Flavonoids from Fig (Ficus carica Linn.) Leaves: The Development of a New Extraction Method and Identification by UPLC-QTOF-MS/MS

Flavonoid-rich leaves of the Ficus carica L. plant are usually discarded as waste. In this work, ultrasonic enzyme-assisted aqueous two-phase extraction (UEAATPE) was proposed as an innovative method to estimate the total flavonoids present in F. carica L. leaves. Total flavonoids were analyzed qualitatively and quantitatively by UPLC-QTOF-MS. At 38% (w/w) ethanol/18% (w/w) ammonium sulfate, we achieved the optimum conditions in which to establish an easy-to-form aqueous two-phase extraction (ATPE) as the final system. The optimal UEAATPE conditions were set at an enzymatic concentration of 0.4 U/g, 150 min enzymolysis time, an enzymolysis temperature of 50 °C, a liquid–solid ratio of 20:1 (mL/g), and 30 min ultrasonic time. The yields of the total flavonoids, i.e., 60.22 mg/g, obtained by UEAATPE were found to be 1.13-fold, 1.21-fold, 1.27-fold, and 2.43-fold higher than those obtained by enzyme-assisted ATPE (EAATPE), ultrasonic-assisted ATPE (UAATPE), ATPE, and soxhlet extraction (SE) methods, respectively. Furthermore, eleven flavonoids from the leaves of the F. carica L. plant were completely identified and fully characterized. Among them, ten flavonoids have been identified for the first time from the leaves of the F. carica L. plant. These flavonoids are quercetin 3-O-hexobioside-7-O-hexoside, 2-carboxyl-1,4-naphthohydroquinone-4-O-hexoside, luteolin 6-C-hexoside, 8-C-pentoside, kaempferol 6-C-hexoside-8-C-hexoside, quercetin 6-C-hexobioside, kaempferol 6-C-hexoside-8-C-hexoside, apigenin 2″-O-pentoside, apigenin 6-C-hexoside, quercetin 3-O-hexoside, and kaempferol 3-O-hexobioside. Therefore, F. carica L. leaves contain new kinds of unidentified natural flavonoids and are a rich source of biological activity. Therefore, this research has potential applications and great value in waste handling and utilization.


Introduction
Plant-derived organic waste mainly includes crop stalks, leftover branches and wood strips, fallen leaves, dry vines, weeds, and nut shells from the production process. Among the large traditional agricultural countries, agricultural organic waste has the following four characteristics: large quantity, poor quality, low price, and harmful properties [1]. In most agricultural organic waste treatment processes, the treatment efficiency is not high, and the environmental damage caused by improper treatment methods is relatively serious [2]. Most agricultural organic waste, due to its relatively abundance, can help protect the environment and save energy while improving comprehensive utilization of agricultural organic wastes. To a total of 0.2 g fig leaf powder, 4.0 g disodium hydrogen phosphate-citric acid buffer solution and 0.3 U/g cellulase were first added in a 10 mL graduated test tube and then mixed evenly by a vortex mixer and placed in a water bath at a certain constant temperature. Then, ammonium sulfate and ethanol were added into the enzymatic slurry and vibrated for 10 min by a vortex mixer to completely dissolve the salt. The suspension was given ultrasonic (100 w) treatment for 30 min. After ultrasonic treatment, the mixture was mixed well and then set at room temperature for 30 min to form an aqueous two-phase system.

Determination of Total Flavonoids
The total flavonoids were determined by the method described in [28]. The extracted solutions (0.3 mL) were transferred to a 10 mL test tube, to which sodium nitrite solution (5%; 0.3 mL) was added. The mixture was allowed to stand for 6 min, and then 0.3 mL of 10% aluminum nitrate solution was added. After another 6 min, this was followed by the addition of 4 mL of 4% sodium hydroxide solution. The absorbance of the mixture was To a total of 0.2 g fig leaf powder, 4.0 g disodium hydrogen phosphate-citric acid buffer solution and 0.3 U/g cellulase were first added in a 10 mL graduated test tube and then mixed evenly by a vortex mixer and placed in a water bath at a certain constant temperature. Then, ammonium sulfate and ethanol were added into the enzymatic slurry and vibrated for 10 min by a vortex mixer to completely dissolve the salt. The suspension was given ultrasonic (100 w) treatment for 30 min. After ultrasonic treatment, the mixture was mixed well and then set at room temperature for 30 min to form an aqueous two-phase system.

Determination of Total Flavonoids
The total flavonoids were determined by the method described in [28]. The extracted solutions (0.3 mL) were transferred to a 10 mL test tube, to which sodium nitrite solution (5%; 0.3 mL) was added. The mixture was allowed to stand for 6 min, and then 0.3 mL of 10% aluminum nitrate solution was added. After another 6 min, this was followed by the addition of 4 mL of 4% sodium hydroxide solution. The absorbance of the mixture was measured at 510 nm using a UV-Vis spectrophotometer (Perkin-Elmer Lambda 25, Waltham, MA, USA). After 15 min, the flavonoid contents in the extracts were determined in comparison to a standard curve that was plotted using rutin. The results were the averages of triplicate analyses. The calibration curve was obtained using rutin as the standard as shown in Table 1. Then, the extraction yield of total flavonoids was calculated according to Equation (1): where Y (mg/g) represents the yield of total flavonoids; C t (mg/mL) represents total flavonoid concentrations in the top phases; and V t (mL) is the volume of the top phases. M t (g) is the total mass of the fig leaf powder.

Experiment Design of UEAATPE
After testing in a single-factor experiment, four factors were selected and combined in the proposed methods to assess the main role of BBD and its interactions. These were X 1 : 36-40% ethanol concentration; X 2 : 16-20% ammonium sulfate concentration; X 3 : 10:1-30:1 (mL/g) liquid-solid ratio; X 4 : 20-40 min ultrasonic time. Y represents the yield of total flavonoids in different ranges.

Comparison of Different Extraction Methods
The extraction yield of total flavonoids was compared with UEAATPE, EAATPE, UAATPE, ATPE, and SE. The mass fractions of ethanol and ammonium sulfate were 38% (w/w) and 18% (w/w), respectively. The other fixed extraction conditions under optimization were enzymatic hydrolysis for 180 min at a 0.4 U/g cellulase concentration, a fixed temperature of 50 • C, an extraction time of 30 min, and a liquid-solid ratio of 20:1 mL/g.

The Analysis of UPLC-QTOF-MS/MS
The supernatant was evaporated until dry with a rotary evaporator (RE-52AA, Shanghai Huxi Instrument, Shanghai, China) under reduced pressure in a 60 • C water bath. The suspended sample was re-dissolved in a methanol solution. Then, before the UPLC-QTOF-MS analysis, the solution was filtered through a 0.22 µm microporous membrane.
The samples were separated on an ACQUITY UPLC HSS T3 (150 mm × 2.1 mm i.d., 1.8 µm, Waters). The column temperature was maintained at 50 • C, and the injection volume was set as 2 µL with 0.3 mL/min as a fixed flow rate. The mobile phase was composed of acetonitrile (A) and 0.1% (v/v) formic acid in aqueous solution (B). The gradient elution conditions were as follows: 0-2 min, 5% A; 2-25 min, 5-40% A; 25-32 min, 40-95% A. The chromatogram was obtained at 254 nm, and semi-quantitative calculations were performed for each compound based on the relative peak area and rutin standard.
In the negative ion mode, m/z 100-1500 was used as the mass spectrum data acquisition condition. Atomizing air (GS1): 55 psi; atomizing air (GS2): 55 psi; source temperature (TEM): 550 • C; source voltage (IS): −4500 V. Level 1 scan: de-cluster voltage (DP) and focusing voltage (CE): 100 V and 10 V. Secondary scan: TOF MS~Product Ion~IDA mode was used to collect mass spectrum data. The CID energy was −20, −40, and −60 V. Before sample injection, a CDS pump was used for mass axis correction to make the mass axis error less than 2 ppm.

Statistical Analysis
The data are presented as mean ± standard deviation (SD). All data for this study were adopted for analysis of variance (ANOVA) to determine significant differences. The significance of such differences between mean values was determined using Duncan's test (p < 0.05). ANOVA and Duncan's multiple range tests were performed with SPSS19 (SPSS, Chicago, IL, USA).

Selection of Ethanol Mass Fraction
Single-factor conditions were fixed to allow analysis of other factors. As shown in Table 2, the effect of the ethanol mass fraction on the yield of total flavonoids was studied. Before the ethanol mass fraction reached 38% (w/w), the yield of total flavonoids was positively correlated with it; then, the yield of flavonoids showed a downward trend with increasing ethanol concentration. The reason for this is that as the mass fraction of the ethanol in the system increased, the concentration of the ethanol in the top phase increased, and the polarity decreased. Under these conditions, it was more conducive to perform the extraction of flavonoids. As the mass fraction of ethanol exceeded 38% (w/w), the polarity of the upper phase decreased. This resulted in the immiscibility of the flavonoids and the salt of the precipitation [30]. At this time, the amount of some fat-soluble organic compounds also increased, which inhibited the leaching of total flavonoids. Therefore, the optimal parameter for the ethanol concentration was 38% (w/w). It can be seen in Table 2 that as the ammonium sulfate mass fraction increased, the yield of flavonoids showed a trend of rising first and then falling. However, the yield of flavonoids changed little in the range 17~22% of ammonium sulfate. When the ammonium sulfate mass fraction reached 18% (w/w), the highest yield of total flavonoids was obtained. This occurred in view of the mass fraction of ammonium sulfate directly affecting the phase ratio in the ATPS system. The volume ratio of the upper and lower phases and the total extraction capacity of the solvent were the factors that affected the extraction rate of total flavonoids [31]. Therefore, we selected 18% (w/w) of ammonium sulfate concentration as the optimal parameter.

Effects of Enzyme Concentration on Flavonoid Yield
When determining the enzyme concentration, economic effects should also be considered. The idea was to attain complete extraction and avoid excessive use of enzymes. It can be seen from Table 2 that the total flavonoid content changed under different cellulase concentrations of 0.3, 0.4, 0.5, 0.6, and 0.7 U/g. In the range of cellulase concentration from 0.30 to 0.4 U/g, the total flavonoid yield showed a positive correlation. However, the same change trend between 0.4 and 0.7 U/g was not clear. The reason for this is that cellulase destroyed the cell wall and released bioactive ingredients [32]. However, a superfluous enzyme concentration can saturate the substrate. Excess enzymes should not be combined with substances that cause waste [33]. In brief, a 0.4 U/g concentration of cellulase was chosen for further experimental optimization.

Effects of Enzymolysis Time on Flavonoid Yield
It can be seen from Table 2 that an obvious trend of flavonoid yield was observed between 90 and 180 min, and then the yield of flavonoids began to decrease after 210 min. Enzymatic digestion of the cell walls appeared to have occurred in the sample and the maximum amount of flavonoids was released in 180 min. Enzymatic hydrolysis time affected the yield of target components, and the enzymatic hydrolysis time was short, which did not allow the target components to fully dissolve. The long enzymatic hydrolysis time not only increased the extraction cost, but also led to an increase in impurities. It may be that as the time increases, some flavonoids are oxidized to form quinone compounds and reduce the yield of the target compounds of the top phase. This indicates that the enzymatic hydrolysis of 180 min is the optimal time to catalyze cell wall hydrolysis. Table 2 shows the changing trend of total flavonoids. As the temperature reached between 35 • C and 50 • C, the extraction yield of flavonoids showed a positive correlation. Generally speaking, enzyme activity is closely related to temperature. Enzyme activity and reaction rate can be increased by raising the temperature of enzyme hydrolysis. However, at temperatures above 50 • C, the extraction yield of flavonoids decreased significantly. It may be that excessive temperatures denature the enzyme. It can be seen that the optimal enzymolysis temperature parameter was set to 50 • C.

Effects of Ultrasonic Time on Flavonoid Yield
Ultrasound is a crucial parameter influencing total flavonoid yield. As shown in Table 2, the parameter range of the extraction time was set from 10 to 50 min. Under the condition of a fixed power of 200 w, the yield of total flavonoids was studied. Before the ultrasound time reached 30 min, the extraction yield of total flavonoids was positively correlated with the ultrasound time. At 30 min, it reached the highest yield of 60.07 mg/g, then the yields of flavonoids decreased with the further increase in ultrasonic time [34]. This was mainly due to the large amount of time taken by the ultrasonic reaction, leading to some flavonoid oxidation or degradation, so the yield of flavonoids decreased. Thus, 30 min of extraction time was selected for the subsequent experiments.

Effects of the Liquid-Solid Ratio on Flavonoid Yield
The effect of different liquid-solid ratios on the yield of flavonoids is shown in Table 2. Economically speaking, an appropriate liquid-solid ratio is very important for flavonoid Appl. Sci. 2021, 11, 7718 7 of 17 extraction. The extraction yield of total flavonoids showed an upward trend as the liquidsolid ratio increased from 10:1 to 20:1 mL/g, and then began to decline after it exceeded 20:1 mL/g. This is because the larger the liquid-solid ratio, the more adverse the penetration of the solvent and solute diffusion, resulting in less sufficient flavone dissolution, and the extraction yield of total flavonoids will thus become lower. However, the addition of dried fig leaf powder can absorb water from the aqueous phase and increase the ethanol concentration. The total amount of leaching decreased with the increase in the amount of powder added due to the decrease in permeability and diffusion capacity. Considering this factor economically, in order to avoid significant wastage of the solvent, the liquid-solid ratio in the optimal test design was 20:1 mL/g. Table 3, the test resulted in 29 uncontrolled runs. The correlation between the response and the independent variable can be visualized with a 3D surface plot. Figure 2 indicates the influence of the ethanol concentration %, (X 1 ), ammonium sulfate concentration %, (X 2 ), liquid-solid ratio mL/g, (X 3 ), and ultrasound time min, (X 4 ), on the yield of total flavonoids and their interaction. Furthermore, it is possible to predict the optimal value of the response and the corresponding experimental conditions through the F value (>7.84) and p value (<0.01), as shown in Table 4. This result shows that the model can describe the total flavonoid yield of UEAATPE well. The equation of the response variables and independent variables is as follows:

As shown in
where Y represents yields of total flavonoids (mg/g); X 1 , X 2 , X 3 , and X 4 , respectively, represent the ethanol concentration (%), ammonium sulfate concentration (%), liquid-solid ratio (mL/g), and ultrasonic time (min). Table 3. Experimental data and total flavonoid extraction analyzed by the Box-Behnken approach.  According to appropriate extraction conditions (independent variables) and actual operation analysis by Design Expert software, all these conditions were modified as follows: 38% (w/w) ethanol/18% (w/w) ammonium sulfate; liquid-solid ratio of 20:1 mL/g, and ultrasonic time of 30 min. Under these conditions, it was possible to obtain 60.22 mg/g flavonoids by UEAATPE. According to the RSM prediction model, the above experimental value matches the fitted value (RSD < 1.72%).

Comparison of Different Methods
The results shown in Figure 3 indicate the yields of flavonoids that were obtained by UEAATPE, EAATPE, UAATPE, ATPE, and SE. Among the above five methods, the highest yield of total flavonoids was extracted by UEAATPE. At the same time, the remaining four methods in order of extraction yield were UAE > EAE > ATPE > SE (53.48, 49.59, 47.46, and 24.79 mg/g, respectively). The extraction yield of total flavonoids with SE was lower than that with the other four methods. Therefore, UEAATPE is a prospective method for the extraction of flavonoids.

Comparison of Different Methods
The results shown in Figure 3 indicate the yields of flavonoids that were obtained by UEAATPE, EAATPE, UAATPE, ATPE, and SE. Among the above five methods, the highest yield of total flavonoids was extracted by UEAATPE. At the same time, the remaining four methods in order of extraction yield were UAE > EAE > ATPE > SE (53.48, 49.59, 47.46, and 24.79 mg/g, respectively). The extraction yield of total flavonoids with SE was lower than that with the other four methods. Therefore, UEAATPE is a prospective method for the extraction of flavonoids.

Identification of Flavonoids in Fig Leaves
The UPLC chromatography of flavonoid extraction by UEAATPE was detected at 254 nm, as shown in Figure 4. The identification and structure elucidation of the compounds from the leaves was completed by UPLC-QTOF-MS/MS in negative ion mode. Approximately 11 peaks were separated in the extract. Subsequently, the ESI-MS 1 and MS 2 were used to identify and characterize the flavonoids from the fig leaves. The mass spectra and fragmentation pathways can be seen in Figures S1-S11 (see the Supplementary Material). The identification of peaks was performed using reference data such as retention time and mass spectrum, as shown in Table 5. According to the structural characteristics, these compounds are all flavonoids. The chemical structure compounds identified are shown in Figure 5.

Identification of Flavonoids in Fig Leaves
The UPLC chromatography of flavonoid extraction by UEAATPE was detected at 254 nm, as shown in Figure 4. The identification and structure elucidation of the compounds from the leaves was completed by UPLC-QTOF-MS/MS in negative ion mode. Approximately 11 peaks were separated in the extract. Subsequently, the ESI-MS 1 and MS 2 were used to identify and characterize the flavonoids from the fig leaves. The mass spectra and fragmentation pathways can be seen in Figures S1-S11 (see the Supplementary Material). The identification of peaks was performed using reference data such as retention time and mass spectrum, as shown in Table 5. According to the structural characteristics, these compounds are all flavonoids. The chemical structure compounds identified are shown in Figure 5. interaction of liquid-solid ratio, mL/g and ammonium sulfate concentration, %; (c) interaction of ultrasonic time, min and ethanol concentration, %; (d) interaction of liquid-solid ratio, mL/g and ammonium sulfate concentration, %; (e) interaction of ultrasonic time, min and ammonium sulfate concentration, %; (f) interaction of ultrasonic time, min and liquid-solid ratio, mL/g.

Comparison of Different Methods
The results shown in Figure 3 indicate the yields of flavonoids that were obtained by UEAATPE, EAATPE, UAATPE, ATPE, and SE. Among the above five methods, the highest yield of total flavonoids was extracted by UEAATPE. At the same time, the remaining four methods in order of extraction yield were UAE > EAE > ATPE > SE (53.48, 49.59, 47.46, and 24.79 mg/g, respectively). The extraction yield of total flavonoids with SE was lower than that with the other four methods. Therefore, UEAATPE is a prospective method for the extraction of flavonoids.

Identification of Flavonoids in Fig Leaves
The UPLC chromatography of flavonoid extraction by UEAATPE was detected at 254 nm, as shown in Figure 4. The identification and structure elucidation of the compounds from the leaves was completed by UPLC-QTOF-MS/MS in negative ion mode. Approximately 11 peaks were separated in the extract. Subsequently, the ESI-MS 1 and MS 2 were used to identify and characterize the flavonoids from the fig leaves. The mass spectra and fragmentation pathways can be seen in Figures S1-S11 (see the Supplementary Material). The identification of peaks was performed using reference data such as retention time and mass spectrum, as shown in Table 5. According to the structural characteristics, these compounds are all flavonoids. The chemical structure compounds identified are shown in Figure 5.      The 11 compounds were divided into flavonoid oxygen glycoside compounds and flavonoid carbon glycoside compounds [35]. As shown in Figure 6, flavonoid glycosides mainly underwent Y − type cleavage; that is, the glycosyl is removed, and the hydroxy is retained, which is represented by Y − . The right subscript of the ion indicates the type of glycosyl. For example, Y H − means the cleavage of the hexose to remove the glycosyl. Flavonoid glycosides mainly underwent the cleavage of the sugar ring (X cleavage). When cleavage occurs, the position of the broken bond on the sugar ring is indicated by the left superscript, and the type of the cleavage glycosyl is indicated by the right subscript. This included hexose (H), pentose (p), and deoxyhexacarbonose (D). For example, 0, 2 X H represents a ring-opening cleavage caused by the breakage of the 0, 2 bond on the hexose [36,37]. The 11 compounds were divided into flavonoid oxygen glycoside compounds and flavonoid carbon glycoside compounds [35]. As shown in Figure 6, flavonoid glycosides mainly underwent Ytype cleavage; that is, the glycosyl is removed, and the hydroxy is retained, which is represented by Y -. The right subscript of the ion indicates the type of glycosyl. For example, YH-means the cleavage of the hexose to remove the glycosyl. Flavonoid glycosides mainly underwent the cleavage of the sugar ring (X cleavage). When cleavage occurs, the position of the broken bond on the sugar ring is indicated by the left superscript, and the type of the cleavage glycosyl is indicated by the right subscript. This included hexose (H), pentose (p), and deoxyhexacarbonose (D). For example, 0, 2 XH represents a ring-opening cleavage caused by the breakage of the 0, 2 bond on the hexose [36,37]. Compounds 1, 2, 7, 10, and 11 were all flavonoid oxygen glycoside compounds [38]. Quasi-molecular ion peaks can appear in negative ion modes of flavonoid oxygen glycoside compounds to cause glycosidic bond cleavage. This is characterized by a neutral loss of the glucosyl C6H10O5 (162), rhamnosyl C6H10O4 (146), or xylose C5H8O4 (132). The aglycon structure can continue to lose -CH3 and -CO2 or RDA cleavage can occur in the C ring of the flavonoid aglycon.
The secondary mass spectrometry, we speculated that there was glucose in the structure of the compound, and the molecular weight of the parent nucleus was 301, which was quercetin. Based on Scifinder and Reaxy database retrieval, Compound 10 was presumed to be quercetin 3-O-hexoside [40]. Compound 1 and Compound 10 had the same mass spectrum fragmentation pathway. The relative molecular mass of Compound secondary mass spectrometry, we speculated that there was glucose in the structure of the compound, and the molecular weight of the parent nucleus was 301, which was quercetin. Based on Scifinder and Reaxy database retrieval, Compound 10 was presumed to be quercetin 3-O-hexoside [40]. Compound 1 and Compound 10 had the same mass spectrum fragmentation pathway. The relative molecular mass of Compound 10 was 464. The fragment ion peak (m/z 301) was caused by the loss of a quasi-molecular ion peak and a neutral fragment of m/z 162. According to the relative molecular mass, the neutral fragment of m/z 162 was preliminarily judged to be a glucose group. Since . According to the relative molecular mass, the neutral fragment of m/z 308 was initially judged to be rutin. According to the literature, the substance was kaempferol 3-O-hexobioside [42].
Compounds 3, 4, 5, 6, 8, and 9 are all flavonoid glycosides [38]. Quasi-molecular ion peaks appeared in negative ion modes for the flavonoid carboglycosides. The negative ion mode showed higher abundance for fragment ions. We observed the ring-opening and cleavage of the sugar ring and the subsequent neutral loss of sugar residues, CO, aglycon loss, CH 3, and other fragment ion peaks. The ring-opening cleavage of the sugar ring is the characteristic cleavage form of carbon glycosides. This mainly occurs in the sugar ring 0,  [44]. The cleavage rules for isochafotaside and kaempferol 6-C-hexoside-8-C-hexoside were the same, but the peak order was different. Kaempferol 6-C-hexoside-8-C-hexoside was in the front; isochafotaside was in the back.

Conclusions
This study indicated that the proposed method could be successfully used for the pretreatment and identification of flavonoids in discarded fig leaves. Overall, an innovative extraction method was developed to extract flavonoids from discarded fig leaves by UEAATPE. The ATPE system of 38% (w/w) ethanol/18% (w/w) ammonium sulfate was established as the final system. Simultaneously, an enzymatic concentration of 0.4 U/g, 150 min enzymolysis, an enzymolysis temperature of 50 • C, a liquid-solid ratio of 20:1 (mL/g), and an extraction time of 30 min were obtained as the optimal UEAATPE conditions. Under these conditions, we obtained 60.22 mg/g flavonoids by UEAATPE. The yield of total flavonoids obtained by UEAATPE was 1.13-fold, 1.21-fold, 1.27-fold, and 2.43-fold higher than the yields obtained by the other four methods (EAATPE, UAATPE, ATPE, and SE), respectively. UEAATPE has been shown to be a promising method in the field of bioactive ingredient extraction. Among the eleven compounds characterized, In the future, this method may be increasingly applied in the development and utilization of agricultural waste, acting as a bond between promoting economic balance and environmental protection.
Supplementary Materials: The following are available online at https://www.mdpi.com/article/ 10.3390/app11167718/s1. Figure S1: The first order mass spectrometry (a), the secondary mass spectrometry (b) and the cleavage pathway (c) of compound 1; Figure S2: The first order mass spectrometry (a), the secondary mass spectrometry (b) and the cleavage pathway (c) of compound 2; Figure S3: The first order mass spectrometry (a), the secondary mass spectrometry (b) and the cleavage pathway (c) of compound 3; Figure S4: The first order mass spectrometry (a), the secondary mass spectrometry (b) and the cleavage pathway (c) of compound 4; Figure S5: The first order mass spectrometry (a), the secondary mass spectrometry (b) and the cleavage pathway (c) of compound 5; Figure S6: The first order mass spectrometry (a), the secondary mass spectrometry (b) and the cleavage pathway (c) of compound 6; Figure S7: The first order mass spectrometry (a), the secondary mass spectrometry (b) and the cleavage pathway (c) of compound 7; Figure S8: The first order mass spectrometry (a), the secondary mass spectrometry (b) and the cleavage pathway (c) of compound 8; Figure S9: The first order mass spectrometry (a), the secondary mass spectrometry (b) and the cleavage pathway (c) of compound 9; Figure S10: The first order mass spectrometry (a), the secondary mass spectrometry (b) and the cleavage pathway (c) of compound 10; Figure S11: The first order mass spectrometry (a), the secondary mass spectrometry (b) and the cleavage pathway (c) of compound 11.