Key words Anredera cordifolia - Basellaceae - adenosine - cordysinin B - dementia - CRE-mediated
transcription
Introduction
AD is the most common neurodegenerative disorder and has become a severe social problem
in advanced countries [1 ]. However, effective preventive and fundamental therapeutic methods for AD have not
yet been established. Notably, Aβ peptide in AD patients’ brains decreases CREB signaling pathway activation to inhibit
hippocampal LTP formation [2 ], and Aβ oligomers inhibit CREB activation in hippocampal neurons [3 ]. Furthermore, the CREB/CRE pathway greatly contributes to LTP, a synaptic memory
model, and memory formation in vivo [[4 ]
[5 ]
[6 ]
[7 ]
[8 ].
Numerous natural resources have simultaneously provided useful pharmacological tools
[9 ] and novel leading compounds for drug development [10 ]. We have reported that Nob, a polymethoxylated flavone from the peel of Citrus depressa , activates the CREB/CRE pathway in PC12D cells or cultured rat hippocampal neurons
to exhibit memory-improving actions in various animal models of dementia [11 ]
[12 ]
[13 ]
[14 ].
There is considerable interest in identifying safe and effective compounds from natural
resources that enhance the function of CREB transcription factor coupled with CRE-mediated
transcription, which could improve memory deficits in AD.
In our survey of natural resources having the increasing action of the CRE-mediated
transcription activity, we found that Anredera cordifolia , like Nob, enhances this activity and reverses memory impairment caused by NMDA receptor
antagonist MK-801 in mice [15 ]. Alternatively, A. cordifolia possesses pharmacologically interesting actions such as anti-obesity, anti-hyperlipidemia,
antihypertensive, antidiabetic, antioxidant, and anti-inflammatory activities [16 ]
[17 ]. Here, we unprecedentedly described that adenosine and cordysinin B are isolated
as active ingredients from A. cordifolia , possessing increasing activity on CRE-mediated transcription. It is also suggested
that these actions are caused by activating adenosine receptors.
Results and Discussion
AD is the most common neurodegenerative disorder showing progressive loss of memory
and cognitive function. Notably, late LTP formation, a synaptic memory model, is blocked
by the Aβ peptide by inhibiting the CREB signaling pathway [2 ]. Using a new strategy in our survey of numerous natural resources activating the
CREB/CRE pathway, we found that A. cordifolia , like Nob from citrus peels, with activating actions on CRE-mediated transcription,
improved MK-801-induced memory impairment [13 ]
[14 ]
[15 ]
[18 ].
A. cordifolia leaves (202 g) were extracted using MeOH (0.4 L) for 5 days at room temperature in
a stationary state to obtain the MeOH extract (47.5 g) by removing the solvent using
a rotary evaporator. The MeOH extracts were fractionated by monitoring the CRE-mediated
transcriptional activity using PC12 cells as illustrated in [Fig. 1 ]. The four layers partitioned with n -hexane, EtOAc, and BuOH had CRE-mediated transcriptional activities [12.2-fold activation
(BuOH layer), 4.4-fold activation (EtOAc layer), 1.5-fold activation (n -hexane layer), and 1.3-fold activation (H2 O layer) at 30 μg/mL]. The BuOH layer (5.8 g) was subjected to silica gel 60 N column
chromatography (φ45 × 420 mm, CHCl3 –MeOH system) to give fractions 1A (CHCl3 :MeOH=13:1, 0.27 L; CHCl3 :MeOH=9:1, 0.25 L), 1B (CHCl3 :MeOH=9:1, 0.15 L), 1C (CHCl3 :MeOH=9:1, 0.1 L; CHCl3 :MeOH=7:1, 0.25 L), 1D (CHCl3 :MeOH=7:1, 0.05 L; CHCl3 :MeOH=5:1, 0.3 L), 1E (CHCl3 :MeOH=5:1, 0.15 L; CHCl3 :MeOH=3:1, 0.15 L), 1F (CHCl3 :MeOH=3:1, 0.15 L), 1G (CHCl3 :MeOH=3:1, 0.10 L), 1H (CHCl3 :MeOH=3:1, 0.15 L; CHCl3 :MeOH=2:1, 0.15 L), 1I (CHCl3 :MeOH =2:1, 0.20 L; CHCl3 :MeOH=1:1, 0.15 L), 1J (CHCl3 :MeOH=1:1, 0.30 L), and 1K (CHCl3 :MeOH=1:1, 0.05 L; MeOH, 0.15 L; MeOH + 0.1% TFA, 0.15 L). Fraction 1D (CHCl3 :MeOH=7:1–5:1, 33 mg) was suspended in 5% MeOH and subjected to ODS open column chromatography
(φ15 × 140 mm, MeOH–H2 O system) at once to give subfractions 2A (H2 O, 0.02 L), 2B (5% MeOH, 0.02 L), 2C (10% MeOH, 0.02 L), 2D (15% MeOH, 0.02 L), 2E
(20% MeOH, 0.02 L), 2F (25% MeOH, 0.02 L), 2G (40% MeOH, 0.02 L), 2H (50% MeOH, 0.02
L), 2I (MeOH, 0.02 L), and 2J (MeOH + 0.1% HCOOH, 0.02 L). Subfraction 2E (20% MeOH,
0.7 mg) was regarded as compound 2 ([Fig. 2 ]). A part of fraction 1H (CHCl3 : MeOH=3:1–2:1, 671 mg of 681 mg) was subjected to silica gel PSQ 100B column chromatography
(φ25 × 160 mm, CHCl3 –MeOH system) to give subfractions 3A-3I. A part of subfraction 3E (CHCl3 : MeOH=5:1, 54 mg of 64 mg) was subjected to HPLC [COSMOSIL Cholester (φ10.0 × 250
mm); eluent: 20% MeOH; flow rate: 5.0 mL/min; UV detection: 254 nm] to give compound
1 (4.4 mg, t
R 4.7 min) ([Fig. 2 ]).
Fig. 1 Scheme for isolation of pharmacologically active compounds from A. cordifolia.
Fig. 2 Chemical structures of compounds 1 (adenosine) and 2 (cordysinin B) from the MEAC.
DMSO and Nob served as the negative and positive controls, respectively, in measuring
CRE-mediated transcriptional activity. As can be seen in [Fig. 3 ], MEAC induces a powerful increasing action on CRE-mediated transcription in PC12
cells. Also, the BuOH layer showed the strongest activity among the four layers and
then was chromatographed on silica gel to afford the active compounds 1 and 2 . The physicochemical properties of the active compounds 1 and 2 properly correspond to those of adenosine and cordysinin B, respectively (Table S1 and S2 ) [19 ]
[20 ]. Cordysinin B was previously isolated from Cordyceps sinensis
[20 ]. Therefore, it is concluded that the major active ingredients of A. cordifolia are adenosine and cordysinin B, respectively.
Fig. 3 Increasing activities of the MEAC and the layer of n -hexane (Hex), ethyl acetate (EtOAc), n -buthanol (BuOH), or water (H2 O) on CRE-mediated transcription in PC12 cells. Data are shown as the mean±SEM (n=4).
For some data points, error bars are smaller than the symbol size.
[Fig. 4 ] indicates that introducing a methyl group into the OH group at the C-2′ position
of adenosine decreases the CRE-mediated transcription activity approximately 100 times.
Therefore, these results suggest that the OH group is important for developing the
activity.
Fig. 4 Concentration-dependent effects of adenosine a , cordysinin B b , and nobiletin c on CRE-mediated transcription in PC12 cells. Data are shown as the mean±SEM (n=4).
For some data points, error bars are smaller than the symbol size.
To demonstrate the mechanisms responsible for CRE-mediated transcription induction
by adenosine and cordysinin B, we used adenosine A receptor antagonists CGS 15943
(a nonselective adenosine A receptor antagonist), DPCPX (an adenosine A1 receptor antagonist), SCH 58261 (an adenosine A2A receptor antagonist), MRS 1754 (an adenosine A2B receptor antagonist), and MRS 1523 (an adenosine A3 receptor antagonist). CRE-mediated transcription induced by adenosine and cordysinin
B was blocked by pretreating with CGS 15943, DPCPX, SCH 58261, or MRS 1754, although
the effect of MRS 1754 on adenosine-induced CRE-mediated transcription was weak ([Fig. 5 ]). In contrast, MRS 1523 did not affect CRE-mediated transcription induced by adenosine
or cordysinin B. These results suggest that adenosine and cordysinin B induce CRE-mediated
transcription, at least in part, by A1 , A2A , and A2B receptors.
Fig. 5 Effects of adenosine receptor antagonists on adenosine- and cordysinin B-induced
enhancement of CRE-mediated transcription in PC12 cells. Data are shown as the mean±SEM
(n=4). *** P<0.001 vs. DMSO-treated control; ### p<0.001 vs. adenosine alone-treated group; +++ p<0.001 vs. cordysinin B alone-treated group.
Conclusively, adenosine and cordysinin B are isolated active ingredients from A. cordifolia and possess the activity to induce CRE-mediated transcription, an important event
for memory formation. It is also suggested that these actions are induced, at least
in part, by adenosine A1 , A2A , and A2B receptors, but not A3 receptor.
Materials and Methods
Plant material
A. cordifolia leaves were collected from the hothouse of Sankyo Co., Ltd. in Fuji, Japan from May
2015 to July 2016. The plant material was identified by Dr. Koji Kajima. A voucher
specimen was deposited at Sankyo Co., Ltd. (SS-U001). A. cordifolia, a perennial Basellaceae native to South America, has been used as a traditional medicine
in China and Japan.
General experimental procedures in chemistry
An ECZ-600 spectrometer (JEOL) was used for NMR spectroscopy and the chemical shift
of the NMR solvent was used as an internal standard. An HPLC system (Shimadzu) comprising
an SCL-10AVP system controller, LC-20AD pump, DGU-12A online degasser, SIL-20A, CTO-10ASVP
column oven, SIL-20A autosampler, SPD-M20A PDA detector, and CLASS-VP software was
used. The following adsorbents were used for purification: silica gel 60 F254 (0.25
mm; Merck) and silica gel 60 RP-8 F254 S (0.25 mm; Merck) for analytical TLC; Silica
gel 60 N (Kanto), silica gel PSQ 100B, and Chromatrex ODS (Fuji Silysia chemical)
for column chromatography; COSMOSIL Cholester (φ10.0 × 250 mm; Nacalai Tesque) for
HPLC.
Reagents
CGS 15943 was obtained from Cayman Chemical. DPCPX, SCH 5826, MRS 1754, and MRS 1523
were purchased from Abcam. Adenosine (purity ≥ 99%) and cordysinin B (2’-O-Methyl
Adenosine) (purity ≥ 98%) were from Sigma-Aldrich and Toronto Research Chemicals,
respectively. Nob was extracted and isolated from C. depressa peels as described previously [11 ]
[12 ]. The purity of Nob was confirmed to be almost 100%.
Culture of rat pheochromocytoma (PC12) cells
PC12 cells were grown in DMEM supplemented with 10% heat-inactivated horse serum (Gibco
by Life Technologies), 5% heat-inactivated fetal bovine serum (Gibco), and 1% penicillin/streptomycin
(Gibco) at 37°C in a humidified atmosphere of 95% air and 5% CO2 .
Measurements of cyclic AMP-response element-mediated transcriptional activity in PC12
cells
Transient transfection and the reporter gene assay were prepared as described previously
[11 ]. PC12 cells cultured in 96-well plates (4 × 104 /well) were transfected for 5 h with 0.2 mg of the reporter plasmid pCRE (Clontech)
and 0.04 mg of the transfection efficiency Renilla luciferase phRG-TK plasmid (Promega) using LipofectAMINE (Invitrogen) according to
the manufacturer’s instructions. After transfection, the medium was replaced with
a fresh medium containing advanced DMEM, 1% horse serum (Gibco), and 1% fetal bovine
serum (Gibco), and the cells were incubated overnight. After incubation, the cells
were stimulated for 5 h with MEAC, adenosine, or cordysinin B. In the reporter gene
assay, commercially available adenosine standard (Sigma-Aldrich, catalog A4036) and
cordysinin B standard (Toronto Research Chemicals, catalog M276150) were used. For
experiments using adenosine receptor antagonists, the cells were preincubated with
CGS 15943(3 μM), DPCPX (0.5 μM), SCH 58261 (3 nM), MRS 1754 (30 μM), or MRS 1523 (1 μM)
for 30 min before stimulation and then stimulated for 5 h with adenosine or cordysinin
B in the presence of an individual antagonist. Luminescence was measured using a Dual-Luciferase
Reporter Assay System according to the manufacturer’s instructions (Promega). All
treatments were performed in quadruplicate on at least three independent cultures.
Statistical analysis
The results are expressed as the mean±SEM. Data were analyzed using one-way analysis
of variance (ANOVA), followed by Tukey-Kramer test, and p<0.05 was considered statistically
significant difference.
Supporting information
1H NMR and MS spectra of compound 1, 1H and 13C NMR and MS spectra of compound 2,
and tables for comparison of the physicochemical properties between compound 1 and
adenosine and those between compound 2 and cordysinin B are available as Supporting
information.
