A Novel Fluoroimmunoassay for Detecting Ruscogenin with Monoclonal Antibodies Conjugated with CdSe/ZnS Quantum Dots

Ruscogenin (RUS) is a steroidal sapogenin found in Ruscus aculeatus and Ophiopogon japonicus with several pharmacological activities. In the work reported herein, a novel method termed competitive fluorescence-linked immunosorbent assay (cFLISA) based on monoclonal antibodies (mAbs) coupled with quantum dots (QDs) was developed for the quick and sensitive determination of RUS in biological samples. The mAbs against RUS were conjugated with CdSe/ZnS QDs by the crossing-linking reagents and an indirect cFLISA method was developed. There was a good linear relationship between inhibition efficiency and logarithm concentration of RUS which was varied from 0.1 to 1000 ng/mL. The IC50 and limit of detection (LOD) were 9.59 ng/mL and 0.016 ng/mL respectively, which much lower than the enzyme-linked immunosorbent assay (ELISA) method. The recoveries in plasma and tissues were ranged from 82.3% to 107.0% and the intra- and inter-day precision values were below 15%. The developed cFLISA has been successfully applied to the measurement of the concentrations of RUS in biological samples of rats, and showed great potential for the tissue distribution study of RUS. The cFLISA method may provide a valuable tool for the analysis of small molecules in biological samples and such an approach could be applied to other natural products.


Introduction
Ruscogenin (RUS), a steroidal sapogenin isolated from Ruscusacu leatus and Ophiopogon japonicus, possesses a variety of pharmacological activities, including anti-inflammatory [1,2], platelet aggregation inhibition [3,4], pulmonary hypertension lowering [5,6], acute lung injury alleviating [7], liver injury attenuating [8] and other effects. Like many other steroidal saponins, there are no chromophoric groups in the structure of RUS, which leads to a lack of ultraviolet (UV) absorption. This factor inherently restricts the potential application of detection techniques involving UV to detect such kinds of steroidal saponins [9,10]. Although there is a large amount of research about the pharmacological functions and mechanism of RUS concerning cardiovascular and cerebrovascular diseases [11,12], the tissue distribution remains unknown. Nowadays, the analytical methods used to analyze RUS including HPLC-ELSD [13] and UPLC-MS/MS [14]. However, these methods are inconvenient due to their tedious sample pretreatment, large consumption of extraction solvents, time-consumption and expensive instruments. In addition to instrumental analysis, an indirect competitive ELISA for RUS was developed. Although this method is uniquely positioned as a simple and quick process compared with the instrumental analysis methods, it still required laborious and time-consuming incubation and washing steps. Additionally, enzyme-based labels suffer from instability due to denaturation and degradation. Therefore, it is necessary to establish a superior method to analyze RUS.
FLISA, is similar in principle to ELISA except that organic dyes take the place of enzymes. In the FLISA method, time-and cost-consuming secondary antibody reactions and subsequent enzyme-substrate reactions can be avoided, which obviously shortens the measurement time. However, there are still some limitations for the traditional FLISA based on the organic dyes such as low photoluminescence quantum yield and poor stability. Therefore, it is essential to search for new luminescent materials to replace the organic dyes to improve the stability of FLISA. QDs are semiconductor fluorescent nanoparticles which have many special properties such as high quantum yields, narrow and symmetric emission spectrum, broad excitation spectrum, large effective Stokes shifts and high photostability which makes them ideal for biological imaging and analysis [15][16][17]. In order to improve the sensitivity and selectivity, QDs are usually coupled with antibodies [16,18,19], which make it possible to complete the test time within one step. Nowadays, FLISA tests based on QDs and antibodies have been widely applied to the detection of low molecular weight analytes such as antibiotics, pesticides, hormones, toxins and so on and the LOD can be as low as pg level [20][21][22][23].
To overcome the problems existing in the detection of RUS, in the work reported here we combined a fluoroimmunoassay and QDs and established a novel cFLISA method to determine RUS with monoclonal antibodies conjugated with CdSe/ZnS quantum dots. This method was successfully applied to study the tissue distribution of RUS in Sprague-Dawley (SD) rats.

The Conjugation
Conjugation is one of the key step in the establishment of the FLISA method. Therefore, it is of vital importance to choose a proper conjugation procedure. Nowadays, many conjugation techniques have been developed [16] such as direct conjugation using active esters, direct conjugation with 4-(N-maleimidomethyl)cyclohexanecarboxylic acid N-hydrosuccinimide ester (SMCC), indirect conjugation using avidin as bridge protein and biotinylated antibodies, and indirect conjugation using modified protein G. As one of the first employed conjugation procedures, the active esters method based on the reagents of EDC and NHS has unique advantages such as being simple, cheap, requiring common reagents, and is frequently performed in an one-pot reaction. The method has been widely employed by many researchers to conjugate QDs and antibodies, such as QDs to specific antibodies against morphine [24], quinoxaline-2-carboxylic acid [25], methyl-3-quinoxaline-2-carboxylic acid [25], ochratoxin A [26,27] and so on. Based on the above, in this paper, we choose the active esters method to conjugate the mAb and QDs.
The carboxylate groups on the surface of CdSe/ZnS QDs can be conjugated with the amino groups of anti-RUS monoclonal antibodies by the EDC coupling method. As shown in Figure 1, the agarose gel electrophoresis image preliminarily verified that mAb was successfully linked to the surface of QD 620 because of the less migration of QD 620 -mAbs compared with the original QD 620 ( Figure 1A). In addition, the FTIR spectra of QD 620 -mAb and mAb showed characteristic amide bond peaks at 1652.95 cm (C=O amide I) and 1433.07 cm (N-C=O amide II), whereas QD 620 just shows the C=O stretch that can be attributed to the modified carboxyl group in the absence of the amide II signal ( Figure 1D). The presence of amide linkages on the QD 620 -mAbs demonstrated the successful covalent coupling between anti-RUS mAb and QD 620 . The fluorescence spectra of conjugates also had a max emission wavelength of 620 nm (λ ex = 400 nm), which was the same as CdSe/ZnS QDs, indicated that the excellent optical properties of QDs coupled with mAbs were still being remained ( Figure 1B). The fluorescence intensity of mAb-QDs was slightly weaker than CdSe/ZnS QDs due to two aspects, the loss of QDs and the fluorescent quenching during the course of reaction [28]. The TEM image ( Figure 1C) showed that the QD 620 -mAb conjugates are spherical crystals, uniform in size (average diameter about 5 nm), and uniformly dispersed without aggregation. The conjugation of QD620 and monoclonal antibody anti-RUS. (A) the gel electrophoretic mobility of QD620 before and after conjugation; (B) the fluorescence spectra of QD620 before and after conjugation (The excitation wavelength was set at 400 nm); (C) the TEM images of QD620-mAbs; (D) the FT-IR spectra of QD620, mAbs, and QD620-mAbs.

Binding Affinity of the Conjugated mAb
For the reason that orientation of the conjugation by EDC and NHS is not controlled, the reaction of the QD with the paratope of the antibody may occur, leading to poor antigen binding. To detect the biological activity of the mAb before and after QDs coupling, the indirect ELISA technique was conducted. In order to keep the recognition efficiency of the mAb during the conjugation process with the QDs, an excess amount of antibody was used. The results are shown in Figure 2. When diluted 400 times, the OD450 value of the conjugates decreased from 2.78 to 0.80 while that of the mAb decreased from 2.79 to 2.25, which indicated that QD620-mAbs still retained bioactivity to interact with their counterparts. On the other hand, the biological activity of the mAbs were somewhat decreased after conjugation since the unbounded mAb were removed by centrifugation during the conjugation process [18]. In addition, the non-oriented reaction between mAb and QDs the gel electrophoretic mobility of QD 620 before and after conjugation; (B) the fluorescence spectra of QD 620 before and after conjugation (The excitation wavelength was set at 400 nm); (C) the TEM images of QD 620 -mAbs; (D) the FT-IR spectra of QD 620 , mAbs, and QD 620 -mAbs.

Binding Affinity of the Conjugated mAb
For the reason that orientation of the conjugation by EDC and NHS is not controlled, the reaction of the QD with the paratope of the antibody may occur, leading to poor antigen binding. To detect the biological activity of the mAb before and after QDs coupling, the indirect ELISA technique was conducted. In order to keep the recognition efficiency of the mAb during the conjugation process with the QDs, an excess amount of antibody was used. The results are shown in Figure 2. (D) the FT-IR spectra of QD620, mAbs, and QD620-mAbs.

Binding Affinity of the Conjugated mAb
For the reason that orientation of the conjugation by EDC and NHS is not controlled, the reaction of the QD with the paratope of the antibody may occur, leading to poor antigen binding. To detect the biological activity of the mAb before and after QDs coupling, the indirect ELISA technique was conducted. In order to keep the recognition efficiency of the mAb during the conjugation process with the QDs, an excess amount of antibody was used. The results are shown in Figure 2. When diluted 400 times, the OD450 value of the conjugates decreased from 2.78 to 0.80 while that of the mAb decreased from 2.79 to 2.25, which indicated that QD620-mAbs still retained bioactivity to interact with their counterparts. On the other hand, the biological activity of the mAbs were somewhat decreased after conjugation since the unbounded mAb were removed by centrifugation during the conjugation process [18]. In addition, the non-oriented reaction between mAb and QDs When diluted 400 times, the OD 450 value of the conjugates decreased from 2.78 to 0.80 while that of the mAb decreased from 2.79 to 2.25, which indicated that QD 620 -mAbs still retained bioactivity to interact with their counterparts. On the other hand, the biological activity of the mAbs were somewhat decreased after conjugation since the unbounded mAb were removed by centrifugation during the conjugation process [18]. In addition, the non-oriented reaction between mAb and QDs may decrease the biological activity of antibody because the QDs occupy some of the antigen-binding sites of the mAb. However, the QDs conjugated mAb still had ideal binding affinity and could be used in the subsequent FLISA assay.

The Calibration Curve
The results are shown in Figure 3, and the calibration curve was then established with a perfect linear relationship (r 2 = 0.9903) between inhibition efficiency and logarithm of RUS concentration when the concentration varied from 0.1 to 1000 ng/mL. Hereinto, the IC 50 value and LOD were 9.6 ng/mL and 0.016 ng/mL, respectively. antigen-binding sites of the mAb. However, the QDs conjugated mAb still had ideal binding affinity and could be used in the subsequent FLISA assay.

The Calibration Curve
The results are shown in Figure 3, and the calibration curve was then established with a perfect linear relationship (r 2 = 0.9903) between inhibition efficiency and logarithm of RUS concentration when the concentration varied from 0.1 to 1000 ng/mL. Hereinto, the IC50 value and LOD were 9.6 ng/mL and 0.016 ng/mL, respectively.

The Comparison of the FLISA and ELISA
The FLISA method is superior to the ELISA one due to the simple steps and short time. In addition to this, the IC50 and LOD of the FLISA assay were much lower than those of the ELISA method according to previous studies [26,29]. In this study, we compared the linear range, IC50 and LOD of the FLISA and ELISA methods. As shown in Figure 4, the FLISA method has a broader linear range. The IC50 value and LOD of ELISA method were 433.4 ng/mL and 5.4 ng/mL, respectively, which are much higher than those of the FLISA method (Table 1). In conclusion, the FLISA based on the QD-mAb exhibited higher sensitivity than the competitive ELISA based on the RUS mAb.

The Comparison of the FLISA and ELISA
The FLISA method is superior to the ELISA one due to the simple steps and short time. In addition to this, the IC 50 and LOD of the FLISA assay were much lower than those of the ELISA method according to previous studies [26,29]. In this study, we compared the linear range, IC 50 and LOD of the FLISA and ELISA methods. As shown in Figure 4, the FLISA method has a broader linear range. The IC 50 value and LOD of ELISA method were 433.4 ng/mL and 5.4 ng/mL, respectively, which are much higher than those of the FLISA method (Table 1). In conclusion, the FLISA based on the QD-mAb exhibited higher sensitivity than the competitive ELISA based on the RUS mAb. The cross reactivity of several analogs of RUS were determined and the results shown in Table 2. Diammonium glycyrrhizinate, sarsasapogenin, notoginsenoside R1 and oleanolic acid which have different skeletons ( Figure 5) showed low cross-reactivity (lower than 0.1%). Diosgenin, just lacking one hydroxyl at the C-1 position in the skeleton compared with RUS only showed 22.3% cross-reactivity. These results indicated that RUS mAb had good specificity to the analogues in FLISA. The cross reactivity of several analogs of RUS were determined and the results shown in Table 2. Diammonium glycyrrhizinate, sarsasapogenin, notoginsenoside R1 and oleanolic acid which have different skeletons ( Figure 5) showed low cross-reactivity (lower than 0.1%). Diosgenin, just lacking one hydroxyl at the C-1 position in the skeleton compared with RUS only showed 22.3% cross-reactivity. These results indicated that RUS mAb had good specificity to the analogues in FLISA.

The Recovery, Precision, Accuracy in Plasma and Tissues of Rats
To reduce the matrix effects, the biological samples were diluted and the calibration curves were obtained by adding the standard substances into the samples. The concentration range of all calibration curves of RUS in plasma and tissues of rats were from 0.05 ng/mL to 5000 ng/mL and almost all the samples had good linear relationship with r 2 greater than 0.99 except the samples of liver and kidney (Table 3). Recovery in plasma and tissues were investigated at the three concentrations of RUS (1500 ng, 100 ng, 10 ng) and the results are summarized in Table 4. The recoveries in plasma and tissue samples ranged from 82.26% to 106.95% with RSD ≤ 13.0%. The intra-and inter-day precision and accuracy data of RUS are presented in Table 5. The precision was expressed as RSD% and accuracy expressed as RE%. These results demonstrated that the precision and accuracy values were within an acceptance range (<15%). The results showed that the FLISA method based on QD-mAb had good linear relations, higher precision and accuracy, and satisfactory recovery, which indicated that the method could be applied to the determination of RUS in biological samples.

The Recovery, Precision, Accuracy in Plasma and Tissues of Rats
To reduce the matrix effects, the biological samples were diluted and the calibration curves were obtained by adding the standard substances into the samples. The concentration range of all calibration curves of RUS in plasma and tissues of rats were from 0.05 ng/mL to 5000 ng/mL and almost all the samples had good linear relationship with r 2 greater than 0.99 except the samples of liver and kidney (Table 3). Recovery in plasma and tissues were investigated at the three concentrations of RUS (1500 ng, 100 ng, 10 ng) and the results are summarized in Table 4. The recoveries in plasma and tissue samples ranged from 82.26% to 106.95% with RSD ≤ 13.0%. The intra-and inter-day precision and accuracy data of RUS are presented in Table 5. The precision was expressed as RSD% and accuracy expressed as RE%. These results demonstrated that the precision and accuracy values were within an acceptance range (<15%). The results showed that the FLISA method based on QD-mAb had good linear relations, higher precision and accuracy, and satisfactory recovery, which indicated that the method could be applied to the determination of RUS in biological samples.      Table 5. Precision and accuracy for determination of RUS in rat plasma and tissues by FLISA Method (n = 6).

RUS Distribution Study
The FLISA method was successfully applied to investigate the tissue distribution of RUS and the study was determined at 0, 0.083, 0.167, 0.5, 1, 3, 6, 12, 24 and 48 h in brain, heart, kidney, liver, lung, spleen, stomach and muscle of SD rats and the oral administration of RUS was at a dose of 100 mg/kg. The results are shown in Figure 6. The RUS was rapidly and widely distributed in various tissues. The concentration of RUS was high in the liver, lung and fat, while obviously low in the heart, brain, kidney and spleen. The phenomenon indicated that extensive metabolism probably happened in the liver and the compound may accumulate in fat. The high concentration of RUS in the lung may suggest the potential target related to the pulmonary disease [5,6]. The low concentration in brain may indicate that the RUS can hardly cross through the blood brain barrier. The result also indicated that the RUS almost completely eliminated after 24 h in all the tissues.

RUS Distribution Study
The FLISA method was successfully applied to investigate the tissue distribution of RUS and the study was determined at 0, 0.083, 0.167, 0.5, 1, 3, 6, 12, 24 and 48 h in brain, heart, kidney, liver, lung, spleen, stomach and muscle of SD rats and the oral administration of RUS was at a dose of 100 mg/kg. The results are shown in Figure 6. The RUS was rapidly and widely distributed in various tissues. The concentration of RUS was high in the liver, lung and fat, while obviously low in the heart, brain, kidney and spleen. The phenomenon indicated that extensive metabolism probably happened in the liver and the compound may accumulate in fat. The high concentration of RUS in the lung may suggest the potential target related to the pulmonary disease [5,6]. The low concentration in brain may indicate that the RUS can hardly cross through the blood brain barrier. The result also indicated that the RUS almost completely eliminated after 24 h in all the tissues.

Apparatus
The fluorescence spectra of QDs were measured by a Varioskan Flash spectrometer (Thermo, Waltham, MA, USA). Absorbance was measured by an Epoch microplate reader (Biotech). The agarose gel images were obtained by gel imaging analysis system (Bio-Rad, Hercules, CA, USA) and the Fourier transform infrared (FTIR) spectra were recorded by a Nicolet Impactor instrument (Thermo). Particle size was determined by transmission electron microscopy (TEM, JEM-1010, JEOL, Tokyo, Japan). Opaque white and the transparent polystyrene 96-well microtiter plates were purchased from Costar (Corning, NY, USA). Deionized water was prepared by a Milli-Q water purification system (Millipore, Billerica, MA, USA).

Animals
SD rats (200 ± 10 g, male) were obtained from Shanghai Jiesijie Laboratory Animal Co. Ltd (Shanghai, China, certificate No. SCXK 2015-0016). Animals were housed in a temperature-controlled environment with a 12-h-light-dark cycle and allowed free access to food and water. Prior to

Apparatus
The fluorescence spectra of QDs were measured by a Varioskan Flash spectrometer (Thermo, Waltham, MA, USA). Absorbance was measured by an Epoch microplate reader (Biotech). The agarose gel images were obtained by gel imaging analysis system (Bio-Rad, Hercules, CA, USA) and the Fourier transform infrared (FTIR) spectra were recorded by a Nicolet Impactor instrument (Thermo). Particle size was determined by transmission electron microscopy (TEM, JEM-1010, JEOL, Tokyo, Japan). Opaque white and the transparent polystyrene 96-well microtiter plates were purchased from Costar (Corning, NY, USA). Deionized water was prepared by a Milli-Q water purification system (Millipore, Billerica, MA, USA).

Animals
SD rats (200 ± 10 g, male) were obtained from Shanghai Jiesijie Laboratory Animal Co., Ltd (Shanghai, China, certificate No. SCXK 2015-0016). Animals were housed in a temperature-controlled environment with a 12-h-light-dark cycle and allowed free access to food and water. Prior to experiments, animals were randomized into different experimental groups and the indices were measured blindly. All procedures for animals were performed in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals and related regulations of China Pharmaceutical University.

Conjugation of QDs with Anti-RUS Monoclonal Antibody
The anti-RUS-mAb was coupled with QD 620 through the method as previously described with some modifications [32,33]. Briefly, 10 µg EDC, 2.2 µg sulfo-NHS, and 0.1 nmol QD 620 were dissolved in 0.5 mL borate buffer (pH 5.0; 0.05 M) and the mixture were reacted at room temperature for 30 min with gentle shaking. Then the solution pH was adjusted to 7.4, and the anti-RUS-mAb was added to the ester-activated QD 620 followed by incubation at room temperature for 1 h. To block the excess carboxyl sites on the QD surface, the solution was mixed with 1000-fold molar of glu-NH 2 and incubation for another 1 h. Finally, the mixture was added by glu-NH 2 to the concentration of 2% (w/v), the pH was adjusted to 4.5. The products were separated by centrifugation at 20,000× g for 1 h at 4 • C and the precipitates were dispersed with borate buffer (pH 8.0; 0.05 M) and stored at 4 • C prior to use.

Identification of the Conjugates
The fluorescence spectra of QDs and conjugates were measured by the Varioskan Flash instrument with the excitation wavelength of 400 nm, an excitation slit of 2 nm, and the emission fluorescence intensity between 450 nm and 700 nm were recorded with an emission slit of 2 nm.
The QD-conjugates were further confirmed with agarose gel electrophoresis. Briefly, 0.25 g of agarose was added to 25 mL of Tris-acetate-EDTA buffer (TAE, pH 8.0) and heated until completely melted and then cooled down at room temperature in a gel tray with a 1.0 mm 11-well comb in forming the 1% (w/v) gel. For each well, 10 µL of the QDs/QD-conjugates samples were added in comb holes. The electrophoresis condition was 110 V, 80 mA for 30 min. The gel then imaged with 0.4 s exposure using the UV gel imaging system (BioRad). The FTIR were recorded within the range of 4000-400 cm −1 . Particle size and morphology were determined by TEM.

Binding Affinity of the Conjugated mAb
To test the biological activity of the mAb before and after QDs coupling, the indirect ELISA technique was applied. The procedure was as follows [29]: briefly, transparent polyethylene plates were coated with RUS-OVA (10 µg/mL, 100 µL/well) which was dissolved in carbonate-bicarbonate buffer (pH 9.6; 0.05 M) and incubated overnight. Then the plates were washed three times with PBST (PBS with 0.05% Tween-20, v/v, 0.01 M; pH 7.4) and blocked with 0.5% gelatin in PBST (150 µL/well) for 1 h at 37 • C. The QD-mAbs was diluted into several series by PBST and added to the wells (100 µL/well). After incubation at 37 • C for 2 h and three washes, the plate was added by goat anti-mouse IgG-HRP (1:1000 in PBST, 100 µL/well) and incubated for 1 h at 37 • C. After three washes, 100 µL/well of TMB solution (1 mg TMB dissolved in 1 mL DMSO and mixed with 9 mL citrate buffer and 10 µL H 2 O 2 ) was added and incubated for 20 min. Lastly, 2 M H 2 SO 4 (50 µL/well) was added to cease the color development and the absorbance was measured with a microplate reader (Thermo) at 450 nm. The FLISA method was as shown in Figure 7. The procedure was performed according to relevant references with some modifications [34]. Briefly, opaque white microtiter plates were coated with 100 µL RUS-BSA (10 µg/mL) in carbonate-bicarbonate buffer (pH 9.6; 0.05 M) and incubated at 4 • C for overnight. After three washes, the plate was blocked with 0.5% gelatin in PBST with 150 µL/well and incubated for 1 h at 37 • C. Then, 50 µL of standard serial dilutions of RUS were added into the plate followed by the 50 µL of diluted mAb-QD conjugates (1:100) and incubated for 2 h at 37 • C.
After three washes, the plate was added 100 µL of PBS (pH 7.4; 0.01 M) into each well and recorded the fluorescence intensity with excitation wavelength of 400 nm and emission wavelength of 620 nm.
2 h at 37 °C. After three washes, the plate was added 100 μL of PBS (pH 7.4; 0.01 M) into each well and recorded the fluorescence intensity with excitation wavelength of 400 nm and emission wavelength of 620 nm.

Competition Curves
Different concentrations of RUS ranging from 0 ng/mL to 10 μg/mL were dissolved in PBST which contained 10% methanol as the standard solution and determined by the method described above. Competitive curves were generated by plotting the logit of inhibition values versus the logarithm of RUS concentration. In the curves, IC50 values were determined as the concentration at which 50% inhibition by binding the antibody to the coating antigen.

The Comparison of the FLISA and ELISA
The comparison of the FLISA and ELISA was conducted by testing the same concentrations of RUS ranging from 0 ng/mL to 10 μg/mL with FLISA and ELISA methods, respectively. Then the competitive inhibition curves were plotted.

Cross Reaction of the Analogues of RUS
The cross reactivity was determined as the IC50 values of analogues of RUS which required for 50% inhibition of the binding of the antibodies to the solid-phase antigen, as compared to RUS itself (100%), which responded to the following formula: Cross reactivity = [IC50 (analogues)/IC50 (RUS)] × 100%.

The Method Assessment
The study of the validation and reliability of the FLISA method including several parameters such as recovery of RUS spiked in the biological samples, intra-and inter-day assay precision and LOD of the assays.

Recovery Analysis
Biological samples of blood, brain, heart, kidney, liver, lung, spleen, fat, spinal cord and muscle were collected from six blank SD rats. The blood samples were centrifuged at 3000 rpm for 15 min at 4 °C to obtain the plasma. All tissues sample were thoroughly rinsed in ice cold saline to eliminate blood and dried with filter paper. An accurately weighted amount of each tissue sample was homogenized in PBS (0.01 M, pH 7.4, 3:1, w/v). The blank samples were spiked with RUS to achieve concentrations of 1500, 100 and 10 ng/mL. Then methanol (4:1, w/v) was added into and vortex-mixed for 30 s. The mixture was centrifuged at 12,000 rpm for 15 min at 4 °C and the supernatant was transferred to another tube. The supernatant was diluted 8 times to yield sample solutions in 10% methanol. The recovery was calculated as the percentage ratio of the determined concentration and added concentration.

Competition Curves
Different concentrations of RUS ranging from 0 ng/mL to 10 µg/mL were dissolved in PBST which contained 10% methanol as the standard solution and determined by the method described above. Competitive curves were generated by plotting the logit of inhibition values versus the logarithm of RUS concentration. In the curves, IC 50 values were determined as the concentration at which 50% inhibition by binding the antibody to the coating antigen.

The Comparison of the FLISA and ELISA
The comparison of the FLISA and ELISA was conducted by testing the same concentrations of RUS ranging from 0 ng/mL to 10 µg/mL with FLISA and ELISA methods, respectively. Then the competitive inhibition curves were plotted.

Cross Reaction of the Analogues of RUS
The cross reactivity was determined as the IC 50 values of analogues of RUS which required for 50% inhibition of the binding of the antibodies to the solid-phase antigen, as compared to RUS itself (100%), which responded to the following formula: Cross reactivity = [IC 50 (analogues)/IC 50 (RUS)] × 100%.

The Method Assessment
The study of the validation and reliability of the FLISA method including several parameters such as recovery of RUS spiked in the biological samples, intra-and inter-day assay precision and LOD of the assays.

Recovery Analysis
Biological samples of blood, brain, heart, kidney, liver, lung, spleen, fat, spinal cord and muscle were collected from six blank SD rats. The blood samples were centrifuged at 3000 rpm for 15 min at 4 • C to obtain the plasma. All tissues sample were thoroughly rinsed in ice cold saline to eliminate blood and dried with filter paper. An accurately weighted amount of each tissue sample was homogenized in PBS (0.01 M, pH 7.4, 3:1, w/v). The blank samples were spiked with RUS to achieve concentrations of 1500, 100 and 10 ng/mL. Then methanol (4:1, w/v) was added into and vortex-mixed for 30 s. The mixture was centrifuged at 12,000 rpm for 15 min at 4 • C and the supernatant was transferred to another tube. The supernatant was diluted 8 times to yield sample solutions in 10% methanol. The recovery was calculated as the percentage ratio of the determined concentration and added concentration.

The Intra-and Inter-Day Precision and Accuracy
The intra-day assay was carried out by the measurement of the spiked samples three times within a day. The inter-day assay measured the spiked samples during three continuous days. The precision was showed as relative standard deviation (RSD) and the accuracy was as the relative error (RE). The value of RE was calculated as the following function: In which x was the added value and µ was the detection value.

The LOD
The LOD was the mean value of six blank samples plus three times standard deviations on the standard curve.

The Tissue Distribution Study
For the tissue distribution study, 60 SD rats randomly were divided into 10 groups (n = 6). Samples of the brain, heart, kidney, liver, lung, spleen, stomach and muscle were collected at designated times (0, 0.083, 0.167, 0.5, 1, 3, 6, 12, 24 and 48 h) following euthanasia by CO 2 . Every tissue sample was thoroughly rinsed in ice cold saline to eliminate blood and other content and blotted dry with filter paper. An accurately weighted amount of each tissue samples were homogenized in PBS (1:3, w/v; 0.01 M; pH 7.4). The plasma and tissue homogenates were added into methanol (1:4, v/v) and vortex-mixed for 30 s. Then the mixture was centrifuged at 12,000 rpm for 15 min at 4 • C and the supernatant was diluted eight times to yield sample solutions in 10% methanol [34].

Statistical Analysis
All experimental data were processed by Microsoft Excel 2010 and Origin Pro 8.0 software. All data was obtained from three independent experiments and presented as the mean ± SD.

Conclusions
A sensitive and rapid FLISA method for the quantitative analysis of RUS was developed and successfully applied for the analysis of plasma and tissue samples in this study. The anti-RUS mAb could be successfully conjugated to CdSe/ZnS QDs by the catalysis of EDC. The conjugates retained biological activity without changing the optical characteristics. The FLISA method showed lower IC 50 and LOD compared with traditional ELISA. The obtained tissue distribution data showed that RUS could be absorbed and distributed quickly in in brain, heart, kidney, liver, lung, spleen, stomach and muscle with a peak concentration of 138.5 µg/mL in liver and minimum concentration of 1.06 µg/mL in the brain.
The tissue distribution study indicated that the mainly metabolization pathway of RUS was in liver and build up in fat, while it could hardly cross the blood-brain barrier. In conclusion, a new FLISA method based on the QD-mAb made the fast and sensitive detection of RUS come true. Furthermore, this study provided a new method for the study of tissue distribution of RUS and this method could be applied to the study of other natural products.