Discovery of 7,9-Disulfatetrahydroberberine as Novel Lipid-Lowering Agents.

Berberine (BBR), a well-known alkaloid, exhibits various pharmacological activities, especially hypolipidemic activity, which has attracted much interest from medicinal chemists in the past decade. However, little progress was made on the structural modification of BBR for improving lipid-lowering activity, mainly due to its unclear biological target and low safety. In this study, a new scaffold of 7,9-disulfatetrahydroberberine was discovered unexpectedly, provided with extremely low cytotoxicity. Hence, a novel series of highly safe 7,9-disulfatetrahydroberberines were designed, synthesized, and evaluated for their hypolipidemic activities. In order to investigate the significance of the 9-position substituent, another new series of 7-sulfatetrahydroberberines were designed and synthesized. Lipid-lowering experiments showed that among these compounds, 5f exhibited the best lipid-lowering activity based on two cell models, 3T3-L1 cells and HepG2 cells. Compared with the blank control, the inhibition rate of compound 5f against total cholesterol was over 60%, the inhibition rate against triglyceride was over 70%, the inhibition rate against low-density lipoprotein cholesterol was approximately 75%, and the inhibition rate against high-density lipoprotein cholesterol was close to 50%, which were far superior to the positive control BBR. This result also verified the feasibility of the development of BBR as a lipid-lowering drug via disubstituted modification at the 7- and 9-position.

Introduction

As a representative of isoquinoline alkaloids, berberine (BBR) has long been used in the treatment of intestinal bacterial infection.[1] However, in recent years, BBR has been reported to exhibit a potent biological activity of lowering blood lipid, the molecular mechanism of which has been studied from many aspects and various angles.[2] Generally, the hypolipidemic activity of BBR can be described according to the following four aspects. As shown in Figure , BBR can increase the expression of LDLR protein by regulating the mRNA level and stability of LDLR, reducing the total cholesterol level in cells.[3,4] Second, BBR can affect the metabolism of fatty acids through the regulation of AMPK protein kinase for reducing the level of triglycerides in cells.[5−7] Third, BBR can alter to some extent the composition of gut flora, promoting the intestinal flora to produce short-chain fatty acids in the intestinal tract of rats, which would reduce the level of blood lipids after being absorbed into the blood;[8] on the other hand, the gut flora can also transform BBR into dihydroberberine, increasing its bioavailability to play a part in lipid-lowering activity.[9] Finally, BBR can decrease the expression of fatty acid translocase Cd36 intestinal cells through the farnesoid X receptor (FXR) signal pathway, promoting cholesterol efflux indirectly.[10,11] Despite the fact that a preliminary understanding of the molecular mechanism of BBR on lipid regulation is revealed, its specific macromolecule target has not been discovered yet. This greatly limits the structure modification of BBR for improving its hypolipidemic activity. Accordingly, few research studies on the structural modification of BBR to increase its lipid-lowering activity have been reported in recent years.

Figure 1

Hypolipidemic mechanism of berberine. BBR can lead to the decrease in TCHO and TG by molecular regulation. Meanwhile, BBR can promote CHO excretion and affect gut flora, thus regulating blood lipid indirectly. TCHO: total cholesterol; TG: triglyceride; CHO: cholesterol; ↓: the decrease in amount.

Based on this situation, a robust QSAR model of BBR used for predicting hypolipidemic activity was developed via computer-aid design means,[12] and meanwhile, dozens of BBR derivatives were designed and synthesized in our laboratory. The results from screening assay showed that 9-O-phenylsulfonylberberine (Figure , compound a) displayed potent activity (data not shown). We sought to obtain sulfanilamide analogs (b) for increasing its pharmacological activity due to the metabolism instability of sulfonate group in vivo. Thus, we expected to produce compound b on the basis of intermediate c; however, as shown in Figure , this process of sulfonation cannot work directly, guessing that the nucleophilic reactivity of the −NH2 group at the 9-position of BBR was very low. The other thought is that the intermediate c can be hydrogenated to compound d and subsequently sulfonated and oxidized to obtain the target compound b. Surprisingly, a series of 7,9-disubstituted sulfonamide derivatives were preferentially prepared. On the one hand, few research studies on disubstituted derivatives of BBR have been reported recently. On the other hand, a number of studies demonstrated that monosubstituted derivatives of BBR at the N7- or 9-position can significantly increase its hypolipidemic activity.[13,14] Therefore, this paper would examine these 7,9-disulfatetrahydroberberines as promising lipid-lowering agents and discuss their hypolipidemic activities carefully.

Figure 2

Discovery of 7,9-disulfatetrahydroberberine. In order to synthesize a class of compounds (b) based on the scaffold (a), an intermediate c was obtained. However, compounds c cannot be sulfonated directly due to the poor reactivity of amine. Another intermediate d was prepared and expected to convert into compounds (b) by the sulfonation and oxidation reaction. Interestingly, a series of 7,9-disulfatetrahydroberberines, rather than compounds (b), were obtained.

Another important problem in the development of BBR as a lipid-lowering agent is its safety. Some study revealed that when the blood concentration of BBR exceeded 0.432 μg/mL, more than 30% of the mice died under the acute toxicity test.[15] Other study showed that BBR can inhibit the CYP3A4 and hERG channel, leading to certain cardiotoxicity.[16] That is to say, BBR can exhibit a certain degree of toxicity; as a result of its poor bioavailability by oral administration, the problem of toxicity seems not prominent. Our in-house experiment showed that a significant number of the BBR derivatives showed significant cytotoxicities at a low concentration, not only against 3T3L normal cells but also against HepG2 tumor cells. Therefore, with the structural optimization of BBR, we should pay attention to the balance between the lipid-lowering activity and toxicity of 7,9-disulfatetrahydroberberines.

Result and Discussion

In this study, 15 kinds of 7,9-disulfatetrahydroberberines were designed, synthesized, and evaluated for their lipid-lowering activities at the cellular level. As shown in Scheme , the nucleophilic attack of benzylamine against BBR (1) at high temperature produced compound 2, which subsequently can be hydrolyzed to compound 3 under acid conditions. Intermediate 4 can be obtained through reduction of compound 3 by sodium borohydride. Finally, a series of the target compounds 5a–5o were prepared via sulfonation (1H NMR spectra shown in the Supporting Information). In addition, in order to discuss the effect of the 7-position substituent on the lipid-lowering activity of compounds 5a–5o, 12 kinds of monosubstituted BBR derivatives at the N7-position were designed and synthesized accordingly. As seen in Scheme , BBR (1) was reduced to tetrahydroberberine 7 by sodium borohydride, and then a dozen of sulfonamide BBR derivatives 8a–8l modified at the 7-position were generated by nucleophilic substitution directly (1H NMR spectra shown in the Supporting Information). In the next section, the above compounds were evaluated for their cytotoxicities and lipid-lowering activities based on two common cell models (3T3-L1 cells and HepG2 cells).

Scheme 1

Reagents and Conditions: (i) 2,4-Dimethylbenzenemethanamine, 120 °C, 8 h; (ii) Concentrated HCl, MeOH, rt, 6 h; (iii) NaBH4, MeOH, Reflux, 1 h; (iv) SOCl2, 80 °C, 2 h; Et3N, CH2Cl2, 12 h

Scheme 2

Reagents and Conditions: NaBH4, K2CO3, MeOH, Reflux, 1 h; K2CO3, CH3CN, 12 h

3T3-L1 cells and HepG2 cells are two common cell models used for screening lipid-lowering agents. 3T3-L1 cells are a strain of Mus musculus fibroblast, which would undergo a pre-adipose to adipose-like conversion and produce triglyceride (TG) when cultured. HepG2 cells are a strain of hepatocarcinoma cells, which not only share the same function of lipid metabolism as normal hepatocyte but also has the same characteristics of high proliferation as tumor cells. Interestingly, BBR not only exhibited significant hypolipidemic activity but also displayed potent antitumor activity, for instance, the inhibition of DNA topoisomerase II.[17] Correspondingly, it is essential to distinguish the antitumor activity of BBR from the cytotoxicity of BBR against HepG2 cells. Our previous in-house screening experiment was performed under a concentration of 25 μM, and the results showed that BBR has a potent cytotoxicity. Hence, we attempted to decrease the concentration by half and examined the cytotoxicity of these compounds at a concentration of 12.5 μM.

As shown in Figure , BBR displayed apparent antiproliferative activity against the two cells, where the toxicity against 3T3-L1 cells is higher than that against HepG2 cells. Without doubt, the toxicity of BBR against HepG2 cells might be attributed to its antitumor activity. For 15 kinds of 7,9-disulfatetrahydroberberines (Figure A), all the other compounds except compound 5n showed no cytotoxicity against these two cells. For 12 kinds of 7-sulfatetrahydroberberines (Figure B), all compounds including compound 7 showed no toxicity to 3T3-L1 cells, indicating that the safety of these compounds is superior to that of BBR. Moreover, compound 8j had a modest inhibitory activity against HepG2 cells, implying that this compound has to some extent antitumor activity. On the whole, the safety of the two series of compounds is significantly better than that of BBR on these two cell models. In addition, given the cytotoxic effect of BBR on the two cells, the drug concentration in the following screening experiment was adjusted to 10 μM.

Figure 3

Assay for cytotoxicity to HepG2 cells and 3T3-L1 cells in vitro by compounds BBR, 4, 5a–5o, and 8a–8l (12.5 μM). (A) Compounds BBR, 4, and 5a–5o. (B) Compounds BBR, 7, and 8a–8l. Data is represented by the mean ± SD of the three independent experiments.

The level of total cholesterol (TCHO) in cells can be used as one of the important indexes of measuring lipid-lowering agents. The TCHO inhibition assay of two series of BBR derivatives (27 compounds in total), together with two intermediates 4 and 7, was carried out strictly. The results (Figure ) showed that most of the compounds had significant inhibitory activities against the TCHO level of cells, whether in the 3T3-L1 cell model or in the HepG2 cell model; moreover, the lipid-lowering effect in 3T3-L1 cells was generally better than that in HepG2 cells; compared with the parent compound BBR, the inhibitory activity of intermediates 4 and 7 was more superior to that of BBR, both of which had an inhibition rate of about 60% in 3T3-L1 cells. In Figure A, except that the inhibition rate of compound 5j in HepG2 cells can reach 70%, the inhibition rate of the other 14 compounds in HepG2 cells was obviously inferior to that of intermediate 4. In 3T3-L1 cells, only two compounds 5a and 5f were better than intermediate 4, while most of the compounds were slightly inferior to intermediate 4. This implied that the structural modification at the N7- or N9-position can indeed increase the TCHO inhibitory activity of BBR. As shown in Figure B, except that compounds 8i, 8j, and 8k displayed significant THCO inhibitory activities, the other compounds had no TCHO inhibitory effect in HepG2 cells. On the contrary, most of the compounds exhibited to some extent cholesterol inhibitory activity in 3T3-L1 cells, which was not only more superior to BBR but also superior to tetrahydroberberine 7, such as compounds 8c, 8d, and 8l. It was inferred that substituting at the N7-position of BBR would lead to the increase in TCHO inhibitory activity. In addition, a series of compounds 5a–5d, 5f–5i, 5k, 5l, and 5n in Figure A were compared with the corresponding compounds with the same substituent (8b–8l) in Figure b, respectively. It would be discovered that compounds 5a, 5d, and 5f–5h were superior to the corresponding compounds 8b, 8e, and 8f–8h, regardless of in 3T3-L1 cells or HepG2 cells. Despite the fact that two compounds 5i and 5k did not show TCHO inhibitory activity in these two cells, overall, disubstituted derivatives were slightly better than monosubstituted derivatives.

Figure 4

Inhibitory rate of compounds 5a–5o and 8a–8l against total cholesterol (TCHO) in 3T3-L1 cells and HepG2 cells when compared to the three controls: BBR, compounds 4, and 7 (10 μM). (A) Compounds BBR, 4, and 5a–5o. (B) Compounds BBR, 7, and 8a–8l. Data is represented by the mean ± SD of the three independent experiments.

In addition, the experimental value of log P of BBR is less than −1.6,[18] resulting in its exceedingly low bioavailability. Structural modification of the 7- and 9-position of BBR can significantly increase its log P; however, the improvement of log P cannot ensure that the lipid-lowering activity of derivatives certainly increased (Figure ). Therefore, the structural modification of BBR based on multi-objective optimization should be considered, rather than only log P. From insights into Figure A,B, the structure–activity relationships showed that the halogen substituent of phenylsulfonyl was beneficial to the rapid increase in TCHO inhibitory activity of the 7,9-disulfatetrahydroberberine scaffold or 7-N-sulfatetrahydroberberine scaffold when comparing 5a–5d, 5f–5h, and 8b–8e with other compounds.

In addition to TCHO, triglycerides (TG) are also an important index used for screening lipid-lowering agents. There are two main sources of TG in plasma: one is derived from exogenous substance, the other is intracellularly synthesized by liver and adipose tissue. Therefore, 3T3-L1 cells and HepG2 cells are two important cell models employed for the detection of endogenous synthesis of TG. Figure displays the reduction of TG amount in cells by these compounds. As shown in Figure A, around half of the compounds had no inhibitory activity against TG, and several compounds of the rest (5j, 5m, 5n, and 5o) showed more modest inhibitory activity when compared with BBR. Compound 4 had an inhibitory rate of 70% against TG in HepG2 cells but had weak inhibitory activity in 3T3-L1 cells. Nevertheless, the introduction of 4-chlorobenzenesulfonyl or 4-bromobenzenesulfonyl at the N7- and 9-position of BBR can significantly increase the TG inhibitory rate of compounds 5f and 5g in 3T3-L1 cells.

Figure 5

Inhibitory rate of compounds 5a–5o and 8a–8l against triglyceride (TG) in 3T3-L1 cells and HepG2 cells when compared to the three controls: BBR, compounds 4, and 7 (10 μM). (A) Compounds BBR, 4, and 5a–5o. (B) Compounds BBR, 7, and 8a–8l. Data is represented by the mean ± SD of the three independent experiments.

In Figure B, compound 7, a known 4H-hydroberberine, displayed potent inhibitory activity against TG in 3T3-L1 cells while exhibiting no activity in HepG2 cells. Similarly, compound 8c showed an extremely high inhibitory rate in 3T3-L1 cells and very low inhibitory rate in HepG2 cells. In general, the inhibitory activity of 7-monosubstituted derivatives against TG was very limited. In contrast, 7,9-disubstituted derivatives possessed better potential of structure optimization for improving their hypolipidemic activity. Additionally, compound 5h and 8h had the same substituent at the N7-position. Interestingly, compound 5h showed relatively high activity in 3T3-L1 cells and low activity in HepG2 cells, while compound 8h was the opposite. Only one of their structural differences was the different substituent at the 9-position. It is known that the molecular mechanism of BBR decreasing TG is mainly involved with the regulation of the oxidation, hydrolysis, and synthesis of fatty acids. However, it was not clear why structural modification of BBR can lead to different lipid-lowering effects in different cells or tissues (3T3-L1 cells were assumed to represent adipose tissue and HepG2 cells represented liver tissue). Therefore, the structural modification around the 7- and 9-position of BBR would contribute to the discussion for different hypolipidemic activities in different kinds of cells.

The expression levels of LDL-C and HDL-C are another two common indexes for screening lipid-lowering agents. BBR can reduce the amount of LDL cholesterol by increasing the expression of LDLR;[3] meanwhile, it can increase the expression level of HDL-C through the LXR/RXR and PKC signal pathway.[19,20] Although some drugs such as fibrates and niacin could increase the expression level of HDL in cells, BBR as a new lipid-lowering agent had significant inhibitory activity against LDL-C and up-regulated expression of HDL-C in cells (Figure ). Compared with the blank control, the inhibition rate of BBR against LDL-C in the two cell models was over 30%, while the increase rate of BBR against HDL-C in the two cell models was also more than 25%. As shown in Figure A,C, most of the compounds showed significant inhibitory activity against LDL-C when compared with the parent compound BBR. More concretely, most of the 7,9-disubstituted compounds showed potent inhibitory activity in both cells, while the 7-monosubstituted compounds showed more significant inhibitory activity in HepG2 cells than in 3 T3-L1 cells. Assuming that the 7,9-disubstituted compounds and the 7-monosubstituted compounds had the same mechanism of lowering LDL-C amount, different structural modifications at the N9-position would result in different TG inhibitory activities in different cells (hepatocytes and adipocytes) as well as TCHO, which would be regarded as a unique feature of structural modifications at the N7- and 9-positions of BBR. The structure–activity relationships demonstrated that no matter the ortho and para substituent or the electron-withdrawing group and electron-donating group, these structural modifications such as compounds 5d, 5f–5h, 5j, 5k, 8e, and 8l can increase LDL-C inhibitory activity, where there was no obvious regularity. Moreover, it should be noted that compounds 5l and 8k shared the same substituent at the N7-position. The former exhibited potent lipid-lowering activity in 3T3-L1 cells, the inhibition rate of which was more than 70%, while the latter had no activity completely. The simple modification at the 9-position of compounds 8k resulted in a sharp change in lipid-lowering activity, which was worthy of further study.

Figure 6

Biological activities of compounds 5a–5o and 8a–8l against low-density lipoprotein cholesterol (LDL-C) and high-density lipoprotein cholesterol (HDL-C) in 3T3-L1 cells and HepG2 cells when compared to the three controls: BBR, compounds 4, and 7 (10 μM). (A) Inhibitory activities of compounds BBR, 4, and 5a–5o against LDL-C. (B) Increased activities of compounds BBR, 4, and 5a–5o against HDL-C. (C) Inhibitory activities of compounds BBR, 7, and 8a–8l against LDL-C. (D) Increased activities of compounds BBR, 7, and 8a–8l against HDL-C. Data is represented by the mean ± SD of the three independent experiments.

Generally speaking, there is no direct relationship between the increase in HDL-C and the decrease in LDL-C in cells. However, by comparison of different subgraphs (Figure A and B, Figure C and D), it would be found that compounds 5f–5i, 5k, 8a, 8b, and 8e can decrease the expression of LDL-C and increase the expression of HDL-C. This indicated that these compounds can significantly enhance the lipid-lowering activity of BBR by multiple kinds of lipid-lowering mechanisms. In HDL-C experiments, most compounds showed moderate activity and were inferior to compound 4 in HepG2 cells except for compounds 5a and 5i. The activities of compounds 5f, 5g, and 5h in the 3T3-L1 cell model were comparable to that of compound 4. Among the 7-monosubstituted derivatives, compounds 8b and 8j showed very high activities, the increased rate of which against HDL-C in HepG2 cells both reached nearly 80% when compared with the blank control. Ssimilar to LDL-C, the 7-monosubstituted derivatives exhibited better lipid-lowering activity in HepG2 cells than in 3T3-L1 cells. Comparing compounds 5f–5h with 8f–8h, respectively, each pair of compounds had the same substituent at the N7-position, but the hypolipidemic activity of the former in 3T3-L1 cells was obviously better than that of the latter, implying that the scaffold of 7,9-disulfatetrahydroberberine had a major advantage in the regulation of LDL-C and HDL-C. In addition, compounds 5d, 5e, 5m–5o, 8d, and 8k displayed very low or no activities. The structure–activity relationships demonstrated that the introduction of the para substituents (5f–5h and 8f–8h) and the meta substituents (5i and 8i) of benzenesulfonyl can significantly increase the expression of HDL-C in cells.

Conclusions

In conclusion, two series of new BBR derivatives, 27 compounds in total, were designed and synthesized. The cytotoxicity experiment and lipid-lowering activity experiment of these compounds were carried out based on two common cell models, 3T3-L1 cells and HepG2 cells. The results from the toxicity test showed that both of the structural modifications at the 7- and 9-position can improve the safety of BBR. The hypolipidemic activity of these compounds can be measured by four common indexes: TCHO, TG, LDL-C, and HDL-C. In TCHO experiments, compounds 5a, 5d, 5f, 5j, 8c, and 8l exhibited potent inhibitory activities; in TG experiments, compounds 5f–5h, 8c, and 8h showed high inhibitory activities; in LDL-C experiments, compounds 5f–5h, 5l, 5m, 8c, and 8l had potent inhibitory activities; in HDL-C experiments, compounds 5f–5h, 8b, 8i, and 8j showed high lipid-lowering activities. Therefore, based on the results of the four hypolipidemic experiments, it was apparent that compound 5f was the best candidate in the four lipid-lowering activity experiments; thus, it can be considered as a new lipid-lowering agent and evaluated by a series of in-depth pharmacology experiments. Compound 5f has a 4-chlorobenzenesulfonyl group at the 7- and 9-position, the molecular weight and predicted log P of which are 674.59 and 5.03, respectively. Although the molecular weight and log P of this compound are not in accordance with the “rule of five”, the structural modification at the 7- and 9-positions should be acceptable to some extent, considering that the poor physicochemical property of BBR itself results in its very low bioavailability. In short, this sort of 7,9-disulfatetrahydroberberine scaffold can achieve the best optimization between the toxicity and hypolipidemic activity of BBR, namely, it can reduce effectively the toxicity of BBR when increasing the lipid-lowering activity. Compound 5f as a representative of these 7,9-disulfatetrahydroberberines would be further studied in subsequent experiments.

Methodology

Biological Assays

Reagents and Cell Culture

HepG2 cells and 3T3-L1 cells were from Nanjing University. 3-(4, 5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) was from Sigma. 6-well and 96-well plates and BCA protein assay kit (P0010) were from Beyotime Biotechnology. A total cholesterol assay kit (TCHO, A111-1), triglyceride assay kit (TG, A110-1), low-density lipoprotein cholesterol assay kit (LDL-C, A113-1), and high-density lipoprotein cholesterol assay kit (HDL-C, A112-1) were purchased from the Nanjing Jiancheng Bioengineering Institute. 3-Isobutyl-1-methylxanthine (IBMX), insulin (INS), and dexamethasone (DEX) were purchased from Sigma Aldrich. HepG2 cells and 3T3-L1 cells were grown in high-glucose Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% fetal bovine serum (FBS), 100 U/mL penicillin, and 100 μg/mL streptomycin and propagated at 37 °C in a humidified atmosphere containing 5% CO2 in air.

Compounds 4, 5a–5o, 7, 8a–8l, and BBR were dissolved in dimethyl sulfoxide to make stock solutions kept at −20 °C. The final concentration of the vehicle in the solution never exceeded 0.1% and had no effects on cell viability.

Assay for Cytotoxic Activity

Nontoxic concentrations of the above compounds were determined according to the MTT test[21] and a concentration of 12.5 μM was chosen to test the effects of BBR analogs. MTT was dissolved at 4 mg/mL in PBS and used essentially as previously described. Briefly, cell lines in the logarithmic phase were seeded at a density of 3 × 103 cells/well in 100 μL of DMEM into 96-well microtiter plates. After 6 h, exponentially growing cells were exposed to the indicated compounds at a concentration of 12.5 μM. After 72 h in final volumes of 200 μL, cell survival was determined by the addition of an MTT solution (20 μL of 4 mg/mL MTT in PBS) for 4 h. After carefully removing the medium, the precipitates were dissolved in 200 μL of DMSO, shaken mechanically for 10 min, and then their absorbance values at a wavelength of 540 nm were taken on a SpectraMax 190 microplate reader (Molecular Devices, America). The survival rate was expressed in percentages with respect to untreated cells.

Protein Extraction and 3T3-L1 Cell Differentiation

The experiment of protein extraction was performed as follows.[21] The HepG2 cells were seeded at a density of 1 × 105 cells/well in 6-well plates with 1 mL of DMEM per well. After 24 h, exponentially growing cells were exposed to the indicated compounds at a concentration of 10 μM. After another 48 h, the HepG2 cells on 6-well plates were rinsed twice with cold phosphate-buffered saline (PBS) and centrifuged (4 °C, 6000 rpm, 30 s) to obtain the protein samples and then were lysed in 100 μL of RIPA lysis buffer (10 mM HEPES, 2 mM EDTA, 0.1% CHAPS, 5 mM DTT, and 1 mM PMSF) on ice for 30 min. The protein concentrations were measured using the BCATM protein quantification kit.

The protocol of 3T3-L1 cell differentiation was as follows.[22] Briefly, 3 T3-L1 cells were seeded at a density of 1 × 105 cells/well in 6-well plates with 1 mL of DMEM per well. The cell shape became round after two days. The culture medium was replaced with differentiation medium (DMEM, 10% FBS, 0.5 mmol/L IBMX, 1.0 μmol/L DEX, and 10 mg/L insulin), and the cells would be cultured for 3 days. Subsequently, the cells were maintained in differentiation medium containing only 10 mg/L insulin for 2 days. The cells were replenished with DMEM every other day. On day 10, over 80% of the 3 T3-L1 cells had differentiated into mature adipocytes. The procedure for protein extraction and quantification of 3T3-L1 was similar to that of HepG2 cells.

Quantification of TCHO, TG, LDL-C, and HDL-C

The HepG2 or 3T3-L1 cell lysates were used to test the levels of TCHO and TG using an assay kit directly. A total of 2.5 μL of cell lysates and 250 μL of working fluid were mixed together at 37 °C for 10 min. Absorbance was quantified at 510 nm with a spectrophotometer. The TCHO and TG values (mmol/gprot) was calculated using the equation [(ODexperimental group – ODblank)/(ODstandard group – ODblank)] × Cstandard group/Cprotein. ODexperimental group, ODstandard group, and ODblank are the mean absorbances of the experimental group, standard group, and only ultrapure water-added group, respectively. Cstandard group and Cprotein are the concentrations of the standard group and protein, respectively. The results were determined through at least three independent experiments.

The HepG2 or 3T3-L1 cell lysates were used to test the levels of LDL-C and HDL-C using an assay kit directly. A total of 2.5 μL of cell lysates and 180 μL of working fluid R1 were mixed together at 37 °C for 5 min. The absorbance OD1 was quantified at 546 nm with a spectrophotometer. Then, 60 μL of working fluid R2 was added and mixed together at 37 °C for 5 min. The absorbance OD2 was quantified at 546 nm with a spectrophotometer. The LDL-C and HDL-C values (mmol/gprot) was calculated using the equation [(ΔODexperimental group – ΔODblank)/(ΔODstandard group – ΔODblank)] × Cstandard group/Cprotein. ΔODexperimental group, ΔODstandard group, and ΔODblank (ΔOD = OD2 – OD1) are the mean absorbances of the experimental group, standard group, and only ultrapure water-added group, respectively. Cstandard group and Cprotein are the concentrations of the standard group and protein, respectively. The results were determined through at least three independent experiments.

Chemistry

Reagents and General Methods

1H NMR spectra were recorded on Bruker AM 400 and 600 MHz spectrometers with tetramethylsilane (TMS) as the internal standard. Electrospray ionization mass spectra (ESI-MS) were recorded using an Agilent 1100 series LC/MSD ion trap mass spectrometer. Melting points (m.p.) were recorded on a SRS OptiMelt-100 full automatic micro melting point instrument. Column chromatography (CC): silica gel (200–300 mesh; Qingdao Makall Group Co., Ltd; Qingdao; China). All reactions were monitored using thin-layer chromatography (TLC) on silica gel plates. Reaction reagents were of analytical reagent grade and purchased from Aladdin.

Synthesis Procedure for 7,9-Disulfatetrahydroberberines

Synthesis of Intermediate 2

The mixture of berberine (5 g, 13.5 mmol) and 2,4-dimethylbenzylamine (2.5 g, 18.5 mmol) was stirred vigorously for 8 h at 120 °C, and the color of the reaction mixture gradually changed from yellow to dark red. The reaction was monitored by TLC; after cooling to room temperature, the excess amine was removed by vacuum filtration, and the residue was washed by acetone (3 × 50 mL). The red product 2 was finally purified by silica gel column chromatography with gradient elution (CH3OH/CH2Cl2 = 1:30 and then 1:10). The yield was about 25%.

9-((2,4-Dimethoxybenzyl)amino)-10-methoxy-5,6-dihydro-[1,3]dioxolo[4,5-g]isoquinolino[3,2-a]isoquinolin-7-ium (2)

ESI-MS m/z: 471.16 (M – Cl)+. 1H NMR (400 MHz, methanol-d4) δ 9.61 (s, 1H), 8.55 (s, 1H), 7.89 (d, J = 8.8 Hz, 1H), 7.69 (d, J = 8.8 Hz, 1H), 7.65 (s, 1H), 7.12 (d, J = 8.3 Hz, 1H), 6.98 (s, 1H), 6.55 (d, J = 2.4 Hz, 1H), 6.44 (dd, J = 8.3, 2.4 Hz, 1H), 6.13 (s, 2H), 4.81 (t, J = 6.3 Hz, 2H), 4.66 (s, 2H), 4.00 (s, 3H), 3.80 (s, 3H), 3.78 (s, 3H), 3.27 (t, J = 6.3 Hz, 2H). 13C NMR (101 MHz, methanol-d4) δ 162.36, 159.90, 151.86, 150.75, 149.86, 147.52, 138.69, 137.83, 134.80, 131.42, 131.31, 124.38, 122.02, 121.45, 120.73, 120.15, 119.95, 109.34, 106.30, 105.41, 103.58, 99.42, 57.50, 57.05, 55.94, 55.79, 28.44.

Synthesis of Intermediate 3

The intermediate 2 (2 g, 4.3 mmol) was dissolved in methanol (10 mL). After the addition of concentrated hydrochloric acid (2 mL), the mixture solution was stirred for 6 h. The reaction was monitored by TLC. Upon completion, the reaction mixture was concentrated under vacuum. The residue was washed by 80% methanol/water solution at least three times. The intermediate 3 was purified from the residue by silica gel column chromatography using methanol and dichloromethane as an eluent (CH3OH/CH2Cl2 = 1:10). The yield was about 75%.

9-Amino-10-methoxy-5,6-dihydro-[1,3]dioxolo[4,5-g]isoquinolino[3,2-a]isoquinolin-7-ium (3)

ESI-MS m/z: 321.09 (M – Cl)+. 1H NMR (400 MHz, DMSO-d6) δ 10.26 (s, 1H), 8.62 (s, 1H), 7.83 (d, J = 8.6 Hz, 1H), 7.74 (s, 1H), 7.31 (d, J = 8.6 Hz, 1H), 7.06 (s, 1H), 6.88 (s, 2H), 6.15 (s, 2H), 4.71 (t, J = 6.0 Hz, 2H), 3.97 (s, 3H), 3.19 (t, J = 6.0 Hz, 2H).

Synthesis of Intermediate 4

The red solid 3 (1 g, 3 mmol) added into methanol (30 mL) was stirred until dissolved. Sodium borohydride (456 mg, 12 mmol) was added slowly into the solution, and the reaction time was about 1 h. The reaction was monitored by TLC. The reaction solution was evaporated under vacuum, and then the residue was dissolved in dichloromethane. The intermediate 4 was purified from this residue by silica gel column chromatography with ethyl acetate and petroleum ether (1:1) as an eluent. The yield was about 60%.

10-Methoxy-5,8,13,13a-tetrahydro-6H-[1,3]dioxolo[4,5-g]isoquinolino[3,2-a]isoquinolin-9-amine (4)

ESI-MS m/z: 325.12 (M + H)+. 1H NMR (400 MHz, chloroform-d) δ 6.74 (s, 1H), 6.70 (d, J = 8.2 Hz, 1H), 6.59 (d, J = 6.7 Hz, 2H), 5.92 (s, 2H), 3.95 (d, J = 14.6 Hz, 1H), 3.84 (s, 3H), 3.66 (s, 2H), 3.57 (d, J = 8.8 Hz, 1H), 3.44 (d, J = 14.6 Hz, 1H), 3.24–3.07 (m, 3H), 2.90–2.79 (m, 1H), 2.72–2.63 (m, 2H).

Synthesis of 7,9-Disulfatetrahydroberberines 5a–5o

At room temperature, the light yellow solid 4 (100 mg, 0.31 mmol) was dissolved in dichloromethane (10 mL). After the addition of triethylamine (100 μL) into the solution, each kind of benzene sulfonyl chlorides (0.62 mmol) was added slowly and stirred for overnight. The reaction was monitored by TLC, and the mixture was evaporated under vacuum. Silica gel column chromatography with gradient elution (pure dichloromethane used as an eluent and then CH3OH/CH2Cl2 (1:100) used as an eluent) was used to separate and purify the 7,9-disulfatetrahydroberberines 5a–5o. The yields of these compounds were 30–50%.

9-((2-Fluorophenyl)sulfonamido)-7-((2-fluorophenyl)sulfonyl)-10-methoxy-5,6,7,8,13,13a-hexahydro-[1,3]dioxolo[4,5-g]isoquinolino[3,2-a]isoquinolin-7-ium Chloride (5a)

ESI-MS m/z: 641.09 (M – Cl)+. m.p. 217–219 °C. 1H NMR (400 MHz, chloroform-d) δ 8.02 (dd, J = 8.8, 5.0 Hz, 4H), 7.25–7.15 (m, 5H), 6.68 (d, J = 7.6 Hz, 2H), 6.58 (s, 1H), 5.92 (s, 2H), 3.81–3.69 (m, 2H), 3.69–3.60 (m, 2H), 3.30 (s, 3H), 3.10–2.98 (m, 1H), 2.83 (dd, J = 33.0, 17.7 Hz, 2H), 2.63 (d, J = 16.1 Hz, 1H), 2.53–2.38 (m, 1H).

9-((3-Fluorophenyl)sulfonamido)-7-((3-fluorophenyl)sulfonyl)-10-methoxy-5,6,7,8,13,13a-hexahydro-[1,3]dioxolo[4,5-g]isoquinolino[3,2-a]isoquinolin-7-ium Chloride (5b)

ESI-MS m/z: 641.09 (M – Cl)+. m.p. 221–213 °C. 1H NMR (400 MHz, chloroform-d) δ 7.80 (t, J = 8.1 Hz, 2H), 7.69 (d, J = 8.0 Hz, 2H), 7.57–7.48 (m, 2H), 7.38 (dt, J = 8.2, 4.2 Hz, 2H), 7.21 (d, J = 8.5 Hz, 1H), 6.69 (d, J = 4.2 Hz, 2H), 6.58 (s, 1H), 5.92 (s, 2H), 3.65 (t, J = 10.0 Hz, 1H), 3.57–3.42 (m, 2H), 3.30 (s, 3H), 3.26 (d, J = 20.3 Hz, 1H), 3.13–2.94 (m, 1H), 2.92–2.72 (m, 2H), 2.61 (d, J = 17.2 Hz, 1H), 2.53–2.39 (m, 1H).

9-((4-Fluorophenyl)sulfonamido)-7-((4-fluorophenyl)sulfonyl)-10-methoxy-5,6,7,8,13,13a-hexahydro-[1,3]dioxolo[4,5-g]isoquinolino[3,2-a]isoquinolin-7-ium Chloride (5c)

ESI-MS m/z: 641.09 (M – Cl)+. m.p. 215–217 °C. 1H NMR (400 MHz, chloroform-d) δ 8.02 (dd, J = 8.9, 5.0 Hz, 4H), 7.24–7.17 (m, 5H), 6.68 (d, J = 7.8 Hz, 2H), 6.58 (s, 1H), 5.92 (s, 2H), 3.76 (d, J = 11.4 Hz, 1H), 3.71–3.60 (m, 2H), 3.30 (s, 3H), 3.29–3.22 (m, 1H), 3.14–2.96 (m, 1H), 2.83 (dd, J = 34.6, 18.5 Hz, 2H), 2.69–2.40 (m, 2H).

9-((2-Chlorophenyl)sulfonamido)-7-((2-chlorophenyl)sulfonyl)-10-methoxy-5,6,7,8,13,13a-hexahydro-[1,3]dioxolo[4,5-g]isoquinolino[3,2-a]isoquinolin-7-ium Chloride (5d)

ESI-MS m/z: 673.03 (M – Cl)+. m.p. 227–229 °C. 1H NMR (400 MHz, chloroform-d) δ 7.99 (d, J = 7.3 Hz, 4H), 7.56–7.48 (m, 4H), 7.18 (d, J = 8.5 Hz, 1H), 6.70 (s, 1H), 6.64 (d, J = 8.5 Hz, 1H), 6.57 (s, 1H), 5.91 (s, 2H), 3.86 (d, J = 15.9 Hz, 1H), 3.50 (dd, J = 22.9, 13.0 Hz, 2H), 3.24 (dd, J = 16.0, 4.1 Hz, 1H), 3.18 (s, 3H), 3.11–2.97 (m, 1H), 2.92–2.71 (m, 2H), 2.60 (d, J = 15.9 Hz, 1H), 2.42 (t, J = 10.1 Hz, 1H).

9-((3-Chlorophenyl)sulfonamido)-7-((3-chlorophenyl)sulfonyl)-10-methoxy-5,6,7,8,13,13a-hexahydro-[1,3]dioxolo[4,5-g]isoquinolino[3,2-a]isoquinolin-7-ium Chloride (5e)

ESI-MS m/z: 673.03 (M – Cl)+. m.p. 231–233 °C. 1H NMR (400 MHz, chloroform-d) δ 7.93–7.83 (m, 4H), 7.64 (d, J = 7.9 Hz, 2H), 7.53–7.44 (m, 2H), 7.23 (d, J = 8.3 Hz, 1H), 6.70 (s, 2H), 6.59 (s, 1H), 5.93 (s, 2H), 3.92 (d, J = 18.5 Hz, 1H), 3.65 (t, J = 9.6 Hz, 1H), 3.52 (d, J = 16.8 Hz, 2H), 3.30 (s, 3H), 3.09 (d, J = 21.0 Hz, 1H), 2.87 (s, 2H), 2.63 (d, J = 16.6 Hz, 1H), 2.48 (s, 1H).

9-((4-Chlorophenyl)sulfonamido)-7-((4-chlorophenyl)sulfonyl)-10-methoxy-5,6,7,8,13,13a-hexahydro-[1,3]dioxolo[4,5-g]isoquinolino[3,2-a]isoquinolin-7-ium Chloride (5f)

ESI-MS m/z: 673.03 (M – Cl)+. m.p. 223–225 °C. 1H NMR (400 MHz, chloroform-d) δ 7.94 (d, J = 8.7 Hz, 4H), 7.51 (t, J = 8.4 Hz, 4H), 7.20 (d, J = 8.5 Hz, 1H), 6.72–6.65 (m, 2H), 6.59 (s, 1H), 5.92 (s, 2H), 3.73–3.60 (m, 2H), 3.48 (dd, J = 21.7, 14.3 Hz, 2H), 3.31 (s, 3H), 3.11–2.78 (m, 2H), 2.76–2.59 (m, 2H), 2.51–2.37 (m, 1H).

9-((4-Bromophenyl)sulfonamido)-7-((4-bromophenyl)sulfonyl)-10-methoxy-5,6,7,8,13,13a-hexahydro-[1,3]dioxolo[4,5-g]isoquinolino[3,2-a]isoquinolin-7-ium Chloride (5g)

ESI-MS m/z: 760.93 (M – Cl)+. m.p. 224–226 °C. 1H NMR (400 MHz, chloroform-d) δ 7.85 (t, J = 8.0 Hz, 4H), 7.68 (t, J = 8.4 Hz, 4H), 7.26 (s, 1H), 7.20 (d, J = 8.5 Hz, 2H), 6.71–6.66 (m, 1H), 6.59 (s, 2H), 5.92 (s, 1H), 3.68–3.60 (m, 2H), 3.57–3.39 (m, 3H), 3.25 (d, J = 16.7 Hz, 1H), 3.12–2.97 (m, 1H), 2.90–2.78 (m, 1H), 2.75–2.59 (m, 2H), 2.52–2.35 (m, 1H).

10-Methoxy-9-((4-nitrophenyl)sulfonamido)-7-((4-nitrophenyl)sulfonyl)-5,6,7,8,13,13a-hexahydro-[1,3]dioxolo[4,5-g]isoquinolino[3,2-a]isoquinolin-7-ium Chloride (5h)

ESI-MS m/z: 695.08 (M – Cl)+. m.p. 206–208 °C. 1H NMR (400 MHz, chloroform-d) δ 8.29–8.07 (m, 4H), 7.93 (d, J = 7.7 Hz, 2H), 7.72 (t, J = 7.6 Hz, 2H), 7.23 (s, 1H), 6.68 (d, J = 16.5 Hz, 2H), 6.59 (s, 1H), 5.93 (s, 2H), 3.79–3.69 (m, 2H), 3.69–3.60 (m, 3H), 3.61–3.43 (m, 2H), 3.20 (s, 3H), 3.03 (d, J = 11.7 Hz, 1H), 2.93–2.78 (m, 2H), 2.68–2.41 (m, 2H).

10-Methoxy-9-((3-(trifluoromethyl)phenyl)sulfonamido)-7-((3-(trifluoromethyl)phenyl)sulfonyl)-5,6,7,8,13,13a-hexahydro-[1,3]dioxolo[4,5-g]isoquinolino[3,2-a]isoquinolin-7-ium Chloride (5i)

ESI-MS m/z: 741.09 (M – Cl)+. m.p. 199–201 °C. 1H NMR (400 MHz, chloroform-d) δ 8.25 (d, J = 8.0 Hz, 1H), 8.17 (d, J = 5.9 Hz, 1H), 8.11 (s, 2H), 7.93 (d, J = 7.7 Hz, 2H), 7.72 (t, J = 7.7 Hz, 2H), 7.23 (s, 1H), 6.71–6.65 (m, 2H), 6.59 (s, 1H), 5.93 (s, 2H), 3.79–3.71 (m, 1H), 3.68–3.61 (m, 2H), 3.27 (d, J = 14.4 Hz, 1H), 3.20 (s, 3H), 3.03 (d, J = 12.7 Hz, 1H), 2.86 (d, J = 11.4 Hz, 2H), 2.62 (d, J = 15.8 Hz, 1H), 2.47 (s, 1H).

10-Methoxy-9-((4-methylphenyl)sulfonamido)-7-tosyl-5,6,7,8,13,13a-hexahydro-[1,3]dioxolo[4,5-g]isoquinolino[3,2-a]isoquinolin-7-ium Chloride (5j)

ESI-MS m/z: 633.14 (M – Cl)+. m.p. 201–203 °C. 1H NMR (400 MHz, chloroform-d) δ 7.90–7.80 (m, 4H), 7.36–7.24 (m, 4H), 7.16 (d, J = 8.5 Hz, 1H), 6.70 (s, 1H), 6.65 (d, J = 8.5 Hz, 1H), 6.57 (s, 1H), 5.91 (s, 2H), 3.83 (d, J = 15.9 Hz, 1H), 3.59–3.41 (m, 2H), 3.25 (d, J = 4.3 Hz, 1H), 3.23 (s, 3H), 3.09–2.98 (m, 1H), 2.90–2.70 (m, 2H), 2.61 (d, J = 15.8 Hz, 1H), 2.46 (d, J = 4.0 Hz, 6H).

10-Methoxy-9-((4-methoxyphenyl)sulfonamido)-7-((4-methoxyphenyl)sulfonyl)-5,6,7,8,13,13a-hexahydro-[1,3]dioxolo[4,5-g]isoquinolino[3,2-a]isoquinolin-7-ium Chloride (5k)

ESI-MS m/z: 665.13 (M – Cl)+. m.p. 202–204 °C. 1H NMR (400 MHz, chloroform-d) δ 7.95–7.88 (m, 4H), 7.16 (d, J = 8.4 Hz, 1H), 6.97 (dd, J = 8.0, 5.5 Hz, 4H), 6.72–6.64 (m, 2H), 6.57 (s, 1H), 5.92 (s, 2H), 3.90 (s, 6H), 3.81 (d, J = 25.5 Hz, 1H), 3.60–3.42 (m, 2H), 3.29 (s, 3H), 3.23 (dd, J = 16.0, 3.7 Hz, 1H), 3.12–2.95 (m, 1H), 2.83 (dd, J = 27.9, 12.8 Hz, 2H), 2.61 (d, J = 16.0 Hz, 1H), 2.44 (t, J = 9.8 Hz, 1H).

9-((4-(tert-Butyl)phenyl)sulfonamido)-7-((4-(tert-butyl)phenyl)sulfonyl)-10-methoxy-5,6,7,8,13,13a-hexahydro-[1,3]dioxolo[4,5-g]isoquinolino[3,2-a]isoquinolin-7-ium Chloride (5l)

ESI-MS m/z: 717.24 (M – Cl)+. m.p. 198–200 °C. 1H NMR (400 MHz, chloroform-d) δ 8.01 (dd, J = 9.0, 5.2 Hz, 4H), 7.24–7.14 (m, 5H), 6.67 (d, J = 7.7 Hz, 2H), 6.56 (s, 1H), 5.90 (s, 2H), 3.85–3.76 (m, 1H), 3.70–3.62 (m, 2H), 3.31 (s, 3H), 3.27–3.20 (m, 1H), 3.12–2.95 (m, 1H), 2.81 (dd, J = 34.5, 18.6 Hz, 2H), 2.62 (d, J = 16.7 Hz, 1H), 2.52–2.37 (m, 1H), 1.33 (s, 18H).

9-((2,4-Dichlorophenyl)sulfonamido)-7-((2,4-dichlorophenyl)sulfonyl)-10-methoxy-5,6,7,8,13,13a-hexahydro-[1,3]dioxolo[4,5-g]isoquinolino[3,2-a]isoquinolin-7-ium Chloride (5m)

ESI-MS m/z: 740.96 (M – Cl)+. m.p. 226–228 °C. 1H NMR (400 MHz, chloroform-d) δ 7.91 (d, J = 8.6 Hz, 3H), 7.50 (t, J = 8.2 Hz, 3H), 7.22 (d, J = 8.6 Hz, 1H), 6.72–6.68 (m, 2H), 6.60 (s, 1H), 5.90 (s, 2H), 3.78–3.67 (m, 1H), 3.46 (dd, J = 21.8, 14.4 Hz, 2H), 3.31 (s, 3H), 3.27–3.20 (m, 1H), 3.15–2.95 (m, 1H), 2.90–2.79 (m, 1H), 2.77–2.59 (m, 2H), 2.50–2.37 (m, 1H).

9-((4-Chloro-3-nitrophenyl)sulfonamido)-7-((4-chloro-3-nitrophenyl)sulfonyl)-10-methoxy-5,6,7,8,13,13a-hexahydro-[1,3]dioxolo[4,5-g]isoquinolino[3,2-a]isoquinolin-7-ium Chloride (5n)

ESI-MS m/z: 763.00 (M – Cl)+. m.p. 187–189 °C. 1H NMR (400 MHz, chloroform-d) δ 8.38 (s, 2H), 8.21 (dd, J = 8.5, 2.2 Hz, 1H), 8.14 (d, J = 7.4 Hz, 1H), 7.78 (d, J = 8.5 Hz, 2H), 7.29 (s, 1H), 6.77 (d, J = 8.5 Hz, 1H), 6.69 (s, 1H), 6.59 (s, 1H), 5.93 (s, 2H), 3.83–3.63 (m, 3H), 3.43 (s, 3H), 3.28 (d, J = 15.4 Hz, 1H), 3.16–3.00 (m, 1H), 2.96–2.72 (m, 2H), 2.65 (d, J = 15.6 Hz, 1H), 2.57–2.43 (m, 1H).

9-((2,4-Dinitrophenyl)sulfonamido)-7-((2,4-dinitrophenyl)sulfonyl)-10-methoxy-5,6,7,8,13,13a-hexahydro-[1,3]dioxolo[4,5-g]isoquinolino[3,2-a]isoquinolin-7-ium Chloride (5o)

ESI-MS m/z: 785.05 (M – Cl)+. m.p. 188–190 °C. 1H NMR (400 MHz, chloroform-d) δ 7.94 (d, J = 8.6 Hz, 3H), 7.49 (t, J = 8.3 Hz, 3H), 7.21 (d, J = 8.3 Hz, 1H), 6.69–6.60 (m, 2H), 6.58 (s, 1H), 5.93 (s, 2H), 3.77–3.68 (m, 1H), 3.46 (dd, J = 21.6, 14.2 Hz, 2H), 3.31 (s, 3H), 3.28–3.23 (m, 1H), 3.13–2.96 (m, 1H), 2.92–2.79 (m, 1H), 2.76–2.56 (m, 2H), 2.51–2.38 (m, 1H).

Synthesis Procedure for 7-Sulfatetrahydroberberines

Synthesis of Intermediate 7

The mixture of berberine (371 mg, 1 mmol) and potassium carbonate (360 mg, 3 mmol) was dissolved in 80% methanol solution (10 mL), and the mixture was heated to reflux for the dissolution of berberine. Sodium borohydride (152 mg, 4 mmol) was added slowly into the reaction solution, which would react for 1 h. The reaction was monitored by TLC and cooled to room temperature. The reaction solution was evaporated under vacuum, and then the residue was dissolved in dichloromethane. The faint yellow intermediate 7 was purified from this residue by silica gel column chromatography with ethyl acetate/petroleum ether (1:1) as an eluent. The yield of tetrahydroberberine 7 was about 60%.

9,10-Dimethoxy-5,8,13,13a-tetrahydro-6H-[1,3]dioxolo[4,5-g]isoquinolino[3,2-a]isoquinoline (7)

ESI-MS m/z: 340.12 (M + H)+. 1H NMR (400 MHz, chloroform-d) δ 6.86 (d, J = 8.4 Hz, 1H), 6.79 (d, J = 8.4 Hz, 1H), 6.73 (s, 1H), 6.59 (s, 1H), 5.91 (s, 2H), 4.24 (d, J = 15.8 Hz, 1H), 3.85 (s, 6H), 3.53 (d, J = 15.5 Hz, 2H), 3.26–3.07 (m, 3H), 2.87–2.74 (m, 1H), 2.70–2.56 (m, 2H).

Synthesis of 7-Sulfatetrahydroberberines 8a–8l

At room temperature, the mixture of tetrahydroberberine 7 (100 mg, 0.3 mmol) and potassium carbonate (124 mg, 0.9 mmol) was dissolved in acetonitrile (10 mL) followed by the addition of each kind of benzene sulfonyl chloride (0.6 mmol). The mixture solution was stirred overnight. The reaction was monitored by TLC, and the reaction solution was filtered under vacuum. The filtrate was evaporated and separated by silica gel column chromatography with methanol and dichloromethane (1:10) as an eluent to afford 7-sulfatetrahydroberberines 8a–8l. The yields of these compounds were 40%–60%.

9,10-Dimethoxy-7-(phenylsulfonyl)-5,6,7,8,13,13a-hexahydro-[1,3]dioxolo[4,5-g]isoquinolino[3,2-a]isoquinolin-7-ium Chloride (8a)

ESI-MS m/z: 477.15 (M – Cl)+. m.p. 172–174 °C. 1H NMR (400 MHz, DMSO-d6) δ 9.88 (s, 1H), 8.94 (s, 1H), 8.21 (d, J = 9.2 Hz, 1H), 8.00 (d, J = 9.1 Hz, 1H), 7.80 (s, 1H), 7.60 (d, J = 2.4 Hz, 1H), 7.58 (s, 1H), 7.34–7.27 (m, 3H), 7.09 (s, 1H), 6.17 (s, 2H), 4.93 (t, J = 6.1 Hz, 2H), 4.09 (s, 3H), 4.07 (s, 3H), 3.23–3.18 (m, 2H).

7-((2-Fluorophenyl)sulfonyl)-9,10-dimethoxy-5,6,7,8,13,13a-hexahydro-[1,3]dioxolo[4,5-g]isoquinolino[3,2-a]isoquinolin-7-ium Chloride (8b)

ESI-MS m/z: 495.14 (M – Cl)+. m.p. 195–197 °C. 1H NMR (600 MHz, DMSO-d6) δ 9.88 (s, 1H), 8.93 (s, 1H), 8.21 (d, J = 9.1 Hz, 1H), 8.00 (d, J = 9.1 Hz, 1H), 7.80 (s, 1H), 7.66 (t, J = 8.5 Hz, 1H), 7.14–7.07 (m, 4H), 6.17 (s, 2H), 4.93 (t, J = 6.4 Hz, 2H), 4.09 (s, 3H), 4.07 (s, 3H), 3.20 (t, J = 6.3 Hz, 2H).

7-((3-Fluorophenyl)sulfonyl)-9,10-dimethoxy-5,6,7,8,13,13a-hexahydro-[1,3]dioxolo[4,5-g]isoquinolino[3,2-a]isoquinolin-7-ium Chloride (8c)

ESI-MS m/z: 495.14 (M – Cl)+. m.p. 199–201 °C. 1H NMR (600 MHz, DMSO-d6) δ 9.88 (s, 1H), 8.93 (s, 1H), 8.51 (s, 4H), 8.20 (d, J = 9.0 Hz, 1H), 8.00 (d, J = 9.1 Hz, 1H), 7.80 (s, 1H), 7.09 (s, 1H), 6.17 (s, 2H), 4.95–4.90 (m, 2H), 4.09 (s, 3H), 4.07 (s, 3H), 3.20 (t, J = 6.3 Hz, 2H).

7-((4-Fluorophenyl)sulfonyl)-9,10-dimethoxy-5,6,7,8,13,13a-hexahydro-[1,3]dioxolo[4,5-g]isoquinolino[3,2-a]isoquinolin-7-ium Chloride (8d)

ESI-MS m/z: 495.14 (M – Cl)+. m.p. 191–193 °C. 1H NMR (600 MHz, DMSO-d6) δ 9.92 (s, 1H), 9.02 (s, 1H), 8.51 (s, 4H), 8.21 (d, J = 9.1 Hz, 1H), 8.03 (d, J = 9.1 Hz, 1H), 7.82 (s, 1H), 7.09 (s, 1H), 6.17 (s, 2H), 4.98–4.91 (m, 2H), 4.09 (s, 3H), 4.07 (s, 3H), 3.23–3.17 (m, 2H).

7-((2-Chlorophenyl)sulfonyl)-9,10-dimethoxy-5,6,7,8,13,13a-hexahydro-[1,3]dioxolo[4,5-g]isoquinolino[3,2-a]isoquinolin-7-ium Chloride (8e)

ESI-MS m/z: 511.11 (M – Cl)+. m.p. 203–205 °C. 1H NMR (600 MHz, DMSO-d6) δ 9.88 (s, 1H), 8.94 (s, 1H), 8.20 (d, J = 9.1 Hz, 1H), 8.00 (d, J = 9.1 Hz, 1H), 7.87 (dd, J = 7.5, 1.9 Hz, 3H), 7.80 (s, 1H), 7.37 (d, J = 1.3 Hz, 1H), 7.09 (s, 1H), 6.17 (s, 2H), 4.93 (t, J = 6.1 Hz, 2H), 4.09 (s, 3H), 4.07 (s, 3H), 3.23–3.17 (m, 2H).

7-((4-Chlorophenyl)sulfonyl)-9,10-dimethoxy-5,6,7,8,13,13a-hexahydro-[1,3]dioxolo[4,5-g]isoquinolino[3,2-a]isoquinolin-7-ium Chloride (8f)

ESI-MS m/z: 511.11 (M – Cl)+. m.p. 207–209 °C. 1H NMR (600 MHz, DMSO-d6) δ 9.90 (s, 1H), 8.97 (s, 1H), 8.51 (s, 4H), 8.21 (d, J = 9.2 Hz, 1H), 8.01 (d, J = 9.1 Hz, 1H), 7.81 (s, 1H), 7.09 (s, 1H), 6.17 (s, 2H), 4.94 (t, J = 6.3 Hz, 2H), 4.09 (s, 3H), 4.07 (s, 3H), 3.21 (t, J = 6.4 Hz, 2H).

7-((4-Bromophenyl)sulfonyl)-9,10-dimethoxy-5,6,7,8,13,13a-hexahydro-[1,3]dioxolo[4,5-g]isoquinolino[3,2-a]isoquinolin-7-ium Chloride (8g)

ESI-MS m/z: 555.06 (M – Cl)+. m.p. 211–213 °C. 1H NMR (600 MHz, DMSO-d6) δ 9.92 (s, 1H), 9.00 (s, 1H), 8.21 (d, J = 9.2 Hz, 1H), 8.02 (d, J = 9.1 Hz, 1H), 7.81 (s, 1H), 7.52 (d, J = 3.6 Hz, 1H), 7.34–7.27 (m, 3H), 7.09 (s, 1H), 6.17 (s, 2H), 4.95 (t, J = 6.3 Hz, 2H), 4.09 (s, 3H), 4.07 (s, 3H), 3.24–3.17 (m, 2H).

9,10-Dimethoxy-7-((4-nitrophenyl)sulfonyl)-5,6,7,8,13,13a-hexahydro-[1,3]dioxolo[4,5-g]isoquinolino[3,2-a]isoquinolin-7-ium Chloride (8h)

ESI-MS m/z: 522.13 (M – Cl)+. m.p. 190–192 °C. 1H NMR (600 MHz, DMSO-d6) δ 9.89 (s, 1H), 8.94 (s, 1H), 8.23–8.16 (m, 5H), 8.00 (d, J = 9.1 Hz, 1H), 7.79 (s, 1H), 7.08 (s, 1H), 6.17 (s, 2H), 4.93 (t, J = 6.1 Hz, 2H), 4.09 (s, 3H), 4.07 (s, 3H), 3.24–3.17 (m, 2H).

9,10-Dimethoxy-7-((3-(trifluoromethyl)phenyl)sulfonyl)-5,6,7,8,13,13a-hexahydro-[1,3]dioxolo[4,5-g]isoquinolino[3,2-a]isoquinolin-7-ium Chloride (8i)

ESI-MS m/z: 545.13 (M – Cl)+. m.p. 191–193 °C. 1H NMR (600 MHz, DMSO-d6) δ 9.89 (s, 1H), 8.95 (s, 1H), 8.46 (s, 1H), 8.20 (d, J = 9.1 Hz, 1H), 8.05–7.97 (m, 3H), 7.80 (s, 2H), 7.09 (s, 1H), 6.17 (s, 2H), 4.93 (t, J = 6.4 Hz, 2H), 4.09 (s, 3H), 4.07 (s, 3H), 3.20 (t, J = 6.3 Hz, 2H).

9,10-Dimethoxy-7-((4-methoxyphenyl)sulfonyl)-5,6,7,8,13,13a-hexahydro-[1,3]dioxolo[4,5-g]isoquinolino[3,2-a]isoquinolin-7-ium Chloride (8j)

ESI-MS m/z: 507.16 (M – Cl)+. m.p. 189–191 °C. 1H NMR (600 MHz, DMSO-d6) δ 9.89 (s, 1H), 8.94 (s, 1H), 8.20 (d, J = 9.1 Hz, 1H), 8.00 (d, J = 9.1 Hz, 1H), 7.80 (s, 1H), 7.51 (d, J = 8.6 Hz, 5H), 6.17 (s, 2H), 4.93 (t, J = 6.4 Hz, 2H), 4.09 (s, 3H), 4.07 (s, 3H), 3.75 (s, 3H), 3.22–3.17 (m, 2H).

7-((4-(tert-Butyl)phenyl)sulfonyl)-9,10-dimethoxy-5,6,7,8,13,13a-hexahydro-[1,3]dioxolo[4,5-g]isoquinolino[3,2-a]isoquinolin-7-ium Chloride (8k)

ESI-MS m/z: 533.21 (M – Cl)+. m.p. 196–198 °C. 1H NMR (600 MHz, DMSO-d6) δ 9.89 (s, 1H), 8.94 (s, 1H), 8.20 (d, J = 9.1 Hz, 1H), 8.00 (d, J = 9.1 Hz, 1H), 7.80 (s, 1H), 7.51 (d, J = 8.6 Hz, 5H), 6.17 (s, 2H), 4.93 (t, J = 6.4 Hz, 2H), 4.09 (s, 3H), 4.07 (s, 3H), 3.22–3.17 (m, 2H), 1.28 (s, 9H).

7-((4-Chloro-3-nitrophenyl)sulfonyl)-9,10-dimethoxy-5,6,7,8,13,13a-hexahydro-[1,3]dioxolo[4,5-g]isoquinolino[3,2-a]isoquinolin-7-ium Chloride (8l)

ESI-MS m/z: 556.09 (M – Cl)+. m.p. 193–195 °C. 1H NMR (600 MHz, DMSO-d6) δ 9.88 (s, 1H), 8.94 (s, 1H), 8.20 (d, J = 9.1 Hz, 1H), 8.00 (d, J = 9.1 Hz, 1H), 7.86 (d, J = 2.0 Hz, 2H), 7.85 (d, J = 2.0 Hz, 3H), 6.17 (s, 2H), 4.93 (t, J = 6.3 Hz, 2H), 4.09 (s, 3H), 4.07 (s, 3H), 3.23–3.17 (m, 2H).