Lentinan with an average molecular weight of 630,000 Da (purity > 98%) was purchased from Yuanfu Pharmaceutical Science and Technology Co. Ltd. (Shanghai, China). Aniline blue was purchased from Nanjing Wanqing Chemical Glassware Instrument Co. Ltd. (Nanjing, China). Glucose, sucrose, maltose and starch were purchased from Sinopharm Chemical Reagent Co. Ltd. (Nanjing, China). Xylan and xanthan gum were purchased from Aladdin Industrial Co. Ltd. (Shanghai, China). Oral health tonic liquid was supplied by Infinitus Co. Ltd. (Guangzhou, China). All reagents were of at least analytical grade and used without further purification. All solutions were prepared with ultrapure water (18.2 MΩ) purified by a Millipore Simplicity System (Millipore, Bedford, USA).

Fluorescence intensity and spectra were recorded by Varioskan Flash Multimode Reader (Thermo Scientific). pH meter (Delta 320) was applied to measure the pH values of solutions.

A series of glycine-NaOH buffer solutions with different pH values were prepared by mixing 0.2 mol/L glycine with NaOH or HCl solutions at different proportions, and the pH values were adjusted by pH meter. 100 μg/mL lentinan stock solution was prepared by dissolving 0.0100 g of lentinan in a 0.5 mol/L NaOH solution and adjusted to pH 10 with 0.2 mol/L glycine, and then diluted with 0.1 mol/L glycine-NaOH buffer solution in a 100 mL volumetric flask. The lentinan stock solution was stored in 4˚C. 0.1% (w/v) of aniline blue stock solution was prepared by dissolving 0.1000 g of aniline blue with 0.1 mol/L glycine-NaOH buffer in a 100 mL volumetric flask and stored in dark to preserve it from light.

Different volumes (0 - 6 mL) of lentinan stock solution were mixed with 4 mL aniline blue solution (0.1%), and the mixtures were diluted to 10 mL with pH 10 glycine-NaOH buffer solution in brown volumetric flasks. The resulted solutions were heated at 80˚C for 15 min and then gradually cooled to room temperature. Afterwards 100 μL mixture per well was added into the 96-well black plate, and the fluorescence intensity was measured by Varioskan Flash Multimode Reader. The samples were excited at 404 nm and the emission was collected at 492 nm. The calibration curve of lentinan was constructed by plotting the relative fluorescence intensity against the corresponding lentinan concentration using linear regression model. The limit of detection (LOD) was determined as the concentration of the lentinan in the solution when the signal to noise ratio (S/N) was approximately 3.

0.4 mL lentinan stock solution was mixed with different volumes (0, 0.2, 0.4, 0.6, 0.8, 1, 1.2, 1.4, 1.6, 1.8, and 2.0 mL) of 1 mg/mL other potentially interferential polysaccharides including glucose, sucrose, maltose, starch, cellulose, xanthan gum, and xylan. Then 4 mL 1% aniline blue stock solution was added, and the mixture was diluted to 10 mL with pH 10 glycine-NaOH buffer solution. The resulted solution was heated at 80˚C for 15 min and gradually cooled to room temperature to perform fluorescence assay. In addition, dimethyl sulfoxide (DMSO) was added to investigate the specific binding between aniline blue and the lentinan with triple helical conformation. 1.0 mg lentinan was dissolved in different volumes of DMSO and diluted to 10 mL using pH 10 glycine-NaOH buffer solution with the DMSO concentration of 0, 2%, 10%, and 50%, respectively. Taken 0.4 mL solution into 4 mL 0.1% aniline blue stock solution, and the mixture was diluted to 10 mL with pH 10 glycine-NaOH buffer solution to perform fluorescence assay.

Different volumes (0, 0.25, 0.50 and 0.75 mL) of 0.1 mg/mL lentinan solution were mixed with 0.25 mL health tonic solution, and then 4 mL 0.1% aniline blue stock solution was added. The solution was diluted to 10 mL with pH 10 glycine-NaOH buffer solution. The established fluorescence method was performed to obtain the recovery to evaluate the accuracy of the method.

Fluorescence spectra of aniline blue in the absence and presence of lentinan were measured to investigate the feasibility of aniline blue fluorescent method to detect lentinan. As shown in Figure 1, the excitation and emission of aniline blue was very weak in aqueous solution, which could be attributed to the intramolecular C-C rotation of aniline blue. After addition of lentinan, strong fluorescence

enhancement was observed which indicated the formation of aniline blue/lentinan complex. The result suggested that upon binding to the cavity of triple helical structure of lentinan, the intramolecular C-C rotation of aniline blue was inhibited which increases the coplanarity and leaded to an enhanced fluorescence for lentinan detection. For the following experiments, 404 nm and 492 nm were chosen as the excitation and emission wavelength, respectively.

Several detection conditions of the aniline blue fluorescent method that may affect the assay of lentinan were investigated, including the concentration of aniline blue stock solution, incubation pH, temperature, and time. When the concentration of aniline blue stock solution was lower than 0.1% (w/v), the fluorescence intensity of aniline blue/lentinan complex increased along with the increasing concentration of aniline blue (Figure 2(a)). When the concentration was higher than 0.1% (w/v), the fluorescence intensity decreased sharply, which may due to the self-quenching effect of aniline blue molecules. Thus, a concentration

of 0.1% (w/v) aniline blue stock solution was selected for the following experiments.

The effect of pH on the fluorescence intensity of aniline blue/lentinan complex was also investigated. The fluorescence intensity was observed in glycine-NaOH buffer solution with the pH ranging from 2 to 12 (Figure 2(b)). The results showed that the fluorescence intensity was very weak under weakly acidic medium. This observation indicated that the acidic pH could influence the intra- and inter-molecular hydrogen bonds of lentinan, which damaged the triple helical structure of lentinan [28] and inhibited the formation of aniline blue/lentinan complex. However, the fluorescence intensity increased conspicuously with the pH from 6 to 12, and the intensity reached a maximum value at pH 10 which was selected as the optimum incubation pH.

The effect of incubation temperature from 20˚C to 90˚C on the fluorescence intensity was examined (Figure 2(c)). The fluorescence intensity increased gradually along with the temperature from 20˚C to 80˚C, suggesting that higher temperature could prompt the molecular motion of aniline blue to bind with lentinan. While the temperature was higher than 80˚C, the fluorescence intensity unexpectedly declined, indicating that excessive temperature may lead to the nonradiative transition probability of aniline blue molecules. For the following experiments, 80˚C was chosen as the optimum reaction temperature.

The influence of time was also investigated. The fluorescence intensity increased rapidly within 10 min (Figure 2(d)). After 15 min, it remained nearly constant, indicating a fast binding between aniline blue and lentinan. Therefore, an optional incubation time of 15 min was selected to ensure a rapid detection of lentinan.

To evaluate the sensitivity of this aniline blue fluorescent approach for quantitative analysis under the optional conditions, the assay was practiced at different amounts of lentinan. As shown in Figure 3, a good linear relationship between relative fluorescence intensity and lentinan concentrations was obtained over the range from 1 to 60 μg/mL (R² = 0.9997). The LOD (the concentration when S/N was approximately 3) was determined to be as low as 0.25 μg/mL, demonstrating the high sensitivity of the aniline blue fluorescent method.

Discrimination of lentinan from a large excess of other saccharides is a critical issue for lentinan detection. We examined the selectivity of the aniline blue fluorescence method for lentinan by detecting the response of lentinan in the presence of potentially interfering saccharides, including glucose, sucrose, maltose, cellulose, starch, xanthan gum, and xylan. The results showed that almost no fluorescence change was observed for the addition of those saccharides (Figure 4), suggesting the presence of these interferences did not affect the quantitative determination of lentinan. Furthermore, DMSO as a powerful hydrogen bond acceptor was added to break the triple helical structure of lentinan to be coil conformation [9] [29] [30] . With the increasing concentration of DMSO from 0 to 50%, the fluorescence intensity decreased significantly (Figure 5), demonstrating

the specificity of binding between aniline blue and triple helical structure of lentinan. Glucose, sucrose, maltose, and cellulose were dissociative in aqueous solution without helical structure. The conformations of starch and xanthan gum were single helix and double helix, respectively. Thus, aniline blue could not bind to the cavity of the single or double helical structure. Although xylan possessed triple helical conformation, the glucan was not β-(1,3)-(1,6). Therefore, the excellent selectivity could be attributed to the specificity of complementary structure between aniline blue and the unique triple helical structure formed by β-(1,3)-(1,6)-glucan of lentinan in aqueous solution, which ensured the selectivity of this approach for lentinan products determination.

In order to examine the performance of this method for practical applications, the recovery test in health tonic solution was conducted. The recovery and RSD were analyzed by the standard addition method, and the results were listed in Table 1. The recovery was 98.9% - 103% with an average recovery rate of 101%. The relative standard deviation was 1.73%. These results indicated that the presence of lentinan in health tonic solution could be precisely detected. Furthermore, aniline blue fluorescent method and phenol-sulfuric acid method as a common measurement were used to determine the original lentinan concentration in 10 samples of health tonic solution, respectively. The results showed that the concentrations of lentinan determined by aniline blue method were 197~236 μg/mL with the average concentration of 216 μg/mL (Table S1 in Supporting Information). As a control group, the concentrations were detected to be 3087 - 3766 μg/mL with the average value of 3369 μg/mL. The much higher obtained concentration could be attributed that phenol-sulfuric acid method detected the hydrolyzed glucose content which could not eliminate other interfering substances like oligosaccharide. Aniline blue fluorescent method was suitable for the determination of triple helical lentinan, demonstrating the applicability of this method in complexed lentinan contained products.