1-(2′-Bromobenzyl)-6,7-dihydroxy- N-methyl-tetrahydroisoquinoline and 1,2-Demethyl-nuciferine as Agonists in Human D2 Dopamine Receptors

Article abstract of DOI:10.1021/acs.jnatprod.9b00921

Certain D₂-like dopamine receptor (DR) agonists are useful therapeutically as antiparkinsonian drugs, whereas D₂-like DR antagonists or partial agonists are proven effective as antipsychotics. Two isoquinoline derivatives, 1-(2′-brom

Full text of DOI:10.1021/acs.jnatprod.9b00921

pubs.acs.org/jnp
Article
1‑(2′-Bromobenzyl)-6,7-dihydroxy‑N‑methyl-tetrahydroisoquinoline
and 1,2-Demethyl-nuciferine as Agonists in Human D2 Dopamine
Receptors
́
Andrea G. Silva, Laura Vila, Patrice Marques, Laura Moreno, Mabel Loza, María-Jesus Sanz,
Diego Cortes,* Marian Castro,* and Nuria Cabedo*
́
Cite This: https://dx.doi.org/10.1021/acs.jnatprod.9b00921
Read Online
sı
*
Metrics & More
Article Recommendations
Supporting Information
ACCESS
ABSTRACT: Certain D2-like dopamine receptor (DR) agonists are
useful therapeutically as antiparkinsonian drugs, whereas D2-like DR
antagonists or partial agonists are proven effective as antipsychotics.
Two isoquinoline derivatives, 1-(2′-bromobenzyl)-6,7-dihydroxy-N-
methyl-tetrahydroisoquinoline (Br-BTHIQ, 1) and 1,2-demethyl-
nuciferine (aporphine, 2), were herein synthesized, and their
dopaminergic affinity in cloned human D2R, D3R, and D4R subtypes
and their behavior as agonists/antagonists were evaluated. They showed
affinity values (Ki) for hD2, hD3, and hD4 DR within the nanomolar
range. The trends in affinity were hD4R ≫ hD3R > hD2R for Br-BTHIQ (1) and hD2R > hD4R > hD3R for 1,2-demethyl-nuciferine
(2). The functional assays of cyclic adenosine monophosphate signaling at human D2R showed a partial agonist effect for Br-BTHIQ
(1) and full agonist behavior for aporphine (2), with half maximal effective concentration values of 2.95 and 10.2 μM, respectively.
Therefore, both isoquinolines 1 and 2 have emerged as lead molecules for the synthesis of new therapeutic drugs that ultimately may
be useful to prevent schizophrenia and Parkinson’s disease, respectively.
enzylisoquinoline alkaloids constitute a large structural
Dopamine (3,4-dihydroxyphenethylamine) is one of the
Bgroup of secondary metabolites that are derived most important neurotransmitters involved in the control of
biosynthetically from tyrosine.1 Condensation of dopamine
essential functions, including locomotor activity, cognition,
emotion, and memory. This catecholamine exerts its effects by
acting on five subtypes of DRs, D1−D5, encoded in humans by
genes DRD1−DRD5, respectively. DRs belong to the seven
transmembrane G protein-coupled receptor (GPCR) families
and are classified into two major classes, D1-like DR (D1 and
D5) and D2-like DR (D2, D3, and D4), based on the ability to
regulate cyclic adenosine monophosphate (cAMP) produc-
tion.12 The D2R subtype has two isoforms, D2S (D2 short) and
D2L (D2 long), which differ in the presence of 29 additional
amino acids in the third intracellular loop, whereas D4R has
many variants.13,14 D2-like DR subtypes share a high degree of
sequence homology (D2R shares 75 and 53% with D3R and
D4R, respectively), with differences only in their tissue
expression pattern, density, and physiological functions.14 For
this reason, despite the crystal structure elucidation of D2R,15
D3R,16 and D4R17 and the greatly improved design of ligands,
it is difficult to obtain selective compounds for each receptor
subtype. D1-like DRs are coupled to Gs proteins, and they
and 4-hydroxyphenylacetaldehyde by norcoclaurine synthase
produces (S)-norcoclaurine, which is considered the first 1-
benzylisoquinoline alkaloid and the common precursor to
isoquinoline derivatives, including 1-benzylisoquinolines and
aporphines.2 They are present mainly in Papaveraceae,
Menispermaceae, Berberidaceae, and Ranunculaceae, among
other plant families.3 Nelumbo nucifera Gaertn., also called
Asian lotus or sacred lotus, belongs to the family
Nelumbonaceae. Its leaves and flower buds are an important
source of aporphines (e.g., nuciferine and N-methylasimilo-
bine) and 1-benzylisoquinoline alkaloids (e.g., armepavine and
N-methylcoclaurine) (Figure 1).3 These alkaloids have
displayed a variety of biological activities, including antiox-
idant,4 antimicrobial,5 antiplatelet,6 antidiabetic, antihyperlipi-
demic,7 and antihypertensive8 effects. Nuciferine has been
described as producing an antipsychotic-like effect in rodents
that is attributed, at least in part, to partial agonist activity on
dopaminergic receptors (DRs).9 In addition, the interaction
with DRs of aporphines and 1-benzylisoquinolines of natural
or synthetic origin has been extensively reported.10 For
instance, apomorphine (Figure 1), a semisynthetic derivative
from morphine with dopaminergic agonist activity and
selectivity for D2-like DR, has some use in treating Parkinson’s
disease.11
Received: September 24, 2019
© XXXX American Chemical Society and
American Society of Pharmacognosy
https://dx.doi.org/10.1021/acs.jnatprod.9b00921
J. Nat. Prod. XXXX, XXX, XXX−XXX
A

Journal of Natural Products
pubs.acs.org/jnp
Article
work, Br-BTHIQ (1) and aporphine (2) were synthesized to
evaluate (i) their dopaminergic affinity for the cloned human
D2R, D3R, and D4R subtypes and (ii) their potential D2-like
DR agonist/antagonist behavior. These results should allow
further lead compounds to be developed toward useful
potential agents in neurological and/or neuropsychiatric
disorders.
RESULTS AND DISCUSSION
■
Chemistry. The synthesis of 1-(2′-bromobenzyl)-6,7-
dihydroxy-N-methyl-tetrahydroisoquinoline (1) and 1,2-de-
methyl-nuciferine (2) was carried out as previously reported.25
The synthetic route is outlined in Schemes 1 and 2. Initially,
Scheme 1. Synthesis of 1-(2′-Bromobenzyl)-6,7-dihydroxy-
a
1,2,3,4-tetrahydro-isoquinoline (Br-BTHIQ, 1)
Figure 1. Natural and semisynthetic isoquinoline derivatives.
increase intracellular cAMP through adenylate cyclase (AC)
activation, which results in the subsequent activation of protein
kinase A (PKA). In contrast, D2-like DRs are coupled to Gi/o
proteins, and they downregulate cAMP generation through AC
inhibition, which leads to diminished PKA activity.12,14 D2-like
DRs are implicated in numerous brain functions, including
regulation of locomotor activity, cognition, and motiva-
tion.13,14 This subfamily is clinically important as therapeutic
targets for antiparkinsonian (agonists) and antipsychotic
(antagonists or partial agonists) drugs.13,14
In the last 2 decades, our group has reported the synthesis
and structure−activity relationship (SAR) studies of isoquino-
line derivatives with affinity for D1- and D2-like DRs in rat
striatal membranes.18−25 These studies were supported by
molecular modeling studies, which revealed the importance of
the catechol group, secondary/tertiary amine, and hydrophobic
substitution pattern in the isoquinoline nucleus to improve the
affinity for each DR subfamily (Figure 2).26−31 In previous
studies into rat striatal membranes, it was determined that both
1-(2′-bromobenzyl)-6,7-dihydroxy-N-methyl-THIQ (Br-
BTHIQ, 1) and 1,2-demethyl-nuciferine (aporphine, 2)
showed high selectivity toward D2-like DR.25 In the present
a
Reaction conditions: (a) 2-BrC6H4CH2COCl, CH2Cl2, 5% aqueous
NaOH, room temperature, 16 h; (b) POCl3, CH3CN, N2, reflux, 1 h;
(c) NaBH4, CH3OH, N2, room temperature,
[(CH3)3COCO]2O, CH2Cl2, room temperature, 2 h; (e) LiAlH4,
tetrahydrofuran (THF), reflux, 4 h; and (f) BBr3, CH2Cl2, N2, room
temperature, 2 h.
2 h; (d)
the appropriate β-phenyl acetamide (1a) intermediate was
prepared under Schotten−Baumann conditions. Next, Bis-
chler−Napieralski cyclodehydration gave the isoquinoline
nucleus bearing an imine function, which was reduced to
give the corresponding brominated BTHIQ (1c). In a first
approach, a reaction sequence of N-protection followed by
carbamate reduction and subsequent O-demethylation gave
brominated N-methyl-BTHIQ (1) (Scheme 1). In a second
approach, the brominated N-Boc-BTHIQ (1d) intermediate
was subjected to a Buchwald−Hartwig cross-coupling reaction
toward carbon−carbon bond formation. Therefore, palladium-
catalyzed direct arylation32 with 2′-(diphenylphosphino)-N,N′-
dimethyl-(1,1′-biphenyl)-2-amine (PhDave-Phos) and Pd-
(OAc)2 generated a good yield of N-Boc nor-nuciferine (2a).
The carbamate-protecting group of compound 2a was reduced
to obtain nuciferine (2b), which was O-demethylated with
boron tribromide to give 1,2-demethyl-nuciferine (2) (Scheme
2).
Affinity of Isoquinolines 1 and 2 for the Human D2R,
D3R, and D4R Subtypes. The in vitro binding affinities of the
synthesized isoquinolines, 1-(2′-bromobenzyl)-6,7-dihydroxy-
N-methyl-tetrahydroisoquinoline (Br-BTHIQ, 1) and 1,2-
demethyl-nuciferine (aporphine, 2) were evaluated at the
cloned human D2R, D3R, and D4R subtypes. For this purpose,
competition radioligand-binding assays were conducted on cell
membranes stably expressing the three cloned human receptor
Figure 2. Structural features of D2-like dopaminergic isoquinolines.
B
https://dx.doi.org/10.1021/acs.jnatprod.9b00921
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
pubs.acs.org/jnp
Article
a
Scheme 2. Synthesis of 1,2-Demethyl-nuciferine (Aporphine, 2)
a
Reaction conditions: (a) PhDave-Phos, Pd(OAc)2, K2CO3, DMA, 130 °C, 4 h; (b) LiAlH4, THF, reflux, 4 h; and (c) BBr3, CH2Cl2, N2, room
temperature, 2 h.
Table 1. Affinity of Isoquinolines 1 and 2 in Human D2R, D3R, and D4R
a
a
a
hD2R
hD3R
hD4R
percent inhibition
at 10 μM (%)
percent inhibition
at 10 μM (%)
Ki
(nM)
percent inhibition
at 10 μM (%)
compound
pKi
Ki (nM)
pKi
Ki (nM)
pKi
Br-BTHIQ (1)
aporphine (2)
haloperidol
6.54 0.08
6.47 0.03
8.21 0.04
ND
286
340
6.22
ND
95
91
103
1
1
1
6.71 0.09
6.08 0.03
8.24 0.15
ND
197
838
5.16
ND
93
93
101
1
1
1
7.86 0.05
6.26 0.06
ND
13.8
551
ND
96
73
ND
1
2
clozapine
ND
ND
7.25 0.04
56.9
96
1
a
Percent inhibition and affinity (pKi and Ki) values were calculated by displacement of the specific binding of the radioligand [3H]-spiperone by
isoquinolines 1 and 2 on membranes from CHO-K1 cells stably expressing the cloned human D2, D3, and D4 receptors. Haloperidol was used as a
reference compound for the hD2R and hD3R assays, and clozapine was used as a reference compound for the hD4R assay. Data are expressed as the
mean standard error of the mean (SEM) (n = 2). ND = not determined.
subtypes.33 The data obtained are depicted in Table 1. The
competition curves of the isoquinolines for the specific binding
of [3H]-spiperone to human D2R, D3R, and D4R are shown in
Figure 3.
terms of their concentration−response curves, Br-BTHIQ (1)
and aporphine (2) exhibited potency within the micromolar
range (EC50 = 2.95 and 10.2 μM, respectively) and achieved an
agonist efficacy at the highest assayed concentration (100 μM)
of 48.4 and 92.7% of the maximal response of the D2R full
agonist quinpirole, respectively (Figure 4A). In view of these
results and with the aim of confirming the partial agonist
profile of Br-BTHIQ (1), the concentration−response curves
of quinpirole (full agonist) and quinpirole in the presence of
10 μM Br-BTHIQ (1) were performed (Figure 4B). Br-
BTHIQ (1) showed an agonistic effect at low quinpirole
concentrations while an antagonistic effect at higher concen-
trations of the full agonist (100 nM quinpirole). Indeed, a shift
of the quinpirole concentration−response curve to the right
resulted in a ≈35-fold increase in the quinpirole EC50 values
(Figure 4B). This response corresponds to that expected for a
full agonist (quinpirole) in the presence of a partial agonist (1-
Br-THIQ).
In summary, two isoquinoline derivatives, 1-(2′-bromoben-
zyl)-6,7-dihydroxy-tetrahydroisoquinoline (1) and 1,2-demeth-
yl-nuciferine (2), were prepared within good yields. Both
compounds 1 and 2 displayed affinity for the hD2, hD3, and
hD4 DR subtypes at nanomolar concentrations. The affinity
order of Br-BTHIQ (1) for the human DR subtypes was hD4R
≫ hD3R > hD2R, whereas aporphine (2) showed a hD2R >
hD4R > hD3R tendency. Br-BTHIQ (1) displayed a partial
agonist effect at D2R, whereas 1,2-demethyl-nuciferine (2)
behaved as a full agonist. When these results are taken
together, both isoquinolines 1 and 2 emerge as potential lead
molecules for the synthesis of new useful therapeutic drugs for
schizophrenia and Parkinson’s disease, respectively.
First, the displacement of the selective radioligand [3H]-
spiperone binding to hD2R, hD3R, or hD4R by Br-BTHIQ (1)
and aporphine (2) was evaluated at a single concentration of
10 μM. Haloperidol for D2R and D3R and clozapine for D4R
were used as reference compounds. The results showed that
both compounds 1 and 2 displayed a high percentage of
inhibition (>70%) of the specific radioligand binding (Table 1)
at the three receptor subtypes. Next, the concentration
response curves of the two isoquinoline compounds were
constructed to determine their affinities (equilibrium dissoci-
ation constants, Ki) at the different receptors. Both Br-BTHIQ
(1) and aporphine (2) were able to fully displace the specific
radioligand binding at hD2R, hD3R, and hD4R in a
concentration-dependent manner (Figure 3). Br-BTHIQ (1),
which contained a flexible appendage of the brominated benzyl
moiety, exhibited a high binding affinity, with Ki values of 286,
197, and 13.8 nM for hD2R, hD3R, and hD4R, respectively, and
notable selectivity for hD4 (Table 1). The semirigid 1,2-
demethyl-nuciferine (aporphine, 2) also displaced the selective
radioligand from its specific binding sites in hD2R, hD3R, or
hD4R at nanomolar concentrations (Ki = 340, 838, and 551
nM, respectively), with a slight selectivity toward hD2R (Table
1 and Figure 3).
Functional Activity of Isoquinolines 1 and 2 at D2R.
The behavior of Br-BTHIQ (1) and aporphine (2) as agonists
or antagonists at human D2R was evaluated using in vitro
functional assays through cAMP signaling in Chinese hamster
ovary (CHO)-K1 cells stably expressing the cloned human D2S
receptor. At the 10 μM concentration, Br-BTHIQ (1) and
aporphine (2) inhibited forskolin-stimulated cAMP production
to different extents (Table 2), as expected for D2R agonists. In
EXPERIMENTAL SECTION
■
General Experimental Procedures. Electron ionization mass
spectrometry (EIMS) and high-resolution electron ionization mass
spectrometry (HREIMS) were determined by a TripleTOF 5600 LC/
C
https://dx.doi.org/10.1021/acs.jnatprod.9b00921
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
pubs.acs.org/jnp
Article
Figure 4. Functional assays of cAMP signaling at human D2R. (A)
Concentration−response (inhibition of forskolin-stimulated cAMP
production) curves of isoquinolines 1 and 2 and the reference agonist
quinpirole in CHO-K1 cells stably expressing the cloned human D2
receptors. (B) Effect of Br-BTHIQ (1, 10 μM) on quinpirole
concentration−response curves. The graphs show data (mean
SEM) of n = 3−4 independent experiments performed in triplicate.
Figure 3. Radioligand displacement curves for isoquinolines 1 and 2
at human D2R, D3R, and D4R subtypes. The graphs show the results
(mean SEM) of n = 2.
column chromatography using silica gel 60 from Merck (40−63 μm).
Solvents were purchased from Scharlab SL (Barcelona, Spain) or
Sigma-Aldrich (St. Louis, MO, U.S.A.) and used without further
purification, unless otherwise noted. Dry and freshly distilled solvents
were used in those reactions carried out under a nitrogen atmosphere.
Synthesis of Isoquinolines 1 and 2. The syntheses and
spectroscopic characterization of Br-benzyltetrahydroisoquinoline
(1) and 1,2-demethyl-nuciferine (2), including the intermediates
1a−1d, 2a, and 2b, are described below.
Table 2. Evaluation of Efficacy and Potency of Isoquinolines
1 and 2 as Human D2R Agonists in Functional Assays of
cAMP Signaling in CHO-K1 Cells Stably Expressing the
Receptors
a
hD2R
Emax at
10 μM (%)
Emax at
100 μM (%)
EC50
(nM)
compound
pEC50
Br-BTHIQ (1)
aporphine (2)
quinpirole
31.9 6.5
50.7 8.6
98.8 0.6
48.4 5.3
92.7 6.5
100
5.53 0.31
4.99 0.22
6.82 0.12
2951
10233
150
β-(3,4-Dimethoxyphenyl)ethyl-(2′-bromophenyl)acetamide (1a).
2-Bromo-phenylacetyl chloride (1.2 mL, 8.07 mmol) was added
dropwise to a solution of 3,4-dimethoxyphenethylamine (2 g, 11.04
mmol) in CH2Cl2 (15 mL) and 5% aqueous NaOH (2.5 mL) at 0 °C.
The reaction was stirred at room temperature overnight and then
extracted with CH2Cl2 (3 × 10 mL). The combination of the organic
phases was washed with brine (2 × 10 mL) and H2O (2 × 10 mL),
dried over anhydrous Na2SO4, and evaporated to dryness. The residue
was purified by silica gel column chromatography (hexane−EtOAc,
a
Efficacy of isoquinolines 1 and 2 at the indicated concentration was
expressed as a percentage of the maximal quinpirole response (%
E
max). Data are the mean SEM of n = 3−4 independent experiments
performed in triplicate.
1
MS/MS system (AB SCIEX, Framingham, MA, U.S.A.) with the
5:5), to afford 2.35 g of amide (1a) as a white powder (77%). H
1
electron ionization (EI) method in the positive mode. H and 13C
NMR (500 MHz, CDCl3): δ 7.48 (1H, d, J = 7.8 Hz, H-3′), 7.20 (2H,
m, H-5′, H-6′), 7.05 (1H, m, H-4′), 6.66 (1H, d, J = 8.1 Hz, H-5),
6.56 (1H, d, J = 1.9 Hz, H-2), 6.52 (1H, dd, J = 8.1 and 1.9 Hz, H-6),
5.38 (1H, br s, CONH), 3.78 (3H, s, OCH3), 3.76 (3H, s, OCH3),
3.60 (2H, s, CH2CO), 3.40 (2H, dd, J = 12.8 and 6.9 Hz, CH2-α),
2.64 (2H, t, J = 6.9 Hz, CH2-β). 13C NMR (125 MHz, CDCl3): δ
169.4 (CO), 149.0 (C-3), 147.6 (C-4), 134.8 (C-1′), 133.0 (CH-3′),
131.6 (CH-6′), 131.1 (C-1), 129.0 (CH-4′), 127.9 (CH-5′), 124.9
(C-2′), 120.5 (CH-6), 111.8 (CH-2), 111.3 (CH-5), 55.9 (OCH3),
55.8 (OCH3), 44.0 (CH2CO), 40.7 (CH2-α), 35.0 (CH2-β).
Electrospray mass spectrometry (ESMS): m/z 380 [M + H]+.
nuclear magnetic resonance (NMR) spectra were recorded with a
Bruker AC-300 or AC-500 spectrometer (Bruker Instruments,
1
Darmstadt, Germany) using CDCl3 (δ 7.26 for H and δ 77.0 for
13C) as the solvent. NMR assignments were performed using
distortions enhancement by polarization transfer (DEPT), correlation
spectroscopy with a 45° mixing pulse (COSY-45), heteronuclear
single-quantum correlation (HSQC), and heteronuclear multiple-
bond correlation (HMBC) experiments. All reactions were monitored
by analytical thin-layer chromatography (TLC) with silica gel 60 F254
(Merck, Darmstadt, Germany). Residues were purified by silica gel
D
https://dx.doi.org/10.1021/acs.jnatprod.9b00921
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
pubs.acs.org/jnp
Article
1-(2′-Bromobenzyl)-6,5-dimethoxy-1,2,3,4-tetrahydroisoquino-
line (1c). A solution of phenylacetamide (1a, 300 mg, 0.82 mmol) in
dry CH3CN (20 mL) was treated with POCl3 (0.37 mL, 4.0 mmol)
and refluxed for 1 h under a nitrogen atmosphere. The reaction
mixture was concentrated to dryness, redissolved in water, and made
basic until pH ≈ 9. Then, the mixture was extracted with CH2Cl2 (3 ×
10 mL), and the combined organic solution was dried over anhydrous
Na2SO4 and evaporated to dryness to give imine (1b), which was
used without further purification. The residue was dissolved in MeOH
(25 mL) and treated with NaBH4 (86 mg, 2.27 mmol) at room
temperature. The reaction mixture was stirred for 2 h. Afterward, H2O
(5 mL) was added, and the organic solvent was removed under
reduced pressure. The aqueous mixture was made basic, extracted
with CH2Cl2 (3 × 10 mL), dried over Na2SO4, and concentrated. The
residue was purified by silica gel column chromatography (toluene−
EtOAc−MeOH−Et3N, 6:3:1:0.1) to obtain 144 mg of THIQ (1c,
50%) as a yellow oil. 1H NMR (500 MHz, CDCl3): δ 7.52 (1H, d, J =
7.7 Hz, H-3′), 7.20 (2H, m, H-5′, H-6′), 7.05 (1H, m, H-4′), 6.65
(1H, s, H-8), 6.53 (1H, s, H-5), 4.20 (1H, dd, J = 9.8 and 3.9 Hz, H-
1), 3.79 (3H, s, OCH3), 3.74 (3H, s, OCH3), 3.28 (1H, dd, J = 13.6
and 6.9 Hz, Ha-α), 3.20 (1H, m, Ha-3), 2.90 (2H, m, Hb-3, Hb-α),
2.68 (2H, t, J = 5.8 Hz, H-4). 13C NMR (125 MHz, CDCl3): δ 147.5
(C-6), 147.1 (C-7), 138.8 (C-1′), 133.0 (C-3′), 132.0 (CH-6′), 130.5
(C-8a), 128.2 (CH-4′), 127.4 (CH-5′), 127.1 (C-4a), 124.9 (C-2′),
111.7 (CH-5), 109.7 (CH-8), 55.9 (OCH3), 55.8 (OCH3), 54.9
(CH-1), 43.1 (CH2-α), 39.5 (CH2-3), 29.4 (CH2-4). ESMS: m/z 362
[M]+.
30 min. The solvent was removed under reduced pressure and
purified by silica gel column chromatography (CH2Cl2−MeOH,
90:1) to obtain 25 mg of catechol-N-methyl-Br-THIQ (1, 55%) as a
yellow oil. 1H NMR (300 MHz, CD3OD): δ 7.56 (1H, dd, J = 7.8 and
1.3 Hz, H-3′), 7.21 (1H, td, J = 7.5 and 1.3 Hz, H-5′), 7.12 (1H, td, J
= 7.5 and 1.3 Hz, H-4′), 7.01 (1H, dd, J = 7.5 and 1.3 Hz, H-6′), 6.53
(1H, s, H-5), 5.86 (1H, s, H-8), 3.92 (1H, m, H-1), 3.31 (2H, m, Ha-
α, Ha-3), 2.90 (3H, m, Hb-α, Hb-3), 2.64 (2H, m, CH2-4), 2.51 (3H,
s, NCH3). 13C NMR (75 MHz, CD3OD): δ 145.4 (C-6), 144.0 (C-7),
139.6 (C-1′), 133.8 (2C, CH-3′, CH-6′), 129.3 (CH-4′), 128.4 (CH-
5′), 128.0 (C-2′), 125.9 (C-8a), 125.1 (C-4a), 116.1 (CH-5), 116.0
(CH-8), 63.4 (CH-1), 46.6 (CH2-3), 42.3 (NCH3), 41.1 (CH2-α),
25.2 (CH2-4). High-resolution electrospray ionization mass spec-
trometry (HRESIMS): m/z 348.0589 [M + H]+ (calcd for
C17H18BrNO2, 348.0594).
tert-Butyloxycarbonyl-nor-nuciferine (2a). A mixture of N-Boc-
THIQ (1d, 451 mg, 0.98 mmol), 2-diphenylphosphino-2′-(N,N-
dimethylamino)biphenyl (PhDave-Phos, 75 mg, 0.20 mmol), Pd-
(OAc)2 (22 mg, 0.098 mmol), and K2CO3 (412 mg, 2.98 mmol) in
N,N-dimethylacetamide (DMA, 5 mL) was heated at 130 °C
overnight under a nitrogen atmosphere. Then, the reaction was
concentrated under reduced pressure, and the residue was purified by
silica gel column chromatography (CH2Cl2−MeOH, 99.5:0.5) to
1
obtain 310 mg of N-Boc-nor-nuciferine (2a, 83%) as a yellow oil. H
NMR (500 MHz, CDCl3): δ 8.42 (1H, d, J = 7.8 Hz, H-11), 7.27
(3H, m, H-8, H-9, H-10), 6.67 (1H, s, H-3), 4.65 (1H, m, H-6a), 4.42
(1H, m, Ha-5), 3.90 (3H, s, OCH3-2), 3.67 (3H, s, OCH3-1), 2.75
(5H, m, CH2-4, CH2-7, Hb-5), 1.49 (9H, s, 3×CH3). 13C NMR (125
MHz, CDCl3): δ 154.6 (CO), 151.9 (C-2), 145.5 (C-1), 137.0 (C-
7a), 131.7 (C-11a), 129.8 (C-4a), 128.4 (CH-11), 128.0 (C-11b),
127.6 (CH-8), 127.5 (CH-10), 126.9 (CH-9), 126.5 (C-11c), 111.4
(CH-3), 79.8 (OC-tBu), 59.9 (OCH3), 55.8 (OCH3), 51.5 (CH-6a),
38.4 (CH2-5), 35.4 (CH2-4), 30.4 (CH2-7), 28.5 (3×CH3). ESMS:
m/z 381 [M]+.
1-(2′-Bromobenzyl)-N-tert-butyloxycarbonyl-6,7-dimethoxy-
1,2,3,4-tetrahydroisoquinoline (1d). A solution of di-tert-butyl
dicarbonate (1.62 g, 7.42 mmol) in CH2Cl2 (10 mL) was added
dropwise to a solution of THIQ (1c, 2.23 g, 6.17 mmol) in CH2Cl2
(10 mL). After stirring at room temperature for 2 h, the solvent was
evaporated under reduced pressure and the residue was purified by
silica gel column chromatography (CH2Cl2−MeOH, 97:3) to obtain
2.76 g of N-Boc-THIQ (1d, 97%) as a brown oil. H NMR (500
Nuciferine (2b).34 A solution of N-Boc-nuciferine (2a, 200 mg,
0.52 mmol) in anhydrous THF (15 mL) was carefully treated with
LiAlH4 (102 mg, 2.69 mmol) under a nitrogen atmosphere at 0 °C.
Afterward, the reaction mixture was heated at reflux for 4 h. The
suspension was cooled at 0 °C, and water (1 mL) was added, followed
by 1 M NaOH solution (1 mL). The suspension was stirred for
additional 1 h at room temperature. The suspension was diluted with
EtOAc (10 mL) and filtered through Celite. The filtrate was washed
with water and brine, dried over anhydrous Na2SO4, and concentrated
under reduced pressure. The residue was purified by silica gel column
chromatography (CH2Cl2−MeOH, 94:6) to afford 120 mg of
nuciferine (2b) (78%) as a yellow oil. 1H NMR (300 MHz,
CDCl3): δ 8.36 (1H, dd, J = 8.4 and 1.9 Hz, H-11), 7.25 (3H, m, H-8,
H-9, H-10), 6.63 (1H, s, H-3), 3.89 (3H, s, OCH3-2), 3.66 (3H, s,
OCH3-1), 3.10 (4H, m, H-6a, Ha-4, Ha-5, Ha-7), 2.67−2.53 (2H, m,
Hb-7, Hb-4), 2.54 (3H, s, NCH3), 2.47 (1H, m, Hb-5). 13C NMR
(125 MHz, CDCl3): δ 151.9 (C-2), 145.1 (C-1), 136.3 (C-7a), 132.1
(C-11a), 128.5 (C-3a), 128.3 (CH-8), 127.8 (CH-9), 127.7 (C-11c),
127.3 (CH-10), 127.0 (CH-11), 126.8 (C-11b), 111.2 (CH-3), 62.3
(CH-6a), 60.2 (OCH3), 55.9 (OCH3), 53.3 (CH2-5), 44.0 (NCH3),
35.2 (CH2-7), 29.3 (CH2-4). ESMS: m/z 296 [M + H]+.
1,2-Demethyl-nuciferine (2). Nuciferine (2b, 0.47 mmol, 130 mg)
in dry CH2Cl2 (5 mL) was stirred at −78 °C. Then, 127 μL of BBr3
(1.32 mmol) was added under a nitrogen atmosphere, and the
resulting mixture was stirred for 2 h at room temperature. The
reaction was cooled at −78 °C to be quenched by the addition of
MeOH (0.5 mL) dropwise and then stirred at room temperature for
30 min. The solvent was evaporated to dryness and purified by silica
gel column chromatography (CH2Cl2−MeOH, 85:15) to afford 115
mg of 1,2-demethyl-nuciferine (2, 98%) as a yellow oil. 1H NMR (300
MHz, CD3OD): δ 8.28 (1H, dd, J = 7.8 and 1.3 Hz, H-11), 7.13 (3H,
m, H-8, H-9, H-10), 6.46 (1H, s, H-3), 3.04 (4H, m, H-6a, Ha-5, Ha-
7, Ha-4), 2.53 (6H, m, Hb-5, Hb-7, Hb-4, NCH3). 13C NMR (75
MHz, CD3OD): δ 146.1 (C-1), 143.1 (C-2), 136.1 (C-11a), 134.0
(C-3a), 129.6 (CH-11), 128.7 (CH-8), 127.8 (CH-9), 127.7 (CH-
10), 125.9, 123.8, 121.2 (3C, C-7a, C-11b, C-11c), 114.3 (CH-3),
1
MHz, CDCl3): δ 7.58 (1H, dd, J = 7.8 and 1.2 Hz, H-3′), 7.15 (3H,
m, H-4′, H-5′, H-6′), 6.81 (1H, s, H-5), 6.62 (1H, s, H-8), 5.35 (1H,
dd, J = 10.7 and 3.9 Hz, H-1), 4.37 (1H, m, Ha-3), 3.86 (6H, s,
2×OCH3), 3.28 (2H, m, Hb-3, Ha-α), 3.03 (1H, m, Hb-α), 2.95 (1H,
m, Ha-4), 2.67 (1H, m, Hb-4), 1.10 (9H, s, 3×CH3). 13C NMR (125
MHz, CDCl3): δ 154.8 (CO), 148.2 (C-7), 147.8 (C-8), 138.5 (C-
1′), 132.9 (CH-3′), 132.3 (CH-6′), 129.3 (C-8a), 128.6 (CH-4′),
127.9 (CH-5′), 126.8 (C-4a), 125.7 (C-2′), 110.1 (CH-8), 110.1
(CH-5), 79.8 (OC-tBu), 56.3 (2×OCH3), 54.1 (CH-1), 43.1 (CH2-
α), 36.5 (CH2-3), 28.5 (CH2-4), 27.82 (3×CH3). ESMS: m/z 462
[M]+.
1-(2-Bromobenzyl)-N-methyl-6,7-dimethoxy-1,2,3,4-tetrahydroi-
soquinoline (1e). A solution of N-Boc-THIQ (1d, 300 mg, 0.65
mmol) in dry THF (6 mL) was carefully treated with LiAlH4 (98 mg,
2.58 mmol) at 0 °C. The mixture was heated to reflux overnight.
Then, the reaction was cooled at 0 °C, quenched with water (1 mL)
dropwise, followed by 1 M NaOH solution (1 mL), and then stirred
at room temperature for 1 h. The suspension was diluted with EtOAc
(10 mL) and filtered through Celite. The filtrate was washed with
water and brine, dried over anhydrous Na2SO4, and concentrated
under reduced pressure to give 85 mg of N-methyl-Br-THIQ (1e,
35%) as a yellow oil, which was used for the next step without further
1
purification. H NMR (300 MHz, CDCl3): δ 7.54 (1H, dd, J = 7.9
and 1.3 Hz, H-3′), 7.16 (1H, m, H-5′), 7.14 (1H, m, H-4′), 7.06 (1H,
m, H-6′), 6.58 (1H, s, H-8), 5.92 (1H, s, H-5), 3.85 (1H, m, H-1),
3.83 (3H, s, OCH3), 3.50 (3H, s, OCH3), 3.29 (2H, m, Ha-α, Ha-3),
2.92 (3H, m, Hb-α, Hb-3), 2.55 (2H, m, CH2-4), 2.53 (3H, s,
NCH3). ESMS: m/z 377 [M + H]+.
1-(2-Bromobenzyl)-N-methyl-6,7-dihydroxy-1,2,3,4-tetrahydroi-
soquinoline (1). A solution of N-methyl-Br-THIQ (1e, 50 mg, 0.13
mmol) in dry CH2Cl2 (3 mL) was treated with BBr3 (38 μL, 0.4
mmol) dropwise at −78 °C under a nitrogen atmosphere. Then, the
reaction mixture was stirred to room temperature for 2 h. The
reaction was cooled at −78 °C to be quenched by the addition of
MeOH (0.5 mL) dropwise and then stirred at room temperature for
E
https://dx.doi.org/10.1021/acs.jnatprod.9b00921
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
pubs.acs.org/jnp
Article
64.0 (CH-6a), 54.5 (CH2-5), 43.4 (NCH3), 36.3 (CH2-7), 28.6
(CH2-4). HRESIMS: m/z 268.1325 [M + H]+ (calcd for C17H17NO2,
268.1332).
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge at
https://pubs.acs.org/doi/10.1021/acs.jnatprod.9b00921.
■
sı
Cell Cultures. CHO-K1 cells stably expressing human D2SR were
generated in-house as described.33 In brief, the CHO-K1 parental cell
line (Leibniz Institute DSMZ−German Collection of Microorganisms
and Cell Cultures GmbH, Braunschweig, Germany) was stably
transfected with the pcDNA3.1 vector (Invitrogen, Carlsbad, CA,
U.S.A.) containing the cDNA of human D2SR using the calcium
phosphate method and selected in medium containing 400 μg/mL
Geneticin (G-418). Cells were maintained in Dulbecco’s modified
Eagle’s medium−GlutaMAX-I (Gibco, Thermo Fisher Scientific,
Madrid, Spain) supplemented with 10% (v/v) fetal bovine serum
(Sigma-Aldrich, Madrid, Spain), 100 units/mL penicillin/0.1 mg/mL
streptomycin (Sigma-Aldrich, Madrid, Spain), 2 mM L-glutamine
(Sigma-Aldrich, Madrid, Spain), and 500 μg/mL Geneticin G418
(Gibco, Thermo Fisher Scientific, Madrid, Spain).
Radioligand-Binding Assays. Competition radioligand-binding
assays for isoquinolines 1 and 2 were conducted in membranes from
CHO-K1 cells stably expressing the cloned human D2SR (isoform D2
short) or the cloned human D3R or D4R (isoform D4.2) (PerkinElmer,
Waltham, MA, U.S.A.), following protocols described previously.33 In
brief, [3H]-spiperone was employed as a radioligand, whereas non-
specific binding was assessed in the presence of 10 μM sulpiride
(D2R), 1 μM haloperidol (D3R), or 25 μM haloperidol (D4R).
Isoquinolines 1 and 2 were evaluated in displacement curves at six
different concentrations ranging from 1 nM to 100 μM or from 0.1
nM to 10 μM. Haloperidol (from 0.1 nM to 10 μM) or clozapine
(from 1 nM to 100 μM) were included in the assays as reference
competitor ligands for both hD2R and hD3R binding (haloperidol) or
hD4R binding (clozapine). The affinities of isoquinolines expressed as
an equilibrium dissociation constant (Ki) were calculated using Prism
6 software (GraphPad, San Diego, CA, U.S.A.), by fitting the data
1H and 13C NMR spectra of synthesized compounds
(1a−1d, 1, 2a, 2b, and 2) (PDF)
AUTHOR INFORMATION
Corresponding Authors
■
Diego Cortes − University of Valencia, Valencia, Spain;
Phone: 34-96-3544975; Email: dcortes@uv.es
́
Marian Castro − Universidad de Santiago de Compostela,
Santiago de Compostela, Spain; Phone: 34-881-815458;
Email: marian.castro@usc.es
Nuria Cabedo − University Clinic Hospital of Valencia,
Valencia, Spain, University of Valencia, Valencia, Spain;
orcid.org/0000-0001-6729-8057; Phone: 34-96-
3544975; Email: ncabedo@uv.es
Other Authors
Andrea G. Silva − Universidad de Santiago de
Compostela, Santiago de Compostela, Spain
Laura Vila − University Clinic Hospital of Valencia,
Valencia, Spain
Patrice Marques − University Clinic Hospital of Valencia,
Valencia, Spain, University of Valencia, Valencia, Spain
Laura Moreno − University of Valencia, Valencia, Spain
Mabel Loza − Universidad de Santiago de Compostela,
Santiago de Compostela, Spain
́
María-Jesus Sanz − University Clinic Hospital of
from competition binding curves to a single binding site competition
model using the equations log EC50 = log(10log K (1 + HotNM/HotKdNM)
)
i
Valencia, Valencia, Spain, University of Valencia,
Valencia, Spain
and Y = bottom + (top − bottom)/(1 + 10X log EC ), where Y is
binding, HotNM is the concentration of radioligand in the assay,
HotKdNM is the equilibrium dissociation constant (Kd) of the
radioligand as determined in saturation binding experiments, and X is
the log molar concentration of the unlabeled compound.
50
Complete contact information is available at:
https://pubs.acs.org/10.1021/acs.jnatprod.9b00921
Notes
cAMP Functional Assay at D2R. The behavior of isoquinolines 1
and 2 as agonists or antagonists of D2R receptors was evaluated in in
vitro assays of cAMP signaling in the CHO-K1 cell line stably
expressing the cloned human D2S receptor employed in radioligand-
binding assays. Cellular cAMP levels were quantified using the
homogeneous time-resolved fluorescence (HTRF)-based cAMP kit
cAMP-Gs Dynamic HTRF Kit (Cisbio, Bioassays, Codolet, France)
following the protocol of the manufacturer. The possible agonist effect
of the compounds was evaluated by their ability to inhibit forskolin-
stimulated cAMP production either at a single concentration (10 μM)
and/or in concentration (from 1 nM to 100 μM)−response curves.
Cells seeded in 384-well plates in stimulation buffer containing 500
μM 3-isobutyl-1-methylxanthine (IBMX) were incubated with the test
compounds for 10 min at 37 °C. Then, forskolin (10 μM) was added,
and the incubation was continued for 5 min. After this time,
intracellular cAMP levels were quantified. Quinpirole (from 10 pM to
100 μM) was included as a control agonist in these assays. When the
effects of Br-BTHIQ (1, 10 μM) on quinpirole concentration−
response curves were investigated, compound 1 and quinpirole were
added simultaneously to the cells and the assay proceeded as
described above. For assessment of a possible antagonist effect, the
test compound was added to the cells 5 min prior to the addition of
the reference agonist quinpirole (100 nM) and assays were
subsequently carried out as described above. In all cases, basal
cAMP levels were determined in control wells in the absence of
compound, quinpirole, and forskolin.
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This work was supported by Grants SAF2014-57138-C2-1-R,
SAF2014-57845-R, SAF2017-89714-R, CP15/00150, and
PI18/01450 from the Spanish Ministry of Economy and
Competiveness, the Carlos III Health Institute, and the
European Regional Development Fund. Nuria Cabedo was
funded by the Miguel Servet Program, Carlos III Institute of
Health, co-funded by the European Fund for Regional
Development Fund and the European Social Fund. Patrice
Marques was funded by a pre-doctoral grant from the Spanish
Ministry of Economy and Competitiveness (FPI). Andrea G.
Silva was funded by a pre-doctoral grant from Xunta de Galicia
(Spain).
REFERENCES
■
672.
(1) Hagel, J. M.; Facchini, P. J. Plant Cell Physiol. 2013, 54, 647−
(2) Stadler, R.; Kutchan, T. M.; Zenk, M. H. Phytochemistry 1989,
28, 1083−1086.
(3) Menendez-Perdomo, I. M.; Facchini, P. J. Molecules 2018, 23,
2899.
(4) Cho, E. J.; Yokozawa, T.; Rhyu, D. Y.; Kim, S. C.; Shibahara, N.;
Park, J. C. Phytomedicine 2003, 10, 544−551.
F
https://dx.doi.org/10.1021/acs.jnatprod.9b00921
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
pubs.acs.org/jnp
Article
̀
(5) Agnihotri, V. K.; ElSohly, H. N.; Khan, S. I.; Jacob, M. R.; Joshi,
V. C.; Smillie, T.; Khan, I. A.; Walker, L. A. Phytochem. Lett. 2008, 1,
89−93.
(30) Newman-Tancredi, A.; Heusler, P.; Martel, J.-C.; Ormiere, A.-
M.; Leduc, N.; Cussac, D. Int. J. Neuropsychopharmacol. 2008, 11,
293−307.
(31) Gerlach, M.; Double, K.; Arzberger, T.; Leblhuber, F.;
Tatschner, T.; Riederer, P. J. Neural. Transm. 2003, 110, 1119−1127.
(32) Lafrance, M.; Blaquiere, N.; Fagnou, K. Eur. J. Org. Chem. 2007,
2007, 811−825.
(33) Selent, J.; Marti-Solano, M.; Rodriguez, J.; Atanes, P.; Brea, J.;
Castro, M.; Sanz, F.; Loza, M. I.; Pastor, M. Eur. J. Med. Chem. 2014,
77, 91−95.
(34) Guinaudeau, H.; Leboeuf, M.; Cave, A. Lloydia 1975, 38, 275−
338.
(6) Chen, K.-S.; Ko, F.-N.; Teng, C.-M.; Wu, Y.-C. J. Nat. Prod.
1996, 59, 531−534.
(7) Zhang, C.; Deng, J.; Liu, D.; Tuo, X.; Xiao, L.; Lai, B.; Yao, Q.;
Liu, J.; Yang, H.; Wang, N. Br. J. Pharmacol. 2018, 175, 4218−4228.
(8) Morales, M.; Bustamante, S.; Brito, G.; Paz, D.; Cassels, B. K.
Phytother. Res. 1998, 12, 103−109.
(9) Farrell, M. S.; McCorvy, J. D.; Huang, X.-P.; Urban, D. J.; White,
K. L.; Giguere, P. M.; Doak, A. K.; Bernstein, A. I.; Stout, K. A.; Park,
S. M.; Rodriguiz, R. M.; Gray, B. W.; Hyatt, W. S.; Norwood, A. P.;
Webster, K. A.; Gannon, B. M.; Miller, G. W.; Porter, J. H.; Shoichet,
B. K.; Fantegrossi, W. E.; Wetsel, W. C.; Roth, B. L. PLoS One 2016,
11, No. e0150602.
̀
(10) Cabedo, N.; Berenguer, I.; Figadere, B.; Cortes, D. Curr. Med.
Chem. 2009, 16, 2441−2467.
(11) Millan, M. J.; Maiofiss, L.; Cussac, D.; Audinot, V.; Boutin, J.
A.; Newman-Tancredi, A. J. Pharmacol. Exp. Ther. 2002, 303, 791−
804.
(12) Kebabian, J. W. Life Sci. 1978, 23, 479−483.
(13) Missale, C.; Nash, S. R.; Robinson, S. W.; Jaber, M.; Caron, M.
G. Physiol. Rev. 1998, 78, 189−225.
(14) Beaulieu, J. M.; Espinoza, S.; Gainetdinov, R. R. Br. J.
Pharmacol. 2015, 172, 1−23.
(15) Wang, S.; Che, T.; Levit, A.; Shoichet, B. K.; Wacker, D.; Roth,
B. L. Nature 2018, 555, 269−273.
(16) Chien, E. Y. T.; Liu, W.; Zhao, Q.; Katritch, V.; Won Han, G.;
Hanson, M. A.; Shi, L.; Newman, A. H.; Javitch, J. A.; Cherezov, V.;
Stevens, R. C. Science 2010, 330, 1091−1095.
(17) Wang, S.; Wacker, D.; Levit, A.; Che, T.; Betz, R. M.; McCorvy,
J. D.; Venkatakrishnan, A. J.; Huang, X.-P.; Dror, R. O.; Shoichet, B.
K.; Roth, B. L. Science 2017, 358, 381−386.
(18) Cabedo, N.; Protais, P.; Cassels, B. K.; Cortes, D. J. Nat. Prod.
1998, 61, 709−712.
(19) Andreu, I.; Cortes, D.; Protais, P.; Cassels, B. K.; Chagraoui, A.;
Cabedo, N. Bioorg. Med. Chem. 2000, 8, 889−895.
(20) Cabedo, N.; Andreu, I.; Ramírez de Arellano, M. C.; Chagraoui,
A.; Serrano, A.; Bermejo, A.; Protais, P.; Cortes, D. J. Med. Chem.
2001, 44, 1794−1801.
(21) Andreu, I.; Cabedo, N.; Torres, G.; Chagraoui, A.; Carmen
Ramírez de Arellano, M.; Gil, S.; Bermejo, A.; Valpuesta, M.; Protais,
P.; Cortes, D. Tetrahedron 2002, 58, 10173−10179.
(22) El Aouad, N.; Berenguer, I.; Romero, V.; Marín, P.; Serrano, A.;
Andujar, S.; Suvire, F.; Bermejo, A.; Ivorra, M. D.; Enriz, R. D.;
Cabedo, N.; Cortes, D. Eur. J. Med. Chem. 2009, 44, 4616−4621.
(23) Berenguer, I.; Aouad, N. E.; Andujar, S.; Romero, V.; Suvire, F.;
Freret, T.; Bermejo, A.; Ivorra, M. D.; Enriz, R. D.; Boulouard, M.;
Cabedo, N.; Cortes, D. Bioorg. Med. Chem. 2009, 17, 4968−4980.
́
(24) Parraga, J.; Cabedo, N.; Andujar, S.; Piqueras, L.; Moreno, L.;
́
Galan, A.; Angelina, E.; Enriz, R. D.; Ivorra, M. D.; Sanz, M. J.;
Cortes, D. Eur. J. Med. Chem. 2013, 68, 150−166.
(25) Moreno, L.; Cabedo, N.; Ivorra, M. D.; Sanz, M.-J.; Castel, A.
L.; Alvarez, M. C.; Cortes, D. Bioorg. Med. Chem. Lett. 2013, 23,
́
4824−4827.
(26) Suvire, F. D.; Cabedo, N.; Chagraoui, A.; Zamora, M. A.;
Cortes, D.; Enriz, R. D. J. Mol. Struct.: THEOCHEM 2003, 666−667,
455−467.
(27) Andujar, S. A.; Tosso, R. D.; Suvire, F. D.; Angelina, E.;
Peruchena, N.; Cabedo, N.; Cortes, D.; Enriz, R. D. J. Chem. Inf.
Model. 2012, 52, 99−112.
(28) Andujar, S.; Suvire, F.; Berenguer, I.; Cabedo, N.; Marín, P.;
Moreno, L.; Ivorra, M. D.; Cortes, D.; Enriz, R. D. J. Mol. Model.
2012, 18, 419−431.
(29) Browman, K. E.; Curzon, P.; Pan, J. B.; Molesky, A. L.;
Komater, V. A.; Decker, M. W.; Brioni, J. D.; Moreland, R. B.; Fox, G.
B. Pharmacol., Biochem. Behav. 2005, 82, 148−55.
G
https://dx.doi.org/10.1021/acs.jnatprod.9b00921
J. Nat. Prod. XXXX, XXX, XXX−XXX