Synthesis of Rottlerone Analogues and Evaluation of Their α-Glucosidase and DPP-4 Dual Inhibitory and Glucose Consumption-Promoting Activity

Our previous study found that desmethylxanthohumol (1) inhibited α-glucosidase in vitro. Recently, further investigations revealed that dehydrocyclodesmethylxanthohumol (2) and its dimer analogue rottlerone (3) exhibited more potent α-glucosidase inhibitory activity than 1. The aim of this study was to synthesize a series of rottlerone analogues and evaluate their α-glucosidase and DPP-4 dual inhibitory activity. The results showed that compounds 4d and 5d irreversibly and potently inhibited α-glucosidase (IC50 = 0.22 and 0.12 μM) and moderately inhibited DPP-4 (IC50 = 23.59 and 26.19 μM), respectively. In addition, compounds 4d and 5d significantly promoted glucose consumption, with the activity of 5d at 0.2 μM being comparable to that of metformin at a concentration of 1 mM.


Introduction
Diabetes mellitus is a multi-gene metabolic disease that is characterized by hyperglycemia that is caused by a relative (Type 2 Diabetes) or absolute (Type 1 Diabetes) insulin deficiency. The number of diabetes patients has been increasing rapidly in recent years due to changes in people's living standards and dietary habits. The number of adult diabetes patients worldwide is about 463 million, and this number may reach 578 million by 2030, according to the statistical results of the International Diabetes Federation (IDF) in 2019 [1]. Diabetes is becoming one of the most widespread health burning problems in the elderly [2].
α-Glucosidase plays an important role in the physiological process of postprandial digestion and the absorption of carbohydrates in food in patients with Type 2 Diabetes, and it is the key enzyme involved in the hydrolysis of dietary carbohydrates [3]. α-Glucosidases specifically hydrolyze the α-1,4-glucopyranosidic bond to release glucose. Therefore, the inhibition of digestive α-glucosidase is one therapeutic approach to slowing down carbohydrate digestion and glucose absorption, thereby stabilizing blood glucose level and preventing hyperglycemia in diabetic patients. α-Glucosidase inhibitors have become a better choice in the case of Asian diabetic patients that are associated with high consumption of carbohydrates in their staple diet [4]. Dipeptidyl peptidase-4 (DPP-4) inhibitors enhance the plasma level of active glucagon-like peptide-1 (GLP-1) and glucose-dependent insulinotropic polypeptide (GIP), which results in increased insulin 4) inhibitors enhance the plasma level of active glucagon-like peptide-1 (GLP-1) and glucose-dependent insulinotropic polypeptide (GIP), which results in increased insulin secretion and decreased glucagon secretion. It has been reported that the addition of a DPP-4 inhibitor to patients with type 2 diabetes that is inadequately controlled by an α-glucosidase inhibitor achieved better glycemic control without further increasing the risk of weight gain and hypoglycemia [5]. Thus, the development of dual inhibitors of α-glucosidase and DPP-4 may provide a novel approach to the discovery of new antidiabetic agents. To the best of our knowledge, such α-glucosidase and DPP-4 dual inhibitors have never been reported. It should be noted that the development of new α-glucosidase inhibitors has been largely neglected, since the introduction of acarbose, voglibose, and miglitol leave much room for the improvement of anti-diabetic therapies [3].
Desmethylxanthohumol (1, Figure 1), isolated from hops, has been found to display anticancer and antioxidant activity [6,7]. In a previous study, we found that desmethylxanthohumol inhibits α-glucosidase in vitro [8]. Furthermore, our recent investigations revealed that dehydrocyclodesmethylxanthohumol (2) and its dimer analogue rottlerone (3) [9] exhibited more potent α-glucosidase inhibitory activity than 1 (Figure 1). Therefore, we prepared a series of rottlerone derivatives 4a-d and 5a-d for the purpose of studying structure-activity relationships and improving the α-glucosidase inhibitory potencies of compounds 1-3. In addition, the synthesized compounds were also evaluated for DPP-4 inhibition with the goal of identifying potential dual-targeted anti-diabetic drugs.

Synthesis and Characterisation of Rottlerone Analogues
The synthesis of rottlerone analogues 4a-d, 5a-d, and 12-13 was accomplished, as shown in Scheme 1. The mono-MOM protected 2',4',6'-trihydroxyacetophenone (6) was synthesized from 2',4',6'-trihydroxyacetophenone according to a known procedure [10]. Compound 6 was converted to 7 by cyclization with 3-methyl-2-butenal, followed by removal of the MOM protection to provide 8. Conversion of the latter to dimer 9 was achieved by reaction with formaldehyde [11]. The rottlerone derivatives 4a-h were then obtained by base-catalyzed aldol condensation of 9 with the corresponding substituted benzaldehydes. Deprotection of 4e-h using 2M hydrochloride acid afforded 5a-d. Finally, methylated rottlerone derivatives 12 and 13 were obtained by aldol condensation of compound 9 with benzaldehyde followed by O-methylation using CH2N2 in DCM.

Rottlerone Analogues Inhibited α-Glucosidase and DPP-4
The α-glucosidase and DPP-4 inhibitory activities of the synthesized compounds were evaluated. Acarbose and sitagliptin, clinically useful α-glucosidase and DPP-4 inhibitors, respectively, were chosen as the reference compounds for activity comparison. Dimer 9 was first evaluated and compared with the inhibitory activity of compound 2 and rottlerone (3), as shown in Table 1. Interestingly, replacement of the styrene moiety of rottlerone by a methyl group (9) led to a reduction in α-glucosidase inhibition. The varia-

Rottlerone Analogues Inhibited α-Glucosidase and DPP-4
The α-glucosidase and DPP-4 inhibitory activities of the synthesized compounds were evaluated. Acarbose and sitagliptin, clinically useful α-glucosidase and DPP-4 inhibitors, respectively, were chosen as the reference compounds for activity comparison. Dimer 9 was first evaluated and compared with the inhibitory activity of compound 2 and rottlerone (3), as shown in Table 1. Interestingly, replacement of the styrene moiety of rottlerone by a methyl group (9) led to a reduction in α-glucosidase inhibition. The variation of the electronic properties of para substituents on the phenyl group did not significantly alter the α-glucosidase inhibitory activity (4a and 4b vs. 3). However, the introduction of a hydroxy group at this para position (5a) resulted in a 10-fold improvement in activity. Furthermore, the α-glucosidase inhibitory activities were found to be two-fold and four-fold higher than 5a in the case of the o,p-dihydroxyphenyl and o,m,p-trihydroxyphenyl derivatives 5b and 5d, respectively. The lower activity of the o,m-dihydroxy derivative 5c is perhaps due to the formation of an intramolecular hydrogen bond between the adjacent hydroxy groups. Notably, compound 5d also showed a modest DPP-4 inhibitory activity with an IC 50 = 26.19 µM. Altogether, these results revealed that free phenolic hydroxy groups are crucial for obtaining the desired activities. Compounds 4c and 4d, in which the phenyl ring is replaced by a furan or a thiophene ring, possessed three-fold and 26-fold higher α-glucosidase inhibitory activity than 3. Noteworthy, compound 3 demonstrated some degree of DPP-4 inhibitory activity (Table 1). This suggests that the phenyl ring of the rottlerone analogues can be advantageously replaced by a heterocyclic ring and it opens the way to a wide variety of further possible structural modifications to improve activity.
The methylation of the phenol hydroxyls on the 2H-chromene nucleus was also found to have considerable influence on α-glucosidase inhibitory activity. In particular, the trimethylated derivative 13 demonstrated a 10-fold enhancement of this activity when compared to the demethylated analogue 12 and comparable activity to that of 4d and 5d. We hypothesize that the 7-methoxy group prevents the formation of an H-bond between the 7-and 7 -hydroxyls, resulting in a vastly different steric environment favorable to α-glucosidase inhibitory activity.

The Enzyme Kinetics of Compounds 4d and 5d
The most active α-glucosidase inhibitors of our series, 4d and 5d, were then used for kinetic analyses in order to shed light on their mechanism of action. The plots of the velocity versus enzyme concentrations in the presence of different concentrations of 4d and 5d gave a family of parallel straight lines (Figure 2a,b). Increasing the inhibitor concentrations did not affect the slopes of the lines, indicating that 4d and 5d are irreversible inhibitors. Moreover, the value of 1/V increased with increasing concentrations of 5d, but the intercept on the X-axis and Y-axis remained constant. This result indicates that compound 5d is a mixed competitive α-glucosidase inhibitor. intercept on the X-axis and Y-axis remained constant. This result indicates that compound 5d is a mixed competitive α-glucosidase inhibitor.

Glucose Consumption of Compounds 4d and 5d
In order to further study the anti-diabetic activity of compounds 4d and 5d, their effects on cell-based glucose consumption were evaluated using metformin as a positive control. Compound 5d was observed to significantly promote glucose consumption at 0.2 and 0.04 μM, while compound 4d showed similar potency at 0.2 μM, as shown in Figure  3.

Glucose Consumption of Compounds 4d and 5d
In order to further study the anti-diabetic activity of compounds 4d and 5d, their effects on cell-based glucose consumption were evaluated using metformin as a positive control. Compound 5d was observed to significantly promote glucose consumption at 0.2 and 0.04 µM, while compound 4d showed similar potency at 0.2 µM, as shown in Figure 3. intercept on the X-axis and Y-axis remained constant. This result indicates that compound 5d is a mixed competitive α-glucosidase inhibitor.

Glucose Consumption of Compounds 4d and 5d
In order to further study the anti-diabetic activity of compounds 4d and 5d, their effects on cell-based glucose consumption were evaluated using metformin as a positive control. Compound 5d was observed to significantly promote glucose consumption at 0.2 and 0.04 μM, while compound 4d showed similar potency at 0.2 μM, as shown in Figure  3.

Chemistry
All of the reagents and solvents were commercially available and used without further purification. The 1 H-NMR and 13 C-NMR spectra were recorded on a Bruker AM-400 NMR spectrometer (Billerica, Middlesex, MA, USA) in CDCl 3 or d 6 -DMSO (dimethyl sulfoxide). Mass spectra were obtained on a Q-TOF mass spectrometer (Agilent, Santa Clara, CA, USA) and high resolution mass spectra were obtained on a hybrid IT-TOF mass spectrometer (Shimadzu LCMS-IT-TOF, Kyoto, Japan). Compound purity was determined by UV absorbance at 254 nm during tandem liquid chromatography/mass spectrometry (Agilent, Santa Clara, CA, USA). All of the final products had a purity of ≥95%, as determined using this method.

General Procedure for the Synthesis of Compounds 5a-5d
Compounds 4e-4h (0.10 mmol) in MeOH (3 mL) and THF (3 mL) was added 2 N HCl (1 mL) at 0 • C. After addition, the mixture was heated to 45 • C for 24 h. The mixture was poured in cold NaHCO 3 (aq, 10 mL) and then extracted with EtOAc (3 × 20 mL). The combined organic layers were washed with brine, dried over Na 2 SO 4 , and then concentrated under reduced pressure. The crude product was purified by column chromatography on silica (petroleum ether/ethyl acetate 4:1) to afford a compounds 5a-5d. A solution of KOH (20% aq, 20 mL) and ether (20 mL) was added methyl nitrosourea (1 g) at 0 • C and then stirred for 10 min. Compound 9 (1.0 mmol) in CH 2 Cl 2 (10 mL) was dripped above ether solution at 0 • C. The reaction mixture was stirred for 6 h at r.t. The reaction was quenched with glacial acetic acid (0.1 mL) and then extracted with EtOAc (3 × 100 mL). The combined organic layers were washed with brine, dried over Na 2 SO 4 , and then concentrated under reduced pressure. The crude product was purified by column chromatography on silica (petroleum ether/ethyl acetate 100:1 and 50:1) to afford compounds 10 and 11.

Kinetics of Enzyme Inhibition
The kinetics of enzyme inhibition was performed with the same assays detailed above, in the presence of different concentrations of compounds 4d and 5d, substrate, and enzyme, respectively. The inhibitory kinetics of 4d and 5d on α-glucosidase was analyzed using the Lineweaver-Burk plot of the substrate concentration and velocity.

Glucose Consumption Assay
Briefly, the HepG2 cells were cultured in Dulbecco's-modified Eagle's medium (DMEM) containing 25 mmol/L D-glucose, 10% heat-inactivated fetal bovine serum (FBS), 10 U/mL penicillin, and 10 mg/mL streptomycin at 37 • C, 5% CO 2 atmosphere. The culture solution was replaced every other day and then passaged once for 2-3 d. The cells were seeded into 96-well plate with twenty-four wells left as blanks. After reaching 70-80% confluence, the cells were washed with PBS twice and the medium was replaced by DMEM containing 25 mmol/L D-glucose to hunger for 24 h. The cells were washed with PBS twice and the medium was replaced by RPMI-1640 containing 11.1 mmol/L glucose that was supplemented with 0.2% bovine serum albumin (BSA). To the medium was then added 1 mmol/L metformin or purified product at different concentrations, and DMSO was used as the blank control. The glucose concentration in the medium determined a DMSO after 24 h treatment. Glucose consumption = glucose concentrations of blank wells − glucose concentrations of plated wells. The sulforhodamine B (SRB) assay was used to adjust the glucose consumption values. The SRB assay is based on the measurement of cellular protein content. After an incubation period, the cells were fixed with 10%, 4 • C trichloroacetic acid, stained for 1 h at 4 • C, and then washed with purified water. After drying, 60 µL SRB (3 mg/mL) was added into each well and then left at room temperature for 30 min. SRB was dissolved by 1% acetic acid before air drying. Bound SRB was solubilized with 150 µL Tris-base (10 mmol/L, pH 10.5) solution for OD determination at 546 nm while using a microplate reader [13].

Conclusions
A series of rottlerone analogues were synthesized and evaluated for their α-glucosidase and DPP-4 inhibitory activities. Compounds 4d and 5d were identified as new potent α-glucosidase inhibitors and moderate DPP-4 inhibitors. The kinetic analysis showed that compounds 4d and 5d are irreversible α-glucosidase inhibitors, and that 5d acts in a mixed competitive inhibitory mode. In addition, compounds 4d and 5d showed glucose consumption-promoting activity in HepG2 cells. Of these two compounds, the activity of 5d at 0.2 µM was comparable to that of metformin at a concentration of 1 mM. In view of these significant results, compound 5d, which combines both α-glucosidase and DPP-4 inhibitory activities, presents great potential as a lead compound for the development of a novel dual inhibitor treatment strategy for type 2 diabetes.

Conflicts of Interest:
The authors declare no conflict of interest.
Sample Availability: All samples are available from the authors.