Introduction

Nitrite, a chemically active substance, exists widely in the environment and food products. In order to suppress the propagation of Clostridium botulinum and to improve the flavor of meat, it has been widely used in meat preservation and processing (Li et al. 2003). Excessive amount of nitrite in food can be fatal, especially for pregnant women and infants. It can interfere with the body’s oxygen delivery system, decreasing the ability of hemoglobin to carry oxygen (Seike et al. 2004; Cotton and Wilkinson 1988). It can also react with the secondary amines and amides in the stomach to form carcinogenic N-nitrosamines (Burden 1961). These hazards of nitrite make its determination and quantification of great interest.

A large number of analytical methods for the detection of nitrite have been developed, including spectrophotometric (Burakham et al. 2004; Nagaraja et al. 2010; Senra-Ferreiro et al. 2010; Tsikas 2007), electrochemical (Menart et al. 2015; Azad et al. 2014; Ramdane-Terbouche et al. 2014; Wang et al. 2013), capillary electrophoresis (Della Betta et al. 2014; Troška et al. 2013; Merusi et al. 2010; Wang et al. 2012b; Tanaka et al. 2004), chromatographic (Wang et al. 2012a; He et al. 2011; Akyüz and Ata 2009; Pagliano et al. 2014), and spectrofluorimetric methods (Lee and Field 1984; Guo et al. 2013; Liu et al. 2009; Huang et al. 2006; Wang et al. 2000). Each of them has its own merits, but each method also offers some drawbacks. Spectrophotometry is the most widely used method for the determination of nitrite. The most common approach to the spectrophotometric detection of nitrite is the Griess method which involves a diazotization-coupling procedure. However, this method suffers from poor sensitivity and interference from other chromophores and anions. Chromatography and capillary electrophoresis are always expensive and time-consuming; thus, the wide utilization of these methods is largely limited.

In comparison to these methods, spectrofluorimetry possesses more merits due to its convenience, simplicity, high sensitivity and selectivity, low limits of detection, and low-cost. Spectrofluorimetric methods are mainly based on the reaction of nitrite with various fluorescent probes such as 5-aminofluorescein (Axelrod and Engel 1975), 4-hydroxycoumarin (Ohta et al. 1986), unsymmetrical rhodamine (Liu et al. 2009), and murexide (Biswas et al. 2004). Probably the most successful fluorescent probes for detection of nitric oxide (NO) and nitrite have been aromatic vicinal diamines such as 5,6-diamino-1,3-naphthalene disulphonic acid (DANDS) (Wang et al. 2000), 2,3-diaminonaphthalene (DAN) (Damiani and Burini 1986; Tarigh and Shemirani 2014; Wiersma 1970), and diaminofluoresceins (DAFs) (Kojima et al. 1998; Kojima et al. 1999). It is believed that these probes can effectively react with NO in the presence of oxygen or with nitrite ion in acidic conditions to yield highly fluorescent triazole compounds (Itoh et al. 2000; Jourd’heuil 2002; Nagano and Yoshimura 2002; Zhang et al. 2004; Zhang et al. 2003).

4-Amino-5-methylamino-2′,7′-difluorofluorescein diacetate (DAF-FM DA), firstly synthesized by Nagano and co-workers in 1999 (Kojima et al. 1999), is the latest generation of fluorescent probes for quantitative detection of nitric oxide. DAF-FM DA is even more useful than other diaminofluoresceins since it is more sensitive for NO with a detection limit of 3 nM. The fluorescence intensity of the triazole form of DAF-FM DA is essentially independent of pH above pH 5.8 (Kojima et al. 1999). Furthermore, DAF-FM DA has been commercially available and frequently used for the detection of nitric oxide in living cells (Nagano and Yoshimura 2002). To the best of our knowledge, there is no report about its application in detection of nitrite. Considering the fact that NO could be generated from nitrite under acidic conditions, we speculated that the determination of nitrite with DAF-FM DA should be applicable and valuable. In this paper, the feasibility of this speculation was confirmed. In acidic medium, DAF-FM DA reacted with nitrite ion to yield highly fluorescent triazolofluorescein (DAF-FM T) and its fluorescence enhancement was directly proportional to the concentration of nitrite ion in the solution. The detection limit was estimated to be 3.3 × 10−8 mol L−1 (S/N = 3). It was demonstrated that DAF-FM DA could be applied to the determination of nitrite in real food samples such as meat products and pickled vegetables, with good precision, accuracy, and reproducibility.

Experimental

Apparatus

All fluorescence measurements were performed with a Cary Eclipse Fluorescence Spectrophotometer (Agilent Technologies, USA) equipped with a 1 cm × 1 cm quartz cell. The slit widths in terms of wavelength were 5 nm for excitation and emission, respectively. A FEP20-FiveEasy Plus pH meter (Mettler-Toledo Instruments (Shanghai) Co., LTD.) was used for the measurements of the pH. A thermostat bath model WB-2000 (Changchengkegongmao Co., Zhengzhou, China) maintained at the desired temperature was used for the experiments. All solutions were prepared in ultrapure water (R = 18.2 MΩ) purified by a Milli-Q Gradient system (Millipore, Molsheim, France). A Cary 50 UV–visible Spectrophotometer (Agilent Technologies, USA) with 1 cm matched quartz cell was used for measuring the absorbance. A LC-20A liquid chromatography (Shimadzu, Kyoto, Japan) with a Shimadzu LC-20AT pump, a Shimadzu SPD-20A UV–VIS detector (operated at 220 nm) and a SIL-20A autosampler was employed. A Waters Symmetry C18 column (150 mm × 4.6 mm i.d., 5 μm particle size) and mobile phase consisted of 12 mM tetrabutylammonium hydroxide, 0.125 vol.% methanol, and 10 mM KH2PO4 (pH 7.00) at a flow rate of 1.2 mL min−1 were used for separation of nitrite. Data acquisition and processing were accomplished with Shimadzu LC solution software.

Reagents

Chemicals of analytical reagent grade and Milli-Q water were used throughout.

A 1.0 × 10−3 mol L−1 stock solution of nitrite was prepared by dissolving 69.0 mg of sodium nitrite (pre-dried at 110 °C for 4 h) in Milli-Q water. A small amount of sodium hydroxide (about 5.0 mg) was added to prevent its decomposition. Twenty drops of chloroform (0.3 mL) were also added to inhibit the bacterial growth and thus make the nitrite solution stable (Lew 1977). The resulting solution was made up to the mark in a 1000-mL volumetric flask. This standard stock solution was prepared weekly and kept in a refrigerator at 4 °C. Working solutions were prepared freshly by an appropriate dilution of the stock solution.

Hydrochloric acid (0.4 mol L−1) was prepared from concentrated hydrochloric acid.

A 1.0 mol L−1 Tris–HCl buffer solution (pH 8.0) was purchased from Beijing Solarbio Company (Beijing, China).

The 5.0 × 10−5 mol L−1 DAF-FM DA solution was prepared by diluting 40 μL of 5 mM DAF-FM DA solution (Beyotime Institute of Biotechnology, Haimen, Jiangsu, China) to 4 mL with the corresponding diluents (Beyotime Institute of Biotechnology, Haimen, Jiangsu, China).

The salts including NaH2PO4, Na2HPO4, Na3PO4, NaNO3, NaHCO3, Na2CO3, Na2SO4, NaBr, CH3COONa, NH4Cl, KCl, LiCl, CaCl2 · 2H2O, Zn(CH3COO)2, FeCl3 · 6H2O, FeCl2 · 4H2O, CuCl2 · 2H2O, and MgCl2 · 6H2O were purchased from Beijing Chemical Works.

Four kinds of meat products (ham, sausage, beef, and salted fish) and three kinds of pickled vegetables (Korean Kimchi, Chinese sauerkraut, and preserved szechuan pickle) were randomly collected from local supermarket in Beijing, China.

p-Aminobenzenesulfonic acid solution (4 g L−1) and N-(1-naphthyl)-ethylenediamine solution (2 g L−1) were prepared by dissolving 0.4 g p-aminobenzenesulfonic acid or 0.2 g N-(1-naphthyl)-ethylenediamine in 100 mL water.

Tetrabutylammonium hydroxide was used as the ion-pairing agent for the HPLC separation of nitrite, and it was obtained as a 0.4-mol L−1 water solution from Acros (NJ, USA).

Procedure

Fifty microliters of 5.0 × 10−5 mol L−1 DAF-FM DA solution and 0.5 mL of 0.4 mol L−1 hydrochloric acid were added to a 10-mL colorimetric tube. An appropriate quantity of working solution of 1.0 × 10−5 mol L−1 sodium nitrite or sample solution was added. The mixture was diluted to 1.0 mL with Milli-Q water and mixed well. Then the tube was placed in a thermostat water bath at 60 °C for 30 min, cooled to room temperature, and made up to 3.0 mL with 1.0 mol L−1 Tris–HCl buffer solution (pH 8.0). The final pH of the solution was 7.8. The fluorescence intensity of the solution was measured at 517 nm, with excitation at 495 nm. All fluorescence intensity measurements were corrected with a blank.

Sample Preparation

Meat products were purchased from local supermarket in Beijing and triturated. Ten grams of each sample was weighed and placed in a beaker, then 25 mL of Na2B4O7 · 10H2O solution (5 %) and 350 mL of hot water were added. The mixture was placed in a hot water bath (80 ~ 90 °C) for 15 min, then cooled to room temperature, and transferred to a volumetric flask. Ten milliliters of potassium ferrocyanide aqueous solution and 10 mL of zinc acetate aqueous solution were added to precipitate the protein, and the volume was made up to 500 mL, followed by filtration through filter paper. The filtrate was stored at about 4 °C. Each time, an 80-μL portion of the filtrate was determined with the method described above.

Pickled vegetables were cleaned and dried with a blower in room temperature and crushed before use. Then 10.0 g of pickled vegetables was weighed and placed in a beaker, then 25 mL of Na2B4O7 · 10H2O solution (5 %) and 350 mL of hot water were added. The mixture was placed in a hot water bath (80 ~ 90 °C) for 15 min, then cooled to room temperature, and transferred to a volumetric flask. Ten milliliters of potassium ferrocyanide aqueous solution and 10 mL of zinc acetate aqueous solution were added, and the volume was made up to 500 mL, followed by filtration through filter paper. The filtrate was stored at about 4 °C. Each time, a 20-μL portion of the filtrate was determined with the method described above.

Results and Discussion

Spectral Characteristics

The proposed method is based on the diazotization reaction of DAF-FM DA with nitrite in HCl solution as shown in Scheme 1.

Scheme 1
scheme 1

Interaction of DAF-FM DA with nitrite in HCl solution to form DAF-FM T

Full size image

The excitation and emission spectra of DAF-FM DA in the presence and in the absence of nitrite were shown in Fig. 1. The excitation and emission maxima were found to be at 495 and 517 nm, respectively. Enhancement of fluorescence intensity was found when nitrite was added. Similar enhancement of fluorescence intensity was also found when DAF-FM DA serves as a fluorescent probe for nitric oxide (Kojima et al. 1999). This result suggested that DAF-FM DA was a potential probe for determining trace amount of nitrite in acidic conditions.

Fig. 1
figure 1

Fluorescence excitation and emission spectra of DAF-FM DA in the absence (1 and 2) and presence (1′ and 2′) of nitrite. C DAF-FM DA = 5.0 × 10−6 mol L−1; C HCl = 0.2 mol L−1; C nitrite = 1.0 × 10−6 mol L−1; reaction time, 30 min; reaction temperature, 60 °C

Full size image

Optimization of Reaction Conditions of DAF-FM DA with Nitrite

In order to find the optimum conditions for the method, the effects of variables such as solution acidity, interaction time, and temperature on the fluorescence enhancement of DAF-FM DA in the presence of nitrite were then systematically investigated.

Effect of Reaction Acidity

The reaction between DAF-FM DA and nitrite takes place in acidic medium. In the present work, HCl solution was chosen to obtain an acidic medium. The effect of acidity on the fluorescence enhancement was investigated by varying the concentration of HCl from 0.05 to 0.50 mol L−1. As depicted in Fig. 2, the fluorescence enhancement increased with the increase of HCl concentration from 0.05 to 0.15 mol L−1 and then kept constant above 0.15 mol L−1. Based on this observation and consideration of reasonable time necessary for a speedy analysis, 0.20 mol L−1 HCl solution was chosen in the subsequent experiments.

Fig. 2
figure 2

Effect of HCl concentration on the fluorescence enhancement. C DAF-FM DA = 2.5 × 10−6 mol L−1; C nitrite = 8 × 10−7 mol L−1; reaction time, 30 min; reaction temperature, 60 °C

Full size image

The pKa value of the phenolic OH group of DAF-FM was 4.38 ± 0.05, and the fluorescence intensity was stable above pH 5.8 (Kojima et al. 1999). Finally, the pH of the system was adjusted to 7.8 with 1.0 mol L−1 Tris–HCl buffer solution (pH 8.0).

Effect of Reaction Times and Temperatures

At the very low concentrations where the measurements were carried out, the reaction between DAF-FM DA and nitrite was found to be time- and temperature-dependent. Hence, the completion of the reaction was obtained by placing the colorimetric tube in a hot water bath. The effects of reaction time and temperature on the reaction of DAF-FM DA with nitrite were shown in Fig. 3. It was found that a higher temperature was necessary to get obvious fluorescence enhancement. But when the temperature was raised from 60 to 80 °C, the fluorescence enhancement was almost constant. At 60 ~ 80 °C, the fluorescence enhancement was almost unchanged after 30 min of reaction. Therefore, the optimized conditions for the analyzing process were found to be at 60 °C for 30 min.

Fig. 3
figure 3

Effect of time and temperature on the fluorescence enhancement. C DAF-FM DA = 2.5 × 10−6 mol L−1; C HCl = 0.20 mol L−1; C nitrite = 8 × 10−7 mol L−1

Full size image

Linearity and Detection Limit

Under the optimized conditions, fluorescence spectra of DAF-FM DA in the absence and in the presence of nitrite were shown in Fig. 4 in the concentration range of 0 to 1.5 × 10−6 mol L−1. The fluorescence enhancement was linear over nitrite concentration. Based on it, a linear calibration curve could be constructed in the range of 5.0 × 10−8 to 1.5 × 10−6 mol L−1. The concentration of nitrite could be calculated from the linear regression equation: ΔF = 220.8096 c + 5.1622 (R 2 = 0.9975), where ΔF is the fluorescence enhancement and c is the concentration of nitrite. The relative standard deviation (RSD) for six replicate determinations is 1.3 % for 5.0 × 10−7 mol L−1 nitrite. The limit of detection (LOD) is 3.3 × 10−8 mol L−1, which was evaluated using 3σ/s, where σ is the standard deviation of the blank signals and s is the slope of the linear calibration plot.

Fig. 4
figure 4

Emission spectra for DAF-FM DA in the presence of nitrite (×10−6 mol L−1). C DAF-FM DA = 2.5 × 10−6 mol L−1; C HCl = 0.20 mol L−1; reaction time, 30 min; reaction temperature, 60 °C; (a) 0; (b) 0.05; (c) 0.10; (d) 0.15; (e) 0.20; (f) 0.30; (g) 0.40; (h) 0.60; (i) 1.00; (j) 1.50

Full size image

In comparison with previously reported spectrofluorimetric methods in determination of nitrite, such as neutral red (Li et al. 2003), 5-aminofluorescein (Axelrod and Engel 1975), 4-hydroxycoumarin (Ohta et al. 1986), ADMND (Chen et al. 2007), and RB-PDA (Xue et al. 2012), the current method is more sensitive and simple.

Interference of Foreign Ions

In order to evaluate the selectivity of this new method, the effects of various common coexisting inorganic ions on the determination of nitrite were also studied by adding various foreign species into the solutions containing 0.5 μmol L−1 nitrite. The tolerance limits for the ions studied were summarized in Table 1. Most of anions can be allowed at concentrations of 2500 μmol L−1. The 1000 μmol L−1 of NH4 +, K+, Li+, Ca2+, and Zn2+ and 10 μmol L−1 of Fe3+, Fe2+, Cu2+, and Mg2+ had negligible interference. The results indicate that the selectivity of the present method is good.

Table 1 Effect of foreign ions
Full size table

Determination of Nitrite in Real Food Samples

To evaluate the viability of the proposed method for routine analysis, the proposed method was applied to determine nitrite in four kinds of meat products and three kinds of pickled vegetables. And for each kind of food, two different samples randomly collected in markets were analyzed. The recovery test was carried out by sparking the samples at one level, and the recoveries of nitrite ranged from 86.4 to 102.9 %. All samples were analyzed in sextuplicate, and the obtained RSD was below 5.1 %. In order to further validate the accuracy of the proposed method, all the samples were also analyzed simultaneously by Griess method and ion-pairing HPLC method. The results were given in Table 2. It can be seen that there were no significant differences between the results obtained from the proposed method and the two reference methods, indicating that the proposed method was reliable.

Table 2 Determination of nitrite in food samples by the proposed method
Full size table

Conclusions

A novel and sensitive spectrofluorimetric method for the determination of trace amount of nitrite with DAF-FM DA was developed. The mechanism for the fluorescence enhancement involves the formation of triazolofluorescein. The present method exhibited good selectivity, high sensitivity, and avoidance of coexisting substances interferences. The detection limit of this new method was lower or comparable to most of reported spectrofluorimetric methods. It was also proved useful in the detection and quantification of nitrite in real food samples.