Dereplication of Components Coupled with HPLC-qTOF-MS in the Active Fraction of Humulus japonicus and It’s Protective Effects against Parkinson’s Disease Mouse Model

Humulus japonicus is an annual plant belonging to the Cannabacea family, and it has been traditionally used to treat pulmonary tuberculosis, dysentery, chronic colitis, and hypertension. We investigated the active components against Parkinson’s disease from H. japonicus fraction (HJF) using high performance liquid chromatography (HPLC) coupled with quadruple-time-of-flight mass spectroscopy (qTOF-MS) and NMR. Fourteen compounds were isolated from HJF, including one new compound, using HPLC-qTOF-MS and NMR. The major compounds of HJF were luteolin-7-O-glucoside and apigenin-7-O-glucoside, and there was approximately 12.57- and 9.68-folds increase in the contents of these flavonoids compared to those of the 70% EtOH extract. Apigenin and luteolin exhibited the strongest inhibitory effects on monoamine oxidase (MAO) B enzyme activity. In animal studies, limb-use behavior was significantly reduced by unilateral 6-OHDA lesion and ipsilateral rotations. These results indicated that oral administration of 300 mg/kg HJF resulted in the improvement of motor asymmetry and motor impairment in unilateral 6-OHDA-lesioned mice. HJF, including active components leads to an improvement of motor behavior in a Parkinson’s disease mouse model.


Introduction
Parkinson's disease (PD) is a common neurodegenerative disorder that caused motor problems such as resting tremor, muscle rigidity, and postural instability [1]. PD pathogenesis is associated with genetic disposition, age and environmental factors, mitochondrial dysfunction, oxidative damage, inflammation, and microgliosis [2,3]. Oxidative damage by the generation of reactive oxygen species (ROS) causes neuronal membrane damage, including damage to membrane proteins, unsaturated lipids, and DNA [4]. Recent studies have suggested that the side chains of membrane proteins and lipids are modified by ROS, and a reduction in membrane unsaturation is associated with decreased membrane
The .68-folds increase in the contents of these flavonoids compared to those of the 70% EtOH extract. Among 14 compounds, the contents of minor compounds eugenyl-β-d-glucopyranoside, vitexin, luteolin and apigenin in 70% ethanol extract were 0.40, 1.90, 0.24, and 0.29 µg/mg. In HJF, they were increased to 2.70, 10.4, 0.50, and 0.40 µg/mg, respectively. At the concentration measured, other compounds were not detected due to the limit of peak area absorption.
We tested the MAO B inhibitory activities of 14 isolated compounds (20 µM) isolated from H. japonicus. Safinamide, a reversible MAO B inhibitor, was used as a positive control in the experiment [30]. DMSO was used as a control group to exclude the influence of the solvent. As shown in Figure 4, isolated compounds 12, 13, and 14 namely vitexin, apigenin and luteolin at 20 µM exhibited the greatest inhibition of MAO B, with enzyme inhibitory activities ranging from 58% to 79%, while compounds 3 and 11 showed weak inhibitory activities compared to the control group.
To determine the effect of HJF on Parkinson's disease, we used a mouse model with a unilateral 6-OHDA lesion. The unilateral dopaminergic neuronal death induced by 6-OHDA lesions in mice caused limb-use asymmetry in the cylinder [31]. This limb-use behavior was significantly reduced by unilateral 6-OHDA lesions in both vehicle-treated mice and HJF-treated mice ( Figure 5A). However, the HJF-treated group showed a much greater use of the contralateral limb compared to the vehicle group ( Figure 5A). 21 days after surgery, d-AMPH-induced rotational behavior was measured in the vehicle group and the HJF group. Unilaterally-caused dopaminergic cell death in the midbrain resulted in asymmetric rotational behavior in the mice. Figure 5 shows the rotational behavior of mice by following an injection of 5 mg/kg d-AMPH. 6-OHDA-lesioned mice displayed robust ipsilateral rotation in response to d-AMPH, while HJF-treated mice significantly decreased ipsilateral rotations ( Figure 5B,C). These results indicate that oral administration of 300 mg/kg HJF leads to improved motor asymmetry and motor impairment in unilateral 6-OHDA-lesioned mice. As oxidative damage in the brain provokes mitochondrial dysfunction and damage, oxidative stress is considered key to the pathogenesis of Parkinson's disease (PD). The 6-OHDA neurotoxin leads to the formation of reactive oxygen species (ROS) and causes cell death in neurons of an experimental animal model of PD. A previous study of 70% EtOH extract of H. japonicus demonstrated neuroprotective activity in 6-hydroxydopamine (6-OHDA) animal model, and alleviated dopaminergic cell death and fiber loss caused by 6-OHDA [17]. However, identification and characterization of active compounds against PD from the 70% EtOH extract of H. japonicus have not been reported. MAO-B inhibitory activity-guided fractionations from H. japonicus lead to the isolation of one new and 13 known compounds.    Compounds 2 and 10 were reported to have ACE (angiotensin I converting enzyme) inhibition activity and may decrease oxidative stress and inflammation [22]. Megastigmane glucoside, compound 8, showed a cell protective effect on benzo[a]pyrene-induced cytotoxicity [32]. Flavonoids, luteolin-7-O-glucoside (LG, 11), apigenin-7-O-glucoside (AG, 3), vitexin, luteolin and apigenin with MAO-B inhibitory activity were isolated from HJF. LG and AG were major components of HJF, and there was a 12.57-and 9.68-fold increase in the HJF compared to those in 70% EtOH extracts of H. japonicus. Luteolin and apigenin have been reported to have microglia-mediated inflammatory effects by regulating the induction of CD40 in response with IFN-γ in N9 and murine-derived primary microglial cells [33]. Luteolin also reduced the pathology of Aβ-amyloid deposit due to traumatic brain injury (TBI) in animal models. In addition, luteolin also significantly inhibited GSK activation, tau phosphorylation and microglial-induced release of inflammatory cytokines [34]. The metabolism of flavonoids was studied in previously cultured intestinal Caco-2 cells and rat intestines. Flavonoid glycosides generally undergo deglucosylation by the lactase phlorizin hydrolase (LPH) and gut enzymes or bacteria localized to the intestinal barrier [35]. LG hydrolyzes to luteolin and produces aglycone through intestinal enzymes or bacteria. Metabolized luteolin from LG was absorbed in the digestive tract without any conjugated forms including glucuronides. The bioavailability of luteolin-7-O-glucoside was approximately 10%, and biotransformed luteolin was detected. In a pharmacokinetics study with rats, the ratio of biotransformation into luteolin was approximately 48.78% [36]. AG is also hydrolyzed enzymatically to apigenin by β-glucosidases in human small intestine [37]. The bioavailability of AG in a germ-free rat model suggested that the major metabolite of AG is apigenin [38]. Therefore, the metabolic pathway of AG may have a pattern similar to that of LG. Humulones and lupulone which were predicted by HPLC-qTOF-MS have been reported antioxidant effects in the literature [39] and are expected to have synergistic effects in H. japonicus.
Previously, 70% EtOH extract of H. japonicus showed neuroprotective effects on dopaminergic neurons in 6-OHDA-lesioned mice [17]. Oral administration of 70% EtOH extract of H. japonicus at 300 mg/kg showed decreased d-AMPH-induced ipsilateral rotations, but these decreases were not significant. In this study, a strong motor improvement effect was observed in the 6-OHDA-lesioned PD model when the same dose of HJF was administered. The right forelimb damaged by unilateral 6-OHDA injection was effectively improved ( Figure 5A). Moreover, rotations under the presynaptically active d-AMPH were markedly suppressed by the administration of 300 mg/kg HJF ( Figure 5B,C). Although dopaminergic neuron death in the midbrain was not investigated in this study, rotation intensity was correlated with the degree of nigrostriatal denervation [40]. Thus, dereplication of the active constituents from HJF using HPLC-qTOF-MS and NMR resulted in purified compounds (1-14) and predicted compounds (15)(16)(17)(18)(19)(20)(21)(22)(23). In particular, the HJF showed a significant increase in contents of AG and LG compared to the 70% EtOH extract of H. japonicus. These results suggest that flavonoids may have a neuroprotective effect in various neurodegenerative diseases including Parkinson's disease (PD).

General Information
Biologically active fraction of H. japonicus against Parkinson's disease (HJF) was analyzed by an HPLC-qTOF-MS system consisting of a 1260 HPLC system and a 6530 qTOF-MS system (Agilent Technol., Santa Clara, CA, USA). Semi-preparative HPLC was performed using a Gilson HPLC system with a 321 pump and a UV/VIS-155 detector. An RS Tech Optima Pak C 18 column (10 × 250 mm, 10 µm particle size, Korea) was used as the HPLC column. NMR spectra for 1D ( 1 H and 13 C NMR) and 2D (NOESY, HSQC and HMBC) were collected using a Bruker 400 MHz, 500 MHz and a JEOL JNM-ECA 600 MHz NMR spectrometer. ZEOprep 60 Silica gel (40−63 µm particle size), Cosmosil 75C 18 -prep, Diaion TM Ion exchange resin for HP-20, and GE Healthcare Sephadex TM LH-20 (18-111 µm) were used for column chromatography. Thin layer chromatography was performed using silica gel 60 F 254 and RP-18 F 254 plates. All solvents for extraction and isolation were of analytical grade.

Plant Material
The aerial part of H. japonicus was collected in Gangwon province in the Republic of Korea in 2014 and was botanically identified by Professor W.K. Oh. The voucher specimen (SNU2014-08) was deposited at the College of Pharmacy, Seoul National University, Seoul, Korea.

In vitro MAO-B Inhibition Assay
The reaction was initiated by incubating the test compound at the desired concentrations with kynuramine (Sigma Chemical Co., St Louis, MO, USA) in 0.1 M potassium phosphate buffer (pH 7.4) in a 96-well plate. After incubation at 37 • C for 10 min, 50 µL of MAO B enzyme (Sigma Chemical Co.) in 0.1 M potassium phosphate buffer was added to each reaction well. The reaction progress was terminated with the addition of 75 µL of 2 M sodium hydroxide after 20 min. The fluorescence intensity of 4-hydroxyquinoline was measured at an excitation wavelength of 310 nm and an emission wavelength of 400 nm using a SpectraMax GEMNI XPS microplate reader (Molecular Devices Corporation, Sunnyvale, CA, USA) [41]. The inhibition ratio is expressed as the activity percentage compared to the vehicle control.

Animals
Nine-week-old male C57BL/6J mice were provided by the KRIBB and housed in a 22 ± 1 • C and humidity-controlled (50-60%) environment under specific pathogen-free conditions under 12-h light-dark cycles. The mice were randomly divided into two groups: vehicle control (0.5% carboxymethyl cellulose, n = 10) and HJF (300 mg/kg/day, n = 13). The vehicle or HJF was administrated daily by oral gavage for three days before 6-OHDA lesion and 21 days after 6-OHDA lesion. All animal experiments were approved by the Institutional Animal Care and Use Committee (IACUC) of KRIBB and were performed in accordance with the institutional guidelines of KRIBB.

6-Hydroxydopamine (6-OHDA) Lesion
The procedure for 6-OHDA lesioning was previously described [31]. Mice were anesthetized with a mixture of ketamine hydrochloride (Yuhan corporation, Seoul, Korea) and xylazine hydrochloride (Bayer Korea) and mounted in a stereotactic frame (Stoelting Europe, Dublin, Ireland) equipped with a mouse adaptor. Mice were pretreated with desipramine hydrochloride [Sigma Aldrich, 25 mg/kg, intraperitoneal (i.p.)] 30 min before surgery to prevent noradrenergic neuronal damage, and unilaterally injected 6-OHDA in a volume of 3 µL (Sigma-Aldrich Co. LLC, 2 µg/µL diluted in saline containing 0.2% ascorbic acid) into the left dorsal striatum at the following coordinates: anteroposterior (AP), 1.2 mm; lateral, −1.8 mm; and dorsoventral, −3.6 mm. The mice were kept on a warming plate until they awoke from the anesthesia and were subsequently returned to their cages until they were used. To avoid dehydration, the mice received glucose (JW Pharmaceutical, Korea) in saline (10 mL/kg, s.c.) during the procedure. Additionally, in the evening of the first week after surgery, the food pellets were soaked in water and were placed in a shallow vessel on the floor of the cages.

Cylinder Test
Individual mice were placed into a transparent acrylic cylinder (diameter, 20 cm) and recorded for 5 min. Tests were conducted before and 7 days after 6-OHDA injection. The number of times that the right and left forelimbs contacted the wall was calculated by the blind observer. Use of the impaired (right) forelimb was expressed as a percentage of the total number of supporting wall contacts [31].

D-AMPH-induced Rotation Test
Dextro-amphetamine (d-AMPH, USP, Rockville, MD, USA, 5 mg/kg, i.p.)-induced rotation was measured 21 days after 6-OHDA injection. The ipsilateral turning behavior induced by d-AMPH administration was recorded for 60 min. The number of ipsilateral rotations was analyzed by a SMART video tracking system (Panlab, Spain).