Zhu B, Yan L, Ma Z, Du B, et al. A colorimetric and fluo-rescent chemodosimeter for discriminative and simultaneous quantification of cysteine and homocysteine. Dyes Pigment 2012;95:275e9.(d) Hu ZQ, Zhuang WM, Li M, Liu MD, Wen LR, Li CX. Highly sensitive and selective fluo-rescent chemodosimeter for Hg2þbased on thiorhodamine 6Geamide and its application for biological imaging. Dyes Pigment 2013;98:286e9;

(e) Huang L, Liu Y, Ma C, Xi P, Kou W, Zeng Z. Aﬂuorescent probe for the determination of Ce4þin aqueous media. Dyes Pigment 2013;96:770e3.

[15] Du J, Hu M, Fan J, Peng X. Fluorescent chemodosimeters using“mild” chemical events for the detection of small anions and cations in biological and envi-ronment media. Chem Soc Rev 2012;41:4511e35.

[16] (a) Maity D, Govindaraju T. A differentially selective sensor withﬂuorescence turn-on response to Zn2þand dual mode ratiometric response to Al3þin aqueous media. Chem Commun 2012;48:1039e41;

(b) Choi MG, Moon JK, Bae J, Lee W, Chang SK. Dual signalling of hydrazine by selective deprotection of chloroﬂuorescein and resoruﬁn acetates. Org Biomol Chem 2013;11:2961e5;

(c) Xu Y, Li B, Li W, Zhao J, Sun S, Pang Y.“ICT-not-quenching” near infrared ratiometric ﬂuorescent detection of picric acid in aqueous media. Chem Commun 2013;49:4764e6.

[17] Li K, Xu HR, Yu KK, Hou JT, Yu XQ. A coumarin based chromogenic and ratiometric probe for hydrazine. Anal Methods 2013;5:2653e6.

[18] (a) Chen X, Xiang Y, Li Z, Tong A. Sensitive and selectiveﬂuorescence deter-mination of trace hydrazine in aqueous solution utilizing 5-chlorosalcicylalehyde. Anal Chim Acta 2008;625:41e6;

(b) Fan J, Sun W, Hu M, Cao J, Cheng G, Dong H, et al. An ICT-based ratiometric probe for hydrazine and its application in live cells. Chem Commun 2012;48: 8117e9;

(c) Hu C, Sun W, Cao J, Gao P, Fan J, Song F, et al. A ratiometric near-infrared ﬂuorescent probe for hydrazine and its in vivo applications. Org Lett 2013;15: 4022e5.

[19] Khan MN. Kinetic evidence for the occurrence of a stepwise mechanism in hydrozinolysis of phthalimide. J Org Chem 1995;60:4536e41.

[20] (a) Duke MR, Veale EB, Kruger PE, Gunnlaugsson T. Colorimetric and ﬂuo-rescent anion sensors: an overview of in the recent developments in the use of 1,8-naphthalimide-based chemosensors. Chem Soc Rev 2010;39:3936e53; (b) Georgiev NI, Bojinov VB, Nikolov PS. The design, synthesis and photo-physical properties of two novel 1,8-naphthalimideﬂuorescent pH sensors based on PET and ICT. Dyes Pigment 2011;88:350e7;

(c) Wang M, Xu Z, Wang X, Cui J. Aﬂuorescent and colorimetric chemosensor for nitric oxide based on 1,8-naphthalimide. Dyes Pigment 2013;96:333e7. [21] March J. Advanced organic chemistry: reactions, mechanisms and structures.

2nd ed. Tokyo: McGraw-Hill Kogakusha, Ltd.; 1977. p. 388.

[22] (a) Bolhoter WA, Sheehan JC. An improved procedure for the condensation of potassium phthalimide with organic halides. J Am Chem Soc 1950;72: 2786e8;

(b) Gibson MS, Bradshaw RW. The Gabriel synthesis of primary amines. Angew Chem Int Ed 1968;7:919e30.

[23] Frisch MJ et al. Gaussian, Inc.: Wallingford CT; 2009.

[24] Tian H, Gan J, Chen K, He J, Song QL, Hou XY. Positive and negative ﬂuorescent imaging induced by naphthalimide polymers. J Mater Chem 2002;12:1262e7.

[25] (a) Shortreed M, Kopelman R, Kuhn M, Hoyland B. Fluorescentﬁber-optic calcium sensor for physiological measurements. Anal Chem 1996;68:1414e8; (b) Umar A, Rahman MM, Kim SH, Hahn YB. Zinc oxide nanonail based chemical sensor for hydrazine detection. Chem Commun 2008:166e8. [26] (a) Ing HR, Manske RHF. A modiﬁcation of the Gabriel synthesis of amines.

J Chem Soc 1926:2348e51;

(b) Arifﬁn A, Khan MN, Lan LC, May FY, Yun CS. Suggested improved method for the IngeManske and related reactions for the second step of Gabriel synthesis of primary amines. Synth Commun 2004;34:4439e45.